Studying and Conserving Paintings

Studying and Conserving Paintings
Occasional Papers on the Samuel H. Kress Collection

Studying and Conserving Paintings
Occasional Papers on the Samuel H. Kress Collection
in association with
The Conservation Center of the Institute of Fine Arts,
New York University
A private university in the public service
Archetype Publications Ltd.
6 Fitzroy Square
London w1t 5hj
www.archetype.co.uk
tel: 44(207) 380 0800
fax: 44(207) 380 0500
in association with
The Conservation Center of the Institute of Fine Arts, New York University
New York, New York
isbn 1-904982-06-9
©2006 Archetype Publications Ltd., The Conservation Center of the

Institute of Fine Arts, New York University, and the authors
Design, typography, and production by Gail A. Cimino

Color separations and printing in the U.S.A. by Thames Printing, Norwich, Connecticut
On the cover: detail of View of the Molo, Canaletto, ca. 1725.

Oil on canvas, 26 1/2 × 32 3/4 in. (67.3 × 83.2 cm).
Columbia Museum of Art, Columbia, South Carolina,
Samuel H. Kress Collection (cma 1954.44).
The identification of the medium throughout this volume in some cases is conjectural,
based on its visual characteristics and published analyses of other works by the artist.
iv
Mario Modestini ca. 1950 treating Hagar and the Angel by Bernardo Strozzi, today found in the Kress Collection
at the Seattle Art Museum. Mr. Modestini’s decades of service to the Kress Collection and his vast knowledge
have inspired us all.
v
Contents
Overview
3 The Samuel H. Kress Collection: Conservation and Context
Marilyn Perry
7 The Samuel H. Kress Program in Paintings Conservation at the
Conservation Center of the Institute of Fine Arts, New York University
Margaret Holben Ellis
11 Introduction to the Volume
Michele Marincola
15 Acknowledgements
Michele Marincola
Historical Papers
19 Philosophies and Tastes in Nineteenth-Century Paintings Conservation
Wendy Partridge
31 Stephen Pichetto, Conservator of the Kress Collection, 1927– 1949
Ann Hoenigswald
43 Mario Modestini, Conservator of the Kress Collection, 1949– 1961
Dianne Dwyer Modestini with Mario Modestini
Technical Studies & Treatment

65 A New Leaf: Recent Technical Discoveries in the Goodhart Ducciesque
Master’s Madonna and Child with Four Saints
Jennifer Sherman
79 Botticelli’s Nativity
Charles R. Mack
95 The Re-use of a Desco da Parto
Mika Okawa & Dianne Dwyer Modestini
vi
99 The Triumphs of Petrarch : An Analysis of a Renaissance Decorative Cycle
Wendy Partridge
113 A Portable Triptych in El Paso
Dianne Dwyer Modestini
129 Guidoccio Cozzarelli’s Scenes from the Life of the Virgin
133 School of Pietro Perugino, Saint Sebastian
Annette Rupprecht & Sheri Francis Shaneyfelt
145 The Master of the Manchester Madonna: Restoration, Technique, and a Context
for Attribution
Molly March
165 Portrait of a Lady and Techniques in the Late Paintings of Nicolaes Maes
Laurent Sozzani with Christopher McGlinchey
189 View of the Molo : A Canaletto Attribution Reinstated
Elise Effmann
197 Canaletto Paints the Molo from the Ponte della Paglia
Katharine Baetjer
207 View of the Grand Canal with Dogana and Guardi Studio Practices
Helen Spande
Appendices & Index

215 Appendix i: Kress Collection Paintings Treated in the Samuel H. Kress
Program in Paintings Conservation at the Conservation Center of the
Institute of Fine Arts, New York University
222 Appendix ii: The Distribution of the Kress Collection

225 Index
vii
Overview
The Samuel H. Kress Collection:
Conservation and Context
Marilyn Perry
President, Samuel H. Kress Foundation
O ld Master paintings survive to us as fragile remnants of former

worlds, where once they served a specific purpose—to evoke religious
devotion, to capture the image of a beloved face, to illustrate a moral
tale on a piece of furniture—in lives and settings from which they are
long removed. Centuries of changing owners and evolving uses obscure
their history and dim their original luster and meaning. Eventually, they are
vulnerable to being treasured in a kind of half-light, as lost and mysterious
objects of faded beauty and mystery.
To re-illuminate Old Master paintings requires the professional skills of art
conservators and art historians. Painstaking conservation relieves a picture of
grime and damage, recapturing the appearance the artist intended and often
explaining the means by which it was achieved. Painstaking historical research
reveals elements of the identity of the work of art, including why and where it
was painted, by whom and for whom, how it was viewed by its first audience.
Together, the conservator and the historian craft what might be called the
biography of the object, bringing it out of the shadows and into the realm of
understanding where its uniqueness can touch our lives.
The papers in this book record these processes, with an emphasis on the
elements of discovery that accompany paintings conservation. They are pub-
lished to share these discoveries, and to mark the first decade of the Kress
Program in Paintings Conservation, a program of advanced training in Old
Masters conservation sponsored by the Kress Foundation at the Conservation
Marilyn Perry 3
Center of the Institute of Fine Arts of New York photography, publications, and conferences that
University. All of the paintings discussed in this sustain careers dedicated to European art.3
volume are today enfolded in the Samuel H. Kress These constituencies often converge, as in
Collection, and have, in effect, arrived at the end the Kress Program in Paintings Conservation,
of their historical wanderings as part of the where the dual purpose has been to support the
permanent holdings of one of the eighteen Kress advanced training of talented paintings conser-
Regional Collections. They are selected for pres- vators and to provide appropriate conservation
entation on the basis of interest for the serious treatment for Old Master paintings in the Kress
student of the field, from more than 100 paintings Regional Collections. Selected works of art are
in the Kress Collection that have thus far been shipped to the Conservation Center (shipping,
treated through the program. insurance, and photography are paid by the
The publication is divided into two comple- museum) for conservation treatment by Kress
mentary parts. The first provides a broad and Conservation Fellows under the supervision of
valuable overview of historic approaches to the Dianne Dwyer Modestini, the consultative con-
conservation of European paintings, including servator of the Kress Collection, and the active
two papers related to the conservation history of interest of Mario Modestini, who guided the
the Kress Collection, which was formed between Foundation’s conservation program in the 1950s.
1929 and 1961. The remainder of the book is For the Fellows, the opportunity to work closely
devoted to information gained in the process of with experts on a significant range of European
conserving specific Italian and Dutch Old Mas- paintings offers unparalleled hands-on profes-
ters, a combination of scientific and humanistic sional experience. As a further component, the
research that opens new avenues for understand- program maintains an inventory of the condition
ing the work of art, its history, and the original of objects in the Kress Collection, insuring a
context to which it belonged. All of the papers consistency of approach based on previous con-
were contributed by individuals associated with servation history that is also unusual. The rigor-
the Kress Program in Paintings Conservation ous nature of the training and the emphasis on
during its first ten years. The book is sponsored art historical research as well as scientific testing
by the Kress Foundation in celebration of these have resulted in the recovery of many beautiful
achievements, and also in demonstration of the European paintings, some of which—as indicated
vision that guides Kress philanthropy. in the papers that follow—have also recovered
The Samuel H. Kress Foundation has focused significant elements of their history and meaning.
on European art and architecture for seventy-five Which brings us full circle to the Kress Foun-
years. Across the United States, museums possess dation’s larger programmatic goals. As custodians
masterpieces from the vast Kress Collection— of European art, we recognize the value inherent
more than 3,000 European paintings, sculptures, in comprehending as much as possible about the
bronzes, drawings, and works of decorative art.1 original place and purpose for which centuries-
In Europe and the Mediterranean region, the old works of art were created. Kress grants are
Foundation has regularly sponsored the preser- directed, from many points of view, toward this
vation of archaeological sites and architectural end, which we call The Art of Europe in Context.
monuments2 —i.e., the settings for which por- The papers in this volume are a signal contribu-
table art was created. And since the early 1960s, tion. We salute all of the individuals who have
a program of fellowships and broad support made the first decade of the Kress Program in
for the essential tools of academic and scientific Paintings Conservation an enduring success. The
research has underwritten the training of more paintings, the museums, the conservators, and the
than 4,000 art historians, conservators, and public have all benefited.
preservationists, and the archives, databases,
4 Overview
Notes
1. For a history of the creation and distribution of the Kress
Collection, see Marilyn Perry, “The Kress Collection” in
Chiyo Ishikawa et al. (eds.), A Gift to America: Masterpieces of
European Painting from the Samuel H. Kress Collection (exhib.
cat.). New York: Harry N. Abrams, 1994, pp. 12–39.
2. From 1929 to the mid-1980s, the Foundation sponsored
the preservation of sites in Italy, Greece, Germany, France,
Spain, and Ireland. In 1987, the Kress Foundation
European Preservation Program was created with the
World Monuments Fund to offer competitive incentive
grants. To date, this program has aided more than 250 sites
in 49 countries. Surveys of the funded projects are avail-
able from the Kress Foundation.
3. An overview of the first forty years of the program—
Launching Careers in Art History and Conservation: The Kress
Fellowship Program –—was published by the Kress
Foundation in 2003.
1
Marilyn Perry 5
The Samuel H. Kress Program
in Paintings Conservation at the
Conservation Center of the Institute
of Fine Arts, New York University
Margaret Holben Ellis
Professor of Conservation
Sherman Fairchild Chairman, –
I n order to take full advantage of its location in New York City,

the curriculum of the Conservation Center of the Institute of Fine Arts,
New York University, needs to be responsive to the constantly changing
resources around it: the availability of visiting conservators, the arrival
of a new piece of scientific equipment in a nearby laboratory, and the
never-ending succession of museum exhibitions and their adjunctive “special
projects,” for which an extra set of hands or the inquiring mind of a conserva-
tion student is needed. Not only must courses at the Center adapt to local
developments and opportunities, but they also need to accommodate broad
philosophical shifts in the conservation profession at large. Such was certainly
the case at the Conservation Center in the late 1980s and specifically within the
discipline of paintings conservation, which had weathered a decade of highly
charged debate over the extent and nature of picture cleaning. In general, it
was argued that while the so-called “objective” method of cleaning paintings
minimized the imposition of passing aesthetic tastes on how a picture looked,
one example being the sweet-faced Italian Renaissance Madonnas favored
by restorers and their clients at the turn of the nineteenth century, the result-
ing visual discord between damaged and intact passages was distracting and
even farther removed from the painting’s original appearance, much less one
appropriate to its age. In other words, because a picture does not change in
appearance via any consistent progression, the “what you see is what you get”
end product of such a cleaning rarely invokes the originally intended effect.
Margaret Holben Ellis 7

Instead, the lines of reasoning went, a conservator Thus, it was in the fall semester of 1990, that
should partially or selectively remove discolored a pilot course for the conservation of Kress
varnish and more fully reintegrate discordant Collection paintings was incorporated into the
passages through inpainting and the limited appli- curriculum of the Conservation Center. Several
cation of glazes and toned varnishes. The final paintings from Kress Regional and Study Collec-
“look” of the painting, therefore, reflects the tions came to East 78th Street to be cleaned
prevailing aesthetic of the period in which it was and treated by graduate students under the super-
created, its original function, under what condi- vision of Dianne Dwyer Modestini. Her intimate
tions and circumstances it was viewed, and most knowledge of the Kress paintings combined with
importantly bears witness to the fact that the her considerable experience in training young
painting has aged in a unique and irreversible way, conservators enabled her to tailor treatments to
but can still be enjoyed as a unified work of art. fit each student’s level of skill. Continued collab-
In acknowledging this shift in the profession, oration with the Institute’s art history faculty was
John Brealey, a leading proponent of the “human- bolstered by the presence at the Conservation
ist” approach to cleaning paintings, began teach- Center of Kress Collection paintings, which lent
ing graduate students at the Conservation Center themselves to object-based study. Cooperating
soon after his appointment in 1975 as Chairman museums, in return, not only received treatments
of the Paintings Conservation Department of the of the highest quality, but also all supporting
Metropolitan Museum of Art. In 1980, Dorothy documentation, including art historical and tech-
Mahon, also of the Metropolitan Museum of nical analyses.1
Art, joined the faculty. In 1986, the Conservation At the end of this trial course, it was con-
Center retained Charles von Nostitz, a leading cluded that, while the treatments were eminently
conservator of European Old Master paintings, successful, two semesters were required for the
to assist Professor and Chairman Lawrence students to complete their work. Accordingly, a
Majewski in the instruction of paintings conser- proposal was submitted to the Foundation
vation. Dianne Dwyer Modestini quickly followed requesting that a Post Graduate Fellowship be
in 1989, and in 1990 George Bisacca, an expert in established in order to provide full-time on-site
the structure of panel paintings at the Metropoli- supervision for two consecutive semesters. On
tan Museum of Art, began to teach in the newly May 29, 1991 the Board of Trustees of the Kress
revamped course, Technology and Structure of Foundation approved the establishment of a Post
Works of Art, part of the “core” curriculum of Graduate Fellowship for Advanced Training in
the Conservation Center. Paintings Conservation, a position whose roles
Concurrently, as described in Dr. Marilyn and responsibilities would be transformed over
Perry’s foreword, the Kress Foundation had the following decade to meet the changing needs
expressed concerns as early as 1983 regarding the of the program.2, 3 Kress Post Graduate Fellows
condition of the paintings that had been distrib- have been responsible for assisting other conser-
uted to Kress Regional and Study Collections vation faculty, teaching small topical workshops,
around the United States. While many remained organizing new courses, overseeing the mechanics
sound and required only superficial cosmetic of arrival, uncrating and documentation of Kress
attention, many older restorations had made the Collection paintings, assisting in the selection of
works unexhibitable because of thick, yellowed paintings for treatment, monitoring Kress Reg-
varnishes and unsightly passages of discolored ional and Study Collections, and recommending
retouching. Structural problems were also iden- the purchase of several pieces of equipment to
tified in wood panel paintings, such as flaking facilitate treatments and insure the safety of the
paint, which had been exacerbated by fluctuating students.4 In return, the Fellows have attended
environmental conditions. numerous lectures both in conservation and art
8 Overview
history, gained experience in panel work on Kress Regional and Study Collections are well
selected Kress paintings, and traveled abroad to documented. Word of the program has spread
attend workshops and undertake related studies.5 throughout the museum and conservation com-
Students enrolled in the Kress Program in munities with the result that well-qualified under-
Paintings Conservation have benefited directly graduates and Fellowship candidates are applying
from the noteworthy paintings in Kress Regional to the Conservation Center to avail themselves
and Study Collections. Their treatments and of this unparalleled educational opportunity.
research have led to presentations at graduate stu- Today, the results of the Foundation’s unflagging
dent and professional conferences and have also commitment to the Kress Collection and its long-
served as topics for Qualifying Papers required standing support of the Conservation Center
for the Institute’s rigorous Master’s Degree are visually delighting museum visitors across
in Art History. Especially meaningful for the the country, informing technical art history and
students has been the presence of the beloved connoisseurship studies, and educating the eyes
Mario Modestini, who has shared his extensive and hands of the conservators who will preserve
knowledge and expertise on Italian paintings, similar artistic treasures in the future.
from the primitives through the Renaissance and,
in particular, on gold-ground tempera paintings, Notes
a rare attribute among American conservators 1. Museums are responsible for costs associated with packing,
shipping, insurance, materials, and photography.
and only slightly less so worldwide. This aspect 2. Annette Rupprecht 1991– 1995
of the program is especially relevant to the Kress Jennifer Sherman 1994– 2000
Collection since the majority of over 500 early Friederike Steckling 1998– 2000
Molly March 1999– 2002
Italian paintings have gold grounds or some form Sue Ann Chui 2001– 2002
of gold embellishment. The students and Fellows Nica Gutman 2002– 2005
have been privileged to work alongside one of the 3. In order to better reflect the responsibilities of the Kress
Post Graduate Fellow the title of the position has been
greatest restorers of the past century. changed to Associate Conservator for the Kress Program in
Graduates of the Kress Program in Paintings Paintings Conservation.
Conservation have gone on to work in museums 4. The Samuel H. Kress Foundation generously funded
equipment purchases separately.
and private studios both in this country and 5. The Samuel H. Kress Foundation provided financial assis-
abroad, including the Museum of Modern Art, tance to the Fellows for supplemental studies.
Los Angeles County Museum, Guggenheim Mus-
eum, Metropolitan Museum of Art, Brooklyn
Museum of Art, Walters Art Museum, Baltimore
Museum of Art, North Carolina Museum of
Art, Kimball Art Museum, J. Paul Getty Museum,
National Gallery of Art, Rijksmuseum (Amster-
dam), Öffentliche Kunstsammlung (Basel), Fon-
dation Beyeler (Basel), Hamilton Kerr Institute
(Cambridge), Uffizi Galleries (Florence), Louvre
(Paris), and the Kunsthistorisches Museum
(Vienna), among others.
Now in its fifteenth year, the Samuel H. Kress
Program in Paintings Conservation has become an
essential component of the Conservation Center’s
curriculum. Its benefits to graduate art history
and conservation students, Institute of Fine Arts
faculty, Post Graduate Fellows, and participating
Margaret Holben Ellis 9

Introduction to the Volume
Michele Marincola
Sherman Fairchild Chairman and Professor of Conservation
T his collection of fifteen papers will, we hope, be the first in a series

of occasional papers devoted to technical studies of the Kress Collection.
Recent graduates and faculty of the Conservation Center of the Institute
of Fine Arts, as well as scholars in the fields of art history and paintings
conservation have contributed to this volume. Much of their research
was prompted by questions that arose during treatment of the paintings at the
Conservation Center, and the resulting papers are grouped around two major
themes of interest to the field of paintings conservation. The first, the history
of conservation, is a topic whose increasing exploration gives an indication of
the maturation of our profession. Conservation history as a formal study has
been taken up by a number of researchers in the United States and abroad;
seminal texts in the development of modern conservation theory are to be
found in the 1996 volume of Readings in Conservation published by the Getty
Conservation Institute.1 Since 1975 there have been publications from the
Working Group of the International Council of Museums–Committee for
Conservation (icom-cc) devoted to conservation history, and more recently,
the topic was a focus of the 2003 American Institute for Conservation annual
meeting, inviting papers in the field of paintings by (among others) Elizabeth
Darrow and Wendy Partridge.2 Certain to become a definitive textbook on
the subject is the new Readings in Conservation anthology, Issues in the Conservation
of Paintings, which contains a wide range of historical and contemporary writ-
ings.3 In Europe, N.S. Brommelle,4 Michael von der Goltz,5 Christine Sitwell,6
Michele Marincola 11
and Sylvie Béguin de Sudurat, among many collection formed by Samuel H. and Rush Kress.
others, have published significant accounts of As Hoenigswald writes, Pichetto fulfilled a
paintings conservation history. Here in the U.S. number of roles for the Kress brothers: restorer,
the oral history projects of the International acquisitions advisor, connoisseur, researcher, col-
Museum of Photography and Film7 and the lections care manager and installation designer.
Foundation of the American Institute for Con- Mario and Dianne Dwyer Modestini offer the
servation have gathered the professional histories next chapter in the history of conservation at the
of American conservators and scientists. In the Samuel H. Kress Foundation: Mario’s role as its
specific field of paintings, the Yale University curator and conservator, called to New York
symposium concerning the treatment history of by Rush Kress after Pichetto died in 1949. In a
the Jarvis Collection,8 and the conservators Jean charming, first-person narrative told to his wife
Portell, Rebecca Rushfield, Eric Gordon,9 Katie Dianne, who is paintings conservator for the
Swerda10 and Joyce Hill Stoner, have all made Kress Collection and adjunct professor at the
important contributions to the record of Amer- Conservation Center, Mario, the “lone survivor
ican conservation history. This volume offers three of those years,” recounts his life in New York,
papers that further expand our critical knowledge Washington and the Pocono Mountains working
in this area, as reflected through the lens of the on the Collection. From his first amusing descrip-
early Italian picture collection of the Samuel H. tion of cleaning a Paolo di Giovanni Fei to “show
Kress Foundation. what he could do”—using a mixture of Pond’s
Wendy Partridge (ifa /cc 1999), a paintings cold cream, Marseilles soap and linseed oil—
conservator and scholar of conservation history, to his collaborations with the scientist Robert
presents a comparison of two nineteenth-century Feller on field trials of new inpainting media
philosophies towards the treatment of early and varnishes, we are captivated by his ingenuity,
Italian paintings that will prove useful to readers immense skill and professional modesty.
who wish to look beneath the overblown rhetoric The second gathering of papers explores the
of recent restoration controversies. Partridge’s interrelated themes of technical study and treat-
well-balanced paper contrasts Eastlake’s approach ment. The Kress paintings discussed in this sec-
to paintings as aesthetic objects requiring inter- tion are by Italian masters, with one exception—
pretation (cleaning and sometimes extensive a paper on the techniques in the late paintings
retouching), with Cavalcaselle’s reluctance to of Nicolaes Maes. They are presented in chrono-
compensate for loss lest the inpainting interfere logical order, beginning with a panel from the
with the historical record of the object. These Trecento and ending with Guardi’s View of the
issues continue to engage museum curators and Grand Canal with Dogana. In most cases, a discovery
conservators today when they contemplate the about or reconsideration of a picture occurred
treatment of a painting. because it was undergoing examination and treat-
Ann Hoenigswald and Dianne Dwyer Modestini ment at the Conservation Center as part of a class
present two important chapters in American in the Kress Program in Paintings Conservation.
paintings conservation history that relate directly The students established the condition of the
to the Kress Collection. Hoenigswald has written paintings and identified materials used in their
the first history of Stephen Pichetto and his making and subsequent repair, both necessary
role as a picture restorer in New York City and steps in formulating a treatment approach. Anal-
Washington, D.C. during the first half of the ysis was also carried out in the service of techni-
twentieth century. Her account of this restorer, cal connoisseurship, or, as it is sometimes called,
who operated a large and active workshop that technical art history,11 the discipline within
left its mark on thousands of pictures, fills a art history in which physical data gathered
significant gap in the history of the painting from works of art are applied to the study of
12 Overview
workshop practice, authorship, function or origi- the original function of the panels, discuss their
nal context, and authenticity. As Elise Effmann possible attribution, and explore occasions for
(ifa /cc 2000) writes in her paper “View of the Molo: their commission. J.J.G. Alexander, Sherman
A Canaletto Attribution Reinstated,” technical Fairchild Professor of Fine Arts at the Institute of
studies of individual paintings and artists’ tech- Fine Arts, saw the paintings at the Conservation
niques have “been important in establishing a Center and identified the tiny coats of arms con-
clearer chronology and also in resolving issues of cealed in two panels as belonging to the Gonzaga
attribution.” Evidence uncovered during cleaning, and Sforza families, thereby finding internal evi-
knowledge accrued from prolonged observation, dence for the circumstances of its commission.
or a visit from a local scholar, have led the authors Comparison of otherwise hard-to-see details
to draw new conclusions about the paintings. in related pictures can often augment traditional
Jennifer Sherman (ifa /cc 1997), paintings methods of art historical analysis. Dianne Dwyer
conservator and adjunct professor at the Conser- Modestini presents her detailed study of the
vation Center, combines our two themes in one Sienese mid-fifteenth-century Kress triptych in El
paper. In her discussion of the Trecento polyptych Paso within the context of four similar portable
Madonna and Child with Four Saints in the Birming- triptychs. Deftly considering the condition, paint-
ham Museum of Art attributed to the Goodhart erly quality, painting technique and punchwork
Ducciesque Master, Sherman presents a study designs in each work, she looks afresh at a thorny
of technique and materials that shed light on its problem of attribution among closely related
now-lost, original appearance. In the course of pictures. Professor Modestini also contributes a
her paper she considers the more recent history of short note on new thoughts about original con-
the picture and the possible role of Icilio Federico text and painting methods that were made possi-
Ioni (or Joni), a highly skilled nineteenth-century ble by the opportunity to study a painting during
gilder, restorer and forger. The art historian a conservation treatment—Guidoccio Cozzarelli’s
Charles R. Mack takes as a point of departure the Scenes from the Life of the Virgin (Lowe Art Museum,
cleaning by Mario and Dianne Dwyer Modestini University of Miami, Coral Gables, Florida).
of the transferred fresco of the Nativity by Annette Rupprecht, paintings conservator and
Botticelli (Columbia Museum of Art, Columbia, Sheri Francis Shaneyfelt, art historian, consider the
South Carolina), and uses it to re-evaluate the partially-preserved signature found during clean-
painting and the artist’s workshop practice. In this ing on Princeton University’s Saint Sebastian, attrib-
process, he allows us to see the painting better, uted to the School of Perugino, and tentatively
with more accuracy: established are Botticelli’s attribute the work to Eusebio de San Giorgio.
primary role in the execution of the fresco, the In her paper, paintings conservator Molly March
participation of workshop assistants (typical for (ifa /cc 2002) presents a careful reading in its
Botticelli, even on so small a work) and the hand cleaned state of the brushstrokes, color, and
of later restorers. Dianne Dwyer Modestini and layering structure of the Kress Foundation’s Virgin
Mika Okawa present new information uncovered Reading with Christ Child and Saint John; her treatment
during treatment about the original appearance of this picture attributed to the Michelangelo
and later re-use of an early Desco di Parto (Birth Associate led her to consider this long-overlooked
Tray). Wendy Partridge’s account of the cleaning painting in the context of recently cleaned pic-
of six decorative panels based on The Triumphs of tures attributed to Michelangelo or his circle.
Petrarch from the Denver Art Museum revealed the Certain technical details that emerged, such as
richness of the original painting, obscured by the distinctive hatch marks visible on this and two
layers of thick, yellowed varnish and discolored other pictures attributed to the same hand, will
retouching. As she writes in her paper, the result- help scholars assign other paintings, and perhaps
ing clarification of detail allowed her to determine even a name, to this anonymous master.
Paintings conservator Laurent Sozzani (with most innovative and rewarding research into the
Christopher McGlinchey, Museum of Modern history of art and conservation.
Art conservation scientist and adjunct professor at
the Conservation Center) examines a late portrait Notes
by the Netherlandish painter Nicolaes Maes in 1. Nicholas Stanley Price, M. Kirby Talley, Jr., and Alessandra
Melucco Vaccaro (eds.), Readings in Conservation: Historical
the Columbia Museum of Art. In his thorough and Philosophical Issues in the Conservation of Cultural Heritage.
examination of this and other pictures by Maes, Los Angeles, CA: Getty Conservation Institute, 1996.
Sozzani recreates the artist’s portrait painting 2. Elizabeth Darrow, “Alchemical transformations: Venice
as laboratory for the restorer’s art and science” (General
process in its skillful economy and presents an Session, p. 4), and Wendy Partridge, “Retouching paint-
unusual use of a uniform red glaze applied to the ings in Europe from the 15th through the 19th centuries:
background, perhaps unique to this painter, and debates, controversies, and methods” (Paintings Specialty
Group Session, p. 29), in Abstracts of Papers Presented at the
explains how Maes’s rapid application brought st AIC Annual Meeting, Arlington, Virginia, June –, .
intense color and depth to his abbreviated model- 3. David Bomford and Mark Leonard (eds.), Readings in Con-
ing of forms. servation: Issues in the Conservation of Paintings. Los Angeles,
CA: Getty Conservation Institute, 2004.
Elise Effmann’s detailed technical study of the 4. N.S. Brommelle, “Material for a history of conservation:
materials and methods of View of the Molo is an the 1850 and 1853 reports on the National Gallery,” Studies
excellent example of research completed by a con- in Conservation, Vol. 2, No. 4 (1956), pp. 176–88.
5. Michael von der Goltz, “Is it useful to restore paintings?
servator trained in art history, and complements Aspects of a 1928 discussion on restoration in Germany
perfectly the art historical consideration offered and Austria” in Preprints of the ICOM-CC th Triennial Meeting,
by Katharine Baetjer. The physical proximity of Lyons, August – September , . London: James & James,
1999, pp. 200–205.
the View of the Molo afforded Baetjer, a curator at 6. Christine Sitwell and Sarah Staniforth (eds.), Studies in
the Metropolitan Museum of Art, the opportu- the History of Painting Restoration. London: Archetype Pub-
nity to look anew at the picture and reconsider its lications, 1998.
7. P. Maynes and Grant B. Romer, “Documenting conserva-
demotion in the mid-twentieth century to a lesser tion through oral history: a case study” in Preprints of the
artist. This re-examination led her to compare the ICOM-CC th Triennial Meeting, Rio de Janeiro, September –,
Kress painting with another picture of the same . London: James & James, 2002, pp. 172–5.
8. Patricia Sherwin Garland, Early Italian Paintings: Approaches
view in Turin, and draw new conclusions about to Conservation. New Haven, CT: Yale University Art
the authorship of the two canvases and their Gallery, 2003.
place in the artist’s oeuvre. In the final paper, 9. Eric Gordon, “The restoration history of a late 15th-
century German altarpiece and how it reflects philosophi-
a short study by paintings conservator Helen cal trends in conservation” in Preprints of the ICOM-CC th
Spande (ifa /cc 2003) supports the attribution to Triennial Meeting, Rio de Janeiro, September –, .
Francesco Guardi of View of the Grand Canal with London: James & James, 2002, pp. 346–51.
10. Katie Swerda, “In their own words: painting conservation
Dogana in the Columbia Museum of Art through and restoration in 19th-century America according to the
a careful compilation of details not visible to artists and restorers who recorded their practices and
the naked eye; for example, X-radiography of beliefs” in Preprints of the ICOM-CC th Triennial Meeting,
Rio de Janeiro, September –, . London: James &
the picture disclosed an entirely unrelated image James, 2002, pp. 198–202.
under the one seen today, evidence of materials 11. David Bomford, “The purposes of technical art history,”
recycling and workshop frugality seen on other IIC Bulletin, No. 1 (February 2002), pp. 4–7; see also Erma
Hermens (ed.), Looking through Paintings: The Study of Painting
paintings by the same artist. Techniques and Materials in Support of Art Historical Research.
This collection of papers will serve not only to London: Archetype Publications, 1998.
re-acquaint us with some of the Kress Collection’s 1
Italian paintings at a level of detail not offered
before, but as a model for collaboration between
art historian and conservator, student and mentor,
or scientist and technical art historian. Such inter-
disciplinary alliances offer the best hope for our
14 Overview
Acknowledgements
M any people contributed to the success of this volume, from its concep-
tion through the final proofs of the color plates. Bruce Cole, formerly
of Indiana University and now Chair of the National Endowment for the
Humanities, has an in-depth knowledge of works in the Kress Collection
and has collaborated with colleagues and students over the years on
several projects related to the Collection. His suggestions in the early planning
stages were critical to the development of the project. Each of the authors
delivered manuscripts in a timely fashion and answered queries from the edi-
torial team with patience and grace. Jean Dommermuth assisted with the early
organization of the project. Constance Lowenthal was responsible for the criti-
cal editing of each paper. Connie worked closely with the authors to clarify
their arguments while retaining their authorial voice; her good sense and excel-
lent editorial skills have made this book a far better one than it would have been
otherwise. Dianne Dwyer Modestini, who plays a central role in the volume as
author of four of its papers, also assisted with many of the other papers from
the very beginning, suggesting fruitful lines of inquiry and carefully reading
the final texts. Many registrars, curators and museum staff helped us to obtain
photographs and permission to publish. In Jim Black, head of Archetype
Publications, we have found the ideal partner and distributor. His sound advice
and marketing expertise have helped us to make the book widely accessible. The
greatest thanks, however, are due to the Samuel H. Kress Foundation, its Board
and extraordinary staff, which for so many years have supported conservation
education at the Institute of Fine Arts. The Kress Program in Paintings Conser-
vation and its graduates are a powerful legacy to leave for the future. The book’s
elegant design is owed to Gail Cimino, who is responsible for Graphic Design
and Special Projects at the Kress Foundation. Gail took this project firmly in
hand, shepherding its editing through the final stages, designing the layout and
cover, and ensuring that the color of images matched, as closely as possible, that
of the original paintings. One other person, in particular, merits special men-
tion. As in so many of her projects, Lisa Ackerman, Executive Vice President
of the Kress Foundation, skillfully helped us all to produce our very best. Her
vision, keen intelligence, steady determination and good humor are the talents
that lie behind this book, from initial concept to realization.
Michele Marincola
Sherman Fairchild Chairman and Professor of Conservation
September 
Historical Papers
Fig. 1. Madonna and Child with the Infant Saint John, Giuliano Bugiardini, 1523–25, tempera and oil
on cradled wood panel, 44 1/2 × 32 in. (113 × 81 cm). Allentown Art Museum, Allentown, PA.
18 Historical Papers
Philosophies and Tastes in Nineteenth-
Century Paintings Conservation
Wendy Partridge
I n nineteenth-century Europe many of the difficult philosophical

questions of the conservation of paintings were articulated and debated
in print for the first time. The first of numerous books specifically on
the restoration of paintings, Christian Köster’s Über Restaurierung Alter
Ölgemälde, was published in 1827. Previously, limited information on
restoration had been available only in treatises on the art of painting.1 These
new writings argued the level to which cleaning should be taken and approaches
to the compensation of losses. These issues were related respectively to a
redefinition of patina in the nineteenth century2 and to the rise of the field of
connoisseurship. Major public controversies concerning restoration began to
surface at the end of the eighteenth century. In 1796 the Louvre organized an
exhibition of half-cleaned paintings to convince the public that the appearance
of the paintings was in fact improving with treatment.3 Between 1846 and 1853
the National Gallery, London was at the center of a cleaning controversy that
resulted in the House of Commons appointing a select committee to conduct
a public inquiry into management practices and cleaning procedures at the
Gallery. The committee’s interviews with artists, collectors, connoisseurs, and
restorers resulted in a 1,100-page report. Although restoration controversies
were not new in Europe, they were a new phenomenon in the public domain
and a direct result of the recent formation of national museums.4
In this environment, the restoration of early Italian paintings seems to have
been especially controversial. Since these paintings had been “rediscovered”
Wendy Partridge 19
after having been mostly ignored by connoisseurs,5
their restoration, like their collection and inter-
pretation, was often a subject of heated debate. In
this paper I will focus on only a handful of influ-
ential collectors, connoisseurs, and restorers of
early Italian paintings whose restorations embod-
ied two representative and contrasting approaches.
The attitudes of the first director of the
National Gallery, London, Sir Charles Eastlake
(1793–1865), and his restorer, Giuseppe Molteni
(1799–1867), like Eastlake an academically trained
painter, and the director of the Brera Gallery in
Milan during his last six years, will be contrasted
with those of the art historian Giovanni Battista
Cavalcaselle (1819–1897). Cavalcaselle studied
painting at the Accademia in Venice from 1835 to
1840 and with J.A. Crowe (1825–1896) wrote the
enormously influential New History of Italian Paint-
ing (1864) and A History of Painting in North Italy
(1871). In the 1870s he was appointed director of
the art department for the Ministry of Public
Education of the Italian State. In this capacity
Cavalcaselle was responsible for major conserva-
tion projects at San Francesco in Assisi, the Arena
Chapel in Padua, and the Camera degli Sposi in
Mantua, among others. His approach to these
restoration campaigns could be characterized as
archaeological, unlike Eastlake and Molteni’s
tendency to make additions and “corrections”
to paintings to bring them into conformity with
contemporary taste and the requirements of
nineteenth-century collecting. While Cavalcaselle
was primarily concerned with issues of stability
and retaining visible distinctions between original
and restoration, he was not immune from aspects
of the taste of his times, as we shall see below. In
general, Molteni and Eastlake saw, and therefore Fig. 2. Saint Michael, Piero della Francesca, 1470, oil on
poplar panel (identified), 52 3/8 × 23 3/8 in. (133 × 59.4 cm).
conserved, paintings primarily as aesthetic objects National Gallery, London. With the 19th-century
while Cavalcaselle tended to view and treat works restorations.
of art more as historical documents.
Eastlake’s circle consisted of the archaeologist
Sir Austen Henry Layard (1817–1894) whose
collection of Italian paintings was bequeathed
to the National Gallery, London in 1916, and
Giovanni Morelli (1819–1891), the Italian collector,
connoisseur, and writer of the seminal book on
attribution Italian Painters: Critical Studies of Their
Works. Morelli and Otto Mündler, the traveling
agent for the National Gallery, London, advised
Layard and Eastlake on the availability of paint-
ings. From the mid-1850s on, this group met
regularly in Molteni’s Milan studio where paint-
ings were examined, cleaned, attributed or reattrib-
uted, and often restored while waiting for export
licenses.6 Molteni’s restorations, then, were related
to the demands of the art market and collecting.
They often involved significant intervention and
overpainting, a reflection of Eastlake and Layard’s
discomfort with precise aspects of early, and non-
canonical, Italian painting.
The art market played a significant role in
restoration done for collectors, and in the nine-
teenth century we see a continuation of practices
that began with the formation of collections in
the sixteenth and seventeenth centuries. The poor
condition of a painting could be concealed, as it
was in Pisanello’s Virgin and Child with Saint George
and Saint Anthony Abbot, purchased by Eastlake in
1858. Eastlake described the painting in his note-
book as having a “blue sky almost rubbed to the
ground. The armour and dress of St. George once
beautifully finished but now almost obliterated.”7
The present, pristine appearance of the painting
is the result of Molteni’s interventions. Early
Italian paintings were also reformatted to hide
the irregular contours that showed they were
often fragments of multi-panel religious fur-
nishings. The Crivelli Pietà in the Metropolitan
Museum of Art is an early example of a fifteenth-
century painting in a seventeenth-century, stan-
dard Barberini frame. Reformatting probably
occurred more frequently in the nineteenth
Fig. 3. Saint Michael (fig. 2), with the 19th-century century than before, as a result of the growing
restorations removed.
demand from new museums and galleries.8 For
example, Piero della Francesca’s Saint Michael,
now at the National Gallery, London, was part
of an altarpiece where the central panel (now
lost) is thought to have been a Coronation of
the Virgin.9 The step and drapery on Saint
Michael’s right side had to be overpainted by
Molteni to disguise a fragmentary appearance
(figs. 2 and 3).
Wendy Partridge 21
A second nineteenth-century phenomenon, paintings had been whitewashed at the end of the
at least in Eastlake’s circle, was the professional eighteenth century, and in 1826 the Peruzzi family
removal of discolored varnish and old restorations was planning a new decorative cycle. However,
to determine attribution. This was connected to with the growing popularity of the “primitives,”
the emerging field of connoisseurship. An Ador- the family decided in 1840 to see if Giotto’s old
ation of the Kings had been attributed to Mantegna, mural cycle could be recovered.12 The wall paint-
but after Molteni’s cleaning, Layard attributed it ings were mostly not true fresco, but painted a
to Bramantino with Morelli and Eastlake concur- secco in a less stable glue medium. They began to
ring.10 Or, again, Mündler in 1862 wrote to Mol- suffer losses during the Renaissance and were
teni concerning a Virgin and Child with Infant Saint probably first restored as early as the last quarter
John and Other Saints that he believed to be by of the fifteenth century.13 Marini left the earlier
Mantegna despite the objections of both Morelli restorations intact and reconstructed only one
and Cavalcaselle. Mündler told Molteni, “you head of a bearded worthy from the Ascension of the
alone in the world can give life to [a painting] Evangelist and the torso of Saint Elizabeth from
extinguished by a very wicked restoration which The Birth of the Baptist. He also reinforced the
is hiding the author.”11 modeling and outlines of the pale images, result-
A dramatic example of a restorer revealing ing in a hardening of expression. The face of the
paintings that had had their “life extinguished” viol player in The Feast of Herod, for example, was
was Antonio Marini’s work on Giotto’s wall etherealized in the nineteenth-century restoration
paintings in the Peruzzi Chapel in Florence. The with a bow mouth and upraised eyes defined by
Fig. 4. Feast of Herod, Giotto, 1320, fresco. Peruzzi Chapel, Fig. 5. Feast of Herod (fig. 4), detail, with the 19th-century
Santa Croce, Florence. Detail, with the 19th-century restorations removed.
restorations.
the new outlines (figs. 4 and 5). Although the
restorer could have argued that he was merely
replacing lost original work, the overall result
seems to emphasize outline in a manner particular
to its time. The even, regular contours are remini-
scent of William Ottley’s linear illustrations
for Séroux d’Agincourt’s Histoire de l’Art par les
Monuments (1823), Carlo Lasinio’s Pitture a Fresco
del Campo Santo di Pisa (1832), or John Flaxman’s
“primitivizing” illustrations of Dante that pay
homage to early Italian painting.
A third phenomenon associated with paintings
restored under the supervision of Eastlake, Layard,
and other nineteenth-century collectors has to do
with the “corrections” made over original, undam-
aged paint. Morelli described Molteni in 1865 as: Fig. 6. Saint Jerome, Cosimo Tura, ca. 1470, oil and tempera
on wood panel (identified), 39 3/4 × 22 1/2 in. (101 × 57.2. cm).
a truly outstanding restorer, endowed as he is National Gallery, London. Detail during cleaning.
with a fine artistic sensibility and a passion for
ancient art. But because he is a pupil of our
Academies he occasionally takes part, just as it—This might be rectified by making the
your excellent Director of the National Gallery gilding of the nimbus a little more conspicu-
[Eastlake] often does, in the battle of the ous—the hair might also be brought down an
Academies to correct the naïve inaccuracies of inch & half on the forehead & the top of the
the Old Masters, which are almost always the head reduced—the nimbus would then also
result of their engaging easy-going manner. The require to be brought lower. The same defect
naïve imprudence of genius will never be under- (too much forehead & skull) is observable in
stood by the pedantry of our academicians.14 the little St. John & might be rectified—his
body is also a little too thick.17
In the nineteenth century, generally, there was
a willingness to add to a painting if it was felt to He did not purchase the painting, and these
improve its appearance. In 1837 Giovanni Bedotti changes were never made. Two paintings Eastlake
wrote in one of the nineteenth-century restora- did purchase and that Molteni restored were
tion books, De la Restauration des Tableaux, that to Cosimo Tura’s Saint Jerome and Cima da Conegliano’s
find a buyer the restorer might have to correct the David and Jonathan. Tura was a problematic painter
“errors” of the painter, although he should be for Crowe and Cavalcaselle as well as for Morelli.
careful to leave the characteristics of the painter’s Morelli described him as “morose,” “grotesque,”
style and period if possible.15 Since early Italian and a “hard, dry and angular painter, but often
paintings were often considered “feeble” and very impressive,”18 and Crowe and Cavalcaselle
problematic,16 it is understandable that they wrote on Tura’s work that:
especially were seen to need correction. He had no idea of selection; leanness, dryness,
Eastlake was willing to make corrections where paltriness, overweight of head and exaggerated
the draughtsmanship in the figures seemed prob- size of feet and hands, were almost invariable
lematic. In 1862, he was considering a Giuliano accompaniments of his pictures. In most of
Bugiardini Madonna and Child with the Infant Saint them it would seem as if well-fed flesh had
John for purchase (see fig. 1) and noted: become withered by want of nutrition … 19
The head of the infant C[hrist] is so placed Eastlake seemingly concurred as Saint Jerome’s
under a palm tree that the tree seems part of raised arm has been widened (fig. 6) and his
Wendy Partridge 23
Fig. 7. Madonna and Child; Saint John the Baptist; Saint Jerome, Sano Fig. 8. Madonna and Child; Saint John the Baptist; Saint Jerome (fig. 7),
di Pietro, ca. 1450–55, tempera on wood panel, 17 3/8 × 12 5/8 in. during treatment, with the 19th-century restorations removed.
(44.1 × 32.1 cm). The Metropolitan Museum of Art, New York,
NY. Portable triptych, with the 19th-century restorations.
exposed bony knee resting on the ground made Christ Child’s robes and around both figures’ eyes
less angular.20 Similarly, in the Cima, Jonathan’s and along the edges of the noses.25 The effect
thin leg has been made more massive and mus- seems to be a sweetening of the expressions and
cular.21 Not to suggest a continual correspon- a more regularized physiognomy.
dence in taste between Eastlake and Crowe and Finally, an examination of Bramantino’s Ador-
Cavalcaselle, it is interesting to note that the ation of the Magi (an 1862 Layard purchase from
latter writers, while favorably disposed to Cima, the Manfrin Collection in Venice that he sent to
remarked that he did not have Giovanni Bellini’s Molteni for restoration) is also revealing. Among
“largeness or breadth of the shape in figures.”22 other changes, Molteni extended and regularized
If Eastlake and other nineteenth-century col- the shadow falling on the building behind the
lectors were disturbed by weak draughtsmanship, Virgin and repainted the left side of the broken
they also appeared to have been bothered by the doorway to disguise a difference of color on the
notion that fifteenth-century works were, to quote lintel.26 Both changes tended to make the play of
William Ottley in 1826, “commonly deficient in light and shade across the building more rational.
the breadth of chiaroscuro …”23 The Sano di A final type of treatment Molteni used for
Pietro triptych from the Costabili Collection Eastlake and his circle was the application of a
owned by the Metropolitan Museum was proba- pigmented varnish. This tended to mute the
bly restored by Molteni’s pupil Luigi Cavenaghi colors by reproducing the look of an aged var-
(1844–1918) sometime before it was put up for nish, believed by many theoreticians to impart
sale in 1885.24 The restorer had recreated lost harmony to paintings. Köster, in Über Restaurierung
modeling on the face of the Madonna and to Alter Ölgemälde, wrote that disharmonies could be
the necks of both the Madonna and Christ Child compensated for by leaving dirt and old varnish
making the figures more three-dimensional. He on pictures and that thanks to this patina, “a pic-
also reinforced outlines, shortened such anatomi- ture could become even more harmonious than
cal oddities as John the Baptist’s long toes, and when made by its creator.”27 In 1837 Bedotti con-
for reasons that are difficult to understand, curred, explaining that “to clean a picture well,
changed Saint John’s hand holding the banderole one must know how to paint since a true artist
(figs. 7 and 8). The figures in a Giovanni Bellini in cleaning a picture is often forced to use the
Madonna and Child were also given additional grime which covers it to give harmony and effect
modeling around the same time, especially in the to the painting …”28 In 1846 in a letter to The
Times concerning the National Gallery, London’s
cleaning of paintings, John Ruskin lamented the
cleaning of Rubens’s War and Peace since with the
old varnish, the painting had:
mellowed by time into more perfect harmony
than when it left the easel, enriched and warmed
without losing any of its freshness or energy.
The execution of the master is always so bold
and frank as to be completely, perhaps, even
most agreeably seen under circumstances of
obscurity … 29
Conveniently, a pigmented varnish also helped

hide abrasions and damages to a painting.30
Harmony was an important nineteenth-century
critical concept, and Crowe and Cavalcaselle often
praised a work by noting its “soft harmony in
colours” and dismissed works with “violent con-
trasts.”31 It was also a quality Eastlake thought
early Italian paintings often lacked, writing in 1853: Fig. 9. Muse, Cosimo Tura, ca. 1455–60,
oil with egg tempera on poplar panel
a large portion of those early pictures are full (identified), 45 3/4 × 28 in. (116.2 × 71.1 cm).
of affectation and grimace; and many persons National Gallery, London. Before varnish
who have, or fancy they have, a taste for those removal.
pictures are insensible to essential elements of
painting, such as beauty of arrangement, har-
mony of coloring, and natural action and
expression.32
If there was not enough harmony, Eastlake

seemed to have been quite willing to add patina.
In 1861 he wrote to the restorer Raffaelle Pinti
concerning a Cima Virgin and Child:
The Cima da Conegliano would be improved,
not by removing anything but by first lowering
the tone of the Child’s head more nearly to the
neutral tone of the rest of his figure; and after-
wards by slightly warming the whole picture.
The obvious defect now is the difference in tone
between the Child’s head and body.33
There is no indication that this was ever done.

In Tura’s Muse however, that Layard bought in
1866 from the Costabili Collection, it appears
that Molteni used a pigmented varnish to tone
down the pink robe’s green lining, diminishing
the contrast between the complementary colors
(figs. 9 and 10).34 The attraction to patina found
an extreme manifestation in Gaetano Biachi’s Fig. 10. Muse (fig. 9), after varnish removal.
Wendy Partridge 25
attempt, while restoring the Bardi Chapel, to any retouching or reconstruction.41 Not only in
create the look of an aged varnish on Giotto’s Assisi, but in other sites in Italy this was the case,
frescoes.35 and in 1831 the local arts commission in Lucca
In contrast to Eastlake’s restoration practices instructed the restorer Michele Ridolfo to leave
was Cavalcaselle’s more archaeological approach. the large lacunae in frescoes by Amico Aspertini
Cavalcaselle opposed the integration of losses if these areas could not be reconstructed accu-
in the artist’s style or any sort of reconstruction rately.42
or additions; in 1877 he wrote regulations for Both Cavalcaselle and Morelli were involved in
restoration work undertaken by the State: the 1867–77 restoration campaign of Mantegna’s
Camera degli Sposi, and their points of view are
It does not matter if you recognize a restoration, interesting to compare. Cavalcaselle originally
in fact, you should be able to recognize it, since
vetoed a campaign in Mantua since both the
what is necessary is respect for the original work
at least for works belonging to the State. A lie, intonaco and paint layer were sound. Morelli, how-
even a beautiful lie, must be avoided. Scholars ever, felt strongly that the paintings would be
should be able to recognize in a restored picture improved if the overpaint from a past restoration
what is original and what is new … 36 effort was removed and if the faded colors of the
festoons and the illusionistic wall hangings were
Cavalcaselle was not the first person to articu- “refreshed.” Morelli and Cavenaghi were appoint-
late this view, and his position was the less pop- ed by the Minister of Education to undertake the
ular side of an ongoing debate. The head of the restorations. When the government fell shortly
Accademia in Venice, Pietro Selvatico, with whom after the appointment, Morelli lost his position,
Cavalcaselle had studied in Padua between 1840
and 1844,37 recommended in 1842 restricting the
treatment of paintings to structural stabilization.
The Florentine restorer Ulisse Forni in his 1866
book Manuele del Pittore Restauratore criticized this
recommendation. Forni countered that Selvatico
had advocated leaving paintings in ruins and
therefore making them impossible to appreciate.38
The restoration work supervised by Cavalcaselle
at the Arena Chapel (1868–71) and Assisi (1872–73),
not surprisingly, focused on stabilization not
reconstruction. In 1871 Cavalcaselle wrote that at
Assisi, “the work to be done comes down to
securing the intonaco which is threatening to fall
and stabilizing the paint which is separating from
the intonaco.”39 To prevent continuing water
damage to the frescoes, Cavalcaselle also urged
that the roof be repaired, the outside walls replas-
tered, and the windows sealed. There was no
provision for reconstruction of lacunae, and
losses were toned back with a neutral water-
color.40 There were precedents for this type of
treatment, and as early as 1836 various government
commissions were working to prevent further Fig. 11. Jacob Deceiving Isaac, Giotto, ca. 1290, Upper Church,
deterioration of the frescoes while prohibiting St. Francis, Assisi. After the Cavalcaselle restoration campaign.
and Cavalcaselle agreed to take over.43 This did project was, “to conserve what has remained of
not improve Morelli and Cavalcaselle’s often the old, restoring … to its primitive character
inimical relationship, and Morelli wrote highly even that part disfigured by additions and later
critical remarks concerning Cavalcaselle’s restora- changes.”47 In practice this involved a proposal
tion, accusing him of having destroyed the paint- to remove any Renaissance or Baroque additions
ings.44 Some of Morelli’s animosity was probably to the church, a re-gothicization common in
related to sloppy workmanship and poor materials projects all over Europe at the time. When the
used by Cavalcaselle’s restorers,45 but his criticism appearance of frescoes from the Upper Church
also appears to have been based on a different after Guglielmo Botti’s restoration in 1872 is
conception of how restored paintings should compared to a mid-twentieth-century and a
look. In 1912 Cavenaghi wrote of Cavalcaselle’s 1978–79 campaign, Botti’s restoration (supervised
restoration at Mantua that the system of “using by Cavalcaselle) seems to underscore the damaged
tints similar to the dominant color … forgot that and worn look of the images (figs. 11, 12, and 13).
restoration is an art and not a mechanical opera- While all three campaigns are similar in avoiding
tion,”46 probably expressing the by-then-deceased reconstruction, Botti’s restoration eschewed any
Morelli’s opinion as well. attempt to use his watercolor tone to integrate
Although Cavalcaselle’s restoration choices were the image.
archaeological, they also seem to have been related In conclusion, Cavalcaselle and Eastlake were
to a Romantic appreciation for pure, primitive scholars who cared passionately about early Ren-
simplicity and even a taste for the picturesque aissance painting and felt that they were present-
ruin. Cavalcaselle’s vision of the Assisi restoration ing these works in the best possible light. Because
Fig. 12. Jacob Deceiving Isaac (fig. 11), after a mid 20th-century Fig. 13. Jacob Deceiving Isaac (fig. 11), after the restoration cam-
restoration campaign. paign of 1978–79.
Wendy Partridge 27
of their different views concerning the nature of 8. Alessandro Conti, “Vicende e cultura del restauro” in
Federico Zeri (ed.), Storia dell’Arte Italiana . Part :
the paintings, however, they restored them in radi- Situazioni, Momenti, e Indagini, Vol. : Conservazione, Falso,
cally different ways. Furthermore, their concerns Restauro. Turin: G. Einaudi, 1981, pp. 68–9.
about levels of cleaning and loss compensation 9. Allan Braham, “The ‘improvement’ of Pre-Raphaelites:
case histories of some fifteenth-century Italian panels,”
still have not been definitively resolved, since there Apollo, Vol. 101 (May 1975), p. 360.
are usually no easy answers, and decisions often 10. Jaynie Anderson, “Layard and Morelli” in F.M. Fales and
can only be made on the basis of taste and aes- B.J. Hickey (eds.), Austen Henry Layard tra l’Oriente e Venezia.
Rome: L’Erma di Bretschneider, 1987, p. 114.
thetic judgment. Generation upon generation has 11. “Voi solo il unico al mondo potete renderli la vita estinta
reinterpreted works of art both in writing and da scelleratissimo restauro intanto che mascherà l’autore
through restoration, and it is difficult to maintain …” The Travel Diaries of Otto Mündler – (ed. C.T.
Dowd, introduction by Jaynie Anderson), The Walpole Society,
that a particular approach will ever be definitive. Vol. 51 (1985), p. 62.
12. Leonetto Tintori and Eve Borsook, Giotto: The Peruzzi
Acknowledgements Chapel. New York: Harry N. Abrams, 1961, pp. 32–4.
This paper began as a seminar report for a 13. Ibid., pp. 15 and 29.
14. Anderson 1993 (cited in note 7), p. 546.
wonderful class taught by Keith Christiansen. 15. Conti 1981 (cited in note 8), p. 77.
I would like to thank Keith Christiansen for 16. Séroux d’Agincourt remarked, “However useful some
his advice and encouragement as well as Dianne knowledge of the works of the earliest stages of the
Renaissance may be for the History of Art, it would be
Dwyer Modestini, Margaret Holben Ellis, Jennifer dangerous to pursue our studies of them too far: above
Sherman, and Sue Ann Chui for their support all, we must avoid the kind of enthusiasm felt by certain
and thoughtful criticisms during the many drafts modern artists for these experiments, which are still
too feeble from every point of view to serve as models.”
of this paper. Constance Lowenthal’s editorial Haskell 1976 (cited in note 5), p. 46.
suggestions were also superb. 17. The Travel Diaries of Otto Mündler (see note 11), p. 23.
18. Giovanni Morelli, Italian Painters: Critical Studies of their
Works, the Munich and Dresden Galleries. London: John
Wendy Partridge is Associate Paintings Conservator at the Murray, 1907, p. 128.
Intermuseum Conservation Association in Cleveland, Ohio. 19. J.A. Crowe and G.B. Cavalcaselle, A History of Painting in
She earned a degree in Paintings Conservation and an M.A. North Italy from the Fourteenth to the Sixteenth Century, Vol. 2.
London: John Murray, 1912, p. 228.
in Art History from the Conservation Center, Institute of 20. Jill Dunkerton, “Cosimo Tura as painter and draughtsman:
Fine Arts, New York University. the cleaning and examination of his Saint Jerome,” National
Gallery Technical Bulletin, Vol. 15 (1994), p. 43.
21. Braham 1993 (cited in note 9), pp. 366–7.
Notes 22. J.A. Crowe and G.B. Cavalcaselle, A History of Painting in
1. Monica Bianchi, “Patina: appunti per una definizione” North Italy from the Fourteenth to the Sixteenth Century, Vol. 1.
in Antonio Boschetto (ed.), Itinerari: Contributi alla Storia London: John Murray, 1912, p. 241.
dell’Arte in Memoria di Maria Luisa Ferrari, Vol. 3. Florence: 23. William Ottley, A Series of Plates Engraved after the Paintings
Studio per Edizioni Scelte, 1984, p. 106. and Sculptures of the Most Eminent Masters of the Early Florentine
2. Ibid., pp. 109–10. School. London, 1826, Introduction.
3. Philip Hendy, An Exhibition of Cleaned Pictures (–). 24. Anderson 1993 (cited in note 7), p. 546.
London: National Gallery, Printed for the Trustees, 1947, 25. Conti 1981 (cited in note 8), plates 84–5.
p. xviii. 26. Jill Dunkerton, “The technique and restoration of
4. Jaynie Anderson, “The first cleaning controversy at the Bramantino’s Adoration of the Kings,” National Gallery Technical
National Gallery 1846–1853” in Victoria Todd (ed.), Bulletin, Vol. 14 (1993), p. 44.
Appearance, Opinion, Change: Evaluating the Look of Paintings. 27. Conti 1981 (cited in note 8), p. 77.
London: ukic, 1990, p. 3. 28. “Pour bien nettoyer un tableau, il faut savoir peindre, car le
5. Francis Haskell, Rediscoveries in Art: Some Aspects of Taste, veritable artiste, tout en nettoyant un tableau, est souvent forcé
Fashion, and Collecting in England and France. Ithaca, ny: de se servir de la crasse qui le couvre pour donner de l’harmonie et
Cornell University Press, 1976, p. 45. de l’effet à la peinture …” Bianchi 1984 (cited in note 1),
6. Jaynie Anderson (ed.), “Dietro lo pseudonimo” in p. 111.
Giovanni Morelli, Della Pittura Italiana: Studii Storico-Critici. 29. Anderson 1990 (cited in note 4), p. 4.
Le Gallerie Borghese e Doria-Pamphili a Roma. Milan: Adelphi, 30. Alessandro Conti, “Giovanni Morelli ed il restauro amato-
1991, pp. 529–30. riale” in Giacomo Agosti, Maria Elisabetta Manca, and
7. Jaynie Anderson, “The rediscovery of Ferrarese Renais- Matteo Panzeri (eds.), Giovanni Morelli e la Cultura dei
sance painting in the Risorgimento,” The Burlington Magazine, Conoscitori; Atti del Convegno Internazionale, Bergamo, - Guigno,
Vol. 135, No. 1085 (August, 1993), p. 545. . Bergamo: P. Lubrina, 1993, p. 167.
31. Crowe and Cavalcaselle 1912 (cited in note 22), pp. 24 Photography Credits
and 244 or Crowe and Cavalcaselle 1912 (cited in note 19), Fig. 1, p. 18. Allentown Art Museum, Allentown, PA, Samuel
p. 229, for example. H. Kress Collection, 1960 (1960.10kb).
32. Haskell 1976 (cited in note 5), p. 54. Eastlake was not Figs. 2 and 3, pp. 20-21. ©National Gallery, London.
alone in this opinion. The Accademia di Belli Arti in Pisa Fig. 6, p. 23. ©National Gallery, London.
wrote in 1859 concerning the use of a wax consolidant on Fig. 7, p. 24. The Metropolitan Museum of Art, New York,
frescoes by Benozzo Gozzoli at the Campo Santo; they NY, Gift of Irma N. Straus, 1964 (64.189.4). Photograph
acknowledged that the wax had altered the colors throwing ©The Metropolitan Museum of Art, all rights reserved.
the “general harmony” out of balance, but felt this was Fig. 8, p. 24. The Metropolitan Museum of Art, New York,
acceptable since, “ora chi non sa che il pregio principale della NY, Gift of Irma N. Straus, 1964 (64.189.4). Photograph
pittura di quel secolo consiste più nella purezza del disegno, nella ©The Metropolitan Museum of Art, Department of
verità dell’espressione, che nel majestero del chiaorscuro o nel artificio Paintings Conservation, all rights reserved.
del colorito …” Conti 1981 (cited in note 8), pp. 79–80. Figs. 9 and 10, p. 25. ©National Gallery, London.
33. Jaynie Anderson, “A ‘most improper picture’: transforma- 1
tions of Bronzino’s erotic allegory,” Apollo, Vol. 139, No. 384
(February 1991), p. 22.
34. Tura’s Muse was examined at the National Gallery, London
before a conservation campaign in the 1980s. It was dis-
covered that the varnish had been deliberately tinted with
blackish and red-brown pigment particles, possibly by
Molteni. Layard had sent the Muse to Molteni in 1866.
Although no treatment records are extant, Molteni’s
correspondence and other letters referring to his work
mention artificial patination of paintings using a varnish
pigmented with Cassel earth. Jill Dunkerton, Ashok Roy,
and Alistair Smith, “The unmasking of Tura’s Allegorical
Figure: a painting and its concealed image,” National Gallery
Technical Bulletin, Vol. 11 (1987), pp. 9–13.
35. Conti 1981 (cited in note 8), p. 78.
36. “Poco importa che si conosca il restauro, che anzi lo si dovrebbe
conoscere, ma quello che è necessario si è che sia rispettato l’originale
della pittura almeno nelle opere appartenenti allo Stato. La bugia, anco
detto con bel garbo deve essere tolta di mezzo. Lo studioso potrà
conoscere da un dipinto restaurato a questa maniera quello che è origi-
nale da quello che è nuovo …” Donata Levi, Cavalcaselle, Il
Pionere della Conservazione dell’Arte Italiana. Turin: G. Einaudi,
1988, pp. 350–51.
37. Ibid., p. 7.
38. Ulisse Forni, Manuele del Pittore Restauratore. Florence, 1866,
pp. 12–13.
39. “Il lavoro da farsi si riduce a fermar gli intonachi che minacciano a
cadere, ed assicurare il colore che si isola dall’intonaco stesso.” Levi
1988 (cited in note 36), p. 337.
40. Ibid., pp. 337–9, 347.
41. Irene Hueck, “La Basilica Francescana di Assisi nell’otto-
cento: alcuni documenti su restauri progettati ed interventi
eseguiti,” Bolletino d’Arte Ser. vi, 66 (1981), p. 144.
42. Conti 1981 (cited in note 8), p. 80.
43. Michele Cordaro, “Vicende conservative dei dipinti
murali” in Michele Cordaro (ed.), Mantegna: La Camera degli
Sposi. Milan: Electa, 1992, pp. 334–6.
44. Giovanni Morelli, Italian Painters: Critical Studies of their Works.
The Borghese and Doria Pamfili Galleries in Rome. London: 1900,
p. 82.
45. Codaro 1992 (cited in note 43), p. 237.
46. “passare tinte approssimative al colore dominante … dimenticava che il
restaurare è un’arte e non un’operazione meccanica.” Conti 1981
(cited in note 8), p. 169.
47. “Conservare quanto è rimasto d’antico restituendo … al suo carattere
primitivo anche quella parte deturpata dalle agguinti o mutamenti pos-
teriori.” Levi 1988 (cited in note 36), pp. 340–41.
Wendy Partridge 29
Fig. 1. Madonna and Child with Saint Mary Magdalene and Saint Catherine, Pietro Lorenzetti, ca. 1330/40, tempera on wood
panel transferred to canvas, center panel 43 1/2 × 23 1/4 in. (110.5 × 59.1 cm); side panels each 40 × 19 1/2 in. (101.6 × 49.5 cm).
National Gallery of Art, Washington, D.C.
Stephen Pichetto, Conservator of
the Kress Collection, 1927– 1949
Ann Hoenigswald
I tion
n 1949 Rush Kress proposed the formation of a “central organiza-
of all American museums where owners and responsible keepers of

paintings could apply for advice in order to keep cultural treasures …
from falling into bad hands.” He went on to suggest that “if and when
a central institute for the technical care of works of art and especially of
paintings … will come into being this institution should bear the name of Mr.
Stephen Pichetto.”1
Rush Kress, following the lead of his brother Samuel H. Kress, focused
not only on the acquisition of art but also recognized that as collectors
they had a responsibility which extended to the care of the paintings and
sculpture as well. Although other collectors hired restorers to look after their
works of art, it was quite unusual that a twenty-year relationship was formed
which extended well beyond the normally defined responsibilities of a restorer.
Stephen Pichetto served Samuel and Rush Kress in a unique capacity.
Stephen Pichetto, the second child of recently arrived Italian immigrant
parents, was born in New York City in 1887. His father, Luigi, had arrived in
the United States from Genoa in 1882, two years after Fortunata who would
become Stephen’s mother.2 Although it has been suggested that Pichetto came
from a long line of old-world restorers, his father supported the family as a
chef.3 Almost nothing is known about Pichetto’s youth in New York City or
his training as a restorer. He was purported to have graduated from the selec-
Ann Hoenigswald 31
tive Townsend Harris High School and City one another—and mentioned to Rush Kress that
College as well as to have enrolled at the Art Pichetto was “one of the very few who can
Students League, but there are no records of his vouchsafe an opinion with absolute competence.”
attendance.4 His training as a restorer is equally Contini-Bonacossi went on to say, however, that
undocumented. Pichetto may have been intro- “I have never allowed him or others to guess that
duced to restoration by an uncle who lived briefly I have this intimate opinion of him, and I think
with the family. It has been suggested that this it good tactics to keep it so.”8 Until Pichetto
uncle took young Stephen with him to Italy where edged out “‘The Count,’ as he was called, as Sam
he introduced him to Italian art and possibly Kress’s principal advisor,”9 Pichetto and Contini-
made connections with Italian restorers. His Bonacossi worked closely together, but their cor-
family has remarked that he traveled in Europe respondence clearly reveals that they shared with
and studied the techniques of the Old Masters at one another only what they assumed the other
the National Gallery, London, but there is no wanted to hear. Theirs was a business partnership
confirmation of this.5 and unlike the relationship of genuine friendship
Until 1908, when he is listed in the New York based on respect and mutual commitment that
City Business Directory as a restorer with an Pichetto had with Kress.
establishment on East 28th Street, there is nothing On February 27, 1929 Pichetto testified on
official that links Stephen Pichetto to the field of behalf of Joseph Duveen in the trial of Hahn
restoration.6 However, owning a business at the v. Duveen, a lengthy case that hinged on the
young age of twenty-one implies that he must authenticity of a disputed painting attributed to
have had financial backing or his own resources. Leonardo da Vinci. Described as “a lowbrow and
Who his clients were at this early date is unknown. highbrow circus,”10 the trial involved every leading
Except for listings in the directories—at various name in the art world. Merely being associated
times as a restorer, an artist, and an art dealer— with the “season’s greatest extravaganza”11 meant
there is little information about him, his clients that one had secured a visible position and would
or his connections. However by the late 1920s his enjoy the free publicity that resulted from the
combination of a restorer’s skills, business acu- media attention. It was perhaps his association
men, and probably most important the requisite with Duveen that gave Pichetto his taste for the
personality, allowed him to become very success- good life and the confidence to believe that he
ful and to secure a prominent position in the art could achieve it. Duveen used the services of
world. This decade saw Pichetto working as a many restorers, but Pichetto was among his
restorer for the dealer, Joseph Duveen, being favorites, and they both benefited significantly
named consultant restorer at the Metropolitan from the relationship working in an era and in a
Museum of Art, and entering the circle of Count trade, “as Duveen practised it, that even a restorer
Alessandro Contini-Bonacossi, a Florentine dealer who worked for Duveen could leave a fortune.”12
and collector. He also began an association with Although it was and is not unheard of for
many of the major American collectors of his day, a restorer to associate himself with dealers, it
including Mellon, Lehman, Heinemann, Dale, required a certain personality to establish and
Walters, Warburg, Lewisohn, and Guggenheim then maintain equal footing. Pichetto’s demeanor
among others. Most importantly at this point, and appearance—always wearing a three-piece
however, he met Samuel H. Kress. Perhaps the suit and sporting a hat—and his practices of
introduction to Kress came through Duveen, who arriving at work in a chauffeur-driven car, staying
presumably knew Kress, and it may also be specu- at the best hotels, and riding in the drawing-room
lated that Kress met Contini-Bonacossi through compartment while his men were given berths on
Pichetto.7 Contini-Bonacossi respected Pichetto— the train implied the position he had reached and,
possibly because they benefited financially from what is more important, the image he wished to
project. John Walker’s description was probably Pichetto’s role extended far beyond that of a
not far off when he described Pichetto as: restorer; Kress discovered in him a confidant
and a connoisseur. There is little doubt that his
… a large well-fed bullfrog, perfectly tranquil but primary responsibility was to provide counsel on
ready to snap at any insect which might fly by.
the purchase of paintings based on their condi-
He had a cigar, lighted or unlighted always in his
enormous mouth. He would get up, invariably tion and undertake any necessary restoration, but
with an amiable smile and take me through room it becomes clear from reading their correspon-
after room where assistants are cleaning, inpaint- dence that Pichetto was the person through
ing, relining or cradling to point out some new whom all decision making was directed including
Kress acquisition.13 art historical advice, information on provenance,
iconography, attribution, and even the final
Described as a man who was “overbearing and approval for the titles of paintings. In the elabo-
threw his weight around,”14 Pichetto claimed he rate rating system of the Kress Collection, leading
had “restored tens of thousands of paintings art historians ranked the paintings, but Pichetto
during the past 25 years at the Metropolitan cast the deciding vote. When John Walker, chief
Museum of Art.”15 He evidently maintained a curator at the National Gallery of Art or David
very high opinion of himself when he boasted Finley, its director, wrote to Kress the mail always
that he had: went through Pichetto who became the conduit
succeeded in making many discoveries including for all art-related correspondence. Pichetto was
absolutely permanent pigment colors; materials responsible for the more mundane details as
for protecting the same; varnish which will not well—insurance valuations, temperature and
change color and materials and methods for humidity standards, and packing specifications.
cleaning, all of which I am the only one who He had the final say on display and installation,
possesses this knowledge.16 which he planned by arranging small maquettes
of the art to work out the hanging; he dictated
Among all his other responsibilities, clients, and the galleries’ wall color, lighting, and decided on
connections, Pichetto’s association with Samuel the use of marble trim, frames, and the infamous
H. Kress and his brother Rush Kress was probably Kress shadow boxes. It was Pichetto who designed
the most fulfilling and rewarding, financially, many of the frames and painstakingly selected the
professionally, and personally. Samuel H. Kress quality of velvet, identifying which frames would
allowed Pichetto access to some of the greatest be bordered in green and which in red velvet.
Italian paintings in America, and Pichetto recip- Until Samuel Kress suffered a stroke in 1946 and
rocated by providing Kress with service on many was disabled for nine years before his death, it
levels. Samuel H. Kress claimed that, “when seems that Pichetto was a key player in the pur-
selecting, I made certain never to acquire a paint- chase and care of the vast paintings collection.
ing that was so affected that it interfered with the Pichetto was the final arbiter on Kress publi-
original conception of the artist,”17 and in offer- cations as well. On offering its paintings to the
ing his generous gift of paintings to the National National Gallery of Art, the Samuel H. Kress
Gallery of Art stated that his intent was not Foundation added “terms and agreements” that
only to deliver these treasures to the Gallery but had to be ratified “before the gift would be con-
to “place them in the best possible condition.”18 summated.” The most important point was that
This, of course, was Pichetto’s contribution. “the Foundation [had] the right to require the
According to John Walker, chief curator and later employment by the Trustees of the National
director of the National Gallery of Art, it was Gallery of Art of Stephen S. Pichetto for … any
Pichetto’s efficiency and businesslike methods restoration work.”19 When the 1946 catalogue
that appealed to Samuel H. Kress. However, was being written, the ultimate decisions includ-
Ann Hoenigswald 33
ing the quality of the leather binding and the itan Museum of Art (fig. 2).20 In 1928 he was
distribution list were Pichettto’s responsibility. In named consultant restorer, a title he held until his
honor of the dedication of six new Kress galleries death in 1949.21 Even after 1941, when Murray
at the National Gallery of Art, Pichetto delivered, Pease was appointed Technical Advisor for Con-
presumably at Kress’s request, the opening remarks. servation of Works of Art, Pichetto maintained
Similarly Pichetto contributed an article celebrat- his position, albeit occupying a separate and dis-
ing the Collection in the September 1939 issue tant space in the building. Pichetto demonstrated
of American Collector. Pichetto even conferred his respect for the Metropolitan Museum in 1948
with doctors during Kress’s illness. On occasion when he contributed funds earmarked for the
the two men vacationed together as well. Kress construction of a restoration studio at the Mus-
reaffirmed his respect for Pichetto by appointing eum in honor of the institution’s seventy-fifth
him a trustee of the Kress Foundation in 1936 and anniversary. The Museum reciprocated by electing
curator of the Samuel H. Kress Collection at the him a Fellow in Perpetuity.
National Gallery of Art in 1947. John Walker In 1939 Pichetto assumed the position of con-
claimed that Pichetto had a greater influence on sultant restorer at the National Gallery of Art.
Kress than anyone else. Although the appointment seemed similar to the
During his association with Kress, Pichetto role he held at the Metropolitan, the conditions
officially began his appointment at the Metropol- for his appointment were unique. One of the
Fig. 2. Stephen Pichetto in the restorer’s studio at the Metropolitan Museum of Art, December 5, 1936.
most important directives of the proposed gift of restorer, Pichetto played a substantial role in creat-
the Kress Collection was that “the paintings and ing a new museum for the nation. Pichetto became
sculpture should be kept in the best condition.”22 one of the important forces behind the institution
The Board of Trustees at the National Gallery of that would not open to the public for another
Art was informed that: two years. Pichetto was responsible for preparing
the paintings for exhibition; he also arranged and
unless Kress could be assured that arrangements planned the construction of an elaborate restora-
could be made, he would be unwilling to pro- tion studio in Washington. He required a large
ceed … Mr. Kress expressed the desire that, if
possible, Mr. Stephen Pichetto, a well known space with rooms dedicated to specific tasks,
and thoroughly qualified restorer … be retained insisted that the area be air-conditioned and that
for this purpose.23 the wall color be warm gray. Precise and lengthy
lists of equipment were proposed, including twelve
John Walker remarked that he had no choice. He presses for the purpose of lining and cradling.
initially disliked and mistrusted Pichetto but later As nothing else could, these numbers reflected
in life concluded that he and Pichetto were, in fact, the level of activity! His responsibilities at the
working towards similar goals and that Pichetto National Gallery of Art extended to managing
had helped the National Gallery of Art “far more the wartime evacuation of paintings to Biltmore
than I realized.”24 In addition to his work as a House in North Carolina in January 1942 (fig. 3),
Fig. 3. Stephen Pichetto, at left, moving paintings from the National Gallery of Art to Biltmore House for safekeeping, January
1942.
Ann Hoenigswald 35
establishing the packing and transit guidelines and in his long letters to the staff at the National
preparing the storage facilities with appropriate Gallery of Art. Kress insisted that paintings
temperature and humidity standards.25 on wood required the greatest care and needed
It must be kept in mind that the National special treatment. He stressed that temperature
Gallery of Art and the Metropolitan Museum and humidity changes could be very detrimental
were secondary centers of activity to Pichetto’s to a painting. Kress also emphasized that the gift
private studio, which occupied an entire floor to the nation included “X-ray shadowgraphs” (as
of the Squibb Building in New York. Very pru- he and Pichetto called them) “for their educa-
dently he never joined the permanent staff of the tional value” and reports showing the physical
Metropolitan or the National Gallery of Art but condition of paintings. He mentioned the impor-
maintained the title of consultant restorer. Pre- tance of appropriate frames and even insisted that
sumably this allowed him to avoid conflict of paintings on wood panels should be packed in
interest, for he was simultaneously juggling work boxes marked with arrows indicating the direction
for major collectors and other museums in New of the grain and shipped in like direction. Surely
York and elsewhere, bearing responsibility for these instructions came from Pichetto.
modern paintings as well as Old Masters and even Stephen Pichetto, however, did not work alone.
consulting with artists. Of course, these arrange- The business and the large staff that Pichetto
ments also allowed him to draw several salaries at employed were run efficiently under the watchful
the same time. eye of Marguerite Lewis, his office manager and
It appears, however, that Pichetto’s most administrator. Three men—Steven Story, Dan
devoted attention was reserved for Samuel and, Coppari, and Paul Kiehart—did retouching in
to a lesser extent, Rush Kress. They depended on one room along with Rose Mary Sullivan who
one another’s expertise and respected each other’s consolidated flaking paint for eight hours a day!
eye and individual skills. Moreover they did not In an adjacent room were his woodworkers, Joe
seem to be in competition nor harbor any jeal- McCarthy and Angelo Fatta. They worked on
ousy towards one another; this was unlike the frames, inlays etc., and attached cradles to many
relationship Pichetto had with art historians and of the wooden panels that were treated in his
his fellow restorers. The art historians appeared to studio. Henry Hecht and Girard Roggeman car-
resent Pichetto because he had intimate access to ried out linings. Frank Sullivan, who worked at
collectors and their paintings yet did not share the the National Gallery of Art after Pichetto’s death,
academic credentials or social pedigree of the art was remembered by Paul Kiehart as having no
historians, then considered almost a prerequisite specific responsibilities. Most of the staff came
in the field. Restorers were competing with one to Pichetto with crafts skills or from art schools.
another for the same jobs and clients and cer- After the war several of them went on to study
tainly resented Pichetto’s success. Pichetto was at the Art Students League on the gi bill.
remembered as a man of strong will and ego, and Recognizing that “unscientific cleaning is the
although respected, he was not well liked by his most serious thing that can happen to a painting
colleagues. His clients, however, felt differently. because it cannot be corrected,”27 according to
Pichetto’s business acumen and perhaps his ego as Paul Kiehart again, Pichetto himself took all
well allowed him to become a very wealthy man responsibility for the cleaning of pictures. (For
and even to refer to himself with some satisfac- a different view, see Mario Modestini’s paper in
tion as “the greatest restorer.”26 this volume.) He worked in what was described
Kress and Pichetto respected one another’s as an elegant office/studio that was presumably
attention to detail, and each may have admired furnished to appeal to his clients. He was sur-
the other’s keen business sense. Samuel H. Kress rounded by upholstered chairs and several easels
often included what he had learned from Pichetto with paintings artistically displayed.28 Declaring
that he “did not want to camouflage the damaged
portions rather to retouch the missing portions
with local color,”29 Pichetto relied heavily on his
three inpainters whose method was to apply
colors in Winsor and Newton watercolors or in
egg tempera, coat with French varnish (shellac)
and glaze with dry colors in dammar varnish.
Inpainting palettes included only seven colors,
and varnishing was done with dammar; yet the
supply books also list the purchase of light and
dark varnish, oil varnish, restoring varnish, soft
varnish, Murphy varnish, and “xx” varnish.30
Although there is no doubt that much of
Pichetto’s inpainting was overdone, his treatments
were often well intended, and he claimed his goal
was visual balance. Regarding the treatment of a
Lorenzetti triptych (see fig. 1), Pichetto advised
Fig. 4. Notes and a sketch of Saint Paul Contini-Bonacossi that he would not clean the
by Antonio Veneziano (now identified as gold background for although it would make it
Lorenzo di Nicolò) from the daybook of
Paul Kiehart, February 27, 1943. Specific more brilliant, “it might lose its present subdued
inpainting procedures are detailed. tone, which blends so well with the rest of the
painting.”31
Pichetto’s studio has been criticized for being
financially driven and factory-like, but although
many believed that he never kept reports, there
remain, in fact, very valuable records. Louis de
Wild claimed that one never knew how much of
the studio restoration was Pichetto’s own work,
but in fact numerous daybooks identify precisely
who did what (figs. 4 and 5).32 In addition, exten-
sive photography documents the condition before,
during, and after treatment, and photographs were
often made in both light and dark conditions
to record different information. Pichetto also
requested raking light images, photographs of
the reverse or the edge of a panel or the tacking
margins. Ultraviolet and infrared images were also
made as well as X-radiographs of nearly every
painting in the Kress Collection.33 Although
Pichetto at one point had his own X-radiographic
equipment, most of this work was done by Alan
Burroughs34 who had a very close relationship
with Pichetto.35 Apparently it was at Pichetto’s
Fig. 5. Saint Paul, Lorenzo di Niccolò, request that Kress financed the X-radiography
ca. 1385, tempera on wood panel, 42 1/2 ×
17 3/8 in. (107.3 × 44.2 cm). Fine Arts project when Burroughs was no longer on the
Museums of San Francisco. staff at the Fogg Art Museum. Despite existing
Ann Hoenigswald 37
Fig. 6. Telegram from Rush Kress to Alessandro Contini-Bonacossi announcing Pichetto’s death.
reports and photographs, the record keeping was of the Kress Foundation in addition to being the
uneven. Presumably this was done intentionally. advisor and confidant to Samuel H. and Rush
There were certain things that Pichetto selected Kress. This heavy workload may have led to his
not to document; however John Walker’s com- death; on January 20, 1949 he died suddenly of a
plaint was only partially justified when he claimed massive heart attack at the age of sixty-one while
that there were no written records at all and that hosting some Italian dealers who had brought
he wished that Pichetto would “keep the type of paintings for Kress’s consideration.
report made out by the Metropolitan Museum.”36 His funeral, held at Saint Patrick’s Cathedral in
Walker’s concern was that “the reports would New York, reflected his importance. The flurry of
protect the present staff against criticism by future telegrams crossing the ocean immediately after his
curators and restorers.”37 Pichetto responded that death also conveyed the weight of the loss. Many
he preferred to devote his time to actual work people believed that an enormous void had been
on the paintings rather than to elaborate records. left in the art world. In a letter to Marguerite
By 1949 Stephen Pichetto held concurrent Lewis, Alan Burroughs remarked that Kress “must
positions as consultant restorer at the Metro- be under terrific pressure without Stephen by his
politan Museum and at the National Gallery of side,” and she replied that “He tells everyone he is
Art. He had an extremely successful private prac- lost.”38 Within hours of acknowledging Pichetto’s
tice and was actively involved in many activities death, John Walker sent urgent wires to Bernard
Berenson requesting proposals for a suitable the Foundation recognized their indebtedness to
replacement for the National Gallery of Art posi- Pichetto as well. Pichetto held that preservation
tion. Berenson replied the following morning that was more important than restoration,41 and he
they shouldn’t make a decision too quickly or left Samuel H. Kress with this legacy. Rush Kress
hire someone trained on Dutch or Flemish paint- claimed that:
ings (presumably implying Marchig or de Wild)
our objective is to supply for the first time in
because “such restorers are apt to skin an Italian
the history of art a complete record of our
picture before they know what they are doing restoration work from the beginning to the
and a picture once skinned can be faked up but end so as to have a carefully worked out chap-
will never be itself again.”39 Simultaneously ter in our foundation books on the subject
Rush Kress was imploring Contini-Bonacossi to restore or destroy.42
suggest a restorer (fig. 6). Within weeks Contini-
Bonacossi responded to Kress with the name Largely because of the influence of Stephen
of Mario Modestini, who was described as Pichetto, the Kress Foundation has remained
having “the temperament of a master and deeply committed to the treatment of works of
without exaggeration the finest restorer in the art, conservation education and research.
world.”40 Modestini arrived in the United States
to assume Stephen Pichetto’s role at the Kress Acknowledgements
Foundation, but he did not replace Pichetto Attempting to understand Stephen Pichetto’s
at either the Metropolitan Museum or the connection to the Kress Collection has been like
National Gallery of Art. piecing together a puzzle. There was no single
Stephen Pichetto’s son-in-law, Paul Andrepont, source of information, and much of what I
assumed the task of continuing the private discovered contradicted earlier findings. Many
business, but abandoned the plan quickly. Subse- of the comments were based on conjecture
quently Marguerite Lewis offered the client list or affected by personal bias. The research, there-
and her assistance to Kiehart, Story, and fore, was dependent on the generosity of numer-
McCarthy, but they too were unable to make the ous people opening archives, sharing letters,
business flourish. The skilled and experienced memories, and reminiscences so I could consult
hands, the able administration, the existing capital primary sources.
equipment, and the impressive client list alone The Conservation Department at the National
couldn’t keep the business afloat. Without Gallery of Art was very supportive of this proj-
Pichetto there was no operation. ect, and I am particularly grateful to my “research
Pichetto’s role at the Metropolitan and the accomplices,” Elizabeth Walmsley and Renée
National Gallery of Art was probably more Lorion. Anne Halpern and Maygene Daniels
important than has been recognized, but his name always kept the name Pichetto in mind when they
is justifiably linked more intimately to Samuel H. found archival references and passed them on to
and Rush Kress and the Kress Foundation. It may me. I am indebted to Lisa Ackerman and Dr.
well have been that he preferred to commit him- Marilyn Perry at the Samuel H. Kress Foundation:
self to individuals rather than institutions. Per- they allowed access to their files and answered
haps it allowed him more autonomy; perhaps he numerous questions. Moreover, the Kress Founda-
disliked or felt uncomfortable with the blatant tion provided financial support to Renée Lorion.
snobbishness of the museums. Certainly Pichetto Mario and Dianne Dwyer Modestini were
allied himself to the museum world and benefited especially helpful particularly where our material
from the contacts and credibility it afforded him, dovetailed. Dorothy Mahon at the Metropolitan
but his most visible devotion was to Kress. Museum provided a crucial research link, and
For their part, the Kress brothers and ultimately I am extremely indebted to her. Jeanie James,
Ann Hoenigswald 39
George Bisacca, and Keith Christiansen provided traveled on occasion with Kress.
6. In a sworn affidavit from July 1917, written to be excused
assistance as well. Francesca Bewer at the Fogg from the draft, Pichetto stated he had been “engaged in
Museum and the Straus Center for Conservation the restoration and preservation of priceless works of art
generously opened their archives and pointed out since 1901.” Duveen Brothers records, 1876–1981, Getty
Research Institute, Research Library/Special Collections
many important details. Teresa Hensick and Ron and Visual Resources.
Spronk answered additional questions. E. Peters 7. In “The Kress brothers and their bucolic pictures” (in
Bowron and Joyce Hill Stoner provided material Chiyo Ishikawa et al. (eds.), A Gift to America: Masterpieces
of European Painting from the Samuel H. Kress Collection, New
and shared my enthusiasm for the project. Access York, Harry N. Abrams, 1994), Edgar Peters Bowron sug-
to the Duveen Archives was facilitated by Mark gests that Kress introduced Pichetto to Contini-Bonacossi.
Henderson at the Getty Research Institute and Additional examination of the chronology of correspon-
dence suggests that it was the other way around. This
by Susan Roeper at the Library of the Clark would not be out of character as Pichetto was known to
Art Institute. make important introductions such as introducing his
I owe most, however, to those who knew client Chester Dale to David Finley, director of the
National Gallery of Art. This resulted in the important
Pichetto, either directly or indirectly, and who gift of modern paintings to the Gallery.
graciously retold their stories to me. Caroline 8. Correspondence from Alessandro Contini-Bonacossi to Rush
Wells, Kate Lefferts, Henrietta Suhr, and Daniel Kress, January 10, 1949. National Gallery of Art Archives.
9. John Walker, Self-Portrait with Donors. Boston and Toronto:
Butler were particularly helpful. My utmost Little, Brown and Company, 1974, p. 142.
appreciation, however, goes to Robert Feller, 10. S. N. Behrman, Duveen: The Intimate Portrait of a Fabulous Art
Paul Kiehart, and Maura Dillon Pichetto. They Dealer. New York: Harmony Books, 1951, p. 71.
11. The Arts, Vol. xv, pp. 183–4
thought they were merely recalling old memories. 12. Ibid. p. 134.
In truth, they allowed me to understand an 13. Walker 1974 (cited in note 9), p. 143.
important part of conservation history. 14. Transcribed telephone conversation between Joyce Hill
Stoner and Louis de Wild, October 9, 1977. Transcription
housed at the faic Archives, Winterthur Museum and
Ann Hoenigswald is Senior Conservator of Paintings at Library.
the National Gallery of Art in Washington, D.C. where 15. New York Times, February 27, 1929
16. Duveen Brothers records, 1876–1981, Getty Research Insti-
she has been since 1977. She completed her undergraduate tute, Research Library/Special Collections and Visual
degree in Art History at the University of Pennsylvania Resources.
and received both a Certificate in Conservation from the 17. Proposed letter from Samuel H. Kress to be sent to David
Finley, March 13, 1941. David Finley Papers, Manuscript
Intermuseum Conservation Association and also an M.A. Division, Library of Congress.
in Conservation from Oberlin College. 18. Unsigned letter to Board of Trustees of the National
Gallery of Art, June 1, 1939. National Gallery of Art,
Conservation Department Archives.
Notes 19. Letter from Rush Kress to David Finley, March 13, 1941.
1. William Suida. National Gallery of Art Memorandum, David Finley Papers, Manuscript Division, Library of
February 18, 1949. Congress.
2. 1900 Census data, United States Census Information. 20. In testimony given in the Hahn v. Duveen case in 1929,
National Archives and Records Administration. The date Pichetto claimed that he had been at the Metropolitan
varies according to the source. The 1900 census, the 1910 Museum for twenty-five years, but there is no documenta-
census, immigration papers, and legal documents support tion of his relationship with the Museum when Pichetto
very different dates. was seventeen years old. New York Times, February 27, 1929.
3. For examples of other stories which were perpetuated 21. John Walker claimed that Pichetto was “inexplicably termi-
see: Ann Hoenigswald with Renée Lorion and Elizabeth nated” at the Metropolitan prior to assuming his position
Walmsley, “Stephen Pichetto and conservation in at the National Gallery of Art. This, of course, was not
America” in Andrew Oddy (ed.), Past Practice–Future true, but Pichetto may have led Walker to believe this was
Prospects, British Museum Occasional Paper No. 145. the case so as to suggest that he was devoting his full
London: British Museum, 2001. attention to the Gallery.
4. These erroneous credentials were repeated in his obituar- 22. Minutes of the Board of Trustees Meeting, May 26, 1939.
ies, official museum biographies, and within his family. It National Gallery of Art, Conservation Department
is assumed that Pichetto himself encouraged these myths. Archives.
5. Other than a trip in 1914 with his mother, there is no 23. Walker 1974 (cited in note 9), p. 142.
record of his journeys. One would like to assume that he 24. Unsigned letter to the Board of Trustees of the National
Gallery of Art, June 1, 1939. National Gallery of Art, Photography Credits
Conservation Department Archives. Fig. 1, p. 30. National Gallery of Art, Washington D.C.,
25. Apparently Pichetto had strong opinions on the appropri- Gift of Frieda Schiff Warburg in memory of her husband,
ate temperature and humidity levels for paintings and was Felix M. Warburg. Photograph © 2002 Board of Trustees,
called upon frequently to make suggestions. In 1941, he National Gallery of Art, Washington D.C.
established his guidelines for these as well as the display, Fig. 2, p. 34. ©The Metropolitan Museum of Art, New
handling, and packing in a letter to Collas Harris of the York, NY.
Committee on the Conservation of Cultural Resources. Fig. 3, p. 35. ©National Gallery of Art, Washington, D.C.,
His suggestions were particularly practical and well Gallery Archives.
thought out. National Gallery of Art, Conservation Fig. 4, p. 37. ©National Gallery of Art, Washington, D.C.,
Department files. Gallery Archives (Conservation Department).
26. Transcribed telephone conversation between Joyce Hill Fig. 5, p. 37. Fine Arts Museums of San Francisco, San
Stoner and Louis de Wild, October 9, 1977. Transcript Francisco, CA, Gift of the Samuel H. Kress Foundation
housed at the faic Archives, Winterthur Museum and (61.44.4).
Library. Fig. 6, p. 38. ©National Gallery of Art, Washington, D.C.,
27. Text of New York University radio broadcast, October 22, Gallery Archives.
1944. 1
28. P. Kiehart, 1994. Correspondence from Paul Kiehart to the
author, December 1994.
29. New York University radio broadcast, October 22, 1944.
30. This may have been Maximilian Toch’s “famous matte var-
nish” as Louis de Wild referred to it. De Wild described a
meeting he had with Toch and Pichetto in New York and
mentioned that Toch and Pichetto were great friends and
probably shared ideas about painting materials. Transcribed
telephone conversation between Joyce Hill Stoner and
Louis de Wild, October 9, 1977. Transcript housed in the
faic Archives, Winterthur Museum and Library.
31. Correspondence from Stephen Pichetto to Alessandro
Contini-Bonacossi, December 30, 1926. National Gallery
of Art, Conservation Department Archives.
32. Each man maintained extensive daybooks detailing his
activities. Rather than providing lengthy reports on every
painting, the treatment records were kept by each individ-
ual, documenting his or her work.
33. Many of the photographs are stamped “Pichetto Archives”
on the reverse. This is not an archive in the true sense of
the word. It is assumed that this label was applied after his
death and may appear only on photographs at the
National Gallery of Art.
34. For more information about Alan Burroughs see: Francesca
G. Bewer, “Technical research and the care of works at the
Fogg Art Museum (1900–1950)” in Oddy 2001 (cited in
note 3).
35. Burroughs was the only person who referred to Pichetto
as “Stephen” in his correspondence. Pichetto in reply
addressed him as “Alan” which was equally unusual.
36. John Walker memorandum for the file, July 22, 1948.
37. Ibid.
38. Correspondence from Alan Burroughs to Marguerite
Lewis, April 7, 1949. Her undated reply is on the same
letter.
39. Correspondence from Bernard Berenson to Rush Kress,
January 22, 1949.
40. Correspondence from Alessandro Contini-Bonacossi to
Rush Kress, February 9, 1949.
41. New York University radio broadcast, October 22, 1944.
42. Rush Kress to David Finley, November 21, 1949, from
the David Finley Papers, Manuscript Division, Library
of Congress.
Ann Hoenigswald 41
Fig. 1. Mario Modestini with his team in the Carnegie Hall studio with Queen Zenobia Addressing her Soldiers by
Giovanni Battista Tiepolo. From left to right: Quarantelli, Robert Manning, Bartolo Bracaglia, Giuseppe Barberi,
Mario Modestini, Amleto De Santis, and Angelo Fatta.
Mario Modestini, Conservator of
the Kress Collection, 1949 –1961
Dianne Dwyer Modestini with Mario Modestini
T he sudden death of Stephen Pichetto on January 20, 1949 was a

grave problem for the Samuel H. Kress Foundation. Had the collection
ultimately formed by Samuel Henry Kress and, later, his Foundation
been installed in a single museum in New York, as he once intended, it
would have constituted, if not the greatest, then certainly the most
comprehensive collection of Italian art in the United States, if not the world,
studded with hundreds of masterpieces or “leaders” as the retail magnate was
wont to call them. The story is well known but, as the lone survivor of those
years, I shall briefly recap the events. Shortly before the 1941 opening of the
National Gallery of Art its first director, David Finley, a lawyer and advisor
to Andrew Mellon, visited Kress to ask if the new museum might borrow
400 paintings. Kress agreed, and an initial group, many taken from the walls
of his apartment, was sent to Washington. Samuel himself fell seriously ill
shortly after and died in 1955; it was left to his younger brother, Rush Kress,
to decide how to best fulfill his brother’s wishes. The Foundation’s commitment
to Washington was already strong and the Kress Collection at the National
Gallery of Art replaced the idea of a Kress Museum. Samuel Kress had always
lent works from his collection around the country, and some paintings had
already been given to museums in Denver, Houston and other cities. This
precedent developed into a program to donate an art collection consisting of
twenty-five to forty paintings and often several sculptures to each of eighteen
regional museums.
Dianne Dwyer Modestini with Mario Modestini 43

In 1949, this ambitious art project had just Kress liked to work out the practical details right
begun to be formulated by the Foundation as one away, and so we agreed that I would submit an
of its two principal missions, the other being med- estimate for the work to be done on each paint-
ical research, and it was still on the drafting board ing and invoice each “job,” as Pichetto had. In the
when Pichetto’s untimely death left Rush Kress, meanwhile John Walker, chief curator and later
the Foundation’s president, and Guy Emerson, its director of the National Gallery of Art, suggested
art director and a trustee, in a state of shock. A that as I planned to stay until early April I might
large shipment from Count Alessandro Contini- work on some of the Kress paintings that were
Bonacossi, the Florentine collector and art dealer, left in Pichetto’s studio, as a sort of sample to
languished in storage, and Pichetto’s staff was show them what I could do.
paralyzed. Meanwhile his widow kept the studio I chose a panel that had been recently cradled,
open while bills for rent and salaries, which a Madonna and Child by Paolo di Giovanni Fei, a
the Foundation felt a moral obligation to pay, Sienese artist of the fifteenth century, today in the
mounted. This was not good business. Emerson National Gallery of Art in Washington. It was a
assessed the situation and wisely urged Rush to tempera painting with a gold background, very
proceed slowly as Pichetto had warned them dirty, covered with candle smoke, soot and old
about the danger of paintings being ruined by varnishes. I don’t think it had ever been cleaned.
“careless and incompetent people … in the field The picture was sent to me at the Plaza Hotel
… experts” rumored to have “ruined” many paint- where a large north-facing window provided per-
ings at the Metropolitan and Boston museums.1 fect light. My first problem was to find something
Rush Kress turned to his principal art advisors, to use to soften the black deposits. Normally I
Contini-Bonacossi and Bernard Berenson.2 Both used an unguent that I made up myself from
suggested that he talk to me and as a result a tele- various ingredients according to a recipe from the
gram arrived inviting me to come to New York. manual Il Restauratore dei Dipinti by Count Giovanni
At that moment I was in São Paolo, Brazil where Secco-Suardo consisting of melted animal fat,
I had set up a didactic exhibition for a proposed linseed oil and Marseilles soap.3 Being without
new museum. I had always wanted to visit New my usual materials, I had to improvise, and I
York, and Contini-Bonacossi urged me to go bought a product called Pond’s cold cream that
directly since the situation was critical. women use to remove make-up. I mixed this
I arrived at La Guardia Field on Saturday, with a bit of Marseilles soap and some raw lin-
March 7, 1949. Gualtiero Volterra, Contini- seed oil. I made various tests to see how long it
Bonacossi’s buying agent and a trusted friend of was necessary to leave this creamy emulsion on
Rush Kress, referred to affectionately by both men the painting, removing it with turpentine. In a
as “the maestro” because he had been a child prod- few days I had cleaned the painting and done
igy pianist, was already in New York and came to some minor retouching with tempera colors.
meet me. We stayed at the Plaza Hotel where a The painting was in a very good state. Walker
lovely room with a large bath and a window over- pronounced himself satisfied and told Kress that
looking Central Park cost $8.50 a night. On Monday I had done a beautiful job.
we went to the offices of the Kress Foundation, a I stayed on in New York until the middle of
small space in Pichetto’s 745 Fifth Avenue studio, April and worked on several other paintings. After
a suite of five or six rooms with small windows Volterra’s departure, the research curator of the
and low ceilings. There were paintings everywhere. Foundation, Dr. William Suida, the great Vien-
Those that were finished looked as shiny as if they nese art historian who had resided in the United
had just come off an automobile assembly line. States since the 1930s, befriended me and helped
With Volterra as my interpreter, I met with me in my conversations with Kress. We agreed
Rush Kress and Guy Emerson and eventually the that I would take on the responsibilities of the
other trustees and officers of the Foundation. Kress Collection for part of the year and would
oversee the men who had worked for Pichetto. returning to New York in July, as we had agreed.
Kress was very kind and cordial, which was, in For the moment, not sure how long I would stay
fact, his nature. He seemed very American to me in New York, I did not close my gallery.
and, in some ways, had a taste for simple things. Shortly after my return to Rome, Kress’s sec-
After we had confirmed our arrangement, he retary Fred Geiger began to cable that the work-
invited me to lunch at Horn & Hardart’s, where, room would be ready on April 25th and when will
he said, they made the best coffee in town. The Modestini arrive? After much frantic correspon-
walls were covered with little boxes with glass dence between an impatient Rush Kress and a
doors through which you could see the food concerned Contini-Bonacossi, I finally booked
offered. You inserted the right number of nickels, passage to New York on the Queen Elizabeth
and the door would pop open, and you took to assume my new responsibilities. Among my
whatever meal you had chosen. It was an interest- papers I recently came across a radiogram dated
ing experience, and naturally I never went back July 12, 1949: welcome to america suida and
there again. emerson will meet you at dock r h kress. By
Kress wanted me to move into Pichetto’s studio, August 19th a Rush Kress memo asks whether
but I didn’t like the space. Although it was on the Modestini “needs any more paintings to work
fifteenth floor, the light was poor since the win- on during the next three weeks.” Scrawled pencil
dows were small, which meant that the restorers note: “Now has 30.”
always had to work with electric lamps. While The room at the Foundation quickly became
Gualtiero Volterra was still in New York, after a too small for the avalanche of work arriving
lot of searching, we leased a suite of rooms at from the storage warehouses. It was evident that
221 West 57th Street next to the Art Students the art program devised by the Kress Foundation
League, which would serve both as my studio required my full attention. Kress gave me no peace
and as offices for the Foundation. There was a until I agreed to take a full-time position. Reluc-
big room with good north light from a large tantly, and within a short time, I had to make the
window. The collection at that time consisted decision to close my studio in Rome, which I did
of about 1,300 paintings, some on loan to the with some difficulty and not without regret. Only
National Gallery of Art, some at the Kress apart- a few years before, together with Pietro Maria
ment at 1020 Fifth Avenue, and many in storage Bardi, a critic and expert on contemporary art,
at the Morgan Manhattan and Atlas warehouses. I had opened a gallery and studio of restoration
I returned to Rome to tidy up my affairs before in fifteen rooms in Piazza Augusto Imperatore,
Fig. 2. One of the rooms for restoration of paintings at the Fig. 3. An examination room with equipment for X-radiography
Studio d’Arte Palma, Rome, ca. 1947. and an ultraviolet lamp at Studio d’Arte Palma.

called the Studio d’Arte Palma (figs. 2 and 3). We general practice.
employed a large staff, had the latest equipment As John Walker relates,5 every painting that
and had already mounted important exhibitions was offered to Kress was sent to Pichetto’s studio
of contemporary artists such as Morandi and for examination and approval and, then, if pur-
Manzù. We also held one devoted to seventeenth- chased, returned for cradling, relining, revar-
century Italian painting, not then in vogue, and nishing and so on. Contrary to popular belief,
another that was perhaps the first antique frame Pichetto rarely cleaned any of the Kress paint-
exhibition anywhere. The gallery was enjoying ings as they all came from dealers and had been
great success. Although it was difficult to extract recently restored.6 Normally he would correct
myself from these arrangements, for various a few restorations or add some retouches using
reasons I was ready for a change despite many powdered pigments bound with dammar varnish.
ties to my beloved Rome. Unfortunately he used zinc white that reacts with
My position with the Kress Foundation was dammar to produce zinc dammarate, a chalky
formalized. I was named curator and conservator whitening of the surface.7 This blanching process
of the collection, for which I was paid a salary; had already begun and we were obliged to remove
space, materials, and other costs associated with Pichetto’s restorations as early as the 1950s, occa-
the work on the collection were supplied by the sionally on paintings that I myself had restored
Foundation, while I was responsible for staff for Contini-Bonacossi just after the war. In the
salaries, my living accommodations and personal years that followed, all of the retouches have
expenses. I sent invoices for each restoration, blanched, and the varnishes have discolored.
reframing, construction of shadow boxes, and In Pichetto’s studio every painting on panel
so on. This was very similar to the arrangement was thinned, flattened in a press and cradled. This
the Foundation had with Pichetto. was standard procedure for most paintings that
By May of 1950 we had moved to a large studio came to America since centrally heated interiors
just across the street from the Foundation in the often provoked warping or splitting of panels
tower of Carnegie Hall. Two of my assistants accustomed to the high humidity in European
from the Studio d’Arte Palma came to work with churches and palaces. The Pichetto cradles are
me, Amleto De Santis and Giuseppe Barberi. instantly recognizable: fixed vertical members of
Amleto had been in art school with me and was varnished mahogany and sliding members of clear
a very gifted painter of the Scuola Romana. He pine, lightly waxed, each approximately 3/4 in.
had worked with me for nearly ten years and thick, although the size varied according to the
had become an excellent restorer. I had also inher- scale of the painting. The panels treated this way
ited three of Pichetto’s assistants. Angelo Fatta have remained flat over the years, and for all their
was the carpenter who, under Pichetto’s direction, brutality, the cradles have caused surprisingly few
thinned and cradled all the panel paintings. The reactions in the original wood panels. Some of
cradles were well made but excessively heavy, and the more fragile panels have developed splits
I tried to explain to him that this could cause along the edges of the fixed members and, in
further cracking of the original panel, but he a few instances, panels that originally had pro-
was difficult to communicate with and set in nounced convex warps have continued to flake
his ways. Born in Sicily, he had come to the in areas where the paint was compressed during
United States when he was twenty years old and the flattening process, especially along the joins.
spoke his own dialect, a mixture of Italian and At that time, an alternative commonly applied
Brooklyn English that was incomprehensible to remedy was to transfer paintings from their
me. Henry Hecht, the reliner, and Paul Kiehart, wooden panel supports to either canvas or to
a restorer, also came to work with me.4 From an inert solid support such as Masonite.8 Apart
them I learned about Pichetto’s techniques and from stability, it was also part of an aesthetic: it
was the machine age, and American taste was for between light and dark. This is particularly
flat, mechanically smooth surfaces. Reflecting this important with Baroque paintings. I always stop
preferred look, part of Pichetto’s normal practice before going too deep, and prefer to leave a little
was not only to flatten panels, but also to reline patina. Many times I have been criticized, in
every painting on canvas whether or not it was particular by American dealers, for not having
necessary. New linings were applied directly to cleaned the painting enough. In my opinion,
previous ones. Glue paste adhesive was used; the most paintings in the hands of dealers today are
surfaces were ironed on the front with a fifty- terribly overcleaned.
pound hot iron and put into a press to dry. This In my experience, for varnish removal, solvents
merciless operation flattened the impasto and that evaporate quickly are the safest. Chemicals
brushwork of each painting. Every relined paint- such as dimethyl formamide, benzene, diacetone
ing was furnished with a sturdy new stretcher, alcohol, essential oils and cellosolve stay in the
the edges bound with gray paper tape. paint layer, and their softening action can con-
Whatever surface texture survived was obliter- tinue over a long period of time. I only resort
ated by a thick layer of varnish built up using to those remedies to remove tough old restora-
alternating layers of dammar in turpentine and tions done with oil paint. Occasionally, with
shellac in alcohol, so-called “French varnish.” This much trepidation, I have used a very strong
was trickier than it sounds; the shellac had to be ammonia solution in certain circumstances,
applied quickly without picking up the varnish “stopping” it (an inaccurate term but widely
underneath. Small, flat soft-haired brushes were used) immediately with turpentine or mineral
used for applying the shellac, which was brushed spirits. Again, its rapid evaporation makes it
on in short strokes in one direction. Sometimes safer than other choices. This technique requires
Pichetto built up a sort of dam around the pic- courage, skill and speed.
ture onto which he poured varnish. He often My father was a gilder, a frame maker and a
reframed paintings with modern reproductions restorer of polychrome sculpture. Since I went to
and others, primarily small gold-ground paintings, work in his shop at the age of fourteen, I have
were fitted with shadow boxes lined with antique worked with gold leaf and, in the course of my
velvet. Pichetto contracted this out to the firm of long career, I have had a lot of experience with
D. Matt, which remained in business until Julius gold-ground paintings. Many, like the Paolo di
Lowy purchased it in the late 1980s.9 Giovanni Fei, haven’t been cleaned for years and
I have been asked to describe my approach to are covered with a black crust consisting of oil,
paintings, not an easy task since every painting soot, glue and grime that is extremely difficult to
presents its own problems. Since this is so, on remove. Many of these paintings have been ruined
consideration, the most important thing is to by the use of strong alkaline cleaning agents, such
come to a painting with humility, great respect as the caustic soda so popular in the nineteenth
for the artist as well as a certain fear of touching century, to remove this black carapace. As I have
it with solvents when there is always the risk of already mentioned, I have had great success using
spoiling it. Therefore, I habitually begin by mak- Secco-Suardo’s unguent or some variation of it
ing a small test in a corner, in some unobtrusive to soften the hardened dirt and oil. This requires
place, never making a cleaning test in the center patience, as it does not work immediately. One of
of a painting. Once I have cautiously determined my earliest experiences as a restorer was with the
the mildest solvent possible, the state of the Rospigliosi Collection in Rome, before its disper-
painting and its sensibility, I begin by removing sal at auction in 1931 and 1932.10 Many paintings
the varnish as evenly as possible over the entire from the family’s Palestrina villa had never been
composition, not paying undue attention to the cleaned and were covered with a hardened black
lighter passages, but developing the relationships crust of smoke and soot from the fireplaces that

could only be removed with the pomade. moved, the solvents leach the medium. Obviously
The cleaning of a gold ground is a very deli- paintings of different periods have different re-
cate operation. Anything containing water has a quirements. While early paintings need the thin-
ruinous effect since the gold leaf is bound to the nest varnish possible in order to obtain a matte
bole preparation with a mild gelatin solution, surface in keeping with their original appearance,
easily undermined by moisture. Therefore I also Baroque paintings, particularly those with a dark
avoid solutions containing alcohol. I have found preparation, require a fuller varnish. I never var-
acetone mixtures to be safe. Sometimes I have nish the gold ground. Despite the treacly varnish
used acetone and linseed or mineral oil. Unguents, recipes given in early treatises, I do not believe
as long as they are an emulsion containing mainly that artists, who intrinsically have good taste, ever
oil and just a touch of soap, can also be used liked glossy surfaces on their work.13
safely. Often the punched decoration of the gold I abhor the practice of thinning, cradling and
ground is clogged with dark brown, discolored transferring panel paintings. Even the warping of
varnish, left behind by previous cleaning. Usually a panel is a sign of its age and manufacture, and
I try to remove these deposits, softening them it is wrong to try to change its appearance. Often
with Red Devil waterless paint remover,11 applied I have had frames made to accommodate the cur-
with a tiny brush, and then cleaning them manu- vature of a panel. Sometimes, in situations where
ally under the microscope, dot by punched dot. the gesso layer has completely lost its consistency,
This painstakingly slow method does not harm the glue binder having degenerated from excessive
the gold but requires a delicate touch. humidity over a long period of time, I have had
Sometimes, if a gold-ground painting has not to resort to transfer. Occasionally I have done
been spoiled by harsh cleaning, a gray patina, this myself, although I usually used an expert in
original to the painting, remains, even under sub- Vienna, Wolfgang Kneisel.
sequent layers of varnish. This is the temporary Having had many unpleasant experiences with
varnish described by Cennino Cennini, which was commercial reliners, and having lost several pic-
made of beaten egg whites. Originally clear, the tures to their inexpert hands, both at the Studio
gray tonality has developed over time. When the d’Arte Palma and while I worked for the Kress
paint mixture is lean, dirt may have also been Collection, I supervised all relining myself.
absorbed into the upper layers, adding to this gray Usually we used a mixture of rabbit-skin glue,
cast. I treasure this and never try to remove it, Venetian turpentine, and flour that was brushed
which would also result in eroding the paint layer onto the front of the lining canvas and to the
and possibly losing some of the delicate final reverse of the original. I always faced the painting
modelling. On some gold backgrounds, a similar first with gelatin glue and tissue paper and cush-
layer, which I suspect is the same egg white var- ioned the marble lining table with a thick layer
nish, can be seen applied over the leaf and around of soft cardboard. The stretched lining canvas
the painted contours, suggesting that this was was lined up with the original and ironed from
sometimes done as a separate step to tone down the back using normal electric irons at a low
the brash effect of newly burnished gold.12 setting, continuously checking the front of the
As for varnishing, I dislike thick glossy coat- original to make sure the surface was not being
ings and have always tried to use the minimum. damaged. I never put the paintings in a press or
Many of my restorations have held up remarkably under heavy weights. Early on I learned through
well for over fifty years and I attribute this, in bitter experience that it is extremely dangerous
part, to my practice of minimal varnishing. The to use water-containing glue paste adhesive on
longevity of a restoration is important, not for paintings that have never been relined, especially
the vanity of the restorer, but for the life of the on seventeenth-century and eighteenth-century
painting itself, since every time the varnish is re- canvas with a dark preparation; in this case the
original canvas must be isolated with a coat of lists made for the 1951 Kress exhibition at the
varnish or shellac, otherwise it will shrink and the National Gallery of Art. The art mission of
color will detach.14 the Foundation had been defined: a large Kress
Returning to 1950 and my work for the Kress Collection for the National Gallery of Art and
Collection, the paintings arrived in such numbers smaller ones for eighteen regional museums, the
that, even after the move to Carnegie Hall, we remainder destined primarily for university study
were still strapped for space, and I took a second collections across the United States. It became
studio for woodworking, framing, relining and so clear that to do justice to the reputation of the
on, reserving the tower for cleaning and retouch- Collection, important pictures would be acquired
ing. For the moment our needs appeared to be for all projects. The National Gallery of Art
satisfied. I had brought two more of my Roman encouraged this, and Rush Kress was pleased to
assistants to New York, Claudio Rigosi and hear from John Walker that the market was pro-
Bartolo Bracaglia, and a wonderful frame restorer pitious at that moment in comparison to twenty
from Florence, known only as Quarantelli, a great years earlier when Andrew Mellon was collecting.
character of whom everyone became very fond, In retrospect it really was golden era for buying
particularly Rush Kress, despite the fact they art. The goal was to make “the Kress Collection
could not communicate with each other since unique in history, a national collection, and not a
Quarantelli spoke in a strict Florentine dialect. Washington collection with eighteen or twenty
Our team was complete (see fig. 1). subsidiary collections of inferior quality.”15
The staff of the Kress Foundation consisted of Soon after I arrived Wildenstein offered us two
Dr. Herbert Spencer, the director, Guy Emerson, important paintings from Count Vittorio Cini’s
head of the art project, Mary Davis, adminis- Collection that had been sold during the war to
trator, and Miss Evans, a secretary. There were raise money to ransom Cini from an underground
six or seven trustees, one of them an Italian- cell in Dachau where he was interned because of
American, Andrew Sordoni, with whom I was his opposition to the Fascist regime.16 His mis-
able to exchange a few words in my native lan- fortune was a great boon to us as we were able to
guage. Mr. Geiger was secretary to Rush Kress. buy Botticelli’s portrait of Giuliano de’ Medici and
The trustees met every two or three months and Benozzo Gozzoli’s enchanting The Dance of Salome,
usually I joined them, particularly when they were both now in the National Gallery of Art.
discussing projects regarding the collection and Luck again favored us when Baron Heinrich
acquisitions. I had to learn English to communi- Thyssen of Lugano, in temporary financial diffi-
cate, especially with Kress who was difficult to culty after the war, was forced to sell several
understand as he always talked with a cigar in his paintings from his collection; we acquired the
mouth. There were many things to discuss with Altdorfer triptych, the double-sided panel by
the employees of the Foundation, all of whom Dürer and Memling’s Saint Veronica.
were American, and so I gradually learned to Acquisitions were not always so easily come
speak English. William Suida helped me very by. For the 1951 National Gallery of Art exhibi-
much, especially with Rush Kress, and with time tion and the first three Regional Collections, we
my English improved so much that in meetings scoured the premises of every dealer we knew
Geiger sometimes would turn to me of all people for suitable paintings. A group of twenty-one
to ask what his cigar-chomping boss had just said! paintings was purchased from Contini-Bonacossi,
By the spring of 1950, as Emerson wrote to which included the five large altarpieces from the
Kress, “Things have been moving here!” The Cook Collection at Richmond. It was a period
Regional Gallery Project, as it had been named, of frenetic activity as we called at Wildenstein,
was well underway, with collections being formed Knoedler, Mont, Drey, Duveen, Weitzner,
for San Francisco, Philadelphia, and Seattle and Seligmann, Koetser, Rosenberg and Stiebel, and

nearest town, Newfoundland, population 200, was
five miles away, an unlikely target. The art facility
consisted of three stories. The ground floor was a
bombproof bunker large enough to store the entire
Kress Collection. It was fitted with rolling racks on
which the paintings were arranged by school and
period so Foundation and National Gallery of
Art staff as well as directors of the prospective
Regional Collections could easily examine them
(fig. 4). Above the storage was a large restoration
studio (figs. 5 and 6). There was a carpenter’s shop
for Angelo Fatta, fully equipped with woodworking
machinery, and a separate studio for Quarantelli,
Fig. 4. Mario Modestini and Robert Manning in the storage the framer. The X-radiograph machine was in a
room at Huckleberry Hill, Pennsylvania, in the 1950s.
lead-sealed room in the basement. There was a
photo studio although we lacked a staff photogra-
French & Company as well as any number of pher. Robert Manning, William Suida’s son-in-
smaller galleries to select paintings and sculpture law, had been engaged as my assistant to be in
for prices that now seem absurdly low. charge of the record keeping, and he engaged a
Sifting through and evaluating all these possi- photographer I recall only as Colden to come to
ble acquisitions, keeping in mind what was wanted Huckleberry Hill for several weeks at a time. We
by Washington and the Regional Collections with ourselves had photo equipment and whiled away
whom we were already in contact, was complicated many a winter evening recording the work in
by the holdings in storage with which the new can- progress so as not to lose time waiting for the
didates had to be compared and integrated. Often photographer’s visits. The unsatisfactory Colden
we found that we already possessed a better paint- was ultimately replaced by Angelo Lomeo and his
ing by the artist. The process was cumbersome wife Sonja Bullaty—two real artists who made the
since access to the warehoused pictures and espe- best photographs of paintings I have ever seen.
cially to the decorative arts was difficult. Lists were They became great friends.
vetted and meetings were held, attended by Rush The studio was fully equipped with every con-
Kress, Guy Emerson, Herbert Spencer, and me for ceivable tool for restoration and examination to
the Foundation and John Walker and sometimes facilitate our work: microscopes, a fluoroscope, a
David Finley for the National Gallery of Art. custom-made apparatus consisting of a platform
Rush Kress, who as a businessman liked mounted on a hydraulic lift so we could work on
streamlined operating procedures, was frustrated. oversized paintings, a press for replacing faulty
He conceived a plan to make a single storage and cradles on panel paintings, relining, and so on.
work space on a property called Huckleberry Hill When I hired Gustav Berger, later to become
that he owned in the Pocono Mountains of Penn- famous for his work with adhesives, he built us
sylvania. In addition to the practicality of having one of the first vacuum hot tables for wax relining
the entire collection together, with the beginning according to the Dutch method.
of hostilities in Korea he was infected by the gen- In addition to Robert Manning, somewhat
eral fear of an atomic attack on New York City later, Sandrino Contini-Bonacossi, the nephew
and wished to protect the Kress Collection. By of Count Alessandro, was also engaged by the
October 1, 1951 it was finished, and we went to Foundation and both men were a great help
inspect it prior to the board meeting. assisting me in the overwhelming details of the
Huckleberry Hill was very remote indeed; the Regional Gallery Project. Among their many
Fig. 5. The studio at Huckleberry Hill in the 1950s.
Fig. 6. The studio at Huckleberry Hill in the 1950s.

assignments was the record-keeping function. It records are correct and therefore valuable, but
was essential to know at a glance the location of they cannot be relied upon entirely, and we con-
each painting, its status in terms of the National tinue to find many errors.
Gallery of Art and regional gallery directors’ Shortly after my arrival Rush Kress asked me
choices, its condition, whether or not it had been to come to 1020 Fifth Avenue to look at the
restored, if it needed restoration, reframing, and part of the collection hanging there. He proudly
to have the appropriate photographic documenta- stopped in front of a portrait of a woman and
tion as an aide mémoire. In a memorandum from asked me what I thought of his “Leonardo.” I
Guy Emerson to Rush Kress, the importance of was sorry to have to tell him that it was by
record keeping is stressed and mention is made Giampietrino. It had recently been acquired from
of a card kept for each painting, detailing the Duveen. Kress immediately called the Foundation’s
restoration. Although I have been asked many lawyer, O.V.W. Hawkins of Duer, Strong, and
times about this card file, frankly I do not remem- Whitehouse. We all went off to Duveen’s, at that
ber it; if there was one, it no longer exists, either time still at Fifth Avenue and 57th Street, in a
at the Kress Foundation or in the National Gal- magnificent beaux-arts building by Carrere and
lery of Art Archives. What I do remember is Hastings, now demolished. The negotiation was
that somewhat belatedly, Mary Davis, who had complicated by the fact that Suida had published
become the director of the Foundation, engaged the picture as a Leonardo. Notwithstanding this
Henry Hecht, my former reliner, to make condi- inconvenient detail, since he did not wish to lose
tion and restoration reports, including Pichetto’s an important client, Edward Fowles of Duveen
work, for every painting. The records that he agreed that we could choose something else from
made were all done at more or less the same time, their stock. I spotted a beautiful portrait by Peter
and, as I recall, often from memory. Sometimes Paul Rubens of the Marchesa Brigida Spinola Doria
Henry asked me for clarification, and it’s apparent and a small Madonna and Child, called studio of
from the original handwritten restoration reports Verrocchio that I believed was by the young
that I reviewed some of them and made correc- Leonardo working in the master’s shop. After
tions. The information in these reports was further negotiations, it was agreed that we could
transcribed by Fern Rusk Shapley for the Kress have the two paintings for a small additional
catalogues. She often asked me to explain the payment, and the deal was settled. The Rubens
condition of certain paintings, which, on exami- is today in the National Gallery of Art as is the
nation, seemed not to square with the records. At little Madonna and Child, attributed to Leonardo.
that point, the paintings were already dispersed, I found an antique fifteenth-century tabernacle
and we relied on photographs, X-radiographs, for the Madonna and Child to replace the Ferruccio
and my good but not faultless memory. Time Vannoni frame provided by Duveen.
did not permit me to properly assess the con- When I visited Washington for the first time
dition of all 2,000 and some odd paintings and, with Rush Kress, David Finley and John Walker
not infrequently, egregious errors found their took us on a tour of the galleries. It was the be-
way into print. For example, the little Madonna ginning of a long and strained relationship with
and Child attributed to Leonardo in Verrocchio’s Walker. I noticed several fakes on exhibition, two
workshop is described as “abraded in flesh tones “Vermeers” in the Mellon Collection, and in the
and hair of the Virgin and Child; the mantle and Kress Collection, a Madonna and Child given to
the landscape have suffered from drastic cleaning.” Alesso Baldovinetti that had been bought from
I do not know where this evaluation came from, Duveen for $300,000, a huge sum in 1939 when it
as I consider the small painting, which I bought was purchased on the recommendation of Bernard
from Duveen’s, to be in an excellent state, Berenson who congratulated Samuel Kress on his
although I never cleaned it. In many cases, the acquisition of “one of the most beautiful Renais-
sance paintings in America.” “BB” was nearly a
god for Walker, who had been his pupil, and con-
tinued to manipulate him from Settignano. I saw
that the painting had originally been on panel
and had been transferred to canvas. I was quite
sure that it had come from a famous Italian
dealer and forger, Baron Michele Lazzaroni, who
sold many pictures to Duveen. Lazzaroni usually
bought paintings by minor artists and then had
his restorer in Paris, who was called Verzetta, turn
them into “masterpieces” by some important
Renaissance artist, although sometimes he would
also ruin perfectly good pictures just for the
pleasure of altering them. Walker was extremely
upset by my assertion. To prove my opinion, I
offered to X-ray it, and about a month later it
was sent to New York. Under the “Baldovinetti”
was a quite different Madonna and Child which
seemed to be by Pier Francesco Fiorentino, a pro-
lific imitator of Pesellino. The forger had copied Fig. 7. Madonna and Child, repainted in the style of Alesso
a photograph, printed in reverse, of a famous Baldovinetti, mixed media, transferred from wood panel
Baldovinetti in the Louvre. Even when Walker to canvas, 29 1/2 × 21 1/2 in. (75 × 54.5 cm). National
Gallery of Art, Washington, D.C. Before cleaning.
saw the X-radiograph, he was not entirely con-
vinced, and he asked me to clean the painting.
My work revealed the half-ruined Pier Francesco
Fiorentino that is still in storage at the National
Gallery of Art (figs. 7 and 8).
After the 1951 Kress exhibition in Washington,
a moratorium was declared on new purchases
while, python-like, we digested the enormous
number of paintings already in the collection.
Suddenly, in 1952, John Walker learned that a
Grünewald Crucifixion, privately owned, had been
released for sale in Germany, negotiations with
the authorities there having broken down. Guy
Emerson broached the matter to a skeptical Rush
Kress: “an emergency matter has come up which
I hesitate to lay before you … however our policy
of not buying paintings at the moment always
had the qualification that we must consider
exceptional items when they came on the mar-
ket.”17 The price was $260,000, and if we had
not acted quickly any number of other buyers
would have snapped it up. Kress was not partic- Fig. 8. Madonna and Child, the underlying Pier Francesco
Fiorentino, tempera, transferred from wood panel to
ularly impressed by the photographs, but was canvas, 29 1/2 × 21 1/2 in. (75 × 54.5 cm). National Gallery
ultimately persuaded, and it is the only Grünewald of Art, Washington, D.C. During cleaning.

Fig. 9. Frame exhibition at Studio d’Arte Palma, ca. 1947.
in America. the best for the Kress Collection, and among the
A second Kress exhibition in Washington was paintings offered there were a number of second-
in the works for 1955, and several of the Regional ary works that we did not need.
Collections had opened to great acclaim in 1953. I was very keen that the paintings should all
We again began to actively acquire paintings, have beautiful frames. My father had collected
not only in New York but also in London and antique frames, as did Contini-Bonacossi, who
especially Paris. Walker and I visited the Villa always tried to find an appropriate period frame
Vittoria in Florence, the magnificent residence not only for works in his own collection but also
and private museum of Contini-Bonacossi for the paintings he sold to Samuel and Rush
(now the Palazzo dei Congressi) where we met Kress. Over the years I added to my father’s frame
with the Count and Gualtiero Volterra. I took collection and, as I mentioned earlier, mounted
Walker aside and made suggestions about what the first exhibition ever of antique frames at my
we might choose from the large group of pic- Studio d’Arte Palma in Rome in the late 1940s
tures being offered by Contini-Bonacossi; these (fig. 9). When I closed the gallery in Rome I sold
included a work by the Master of the Badia a my frame collection to Contini-Bonacossi. In 1953
Isola, Titian’s ceiling of Saint John the Evangelist, the Foundation bought about 500 frames from
Bronzino’s portrait of Eleanora of Toledo, and him, including some from my collection. We used
an important Savoldo. Although we tried to be these to reframe paintings whenever possible, not
discreet, Contini-Bonacossi realized what was only for Washington but also in the Regional
going on and, when Walker left, made his dis- Collections.
pleasure clear: he was accustomed to selling the We removed many of the modern frames
entire lot to the Kress Foundation without Duveen had used, especially on the Italian paint-
anyone’s interference. Although he and Volterra ings. Duveen had a wonderful frame maker,
were old friends, my priority was to buy only sometime forger, in Florence, Ferruccio Vannoni,
who designed quirky, beautifully crafted modern Feller’s suggestion.
interpretations of Renaissance models, each one The degree of matte and gloss could be ad-
slightly different. They are interesting in them- justed by locally varnishing with more medium.
selves and immediately proclaim their provenance, The alcohol diluent evaporated very quickly so
which was the intention. Quarantelli, our Floren- that it was possible to build up the restoration
tine framer, was a magician at cleverly adapting without picking up the color that had already
antique frames so they looked as if they had never been put down. At first we added a bit of
been touched. Naturally, it was not possible to use bleached beeswax, although I later abandoned
every frame. Those remaining were given to the that practice, as it was really not necessary. I
Metropolitan Museum and the National Gallery continued to use watercolor for some glazes and
of Art. to patinate the underpaint.
For years I had used egg tempera glazed with The first painting I restored using the pva ayab
drained oils or watercolor for retouching and medium was a Perugino Madonna and Child, now
dammar as a varnish. Although I had always used in the National Gallery of Art. When I saw it
varnishes as thinly as possible on the theory that recently, the restoration had not altered in the
it was the varnish, not the original painting that slightest way. Hanging nearby is a painting by
deteriorated, I still sought a more stable alter- Signorelli, Madonna and Child with Saints and Angels,
native to the traditional materials, all of which that I had restored only a few years earlier using
altered or darkened. I hoped that some of the egg tempera, watercolor and drained oils; those
new synthetic resins might be suitable as var- retouches are now distinctly discolored as are
nishes and retouching mediums. At John Walker’s those of the Mantegna portrait I restored with
suggestion I contacted Dr. Robert Feller, a scien- the same technique. Other paintings in Washing-
tist at Carnegie Mellon University in Pittsburgh, ton restored in the 1950s using pva ayab include
whose work on artists’ materials was funded the severely damaged Allegory by Piero di Cosimo,
by the Mellon Foundation, to collaborate on a varnished with Talens Rembrandt and wax, the
project to find a new retouching medium. Ercole Roberti, The Wife of Hasdrupal and her Chil-
He supplied me with a number of different dren, and the Giovanni Bellini Madonna and Child.
synthetic resins: various methacrylates, polyvinyl All of them still look perfect. This is also true of
alcohol and polyvinyl acetates, all of which were an extensively damaged altarpiece by El Greco in
considered to be stable. We began to use the the Metropolitan Museum, The Vision of Saint John,
new materials in 1953. In the beginning these that I restored in 1956 with the new resins. The
synthetic polymers were quite difficult to handle late Dr. Hubert von Sonnenburg, former Chairman
but with persistence and by altering the solvents of Paintings Conservation at the Metropolitan
and the viscosity of the solution, we finally Museum of Art, who shared my preference for
came up with a satisfactory application tech- thin varnishes, revived the dull surface with a
nique. We settled on a resin, a polyvinyl acetate spray of solvent many years later.
made by Union Carbide with a relatively low I also abandoned dammar varnish in favor of
molecular weight classified by the manufacturer one made from a synthetic resin, a polycyclohexa-
as pva ayab. An adhesives company supplied none condensation resin, known as aw2. I have
it under the name of Palmer’s a70, that is, a always used the commercial formulation made by
seventy-percent solution in acetone that we Talens called Rembrandt Varnish, developed as a
diluted to an eight-percent solution in methyl conservation varnish.18 Feller was experimenting
or ethyl alcohol, approximately the viscosity of with other resins. His methacrylate varnish, called
a retouching varnish. For certain purposes we Mellon 27h, was colorless and had good handling
wanted a more slowly evaporating solvent as an properties. I used it on a number of paintings
additive and chose methyl cellosolve, again on until one day an alarmed Feller sent out a general

alert that new aging tests demonstrated that the synthetic. Each panel had one section covered
resin cross-linked under certain circumstances— with Plexiglas, one with aluminum so it was pro-
that is, the varnish might become insoluble over tected from the light, and one section exposed to
time. The National Gallery of Art was very con- direct sunlight.19 We left these samples under the
cerned by this news. In December 1957 I went to skylights of the National Gallery of Art for eight
Washington to meet with Feller, John Walker, months. The results demonstrated, among other
Guy Emerson, Perry Cott, curator of sculpture, things, that the colors bound with ayab had not
and Frank Sullivan, a former Pichetto assistant altered and that the combination of dammar and
and preparator at the National Gallery of Art, zinc white produced blanching. Over the years
and the entire conversation was taped. Feller Feller identified other causes of blanching. In 1959
explained the cross-linking of 27h, and he urged blanched retouches on a Canaletto, The Piazzetta,
us to remove it from all the paintings that we had were found to result from use of the unstable
already restored; he was extremely nervous, not anatase form of titanium oxide, a photochemically
wanting to be blamed for anything that might active variety of the crystal. This alteration of
happen in the future. In fact, I habitually put titanium has given the pigment an undeservedly
a coat of dammar, Talens Rembrandt Varnish, bad reputation, especially in Germany and Austria.
polyvinyl acetate or beeswax under Mellon 27h, The rutile crystalline form, which one must be
using it only as a final varnish. I had to sign a careful to confirm, is completely stable.
paper that I had been advised of the danger Feller was one of many visitors to Huckleberry
and guarantee that I always used an intervening Hill in the early 1950s. By this time the dispersal
varnish under 27h. So much for artificial aging. of the entire Kress Collection was well underway
Not only has 27h remained soluble in mild as was our principal task, the preparation of hun-
hydrocarbon mixtures, but a bottle has sat on dreds of paintings destined for the National
my windowsill for fifty years and is still water Gallery of Art and the regional museums. Walker
white. On paintings, I have noticed that it has made an occasional day trip, and the directors of
a tendency to become dull and slightly gray, like the Regional Collections came to choose paint-
all the methacrylates. Along with everyone else ings for their collections. For overnight guests,
I abandoned 27h after Feller’s warning. Newfoundland boasted a small hotel. Some of
Pichetto’s often gratuitous restorations contin- our more interested clients, such as Walter Heil of
ued to blanch, an unsightly phenomenon that the M.H. De Young Memorial Museum of San
affected many of the Kress paintings already in Francisco, visited often to see the progress of
the permanent collection of the National Gallery the restoration on their paintings and what was
of Art. On our periodic visits, we glazed the new in the ever-expanding Kress Collection. The
whitened areas but shortly thereafter the blanching winter snows reached two or three feet in height.
returned. The possible causes of blanched and Often we opened the door in the morning to find
darkened varnishes and the materials that were a white wall blocking the entrance, and we could
subject to these alterations were the focus of much not go out until the plow came to clear the drive.
discussion with Feller. In 1959 we made up panels This did not impede our work as our living
using eight pigments mixed with three different quarters were on the third floor, right above the
whites: lead white, zinc white, and titanium white studio. We worked from Monday morning until
in varying proportions. Each of these was painted Friday afternoon when everyone returned to New
out with all the commonly used retouching medi- York for the weekend. Occasionally we would be
ums such as egg tempera, dammar, and so on, as snowed in, cursing the beastly weather.
well as the new ayab, along with some other resins We had a housekeeper and a cook and took
supplied by Feller, and the samples were coated all our meals together at a long table. I was at the
with many different varnishes, both natural and head, and everyone else sat in their accustomed
place. Relations were not always smooth as the museums would come to the Pennsylvania bunker
various personalities conflicted: Angelo Fatta and to look at the collection. Some directors had a
Quarantelli in particular did not get along. Paul preference for a particular school that was not
Kiehart often fanned the flames of this conflict, represented in the collection or that reflected the
and poor Angelo, with his strange Brooklyn ethnic background of their region and would
dialect, was the butt of everyone’s jokes. The request that the Foundation acquire paintings to
atmosphere was a bit like a military barracks; in fill in those gaps. We made such purchases often.
fact, one of our cooks had been an army chef Tintoretto was very popular, and altogether we
who made soup using a piece of lard attached bought fifteen canvases by Jacopo and his studio.
to a string. Once a museum director sent us a Each museum was given approximately forty paint-
present of wonderful filet steaks that our cook ings. Often a director would ask my advice about
reduced to tasteless cardboard. He didn’t last long. the attribution, the condition, and the quality of
There was a nearby trout stream, and in the good the works; this I offered dispassionately, not wish-
weather everyone fished after work. It was not ing to favor one museum over the other. The
wise to venture too far into the woods that were normal procedure began with a visit to storage
inhabited by bears and wildcats. where the paintings hung on numbered sliding
As word of the program spread, many cities screens arranged according to period and school,
applied to the Foundation. Most of them did not easy to locate. This initial examination was fol-
have a museum. One of the requirements was that lowed by lunch with the staff during which the
the recipients provide a suitable space to house the paintings under consideration were discussed. In
collection. When the project was approved, the the afternoon we returned to the storerooms and
directors or representatives of the various regional again looked at paintings, making new selections,
Fig. 10. Visiting the National Gallery of Art in 1951. From left to right: Sandrino Contini-Bonacossi, David Finley, Perry Cott,
Rush Kress, Colonel McBride, unknown, Alessandro Contini-Bonacossi, Fred Geiger, Guy Emerson, Patricia Volterra,
Gualtiero Volterra, Mario Modestini, and unknown.

eliminating some paintings and adding others. whom Kress had never met, offered to give thanks
Rush Kress was obsessed with every detail for the meal and quoted a long passage from the
regarding the display of the Kress Collection at Bible. Kress was a passionate devotee of the scrip-
the National Gallery of Art: how the paintings tures, and he asked his guest if he knew the Bible
were appended, at which angle, the wall color, the well. Hamilton replied that he knew it by heart.
labels on the frames, how didactic material would Kress was jubilant and sent one of the children
be made available since he disliked wall labels, the to get The Book. As he began a line, Hamilton
lighting, and so on. Periodically we would go to finished it. They became great friends after this,
Washington to make an inspection of the Kress and Kress directed me to select a particularly fine
galleries. On several occasions Contini-Bonacossi, group of paintings for the museum in Raleigh. I
whose collection at Villa Vittoria was impeccably knew that they already had a good collection of
displayed and whose advice Kress greatly valued seventeenth- and eighteenth-century paintings
on all such matters, would accompany him to acquired by Valentiner and that they needed a col-
Washington for an inspection tour together with lection of early and Renaissance Italian paintings
a large retinue (fig. 10). He held the Regional and a few important Baroque canvases. Altarpieces
Collections to a similarly high standard. I would from the Cook Collection by Massimo Stanzione
travel to each city, accompanied by Guy Emerson and Domenichino had been offered to the Nat-
and the lighting designer, Abe Feder, to install ional Gallery of Art. Walker, naturally, was not
the Kress galleries, and Kress joined us for the interested. Nor did he want to take a five-panel
opening ceremonies. On certain occasions he was polyptych by Giotto and assistants originally
deeply disappointed, for he had very clear tastes. painted for the Bardi Chapel in Santa Croce that
In Seattle the young curator, Sherman Lee, had we had assembled from different dealers. When
painted the galleries black. Kress was furious and I offered it to him he refused because, he said,
shot off memos to all and sundry. they already had a Giotto. “Mario, we might con-
Rush Kress was not autocratic but instinctively sider taking the central panel but not the four
generous by nature, as for example when he saints.” I bit my tongue but was appalled by his
learned that Contini-Bonacossi’s nephew Sandrino ignorance. A few days later Hamilton came to
was in difficulty, he immediately brought him to talk about the schedule for consigning the paint-
New York to work for the Foundation; however, ings to Raleigh, the catalogue and so on. I said to
he could also be fanatically parsimonious. Once him, “Carl, I had a thought. In order to complete
at Huckleberry Hill, a large number of paintings the Kress Collection in your museum you should
arrived from Contini-Bonacossi, a two-or-three have one painting of world-class importance, the
million-dollar shipment, very elegantly wrapped. Giotto polyptych from the Bardi Chapel.” He
As we eagerly cut the ribbons to open the pack- looked at me in amazement and nearly fainted.
ages, I noticed Kress carefully picking them up Like Berenson, Walker did not understand
and rolling them! painting after the Renaissance. He turned down
One of the most interesting characters to take the great Caravaggio Saint John the Baptist because
part in the regional gallery collections was the Berenson considered it a copy of the painting
legendary Carl Hamilton. He came to Pennsyl- in Naples. On that occasion my distress was so
vania with the director of the Raleigh Museum in great that Rush Kress authorized me to try to
North Carolina, to whom, being a native of that buy the painting anyway. Unfortunately it was
city, he had offered himself as an advisor. His too late as Kansas City had already reserved it.
credentials were impressive, as he had once been Washington has yet to have an opportunity to
a great collector himself. Rush Kress invited add a Caravaggio to the collection. Nor was he
Hamilton to dinner at 1020 Fifth Avenue, where particularly enthusiastic about the Saint Lucy
he always asked someone to say grace. Hamilton, Altarpiece20 or the great François Clouet portrait
of Diane de Poitiers in her bath (A Lady in Her painting of the sixteenth century that has never
Bath). We decided to buy both those paintings been relined. The linen was in good condition.
despite his lack of interest and ultimately, if When the painting arrived in my studio I simply
somewhat reluctantly, he took them. had the edges reinforced with strips of canvas
One morning a woman came to the Foundation and mounted it to a stretcher. I searched among
with a photograph of an unpublished painting our collection of antique frames looking for
by Giovanni Battista Tiepolo, The Empire of Flora, something suitable. I found a sixteenth-century
for which she was asking $15,000. It was in excel- Venetian frame by Luca Mombello, Titian’s frame
lent condition under a bit of yellow varnish. maker, which was about the right size. I had the
After it was cleaned we sent it to Washington. frame sent to the studio and put the painting in
Walker and his curators all agreed that it was by it. To my wonder and amazement, it fit perfectly.
Giandomenico Tiepolo and sent it back to New As paintings were not standard sizes in the six-
York. Disgusted by the Gallery’s response, at teenth century this coincidence was almost spooky.
that moment I was assembling the collection for I was often in and out of New York to visit
Walter Heil, the director of the museum in San dealers, attend Foundation meetings and also
Francisco and a good connoisseur whom I have made frequent trips to Washington. So that I
mentioned as a visitor to Huckleberry Hill. I could work on as many paintings as possible,
showed him the picture and explained why the in 1954 we took a studio at 16 East 52nd Street,
National Gallery of Art had rejected it. “Are where I worked with some of my assistants while
they blind?” he exclaimed. The painting is today we continued to commute to Huckleberry Hill
considered by one and all to be by Giovanni preparing for the 1955 exhibition at the National
Battista and is one of the masterpieces of the Gallery of Art. Finally, after seven years at
Fine Arts Museums of San Francisco. Huckleberry Hill, my assistants went on strike.
One day I received a phone call from Walter The Korean War had ended two years earlier. I
Hoentschel of Knoedler Galleries asking me talked the situation over with Guy Emerson who
to come and see a Titian portrait of the Doge brought it up with Rush Kress. We decided to
Andrea Gritti that they had acquired in Vienna. bring the restorers back to New York and found
My first thought was that it must be a copy of a studio at 250 West 57th Street in the Fisk Buil-
the famous portrait from the Czernin Collection, ding, just across the street from the offices of the
but I decided to look at it anyway. It was in the Foundation at number 221. This arrangement
Morgan Manhattan Storage Warehouse. The made it easier for the directors and curators of
painting had been rolled, fortunately face out, the regional museums to follow the work on
and we laid it out on the floor. It was in excellent their collections. Storage was still at Huckleberry
condition under an old discolored varnish. It Hill where framing and panel work continued
had never been relined, and there was a drawing, to be done. Angelo Fatta the carpenter and the
a study of the Doge, on the back. Astounded, Florentine framer Quarantelli were both men of
I thought that it must have been stolen but a certain age, one with grown children and the
Hoentschel assured me that the director of the other a widower who didn’t mind being in an
Österreichische Galerie in Vienna, Ernst H. out-of-the-way spot. After the move the atmos-
Buschbeck, considered it the work of Palma phere of the studio improved greatly, and there
Giovane, and had granted it an export license. I was a return to the easy, friendly and sociable rela-
immediately called Rush Kress who was at the tionships that we previously enjoyed and which
Foundation that morning and told him he must are essential to any group of people who work
come straight away and bring Suida with him. closely together.
Needless to say, as soon as they arrived we bought Throughout the 1950s it was possible to buy
it on the spot. It is exceedingly rare to find a important Italian Baroque pictures for $8,000

to $15,000 dollars and sometimes for less. We lection in Washington consisted of 365 paintings,
paid $1,200 for the masterpiece by Donato Creti, 82 pieces of sculpture and over 1,300 other
Alexander the Great Threatened by his Father that was works—medals, plaquettes and small bronzes
exhibited in the 1955 exhibition along with the acquired from the Dreyfus Collection. We had
early, Caravaggesque Simon Vouet Saint Matthew. accomplished a great deal and were very proud of
Both have remained in Washington. Rush Kress “The Kress Collection” that we had assembled
was very much in favor of these acquisitions, and dispersed far and wide across the nation
which he called “bucolic” pictures, partly because according to Samuel Kress’s conviction that art
they were a great bargain.21 One of my favorite and beauty were essential for the education of
dealers in New York was David Koetser, who had young Americans and the formation of good
great taste for seventeenth- and eighteenth-century character and values.
Italian paintings and a large stock. We bought I have some regrets. Primarily, of course, that
thirty-two paintings from him despite Walker’s we were not able to form the collection of the
indifference. The 1961 exhibition of recent acquisi- National Gallery of Art as we wished. Also I
tions at the National Gallery of Art included would have liked to have had time to restore many
many of these paintings. Suida and I campaigned other paintings from the Kress Collection, partic-
for several Baroque galleries in Washington. To my ularly those in Washington—something we had
great disappointment, Walker was not willing to always intended to do. Consequently many impor-
take more than two or three of the works we had tant works still have the thick and discolored
lent. So the Regional Collections became the varnishes added by Pichetto, now dull, dusty and
recipients of masterpieces by Tanzio da Varallo, streaky, and full of blanched retouches.
Sebastiano Ricci, Gaulli, Ceruti, Traversi, and Another cherished project often discussed
Magnasco, and other paintings that I had hoped was a Kress institute to train paintings conser-
would go to Washington. vators and, in particular, young Americans,
Numerous and important acquisitions were because, at that time, most restorers came from
also made of non-Italian paintings, among them Europe. Although there are several restorers
two portraits by Jacques Louis David, including whom I consider my pupils, the frenetic activity
his full-length Napoleon, three paintings by from 1949 to 1961 did not allow time to realize
Ingres, canvases by El Greco, van der Hamen, and this dream, a great pity since we had the chance
Zurbarán to name just a few. Our purchases of at that moment to exert great influence on the
Netherlandish paintings were outstanding and of approach to restoration of Old Master paintings
great importance to the National Gallery of Art in this country. It might have been possible to
which possessed only a few works of this school. avoid some of the destruction and controversies
We added German works by Dürer, Cranach, that later ensued.
Holbein, Grünewald, Altdorfer, Baldung Grien, I remained consultant to the Kress Foundation
and Flemish, Netherlandish, and Dutch paintings and also to the National Gallery of Art for many
by Robert Campin, Memling, the Master of the years, working closely with Fern Rusk Shapley,
Saint Lucy Legend, Hieronymus Bosch, Mabuse, Ulrich Middeldorf and Colin Eisler on the Kress
van Orley, Sanraedam, Ruisdael, Jan Steen, Pieter catalogues and visiting Kress Foundation restora-
de Hooch, van Beyeren and, of course, Rubens. tion projects in Europe. The Kress Collection is
Most of these paintings were acquired from of great importance to me to this day.
Knoedler, Wildenstein, Rosenberg and Stiebel, After the deeds of gift had been made to
Seligmann, Mont, Koetser, Mitchell Samuels of Washington and to the Regional Collections,
French and Company, and Schaeffer Galleries. I was moved by a letter from Franklin Murphy,
The final deeds of gift were made and a great the long-time Chairman of the Board of the
exhibition held in Washington in 1961. The col- Kress Foundation and a man I greatly respected.
It is dated April 2, 1962: Dianne Dwyer Modestini is a paintings conservator,
consultant to the Samuel H. Kress Foundation and Adjunct
Dear Mario:
Professor at the Conservation Center of the Institute of
Now that the Kress Gift to the Nation has been Fine Arts, New York University.
consummated and this project draws to a close,
I want to express to you personally and on behalf
of all of the Trustees our enormous gratitude
Mario Modestini was the Curator and Conservator of
for your dedication in making this whole thing the Samuel H. Kress Foundation until 1961. Based in
possible. It is my own view that you have been New York, he continued to restore paintings and advise on
a crucial enzyme in this entire process. Your questions of connoisseurship to an international clientele
competence—indeed, virtuosity in restoration including important museums and private collections.
has been the central fact in this project, and,
in a way, the collection is as much a monument
to you as to anyone else. Notes
Jackals may snarl and vultures may swoop but 1. Memorandum of Guy Emerson to Rush Kress, January 22,
the reality remains serenely unaware of both. 1949, Kress Foundation Archives.
2. Letter ccb to rhk, February 9, 1949. “As I explained to
All of us in the Kress Foundation and, in fact, you at length in my last letter, the choice of a candidate
the American public generally, will always be in who possesses the many necessary qualities restricts the
your debt. horizon considerably. Only one man—in my opinion—
has my complete and unconditional confidence; that is
I was fifty-five years old. For a brief moment the man who has the keeping of my own collection and to
whom I have always entrusted the most important works.
I considered returning to Europe, perhaps to Naturally I have always been very jealous of this man, as
London, but soon had more work than I could I consider him irreplaceable; therefore I have been faced
handle in my studio on 52nd Street and began a with a serious case of conscience … B.B. whom I believe
has always had a very high opinion of the way in which my
new chapter in my professional life. pictures are kept, did not even think I would be willing to
Through my friendships with Mary Davis, suggest him; but when I told him the news he seemed very
Franklin Murphy, and Marilyn Perry, I stayed in pleased and agreed entirely and he has also fully under-
stood and appreciated this solution …This man has
close touch with the Kress Collections and was the temperament of a Master. His technical and artistic
often called on for advice about the dispersed knowledge and his ability to inculcate into others love and
Kress Collection. Shortly after becoming director, care in their work make him substantially quite unique …
I do not think he would be able to dispose of more than
Marilyn Perry wisely decided that a review of the six months of the year for the U.S.”
Regional Collections Project was in order. I was 3. Count Giovanni Secco-Suardo, Il Restauratore dei Dipinti.
most gratified to hear that most of the restora- Milano: Ulrico Hoepli, 1918, p. 557. “N. . Pomata ammollente.
Predni sapon bianco di solda, di uello che serve per uso di toeletta,
tions we had done in the 1950s had held up very parti una; grasso di vitello stato bollito e stacciato, parti due; olio
well. However, the survey revealed that a number d’olivo del migliore, parti tre ed acqua parti sei. Tagliuzza minuta-
of paintings, primarily those that had had no mente. Meti il sapone e il tutto in vaso di tera vetrato, ed a fuoco
moderato, sempre rimescolando, portalo all’ebollizione, e lascialo
attention since early in the twentieth century and bollire sino a che quegli ingredienti si siano bene incorporati, ed
those that had passed through Pichetto’s hands, abbiano formato una pomata di mediocre densità. Versala in vaso
now required work. As I have said, I always regret- di terra o di vetro, lasciala raffreddare, poi coprila e conservala
pe’ tuoi bosogni. Dura indefinitivamente.”
ted that there was not time to put everything in 4. The other Pichetto assistant, Frank Sullivan, was a sort of
order. Since then many Kress paintings from the handyman. By some curious logic, John Walker hired him
Regional and Study Collections have come to the to be the restorer of the National Gallery of Art. Walker
claimed that Sullivan never touched the collection, but
Conservation Center of the Institute of Fine Arts several times I saw him working on Paul Mellon’s English
where I have been happy to follow the restoration. paintings, relining four or five at one time; he spoiled many
At this point in my life it is a delight to see some of them.
5. John Walker, Self Portrait with Donors. Boston and Toronto:
old friends again and to pass on my experience to Little, Brown and Company, 1974, pp. 142–3.
young conservators, and, in particular, the nearly
lost skill of restoring gold-ground paintings.

6. Although this would appear to conflict with Ann 14. This appears to have happened to the recently discovered
Hoenigswald’s paper in this volume, while it seems clear painting by Caravaggio, The Taking of Christ, now in the
that Pichetto did not clean paintings offered to Kress, National Gallery, Dublin.
he must have cleaned paintings for his other clients. 15. See Marilyn Perry, “The Kress Collection” in Chiyo
7. This so-called “Pichetto whitening” was noted quite early Ishikawa et al. (eds.), A Gift to America: Masterpieces of Euro-
by Modestini and further studied with a series of test pean Painting from the Samuel H. Kress Collection. New York:
panels that suggested the role of near-ultraviolet radiation Harry N. Abrams, 1994, pp. 13–35.
in the chalking of the inpainting film. See Robert Feller, 16. Marie Vassiltchikov, Berlin Diaries, –. New York:
“Problems in retouching: chalking of intermediate layers,” Vintage Books, 1988, p. 184. “I had to leave a message at
Bulletin of the American Group -IIC , Vol. 7, No. 1 (1966), the Adlon and in the hall ran into Giorgio Cini. He has
pp. 32–4. Other phenomena, such as chemical interaction come to Berlin to try and bribe the S.S. into setting his
between the basic pigment and acidic binder, may also father, old Count Cini, free. When Italy changed sides
play a role. Zinc pigments also seem to play a role in the last year, the latter (a one-time Minister of Finances of
chalking of original painting. See Karin Groen, “Materieel Mussolini) was arrested and has been in Dachau concen-
onderzoek aan schilderijen met wit mitgeschlagen parijen tration camp, in an underground cell, for the past eight
en de (on) mogelijkheden van restauratie” in Schilderkunst months. He suffers from angina pectoris and is in very
Materielden en Technieken. Amsterdam: Centraal Laboratorium bad shape. The Cinis have millions and Giorgio is willing
voor Onderzoek van Voorwerpen van Kunst en Weten- to pay anything to get him out. He himself has changed a
schap, 1989, pp. 38–46. lot since I last saw him just before the war. He evidently is
8. For example, the restorer William Suhr had a Viennese desperately worried. He adores his father and for many
worker, Eduard Kneisel, in his studio, who routinely trans- months did not know his whereabouts, nor whether he
ferred every painting on panel. Kneisel’s name was cited in still lived. Now he was waiting for some big Gestapo man.
a recent book, The Faustian Bargain, by Jonathan Petropoulos Who knows? With that amount of determination and will
(Oxford University Press, 2000, p. 279), where he is iden- power—and money—he may succeed. He wants them to
tified as working for the Nazis’ art looting program in agree to transfer his father to an S.S. hospital and from
Poland and where the writer says that Kneisel “reportedly” there to Italy.”
later worked for The Frick Collection, which he did not. 17. Memorandum to rhk from ge, November 24, 1952, Kress
9. Pichetto’s prices for this work were very high. A memoran- Foundation Archives.
dum in the files of the Kress Foundation records the sum 18. See: Sarah Fisher, “Talens Rembrandt Picture Varnish”
owed to Pichetto’s estate after his death and the writer’s in Sarah Fisher et al. (eds.), Painting Conservation Catalogue,
(presumably Emerson) consternation at the difference Vol. : Varnishes and Surface Coatings. Washington, D.C.:
between my fees and those of my predecessor, which were aic, 1998, pp. 99–107.
three times greater. 19. Gustav Berger, “Inpainting using ayab medium,” Cleaning
10. Mario Modestini, “Le vendite Rospigliosi: i ricordi di un Retouching and Coating, IIC Brussels Congress, September –,
testimone” in Angela Negro (ed.), La Collezione Rospigliosi: , pp. 150–55.
la Quadreria e la Commitenza Artistica di una Famiglia Patrizia a 20. The Assumption and Coronation of the Immaculately Conceived
Roma nel Sei e Settecento. Rome: Argos, 1999, pp. 91–3. Virgin by the Master of the Saint Lucy Legend and
11. The product is no longer made. Sue Ann Chui, with the Assistants, k-1689.
cooperation of the manufacturer, developed the following 21. See: Edgar Peters Bowron, “The Kress brothers and their
formula: to make Red Devil Liquid #99 from Zip Strip bucolic pictures” in Ishikawa et al. 1994 (cited in note 15).
Trigger Spray: to 100 g of Trigger Spray add 15.4 g meth-
anol, 26.1 g toluene, and 19 g acetone.
12. Painting in the Pinacoteca Nazionale in the Bologna
Photography Credits
gallery where you clearly see gray varnish applied over Figs. 7 and 8, p. 53. National Gallery of Art, Washington,
the gold ground and skirting the contours of the figures. D.C., Samuel H. Kress Collection (1939.1.214).
Newly burnished gold is extremely shiny and metallic and 1
while this may have been acceptable in the fourteenth cen-
tury, I think some of the more sophisticated formal values
that later emerged would have clashed with unpatinated,
highly burnished gold. One approach to key the painted
passages with the gold ground was the development of
elaborately punched, incised and glazed treatments of the
background. An egg-white varnish would have made the
gold slightly matte.
13. See: Helen Glanville, “Varnish, grounds, viewing distance,
and lighting: some notes on seventeenth-century Italian
painting technique” in Historical Painting Techniques, Materials,
and Studio Practice (Preprints of a Symposium, University of
Leiden, The Netherlands, June 26–29, 1995). Los Angeles,
CA: Getty Conservation Institute, 1995, pp. 12–19.
Technical Studies & Treatment
Madonna and Child with Four Saints
Goodhart Ducciesque Master, 1310 – 20
Egg tempera on cradled wood panel
Central panel 30 × 19 1/2 in. (76.2 × 49.5 cm);
side panels, each 24 1/8 × 13 5/8 in. (61.3 × 34.6 cm)
Birmingham Museum of Art, Birmingham, Alabama
61.104 (k-592)
Fig. 1. Madonna and Child with Four Saints, before cleaning and restoration.
Fig. 2. Madonna and Child with Four Saints (fig. 1), after cleaning and restoration.
64 Technical Studies & Treatment

A New Leaf: Recent Technical Discoveries
in the Goodhart Ducciesque Master’s
Madonna and Child with Four Saints
Jennifer Sherman
T his study focuses on a Trecento five-panel polyptych, Madonna and Child

with Four Saints by the Goodhart Ducciesque Master, an accomplished
follower of the Trecento Sienese artist Duccio, whose unique hand has been
recognized in a small corpus of works (figs. 1 and 2). Technical study of this
altarpiece from the collection of the Birmingham Museum of Art
enabled us to learn new information about its original appearance, the decora-
tive gilding materials and techniques utilized, and to discover more about its
conservation history.
Richard Offner identified this anonymous Sienese master and named him
for a Madonna and Child then in the collection of Mr. and Mrs. A.E. Goodhart
of New York.1 That panel, bequeathed to Robert Lehman, became a part of
Lehman’s bequest to the Metropolitan Museum of Art (1975.1.24). The Good-
hart Master was active from around 1310 to 1330. This painter has not been
linked directly through any specific evidence to Duccio or his workshop, but
judging from stylistic and iconographic considerations, his work is strongly
influenced by the precedents established by the great Sienese master. Other
important Sienese artists who influenced him are cited in the literature on
the Goodhart Master and include Ugolino da Siena, Segna Bonaventura, and
Simone Martini.2 It would appear that the Goodhart Master worked in and
around Siena, and was commissioned to provide paintings for some of its
provincial environs.3 From the limited number of extant works attributed to
this artist, it is clear that he excelled when working on a small scale; most of
Jennifer Sherman 65
his surviving paintings are panels from diptychs around 1941,7 so no technical information can be
or other small portable devotional objects. The gleaned by examining the backs. However, there
Goodhart Master delighted in the finer details of are X-radiographs in the Kress Archives that were
these miniature paintings, and there is a tender taken prior to the thinning and cradling.8 The
reverence and naturalistic charm that comes across report notes that “plaster” had been applied to the
in the smaller scale that is somewhat lacking in his backs of all the panels, and that the X-radiographs
larger works. The Birmingham altarpiece is the reveal more about the plaster than the planks
largest commission convincingly attributed to the themselves. It is clear that the Madonna and Child
Goodhart Master.4 Although there are many panel had suffered from considerable worm tun-
charming elements in the Birmingham polyptych, neling, and the damage subsequently filled with
as well as some extraordinary details, there is a gen- this “plaster.” Dowel holes used to align the panels
eral stiffness in the figures in the larger scale that are evident in the early “shadowgraphs,” linking
is absent in his more diminutive paintings. Despite the bishop saint to Saint John, and the Archangel
this minor shortcoming, the Birmingham altar- Michael to Dionysius, confirming the correct
piece is regarded as one of the most accomplished sequence of the panels. Although dowel holes are
(as well as ambitious) examples of the Goodhart evident in the adjacent sides of the two saints
Master’s production. flanking the Madonna, no dowel holes can be seen
The known provenance of the polyptych is in the “shadowgraph” of the Madonna panel.
rather scant and relatively recent. F. Mason X-radiographs taken by the author in 1994 (well
Perkins notes that it had “long remained, com- after the panels had been thinned) do not reveal
paratively unheeded, in the seclusion of an any evidence of the doweling. A faint pattern dis-
ancient Tuscan villa.”5 Samuel H. Kress acquired cernible in the recent X-radiographs confirms that
the altarpiece in 1941 from Count Alessandro all of the panels were covered with fabric prior to
Contini-Bonacossi in Florence, and in 1952 the application of the ground layer.
Kress Foundation gave it to the Birmingham The gesso preparation is fairly thick, as is
Museum of Art in Alabama. typical for tempera panels of the period. In the
The altarpiece is comprised of five panels X-radiographs it is clear the panels had many
with semicircular arched tops. The central panel knots and other irregularities in the wood grain
depicting a half-length Madonna and Child is which were subsequently filled with gesso to pro-
taller and wider than the four flanking panels. vide a smooth surface. In spite of this preparation,
This format, which may have included triangular some areas remained problematic, and gesso and
gables above each panel, was established in the paint losses (subsequently filled) and several consol-
first decade of the Trecento.6 From left to right, idation campaigns were evident in these locations.
the side panels represent an unidentified bishop The backgrounds of the panels were water
saint, Saint John the Baptist, the Archangel gilded, and the haloes of the saints were punched
Michael, and a saint tentatively identified as and tooled. Although the thick, discolored var-
Dionysius the Areopagite. The frame dates to nish obscured the gold grounds to some degree,
the twentieth century. it was apparent that some of the cracks in the
Each of the five panels was made from a single gesso beneath the leaf were quite old, and that
plank, presumably of poplar; no joins are evident many of these cracks extended into the original
in any of the panels. Numerous irregularities in paint layer and were, thus, original. Numerous
the wood grain of the panels have resulted in localized campaigns of repair and regilding of the
chronic flaking of the gesso and paint layers in plain gold backgrounds were evident (fig. 3). The
localized areas. Thin wood strips have been applied areas of regilding are particularly obvious at the
all around each panel, obscuring the original edges. joins where the wood strips had been added to
The panels were thinned and cradled sometime the edges of the panels. The punched and tooled

haloes appeared to be in remarkably good condi-
tion, the tooling crisp and well preserved, with
only a few minor exceptions. In some areas within
the haloes the gilding was a bit abraded, exposing
the underlying red bole.
The egg tempera paint layer was built up in
the traditional manner of the period. Local areas
of color were applied in thin layers consisting
of fine, hatched strokes. In general, each area of
color was kept fairly pure, with variations in
modeling for the lighter tones being achieved by
the addition of lead white to the local color, and
shadow by either the absence of white and/or
with deeper, transparent glazes of local color. A
green verdaccio was applied as a base color for the
flesh tones. The flesh tones applied on top of the
verdaccio are a mixture of yellow ochre, vermilion
and lead white. The brushwork is delicate and
fine, and the transitions between dark and light
are soft and smoothly rendered.
With the exception of some areas of localized
Fig. 3. Madonna and Child with Four Saints (fig. 1), Bishop Saint, loss, the paint layers of the five panels are in a
detail of gold ground and tooled halo, before restoration.
remarkably good state. The subtle modeling of
the flesh tones has survived almost intact with
little abrasion. In addition, some of the finest
brushwork, such as the delicate whiskers of
Dionysius’s beard, remain beautifully preserved
and are a testament to the Goodhart Master’s
refined sensibility, as well as his miniaturistic
predilections.
Despite the overall good state, there is consis-
tent paint loss along the contours of the painted
image where the paint had been applied on top of
the metallic ground. A later, clumsy restoration
extended beyond the original contours and over
the edges of the metal leaf ground, enlarging the
silhouette of each figure (fig. 4). Another signifi-
cant type of paint loss corresponds to irregular-
ities in the poplar planks, evident both as linear
losses related to irregular wood grain and larger,
broader losses related to knots. A third type of
paint/ground loss is noted in patches along the
bottom of each panel; these appear to be from
some sort of water damage.
Fig. 4. Madonna and Child with Four Saints (fig. 1), detail of
Madonna’s and Christ Child’s heads with surrounding gilding, There was much decorative gilding applied to
after cleaning. the painted draperies and attributes of the saints,
Jennifer Sherman 67
some consisting of a yellow metal leaf, some of from a later stage of restoration in which shell
a white metal leaf, and some appearing to be a gold (powdered gold in an aqueous medium) is
combination (perhaps an alloy) of yellow and used in lieu of metal leaf. Shell gold is obvious in
white metal leaf. We observed that some of the these passages, as the individual particles of the
brighter yellow and white metallic mordant leaf ground gold are visible under the microscope. The
gilding appeared to be a later restoration. These areas from this more recent campaign include the
additions were found on the medallion and star on the mantle of the Madonna’s proper right
border of the bishop saint’s cope (fig. 5), as well shoulder, the cuff of her sleeve and the top of the
as his miter, and on the archangel’s sword and mantle’s border directly above it, as well as on
brooch (figs. 6 and 7). All of these overgilded some abraded areas of the lining of her mantle
areas had a distinctive thickness and texture as below her neck. The gold quatrefoils of the bishop
well as a slightly gaudier appearance that distin- saint’s cope were also reinforced with shell gold.
guished them from the more subtle original mor- Originally the decorative leaf consisted of a
dant gilding. (For comparison, note the border of mordant gilt metal leaf with a tone somewhere
the bishop saint’s mantle along the right edge of between gold and silver, and toned or glazed with
the panel. This is the only remaining area of the translucent pigments to enrich its effect (fig. 8).
original unrestored mordant gilding on this panel.) A sample of this leaf was analyzed with scanning
The handling is less refined in these areas of electron microscopy—energy dispersive spec-
overgilding, and there is a rather crude attempt at trometry (sem /eds).9 The results confirmed that
incising pattern into the leaf in the restored areas the material contained both gold and silver,
that is not observed on the original mordant gild- with a proportionately higher amount of silver.
ing. This campaign of restoration is distinguished Examination by sem revealed that the leaf was a
Fig. 5. Madonna and Child with Four Saints Fig. 6. Madonna and Child with Four Saints Fig. 7. Madonna and Child with Four Saints
(fig. 1), detail of later restoration of (fig. 1), detail of the archangel’s hand hold- (fig. 1), detail of the archangel’s proper
incised and glazed mordant gilding on ing sword; compare the original mordant right hand with crude regilding of sword
the clasp on the bishop saint’s cope; gilding on the cuff with the clumsy, thick elements; note the use of a white metal
note the use of white metal leaf. regilding on the sword handle; note also for the sword blade and yellow gold for
the pentimento left from the tarnished the hilt.
zwischgold on the archangel’s hand.

as well as by the interruptions of the old craque-
lure where losses to the gesso had been filled. In
many areas the gilding lying over the old gesso
cracks appears to have been patched with an addi-
tional layer of gold leaf, perhaps at a later time,
with characteristic streaks left by the burnisher as
the leaf was rubbed over an already-compressed
gesso preparation.
The gilded and tooled haloes, demonstrating
typical Trecento Sienese craftsmanship and design,
appeared to be in a remarkably good state in
comparison to the general condition of the gold
Fig. 8. Madonna and Child with Four Saints (fig. 1), detail of well- grounds. Clearly there were some minor repairs
preserved zwischgold from the archangel’s diadem.
and abrasion in the haloes, but overall, they
appeared to be beautifully preserved, with crisp
laminate comprised of a thicker silver substrate, definition in the incised patterns.
upon which a very thin layer of gold had been Microscopic examination of the gold grounds
applied. This material is what was commonly of all five panels revealed several minute metal leaf
referred to in Northern artists’ treatises as fragments that appeared to be silver. Most were
zwischgold, or, in Italy as oro di metà (fig. 9). This located along contours of the painted images.
laminate leaf is distinctive in its appearance, Some were completely blackened, and others had
being neither bright gold, nor cool silver, but a the brownish cast of partially tarnished silver
rich, understated bronzy tone. It should be noted, (silver sulfide) (fig. 10). When scratched, white
however, that the tarnishing of the silver compo-
nent undoubtedly has contributed to the overall
darkening of this leaf; when newly applied, the
oro di metà would have been closer in appearance
to gold leaf.
Traditionally, oro di metà was used as an eco-
nomical alternative to gold leaf.10 It is mentioned
in treatises in three basic forms: one using thin
gold leaf hammered into the silver substrate; Fig. 9. Madonna and Child with Four Saints (fig. 1), cross-section
(200 ×) of zwischgold from the decorated border of the
another in which a mordant of some sort was Madonna’s mantle.
used to bind the two layers together; and a third
that uses a mordant between the two layers, which
are then hammered together. The terms zwischgold
or oro di metà appear to be used interchangeably to
describe any of these techniques. It was not deter-
mined which method was used for the oro di metà
on the Goodhart Master polyptych.
Each of the panels has a gold ground, with
a tooled halo decorated with a simple circular
punch. The gold grounds have undergone numer-
ous campaigns of repair and regilding. The many Fig. 10. Madonna and Child with Four Saints (fig. 1), cross-section
(200 ×) from central panel showing the silver ground along the
repairs and patches are easily identified by their left edge of the mantle below the shoulder; note the restored
different metallic sheens and in some cases colors, gold layer directly on top of silver leaf.
Jennifer Sherman 69
metal leaf was exposed on these fragments. We If silver grounds were part of the artist’s origi-
wondered whether these whitish metallic frag- nal conception, the polyptych takes on a whole
ments noted throughout each panel were stray new aspect. It seems to me that the artist’s palette
bits of some of the decorative gilding from the was clearly designed to complement the cooler,
garments and attributes of the saints, or whether, more subtle silver background. If we envision the
at some point, the panels could have had a white altarpiece with a more subdued silver ground,
metal background. During cleaning more silvery rather than the brighter, brassier yellow-gold
fragments were noted (fig. 11), especially along ground, combined with the rich, subtle contrast
the contours of the figures, underneath old in metallic tone with the mordant-gilt zwischgold
restoration. Further examination with the stereo- decorative elements, we can appreciate what a bril-
binocular microscope enabled us to find more liant colorist the Goodhart Master was (fig. 12).
silver fragments hidden beneath old restoration The cool silver ground set against the rich inter-
and gilded repairs. Elemental analysis (sem /eds) plays of gradations of warm and cool tempera
of a few samples located in representative areas color, coupled with the subtle subdued tones of
confirmed that these fragments were indeed the oro di metà must have created a stunningly har-
silver.11 It gradually became evident that the monious and sublime effect. And what a glorious
“gold grounds” that we now see had originally impression the whole ensemble would have made
been silver. when it first left the artist’s studio, before oxida-
tion began to alter the tone of the silver.
Silver leaf is mentioned in medieval artists’
treatises, but usually with a caveat about the tar-
nishing and blackening of silver that inevitably
spoils the splendid effect of the painting set
against a precious metal background. While silver
grounds for panel paintings are known to exist,
they are somewhat rare (see Appendix). However,
silver leaf is frequently used in Sienese painting
for decorative purposes on spandrels and framing
elements, as well as for specific details within the
painted composition. The incorporation of com-
binations of precious metals, often glazed with
translucent pigments to simulate enamel or to
create other effects, is an important feature of
Sienese painting. The aesthetic of the Sienese
School is characterized by elegance, sinuous line,
jewel-like color, rich patterning, and skillful
manipulation of precious materials.
There are several reasons why silver-ground
paintings are rarely encountered: first, the tarnish-
ing of the silver detracts from the magnificent
effect of the painting against a precious metal
ground. It is important to note, however, that
more silver-ground paintings may have been
created than the small number remaining extant
Fig. 11. Madonna and Child with Four Saints (fig. 1), detail of
blackened silver fragments from silver ground after cleaning would lead us to believe. Easily darkened, even dis-
(from bottom left corner of Saint John the Baptist). figured, by exposure to sulfur in the atmosphere,

Fig. 12. Madonna and Child with Four Saints (fig. 1), digitally re-colored with silver ground.
silver grounds most probably would have been between the background and the haloes was seam-
rubbed down and gilded with the non-tarnishing less and the tooling had clearly been done after
yellow gold leaf. The second reason is that gold the gold leaf had been applied and burnished.
is a more precious metal and thus regarded as a The absence of cracks in the gesso within the
more suitable material for use in sacred works. contours of the haloes made them suspect.
Inherent in the preciousness of gold is its cost, so Ultimately, after much examination under the
commissions of a more humble economic origin binocular microscope and consultation with
often stipulate the use of silver as an alternative Mario Modestini, we postulated that, in the areas
to the more costly gold leaf. However, at any time of the haloes, the original gesso had been carved
after the altarpiece was finished, a parish might out and replaced with a fresh preparation that
have raised funds to replace the tarnished silver would allow the gold to be burnished and tooled.
ground with the more precious gold. Finally, in This new gesso was made perfectly level with the
the early twentieth century, when collecting Italian rest of the background and then, over this hybrid
primitives became fashionable, dealers routinely ground, new bole and leaf were applied. After
replaced or camouflaged the damaged silver and burnishing, the new haloes were incised and
gold backgrounds, sometimes with paint, but punched in the manner of a Ducciesque Trecento
usually with gold, often clumsily applied. artisan; the leaf applied to the freshly gessoed
After considering that these panels originally preparation took on an even, mirror-like burnish,
had silver grounds, we were forced to confront the unlike the surrounding gilded areas which were
issue of the “remarkably well-preserved” punched streaky. The restorer might have copied the origi-
and tooled haloes, executed in gold leaf. While nal haloes although there are numerous contempo-
the punched and tooled haloes appeared to be rary examples that might have served as patterns.
well preserved and their motifs stylistically in The question still remained as to how the old-
keeping with other Sienese works of the period, est cracks in the background, which continued
we were unable to explain why the tiny crack into the paint layer and seemed to be in the gesso
pattern in the original gesso, present in the rest preparation, would still be present if the original
of the background, disappeared in the areas cir- silver grounds had been subsequently gilded
cumscribed by the incised outline. The transition with yellow gold. After studying the yellow gold
Jennifer Sherman 71
grounds of each panel it appeared that the present
gold leaf was applied to the original gesso and
successfully burnished in a masterful, if inexplica-
ble, way. The jagged edges of the contours of the
figures bore witness to the removal by scraping of
the tarnished silver and underlying bole. Grad-
ually we formed a hypothesis to account for the
technical evidence: the original gesso grounds,
after the silver and bole had been scraped off, had
been smoothed down, followed possibly by the
application of a thin layer of new gesso and bole
before regilding. After drying, the underlying
craquelure reasserted itself. The burnisher, as
noted earlier, had left streaky marks because the
underlying gesso preparation had already been
compacted by the burnishing of the original silver
ground. These marks were noted in all areas where
the gesso appeared to be original.
Fig. 13. Virgin and Child, Ambrogio
The procedure described above to create a new Lorenzetti, late 1330s–early 1340s, tem-
gold background is extremely difficult to execute pera on wood panel, 29 3/4 × 17 7/8 in.
with convincing results because even the slightest (75.5 × 45.4 cm). Museum of Fine Arts,
Boston, MA. In its present state.
surface variation is magnified during the gilding
process. It would take an extremely skilled techni-
cian to accomplish such a feat and we tried to
understand when and where this skillfully decep-
tive regilding had been done. A number of gifted
and knowledgeable restorers and artisans working
in Italy toward the end of the nineteenth century
and into the first half of the twentieth had great
success replicating early Italian painting and gild-
ing techniques for use in restoration and in the
creation of complete paintings “in the antique
style.” The most famous of this group is Icilio
Federico Ioni (also spelled Joni).
Ioni, who was born in Siena in 1866 and lived
there for his entire life, first learned gilding and
painting techniques at the hand of his uncle,
Giovacchino Corsi, whose important workshop
in Siena produced antique-style frames, taberna-
cles and other decorative works. Ioni quickly
mastered gilding techniques, and his precocious
drawing talents soon led him to study tempera
painting technique. These skills enabled him to
produce copies of paintings from the Pinacoteca
Nazionale di Siena, as well as concoctions for Fig. 14. Virgin and Child (fig. 13), as
the art market. His mastery and knowledge of restored by Icilio Federico Ioni.

traditional painting techniques attracted the example, we can see that Ioni freely altered dimen-
attention of many prominent collectors, scholars, sions and shapes of paintings so that they would
and dealers specializing in early Italian painting. appear less like fragments of a mutilated altar-
In his autobiography, Ioni describes numerous piece and more like whole works in their own
examples of battered early Italian paintings he right. He was confident enough of his own skill
was commissioned to restore. Along with the to remove all of the regilding surrounding the
legitimate restorations, many of his “antique- tooled haloes and regesso and regild right up to
style” paintings ended up on the international the edges of the original nimbi, with extremely
art market. During the last half of the twentieth successful and convincing results.
century, Ioni’s paintings were identified in collec- Ioni’s hand has been identified in the restora-
tions of many prominent museums in Europe and tion of the Metropolitan Museum of Art’s poorly
the United States. preserved Blumenthal Madonna and Child, given to
Some of Ioni’s activities have been identified Ambrogio Lorenzetti (41.190.26) (fig. 15). In this
with individual paintings, forgeries, and so-called case only a fragment of the original punched and
quadri antichi. Both Frinta and Skaug have con-
nected several paintings with Ioni, based on
evidence found through examination of the
punchwork. Kanter, Pope-Hennessy and Mazzoni
have documented numerous examples of Ioni’s
production, both original works in the “antique
style” and restorations.
The Boston Museum of Fine Arts’ Virgin
and Child by Ambrogio Lorenzetti (39.536) was
restored by Ioni (fig. 13). Kanter states that photo-
graphs of the painting after its early nineteenth-
century restoration document its state prior to
Ioni’s restoration, and he published a photograph
of the painting after Ioni’s restoration when it was
acquired by Daniel Platt (fig. 14).12 The photo-
graphs make it possible to compare the various
states of the painting and to gain an understand-
ing of Ioni’s approach toward restoration and to
appreciate his considerable skill. In the nineteenth
century this picture had a rectangular contour
and punched borders in its regilt ground.13 Ioni
removed the old retouching and regilding and
altered the shape of the panel by adding a curi-
ously conceived steep gable set within a pointed
arch. He then regessoed and regilded the back-
ground, leaving the original punched and tooled
haloes intact, and added an elaborate foliate
incised design in the narrow areas between the
outside of the steep gable and the created contour
of the panel. The paint layer was restored with a Fig. 15. Madonna and Child, Ambrogio Lorenzetti
(the Blumenthal Madonna), tempera on wood panel,
fairly liberal hand, reinforcing the facial features 37 × 22 1/8 in. (94 × 56.2 cm), The Metropolitan Museum
of the figures with a linear emphasis. From this of Art, New York, NY.
Jennifer Sherman 73
tooled haloes survives, located between the heads Another example of Ioni’s restoration work
of the two figures. The rest of the gold ground can be found in the Madonna and Child with Two
has been entirely regessoed and regilded. Here Angels, Saint Francis, and Saint Louis of Toulouse by
again, we see how Ioni’s bravado in applying new Paolo di Giovanni Fei in the collection of the
gesso and leaf up to the fragmentary remains of High Museum in Atlanta (fig. 16). The halo of
old tooled gilding met with considerable success. the angel on the left was restored. Frinta claims
Ioni provided the “missing” tooled decoration on that the design of the modern punch used to
the rest of the haloes and borders in a manner form the “formal cluster” in the restored halo is
consistent with other examples of Ambrogio’s based on the design of the original punch used in
production. In the case of the punch noted in the the better-preserved areas of tooled gilding from
restored portion of the Christ Child’s halo, Ioni the same painting. He identified this particular
actually recreated a punch design based on the punch in several paintings documented as having
Ambrogio Madonna and Child from the Pinacoteca been restored by Ioni, concluding that Ioni
Nazionale in Siena (no. 605).14 restored the Atlanta painting. In this panel, it
is evident that Ioni has gone to the effort of
manufacturing a tool that imitates the surviving
distinctive original punchwork, a telling indication
of his skill and initiative.
After examining some of the work known to
have been executed or extensively restored by Ioni,
it is not unreasonable to suggest that he, or some-
one in his circle of talented artisans, could have
restored the gold-ground haloes of the Birming-
ham altarpiece. Mario Modestini, who visited
Ioni in his studio and saw many of his produc-
tions first-hand, holds the opinion that the tooled
haloes may, indeed, have been executed by this
clever restorer and artist.15 Coming across the
anomalous “clues” during the early phase of treat-
ment forced us to stop periodically and hypothe-
size about what the significance of these bits of
evidence might be. Ultimately we came to under-
stand that the color of the gilded background
of the polyptych had been radically changed, and
that the haloes were masterfully regilded, punched
and incised, sometime in the early twentieth
century by a bold and skillful restorer. These
conclusions added a fascinating and unexpected
dimension to the restoration of this altarpiece.
It is hoped that future discoveries of this type—
finding evidence of silver grounds, or examples
Fig. 16. Madonna and Child with Two Angels, Saint Francis, and of individual restorers making significant changes
Saint Louis of Toulouse, Paolo di Giovanni Fei, ca. 1375, tempera
and silver gilding on cradled wood panel, 70 1/8 × 50 5/8 in. to a work of art—will be published and shared
(178.2 × 128.6 cm). High Museum of Art, Atlanta, GA. with the conservation community.

Acknowledgements 10. Skaug (1994), Vol. 1, p. 57, and Theophilus (1979), p. 156.
11. My thanks, again, to Mark Wypyski and Christopher
For her unstinting help and encouragement McGlinchey for the analysis carried out at the Metro-
with this project, and many others, I would like politan Museum of Art.
to thank my mentor, Dianne Dwyer Modestini. 12. Kanter (1994), p. 90.
13. Ibid, p. 90.
Mario Modestini’s encyclopedic knowledge 14. Frinta (1978), pp. 7–10.
presented an interesting avenue for me to pursue 15. Personal communications with M. Modestini throughout
with this project and his charming presence the course of treatment of the polyptych.
always made it fun. Special thanks are extended
to Lisa Ackerman for her unflagging commitment Bibliography
to the Kress/nyu program, and to Gail Cimino Bomford, David et al. (1989) Art in the Making: Italian Painting
before . London: National Gallery Publications.
who produced this beautifully designed volume. Coor-Achenbach, G. (1955) “Contributions to the study of
Ugolino di Nerio’s art,”Art Bulletin, Vol. xxxvii, pp. 163 f.
Jennifer Sherman received a Diploma in Conservation and Frinta, Mojmír S. (1978) “The quest for a restorer’s shop
of beguiling invention: restorations and forgeries in
an M.A. in Art History from the Conservation Center, Italian panel painting,” Art Bulletin, Vol. 58, No. 1 (March),
Institute of Fine Arts, New York University. She has a pp. 7–23.
private paintings conservation practice in New York special- ––– (1982) “Drawing the net closer: the case of Ilicio [sic]
Federico Joni, painter of antique pictures,” Pantheon,
izing in Old Master paintings, and has been associated Vol. 40, No. 3, pp. 217–23.
with the Kress Program in Paintings Conservation at Joni, Icilio Federico (ca. 1936) Le Memorie di un Pittore di Quadri
New York University for over ten years. Antichi. Florence: Società Editrice Toscana.
––– (1936) Affairs of a Painter [trans. of Le Memorie di un Pittore
di Quadri Antichi]. London: Faber and Faber.
Kanter, Laurence B. (1994) Italian Paintings in the Museum of
Notes Fine Arts, Boston, th–th Century. Boston, MA: Museum
1. The name was coined by R. Offner, the first scholar to of Fine Arts.
isolate and identify paintings by this Sienese master (see Maginnis, Hayden B.J. (2001) The World of the Early Sienese
Wehle 1940, p. 71). Painter. University Park, PA: Pennsylvania State University
2. For a review of literature on the Goodhart Ducciesque Press.
Master and works attributed to him, see Zeri (1980), Mazzoni, Gianni (1988) “Icilio Federico Joni,” Siena tra Purismo
Kanter (1994), Coor-Achenbach (1955), Shapley (1966), e Liberty (exhib. cat.). Milan and Rome.
Pope-Hennessy (1987) and Stubblebine (1979). ––– (2004) Falsi d’autore. Icilio Federico Joni e la Cultura del Falso
3. Stubblebine (1979), Vol. 1, p. 106. tra Otto e Novecento. Siena: Protagon Editori.
4. The altarpiece at Monterongríffoli has been given to the Olivetti, A. and Mazzoni, Gianni (1988) “Entre invention,
Goodhart Master by Coor-Achenbach (1955), pp. 163–4 restauration et vente: la peinture médiévale au début du
and Stubblebine (1979), pp. 109–10. Suida (1959, p. 11) xxe siècle,” Médiévales, Vol. 14, pp. 73–86.
also cites this polyptych. Kanter (1994, p. 81) convincingly Pope-Hennessy, John (1987) Italian Paintings in the Robert Lehman
argues that although there are similarities between the Collection. Princeton, NJ: Metropolitan Museum of Art.
Birmingham and Monterongríffoli altarpieces, the latter Shapley, Fern Rusk (1966) Paintings from the Samuel H. Kress
is “clearly dependent in style on the Goodhart Ducciesque Collection, Vol. : Italian Schools, XIII–XV Century. London:
Master.” Phaidon.
5. (Undated) manuscript opinion of Perkins on reverse of Skaug, Erling S. (1994) Punch Marks from Giotto to Fra Angelico:
photograph of k-592, Kress Foundation Archives. Attribution, Chronology, and Workshop Relationships in Tuscan Panel
6. See, for example, Duccio’s Polyptych No. 28, Siena Pina- Painting ca. –, 2 vols. Oslo.
coteca Nazionale. Maginnis (2001), p. 104. Stubblebine, James H. (1979) Duccio di Buoninsegna and his School,
7. Condition and Restoration Record, k-592, Kress Foun- 2 vols. Princeton, NJ: Thames and Hudson.
dation Archives. Suida, William E. (1959) The Samuel H. Kress Collection.
8. X-ray “shadowgraphs” and report, Alan Burroughs n.d., Birmingham, AL: Birmingham Museum of Art.
k-592, ibid. Theophilus (1979) On Divers Arts: The Foremost Medieval Treatise
9. My thanks to Christopher McGlinchey, now scientist at on Painting, Glassmaking and Metalwork. New York: Dover
the Museum of Modern Art and to Mark Wypyski of Publications.
the Metropolitan Museum of Art, both of whom were Wehle, Harry B. (1940) A Catalogue of Italian, Spanish and
at the Metropolitan when they assisted in carrying out Byzantine Paintings. New York: Metropolitan Museum
the analysis. of Art.
Jennifer Sherman 75
Wyld, Martin and Plesters, Joyce (1977) “Some panels from
Sassetta’s Sansepolcro Altarpiece,” National Gallery Technical
Bulletin, Vol. 1 (September), pp. 3–17.
Zeri, Federico (1980) Italian Paintings: A Catalogue of the Collection
of the Metropolitan Museum of Art—Sienese and Central Italian
Schools. New York: Metropolitan Museum of Art.
Appendix: Panel Paintings with Silver Grounds

This list is hardly exhaustive, and it would be worthwhile to identify
other examples in an attempt to understand whether in some cases the
motive for using silver instead of gold was aesthetic rather than purely
economical as is usually supposed. Future technical study and conser-
vation treatments of gold-ground panels may bring to light more pic-
tures that originally bore silver grounds.
Goodhart Ducciesque Master, Madonna and Child with Four
Saints (k-592), Birmingham Museum of Art, Birmingham,
Alabama (61.104). See Shapley (1966), p. 18.
Master of the Grosseto Madonna, Madonna and Child with
Saint Peter and Saint John the Baptist (k-577), Museo de Arte
de Ponce, Ponce, Puerto Rico (62.0255). See Shapley
(1966), p. 15.
Niccolò di Segna (?), Young Male Saint, art market, whereabouts
unknown. Not seen by the author, but by Mario and
Dianne Dwyer Modestini and, according to them, iden-
tified by Everett Fahy as one panel of a polyptych, part of
which remains in Montalcino.
Paolo di Giovanni Fei, Madonna and Child with Two Angels, Saint
Francis, and Saint Louis of Toulouse (k-187), High Museum
of Art, Atlanta, GA (58.42). See fig. 16 of this paper.
Pietro Lorenzetti, Madonna and Child, Pieve dei SS. Stefano e
Degna, Castiglione d’Orcia (Siena). See Mostra di Opere
d’Arte Restaurate nelle Province di Siena e Grosseto. Genoa:
Sagep Editrice (1979), pp. 49–51.
Bernardo Daddi, The Nativity and the Annunciation to the Shepherds,
Museum of Fine Arts, Boston. See Kanter (1994),
pp. 54–7.
Florentine School (Giovanni Bonsi?), Saint Onufrius, Acton
Collection, Villa La Pietra, Florence.
Sano di Pietro, Madonna and Child, Acton Collection, Villa La
Pietra, Florence.
Workshop of Ugolino da Siena (Olena Master?), Madonna
and Child with St. Peter and St. John the Evangelist, Olena,
near Barberino in Chianti, San Pietro. See Stubblebine
(1979), Vol. 1, p. 91.
Photography Credits
Figs. 1–12, pp. 64, 67, 68, 69, 70, and 71. Birmingham Museum
of Art, Birmingham, AL, Samuel H. Kress Collection
(61.104).
Figs. 13 and 14, p. 72. Museum of Fine Arts, Boston, MA,
Charles Potter Kling Fund (39.536). Photograph © 2005
Museum of Fine Arts, Boston.
Fig. 15, p. 73. ©The Metropolitan Museum of Art, New York,
NY, all rights reserved.
Fig. 16, p. 74. High Museum of Art, Atlanta, GA. Gift of the
Samuel H. Kress Foundation (58.42).
1

Nativity
Sandro Botticelli (Alessandro di Mariano Filipepi), ca. 1475
Mixed media on plaster transferred to cradled canvas
63 1/2 × 54 in. (161.3 × 137.2 cm)
Columbia Museum of Art, Columbia, South Carolina
cma 1954.29 (k-1410)
Fig. 1. Nativity, during 1994–95 conservation.
Fig. 2. Nativity (fig. 1), after cleaning and restoration.

Botticelli’s Nativity
Charles R. Mack
I n 1943, Renaissance art historian, R. Langton Douglas concluded

his evaluation of this detached fresco of the Nativity (figs. 1 and 2), with
these evocative words: “This picture has all the charm, all the gracefulness
of Botticelli’s style at this period. It is the recorded vision of a painter,
a painter who was essentially a mystic, though not without a sensuous
appreciation of the beauties of the present world. This lovely pastoral is the
Christian counterpart of a spring-tide dream of Theocritus. We see the New
Life springing up amongst the ruins of the Old Order, whilst Angels sing the
Adeste Fideles.”1
Despite the poetic phrasings of Douglas, a true appreciation of the Nativity,
a part of the Samuel H. Kress Collection of the Columbia Museum of Art in
South Carolina, since 1954, has been complicated and hampered by a number
of problems involving the history of its ownership, proper attribution, original
location and function, and condition. One scholar apparently believed there to
be two separate works and wrote of both in the same book.2 Such confusion,
actually, is not surprising considering the complicated story of its provenance.
Provenance
A reconstruction of the history of the painting might begin with a tantaliz-
ingly brief statement, signed in Munich on October 28, 1927 by a certain
Franco Steffanoni of Bergamo.3 In this document, Steffanoni stated that he
had once transferred a painting called The Holy Nativity from wall to canvas.
Charles R. Mack 79
Steffanoni went on to say that he had made his a special catalogue to that exhibition.9 During this
identification from a photograph that had been time, the Nativity attracted scholarly attention and
sent to him. He added that the dimensions of the was discussed in several articles focusing on the
fresco he had transferred were 160 × 140 cm. These von Nemeš Collection.10 On June 17, 1913, the
dimensions correspond closely to those of the painting was among a number of works von
painting now in the Columbia Museum (161.3 × Nemeš put up for sale at the Manzi firm of Paris;
137.2 cm or 63 1/2 × 54 in.), and the photograph as item number 4 in the catalogue, it failed to find
sent to Steffanoni showed the Columbia Nativity. a buyer.11 Shortly thereafter, the Nativity passed
There is no doubt that the painting began its life through the hands of Parisian dealers Charles
as a mural and that it was later transferred to a Sedelmeyer and Broux Gilbert, but remained the
canvas support.4 Unfortunately, Steffanoni’s terse property of von Nemeš, who was, it would seem,
testimony provided no information as to when or attempting to dispose of the work with World
where he had performed his task. War I looming.12 The whereabouts of the Nativity
If Steffanoni’s 1927 account is to be credited, during World War I is unclear, but as that conflict
it would mean that he removed and transferred came to an end, the painting was included and
the painting at least forty-two years earlier, since illustrated in a multi-volume history of medieval
a reconstruction of the painting’s provenance and Renaissance painting written by Salomon
points to it having been in the collection of Sir Reinach.13 In 1921, von Nemeš acquired a castle,
William Neville Abdy (1844–1910) of the Elms, Schloss Tutzing, in Upper Bavaria and used it as
Newdigate, Dorking, England by 1885.5 In that a private gallery for his extensive collection. The
year, Abdy lent the work to the Louvre for an Nativity was apparently still in von Nemeš’s
exhibition to benefit the Franco-Prussian War possession throughout the 1920s and may have
orphans of Alsace-Lorraine. The exhibit was spent the decade at Tutzing. It was during that
called Exposition de Tableaux, Statues et Objets d’Art au period that Steffanoni was asked to document his
Profit de L’Oeuvre des Orphelins d’Alsace-Lorraine; Salle involvement; evidently his testimony was part of
des États au Louvre.6 Listed as number 312 on page an effort to authenticate the painting prior to an
89 of the exhibition catalogue, the Nativity bore anticipated sale. At von Nemeš’s death in 1930,
an unsurprising attribution to Botticelli’s pupil, however, the Nativity was still unsold and formed
Filippino Lippi. Eventually, this same painting part of his estate.
was among works from the Abdy estate sold in When von Nemeš died, he was in debt to sev-
London at Christie’s on May 5, 1911 (lot 86).7 By eral banks that had come under the control of the
then it had received its more customary associa- German government and which now proceeded to
tion with the name of Sandro Botticelli. seize and dispose of the collection. The German
At this point, there is a bit of chronological authorities made an unsuccessful effort to sell the
confusion since Museum file records indicate that Nativity at auction on February 29, 1932.14 This
the painting was exhibited at the Szépmũvészeti was a particularly difficult period in the German
Museum in Budapest from 1909 to 1911. How and and world economy, and it is not surprising that a
why it traveled from England to Hungary and buyer could not be found. In connection with the
back to London for the 1911 auction is unclear. 1932 auction (location undocumented but proba-
In any case, the person who acquired the painting bly in Munich), the painting was examined by a
at Christie’s was the well-known international art certain “Professor Graf, Chief Conservator of the
collector and dealer, Marczell von Nemeš.8 Pinakothek here” to evaluate its condition. It was
While in von Nemeš’s hands, the Nativity, along Graf who left the first condition report for the
with other works from his collection, was placed Nativity (see discussion, below).15
on public view from 1912 to 1913 at the Städtische Around 1935, German officials included the
Kunsthalle in Düsseldorf, Germany, and listed in Nativity among the unsold works from the von

Nemeš estate that were placed in storage at the to him. (For more on Vannoni, see Mario and
Kaiser Friedrich Museum in Berlin.16 This tempo- Dianne Dwyer Modestini’s paper in this volume.)
rary connection has led, at times, to incorrect The Nativity’s new owner supplied many works
provenance entries that indicate that the painting of art to Samuel H. Kress, and it was through
was part of the Museum’s collection.17 Duveen that the Nativity, now rather securely
Finally, in 1937, the Nativity was removed from bearing the name of Sandro Botticelli, came to
its Berlin deposit and taken to Munich, where, the United States and eventually into the Kress
once again, it was put on auction, this time suc- Collection in 1946.20 At that point, Kress was
cessfully, at the firm of Julius Boehler.18 The buyer enhancing his donation to the National Gallery
was the famous English art dealer Lord Duveen of of Art, and the Nativity was placed on public view
Millbank. Duveen, of course, was interested in a in Washington from 1946 until 1953. In 1954, after
resale, and it was at his behest that in October of decades of wandering, the Abdy–von Nemeš–
the same year the Nativity was fitted out with an Kress Nativity found its permanent home in South
elaborately carved and handsomely painted and Carolina when it joined twenty-six other paint-
gilded frame commissioned from an Italian crafts- ings in an initial gift from the Samuel H. Kress
man named Ferruccio Vannoni at a cost of 2,500 Foundation to the newly established Columbia
lire.19 Vannoni would seem to have produced Museum of Art, then housed in the renovated
his frame in Italy, working from dimensions sent Thomas Taylor Home (fig. 3).21 To better
Fig. 3. Nativity (fig. 1), in the Ferruccio Vannoni frame as installed in the Columbia Museum of Art (Taylor House location) in 1954.
Charles R. Mack 81
Fig. 4. Nativity (fig. 1), 1962 reinstallation without the Vannoni frame.

approximate the Nativity’s origin as a mural, Van- diaphanously clad angels bears greater similarity
noni’s elaborate frame was removed sometime in to the fresco in Columbia than it does to the
1961 or 1962, and the painting was set into a wall angelic celebrants floating and dancing above
of the gallery (fig. 4). Botticelli’s late (circa 1501) Mystic Nativity in Lon-
don with which the drawing has been associated,
Attribution but an absolute connection cannot be made. The
The fresco of the Nativity has been associated Uffizi drawing has been dated variously within the
with the names of both Sandro Botticelli and his earlier to middle phases of Botticelli’s career, i.e.,
pupil, Filippino Lippi. Sadly, neither of the biog- from the early 1470s to as late as 1490.
raphies of these two artists in Giorgio Vasari’s The composition and stylistic features of the
mid-sixteenth-century Lives of the Artists, makes Nativity in Columbia are most often connected with
mention of the fresco—an understandable omis- a frescoed lunette of the same subject (but with-
sion considering its relatively small size and its out the landscape setting and Florentine youths) in
simplicity of statement. When the Nativity first the church of Santa Maria Novella in Florence.24
entered the literature in 1885, it was ascribed to This fresco has been relocated within the church—
Filippino Lippi. Since then, however, scholars perhaps more than once—and originally may have
have united around the name of Sandro Botticelli, surmounted the famous Adoration of the Magi in
differing primarily in describing it as either auto- the Del Lama funerary chapel.25 This badly pre-
graph or a workshop production. Even those who served fresco reverses the basic composition of
give it to Botticelli have assigned portions of it the Columbia Nativity by placing the Madonna on
(e.g., the three hovering angels) to an assistant, the left and further differs from it by having the
perhaps even Filippino Lippi. Opinions in the young Saint John rush in from the left rear and
Kress Foundation files from Bernard Berenson having Joseph seated in an attitude of slumber.26
(1932), Lionello Venturi (1939), R. Langton Dated between 1475 and 1477, it has been seen as
Douglas (1943), Fern Rusk Shapley (1966), B.B. a stylistic relative of the Columbia Nativity.
Fredericksen and Federico Zeri (1972), and most Based in part on its perceived similarities to
recently from Everett Fahy (who saw the painting the Santa Maria Novella Nativity, the Columbia
after its latest restoration in 1994) have supported fresco has usually been assigned a somewhat ear-
a definite attribution to Botticelli, with Berenson lier date in Botticelli’s chronology, around 1473 to
revising his view in 1963 in favor of a more con- 1475. To support this dating, R. Langton Douglas
servative workshop association, a position taken pointed to a stylistic affinity with the manner
in 1931 by Raimond van Marle.22 of Fra Filippo Lippi.27 Douglas noted that,
Several authorities have associated the Nativity’s although Botticelli had received his first instruc-
style and composition with two works that have tion in Lippi’s shop, he had acquired a more
been attributed to Botticelli or his circle. One of sculpturesque approach through his later connec-
these is a pen and ink drawing, usually thought tion with the bottega of Andrea del Verrocchio.
to be a school work, of three flying angels in the When Fra Filippo’s son, the precocious Filippino,
Gabinetto dei Disegni of the Uffizi.23 There is a apprenticed with Botticelli in 1472, Botticelli was
general resemblance between this group and the motivated to revive the manner of his old master.
angels hovering above the stable in the Nativity, Douglas believed that the Nativity in Columbia is
although the apparent similarity may be mislead- one manifestation of the lyrical Lippi revival
ingly enhanced by the drawing’s semicircular within the evolving style of Sandro Botticelli.
shape that heightens its resemblance to the angels
in Columbia. Whether the drawing was made The Question of the Original Location
as a lunette or was later cut down to that form As noted earlier, the first recorded mention of
is uncertain. Admittedly the fluidity of these the Nativity placed it in the private collection of
Charles R. Mack 83
William Abdy in 1885. Some four decades later, painting. That intact setting provides a visual key
the Italian conservator Franco Steffanoni attested to understanding one context for the Nativity in
to having removed the fresco from what we can Columbia and how it might be better appreciated
assume was its original location and having trans- today. But even if this hypothesis were true, where
ferred it to a canvas support (making possible all were the church and the chapel in which the
its subsequent international travels and its even- Nativity served as an altarpiece?
tual arrival in South Carolina). But where had Given the clearly Florentine character of the
Steffanoni done his work and what, in the first Nativity, Florence would be the logical assumption,
place, had occasioned the transfer from wall to although no record of such an altarpiece has been
canvas, from a fixed to a mobile condition? The preserved. One possible clue, however, is offered
answers to these questions would not only satisfy by the suggested date for Steffanoni’s removal
simple curiosity but would assist in resolving of the fresco and for its appearance in the Abdy
problems of its purpose and attribution. Collection at some point prior to 1885. It was
A search through the old accounts and histo- just at that time that the city of Florence was
ries of Florentine art, including Giorgio Vasari’s initiating a drastic program of urban renewal that
Lives (looking under the various possible artists would obliterate almost its entire central core.28
to whom the painting might be attributed— In the campaign to modernize the city, much of
Botticelli, Filippino Lippi, Botticini, etc.) has not the medieval and Renaissance district around the
produced any record of this Nativity prior to 1885. old market square was razed to be replaced by the
The questions remain. For what purpose could neo-classical Piazza della Repubblica, the central
such a fresco with its comparatively small dimen- post office, and other structures of the late nine-
sions and with such a subject have been originally teenth century. Perhaps, the Columbia Museum’s
commissioned? The subject matter is a common Nativity was frescoed on the wall of one of the
one for an altarpiece, yet frescoes are not generally several churches sacrificed to that massive rebuild-
associated with that particular form. On the other ing campaign.29 Something, but certainly not
hand, altarpieces in fresco may have been more everything, is known of these churches’ architec-
common than is supposed, with many having tural character and furnishings. It is just possible
been destroyed (not being easily movable) during that the Nativity was salvaged from a church
modernization campaigns or, as in the case with sacrificed to this lamentable nineteenth-century
our Nativity, converted to a transportable and modernization of the heart of old Florence, first
salable state. One famous example of an altar- rescued and then sold to an English collector.
piece in fresco, albeit of uncommon type and Of course, there is no proof that the Nativity
with a totally different subject, is the celebrated had a Florentine origin at all. Both Botticelli and
Masaccio Trinity from the mid-1420s, above a Filippino Lippi, the two artists with whom the
memorial altar in the Florentine church of Santa fresco has been most commonly associated, also
Maria Novella. The Botticelli Nativity discussed worked in Rome. Botticelli was there from 1481 to
earlier, still in the same church but in a new loca- 1482, when he worked on the frescoes lining the
tion, is another possible example. Still another walls of the Sistine Chapel. The young Filippino
and more obvious example—one with an identical Lippi was in Rome from 1488 to 1493 while he
theme and offering a parallel to the Columbia was executing the frescoes in the splendid chapel
painting—can still be found in its original loca- of Cardinal Olivieri Carafa in the church of Santa
tion in the church of Santa Maria del Popolo in Maria Sopra Minerva. Either artist could have
Rome. There, between 1485 and 1489, Pinturicchio accepted a small side commission to paint a
painted a fresco of the Nativity above the altar of Nativity on the wall of some Roman church. On
the Cappella della Rovere and framed it in such a the other hand, there is no more evidence for such
way that it appears as if it were a normal panel an altarpiece in Rome than there is in Florence,

considerable wear of the type one might expect
to find in a work exposed to the elements.
It is more than likely that the Nativity, what-
ever its original site or function, received the
attention of informal cleanings and even minor
restorations long before Steffanoni entered the
scene. His interventions were, however, radical in
nature. Thankfully, his approach to the problem
of mural detachment appears to have been com-
petent. After it had been transferred to heavy
canvas (primed on the reverse and reinforced with
a lightweight cradle) and had passed from the
Abdy to the von Nemeš Collection, the Nativity
apparently underwent an energetic restoration, by
whom we do not know.
When Professor Graf, the chief conservator
of the Munich Alte Pinakothek, examined the
painting in 1932 he presented a brief but contra-
dictory report. 32 He stated that the Nativity “has
Fig. 5. Nativity (fig. 1), showing the condition of the fresco been very well preserved,” but that “the restora-
following its transfer to canvas and prior to the first series tions which have been made by Von Nemeš are
of restorations. to be regretted.” Graf went on to note that while
a “moderate and orderly restoration” might be
undertaken, his conclusion was that “after close
and stylistically the Nativity would seem too early scrutiny and experienced consideration, no further
to support a Roman hypothesis for either artist. restoration should be attempted.” If anything were
If an original location and function as an altar- to be done, he recommended nothing more than
piece within a church cannot be substantiated, “touching up the white places with light coloring,
what of some external site? The street corners of in order to make it more attractive to the eye of
Florence, even today, abound with outdoor taber- the spectator.” If this were done, Graf suggested
nacles.30 Some 1,300 are known to have existed. “photographing the picture in its present state, in
Over the years many of these have disappeared for order to show to the eventual buyer the exact state
a multitude of reasons from floods to street repair of the picture, the parts that have been repainted,
to urban renewal. Certainly a number vanished and those which have been restored.”
with the demolition of the old market quarter Until recently, conservation records for
just mentioned. Most of these street-side shrines the Nativity have been meager. In addition to
featured images in fresco. While most of these, Steffanoni’s treatment of the painting after its
judging by the survivors, are considerably smaller transfer to canvas, the more aggressive treatment
in size and less compositionally complex than probably done for von Nemeš, and the possible
the Columbia Nativity, some larger examples sur- work performed under Graf ’s direction, an
vive, and this suggests another possible origin for undated Columbia Museum of Art condition
the Nativity. In support of this “open air” hypoth- report mentions, without elaboration, minor
esis is a file photograph of the fresco; although restorations in 1947 and in 1954, that is subse-
undated, it must have been taken at some point quent to the Nativity’s acquisition by Samuel H.
before the Nativity entered the Kress Collection Kress and then in preparation for sending it
(fig. 5).31 In this photograph the surface shows to Columbia.
Charles R. Mack 85
Fig. 6. Nativity (fig. 1), as presently installed in the new Columbia Museum of Art.

On June 14, 1993, the Columbia Museum of are original; therefore the iconographical signifi-
Art’s catalogue sheet for the painting rated its cance is valid. The figure of St. Joseph, the
condition as good to fair, adding the following bundle in the foreground and the Madonna’s
comments: head and hands are quite well preserved. Her
dress, painted with good quality lapis blue,
chipping of paint has occurred along bottom has lots of restorations but on the whole is
and right side. There are many cracks where in fair state. For the flesh tones, the paint has
chips were lost. Also, there are some areas where been applied as a liquid enamel over which thin
the top layer of paint is missing. It looks as modeling glazes have been floated. It exhibits
though work has been done to keep further a fine craquelure pattern which indicates that
deterioration from occurring. there is a binder, possibly a tempera grassa. The
Child is in good state and the mordant gilding
In March 1994, Kress Foundation conservator is original. Other areas of mordant gilding are
reasonably intact, especially the little curlicues
Dianne Dwyer Modestini visited the Columbia which rain down on the Child from the angelic
Museum of Art to examine the overall condition trio. There has been some reinforcement with
of its collection and to recommend a program shell gold.
of regular maintenance and restoration. Her
report reviewed the condition of the Nativity and Modestini concluded her preliminary obser-
suggested that it be removed to her conservation vations by suggesting that in a new round of
studio in New York for appropriate attention conservation procedures after cleaning, the old
to begin that September. In this initial report, inpainted restorations be removed, and that those
Dianne Dwyer Modestini theorized that the areas suffering from the most damage be restored
major work of restoration on the fresco had been with neutral tones; these areas would include the
done at the beginning of the twentieth century, an three hovering angels and the foreground. Follow-
observation that agreed with Graf ’s notes in 1932. ing her advice, and with the support of the
She went on to observe that: Kress Foundation, the fresco was transported to
New York where cleaning and restoration work
The condition is uneven with some passages well
preserved and others in ruinous state. The three
was carried out by Mario and Dianne Dwyer
angels above are largely reconstructed from exist- Modestini. The Nativity was missed in Columbia
ing fragments which can be glimpsed here and but, as its absence coincided with preparations for
there under crude repaint. The architecture of reinstalling the Museum’s collection in its hand-
the stable is relatively well preserved with only some new quarters on Columbia’s Main Street,
the beam immediately behind the angels com- this seemed the perfect occasion. Work on the
pletely repainted. These elements in the upper Nativity was undertaken in the autumn of 1994,
part of the painting could have been painted and the painting was back in Columbia, fitted out
in buon fresco. The sky is repainted in full. Small
with a new tabernacle-like frame, ready to assume
fragments of the original blue can be located
in a few places. The original blue is a thin wash its pivotal position in the Museum’s collection of
of what appears to be lapis. The donkey is well Renaissance and Baroque art when the building
preserved; the mouth of the ox, the neck, and the was inaugurated in the summer of 1998 (fig. 6).
part of the head in shadow have been repainted.
The distant landscape and the grove of trees Treatment
on the left are well preserved. The two youths
on the left are worn, especially the heads and
The treatment at the New York University Con-
hands, and the costumes have been much, but servation Center was summarized in a report sent
not completely, repainted. The foreground land- to the Columbia Museum on January 5, 1995. In
scape is largely, but not completely, restoration. it, Dianne Dwyer Modestini explained the current
Some parts of the bushes, including the fruit, state (see fig. 1) of the fresco and outlined the
Charles R. Mack 87
steps and procedures taken to stabilize the work Modestini also explained that, “The painting was
and optimize its appearance, first noting that: transferred from plaster, lined to linen, which was
then mounted on some sort of cradled board.”
Of course, the painting is not in good condition
… important parts are well preserved: notably She determined that:
the head and hands of the Madonna, the figure The technique is mixed media on plaster … not
of St. Joseph, the Child, and, somewhat less, the entirely buon fresco. Many passages, especially
young St. John. The two figures on the left are the flesh tones, exhibit a fine craquelure pattern
badly damaged. Other details are well preserved, associated with an aqueous binder, and are
while the foreground and sky and the three hov- minutely executed like a tempera painting. The
ering angels are in ruinous state. Of the angels, cracks and deformations of the original plaster
only the head of the angel on the right is in support are evident throughout and the pattern
good condition. The landscape backgrounds, of the cradle [of Steffanoni] can be seen in
while full of scattered losses and abrasions, are, raking light. Structurally, the painting is stable.
nonetheless, original, that is, not completely
repainted, whereas the grove of trees on the Following a discussion of the particular proce-
left is largely reconstructed. The plants along dures used in the restoration and of the various
the bottom are mostly reconstructed, with large solvents and chemicals used in the cleaning and
areas of loss; however, there is some original. retouching processes, she concluded by saying
Those who saw the painting during its period that her:
of convalescence in New York affirmed the pri- restoration generally treated the painting as an
mary authorship of Botticelli but suggested that easel painting, rather than a fresco for a variety
there was a strong influence present from his of reasons: the poor state, the fact that it was
apprentice, Filippino Lippi. In addition, “…we not painted as a true fresco to begin with, and
have noted,” Modestini wrote, “that there is a the treatment that it had undergone in the past
variation in quality, the principal parts being which has been selectively removed.
superbly drawn and painted, while other elements,
such as the stable, the animals, seem to be by an Results and Discoveries
inferior hand, a studio assistant.” In all probability, What has been the effect of the recent cleaning
this lesser hand was not that of Filippino Lippi and the conservation measures undertaken under
whose abilities matched those of his teacher. the auspices of the Kress Foundation? The most
Fig. 7. Nativity (fig. 1), detail of angels prior to 1994–95 Fig. 8. Nativity (fig. 1), detail of angels following 1994–95
conservation procedures. conservation procedures.

obvious result has been to stabilize its condition
and to enhance its appearance (figs. 7 and 8). In
addition, examination during treatment allowed
for a more secure confirmation of Botticelli’s
primary role in its execution and an opportunity
to distinguish between what is autograph and
where Botticelli’s contemporaries or later restorers
have intervened.
The technical and stylistic understanding that
the recent conservation measures brought to the
Nativity have enabled us to re-evaluate Botticelli’s
working procedures. In 1978 Ronald Lightbown
observed that:
the division between Botticelli’s autograph works
and the paintings from his workshop and circle
is a fairly sharp one. Only in a single major
panel painting (the Trinity Altarpiece in London’s
Courtauld Institute), do we find important parts
executed by assistants … Even in the Sistine
frescoes, where we might expect considerable
traces of help from secondary hands, none has
been convincingly demonstrated.33
Lightbown also noted that “conversely, there

are a very few workshop pictures in which Botti-
celli finished important parts or added finishing
Fig. 9. Nativity (fig. 1), detail showing “golden flames,” follow-
touches …” The close observation recently ing 1994–95 conservation procedures.
afforded the Nativity might necessitate a recon-
sideration of Lightbown’s conclusions. At least
in the case of the Nativity in Columbia, it has
been shown that an essentially autograph fresco the drapery of Mercury in the famous Primavera
by Botticelli, even one of small dimensions, in the Uffizi. Most telling is the use of this motif
could involve the participation of one or more in several of the drawings the master executed
assistants in its execution. to illustrate Dante’s Divine Comedy. They are
Another result has been to reveal and clarify major pictorial elements in Botticelli’s drawings
painted elements that allow for a more intelligent for Inferno Cantos xxvi–xxviii and especially
reading of the various visual meanings within for Paradiso Cantos vi–viii and xxiii–xxvi. The lit-
the seemingly straightforward presentation. One erary context makes the meaning clear in these
such element, previously only barely visible, is the connections: they represent “spirits” or “souls.”
shower of golden flames (fig. 9) that fall upon Thus, in the Columbia Nativity, the newly clari-
the Christ Child from the trinity of angels hover- fied golden flames falling upon the Bambino might
ing above. Such flaming bundles (resembling the be interpreted as a Heavenly descent of the Holy
badges worn by today’s Carabinieri, the Italian Spirit as God is made man,34 a reminder of the
national police) are to be found in other Botticelli Virgin Birth.
compositions: they appear on the shoulder of For the visitor to the Columbia Museum of
Mary in the Madonna of the Book in the Museo Art and for those who will be using its antici-
Poldi Pezzoli in Milan, and they sprinkle across pated catalogue of Renaissance and Baroque art,
Charles R. Mack 89
this conservation effort will have a decided Notes
impact. Not only has the visual integrity of the 1. Letter from R. Langton Douglas, dated March 5, 1943,
in the painting’s files at the Columbia Museum of Art.
Nativity fresco been strengthened, but the infor- 2. Gabriele Mandel, The Complete Paintings of Botticelli (New
mation we have learned will be used to clarify York, Harry N. Abrams, 1967) where it is described on
the presentation. It will now be possible, thanks page 91 as a fresco copy of the Nativity fresco in the Flor-
entine church of Santa Maria Novella, measuring 150 ×
to the careful art historical and scientific reading 250 cm, which “went from the Boehler Gallery, Munich,
of the painting afforded by the Kress-sponsored to the Kress Collection, New York, which transferred it
restoration, to explain the areas of varying quality to the Columbia Museum of Art” and on page 109 (cat.
no. 149), it is illustrated with a line drawing reproduced
within the composition—why, for instance, the from Reinach. In its second appearance in Mandel’s book,
Virgin’s face can be so lovely while that of the ox the painting is described as “The Nativity, formerly Bud-
is so poorly executed. Such aspects as how little apest, Von Nemeš Collection” and listed as a workshop
production, executed in tempera on a wood support. The
of the angels’ original figures do in fact remain same entry also associates it with the Abdy Collection and
can also be pointed out, allowing for a more says that “its present whereabouts are unknown.”
discriminating appraisal of the true qualities 3. The text of this statement reads: “Io sottoscritto Francesco
Steffanoni di Bergamo (Italia), trasponitore di dipinti, dichiaro: di
of Botticelli’s manner and his contributions to aver trasportato il dipinto ramp.te il Sacro Presepio, di cui è oggetto
the history of Renaissance art. The public “con- codesta stessa fotografia, dal muro su tele tel delle dimensioni di
fession” that can be now attached to the Nativity m. . x. .. In fede Franco Steffanoni Munchen,  Ottobre
m.c.m. XX VII.” [Translation: “I, the undersigned Francesco
and to other works in the Columbia Museum Steffanoni of Bergamo (Italy), a specialist in the transfer
of Art that have benefited from Kress-sponsored of paintings, declare: to have transferred the painting rep-
conservation will aid visitors in applying these resenting the Holy Nativity, which is the object in this
photograph, from the wall to canvas whose dimensions
same lessons in appreciation when viewing other are 1.40 × 1.60 meters. In fede Franco Steffanoni Munich,
Old Master paintings in the Museum’s collection October 28, 1927.”] A copy of this document is in the
and elsewhere. painting’s files at the Columbia Museum of Art.
4. The recent restoration of the Nativity concluded that it
Despite what has been learned of the true con- had been transferred to a linen canvas mounted on a
dition of the Nativity in Columbia, the perceptive cradled solid support. Letter in the files of the Columbia
appraisal of R. Langton Douglas six decades Museum of Art from Dianne Dwyer Modestini dated
January 5, 1995.
ago still amplifies our appreciation of Botticelli’s 5. Alessandro Contini-Bonacossi, Art of the Renaissance from
gentle scene. His eloquence has only been the Samuel H. Kress Collection. Columbia, SC: Columbia
strengthened by a better understanding of the Museum of Art, 1962, p. 68; Ronald Lightbown, Sandro
Botticelli: Complete Catalogue. Berkeley and Los Angeles, CA:
painting’s complex history and by the thorough University of California Press, 1978, ii, p. 33.
attention given to its condition. Restored and 6. Contini-Bonacossi 1962 (cited in note 5), p. 68.
handsomely installed as a visual focal point in 7. Lightbown 1978 (cited in note 5), ii, p. 33.
8. Von Nemeš’s first name also appears in the literature
the Renaissance and Baroque galleries of the new spelled as “Marcel” or “Marczell.” His portrait, painted
Columbia Museum of Art, Sandro Botticelli’s in 1928–29 by Oskar Kokoschka, hangs in the Wolfgang-
lovely Nativity continues to captivate. Gurlitt Museum in Linz, Austria.
9. Contini-Bonacossi 1962 (cited in note 5), p. 68.
10. Ibid. Six separate publications dealing with the von Nemeš
Charles R. Mack is William Joseph Todd Distinguished Collection during this period are cited in the files of the
Columbia Museum as having included the Nativity. They
Professor Emeritus of the Italian Renaissance at the are: Gabriel von Terey, Katalog der Sammlung des Kgl. Rates
University of South Carolina where he taught art history Marczell von Nemeš, Budapest (Düsseldorf: Städtische Kunst-
from  to . A specialist in Quattrocento art halle, 1912), No. 3; August L. Mayer, “Die Sammlung
Marczell von Nemeš in Budapest,” Westermann’s Monatshefte
and architecture with a Ph.D. from the University of 133 (December 1912), pp. 495 and 540 (illus.); Georg
North Carolina, his publications include Looking at Biermann, “Die Sammlung Marczell von Nemeš,” Der
the Renaissance: Essays toward a Contextual Cicerone (1912), p. 374 (illus., fig. 5); Gabriel Mourey,
“La Collection Marczell von Nemeš,” Les Arts (June 1913),
Appreciation () and Pienza: The Creation pp. 2–3; François de Miomandre, “Les Idées d’un amateur
of a Renaissance City (). d’art,” L’Art et les Artistes (March 1913), p. 251 (illus.); and

Emile Dacier, “La Collection Marczell de Nemeš,” Revue Foundation form the core of the Columbia Museum’s
de l’Art Ancien et Moderne (June 1913), p. 458. holdings in Renaissance and Baroque art, now more
11. Lightbown 1978 (cited in note 5), ii, p. 33. than 200 in number.
12. Contini-Bonacossi 1962 (cited in note 5), p. 68 and 22. Copies of these opinions are in the Kress Foundation files
Lightbown 1978 (cited in note 5), ii, p. 33. The former lists and at the Columbia Museum of Art. Most of them are
the Sedelmeyer Gallery catalogue as Charles Sedelmeyer, given in summary form in Contini-Bonacossi 1962 (cited
Catalogue of the Twelfth Series of  Paintings by Old Masters in note 5), p. 68. Berenson’s initial attribution was based on
(Paris, Sedelmeyer Gallery, 1913), p. 62, No. 39 (illus.). a photograph of the Nativity on which he wrote “Sandro
13. Salomon Reinach, Répertoire de Peintures du Moyen Age et de la Botticelli, about 1475. B. Berenson.” That of Fredericksen
Renaissance (Paris, 1918), iv, p. 76 (illus.). Thirteen years and Zeri was contained in their Census of Pre-Nineteenth-
later Raimond von Marle included the Nativity (as a Century Italian Paintings in North American Public Collections
Botticelli school piece) in his authoritative The Development (Cambridge, Harvard University Press, 1972); that of
of the Italian School of Painting (The Hague, Nijhoff, 1931), xii, Shapley in her Paintings from the Samuel H. Kress Collection,
p. 272. Vol. : Italian Schools, XIII–XV Century (London, Phaidon,
14. This information is contained in a translated document 1966), p. 117, fig. 318. The most recent observations of
in the Kress Foundation files with a copy at the Columbia Everett Fahy are summarized in the restoration report
Museum of Art. submitted to the Columbia Museum of Art by Dianne
15. Ibid. Dwyer Modestini on January 5, 1995.
16. This is attested to in a Columbia Museum file copy of a 23. See the discussion of this drawing in Lightbown 1978
Western Union Telegram from Paris to the Duveen Com- (cited in note 5), ii, pp. 161–2. Another drawing in the
pany in New York, dated Tuesday, August 8, 1939, that Uffizi collection, not cited in the literature but possibly
reads: “Botticelli Nativity came from Nemeš Collection. related, is a badly damaged rendition in pen and ink of the
When Nemeš died he owed money banks which were Adoration of the Child. See the discussion in Lightbown 1978,
taken over German Government who gave pictures to p. 163. This drawing, over which there is considerable
Kaiser Friedrich Museum who kept them stored several debate as to authorship and dating, depicts an animated
years until sold auction Munich.” Christ Child between a dozing Joseph on the left and an
17. This is the impression given in Lightbown 1978 (cited in adoring Mary on the right. The attitude of the Virgin
note 5), ii, p. 33. resembles that in Columbia, and there is some similarity
18. This is documented in the sales catalogue Kunstwerke aus between the bambini as well.
dem Besitz der Staatlichen Museen (Munich, Julius Boehler, 24. See the discussion in Lightbown 1978 (cited in note 5),
1–2 June 1937), pp. 104–5, No. 654, plate 48. The title ii, pp. 32–3. His treatment of the Columbia Nativity is
of this auction catalogue may have contributed to the appended to this entry. Also see Caterina Caneva, Botticelli:
misunderstanding of the Nativity’s status while in Berlin. Catalogo Completo dei Dipinti (Florence, Cantini, 1990), p. 46.
19. See the copy of a letter in the Kress Foundation files and 25. This was the opinion of Carlo Gamba, Botticelli (Milan,
those of the Columbia Museum of Art, dated October 29, Ulrico Hoepli, 1936), pp. 115–16.
1937, sent from New York and requesting information 26. The general character of Botticelli’s composition in
regarding Vannoni’s prices. Vannoni’s frame was still on the Nativity fresco, as well as the representation of such
the painting when it arrived in Columbia in 1954 for its elements as the wood-beamed stall, might be compared
initial installation in the new Kress wing of the Columbia with a now ruined fresco by Paolo Uccello in the cloister
Museum. Notations on the reverse of file photographs arcade of the former hospital of San Martino della Scala
in the Columbia Museum indicate that the removal of in Florence, dated around 1446. Botticelli would have
Vannoni’s frame and the reinstallation of the Nativity took known Uccello’s fresco since he is documented as having
place between late 1961 and October 1962. The Vannoni worked at the same institution in the spring of 1481, when
frame was subsequently deaccessioned and its present he executed a fresco of the Annunciation for the tomb of
whereabouts are unknown. A photograph published in the hospital’s founder Cione Pollini; Botticelli could have
the State and Columbia Record of December 19, 1965 (copy been familiar with it much earlier.
in Museum files) shows the Nativity in its new frame 27. R. Langton Douglas, letter dated March 5, 1943 in the files
and setting. The Nativity remained so displayed until the of the Columbia Museum of Art.
Columbia Museum moved into its new quarters in 1998. 28. On this project, see Giovanni Fanelli, Firenze: Architettura
20. On this see R. Langton Douglas, “Recent additions to e Città (Florence, Vallecchi, 1973), i, pp. 447–52.
the Kress Collection,” The Burlington Magazine Vol. 88 29. Ibid, ii, pp. 10 and 68. The churches then swept away in
(April 1946), p. 82, plate ivb. the risanamento of old Florence included S. Andrea, S. Pier
21. The former Taylor residence, with its gallery wing Buonconsiglio, S. Tommaso, S. Maria in Campidoglio,
additions, was home to the Columbia Museum of Art S. Leone, S. Miniato fra le Torri, S. Salvatore, S. Ruffillo,
from 1950 until 1998 when the collection was transferred S. Maria degli Vigni, S. Donato dei Vecchietti, and the
to its present location. The initial Kress donation was Oratorio di S. Maria della Tromba. On these destroyed
augmented in 1964 by the gift of forty-four works of art, churches and their furnishings see Walter and Elizabeth
consisting of seventeen paintings, four sculptures, ten Paatz, Die Kirchen von Florenz, 6 vols. (Frankfurt, Kloster-
objects of decorative art, nine textiles, and four pieces mann, 1940–1954). Plans published in vol. 6 show the sites
of furniture. These seventy-one works from the Kress of dozens of churches throughout the city that have either
Charles R. Mack 91
vanished or been converted to secular use. The Columbia
Nativity could have come from one of these or from a
countryside parish in the outskirts of Florence.
30. On these outdoor devotionals, see “i tabernacoli” in Aspetti
Minori di Firenze, ed. Piero Bargellini (Florence, Azienda
Autonoma di Turismo, n.d.), pp. 29–41.
31. This photograph may have been taken in response to
the suggestion made in 1932 by Professor Graf; see the
discussion below.
32. Report copy in Registrar’s files, Columbia Museum of Art.
This document, dated February 29, 1932 is a translation of
Graf ’s report; the translator is not identified.
33. Lightbown 1978 (cited in note 5), i, p. 155.
34. On Botticelli’s use of this motif see Horst Bredekamp,
Sandro Botticelli: La Primavera (Frankfurt, Fischer, 1988),
pp. 40–46. A similar motif can be seen on the Virgin’s
shoulder in the Madonna del Libro in Milan and on
Mercury’s cloak in Botticelli’s Primavera in the Uffizi.
Photography Credits
Figs. 1, 2, 6, 7, 8, and 9, pp. 78, 86, 88, and 89. Columbia
Museum of Art, Columbia, SC, Samuel H. Kress
Collection (cma 1954.29).
Fig. 3, p. 81. Columbia Museum of Art, Columbia, SC,
Samuel H. Kress Collection (cma 1954.29). Museum file
photograph dated fall 1961.
Samuel H. Kress Collection (cma 1954.29). Museum file
photograph dated October 1, 1962.
Samuel H. Kress Collection (cma 1954.29). Undated
Museum file photograph.
1

Venus and Cupid, ca. 1500
Girolamo di Benvenuto
Egg tempera and oil on wood panel
20 1/2 × 20 in. (51.1 × 50.8 cm)
Denver Art Museum, Denver, Colorado
1961.172 (k-222)
Fig. 1. Venus and Cupid, before cleaning and restoration.
Fig. 2. Venus and Cupid (fig. 1), after cleaning and restoration.

The Re-use of a Desco da Parto
Mika Okawa and Dianne Dwyer Modestini
I n 1995 the Girolamo di Benvenuto Venus and Cupid (figs. 1, 2, and 3),
from the Samuel H. Kress Collection of the Denver Art Museum, came
to the Conservation Center of the Institute of Fine Arts, New York
University for treatment. A Sienese painter, Girolamo (1470–1524) was
the son of Benvenuto di Giovanni with whom he collaborated and
whose style he imitated.1 The painting in question is of particular interest as
it is a desco da parto or birth tray, a salver presented to a mother on the birth of
her child; in this case the form is a sixteen-sided polygonal panel painted in
tempera and oil. It had sustained severe damages in the form of deep scratches,
abrasions, losses and stains, probably from its use as a piece of household fur-
niture, but also from deliberate vandalism of the nudes, a common occurrence.
The perfectly legible coat of arms on the bottom has not been identified.
Two interesting features of this object emerged in the course of examination
and restoration. The first has to do with its provenance and critical history.
It was acquired by the Florentine dealer Count Alessandro Contini-Bonacossi
from a Conte della Gherardesca, a member of an old and noble Florentine
family. As Fern Rusk Shapley records:
When it was in the Gherardesca Collection, k-222 was, according to Schubring
… decorated on the back with a standing Cupid in a circular simulated frame.
In a letter of Feb. 19, 1949, R. Mather writes of having seen back and front
as two separate panels before k-222 entered the Kress Collection. An X-ray
made by A. Burroughs soon after it entered the Kress Collections shows a circle
Mika Okawa & Dianne Dwyer Modestini 95

corresponding to the inner circle of the tondo was cradled in Stephen Pichetto’s studio according
frame but no further evidence of the tondo to usual practice, presumably by the carpenter
panel. Discussing the X-rays, Burroughs indi- Angelo Fatta, and fitted into a shadow box with
cated that there was some kind of design at this a Masonite backing. When new radiographs
time on the reverse of the panel: “In spite of
the design on the reverse of this panel,” he says,
were made in the 1950s, the heavy cradle made it
“the paint is well recorded in good condition.” impossible to determine if there was any design
Whatever the design, it seems to have disap- on the reverse. What is more, the very presence
peared when the panel was treated for cradling of a cradle would logically lead to the conclusion
in 1933. Not even the circle shows in an X-ray that the desco lacked its original back. Evidently
made in the 1950s. The present whereabouts of no one saw the painting out of its shadow box
the cupid tondo is unknown.2 and assumed that Zeri was right, that the double-
sided desco had been split and sold as two sep-
In a footnote in the third volume of the Kress
arate paintings before 1932. As late as 1997 in a
Italian paintings catalogues,3 Shapley had iden-
definitive study, I Deschi da Parto e La Pittura del
tified the missing back of the desco:
primo Rinascimento Toscano,4 the presumed verso,
The reverse of this salver has been given to the the Standing Cupid in Castel Sant’Angelo, is illus-
Castel Sant’Angelo, Rome (Menotti Bequest), trated together with the recto in Denver.
as noted by Federico Zeri (in The Burlington When the desco was freed from the shadow box
Magazine, Vol. cix, 1967, p. 477) and listed in with its Masonite backing we were surprised to
the posthumous edition of B. Berenson (Italian
see that it had retained its original back, a simple
Pictures … Central and North Italian Schools, vol. i,
1968, p. 187). design of a large black circle on a white ground,
consistent with Alan Burroughs’s notes. On fur-
Samuel H. Kress acquired the desco in 1932. As ther examination we were amazed to discover that
stated, Alan Burroughs’s X-radiograph revealed a a cradle was concealed within the desco. The panel
“circle” and “some kind of design on the reverse had been sliced in half and the inside hollowed
of the panel.” After the painting was purchased, it out to accommodate a heavy cradle; the two
halves were then rejoined with screws. There was
no doubt that they had once been a single object.
This bizarre undertaking was accomplished with
great skill; a thin, barely perceptible cut separated
the front from the reverse. While we have not had
an opportunity to examine the Castel Sant’Angelo
Standing Cupid that bears devices associated with
the della Gherardesca family, the two deschi cer-
tainly were never part of the same object and may
have entirely different provenances.
Parts of Venus and Cupid were reworked at an
early date: the mordant gilding, the coat of arms,
and the delicately painted flowers in the fore-
ground appear to be by another hand, in a differ-
ent medium (possibly aqueous or an emulsion),
and were all added later. It is impossible to say
with certainty how many years intervened between
the creation of the original image and its later
embellishments. They are certainly antique but
Fig. 3. Venus and Cupid (fig. 1), verso. were painted when the original, which appears to

be in an oil medium, was thoroughly dry. The Notes
later additions are brittle with a different craque- 1. Fern Rusk Shapley, Paintings from the Samuel H. Kress
Collection, Vol. : Italian Schools, XIII–XV Century (London,
lure pattern, and they sometimes cover losses. The Phaidon, 1966), p. 161 and the entry on Benvenuto di
entire coat of arms belongs to this reworking and Giovanni (di Meo del Guasta) by Cynthia Coté, Grove
gives credence to Mather’s observation in 1949 Dictionary of Art (New York, Grove Press, 1996), Vol. 3,
pp. 750–51.
that it was not original. 2. Shapley 1966 (cited in note 1), p. 162.
Noting that the verso of the desco was original 3. Fern Rusk Shapley, Paintings from the Samuel H. Kress
and the recto bore later embellishments including Collection, Vol. : Italian Schools, XVI–XVIII Century. London:
Phaidon, 1973, p. 387.
the alteration of the coat of arms, it might be 4. Cecilia De Carli, Archivi di Arte Antica. Turin: Umberto
hypothesized that the panel had been used as a Allemandi & Co., 1997, no. 60, pp. 200–201.
salver for another birth, perhaps fifty to one hun-
dred years later. As already mentioned, the coat of Photography Credits
arms has not been identified. Perhaps the simple Figs. 1, 2, and 3, pp. 94 and 96. Denver Art Museum,
Denver, CO, Samuel H. Kress Collection (1961.172).
decoration of the reverse, a black circle on a white 1
field, has some significance. There is no doubt
that the confusion, first introduced by Schubring
in 1923, between the Castel Sant’Angelo Cupid
and our desco is finally resolved. Venus and Cupid,
though tampered with, is intact; the coat of arms,
while not original, is antique. With this informa-
tion in the future it may be possible to shed light
on this intriguing instance of mistaken identity
and re-use of a Renaissance ceremonial object in
a later period.
Because of the history of the object, during
the restoration we left the reworked passages
in their entirety and tried to be selective in our
retouching, particularly of the numerous gouges
and depressions that characterize the Renaissance
objects made for household use, such as cassoni,
removing the most disturbing damages but not
trying to make the image perfect.
Dianne Dwyer Modestini is a paintings conservator,

Fine Arts, New York University.
Mika Okawa is a paintings conservator at the Art Conser-

vation Laboratory in Tokyo, Japan. She received an M.A.
in Conservation Science from Tokyo National University
of Fine Arts and Music, where her dissertation subject
was solidification problems in modern paint. She attended
the Conservation Center of the Institute of Fine Arts,
New York University as a special student.
Mika Okawa & Dianne Dwyer Modestini 97

The Triumphs of Love, Chastity, and Death
Follower of Andrea Mantegna, late 15th century
Egg tempera and oil on cradled wood panel
20 × 21 1/4 in. (50.8 × 54 cm); 20 3/4 × 21 3/8 in. (52.1 × 54 cm);
20 5/8 × 21 3/8 in. (52.4 × 54.3 cm)
1961.169.1 (k-12, 13, 15)
Fig. 1. The Triumphs of Love, Chastity, and Death, before cleaning and restoration.
Fig. 2. The Triumphs of Love, Chastity, and Death (fig. 1), after cleaning and restoration (framed together).

The Triumphs of Petrarch: An Analysis of
a Renaissance Decorative Cycle
Wendy Partridge
A series of six paintings from the Kress Collection in the Denver Art
Museum was received at the Conservation Center of the Institute of
Fine Arts, New York University, for treatment in 1995 (figs. 1, 2, 3, and 4).
The series is based on the Triumphs, a long narrative poem by Petrarch
begun in 1340 and still unfinished at his death in 1374, describing
the successive triumphs of Love, Chastity, Death, Fame, Time, and Divinity.
The Triumphs were one of the most popular secular subjects in the Ren-
aissance, illustrated in countless manuscript illuminations, woodcuts, engrav-
ings, tapestries, and paintings. There is surprisingly little concrete imagery in
the poem, and Petrarch described only one chariot, belonging to Love and
drawn by four white horses. Depictions of the Triumphs, however, almost uni-
versally put all six allegorical figures on chariots, each led by a set of different
animals. Chastity, in the Denver panel, is drawn by unicorns, Death by water
buffalo, Fame by elephants, Time by deer, and angels lead Divinity. The earliest
manuscript illuminations of this type are from Florence, date to 1442, and are
possibly by Apollonio di Giovanni.1 The first panel paintings of the subject
also seem to be from Florence and date to about the same period.
The Denver paintings were sent to the Conservation Center because a thick
layer of varnish had yellowed, and there were numerous awkward, discolored
retouchings. In the course of treatment, areas of original composition that
had been overfilled and overpainted were uncovered. Once the paintings were
cleaned and retouchings removed, the cycle revealed a clarity, brilliance of
Wendy Partridge 99
color, and richness of detail. The extensive use of he noted their compositional similarity to the
gold and ultramarine (largely overpainted in the Graz ivory reliefs and identified the Graz ivory
skies by a dark Prussian blue) suggested a luxury chests as produced for the 1477 wedding of Paola
commission of considerable expense. After con- Gonzaga and Leonard von Goerz.11
servation, the quality and sumptuousness of the Representations of the Trionfi, in fact, had
paintings was far more evident. become a popular subject for domestic Florentine
Fern Rusk Shapley attributed the paintings furniture decoration by about 1445. Other exam-
to a follower of Andrea Mantegna.2 Their prov- ples of Trionfi panels include an anonymous set
enance can be traced back only to the late 1870s from the Palazzo Davanzati12 and a Pesellino
at the Castello of Colloredo near Udine.3 After cycle at the Isabella Stewart Gardner Museum
their dispersal to various collectors and dealers dated to about 1445.13 The Trionfi were one of
in the late nineteenth century, the panels were the popular subjects of Pesellino’s shop, and
reunited in the Kress Collection in 1927–28.4 The two other versions are extant. A fragmentary fron-
original function of the cycle, its commission and tal with only Love and Chastity is now in the
artist, and the relationship to Andrea Mantegna National Gallery of Scotland and was probably
and the Gonzaga court are puzzling problems. done by a painter who trained in Apollonio di
The paintings were first published by Joseph Giovanni’s workshop.14 Panels with Love, Chas-
Wastler in 1880. Wastler compared the composi- tity, Death, and Fame are in the Siena Pinacoteca;
tions to remarkably similar depictions found on Schubring tentatively attributed them to Pier
ivory reliefs in the cathedral at Graz (figs. 5, 6, Francesco Fiorentino, while Giovanni Carandente
and 7). By analyzing heraldic devices, he was able gave them to Marco del Buono.15 There are also
to link the Graz pieces to the Gonzaga family Jacopo Sellaio’s large triumphs of Love, Chas-
sometime after 1432.5 He also suggested a possible tity, Time, and Divinity of about 1490,16 and a
connection between the Denver images and a lost Triumph of Chastity by a follower of Botticelli that
series of Petrarchan Triumphs known in 1501 when was probably part of a larger cycle.17 Although it
they were used with Andrea Mantegna’s Triumphs cannot be demonstrated that these works were all
of Caesar as decorations for a temporary theater at commissioned for weddings, the majority prob-
the court of Mantua. ably were. Almost all 173 entries in Apollonio di
The theater decoration was described in a 1501 Giovanni’s account book, for example, seem to
letter from Sigismondo Cantelmo, a Ferrarese relate to marriage commissions.18 From northern
courtier, to Duke Ercole of Ferrara. (See Appendix Italy, there is also a small, gilt pastiglia “cofanetto”
for the full text of the letter.) Cantelmo stated or jewelry box depicting Trionfi that dates between
that the Petrarch cycle was also by Mantegna, and 1450 and 1460 and is believed to be a marriage gift
Wastler and several later scholars believed this for an aristocratic couple, perhaps commissioned
lost cycle was by Andrea’s son Francesco. Wastler by a member of the Este family.19
thought that Francesco had painted the Denver
cycle as a preparatory study for the large-scale Part i: Original Function
theater decorations,6 a hypothesis that now seems It is almost certain that the Kress Triumphs, given
untenable. Later scholars have suggested other their dimensions, format, and subject matter, were
attributions including Francesco Buonsignori,7 part of the decoration for luxurious domestic
Francesco Benaglio,8 Niccolò da Verona,9 and furniture. Each of the Denver panels measures
Girolamo da Cremona.10 approximately 52 × 54 cm. Their X-radiographs
In 1915 Paul Schubring included the panels in reveal that the support consisted originally of two
his book on cassoni, placing them in the context of continuous planks of wood, on each of which
luxury domestic furniture probably produced on three scenes were painted.20 They were probably
the occasion of a wealthy marriage. Like Wastler, either cassoni frontals or spalliera panels. A cassone

The Triumphs of Fame, Time, and Divinity
Follower of Andrea Mantegna, late 15th century
Egg tempera and oil on cradled wood panel
20 1/2 × 21 1/4 in. (52.1 × 54 cm); 20 3/8 × 21 3/8 in. (51.8 × 54.3 cm);
20 1/2 × 21 3/8 in. (52.1 × 54.3 cm)
1961.169.2 (k-10, 11, 14)
Fig. 3. The Triumphs of Fame, Time, and Divinity, before cleaning and restoration.
Fig. 4. The Triumphs of Fame, Time, and Divinity (fig. 3), after cleaning and restoration (framed together).
Wendy Partridge 101

Fig. 5. The Graz cassone, Triumphs of Love, Chastity, and Death, ivory cassone frontal, 1477, Cathedral of Graz, Austria. One of a pair of
cassoni commissioned for the wedding of Paola Gonzaga and Leonard von Georz.
Fig. 6. The Graz cassone, Triumph of Chastity (fig. 5), detail. Fig. 7. The Graz cassone, Triumph of Death (fig. 5), detail.

was a hope chest often used for seating as well as The earliest known record of a pair of chests
storage while the spalliera is a somewhat fluid term with a painted spalliera above comes from Bernardo
that described both wainscoting, set above eye di Stoldo Rinieri’s Riccordi of 1458. In the Riccordi
level into the walls of a room, and painted panels he documented an expenditure for the redecoration
providing backs for furniture. Cassoni, cassapanche of his home prior to his marriage to Bartolommea
(benches), or lettucci (day beds) could all have had di Dietosalvi.27 “A spalliera that is 13 braccia long
spalliere.21 Scholars writing on Renaissance domes- by one and a half braccia high above said chests,
tic furniture have often pointed out that this is a painted with the story of Lorenzo’s tournament,
difficult topic since so little painted decoration is with gilt framing and columns” is also described
preserved in its original context.22 in the 1492 inventory from Lorenzo de’ Medici’s
Each plank would have been approximately palazzo on the Via Larga.28 Or again, Giovanni
52 × 162 cm, somewhat larger than a typical cassone Rucellai listed a present of “a pair of chests with
frontal which averaged about 42 × 149 cm.23 They very rich spalle” for his son’s wedding in 1466.29
could simply have been atypically large cassoni. The The two panels of the Story of Jason at the Metro-
large Graz chests, after all, measure 95 × 190 cm.24 politan Museum were probably spalliere for paired
The Kress cycle’s pattern of damage, however, cassoni.30 At 60 × 150 cm each, they are similar in
is not consistent with that found on many cassoni. size to the Denver panels.
Cassoni, placed on the floor, often sustained a Decorated spalliere could also be associated with
significant number of scratches and abrasions. lettucci and benches as we know from inventories,
Furthermore, since the locking mechanism for records of payment,31 and illustrations. There is,
a cassone was usually located at the center of the for example, a woodcut from Savonarola’s 1499
chest in the molding above the image area, the top Predica dell’Arte del Ben Morire where the design on
center of the painted panel often has a series of the high back of the lettuccio was probably meant
indentations made by keys that struck its surface. to be intarsia. A Birth of the Virgin from Verona of
Although the Kress paintings have sustained a sig- circa 1500 shows a lettuccio depicting a landscape.32
nificant number of losses, these relate to knots in
the wood support rather than to harsh use or dam-
age from banging keys. Ellen Callmann believed,
furthermore, that the horizon somewhat above
center on the Kress paintings makes the images
read better when seen slightly above eye level,25 a
further argument against the cassone panel theory.
The Kress cycle may have originally formed
spalliere paired with cassoni in the manner of the
set painted by Biagio di Antonio in collaboration
with Jacopo Sellaio for the Nerli–Morelli mar-
riage of 1472 (fig. 8). The framework for the two
spalliere has been much restored, and they were not
originally attached to the chests. It appears that
the chests were originally joined by a single long
spalliera since Morelli’s records, apparently account
books, show payment to Sellaio and Antonio for
“a pair of chests and a spalliera and a base …”26
While the Nerli–Morelli cassoni/spalliera is the only
Fig. 8. Morelli cassone, Zanobi di Domenico and Jacopo del
extant example of such a set, it does not seem to Sellaio, 1472, wood, gesso, tempera, and gilding. Courtauld
have been a unique arrangement. Institute of Art Gallery, London.
Wendy Partridge 103

Fig. 9. Triumph of Chastity (fig. 1), detail of the coat of arms. Fig. 10. Triumph of Fame (fig. 3), detail of the coat of arms.
With the exception of the school of Verona of the married couple. Considerable expense was
painting, it should be noted, all mentioned spent on such furnishings.37 The principal bed-
examples of spalliere paired with cassoni, lettucci, or chamber could also be a reception room for hon-
benches come from Florence. This may be a result ored visitors or favored friends,38 its furnishings
of the comparative lack of research on furniture often intended to create an impression. Again
in northern Italy. Fewer may have been made, and using an example from Florence, it has been
what does survive has not been studied. Very little established that the largest concentration of pur-
is known about domestic furniture at the court of chases for the home occurred on the occasion of
Milan under the Sforza or the court of Mantua marriage and that most of these commissions
whose furnishings were completely dispersed in were installed in the bedchamber.39
1708.33 Spalliere do not seem to be an exclusively Had the paintings not been examined at the
Florentine phenomenon—the Sienese also made Conservation Center, the case for their having been
them.34 Although lettucci have been assumed to commissioned for a Gonzaga marriage would have
be mainly Tuscan, there are also records of them rested on the panels’ similarity to Paola Gonzaga’s
in Genoa, Milan, and Ferrara.35 It should be cassoni. The Colloredo provenance is also signifi-
remembered, finally, that wealthy Florence often cant, since in 1721 Carlo Ludovico Colloredo
set trends and provided furniture fashions for the married Eleonara Gonzaga of Vescovato.40 The
rest of Italy.36 Colloredo family, then, could have possessed some
Renaissance Gonzaga furniture.
Part ii: Possible Occasions for the Commission When Professor Jonathan Alexander saw the
The Kress paintings were probably commissioned Kress paintings at the Conservation Center he
in the context of a marriage, regardless of their identified the tiny coat of arms on the tower in
original format. They probably were intended to the background of the Triumph of Chastity as
adorn a bedroom, or at least the private apartments, belonging to the Gonzaga, and in the Triumph of

Fame as the stemma of the Sforza (figs. 9 and 10).41 daughter of Barbara of Brandenburg and the
Although it is perhaps not wise to place too much Marchese Ludovico. Dorotea was also sister to
significance on these shields, heraldic devices were Paola of the Graz ivory cassoni. Galeazzo was first
used in the Renaissance with great care. Their engaged to Dorotea’s sister Susanna, but the
placement near the images of Chastity and Fame match was transferred to Dorotea in 1457 when
would have been eminently appropriate in the Susanna seemed to be becoming a hunchback.
context of a marriage, the first for the bride, the When Dorotea attained her majority in 1463,
second for the groom. Galeazzo’s father, the Duke of Milan, sent a letter
Professor Alexander, moreover, pointed out to Barbara of Brandenburg requesting a medical
a 1489 marriage between Maddalena Gonzaga examination to ensure that his son’s new fiancée
(1472–1490), sister of the Marchese Francesco ii, would not also become hunchbacked. Ludovico
and Giovanni Sforza (died 1510), Lord of refused. In 1465, the second marriage contract
Pesaro.42 (See family tree, fig. 11.) The marriage was also annulled, and Dorotea died in 1467.43
did not last long since Maddalena died in child- According to this scenario, the panels could have
birth in 1490. Another Sforza–Gonzaga engage- been painted sometime between 1457 and 1463.
ment that might have provided an occasion for This agrees with Ronald Lightbown’s dating
the commission was the betrothal of Galeazzo of the panels to the 1450s or 1460s,44 and seems,
Maria Sforza (1444–1476), heir to the Duchy further, to support the proposed attribution
of Milan, and Dorotea Gonzaga (1449–1467), discussed below.
Fig. 11. The Gonzaga family tree.
Wendy Partridge 105

In either case it is necessary to discuss the
small size of the coats of arms. Furniture com-
missioned for marriages tended to contain rather
more visible heraldic devices. We see this in
the numerous Gonzaga imprese found on Paola
Gonzaga’s Graz cassoni or in large Gonzaga/
Montefeltro coats of arms and Montefeltro
devices for a marriage chest commissioned for
Elisabetta Gonzaga (Maddalena’s sister) on the
occasion of her marriage in 1488 to Guidobaldo Fig. 12. Adoration of the Shepherds, illumination from Barbara
da Montefeltro, Duke of Urbino.45 Elisabetta’s of Brandenburg’s Missal, 1461–68, tempera on parchment.
Cathedral of San Pietro, Mantua.
cassone was somewhat atypical, though, since
devices and coats of arms were often found on
the sides and lids of chests. This was true for
the Graz chests or the Conquest of Trebizond cassone the two painters, especially when the Denver
at the Metropolitan Museum, one of the few panels are compared to Girolamo’s early work
painted cassoni to survive almost intact. Coats of in the Missal that he illuminated in Mantua for
arms and heraldic devices are also found in the Barbara of Brandenburg between approximately
decorative pastiglia work framing the Metropol- 1461 and 1468.47 The similarities are not particu-
itan’s King Solomon and the Queen of Sheba and the larly surprising since Girolamo da Cremona could
Conquest of Naples by Charles of Durazzo frontals. be considered a Mantegna follower, especially
The Denver panels, then, could have been framed during his stay at Mantua. He seems to have been
by decorative imprese or paired with chests with a friend (and possibly a student) of Mantegna’s
more prominent heraldic devices similar to the and was probably recommended by him to Barbara
one belonging to Elisabetta Gonzaga. of Brandenburg.48 There is, however, a second
Further information might be found in the hand found in the Missal, working in a style
Mantua archives. In Pietro Torelli’s summary of similar to, but not as accomplished as Girolamo’s.
the documents, there is a category for Sponsali, This second artist seems to have been respon-
Maritaggi e Doti delle principesse di Mantova passate in sible for the Adoration of the Magi and the decorated
altre famiglie, e loro ragioni with headings for “Dorotea initial “I” (figs. 12 and 13).49 The illuminator’s
figlia di Lodovico II march.,” and “Maddalena figlia di rocky landscapes display a more exaggerated
Federico III march., moglie di Giovanni Sforza signore di chiaroscuro than the boulders in the Kress paint-
Pesaro.” There are also “Atti relativi ai matrimoni di ings, although the backgrounds of the two illu-
Elisabetta e Maddalena Gonzaga.”46 A look through minations and the Trionfi are quite similar. The
these records could reveal the type of furnishings blue-green hill dotted with white city walls and
commissioned for the engagements even if there towers in the Adoration is almost identical to the
were no specific records concerning a cycle of blue-green hill in the background of Chastity.
painted Petrarchan Triumphs. The rhythm and shape of the bare trees in the
Adoration remind one of the bare trees in Death.
Part iii: A Possible Attribution Both artists had similar problems in rendering
Caterina Furlan’s 1973 article was the most recent anatomy. The foreshortening of the Virgin in the
attempt to attribute the Denver series. Her attri- Adoration does not quite work. Her head is a bit
bution to Girolamo da Cremona does not seem too big for her body in the same way that the
convincing, since Girolamo was significantly more captive Jupiter’s head in Love is also too large.
accomplished than the anonymous Kress painter. The illuminator allows faces to intersect in a
However, there do seem to be similarities between bizarre manner, a hallmark of the Kress Petrarch

cycle. In the Adoration, one shepherd’s bushy hair
obscures the mouth and chin of the second.
This produces a clumsy effect similar to Time’s
chariot slicing through the faces of his entourage
or the strange intersection of the faces of the old
men leading Time’s procession. By contrast, the
sweetness of the Virgin’s face with its high fore-
head, heavy lidded eyes, and downturned bow
mouth could belong to Fame or to any lady sur-
rounding Chastity.
A final comparison in favor of the anonymous
illuminator is the similarity of his “I” to the
chariots in the Triumphs. Not only are the slightly
naïve perspective and gold decoration alike, but
the figures outlined in black on a gray ground
look very close to those decorating the chariots
of Love and Chastity.
Several art historians who have looked at the
Kress Triumphs at the Conservation Center have
commented on the relatively high quality of the
chariots and columns in contrast to charming, but
less accomplished human figures. This skill with
shell gold coupled with the Kress painter’s fasci-
nation with minute detail have led some scholars
to wonder if he was not perhaps a miniaturist.
However, Professor Alexander has pointed out
that there is very little evidence to document
illuminators working as furniture painters,50
although there is at least the one instance of
Apollonio di Giovanni’s Petrarch illuminations.
My attribution is, of course, highly speculative.
From “Mantegna Follower” we could only say,
slightly more specifically, “Girolamo da Cremona
Follower.” A definite attribution may not be
possible for the Kress panels. The difficulty is
not surprising when we read, in Delle Arti e degli
Artefici di Mantova, Carlo d’Arco’s compilation
of thirty(!) names of pupils and assistants to
Fig. 13. Decorated initial “I” illumination from Barbara Andrea Mantegna culled from documents in
of Brandenburg’s Missal, 1461–68, tempera on parchment.
Cathedral of San Pietro, Mantua. the Mantua archives. This list consists only of
artists for whom no work is known and does not
include, for example, Mantegna’s son Francesco
or Francesco Buonsignori.51
Wendy Partridge 107

Part iv: The Relationship to one of the areas was decorated with six paintings
Andrea Mantegna of the triumphs of Caesar by the hand of the
exceptional Mantegna … Around the stage on
In addition to the question of attribution, the lower pediment (or decorative façade?) were
scholars since Wastler have puzzled over Andrea the Triumphs of Petrarch also by the hand of the
Mantegna’s relationship to this Petrarch painter Mantegna …58
cycle. Although the panels possess a certain
Mantegnesque style that was influencing artists Furthermore, it seems quite likely that the Mar-
throughout northern Italy, could there really be mirolo Triumphs that Ghisolfi mentioned in 1491
any direct connection between the court painter actually depicted the triumphs of Alexander
and the delightful but somewhat naïve Kress the Great. The Bolognese antiquarian and natu-
paintings? Yet, there may have been a relationship ralist Ulisse Aldrovandi described the castle
between the Kress paintings and the Petrarchan of Marmirolo in his Itineraria Mantuae written
theater decorations described in the letter by between 1561 and 1571. He mentioned seeing an
Sigismondo Cantelmo (see Appendix). “Aula magna depicta cum historia triumphi Alexandri
There are numerous records documenting Magni ab Ecc. ti Mantegna.”59
trionfi painted for the Gonzaga, especially for While we cannot rule out the possibility that
the Marchese Francesco ii (born 1466, reigned the Petrarch Triumphs seen by Cantelmo were by
1484–1519), and some scholars have argued that Francesco or another one of Mantegna’s pupils,
the Kress paintings and the Graz ivories must be it does not seem possible to associate them with
based on one of these lost cycles.52 Francesco the Triumphs being painted in 1491 in Marmirolo.
Buonsignori wrote to Francesco ii in an undated It is even possible that Andrea Mantegna himself
letter concerning a Triumph of Fame.53 In a 1493 painted Cantelmo’s Petrarch cycle. In either event,
letter Niccolò da Verona discussed a “trionfo” he a set of Petrarch Triumphs would imply the possi-
had painted for Francesco.54 Perhaps the most bility of sketches and designs. These designs
important reference is in a 1491 letter to Francesco could have found their way to the hands of both
from Bernardo Ghisolfi, the supervisor of the the carver of the Graz ivories and the painter of
decoration at the new palace at Marmirolo. the Kress cycle. Given the Kress artist’s insecuri-
Ghisolfi wrote, “Francesco and Tondo have begun ties with the figure and perspective, it seems hard
to paint the Triumphs on canvas like Andrea to argue, however, that Mantegna was directly
Mantegna. They say that as a result they do them involved in supervising him. We are ultimately
more quickly and that they will be more beautiful left with questions about the relationship of a
and durable …”55 Tondo has not been identified, major artist like Andrea Mantegna to the numer-
but since Francesco Mantegna was known to have ous second-tier artists, craftsmen, and decorators
worked at Marmirolo, most scholars have assumed employed by the courts and about the relationship
that “Francesco” refers to Andrea’s son.56 Wastler of those craftsmen to each other.
and Furlan both believed that the 1491 canvas
paintings might have been a lost Petrarch cycle. Conclusion
Since there are letters concerning transporting The Kress paintings seem to have formed part of
“trionfi” from Gonzaga to Mantua in 1503, 1505, a body of Petrarch cycles painted for domestic
and 1507 and from Marmirolo to Mantua in 1507 interiors. Given their dimensions, damage pat-
to serve as temporary theater decorations,57 it has terns, and perspectival construction, they were
been argued that the Triumphs Cantelmo saw in more likely spalliera paintings than cassoni frontals.
1501 were also from Marmirolo and by Francesco Unfortunately, it is impossible to know the type
Mantegna. Cantelmo clearly stated, however, that of furniture for which they provided the backs.
the Petrarch Triumphs were painted by the same The panels were probably commissioned by
artist who painted the Triumphs of Caesar: the Gonzaga for a Sforza–Gonzaga marriage.

Given the similarities between the hand in the 4. Shapley (1968), pp. 27–8.
5. Wastler (1880), p. 71.
1461–68 Missal and the Kress painter, the earlier 6. Ibid., pp. 61–2, 72.
Sforza–Gonzaga engagement between Gian 7. Schubring (1915), p. 358.
Galeazzo and Susanna/Dorotea seems the more 8. Shapley (1968), p. 27.
9. Carandente (1963), p. 94.
likely candidate for the commission. They may 10. Furlan (1973), pp. 81–90.
have been painted by someone in the circle of 11. Schubring (1915), pp. 357–8. The Graz chests—now
Girolamo da Cremona and might be loosely based reliquaries containing relics of Saints Martin, Vincent,
Maxentia, and Agatha—were given by Pope Paul v
on designs by Andrea Mantegna or one of his to Archduke Ferdinand ii in 1617. Coudenhove-Erthal
pupils, but the master could not have directly (1931), p. 9.
supervised their execution. 12. Carandente (1963), pp. 57–9.
13. Schubring (1915), pp. 271, 279.
Further research on dowries and trousseaus 14. Callmann (1974), p. 60.
for Gonzaga brides could be done in the Mantua 15. Carandente (1963), pp. 67–8.
archives. An examination of extant records of 16. Barriault (1994), pp. 67–8.
17. Schubring (1915), p. 294.
Gonzaga wedding festivities might also be useful. 18. Callmann (1974), Appendix.
Although this paper raises more questions than 19. Molfino and Natale (1991), pp. 226–9.
provides answers, it contributes to the study 20. The X-radiographs are on file at the Conservation Center.
It is not known when they were originally cut into six sep-
of domestic painting in northern Italy and arate pieces. Joseph Wastler described seeing six paintings
in its courtly society, a subject that is seldom in 1880, p. 63.
addressed given the paltry amount of furniture 21. Schiaparelli (1908), pp. 159–60.
22. Vidas (1997) p. 41.
that has survived. 23. Barriault (1994), p. 57.
24. Coudenhove-Erthal (1931), p. 16.
Acknowledgements 25. Ellen Callmann, personal communication, 1998.
I would like to thank Dianne Dwyer Modestini 26. “uno paio di forzieri e per una spalliera e per la predella …”
Callmann (1988), p. 9. The difference between cassoni,
for encouraging me to write this paper. Dianne, forzieri, and coffani is not clear although Peter Thornton
Jennifer Sherman, Friederike Steckling, and thought that the distinction might be based on shape.
Sue Ann Chui provided moral support and many “Cassone” tends to be used as a generic term by most
current scholars. Thornton (1991), pp. 192–3.
insightful suggestions. The paper reads much 27. Callmann (1988), pp. 6, 16.
more clearly thanks to editing by Colin Eisler 28. “Una spalliera di braccia XIII et alta braccia 1/2 sopra a detti
and Constance Lowenthal. chassone, dipintovi drento la storia della giostra di Lorenzo, con
cornicie et colonette messe d’oro.” Schiaparelli (1908), p. 167.
29. “uno paio di forsieri colle spalle molto riche,” Barricault (1994), p. 38.
Wendy Partridge is Associate Paintings Conservator at the 30. Pope-Hennessy and Christiansen (1980), p. 29.
Intermuseum Conservation Association in Cleveland, Ohio. 31. In 1463 Giuliano da Maiano received payment from
Piero degli Alberti for narrative designs for a lettuccio,
She earned a degree in Paintings Conservation and an M.A. for example. Barricault (1994), p. 29.
in Art History from the Conservation Center, Institute of 32. Honorati (1981), pp. 39–41.
Fine Arts, New York University. 33. Alberici (1969), p. 7.
34. Callmann (1996), p. 363.
35. Thornton (1991), p. 152.
Notes 36. Alberici (1969), p. 8.
1. Trapp (1992–93), p. 40. 37. Thornton (1991), pp. 11–12.
2. Shapley (1968), pp. 27–8. 38. Ibid., p. 228.
3. No information could be found on this palazzo. There was 39. Vidas (1997), pp. 41–2.
also a Palazzo Colloredo in Mantua (now the Palazzo di 40. Ganzer (1988), p. 20.
Giustizia) built in the 1620s by the architect A.M. Viani. It 41. Jonathan Alexander, personal communication, 1998.
belonged to the Gonzaga until 1721 when it was transferred 42. Ibid.
by marriage to the counts of Colloredo. The palazzo 43. Conigilio (1967), pp. 66–70, 75.
remained in the Colloredo family until the first half of the 44. Lightbown (1986), p. 464.
nineteenth century. Marani and Perina (1965), p. 196, n. 81. 45. Chambers and Martineau (1981), p. 149.
It is possible that the Kress cycle was in Mantua until the 46. Torelli (1920), pp. 37, 51–2.
early nineteenth century and then was transferred to the 47. Pastore and Manzoli (1991), p. 122.
Colloredo castle in Udine. 48. Hermann (1994), p. 241.
Wendy Partridge 109

49. Pastore and Manzoli (1991), pp. 167–70. Ganzer, Gilberto (1988) “Testimonianze dei Gonzaga in terra
50. Jonathan Alexander, personal communication. friulana,” Antichità Viva 27, pp. 20–22.
51. D’Arco (1857), pp. 45–8. Hermann, Hermann Julius (1994) La Miniatura Estense.
52. D’Essling and Müntz (1902), p. 152; Coudenhove-Erthal Modena: Franco Cosimo Panini.
(1931), pp. 42–3; and Carandente (1963), p. 94. Honorati, Maddalena Trionfi (1981) “A proposito del lettuc-
53. Martindale (1979), document 7, p. 182. cio,” Antichità Viva 20, pp. 39–47.
54. Ibid., document 8. Lightbown, Ronald (1986) Mantegna, with a Complete Catalogue of
55. “Francesco e Tondo insieme ancoro loro commenzaria a depingere the Paintings, Drawings, and Prints. Oxford: Phaidon.
quelli trionfi li quali a lor ge par farli suso tela secondo ha facto mess. Marani, Ercolano and Perina, Chiara (1965) Mantova: Le Arte,
Andrea Mantegna et dicono cusi facendo farano più presto et serano Volume : dall’Inizio del Secolo XV alla Metà del XVI. Mantua:
più belle e più durabile …” D’Arco (1857), p. 30. Istituto Carlo D’Arco per la Storia di Mantova.
56. Lightbown (1986), p. 142. Martindale, Andrew (1979) The Triumphs of Caesar by Andrea
57. Martindale (1979), documents 15–18, 20, pp. 183–4. Mantegna in the Collection of Her Majesty the Queen at Hampton
58. “una delle bande era ornata delli sei quadri del Caesaro triumpho per Court. London: Harvey Miller Publishers.
man del singulare Mantengha … Dintorno alla scena al frontespitio Molfino, Alessandra Mottola and Natale, Mauro (eds.) (1991)
di basso era li triumphi del Petrarcha, ancor loro penti per man del Le Muse e il Principe, Arte di Corte nel Rinascimento Padano,
pN Mantengha …” D’Ancona (1971), p. 382. Catalogo. Modena: Franco Cosimo Panini.
59. Franchini et al. (1979), pp. 194–5, 237. Aldrovandi’s confu- Pastore, Guise and Manzoli, Giancarlo (1991) Il Messale di
sion between Mantegna father and son at a generation’s Barbara. Mantua: Editrice Sintesi.
distance seems more probable than Cantelmo’s. During Petrarca, Francesco (1984) I Trionfi, Guido Bezzola (ed.).
Andrea Mantegna’s lifetime Cantelmo attended a spectacle Milan: Rizzoli.
displaying the wealth and taste of the Gonzaga, a family Pope-Hennessy, John and Christiansen, Keith (1980) Secular
very proud of the achievements of their official painter. Painting in th-Century Tuscany: Birth Trays, Cassone Panels, and
Portraits. New York: Metropolitan Museum of Art.
Schiaparelli, Attilio (1908) La Casa Fiorentina e i suoi Arredi nei
Bibliography Secoli XIV e XV, Vol. 1. Florence: Casa Editrice le Lettere.
Alberici, Clelia (1969) Il Mobile Lombardo. Milan: Electa. Schubring, Paul (1915) Cassoni: Trühen und Trühenbilder der italienis-
D’Ancona, Alessandro (1971) Origini del Teatro Italiano. Turin: chen Frührenaissance, ein Beitrag zur Profanmalerei im Quattrocento.
Bottegha d’Erasmo. Leipzig: Verlag von Karl W. Hiersemann.
D’Arco, Carlo (1857) Delle Arti e degli Artefici di Mantova: Notizie Shapley, Fern Rusk (1968) Paintings from the Samuel H. Kress
Raccolte ed Illustrate con Disegni e con Documenti. Mantua: Collection, Vol. : Italian Schools, XV–XVI Century. London:
Tipographia Ditta Giovanni Agazzi. Phaidon, pp. 27–8.
Barriault, Anne B. (1994) Spalliera Paintings of Renaissance Tuscany: Thornton, Peter (1991) The Italian Renaissance Interior,
Fables of Poets for Patrician Homes. University Park, PA: –. New York: Harry N. Abrams.
Pennsylvania State University Press. Torelli, Pietro (1920) L’Archivio Gonzaga di Mantova. Mantua:
Callmann, Ellen (1974) Apollonio di Giovanni. Oxford: Clarendon Mondadori.
Press. Trapp, J.B. (1992–93) “The iconography of Petrarch in the age
––– (1988) “Apollonio di Giovanni and painting for the early of humanism,” Quaderni Petrarcheschi 9–10, pp. 11–73.
Renaissance room,” Antichità Viva 27, pp. 5–17. Vidas, Marina (1997) Representation and Marriage: Art, Society, and
––– (1996) “Spalliera” in Jane Turner (ed.), The Dictionary of Gender Relations in Florence in the Late Fourteenth through the
Art, Vol. 29. New York: Grove Press, pp. 359–63. Fifteenth Century. Ph.D. dissertation, New York University.
Carandente, Giovanni (1963) I Trionfi nel Primo Rinascimento. Wastler, Joseph (1880) “Mantegna’s Triumphe des Petrarca,”
Turin: Eri, Edizioni rai Radiotelevisione Italiana. Zeitschrift für Bildende Kunst, 15, pp. 61–72.
Chambers, D. S. and Martineau, J.T. (eds.) (1981) The Splendours Weisbach, Werner (1919) Trionfi. Berlin: G. Grote.
of the Gonzaga,  November –  January, , Victoria and Albert
Museum: Catalogue. Milan: Victoria and Albert Museum.
Cole, Alison (1995) Art of the Italian Renaissance Courts. New Appendix
York: Harry N. Abrams. Letter from S. Cantelmo, 1501
Conigilio, Giuseppe (1967) I Gonzaga. Milan: Dall’Oglio. Published in d’Ancona (1971), pp. 381–3.
Coudenhove-Erthal, Eduard Count (1931) Die Reliquienschreine
der Grazer Doms und ihr Beziehung zu Andrea Mantegna. Graz: Eccellentissimo et unico Sig. Mio Com. Essendo stato lo apparato facto
Leykam-Verlag. da questo Ill.mo Sig. Marchese sumptuosissimo et meritamonte da essere
D’Essling, Prince (Victor Massena) and Müntz, Eugène (1902) equiperato ad qual se voglia temporaneo theatro delli antiqui o moderni,
Pétrarque: ses Etudes d’Art, son Influence sur les Artistes, ses Portraits non dubito V. Ex.a per più vie harà inteso l’essere del spectaculo quale sia
et ceux de Laure, les Illustrations de ses Ecrits. Paris: Gazette de stato: non di meno ancor min non voglio mancare dal offitio della mia
Beaux-Arts. debita servitù: certificandola scrivo la verità, quantuncha tanta magni-
Franchini, Dario A. et al. (1979) La Scienza a Corte: Collezionismo ficentia recerchasse chi sapesse meglio scrivere, et exprimendo pengere la
Eclettico, Natura e Imagine a Mantova fra Rinascimento e nobilità et excellentia del prefato spectaculo; la vaghezza del quale con
Manierismo. Rome: Bulzoni. quanta brevità potrò, me sforzarò demostrare ad V. Ill.ma Sig.a. Era la
Furlan, Caterina (1973) “I Trionfi della Collezione Kress, una sua forma quadrangula, protensa alquanto in longitudine: li doi lati l’uno
proposta attributiva e divagazioni sul tema,” Arte Veneta, al altro de rimpecto havevano per ciaschuno octo architravi con colonne ben
Vol. 33, pp. 81–90. conrespondenti et proportionate alla larghezza et alteza de dicti archi: le

base et capitelli pomposissimamento con finissimi colori penti, et de fogliami
ornati, representavano alla mente un edificio eterno ed antiquo, pieno de
delectatione: li archi con relevo di fiori rendevano prospectiva mirabile: la
largheza di ciascheuno era braza quactro vel cerca: la alteza propozionata
ad quella. Dentro nel prospecto eran panni d’oro et alcune verdure, si come
le recitationi recerchavano: una delle bande era ornata delli sei quadri del
Cesare triumphe per man del singulare Mantengha: li doi altri lati discon-
tro erano con simili archi, ma de numero inferiore, che ciascheuno ne
haveva sei. Doj bande era scena data ad actorj et recitatorj: le doe altre
erano ad scalini, deputati per le donne et daltro, per todeschi, trombecti et
musici. Al jongere del’ angulo de un de’ grandi et minorj lati, se vedevano
quactro altissime colonne colle basi orbiculate, le quali sustentavano quactro
venti principali: fra loro era una grocta, benchè facta ad arte, tamen natu-
ralissima: sopra quella era un ciel grande fulgentissimo de varij lumi, in
modo de lucidissime stelle, con una artificiata rota de segni, al moto de’
quali girava mo il sole, mo la luna nelle case proprie: dentro era la rota
de Fortuna con sei tempi: regno, regnavj, regnabo: in mezo resideva la
dea aurea con un sceptro con un delphin. Dintorno alla scena al frontespi-
tio da basso era li triumphi del Petrarcha, ancor loro penti per man del p.e
Mantengha: sopra eran candelierj vistosissimi deaurati tucti: nel mezo era
un scudo colle arme per tucto della C.a M.à; sopra l’aquila aurea bicapi-
tata col regno et diadema imperiale: ciascheuno teneva tre doppieri; ad
ogni lato era le insegne. Alli doi maiorj, quelle della S.tà de N. S. et quelle
della Cesarea Maestà: alli minorj lati quelle del C.o Sig. Re, et quelle
della Ill.ma Sig.a da Venetia; tra li archi pendevano poi quelle de V. Ex.,
quelle del Sig. duca Alberto Alemano: imprese de Sig. Marchese et Sig.a
Marchesana: sopre erano più alte statue argentate, aurate et de più colorj
metallici, parte tronche, parte integre, che assai ornavano quel loco; poi
ultimo era il cielo de panno torchino, stellato con quelli segni che quella
sera correvano nel nostro hemisperio …
Le recitationi sonno state belle et delectevole: Venere fo Philonico:
Sabato il Penulo de Plauto: Domenica lo Hippolito: Lunedì li Adelphi
de Terentio, da persone docte recitate optimamente con grandissima voluptà
et plausi de spectatorj. Per essere qui Monsignor Loys d’ Ars, locumtenente
dell’ Ill.mo Monsig. de Ligni, et non haver viste le doj prime, intendo
vogliono un altra volta pure il Philonico. Se ho mancato di questa in
alcuna cosa, prestissimo supplirò ad bocca con V. Ex.a; in bona gratia de
quella mi recomando. Datum Mantuae Xiij Februarii MD primo.
De V. EX.a
Servitore et Schiavo
Sigismondo Cantelmo.
Photography Credits
Figs. 1, 2, 3, 4, 9, and 10, pp. 98, 101, and 104. Denver Art
Museum Collection, Denver, CO, Gift of the Samuel H.
Kress Foundation (1961.169.1, 1961.169.2).
Figs. 5, 6, and 7, p. 102. Coudenhove-Erthal (1931), figs. 3, 5,
and 10.
Fig. 8, p. 103. Courtesy of the Courtauld Institute of Art
Gallery, London (f.1947.lf.4).
Fig. 11, p. 105. Cole (1995), p. 185.
Figs. 12 and 13, pp. 106 and 107. Pastore and Manzoli (1991),
pp. 167 and 169.
1
Wendy Partridge 111

Adoration of the Shepherds with Saint John the Baptist and Saint Bartholomew
Sienese School, ca. 1440
Egg tempera on wood panel
24 5/8 × 19 3/4 in. (62.6 × 50.2 cm)
El Paso Museum of Art, El Paso, Texas
1961-6/6 (k-1434)
Fig. 1. Adoration of the Shepherds with Saint John the Baptist and
Saint Bartholomew, before restoration.
Fig. 2. Adoration of the Shepherds with Saint John the Baptist and
Saint Bartholomew (fig. 1), after restoration.

A Portable Triptych in El Paso
O ne of the finest and least frequented of the regional Kress Collections

is in the old Mexican border town of El Paso. The largest and most
profitable of the S.H. Kress five-and-dime emporiums was in El Paso
and, in recognition of this fact, the city was designated to receive an
extraordinary group of fifty-eight paintings, chosen by Robert Manning,
assistant curator at the Kress Foundation, excellent connoisseur and loyal son
of Texas.1 The quality of the El Paso collection is even more remarkable
because it was the penultimate collection to be allocated, opening in 1960, a
telling indication of the richness and depth of the Kress reserves even at that
late date when the collections in Washington and seventeen other Regional
Collections had already been formed.
The collection consists primarily of Italian and Spanish masters and also
includes three fine portraits by Van Dyck, Rigaud, and Largillière. The Sienese
School is represented by a damaged but genuine Madonna and Child by Ambrogio
Lorenzetti, a Giovanni di Paolo Assumption of the Virgin,2 and a perfectly pre-
served Madonna and Child with Saints and Angels by Sano di Pietro, comparable
in quality to the National Gallery of Art’s, in its original frame and finely
finished on the reverse with a gilded and tooled emblem of San Bernardino.
Among the Sienese works is an ornate portable triptych (figs. 1 and 2), which
had over the years been attributed to Sassetta, the Osservanza Master, Giovanni
di Paolo, Pellegrino di Mariano, and Sano di Pietro.3 With this lack of consen-
sus, it was finally catalogued simply as Sienese School, circa 1440, exhibited at
Dianne Dwyer Modestini 113

the National Gallery of Art from 1951 to 1956 and
ultimately given to El Paso.
The central panel of the portable triptych has
a particularly elaborate representation of the Ado-
ration of the Shepherds, and is surmounted by a
separate compartment with the Last Judgment.
In the right wing a figure of Saint Bartholomew
holds his knife with the Virgin of the Annun-
ciation in the gable, while on the left Saint John
the Baptist is seen below the Angel of the
Annunciation. The reverse sides of the wings are
painted in faux porphyry inside an earth-green
border with medallions enclosing drawings of
the Madonna on the left and the Crucified
Christ on the right executed in lead white and
red earth on a yellow ochre field (figs. 3 and 4).
The moldings are original, although the tips of
the crockets of the central pinnacle are broken.
X-radiography reveals holes for the dowels
that originally fixed the central panel to a base,
now lost. Fig. 3. Adoration of the Shepherds with Saint John the Baptist and Saint
In the early nineteenth century, the painting Bartholomew (fig. 1), reverse of side panels before restoration.
was recorded in the collection of Max Chabrière-
Arles in Lyons before it was acquired by Harold
I. Pratt of New York.4 While in the Pratt Collec-
tion, it was exhibited in the Loan Exhibition
of Italian Primitives, at the Kleinberger Galleries
in New York in November 1917, catalogued by
Osvald Siren and M.W. Brockwell as Sassetta5
and in the 1939 New York World’s Fair Exhibi-
tion, number 350, catalogued by G.H. McCall,
again as Sassetta.6 It was bought by Wildenstein
and exhibited in Italian Paintings, 1947, catalogue
number 24, as Sano di Pietro.7
Attributions to the Master of the Osservanza
provoke the most contentious arguments8 in the
study of fifteenth-century Sienese painting. The
intricate and often contradictory art historical
vicissitudes of this artist or group of painters,
first identified by Roberto Graziani in 1948 fol-
lowing a suggestion of Roberto Longhi,9 and
named after a tripartite altarpiece in the Church
of the Osservanza near Siena, arise not only from
the total lack of documentation about this gifted
painter but also from inconsistencies in the body Fig. 4. Adoration of the Shepherds with Saint John the Baptist and Saint
of work assigned to him. The critical history has Bartholomew (fig. 1), reverse of the side panels after restoration.

been succinctly summarized recently by Cecilia A third hypothesis was put forward by Alessi
Alessi10 in the Grove Dictionary of Art: and Scapecchi (1985). They established that the
Osservanza panel was painted for S Maurizio,
Longhi recognized that two triptychs, formerly Siena, and that the date on the painting, 1436,
attributed to Sassetta, were the work of another refers to the foundation of the chapel by its
hand. The Virgin and Child with SS Jerome and patron, the grocer Manno d’Orlando (d 1442),
Ambrose (Siena, Osservanza) and the Birth of and not necessarily to the year in which the
the Virgin (Asciano, Mus. A. Sacra), formerly altarpiece was painted. They suggested that the
in the Collegiata, Asciano, both have a stylistic Osservanza Altarpiece and the Birth of the
affinity with Sassetta’s works but, in terms of Virgin date from the late 1440s and that the
narrative expression, still belong to the Late Master was active from the 1440s to the 1470s
Gothic tradition. Longhi observed that a further and was influenced by developments in Floren-
group of paintings was closely related to tine painting of that date, particularly by
these works. This included the predella of the Fra Angelico and Uccello. They further pro-
Osservanza Altarpiece (Siena, Pin. N., 216), a posed that the Master could be identified with
predella of St Bartholomew (Siena, Pin. N.), Francesco di Bartolmeo Alfei, a well-documented
scenes of the Passion (Rome, Pin. Vaticana; artist who was associated with Sano di Pietro but
Philadelphia, PA, Mus. A.; Cambridge, MA, whose work has not been identified. While Pope-
Fogg) and the scenes from the Life of St Hennessy (1987) did not accept the identification
Anthony Abbot (dispersed; e.g. panels in of the Master with Alfei, he accepted Alessi’s
Washington, DC, N.G.A.; New York, Met.; and Scapecchi’s attribution of additional works
Wiesbaden, Mus. Wiesbaden) previously also to the Master. These include the Virgin and
attributed to Sassetta. These last panels are Child (New York, Met.) and two paintings
difficult to integrate into the group. The full- of the Virgin of Humility (Altenburg, Staatl.
length painting of St Anthony Abbot (Paris, Lindenau-Mus.; New York, Brooklyn Mus. A.).
Louvre), which scholars have attempted to
integrate with the small scenes from the saint’s The El Paso triptych is mentioned only a few
life into a multipartite altarpiece, seems to times in the literature on Sassetta and the Master
come from another altarpiece. of the Osservanza. Its location is often incorrectly
described, and it has not been illustrated in any
Graziani named the painter the Master of the
Osservanza after the altarpiece in that church publication about Sassetta or his followers.11
and reconstructed his oeuvre around this work, Since my first visit to El Paso, I have been in-
ranging between the Pietà with St Sebaldus and trigued by this delightful object, particularly by
a Devotee (Siena, Monte Dei Paschi priv. col.), the fanciful depiction of the Adoration of the
datable 1432–3, and the painted cover of the Shepherds, and in the course of my travels, I
Gabella (tax records) showing the Archangel have had occasion to compare it with the gener-
Michael (Siena, Pal. Piccolomini, Archv Stato), ally accepted works by the Osservanza Master
dated 1444. Graziani proposed that the Master and Sano di Pietro. At some point I focused
took as his models Giovanni da Milano, Gregorio
di Cecco and Masolino, thereby combining
on an unusual punch mark common to most
Sienese and Florentine stylistic elements. of the small-scale works of the painter called
Graziani’s theory was accepted by Zeri, Carli, the Osservanza Master but not used by Sano di
Volpe, Laclotte, Benati, Angelini and Christiansen. Pietro after he became an independent artist,
beginning in 1444 with the signed and dated
A different theory was proposed by Berenson, altarpiece of the Gesuati.
who suggested that the Master’s oeuvre was the
early work of Sano di Pietro, known to have
An opportunity to study the work closely
been active from 1428 but whose earliest dated came in 1999, when it arrived at the Conservation
work is the Gesuati Polyptych of 1444 (Siena, Center to have a broken hinge replaced and a
Pin. N.). This was accepted by Brandi (1949), treacly varnish removed. It was encased in a
Pope-Hennessy (1956), Torriti and Boskovits. shadow box, which had been manufactured for

it in the 1950s,12 shortly after its acquisition by
the Samuel H. Kress Foundation. Professor
Mina Gregori, confirming my long-held belief,
immediately attributed it to the Osservanza
Master.13 This publication seemed an opportune
vehicle to present the results of my examination
and, above all, provide good reproductions of the
El Paso painting alongside four other portable
triptychs which have all, at one time or another,
been attributed to the Master of the Osservanza.
The central panels of the other four triptychs
represent the Madonna and Child, while the
wings, like k-1434, bear standing saints with
annunciation figures in the gables. They are in Fig. 5. Annunciation to the Shepherds, Sano di Pietro, after 1450,
tempera on wood panel, 21 1/4 × 27 1/8 in. (54 × 69 cm).
the collections of the Pinacoteca of Siena, the La Pinacoteca Nazionale di Siena, Siena.
Chigi-Saracini Collection, the Diocesan Museum
of Pienza, and the Rijksmuseum Meermanno-
Westreenianum in The Hague. Longhi’s famous
1940 footnote14 mentions the two Siena triptychs, In 1957 Enzo Carli published his essential
and Graziani writing in 194815 dates the first monograph, Sassetta and the Maestro dell’Osservanza,17
two to around the time of the Birth of the Virgin revising Graziani’s dating (and his own initial
from Asciano, and accepts the Meermanno- notion). Arguing that the Asciano Altarpiece
Westreenianum triptych as rather later, after depends from Sassetta’s scenes from the life of
1440, and under the influence of Vecchietta. The Saint Francis in the Borgo San Sepolcro Altar-
El Paso work is not mentioned, and Graziani piece (1437–44), he assigns also to this period
seems not to have been aware of its existence. the triptychs 177 of the Siena Pinacoteca and
The attribution of the Saint Anthony series is the Chigi-Saracini Collection. Carli agrees with
as convoluted as that of the five triptychs. Initially Graziani that the Meermanno-Westreenianum
they were given to Sassetta, and then, by Longhi triptych is of a slightly later date. Equivocating
and Graziani, to the Osservanza Master. Pope- over the dating of the Saint Anthony panels, he
Hennessy (1956) postulates that the panels are argues for an Osservanza Master still under the
the work of three different painters.16 ‘Artist A’ is influence of Sassetta’s Arte della Lana altarpiece,
Sassetta himself, entirely responsible for the Berlin that is between 1426 and 1432, and sees the hands
Mass and collaborator with ‘Artist B,’ his assistant, of two different artists where Pope-Hennessy saw
on other panels from the series. ‘Artist B’ may three. However Carli rejected the Vico di Luca
be one of Graziani’s candidates for the undocu- hypothesis stating, “I wouldn’t be surprised if
mented Master of the Osservanza, Vico di Luca, Artist B is early Sano di Pietro and some day it
who is also the author of the Passion predellas, might be proven.”18 He excludes the Pienza trip-
the “Pratt Triptych” (by that time in Washing- tych from the oeuvre of the Osservanza Master,
ton), and the Meermanno-Westreenianum trip- considering it by an inferior hand, and gives “the
tych. A third figure, ‘Artist C,’ is Sano di Pietro, charming triptych from the Pratt Collection,
and to him Pope-Hennessy assigns the Osservanza New York (sic)” to the same painter whom he
and Asciano altarpieces, the Serristori Pietà, and baptizes the “Master of Pienza,” noting the
“three interrelated triptychs in the Siena gallery influence of Giovanni di Paolo in the figure of
(no. 177), the Chigi-Saracini Collection and the Saint John the Baptist. In a meandering and
Museo d’Arte Sacra at Pienza.” inconclusive rumination about the scene of the

Annunciation to the Shepherds in the El Paso shepherds, with their great winter mantels, are
triptych he compares it to: warming themselves, and ecstatically greeting the
sudden apparition of the angel in a cloud of gold
the enchanting panel No.  in the Pinacoteca against the blue sky. Over the barren and gray
of Siena [fig. 5] a work by Sano di Pietro [for landscape of the crete senesi, punctuated by only a
which] one could consider a date of around few trees, a red turreted city sprouts, rendering
1450; but the reverse is more likely—that Sano even more magical the idea of the apparition.19
based his work on that of the Master of Pienza.
While not an exact copy, the composition
Referring to this painting by Sano di Pietro, clearly, as Carli observed, derives from the El Paso
which measures 54 × 69 cm, Piero Torriti writes: painting: the angel, the cloud of gold (in our case
Not a work of refined execution, in fact, one real gold leaf and not yellow paint), the flock of
would have to say that it was rather grossolana black and white sheep in their enclosure, the magi-
(and perhaps for this reason later than the tradi- cal turreted towns scattered about, the same type
tional dating of 1450) but of great fascination of dog. Unfortunately Carli neither reproduces the
because of the extraordinary interpretation of El Paso triptych nor follows this intriguing line
the event: in the center, closed in a paddock, is a of reasoning about its charming landscape back-
group of hairy black and white sheep, crowded ground which, both stylistically and technically, is
one against the other in such a way that the flock
becomes simply a black and white patch. The
closely related to the fanciful and much admired
landscape in the background of Saint Anthony
Tempted by Gold in the Lehman Collection (fig. 6).
Cecilia Alessi and Pietro Scapecchi20 argue that
the two portable triptychs that Carli assigned to
the Master of Pienza (ours and the Pienza trip-
tych), are in fact by the Master of the Osservanza,
painted around 1445 and reflecting the influence
of Sassetta’s Borgo San Sepolcro Altarpiece; they
point out the similar figures of the two lateral
saints in both triptychs.21 Dr. Alessi, in a verbal
communication, pointed out the strong influence
of Giovanni di Paolo, to whom the young artist
may have been apprenticed, indicating that,
according to her reconstruction, it is among the
earliest of his works.
Finally, Keith Christiansen noted: “perhaps
most illuminating are four small portable trip-
tychs in the Pinacoteca Nazionale and the
Chigi-Saracini Collection, Siena; the Museo
della Cattedrale, Pienza; and the Rijksmuseum
Meermanno-Westreenianum, The Hague.”22 For
the fifth portable triptych, The Adoration of the
Shepherds with Saints John and Bartholomew in the Kress
Collection in El Paso, Christiansen followed Carli
in attributing it, together with the Pienza triptych,
to a different hand, the same one responsible
Fig. 6. Saint Anthony Tempted by Gold, Master of the Osservanza,
mid-1430s, tempera on wood panel, 18 3/4 × 13 5/8 in. (47.6 × for the predella with scenes from the life of the
34.6 cm). The Metropolitan Museum of Art, New York, NY. Virgin in the Vatican. While the Pienza and El

by Alessi and Scapecchi. Pope-Hennessy appears
to have held the same view as he assigned this
work to ‘Artist B,’ possibly Vico di Luca, Sassetta’s
assistant in the Death of Saint Anthony and the same
painter responsible for the five Passion predellas,
implicitly assigning the El Paso work to the artist
or group of artists now accepted as the Osser-
vanza Master. Carli and Christiansen consider
both the El Paso triptych and the one in Pienza
to be by a different hand called the “Master of
Pienza,” although they do not concur about other
works assigned to this artist.
Let us for a moment turn our attention to the
five triptychs under discussion.
1. Triptych 177, Pinacoteca di Siena (fig. 8)

The central panel depicts the Madonna and
Child seated on a cushion. Four angels hold a
cloth of honor, once silver leafed, as was the
Fig. 7. Madonna and Child with Saint Catherine of Alexandria and
Saint Peter, Follower of Sassetta, 1460 / 70, tempera on wood Madonna’s robe. Over her dress the blue mantle
panel, 19 × 16 1/2 in. (48.2 × 41.9 cm) including molding. Spencer covers her head and cascades from a clasp at the
Museum of Art, The University of Kansas, Lawrence, KS. neck. In the pinnacle is a figure of Christ Blessing.
An attenuated figure of Saint John the Baptist
occupies the right wing with the Virgin of the
Paso triptychs are clearly by the same hand, Annunciation, seated on the ground, in the gable
especially evident in the lateral wings, it is diffi- above, while on the left is Saint Catherine of
cult to connect either of them to the Vatican Alexandria, in a dress that was once silver leaf,
predellas, which, at least to this viewer, are instead surmounted by the Angel of the Annunciation
stylistically closely related to the painter of the who leans back in a somewhat awkward pose in
portable triptych in the Spencer Museum of Art order to fit into the triangular space. Both saints
in Lawrence, Kansas23 (fig. 7), another in the stand on marbled floors. The moldings are bat-
Siena Pinacoteca, number 158, a Madonna of tered but original, as are the hinges. Each com-
Humility flanked by Saint Catherine on the right partment has three crockets per edge and a finial.
and Saint John the Baptist on the left, attributed There is no base. The wings are well preserved
to an unknown painter working in the middle of though covered with grime and discolored var-
the fifteenth century, and perhaps the Vatican nish. The central panel has suffered: the azurite
Scenes from the Life of the Virgin. The Kansas triptych, robe of the Madonna has many losses, the fore-
presently catalogued as Follower of Sassetta, ground is illegible, and the Child is severely
has been attributed to Pellegrino di Mariano by scarred, while the head of the Madonna, the
Fiocco, Longhi, Suida, Venturi, tentatively by angels, and the Blessing Christ are all well pre-
Berenson, and by Pope-Hennessy to a follower of served as is the gold ground throughout. The
Sassetta who shows some influence of Pellegrino punch work, though simple, is finely executed.
di Mariano. Four different punches are used: a rosette, two
In summation, the El Paso triptych has been circle punches of different sizes, and an odd
attributed unequivocally to the Osservanza Master punch, an imperfectly cut diamond or quatrefoil

shape, slightly longer than it is wide that could
perhaps best be described as an irregular man-
dorla. The garments of the angels are sgrafittoed,
making a pattern with the underlying gold ground.
Unfortunately it was not possible to see the back
of the doors.
2. The Chigi-Saracini Triptych (fig. 9)

The central panel depicts the Madonna seated
in a three-quarter pose similar to that of the
Siena triptych 177, but her head, with elaborately
braided hair, is uncovered, and she holds a leaping
Child against a white cloth. Four angels stand
slightly behind her in attitudes of prayer, while in
the cusp, a figure of God the Father emerges from
a starry firmament. The floor is covered with a
carpet of oriental design. John the Baptist, nearly
a twin to the figure in the Siena Pinacoteca, and a
Virgin of the Annunciation above, seated on a
bench but otherwise similar to number 177, occupy Fig. 8. Madonna and Child with Saint Catherine of Alexandria
and Saint John the Baptist, No. , Master of the Osservanza,
the right wing. On the left, a monastic saint, later tempera on wood panel, 21 7/8 × 18 1/4 in. (55.5 × 46.5 cm).
replaced by a figure of Saint Catherine, was once La Pinacoteca Nazionale di Siena, Siena.
depicted below the Angel of the Annunciation,
a different model this time, slightly smaller in
scale, with elegantly designed raised wings, better
adapted to the space. Both saints stand on mar-
bled floors. As expected, the azurite robe of the
Madonna has suffered, and there are numerous
scattered losses to the foreground of the central
panel making it barely legible; of the monk in
the left wing only traces remain. The painting has
recently been cleaned and restored. The carpentry
is identical to the Siena triptych and, of the five,
this is the only one which retains its original
base. Apart from some flaked losses in the central
panel, the gold ground is well preserved, the
punch work finely executed and only slightly
more elaborate than number 177. The same four
punches are used in a different variation and there
is a fifth punch, a small star (also used by Sano
di Pietro, see below), identical to one used in
the Saint Anthony series along the top of the
uppermost panel and which also occurs in the
Fig. 9. Madonna and Child with Saint Francis and Saint John the
Asciano Altarpiece. The reverse of the doors is Baptist, Master of the Osservanza, tempera on wood panel,
faux porphyry with a green earth border. 25 × 21 5/8 in. (63.5 × 55 cm). Collezione Chigi-Saracini, Siena.

3. Pienza Triptych (fig. 10)
A Madonna of Humility seated on a beautifully
preserved oriental carpet occupies the central
panel. Again she is seen in three-quarter view,
but facing right rather than left. A Christ Blessing
surrounded by half figures of angels and proph-
ets hovers above. The right wing depicts Saint
Catherine of Alexandria dressed in an elaborately
worked gown of sgraffitoed and incised gold and
painted decoration. Saint John the Baptist is on
the left, facing left and in a slightly different pose
than in the other two triptychs. The Angel of
the Annunciation is similar to that of the Chigi-
Saracini Altarpiece, while the Virgin is of a differ-
ent type than in the other two paintings, more
elegant with finer garb. Again, the carpentry is
identical. The gold ground is well preserved, and
we observe the same punches: the rosette, the
irregularly cut mandorla, and two circle punches
Fig. 10. Madonna and Child with Saint Catherine of Alexandria of different sizes. Of the three described thus far,
and Saint John the Baptist, Pienza, tempera on wood panel, this painting is the best preserved. The reverse of
25 1/4 × 23 1/4 in. (64 × 59 cm). Museo Diocesano, Pienza.
the wings are decorated with faux porphyry with
an elegant central geometric design (fig. 11). The
base is modern.
4. The Hague Triptych (fig. 12)

In the central compartment the Madonna is
posed in three-quarter view on a draped throne
facing right. She is flanked by Saint Catherine of
Alexandria on the left and a female saint on the
right, possibly Saint Lucy.24 Two angels stand in
prayer behind the throne and a figure of God the
Father is in the gable above. The Christ Child
holds a scroll with the word “ego.” A carpet of
similar design to that in the Pienza triptych covers
the floor. Saint Ansanus, one of Siena’s patron
saints, stands in the left wing holding the balzana
in his right hand. Saint Lawrence with his grill
occupies the right wing, and the Angel and Virgin
of the Annunciation are in their customary posi-
tions in the left and right gables, respectively, of
the wings. The angel is the same figure used in the
Pienza triptych, while the Virgin is rather solemn,
seated on a bench in a simple painted interior
Fig. 11. Madonna and Child with Saint Catherine of Alexandria suggesting a loggia. The carpentry is identical to
and Saint John the Baptist (fig. 10), reverse of side panels. the other three triptychs. The backs of the wings

the island in the lake similar to those depicted
in the Lehman Saint Anthony with a slight resem-
blance to the hermitage itself.25 In the foreground
the Madonna sits in front of the manger with
Saint Joseph behind and to her right. Over the
newborn Child, Christ’s “xp” monogram hovers
accompanied by six cherubim with the Holy
Spirit below. The ox and ass are represented on
the far left, and a saddle and a double pouch of
linen decorated with red embroidery rest on the
ground. Three shepherds stand to the right of
Saint Joseph, accompanied by a small white
sheepdog. At the extreme right behind the shep-
herds is a city gate. On the hillside beyond, on a
smaller scale, the same three shepherds and their
sleeping dog gather around the comfort of a fire
fashioned of painted and sgrafittoed gold leaf,
gazing towards the heavens at an angel in a golden
nimbus wrought, like the fire, in gold leaf, painted
and sgrafittoed. Black and white sheep occupy a
Fig. 12. Madonna and Child with Saint Catherine of Alexandria
and a Female Saint, Attributed to the Master of the Osservanza, pen enclosed by white lattice. A crane looms large
tempera on wood panel, 24 × 20 7/8 in. (61 × 53 cm). Museum in contrast to the small trees. A red bridge crosses
Meermanno-Westreenianum, The Hague. a stream, and the dark blue sky is streaked with
white clouds over the distant hills. It is a compo-
sition of infinite refinement and is the scene that
are painted to resemble porphyry, and each has a so impressed Carli and Torriti. Sano di Pietro
painted shell in the center, which may refer to the later borrowed it for his much larger and coarser
Sienese district, the Contrada di Nicchio. Two holes panel of around 1450 in the Pinacoteca of Siena
bored into the wood of the central panel suggest (see fig. 5).26
dowels for a base such as that still retained by the The upper compartment is unique to this
Chigi-Saracini triptych. The gold background is triptych, otherwise the carpentry of the moldings
well preserved. None of the costumes displays the and the crockets are the same as in the other four.
elaborate Sassettesque sgrafitto and incising of This unusual element depicts a miniature Last
the gold that are an important feature in the other Judgment complete with trumpeting angels, souls
triptychs. The punches used are also somewhat rising from their graves, and vicious devils prod-
different: a Gothic arch motif along the borders, ding the damned into their ghastly realm. The
one circle punch, and the familiar rosette. The figure of Christ would seem to derive from
punching is slightly irregular in places and sug- Sassetta’s Last Supper, part of the dismembered
gests an unsure hand. Arte della Lana altarpiece.
The pose of the Virgin of the Annunciation
5. The El Paso Triptych (see figs. 1, 2, 3, and 4) (fig. 13) is a direct quote—with the exception
The subject of the central panel is a complex of the position of the hands—from the Pala
scene representing the Adoration of the Shep- dell’Osservanza (fig. 14), but set into rose and white
herds and the Annunciation to the Shepherds set marble domestic architecture with a plaid bed
in a fully developed landscape that recalls the one cover and a glimpse of a garden in the distance.
around San Leonardo al Lago, the buildings on The angel is a new variation from those in the

Fig. 13. Adoration of the Shepherds with Saint John the Baptist and Saint Fig. 14. Pala dell’Osservanza, Master of the Osservanza, tempera
Bartholomew (fig. 1), detail of the Virgin Annunciate. on wood panel. Church of the Osservanza, Siena, Italy. Detail
of the Virgin Annunciate.
other triptychs and occupies the ample space with The facial types in the Hague triptych are more
great conviction. The robes are elaborately worked closely related to these two triptychs, although
in sgrafittoed brocade; the pricked fabric, which not as finely executed. The Pienza and the El Paso
also comes from Sassetta, and a charming detail triptychs are clearly related stylistically and tech-
of a gilded vase with flowers fills an empty corner. nically and admittedly are populated by more
As already stated, old dowel holes at the bot- swarthy figures dominated by a dark verdaccio
tom indicate a lost base. The backs of the doors underpaint in the flesh tones, and executed with
are painted in fictive porphyry with central medal- coarse brushwork, at times summarily applied.
lions containing small sketches of the Mater This style can also be observed in some of the
Dolorosa and the Crucified Christ. When closed predellas of the Passion series, particularly in the
the gable with its scene of the Last Judgment Vatican Flagellation, and in the panels of the Saint
projects above the doors. Anthony series. The figures in the Death of the
There are technical and stylistic differences Saint, Saint Anthony Distributing his Wealth to the Poor
among the five portable triptychs. The two in (National Gallery of Art, Washington) (fig. 15)
Siena, in Palazzo Chigi-Saracini and in the and Saint Anthony Tempted by the Devil in the Guise of a
Pinacoteca, are somewhat more refined, with Woman (Yale University Art Gallery) are all close
minute and carefully integrated brushwork, pale cousins of the shepherds (fig. 16) in the El Paso
flesh tones, and sweet facial types resembling triptych and are not painted with more refine-
Sassetta’s early work. This style also characterizes ment. Nor would Saint Anthony Tempted by Gold find
the predella of the Osservanza Altarpiece and the himself out of place in the El Paso painting, and,
smaller scenes of the Asciano Birth of the Virgin. as has already been pointed out, the landscape,

Fig. 15. Saint Anthony Distributing his Wealth to the Poor, Master of Fig. 16. Adoration of the Shepherds with Saint John the Baptist and Saint
the Osservanza, tempera on wood panel, 18 5/8 × 13 5/8 in. (47.5 × Bartholomew (fig. 1), detail.
34.5 cm). National Gallery of Art, Washington, D.C. Detail.
technically and in the poetry of its invention, is

closely related to the Lehman panel. The water
(figs. 17 and 18) for example, in both panels is a
dark brownish green that might be mistaken for
earth were it not for the bridge and the boat.
Close examination shows it to have a pale blue
underpaint and what would have been a trans-
lucent bright green copper resinate glaze, now
altered. Because it is painted in a different med-
ium, the water in the Cambridge miniature of the
Burial of Saint Monica and Saint Augustine Departing for
Fig. 17. Saint Anthony Tempted by Gold (fig. 6), detail of San
Africa gives some idea of the original appearance Leonardo al Lago.
of similar passages in the other two panels.
The Berlin Saint Anthony at Mass is superior in
perspective and execution to the other panels of
the Saint Anthony series, and as Pope-Hennessy
observed, very close to Sassetta himself, with
its sophisticated lost profiles and the delicate
tempera hatching in the lapis lazuli drapes of
the officiating priest. Comparing the awkward,
reworked, and unsuccessfully rendered tile floor
in the Death of the Virgin (Villa I Tatti), a compart-
ment of the Asciano Altarpiece, with the perfect
realization of the patterned floor in the Berlin
painting, it is hard to believe that the same painter Fig. 18. Adoration of the Shepherds with Saint John the Baptist and Saint
was responsible for both works. Bartholomew (fig. 1), detail of San Leonardo al Lago.

It should also be noted that the different states and has been cut unevenly in such a way that it
of the various panels of the Saint Anthony series is easily recognizable; this odd shape is difficult
do not allow for comparison on the basis of their to accurately classify but can be described as an
palette. Three of the Washington panels are cov- irregularly cut mandorla. It is present not only in
ered with thick coatings of yellowed dammar and these four triptychs but also in the predella of
shellac, severely distorting the original colors, the Pala dell’Osservanza and is used as an outline
which are both cooler and more brilliant. This for the nimbus of the Madonna and Child with
is true to a slightly lesser degree of the Death of Four Angels atop the Asciano Altarpiece. The star
Saint Anthony, which has a thinner varnish that has punch observed in the Saint Anthony series and
significantly discolored since it was applied in the the Hague triptych is also used in Asciano. Keith
mid-1940s when the painting appeared on the art Christiansen has noted that:
market. The other panels of the series have all
been cleaned in the recent past. a number of the punches employed in tooling
the gold in the Osservanza altarpiece and in the
The punch work may have some significance. Birth of the Virgin in Asciano recur in the Saint
As already described, the painter of four of the George, the Lehman Madonna and Child, and in
portable triptychs—those in the Pinacoteca of Sano’s signed and dated Gesuati altarpiece in the
Siena (no. 177), the Chigi-Saracini Collection, the Siena Pinacoteca.27
Diocesan Museum in Pienza, and the El Paso
Museum of Art—used the same four punches Both the star punch and the rosette used in
(figs. 19 and 20). Three of them are common Asciano and in the Saint Anthony cycle can be
designs—the rosette and the two different-sized found in other works by Sano di Pietro as late as
circles. The fourth punch is an unusual shape the 1447 Madonna and Child with Saints Bartholomew
and Lucy, the 1449 polyptych of Scrofiano where
they decorate the mitre and gilded border of
Saint Biagio in the predella. Strangely, for what
it is worth, in the Santa Bonda altarpiece dated
between 1450 and 1452 the original rosette punch
is used in the predella scenes of the Last Supper
and the angels, while a slightly different rosette
punch, larger and more accurately cut, is used in
some of the other predella scenes. This larger
rosette is also used in the altarpiece of Saints
Cosmos and Damian from the convent of the
Gesuati of San Girolamo.
The irregular mandorla punch seems to be
used only in those works that are associated with
the Osservanza Master, and I have not seen it
in any paintings done independently by Sano
di Pietro after 1444. However, a similar punch,
again, like Sano’s new rosette, larger and more
regularly cut, can be found in several smaller
paintings by Sano di Pietro, such as The Cruci-
fixion in Washington.
Another technical feature of some of the small-
Fig. 19. Adoration of the Shepherds with Saint John the Baptist and Saint scale works attributed to the Osservanza Master
Bartholomew (fig. 1), detail of punches. is the presence of significant pentimenti, unusual

flesh tones; it has been removed from the more
prominent areas. The azurite robe of the Madonna
has been scraped down and some of the painter’s
corrections, especially around the ox in the fore-
ground, have flaked off or have been scraped away
in the past. The gold leaf is worn in places: the
haloes of Saint John, the Madonna and Saint
Joseph and, sadly, part of the golden nimbus
Fig. 20. Adoration of the Shepherds with Saint John the Baptist and Saint
Bartholomew (fig. 1), detail of punches. radiating from the Child, which is also incised
and painted. The head of Christ in the medallion
on the back of the right door is missing, and the
backs of the doors are generally a bit battered.
in this period. In The Way to Calvary (Philadelphia And, as mentioned earlier, the original base is
Museum of Art), the building on the right has gone. However, much of the altarpiece is in lovely
been painted over a mountain. In The Death of the condition, and the decision was made to restore
Virgin (I Tatti), the landscape and fence in the it as completely as possible, including replacing
background have been moved. In the Pienza the missing gold with new leaf on a wax mordant.
triptych there is a pentiment of the robe of the The reverse of the wings were also restored, with
Baptist, as there is in the El Paso triptych, which the exception of the missing head of Christ, and
also has a change in the position of the finger of new hook fasteners, made in imitation of antique
the right hand (fig. 21) and various other correc- ones, were inserted to replace the unsuitable mod-
tions of contours. ern brass hinges.28 The central panel had a typical
As far as the state of the El Paso triptych is Kress cradle but had not been thinned. This un-
concerned, while it is quite well preserved for necessary secondary support was removed.
a painting of this period, it has been slightly Whatever conclusions may be drawn about
abraded in the past. An old varnish or patina its authorship, the publication of good color
can be seen in between some of the thickly clot- images of this long neglected work will bring
ted paint of the brushwork, especially in the it the attention it merits and allow scholars to
compare the five triptychs.

Notes
1. The extensive Suida-Manning collection is now in the Jack
Blanton Museum at the University of Texas, Austin.
2. Part of the Coronation of the Virgin altarpiece in the Robert
Lehman Collection at the Metropolitan Museum of Art.
3. According to Shapley (1966), the portable altarpiece
should be dated probably after 1440. It has been attrib-
uted to Sassetta (Siren 1917, Van Marle 1927); follower
of Sassetta, the Vatican Master, or Vico di Luca (Pope-
Hennessy 1956); early Giovanni di Paolo (Berenson 1932);
Master of Pienza (Carli 1957 and Zeri 1954; the latter iden-
Fig. 21. Adoration of the Shepherds with Saint John the Baptist and Saint tifies this painter as the Pseudo Pellegrino di Mariano);
Bartholomew (fig. 1), detail of pentimento. Sano di Pietro (Volpe 1958, comparing it to an Assumption

in the Siena Pinacoteca; and Berenson in 1947, verbally, 22. Christiansen et al. (1988), p. 107
to Wildenstein); Master of the Osservanza (Alessi and 23. Shapley (1966). k-444, Follower of Sassetta, Madonna and
Scapecchi 1985); Master of Pienza (Christiansen 1988). Child with Saints, portable triptych, 19 × 16 1/2 in. (48.2 ×
4. Harold Irwin Pratt, Brooklyn, 1877 – New York, 1939. 41.9 cm), including the molding. Very good condition.
The youngest son of Charles Pratt (1830–1891), John D. Frame regilded. Given in the past to Pellegrino di Mariano
Rockefeller’s partner in Standard Oil of New Jersey, (in ms) by Fiocco, Longhi, Suida, and Venturi; to
founder of the Pratt Institute. Harold I. Pratt built and Pellegrino tentatively by Berenson; school of Sassetta by
lived in a mansion on the southwest corner of Park Avenue Perkins. Pope-Hennessy (1939) to follower of Sassetta who
and 68th Street donated in 1944 by his widow to the had worked in Sassetta’s bottega and shows some influence
Council on Foreign Relations, of which Pratt had been a of Pellegrino di Mariano. Provenance: Achillito Chiesa,
member since 1923. The Council converted the residence Milan, Alessandro Contini-Bonacossi, Kress, 1936.
into offices, meeting rooms, and a library. Presumably the Exhibited at the National Gallery of Art, Washington,
collection was acquired by Wildenstein around the time of D.C. from 1941–52 (fig. 8).
Mrs. Pratt’s gift, and three paintings were bought by the 24. Van Os et al. (1989), p. 127.
Kress Foundation: the Sienese triptych, a Madonna and Child 25. The hermitage of San Leonardo al Lago is recorded as
by Andrea Solario now in Columbia, South Carolina, and early as 1119, and a church existed by 1168. Eventually the
the Piero di Cosimo in Honolulu. church was joined to the Augustinian Order. The Beato
5. Shapley (1966), p. 147. Agostino Novello retired to the hermitage and became
6. Ibid. a cult figure after his death in 1309, and the church pros-
7. The attribution to Sano di Pietro was likely made by pered. Around 1350 it was rebuilt in the Gothic style,
Bernard Berenson since he was Wildenstein’s expert in frescoed by Lippo Vanni in 1370 and decorated with
that period. other images. In 1366 the hermitage offered refuge to the
8. That the problem is still controversial one only has to population of nearby Santa Colomba during a siege and
read Kanter’s note in his review of the London National was fortified with a wall and two towers, one round and
Gallery’s new catalogue of fifteenth-century Italian paint- one square in plan. Giovanni di Paolo painted a Cruci-
ings in the February 2004 issue of The Burlington Magazine, fixion for the chapter room, his only fresco. During the
p. 107: “ng 5114: This panel of the Birth of the Virgin (fig. 34) fourteenth and fifteenth centuries, it was accorded particu-
is by Sano di Pietro, not the Osservanza Master. lar reverence as attested by the 1460 papal visit of Pio ii.
Confusion between these two painters still allows for their It was once surrounded by water, thus the reference to a
identity as a single hand to be entertained as a serious pos- lake in its name. Now drained, it stands on a hilltop in
sibility, but is ruled out by their having worked side by the middle of fields of grazing sheep.
side and in distinctive styles and techniques on the St. 26. The hilltop scene of the Annunciation to the Shepherds meas-
Anthony Abbot series.” In sharp disagreement, Boscovits ures 54 × 69 cm.
(2003) attributes all four panels from that same series in 27. Christiansen et al. (1988).
the National Gallery of Art, Washington to Sano di 28. Giovanni Marussich replaced the hinges and removed the
Pietro. cradle.
9. Longhi (1975), p. 60.
10. Turner (1988), pp. 738–40.
11. Kress Acquisitions, –, National Gallery of Art, Bibliography
Washington; Fern Rusk Shapley, The Samuel H. Kress Alessi, C. and Scapecchi, P. (1985) “Il Maestro dell’Osservanza:
Collection, El Paso Museum of Art, 1961, cat. no. 5; Shapley Sano di Pietro o Francesco di Bartolomeo,” Prospettiva, No.
(1966). All three illustrations are mediocre quality black 42 (July), pp. 13–37.
and white overall shots. Angelini, A. (1986) Sassetta e I Pittori Toscani tra XIII e XV Secolo
12. Invoice from Robert M. Kulicke, Inc., Framemakers, Sep- (exhib. cat.). Siena: Pinacoteca of Siena, Chigi-Saracini
tember 30, 1955, where, interestingly, k-1434 is referred to Collection.
as Master of Osservanza. It is likely that Mario Modestini Berenson, B. (1932) Italian Pictures of the Renaissance; a List of the
believed it to be by that artist since he would have been Principal Artists and their Works, with an Index of Places. Oxford:
working directly with Kulicke. Clarendon.
13. Verbal communication based on her observations in front ––– (1946) Sassetta: Un Pittore Senese della Leggenda Francescana.
of the original. Milan: Electa, pp. 51–2.
14. Longhi (1975). Boskovits, M. and Brown, D.A. (2003) Italian Paintings of the
15. Graziani (1948), pp. 75–88. Fifteenth Century. Washington, D.C.: National Gallery of
16. Pope-Hennessy (1956). Art.
17. Carli (1957). Brandi, C. (1946) Introduzione alla Pittura Senese del Primo
18. Ibid., in ‘Nota’ following p. 121. Quattrocento, Rassegna, Italia. Siena: Deputazione Toscana
19. Torriti (1977), p. 276 (author’s translation). di Storia Patria.
20. Alessi and Scapecchi (1985). ––– (1949) Quattrocentisti Senesi. Milan: Ulrica Hoepli [for a
21. The rest of their argument has not found acceptance with review see Pope-Hennessy 1951].
other scholars. It depends on an interpretation of new Carli, E. (1957) Il Sassetta e il Maestro dell’Osservanza. Milan: Aldo
archival material about the two patrons who commissioned Martelli, pp. 89–121.
the Osservanza and Asciano altarpieces. Christiansen, K., Kanter, L. and Strehlke, C.B. (1988) Painting

in Renaissance Siena, – (exhib. cat.). New York: Graziani, A. (1948) “Il Maestro dell’Osservanza,” Proporzioni, ii.
Metropolitan Museum of Art. Florence: Sansoni, pp. 75–88.
Longhi, R. (1940) “Fatti di Masaccio e Masolino,” Critica
d’Arte v. Florence: Sansoni, pp. 145–91.
––– (1975) “Fatti di Masaccio e Masolino,” Opere Complete di
Roberto Longhi, viii. Milan: Sansoni.
Loseries, W. (1987) “La Pala di San Giorgio nella Chiesa di
San Cristoforo a Siena: Sano di Pietro o il ‘Maestro
dell’Osservanza’?” Prospettiva, No. 49 (April), pp. 61–74.
Panders, J. (1997) “Some examples of fifteenth-century Sienese
underdrawing” in Van Asperen de Boer (ed.), Album
Discipulorum. Zwolle: Waanders Uitgevers.
Pope-Hennessy, J. (1939) Sassetta. London: Chatto and Windus.
––– (1951) “Review of Quattrocentisti Senesi by C. Brandi,” Art
Bulletin xxxi-ii, pp. 141–3.
––– (1956) “Rethinking Sassetta,” The Burlington Magazine,
xcviii, pp. 364–70.
––– (1987) Italian Paintings in the Robert Lehman Collection.
Princeton, NJ: Metropolitan Museum of Art.
Shapley, Fern Rusk (1966) Paintings from the Samuel H. Kress
Collection, Vol. : Italian Schools, XIII–XV Century. London:
Phaidon.
Siren, O. (1917) Loan Exhibition of Italian Primitives (exhib. cat.).
New York: Kleinberger Galleries.
Torriti, P. (1977) La Pinacoteca Nazionale di Siena: I Dipinti dal XII
al XV Secolo. Genoa: Sagep Editrice.
Turner, J. (ed.) (1998) The Dictionary of Art, Vol. 20. New York:
Grove Press.
Van Os et al. (1989) The Early Sienese Paintings in Holland.
Florence: Centro Di.
Van Marle, R. (1927) The Development of the Italian Schools of
Painting. The Hague: M. Nijhoff.
Volpe, C. (1958) Arte Antica e Moderna. Bologna: Zanichelli.
Zeri, F. (1954) “Il Maestro dell’Osservanza: una crocefissione,”
Paragone, Vol. 49, pp. 43–4.
Photography Credits
Figs. 1, 2, 3, 4, 13, 16, 18, 19, 20, and 21, pp. 112, 114, 122, 123,
124, and 125. El Paso Museum of Art, El Paso, TX,
Samuel H. Kress Collection (1961-6/ 6).
Figs. 6 and 17, pp. 117 and 123. Photograph ©The
Metropolitan Museum of Art, New York, NY.
Fig. 7, p. 118. Spencer Museum of Art, University of Kansas,
Lawrence, KS, Samuel H. Kress Study Collection
(1960.0045).
Fig. 14, p. 122. Photograph by Scala ©Art Resource, Inc.,
New York, NY.
Fig. 15, p. 123. ©National Gallery of Art, Washington, D.C.
1

Scenes from the Life of Virgin, ca. 1480–90
Guidoccio Cozzarelli (Guidoccio di Giovanni di Marco Cozzarelli)
Egg tempera on cradled wood panel
26 3/4 × 21 1/4 in. (68 × 54 cm)
Lowe Art Museum, University of Miami, Coral Gables, Florida
61.022.000 (k-1286)
Fig. 1. Scenes from the Life of the Virgin, before cleaning and
restoration.
Fig. 2. Scenes from the Life of the Virgin (fig. 1), after cleaning
and restoration.

Guidoccio Cozzarelli’s Scenes from the
Life of the Virgin
T his unusual composition was exhibited in “Painting in Renaissance

Siena” at the Metropolitan Museum of Art in 1988; on that occasion a
number of intriguing questions were raised about its original context. In
1990 the painting was cleaned and restored at the Conservation Center in
New York (figs. 1 and 2). As Laurence Kanter noted in the exhibition
catalogue, the grain of the wood is vertical, so the painting could not have
belonged to a predella but was most likely part of an altarpiece although no
related fragments have as yet come to light.1 The restoration did not yield any
information about its position in the altarpiece as all four edges have been cut
and narrow wood strips were added to all sides. The fragment is made up of
two planks, that on the right measuring approximately 30 cm, a reasonable size
for a full board. Taking into consideration the elements of its composition,
it could be posited that it belonged on the right edge of an altarpiece.
The painting is generally slightly worn; certain passages, especially the
details of the brickwork in the townscape, were freely reinforced during an
old restoration.2 Despite the wear, the fragment is essentially in good con-
dition for a work of this period, and there can be no question that the fore-
ground architecture is of a piece with the rest. In his catalogue entry Kanter
proposed that it was a modern restoration disguising the original appearance
as it was recorded in a photograph of the painting while it belonged to
R. Langton Douglas.3 As Shapley notes in the Kress catalogue, “the stylized
dolphin and vase border at the bottom is unusual in a panel painting but

would be normal in a miniature of the period.
Close parallels are offered in some of Cozzarelli’s
illuminations of 1480–81, now in the Piccolomini
Library.”4 This set of choir books (under discus-
sion are three separate books containing five
illustrations by Cozzarelli) was originally com-
missioned for the Ospedale della Scala from
Guidoccio Cozzarelli and Bernardino di Michele
Cignoni and includes Antiphonaries 6f, 15q and
26r with five miniatures by Cozzarelli illustrating
the story of Isaac and Jacob, the Assumption, the Birth
of Saint John the Baptist, the Nativity of the Virgin and A
Bishop Leading a Procession into a Church.5 (Antipho-
nary 7g incorporates a signed illustration of Moses
and the Burning Bush by Cignoni whose work con-
sistently demonstrates a hand less refined than
Cozzarelli’s and a predilection for a different set
of decorative motifs.6) The border design of dol- Fig. 3. Isaac Blessing Jacob, Guidoccio Cozzarelli, tempera on
phins as well as the central column terminating parchment, Biblioteca Piccolomini, Siena.
in a dolphin frieze in the page Isaac Blessing Jacob
(fig. 3) are related to the architectural details in
our painting as are similar motifs decorating the A curious anomaly in Scenes from the Life of the
border of the Bishop Leading a Procession into a Church Virgin is the relationship of the existing architecture
(fig. 4), while similar treatment of the architecture to a larger composition. It has been suggested,
is seen in two cassone panels by Cozzarelli, The presumably by comparison with Cozzarelli’s
Return of Ulysses in the Musée de Cluny and The Pannilini altarpiece Madonna and Child with Saints
Legend of Cloelia in the Metropolitan Museum. Simon and Thaddeus in San Bernardino in Sinalunga8
Kanter suggested that the Coral Gables frag- where the Madonna appears seated on a carved
ment and The Birth of the Virgin, illustrated by van throne placed in front of a dolphin-decorated
Marle while it was in the collection of the Baron frieze, that the monumental frieze and pilaster in
Michele Lazzaroni (obviously much repainted by the Coral Gables painting formed part of a simi-
Verzetta, Lazzaroni’s restorer,7 and later offered for lar throne.9 This enticing hypothesis is difficult
sale by the Galleria Gilberto Algranti, May 5 – 30, to sustain since the pilaster sits directly on top of
1971) might be part of the same composition. the frieze and in the same plane, while the frieze
Kanter’s notion is appealing not only because of must continue to complete the dolphin-vase-
the markedly similar figure types but also because dolphin motif. It is impossible to imagine how
of the similar treatment of the incised halo of these two elements could become the arms of
Saint Joseph in our picture and that of Saint a presumed throne and would exclude standing
Elizabeth in the ex-Lazzaroni painting. If however, saints whose heads would project above the frieze
as recorded, the Lazzaroni panel measures 67 × as in the Sinalunga Altarpiece. With the existing
58 cm, then the Coral Gables painting’s dimen- information it is only possible to conjecture how
sions, 68 × 54 cm, make this association problem- these architectural elements developed in the lost
atic since there is a notable difference in the scale altarpiece, but it is more likely that they formed
of the figures—nearly twice as large in the paint- part of an architectural background, a palace or
ing formerly in the Lazzaroni Collection in which urban scene. This would imply that the principal
the figures occupy most of the composition. subject of the altarpiece was not a Madonna and

Notes
1. K. Christiansen, L. Kanter and C.B. Strehlke, Painting
in Renaissance Siena, – (exhib. cat.). New York:
Metropolitan Museum of Art, 1988, pp. 283–5.
2. The provenance is from Fern Rusk Shapley, Paintings from
the Samuel H. Kress Collection, Vol. : Italian Schools, XIII–XV
Century (London, Phaidon, 1966), p. 159. The supposition
about the date of the removal of the restoration is my
own. Dr. Friedrich Lippmann, Berlin (sold, Rudolph
Lepke, Berlin, Nov. 26–27, 1912, catalogue by M.J.
Friedländer, No. 35, as Domenico Cozzarelli, bought by
Kleinberger); R.L. Douglas, London (purchased, 1922)—
exhibited: “Antiquities and Works of Art,” Olympia,
London, July 19 – August 1, 1928, no. x-21, as Guidoccio
Cozzarelli; London market; Contini-Bonacossi, Florence;
Kress acquisition, 1939. The misleading classical ruin must
have been removed during a restoration done between 1928
and 1939, possibly in Florence by one of Contini-Bonacossi’s
restorers, before its purchase by Samuel H. Kress.
3. Exhibition Olympia, London, July 19 – August 1, 1928,
no. x-21.
4. Shapley 1966 (cited in note 2).
5. Raimond Van Marle, The Development of the Italian Schools
Fig. 4. Bishop Leading a Procession into a Church, Guidoccio of Painting, Vol. xvi. The Hague: Martinus Nijhoff, 1937,
Cozzarelli, tempera on parchment, Biblioteca Piccolomini, Siena. pp. 368–9.
6. Mario Cignoni, Bernardino Cignoni, Maestro Miniatore del
Rinascimento (Casole d’Elsa /– Siena ). Florence:
Studio per Edizioni Scelte, 1996.
Child but another episode connected with the life 7. Van Marle 1937 (cited in note 5), p. 365.
of the Virgin, and not a Nativity or Adoration of 8. Ludwig Pardekooper, “Due familigie rivali e due pale di
Guidoccio Cozzarelli per Sinalunga,” Prospettiva, No. 72
the Magi which are depicted in rustic settings. (October 1993), pp. 51–65.
A curious aspect of the painting’s technique 9. Shapley 1966 (cited in note 2).
is worth noting. Both Cozzarelli and his master, 10. The Crucifixion, The Carrying of the Cross, and The Deposition
(three paintings) attributed to Guidoccio Cozzarelli by
Matteo di Giovanni, often used the method of Cesare Brandi in 1931, are painted with white and yellow
making a complete underpainting in grisaille and earth monochrome. The background, as Brandi points
finishing the painting by glazing over the mono- out, is a red similar to a bole preparation, and the only
other color is the red blood flowing from Christ’s wounds.
chrome rendering with washes of semi-transparent Originally published in Daedalo, and reprinted in Lecceto
color. The gray appearance of the heads in many e gli Eremi Agostiniani in Terra di Siena (Monte dei Paschi di
of their works is due to the removal in a past Siena, 1990), pp. 315–27.
cleaning of the delicate pink final layer. In the
Coral Gables fragment the final glaze had been Photography Credits
Figs. 1 and 2, p. 128. Lowe Art Museum, University of
much abraded, and an attempt to suggest it was Miami, Coral Gables, FL, Samuel H. Kress Collection
made during the restoration. The worn affreschi in (61.022.000).
the former dining hall or cenacolo of Sant’Agostino 1
in Monticiano attributed to Cozzarelli by Cesare
Brandi in 1931 demonstrate the artist’s affinity for
monochrome painting.10


Saint Sebastian
School of Pietro Perugino, ca. 1500
Oil and egg tempera on wood panel, transferred to cradled Masonite panel
29 7/8 × 20 3/4 in. (75.9 × 52.7 cm)
Princeton University Art Museum, Princeton, New Jersey
(k-1557)
Fig. 1. Saint Sebastian, before cleaning and restoration.
Fig. 2. Saint Sebastian (fig. 1), after cleaning and restoration.

School of Pietro Perugino, Saint Sebastian
Annette Rupprecht & Sheri Francis Shaneyfelt
T he Kress Collection Saint Sebastian (figs. 1 and 2) is clearly executed in

the manner of Pietro di Cristoforo Vannucci (Città della Pieve ca. 1450–
Fontignano 1523), called Pietro Perugino, since for many years he lived and
worked in the city of Perugia.1 The painting has a distinguished provenance.
According to an inscription formerly on the verso it was commissioned
by the Oddi family of Perugia:
questo s. sebastiano è stato da raffaelo sanzio da urbino
dipinto per i signori conti degli oddi perugia. i.a.d.s.p.
By 1847 it was in the Edward Solly Collection, London, then bought at
Christie’s, London, by Lord Northwick, Thirlestane House, Cheltenham,
Gloucestershire; in 1873 it was sold through Sir J. Charles Robinson and
entered the Cook Collection, Richmond, Surrey from which it was acquired
after 1945 by the Florentine dealer and collector Count Alessandro Contini-
Bonacossi who sold it to Samuel H. Kress in 1948.
In 1994 the Saint Sebastian was requested for loan by the Grand Rapids Art
Museum for its upcoming exhibition Pietro Perugino: Master of the Italian Renais-
sance 2 and the work was sent to the Conservation Center of the Institute of
Fine Arts of New York University for examination and treatment. Although
basically in good condition, the painting had minor flaking requiring consoli-
dation, and the varnish was thick and significantly discolored. While the treat-
ment itself proved straightforward, the technical examination of the painting
Annette Rupprecht & Sheri Francis Shaneyfelt 133

prior to its treatment and research at the Kress
Archive into its restoration history revealed infor-
mation that has reopened the discussion of the
painting’s authorship.
At the time of its acquisition from the Contini-
Bonacossi Collection in 1948, this painting was
still on its original support, a wood panel with
vertically oriented grain that subsequently had
been cradled.3 According to Mario Modestini,
exposure to excessive humidity during the summer
had caused the paint to lift and it was necessary
to transfer the paint layer because the glue binder
had denatured, causing the gesso preparation to
disintegrate to powder.4 It was transferred from
wood panel to a plain, closed-weave canvas5 and
marouflaged onto a single piece of Masonite,
approximately 1/8 in. larger than the image in all
four directions. The Masonite was then veneered
and cradled.
The transfer procedure provided a rare oppor-
tunity to view and document the reverse of the Fig. 3. Saint Sebastian (fig. 1), photograph of the
reverse of the painting during the 1950 transfer,
paint layer; it clearly revealed an underdrawing of digitally enhanced to accentuate the incised lines.
lines incised into the original gesso ground laying
out the initial contours of the figure. Photographs
recording the incised lines were taken during the
treatment. Although the original negatives and
photographic prints were missing, we were able
to make use of photocopies in the Kress Archives,
including an overall image and a detail of the face.
Since then, eight of the original negatives have
been located in the Department of Image Col-
lections at the National Gallery of Art (fig. 3).6
The presence of such incisions indicates that the
Sebastian figure was perhaps derived from another
image of the saint, possibly from a work by the
master Pietro Perugino himself, or from a picture
originating in his workshop. Indeed, the use of
cartoons in the repetition of particular figures
and even entire compositions by Perugino and his
numerous collaborators is well accepted by schol-
ars of the Perugino School.7
When a diagram of the incisions was reversed
and superimposed onto the painted image of
the Kress Saint Sebastian, it was apparent that the
saint’s hips had been considerably shifted to the Fig. 4. Saint Sebastian (fig. 1), diagram of incised lines
right between the drawing stage and the final (fig. 3), reversed and superimposed onto the painting.

painting (fig. 4). In addition, the original incised
drawing indicated a loincloth with folds, similar
to Perugino’s full-length Saint Sebastian in the
Musée du Louvre, Paris, of circa 1490 (fig. 5),
and the closely related figure of the saint in his
Madonna and Child Enthroned with Saints John the Baptist
and Sebastian, in the Galleria degli Uffizi, Florence,
of 1493 (fig. 6).8 In the Kress painting, a smooth
violet ribbon has replaced the original draped
loincloth seen in Perugino’s pictures. Furthermore,
the fingertips of the Kress figure’s proper right
hand had been incised and, like the Saint
Sebastians in Perugino’s works, the fingertips were
rendered in the painted composition. The Louvre
representation of the saint closely resembles the
Kress Saint Sebastian, particularly in the construc-
tion and posture of the figure, the upward tilt of
Fig. 6. Madonna and Child Enthroned with Saints John the Baptist
and Sebastian, Pietro Perugino, 1493, oil and egg tempera on
panel, 70 1/8 × 64 5/8 in. (178 × 164 cm). Galleria degli Uffizi,
Florence, Italy.
the head, and the pious heavenward gaze. The

striking similarity between the Sebastians in the
Louvre and Uffizi compositions suggests that the
figures may have been derived from the same car-
toon. Furthermore, these two figures are certainly
ancestors to the Kress Saint Sebastian, which should
be dated just a few years later, to about 1500.
Striking similarities were also found when the
overlay of incised lines in the Kress image was
superimposed onto a photograph of the full-
length Saint Sebastian depicted in Saint Sebastian
with Two Archers, a painting formerly in the Kaiser-
Friedrich Museum, Berlin (figs. 7 and 8).9 The
incised image of the Kress panel was found to
be virtually identical to the painted image of the
Berlin picture. The only slight difference is that
the fingertips of Saint Sebastian’s proper right
hand are not rendered in the Berlin painting.
Both Saint Sebastians have smooth ribbons across
their loins and are pierced by a single arrow in
their proper right, lower groin. Although the
Fig. 5. Saint Sebastian, Pietro Perugino, ca. 1490, oil on wood, arrows enter in slightly different positions and
69 1/4 × 45 5/8 in. (176 × 116 cm). Musée du Louvre, Paris, France. at different angles, the shafts are similar.

Fig. 7. Saint Sebastian with Two Archers, Master of the Greenville Fig. 8. Saint Sebastian with Two Archers (fig. 7), diagram of incised
Tondo (formerly attributed incorrectly to Giannicola di lines in Saint Sebastian (fig. 2), reversed and superimposed onto
Paolo), formerly Kaiser-Friedrich Museum, Berlin, Germany; the painting.
present location unknown.
The Kress Saint Sebastian and Saint Sebastian with while a single arrow pierces his groin, clearly iden-
Two Archers are clearly related not only to one tifying him as Sebastian. The particular sinuous,
another, but also to Perugino’s two aforemen- graceful stance of the figure, his tilted head and
tioned paintings in the Louvre and Uffizi Gallery. wistful upward gaze, in addition to the apparent
These similar representations of Saint Sebastian lack of pain all recall the manner of Perugino.
indicate the common replication of images within Indeed, Perugino’s compositions are generally
the Perugino workshop and the consequent neces- marked by a balanced, often symmetrical arrange-
sity of keen connoisseurship in the study of ment of figures and forms, crisp, clear illumina-
Perugino School paintings. A close stylistic exami- tion, pastoral landscapes, and an atmosphere of
nation of the Kress picture will provide not only contemplative solemnity, without a display of
a better understanding of its relationship to the overt emotion which would disrupt the tranquil
manner of Perugino, but also its position within stability of the scene.
the Perugino School, and its possible authorship. The most striking aspect of the Kress Saint
In our painting, a half-length Saint Sebastian Sebastian is its dark background, a rich black,
stands before a column to which his hands are uncommon in Peruginesque painting, and perhaps
bound, against a black background which lends a a special request of the patron. Whatever its ori-
dramatic air to the scene. The saint is naked but gin, this feature serves a distinct purpose, drama-
for the violet ribbon which serves as a loincloth, tizing the solitary figure of the saint, elevating it

to the status of an icon. Set off in this manner, career, for according to archival records, his
Sebastian captivates the viewer’s attention, invok- Florentine bottega was in operation for more than
ing reverence and contemplation, appropriate twenty years, from 1487 to 1511, while that in
for the devotional image of a Christian martyr, Perugia is documented from 1501 to 1513.12 The
alluded to by his pierced groin, as well as his early attribution of this painting to Raphael was
nearly naked body, which is presented to the based primarily upon the aforementioned inscrip-
viewer much like the dead Christ in images of tion formerly located on the verso of the work:
the Pietà. Sebastian’s body is revealed and meticu-
questo s. bastiano è stato da raaelo
lously described by the divine light streaming in sanzio da urbino dipinto per i signori
from the left, and in return, this radiance is the conti degli oddi perugia. i.a.d.s.p.
recipient of his pious, tender gaze. Indeed, the
dramatic contrast between the illuminated saint which indicates Raphael as its author, and the
and the obscure darkness from which he emerges noble Oddi family of Perugia as its original
even foreshadows the theatrical tenebrism of the owners.13 This inscription is dated to either the
Baroque master Caravaggio. seventeenth or eighteenth century by the majority
Saint Sebastian was a popular subject during of sources, including Tancred Borenius, Herbert
the Renaissance, not only in his role as a Chris- Cook, W.E. Suida, and Fern Rusk Shapley,
tian martyr, for he was also invoked as a protector whereas Roberto Longhi dated the inscription to
from plague, and was thus commonly depicted the sixteenth century, i.e., closer to the painting’s
together with Saint Roch. Furthermore, Sebastian time of execution. The last five letters of the
provided an opportunity for representing the inscription, “i.a.d.s.p.” are as yet undeciphered.14
standing male nude, which was a challenge of The attribution to Raphael, Perugino’s most
great interest to Renaissance artists, and in this famous pupil, was maintained while the work was
regard, he was perhaps the equivalent of the in the Solly and Northwick collections (until
Classical Apollo. Saint Sebastian was martyred 1873), and was then resurrected by Cook in 1913.
under the Roman Emperor Diocletian in circa Likewise, both Longhi and Suida considered it
300 a.d., and according to his legend, he had been an early work by Raphael while in the studio of
sentenced to execution at the hands of archers, Perugino. The attribution to Giannicola di Paolo
but survived the arrow wounds, and was then was proposed by Borenius in 1913, and followed by
clubbed to death.10 In the Kress picture, Sebastian Umberto Gnoli in 1918, and Raimond van Marle
is pierced by a single arrow, and despite the blood in 1933.15 However, ascribing the painting to either
that flows forth from the wound, he apparently Raphael or Giannicola di Paolo has since lost
suffers no pain. Not only is the lack of pain char- favor in the literature.
acteristic of the Peruginesque style, it perhaps Eusebio da San Giorgio was suggested by
also refers to the triumph of the righteous, here Bernard Berenson as the possible author of the
exemplified by Saint Sebastian, over the wicked, Kress Saint Sebastian, and such a proposition merits
whose presence is implied by the executioner’s further consideration.16 Eusebio was a frequent
arrow.11 assistant to both Perugino and the Perugian
The Kress Saint Sebastian was clearly executed painter Bernardino Pinturicchio (ca. 1456/60–1513)
by a Peruginesque painter, and has been attribu- and collaborated with these masters in the execu-
ted to several artists in this circle, including tion of several important commissions. A number
Raphael Santi (1483–1520), Giannicola di Paolo of scholars attribute the Tezi altarpiece of 1500, in
(ca. 1460/65–1544), Eusebio da San Giorgio the Galleria Nazionale dell’Umbria, Perugia, to
(doc. 1480–1540), and the Master of the Green- the combined efforts of Perugino and Eusebio
ville Tondo. Perugino directed sizeable workshops (fig. 9).17 It depicts the Madonna and Child with
in both Florence and Perugia during his prolific Saints Nicholas of Tolentino, Bernardino of Siena, Jerome

examination. If this inscription was originally
“sa-gio,” with a stroke mark above the “a-” signi-
fying an abbreviation, then this would indicate
“san giorgio,” and thus the name of the painter
Eusebio da San Giorgio.18 Indeed, it is possible
that the inscription on the loincloth is Eusebio’s
signature, however, due to its abraded condition,
this is not a certainty. It is important to note that
an attribution to the Perugian painter Eusebio
coincides with the Perugian provenance of the
painting, as was indicated in the previously men-
tioned lost inscription.
The Kress Saint Sebastian has also been recently
attributed to the Master of the Greenville Tondo,
an unidentified Perugino follower named for the
Madonna and Child with Angels in the Bob Jones
University Collection in Greenville, South Caro-
lina, a Peruginesque work usually dated to circa
1500 (fig. 10).19 This painting depicts the Madonna
and Child seated, as if enthroned, upon a stepped
platform, and flanked at the left and right by two
splendidly clothed angels.20 The rightmost angel
Fig. 9. The Madonna and Child with Saints Nicholas of Tolentino, bears a striking likeness to the Kress Sebastian, as
Bernardino of Siena, Jerome and Sebastian, “The Tezi Altarpiece,” the faces and features of the two figures are very
Pietro Perugino with Eusebio da San Giorgio, 1500, oil and similar, and were possibly executed in the same
egg tempera on panel, 69 7/8 × 52 3/4 in. (177.5 × 134 cm).
Galleria Nazionale dell’Umbria, Perugia. workshop. Furthermore, the particular golden
color of this angel’s hair, and the manner in
which the long hair of both of the angels falls
to the sides of their faces is much like that of
and Sebastian, and the figure of Saint Sebastian in Sebastian.21 As mentioned previously, the Kress
this work is similar in many respects to the Kress Sebastian closely resembles the image of the saint
Sebastian, as seen in the construction of the in the lost Saint Sebastian with Two Archers, formerly
upper torso, the handling and tautness of the in Berlin, and recently attributed to the Master
flesh, the position of the head and upward gaze, of the Greenville Tondo.22 The figure of Saint
and especially the treatment of the features and Sebastian in this picture is remarkably similar to
facial expression. the Kress saint in appearance, including the con-
In addition to stylistic considerations, there is tour and posture of the body, the location of the
further evidence that points to Eusebio da San arrow in the lower proper right groin, and the
Giorgio as the author of the Kress painting. Con- treatment of the face and features. Thus, the two
servation of the picture revealed remnants of a figures are quite possibly related, and they may
simple mordant-gilt halo as well as a decorative have been created in the same workshop using the
pattern on the violet ribbon loincloth, includ- same cartoon.
ing a partially effaced perpendicular gold-leaf The problems posed by the Saint Sebastian in the
inscription which can be interpreted as either Kress Collection at the Princeton University Art
“sagio” or “sacio.” Unfortunately, the third gilt Museum truly illustrate the complexity of con-
letter remained illegible even with ultraviolet light noisseurship associated with the Perugino School.

with the Madonna and Child with Angels and the Saint
Sebastian with Two Archers, both attributed to the
Master of the Greenville Tondo. However, based
on the existing evidence, the most appropriate
attribution of the Kress picture is to the School
of Perugino, possibly Eusebio da San Giorgio,
due to the presence of the inscription on the
loincloth, which may specifically indicate this
painter. In sum, the Kress Saint Sebastian is a
highly characteristic painting from the Italian
Renaissance, as it represents a very efficient and
productive workshop system.
Annette Rupprecht is a private paintings conservator in New

York City. She was previously appointed the first Samuel
H. Kress Fellow at the New York University Conservation
Fig. 10. The Madonna and Child with Angels, Master of the Center after having worked with Herbert Lank in London.
Greenville Tondo, ca. 1500, oil on panel, Bob Jones She holds a Fellowship Diploma in Conservation from the
University, Greenville, SC.
Hamilton Kerr Institute in Cambridge and an M.A. in
Paintings Conservation from the Cooperstown Program.
As indicated above, Pietro Perugino operated large Sheri Francis Shaneyfelt holds a B.S. in Biology from
workshops in both Florence and Perugia where Centre College, an M.A. in Art History from Vanderbilt
numerous artists were trained to faithfully repro- University, and a Ph.D. in Art History from Indiana
duce the style of the master. The frequency with University at Bloomington, with a specialty in Italian
which cartoons and model drawings were used Renaissance and Baroque Art. Her primary field of
and re-used in the creation of standardized fig- research is the art of Pietro Perugino and his school, and
ures and compositions within this school is well she is a professor of Art History at the Umbra Institute
known and documented, and this, together with in Perugia, Italy.
the abundance of Peruginesque paintings through-
out the world, complicates matters further. This
Notes
particular difficulty in connoisseurship has been 1. For the most comprehensive treatment of the life and
recognized for centuries, for even Giorgio Vasari, works of Pietro Perugino refer to Scarpellini (1991),
in his Vita of Raphael from 1568, commented and Canuti (1931). See additionally Bombe (1914); Gnoli
upon the striking similarity of this young artist’s (1923 a, b, and c); Camesasca (1959); Castellaneta and
Camesasca (1969); Becherer et al. (1997); Garibaldi (1999);
work with that of his master: Garibaldi and Mancini (2004).
2. Exhibition curated by Joseph A. Becherer, November
It is a very notable thing that Raffaello, studying 1997 – February 1998, catalogue cited above in note 1.
the manner of Pietro, imitated it in every respect 3. The grain of the wood panel is visible in the X-radiograph
so closely, that his copies could not be distin- composite made at nyu. There is an invoice in the Kress
Archive from J. Newcombe for altering this painting’s
guished from his master’s originals, and it was sixteenth-century frame to accommodate a cradle.
not possible to see any clear difference between 4. According to the Samuel H. Kress Art Collection Data,
his works and Pietro’s.23 Condition and Restoration Record for this painting:
“1950—Flaking pigment in many areas necessitated trans-
Indeed, there are definite likenesses between the ferring to new support. M. Modestini removed cradle
and wood panel and calcium preparation from back of
Kress Saint Sebastian and the style of Eusebio da painting. Applied new preparation to back of picture
San Giorgio, in addition to significant comparisons and attached painting to calcium and glue mixture over

Masonite. Leveled missing areas with preparation mixture p. 240; Shaneyfelt (2000), Vol. , p. 560. For the Raphael
and inpainted with dry colors and egg tempera medium. attributions, see Borenius (1913), Vol. i, p. 67, no. 58; and
French varnish isolator. Damar varnish coating.” Catalogue of Pictures at Doughty House, Richmond, Surrey: Collec-
5. Visible in the X-radiograph composite made at nyu. tion of Sir Herbert Cook (1932), p. 72, no. 58; R. Longhi, ms.
6. In 2005, Elizabeth Walmsley, Painting Conservator at the opinion of 1948, Kress Archives; Suida (1950), p. 8, no. 4.
National Gallery of Art, located the negatives taken during 15. The Kress Saint Sebastian was attributed to Giannicola di
the 1950 transfer procedure in the Kress Collection Archive, Paolo by Borenius (1913), Vol. i, p. 67, no. 58, and by the
Department of Image Collections, National Gallery of same scholar in Crowe and Cavalcaselle (1914), pp. 458–9;
Art, Washington, D.C. The Archive includes early treat- Gnoli (1918), p. 37; Gnoli (1923 c), p. 140; Van Marle (1933),
ment photos of many non-NGA Kress paintings. p. 430. For further information on the Perugian artist
7. For sources, refer to Hiller von Gaertringen (1997), Giannicola di Paolo, refer to Shaneyfelt (2000).
pp. 223–30; Hiller von Gaertringen (1998), pp. 53–69; 16. Refer to Berenson (1932), p. 178; Berenson (1936), p. 154;
Hiller von Gaertringen (1999). see also Becherer et al. (1997), p. 243.
8. The Louvre Saint Sebastian is discussed by Scarpellini (1991), 17. Scholars assigning the Tezi altarpiece to Perugino and
pp. 86–7, cat. no. 53; the work is on a panel, 176 × 116 cm. Eusebio include Berenson (1932), p. 178; Berenson (1968),
For further information on the Uffizi painting, refer to Vol. i, p. 123; Scarpellini (1991), p. 102, cat. no. 110; Becherer
Scarpellini (1991), p. 86, cat. no. 51; the picture is likewise et al. (1997), p. 243; Shaneyfelt (2000), Vol. ii, p. 553. For
on a panel, 178 × 164 cm, and is signed and dated by the further information on this altarpiece, see additionally
artist. The Uffizi Sebastian was lauded by Vasari in his Santi (1989), pp. 103–4, cat. no. 88; Garibaldi (1999), p. 132,
1568 Lives of the Most Excellent Painters, Sculptors and Architects cat. no. 57; Garibaldi and Mancini (2004), pp. 270–73.
as “a St. Sebastian worthy of the highest praise;” see Vasari 18. The inscription was recorded with an abbreviation mark
(1568), Vol. i, p. 593. There are indeed numerous represen- above the “a- ” by Suida (1950), p. 8, no. 4, and Shapley
tations of Saint Sebastian in the works of Perugino and (1968), p. 100. Thus, the “sa- ” of the inscription would
his followers; for sources, refer to note 1 above. indicate the word “San” or “Santo,” i.e., “Saint.” If
9. This painting, also known as the Martyrdom of Saint Sebastian, “sagio” is the correct reading of this inscription, then it
was formerly attributed erroneously to Giannicola di Paolo would indicate “san gio,” which in its complete form
and most recently to the Master of the Greenville Tondo. is “san giorgio,” and thus the name of the painter
The work was sold through Julius Böhler, Munich, June Eusebio da San Giorgio. The authors would like to thank
1937, no. 669, plate 54, and in the sale catalogue its meas- Maria Rita Silvestrelli of the Università per Stranieri di
urements are given as 146 × 106 cm. The painting had Perugia, and Alberto Maria Sartore of the Archivio di
previously been in the Somzée Collection, Brussels (sold Stato, Perugia, for their assistance in deciphering this
in 1904, no. 384) and subsequently in the Sedelmeyer inscription.
Collection, Paris. The present location of this work is 19. The Kress Saint Sebastian was attributed to the Master
unknown, and it is assumed that it was lost in the 1945 of the Greenville Tondo by Everett Fahy, in a letter of
fire in Berlin. William E. Suida noted this particular December 1966 in the Kress Archives; by Shapley (1968),
comparison in the k-1557: Figure 241 entry of Paintings pp. 100–101; and by Todini (1989), Vol. i, p. 198. For infor-
and Sculpture from the Kress Collection, 1956. mation on the Master of the Greenville Tondo, refer to
10. For further discussion of Saint Sebastian, refer to De Todini (1989), Vol. i, pp. 197–8; Pepper (1984), pp. 20,
Voragine (ca. 1260, trans. by Ryan 1993), Vol. 1, pp. 97–101; 317–18, cat. no. 14.1; Zeri (1976), p. 179, cat. no. 118 (a dis-
Farmer (1992), p. 429; Hall (1979), pp. 276–7. cussion of several paintings commonly attributed to the
11. This is referred to more explicitly in an inscription found Greenville master, without specific indication of the artist);
at the base of Perugino’s aforementioned Louvre Saint Shapley (1968), pp. 100–101; Bob Jones University (1962),
Sebastian, “sagittae tuae infixe sunt michi,” which trans- p. 54, cat. no. 24. It is important to note that due to the
lates as “Your arrows have been thrust into me.” This large number of paintings assigned to this particular mas-
phrase derives from Psalm 37:14–15 (King James Version): ter (between thirty and forty), it is highly doubtful that
“The wicked have drawn out the sword, and have bent one hand could have executed all these works, especially
their bow, to cast down the poor and needy, and to slay considering the discrepancies in style. The Greenville tondo
such as be of upright conversation. Their sword shall enter has not been assigned to Eusebio da San Giorgio in the lit-
into their own heart, and their bows shall be broken.” erature, nor does it appear to be by his hand; other than
12. For further information and documentation concerning the aforementioned similarity with the Kress Saint Sebastian,
Perugino’s workshops in Florence and Perugia, refer to the tondo is not stylistically characteristic of Eusebio.
Coonin (1999), pp. 100–104; Scarpellini (1991), pp. 37, 55, 20. The Madonna and Child in the Greenville Madonna and
64–6; Canuti (1931 [1983]), Vol. ii, pp. 302–5, doc. nos. Child with Angels was clearly derived from a commonly used
540–48. Perugino prototype of the Madonna and Child Enthroned,
13. The Oddi were also important patrons of Raphael, as and was adapted to fit this particular setting; thus, the
they commissioned his Coronation of the Virgin, in the noticeable absence of a throne seems rather awkward.
Vatican Museums, for their chapel in San Francesco al 21. There is also a similarity in the construction of the upper
Prato, Perugia, in 1503. For further information, refer to torso of both the rightmost angel in the Greenville tondo
Oberhuber (1977); Jones and Penny (1983), pp. 14–16. and Sebastian, and in the placement and position of the
14. Refer to Borenius (1913), p. 67, no. 58; Suida (1950), p. 8, head upon the neck, indicating that these two figures may
no. 4; Shapley (1968), pp. 100–101; Becherer et al. (1997), have originated from a common source, likely within a

Peruginesque workshop. An additional Perugino School Descriptive Catalogue of the Collection of Italian Pictures of the Rafaello
painting attributed to the Master of the Greenville Tondo Period of the Late Edward Solly, Esq. (Solly Sale, May 8, 1847).
in which a comparable angel can be observed, especially De Voragine, Jacobus (ca. 1260) Legenda Sanctorum. Translated
in the treatment of the face and features, is the Madonna by William Granger Ryan (1993), The Golden Legend:
Adoring the Child with two Music-Making Angels in a Landscape, Readings on the Saints, 3 vols. Princeton, NJ: Princeton
in the Pinacoteca Estense, Modena; refer to Shapley University Press.
(1968), pp. 100–101; Pepper (1984), p. 317; Todini (1989), Farmer, David Hugh (1992) The Oxford Dictionary of Saints,
Vol. i, p. 198; Vol. ii, fig. 1333. 3rd ed. New York and Oxford: Oxford University Press.
22. Saint Sebastian with Two Archers was attributed to the Master Fredericksen, Burton B. and Zeri, Federico (1972) Census of
of the Greenville Tondo by Shapley (1968), p. 100; Pepper Pre-Nineteenth-Century Italian Paintings in North American Public
(1984), p. 317; and Todini (1989), Vol. ii, p. 197. Collections. Cambridge, MA: Harvard University Press.
23. Quoted from Giorgio Vasari (1568), trans. by Gaston du Garibaldi, Vittoria (1999) Perugino: Catalogo Completo. Florence:
C. de Vere (1996), Vol. i, p. 711. Octavo.
Garibaldi, Vittoria and Mancini, Francesco Federico (eds.)
(2004) Perugino: Il Divin Pittore. Milan: Silvana Editoriale.
Bibliography Gnoli, Umberto (1918) “Giannicola di Paolo, nuovi docu-
Becherer, Joseph A. et al. (1997) Pietro Perugino: Master of the menti,” Bollettino d’Arte Vol. 12, Nos. 1–4, pp. 33–43.
Italian Renaissance. Grand Rapids, MI: The Grand Rapids ––– (1923 a) I Documenti su Pietro Perugino. Perugia: Unione
Art Museum. Tipografica Cooperativa.
Bellucci, Roberto and Frosinini, Cecilia (2003) “The myth ––– (1923 b) Pietro Perugino. Spoleto: Claudio Argentieri,
of cartoon re-use in Perugino’s underdrawing: technical Edizione d’Arte.
investigations” in The Painting Technique of Pietro Perugino, ––– (1923 c) Pittori e Miniatori nell’Umbria. Spoleto: Claudio
called Il Perugino: Proceedings of the LabS TECH Workshop. Edited Argentieri, Edizione d’Arte.
by Brunetto Giovanni Brunetti, Claudio Seccaroni, and Hall, James (1979) Dictionary of Subjects and Symbols in Art,
Antonio Sgamellotti. Supplement to Kermes 53, pp. 71-80. revised ed. New York: Harper and Row.
Berenson, Bernard (1932) Italian Pictures of the Renaissance: A List Hiller von Gaertringen, Rudolf (1997) “On Perugino’s
of the Principal Artists and their Works with an Index of Places. re-uses of cartoons” in Roger van Schoute and Hélène
Oxford: Clarendon Press. Verougstraete (eds.), Le Dessin Sous-jacent et la Technologie dans
––– (1936) Pitture Italiane del Rinascimento: Catalogo dei Principali la Peinture. Colloque XI, –  Septembre . Dessin Sous-jacent
Artisti e delle Loro Opere, con un Indice dei Luoghi, translated by et Technologie de la Peinture: Perspectives. Louvain-la-Neuve:
Emilio Cecchi. Milan: U. Hoepli. Collège Erasme.
––– (1968) Italian Pictures of the Renaissance: A List of the Principal ––– (1998) “L’uso e il riuso dei cartoni nell’opera del
Artists and their Works with an Index of Places. Central Italian and Perugino. La ripetizione della formula perfetta” in
North Italian Schools, 3 vols. London and New York: Phaidon. L’Ascensione di Cristo del Perugino. Cinisello Balsamo: Silvana.
Bob Jones University (1962) The Bob Jones University Collection of ––– (1999) Raffaels Lernerfahrungen in der Werkstatt Peruginos:
Religious Paintings. Catalogue of the Art Collection. Vol. i: Italian Kartonverwendung und Motivübernahme im Wandel. Munich:
and French Paintings. Greenville, SC: Bob Jones University. Deutscher Kunstverlag.
Bombe, Walter (1914) Perugino: Des Meisters Gemälde in  Jones, Roger and Penny, Nicholas (1983) Raphael. New Haven
Abbildungen. Berlin and Stuttgart: Deutsche Verlags-Anstalt. and London: Yale University Press.
Borenius, Tancred (1913) A Catalogue of the Paintings at Doughty Martin, Elisabeth and Rioux, Jean Paul (2003) “Comments on
House, Richmond, and Elsewhere in the Collection of Sir Frederick the technique and the materials used by Perugino, through
Cook. Vol. i: Italian Schools, edited by Herbert Cook. the study of a few paintings in French collections” in The
London: William Heinemann. Painting Technique of Pietro Perugino, called Il Perugino: Proceedings
Burger, William (1929) “A tondo by Sinibaldo Ibi,” The of the LabS TECH Workshop. Edited by Brunetto Giovanni
Burlington Magazine 55, pp. 87–9. Brunetti, Claudio Seccaroni, and Antonio Sgamellotti.
Camesasca, Ettore (1959) Tutta la Pittura del Perugino. Milan: Supplement to Kermes 53, pp. 43-56.
Rizzoli. National Gallery of Art (1956) Paintings and Sculpture from the
Canuti, Fiorenzo (1931) Il Perugino, 2 vols. Siena: Editrice d’Arte Kress Collection: Acquired by the Samuel H. Kress Foundation,
“La Diana” [reprinted 1983, Foligno: Editoriale Umbra]. –. Washington, D.C.: National Gallery of Art.
Castellaneta, Carlo, and Camesasca, Ettore (1969) L’Opera Oberhuber, Konrad (1977) “The Colonna Altarpiece in the
Completa del Perugino. Milan: Rizzoli. Metropolitan Museum and problems of the early style of
A Catalogue of Pictures at Doughty House, Richmond, Surrey: Collection Raphael,” Metropolitan Museum of Art 12, pp. 77-8.
of Sir Herbert Cook (1932) London: William Heinemann. Pepper, D. Stephen (1984) Bob Jones University Collection of Religious
Coonin, A. Victor (1999) “New documents concerning Art: Italian Paintings. Greenville, SC: Bob Jones University.
Perugino’s workshop in Florence,” The Burlington Magazine Santi, Francesco (1989) Galleria Nazionale dell’Umbria: Dipinti,
Vol. cxli, No. 1151, pp. 100–104. Sculture e Oggetti dei Secoli XV–XVI. Cataloghi dei Musei e Gallerie
Crowe, Joseph Archer and Cavalcaselle, Giovanni Battista d’Italia. Rome: Istituto Poligrafico e Zecca dello Stato.
(1914) A History of Painting in Italy, Umbria, Florence and Siena Scarpellini, Pietro (1991) Perugino, revised ed. Milan: Electa.
from the Second to the Sixteenth Century. Vol. v: Umbrian and Shaneyfelt, Sheri Francis (2000) The Perugian Painter Giannicola
Sienese Masters of the Sixteenth Century, edited by Tancred di Paolo: Documented and Secure Works. Ph.D. dissertation,
Borenius. New York: Charles Scribner’s Sons. Indiana University.

Shapley, Fern Rusk (1968) Paintings from the Samuel H. Kress
Collection, Vol. : Italian Schools, XV–XVI Century. London:
Phaidon.
Suida, William E. (1950) “The Samuel H. Kress Collection,”
The Philadelphia Museum Bulletin 46, p. 8.
Todini, Filippo (1989) La Pittura Umbra dal Duecento al Primo
Cinquecento,  vols. Milan: Longanesi.
Van Marle, Raimond (1933) The Development of the Italian Schools
of Painting. Vol. xiv: The Renaissance Painters of Umbria. The
Hague: Martinus Nijhoff.
Vasari, Giorgio (1568) Le Vite de’ Più Eccellenti Pittori, Scultori et
Architettori. Florence. Translated by Gaston du C. de Vere
(1996), Lives of the Painters, Sculptors and Architects,  vols.
New York: Alfred A. Knopf.
Zeri, Federico (1976) Italian Paintings in the Walters Art Gallery,
Vol. i. Baltimore, MD: The Walters Art Gallery.
Photography Credits
Figs. 1 and 2, p. 132. Princeton University Art Museum,
Princeton, NJ, Samuel H. Kress Collection.
Fig. 3, p. 134. Photograph ©Kress Collection Archive,
Department of Image Collections, National Gallery
of Art, Washington, D.C. Digital enhancement by
Elizabeth Walmsley.
Fig. 5, p. 135. Photograph by Erich Lessing ©Art Resource,
Inc., New York, NY.
Fig. 6, p. 135. Photograph by Scala ©Art Resource, Inc.,
New York, NY.
Figs. 7 and 8, p. 136. Photograph ©Department of Image
Collections, National Gallery of Art, Washington, D.C.
Fig. 9, p. 138. Courtesy of the Soprintendenza per i Beni
Architettonici, il Paesaggio, il Patrimonio Storico, Artistico
e Etnoantropologico dell’Umbria, Perugia, Italy.
Fig. 10, p. 139. From the Bob Jones University Collection,
Greenville, SC.
1

Madonna and Child with Saint John
Michelangelo Associate, ca. 1500
Egg tempera and perhaps oil on thinned wood panel
27 1/2 × 19 1/8 in. (69.9 × 48.6 cm)
Samuel H. Kress Foundation, New York, New York
(k-1569)
Fig. 1. Madonna and Child with Saint John, before cleaning and
restoration.
Fig. 2. Madonna and Child with Saint John (fig. 1), after cleaning
and restoration.

The Master of the Manchester Madonna:
Restoration, Technique, and a Context
for Attribution
Molly March
T he identity of the “Master of the Manchester Madonna” has been a

topic of art historical debate for the better part of fifty years.1 Recently,
an exhibition publication entitled Making and Meaning: The Young Michelangelo
acknowledged this body of scholarship and advanced it by interpreting
specific observations of painting technique.2 The study principally
addressed material aspects of the National Gallery, London’s Virgin and Child
with Saint John and Angels (“The Manchester Madonna,” fig. 3) and Entombment
(fig. 4) with the intention of adding these unfinished paintings to
Michelangelo’s oeuvre. The authors employed works of an artist close to
Michelangelo, formerly the “Master of the Manchester Madonna” and here
called Michelangelo Associate, as a foil for the two London pictures attributed
to the young Michelangelo in the exhibition. As a result, the distinctive style
and technique of this Michelangelo Associate were briefly explored in the
Making and Meaning project, providing a valuable launching point for further
investigation.
This paper is intended to set down selected details of the Michelangelo
Associate’s painting technique as encountered during the restoration of the
Kress panel Madonna and Child with Saint John (figs. 1 and 2). Optimally, this
information will join with similar accounts to further illuminate the path
towards attributing a small group of pictures traditionally considered together
with the Kress painting: Madonna and Child with Saint John the Baptist (fig. 5);
Madonna and Child (fig. 6); Pietà (Galleria Nazionale d’Arte Antica di Palazzo
Molly March 145

Fig. 4. The Entombment, Michelangelo, ca. 1501, oil on wood
panel (identified), 63 3/8 × 59 in. (161.7 × 149.9 cm). National
Gallery, London.
Fig. 3. Virgin and Child with Saint John and Angels (“The Manchester
Madonna”), Michelangelo, ca. 1497, egg tempera on wood panel,
41 1/2 × 30 1/2 in. (105.4 × 76.8 cm). National Gallery, London.
Fig. 5. Madonna and Child with Saint John the Baptist, Michelangelo Fig. 6. Madonna and Child, Michelangelo Associate, ca. 1505,
Associate, ca. 1498, tempera on wood panel, 26 in. (66 cm) dia. tempera and oil on panel, 14 5/8 × 11 3/4 in. (37 × 30 cm).
Gemäldegalerie der Akademie der bildenden Künste, Vienna. Pinacoteca Ambrosiana, Milan.

Barberini, Rome, Inv. Nr. 948); and Madonna and The collection of colors used in the Kress,
Child (ex-Baden bei Zurich3).4 Vienna, and Ambrosiana paintings are strikingly
During the last half century, scholars have similar in hue, location within each picture, and
attributed these works alternatively to an Umbro- their technical application. Most notably, the
Bolognese or Florentine-Ferrarese master, among mauve- and lavender-colored architectural planes
other descriptive combinations, all meant to discovered in the cleaned state of Madonna and
recognize the artist’s eccentric style in translating Child with Saint John in New York are virtually
Michelangelo’s designs.5 Most authors agree identical in tone to contextual walls in the Vienna
that the painter’s training may have encompassed tondo image.8 In addition, the darkened blue of
workshop practices from more than one region of the Kress Virgin’s mantle, though in compromised
Italy. The Making and Meaning study lent credence condition, is quite similar in hue to the deep blue
to this hypothesis through technical analysis: of the Madonna’s robe in the Ambrosiana paint-
Jill Dunkerton, Senior Restorer, Conservation ing, as well as to the underlying or less retouched
Department, National Gallery, London, compares passages of blue robe in the Vienna tondo.
the Michelangelo Associate’s choice and applica- Furthermore, the generalized, mound-like forms
tion of materials in Vienna’s Virgin and Child with of hills in the landscape of the Kress painting
Saint John the Baptist to painting practices believed are much in keeping with the simple bluish-green
to be characteristic of the area around Ferrara, shapes portraying distant mountains in the cor-
specifically as exemplified by the works of Cosimo ners of landscape in the Vienna tondo and in the
Tura.6 The potentially regional qualities of this
Michelangelo Associate’s distinctive painting
technique as seen in three pictures, Virgin and
Child with Saint John the Baptist (Vienna), Madonna
and Child (Milan), and Madonna and Child with
Saint John (New York) will be touched upon below
in the context of discussing treatment of the
Kress panel.
A Signature Palette and

Method of Pigment Application
Madonna and Child with Saint John exhibited chronic
lifting of paint along two periodically active verti-
cal joins in the panel support. It was this persist-
ent problem—as well as the obscuring presence
of several layers of very discolored varnish—that
prompted a decision to undertake the restoration
of this picture despite the worn condition of the
image (see figs. 1 and 2).7 Cleaning the Kress
painting revealed a brilliant palette of jewel-like
colors that had been completely suppressed by the
picture’s coatings of darkened resin (fig. 7). The
hues uncovered are quite important to firmly link-
ing the Kress panel to the Vienna tondo and the
Ambrosiana’s Madonna and Child, as well as to colors
uniquely employed by Domenico Ghirlandaio and
his followers for the creation of flesh tones. Fig. 7. Madonna and Child with Saint John (fig. 1), cleaned state.
Molly March 147

upper-right background of the Ambrosiana’s
Madonna and Child.9
The figures in the Kress, Vienna, and Ambro-
siana paintings all exhibit skin that is virtually
“opalescent” in appearance.10 This effect is a
result of both the artist’s selection of hues and
method of applying them. The Michelangelo
Associate’s flesh tones are based on a limited
palette of pure, strident colors very similar to
those employed by Domenico Ghirlandaio and
his workshop, namely an unusually blue-green
underpaint modified by cool white and shades
of salmon pink. Flesh painted with these bright
tones has a markedly different appearance than
flesh created with a palette based on more muted
hues, such as the light yellow-green (or cream)
preparation modeled by warm white highlights,
hints of rose and yellow-brown middle tones, and
earth-colored shadows that can be seen in the
works of artists such as Andrea del Verrocchio,
Sandro Botticelli, and others.11
The Michelangelo Associate’s use of a triad
Fig. 8. Portrait of a Woman, Domenico Ghirlandaio, ca. 1490,
of brilliant hues for the modeling of flesh most tempera on wood panel, 22 1/2 × 17 3/8 in. (57.1 × 44.1 cm).
likely derives from contact with Ghirlandaio or The Metropolitan Museum of Art, New York, NY.
one of his pupils, such as Pinturicchio, Granacci,
or Michelangelo himself, all of whose early
works display this method of coloring skin to
some degree.12 Nevertheless, this relationship to (fig. 8) are all given the same pictorial weight.
Ghirlandaio’s practice does not necessarily tie the In this image, though hatches of white are more
Michelangelo Associate exclusively to Florence, plentiful in areas of highlight, and pink strokes
as all the artists mentioned worked in Rome with hints of the green preparation color domi-
for some time near the close of the fifteenth nate in the shadows, the finished flesh is formed
century.13 The modeling of flesh in the Kress, by a film that is consistent in opacity and appar-
Vienna, and Ambrosiana paintings can be dis- ent thickness from highlight to shadow. Though
tinctly separated from that in Ghirlandaio’s the Metropolitan picture may be an exaggerated
pictures by the Michelangelo Associate’s unique example, the consistency typical of Ghirlandaio’s
manner of unevenly disposing pigment across surface is not present in the flesh passages of the
flesh passages, and his use of a translucent, brick- three Michelangelo Associate works under dis-
red color for shaded regions of skin.14 cussion.15 For example, where flesh is rendered
In the work of Ghirlandaio and his followers, in the Kress picture, brick-colored areas of deep-
the opaque white or light pink strokes that create est shade are thin and smooth, as are the brick
highlights typically extend much further into the and salmon passages of semi-shade. The density
middle tone and shade areas of a form than they and low relief of the Michelangelo Associate’s
do in works by the Michelangelo Associate. For hatching strokes notably increase at the perimeter
example, the hatching strokes that describe flesh and into the center of skin highlights, as will be
in Domenico Ghirlandaio’s Portrait of a Woman further described below.

The exact sequence of pigment application in tary pink and green employed; the pearlescent
the flesh passages of the Kress Madonna and Child aspect introduced by scumbling cool white over
with Saint John was not determined by examining a lower-valued green; and the presence of signifi-
cross-sections of original paint samples. However, cant gaps between the artist’s uppermost hatching
close study of these areas during the retouching strokes which allow the color of the underlying
process, with and without the aid of a stereo- layer to participate in the final effect. In addition,
binocular microscope, suggested that the the tips of the Michelangelo Associate’s cool
Michelangelo Associate created flesh in the white highlight strokes are intermeshed with the
following manner. A relatively flat layer of milky upper ends of opaque pink middle-tone strokes,
blue-green is applied to the white gesso within and the tips of these opaque pink strokes are
a figure’s perimeter. This underpaint is modified interlaced with the scattered, deep-pink colored
by a translucent, brick-red glaze over sections hatches faintly visible over the brick-red, under-
destined to be middle tone and shadow, leaving glazed shade. As every point of intermeshing
the green in reserve for areas of intended high- stroke ends, a new hue is suggested optically, and
light. In the middle shade, hints of form are these implied transitional tones play an indispen-
subtly picked out with a few relatively liquid, sable role in the Michelangelo Associate’s delicate
curved hatching strokes of a semi-opaque deep rendering of form. This artist’s technique of
pink over the smooth, translucent, brick-red creating the illusion of volume in his figures is so
glaze. A narrow margin of similarly shaped, short economical and precise in its execution, it seems
strokes in a more opaque light pink initiates an probable that he painted in the presence of a
abrupt transition from middle tone into light. model or prototype. Furthermore, the nature of
Finally, short, curved hatches of lean, opaque the Michelangelo Associate’s technical economy
white directly over the reserved green underpaint in these passages might suggest that he was accus-
create the volumetric areas of flesh in highest tomed to working in another medium such as
light. With this system, the final hatched strokes sculpture, enamels or metalwork. In any case,
forming the strictly localized flesh highlights are there are no visible signs of working out a design
slightly raised or in relief on the picture’s other- during the painting process and in fact, no room
wise porcelain-smooth surface (figs. 9 and 10). in the crisply realized yet thinly executed surface
Flesh passages created in this way have an layers in which to do so. Before he began painting,
optically scintillating quality resulting from this artist knew precisely what he wanted to show
several factors: the interplay of the complemen- and exactly how to achieve his end.
Fig. 9. Madonna and Child with Saint John (fig. 1), detail of Fig. 10. Madonna and Child with Saint John (fig. 1), detail of
hatching of flesh tones, Christ Child’s foot (cleaned state). hatching of flesh tones, Christ Child’s chest (cleaned state).
Molly March 149

As was suggested above, a more specific articu- employ the brilliant hues that are a hallmark of
lation of the regional roots of the Michelangelo the Ghirlandaio-derived system of modeling.16
Associate’s painting technique will be fundamental If the palette of the Michelangelo Associate’s
to further study of his identity. To this end, it is flesh tones in the Kress panel cannot be linked to
important to state that the Michelangelo Associ- the work of artists such as Cosimo Tura, his
ate’s coloration of flesh has much more in com- physical application of pigments to the panel in
mon with the work of Domenico Ghirlandaio flesh and drapery areas might be. The
(see fig. 8) than with that of Cosimo Tura (fig. 11), Michelangelo Associate’s predilection for model-
a suggested source for the Michelangelo Associate ing that shows thickly applied, localized high-
pictures’ extra-Florentine or Ferrarese elements. lights immediately juxtaposed to relatively
Cosimo Tura’s flesh tones often have a glowing, smooth, thin middle tones and shadows seems
pearl-like quality resulting from his use of strong related in its technical execution to works pro-
highlights scumbled over a darker underlayer duced by Cosimo Tura and
(and perhaps the presence of a heightened under- his followers (see figs. 9 and 11). This can be seen
drawing); however his flesh passages are much in the X-radiograph of the Kress Madonna and
more somber in overall tonality, and do not Child with Saint John (fig. 12).17 In a similar vein,
the Michelangelo Associate’s positioning of
strong lights on the edges of drapery folds may
be connected to the visual example of the Paduan
Andrea Mantegna’s art—through the Tura circle
or directly.18
Autograph Hatches and

Approaches to Restoration
During the process of thinning darkened varnish
layers from the Kress Madonna and Child with Saint
John, semi-opaque scumbles of an aged restoration
were also cleared from the painting’s surface. The
removal of these restorer’s touches from the inter-
stices of original brushwork recovered a surface
that is alive with eccentrically placed, hatched
strokes. The Michelangelo Associate’s use of these
hatchings is strictly limited to the Virgin’s mantle
and to passages of flesh. The rest of the painting
is executed in fluid, blended brushwork that is
barely detectable.19
The function and execution of these strokes
are different in the mantle and skin areas. Inter-
mittently placed, unblended hatches of crisp
white on the blue mantle highlight the edges of
drapery folds and summarily suggest selected
planes in between these fold ridges. A very similar
effect can be seen in the Metropolitan Museum’s
Ghirlandaio, Portrait of a Woman, where light falling
Fig. 11. Virgin and Child with a Female Martyr and Saint Jerome,
Cosimo Tura, oil on canvas mounted to wood, 60 1/4 × 43 1/2 in. on the sitter’s upper sleeve is briefly indicated
(153 × 110.5 cm). Musée Fesch, Ajaccio, Corsica, France. with pale pigment (see fig. 8). The relative

isolation of the white hatches on the Kress
panel’s abraded blue mantle gives these strokes a
quality of decorative accents, though they were
once obviously more integrated with the fabric
portrayed (fig. 13). In the flesh passages, the
Michelangelo Associate’s use of hatched strokes
is more extensive. Here, a profusion of roughly
parallel yet variously angled white and pale pink
hatches concisely structure a mannered, bulbous
musculature underlying highlighted skin (see
figs. 9 and 10).
In approaching the retouching of the Kress
Michelangelo Associate painting, the seemingly
signature quality of the peculiar hatching strokes
in the blue drapery and flesh prompted trips
to study the surfaces of the Vienna tondo, the
Barberini Pietà, and the Ambrosiana panel to look
for similarly hatched passages. Travel was also
Fig. 12. Madonna and Child with Saint John (fig. 1), undertaken with the hope of locating a model on
X-radiograph, detail. which to base any reconstruction of form in the
Kress panel’s quite fragmentary blue mantle.20
Careful study of these related pictures’ surfaces,
albeit in gallery conditions for the latter two,
revealed that hatching strokes remarkably close in
appearance to those found in the Kress painting
are present on all three panels. Furthermore, in
the Vienna tondo and the Ambrosiana Madonna
and Child, the individual sizes, shapes and spacing
of the hatching strokes not only bear a striking
resemblance to those in the Kress painting, but in
each picture they are similarly concentrated in the
flesh and blue mantle passages.
The virtually identical hatching strokes dis-
covered in the Vienna and Ambrosiana images
indicated that the restoration of the Kress panel
obviously should retain the legibility of this
signature hatching technique while visually
reintegrating areas of loss with extant original
passages. These calligraphic hatching strokes are
so distinctly recognizable that they may one day
serve—in combination with other material and
documentary evidence—to facilitate the discovery
of this artist’s identity.
During the treatment of the Kress painting,
Fig. 13. Madonna and Child with Saint John (fig. 1), detail of maintaining the integrity and visibility of these
hatching of blue mantle highlights (cleaned state). signature hatches in the blue drapery and figures’
Molly March 151

flesh was also important to the process of restora- National Gallery, London. The Pietà is currently
tion itself. The Kress Virgin’s blue mantle is so displayed in a modern frame within a Plexiglas
extensively damaged that, in most areas, only the vitrine that was presumably created to protect
fragmentary white accent hatchings remain to and/or transport this multi-planked panel.
suggest the original placement of drapery folds Retouching of losses to the Barberini painting
(see figs. 7 and 13). In passages of flesh, the was carried out in selezione cromatica, the Italian
Michelangelo Associate’s signature hatchings method of compensating areas of loss with painted
effectively governed the retouching process by lines of diverse colors that blend optically into a
their large scale relative to the dimension of each single tone when viewed from a distance. The
figure. As was noted above in the discussion of restoration is in the Roman style of rigatini: the
painting technique, the length, individual shape lines are painted in a rigidly vertical orientation
and spacing of each extant hatched stroke is (rather than being directionally placed to suggest
indispensable to the illusion of volume in the form in areas of loss, as is the practice in Flor-
Kress figures’ musculature, as well as to the ence). Though the intention of this broken-stroke
suggestion of form in the Virgin’s blue mantle. retouching technique is that restorations can be
It was necessary to study the autograph hatches separated from passages of original paint upon
surrounding losses, particularly in the flesh, on close scrutiny, in the case of the Barberini Pietà,
a stroke-by-stroke basis before broken or inter- which is rendered in a restricted grisaille palette, this
rupted forms could be connected across voids in mode of retouching conflicts aesthetically with
the paint layer without any illusionary compro- the painting’s unique, original hatching technique.
mise to the continuous net of flickering, original The 1994 restoration of Vienna’s Virgin and
strokes that economically create volume. Child with Saint John was accomplished with
It proved particularly valuable to have seen retouching that is meant to be invisible upon
the other paintings attributed to this Michelangelo casual inspection, presenting an integral image
Associate when considering the reintegration of for the viewer’s enjoyment.21 Losses and surface
local losses to the network of hatches forming abrasion in the Vienna tondo have been retouched
the flesh. The first-hand observation of related to quite an advanced level of finish. During the
works ultimately informed our decision regarding treatment of this picture in preparation for its
how far to close—or to what degree to retouch— exhibition in London, the restorer appears to
abraded flesh passages in the Kress panel. It was have chosen to knit together the Michelangelo
important to consider whether or not to replace Associate’s hatched strokes, placing translucent
the previous restorer’s milky touches that were scumbles in between them, thereby producing
removed with the discolored varnish. These touches, a tonally even or smooth surface in the flesh
intended to unify or smooth the appearance of passages and some areas of the blue mantle in
the figures’ skin, had been applied over hints of the Vienna painting.22
green earth underpaint showing in between the The Madonna and Child in the Pinacoteca
lattice of original white and pink hatches in the Ambrosiana was last restored employing both
Kress picture. First-hand study of the Barberini, intentionally visible and less apparent methods of
Vienna, and Ambrosiana paintings not only pro- loss compensation. A large part of the original
vided similar original surfaces to study, but also background in the upper left corner of this paint-
presented different ways of approaching the resto- ing no longer exists. This area of the Ambrosiana
ration of pictures created with the Michelangelo picture is currently toned with a simple beige
Associate’s unusual painting technique. color, a retouching technique known in Italy as
The Barberini Pietà and Vienna tondo were neutro, the filling of areas of total image loss with
both restored quite recently, the latter just prior a tone intended as neutral. The rest of the picture
to the 1994 Young Michelangelo exhibition at the has been loosely retouched in colors that currently

do not exactly match the surrounding original increasingly possible to see the faint, hot pink
pigment. Though the transparency and colorless- strokes delicately suggesting form on top of this
ness of the current varnish on this painting allow layer in the middle tones. Where broken, these
one to appreciate the Michelangelo Associate’s original pink strokes were reconnected, but no
characteristic palette, the imprecision of the further retouching was applied to these areas.
retouches in placement and hue compromises Finally, the palest pink and pure white highlights
the legibility of the original hatching brushwork, of the flesh were retouched by discreetly connect-
particularly in the figures’ flesh. ing points where original hatches of paint had
The study of these three diverse approaches to obviously been interrupted by pigment loss mid-
restoring paintings attributed to the Michelangelo stroke. Due to the Michelangelo Associate’s abbre-
Associate reconfirmed the fragility of this artist’s viated mode of indicating form, the overall shape
particular mode of image making and its vulner- of each highlight in his figures’ flesh is of crucial
ability to aesthetic compromise. Tonal transitions importance to the intended illusion. Thus, during
in flesh passages of the Vienna, Ambrosiana, and the final retouching of the flesh highlights, much
Kress paintings have all been affected by strong time was spent studying the original strokes at the
cleaning in the past, with some loss to middle edges of the lighted areas, and retouching along
tone and mid-shade pink hatchings which may these margins was intentionally minimal in order
contain the often sensitive vermilion. Due to to preserve an impression of the original, though
the abraded state of these pictures, it is unclear abraded, transitions and isolated shapes building
whether or not the opalescent white flesh high- the eccentric musculature of this artist’s figures.
lights were once covered with a now faded or lost The passage showing the most extensive paint
glaze. In addition, the degree to which the skeletal loss in Madonna and Child with Saint John is the blue
hatching strokes in the flesh tones were originally mantle of the Virgin (see figs. 7 and 13). The frag-
incorporated with their surround by scumbles mentary state of the Kress mantle and numerous
covering interstices in the extant brushwork lattice past campaigns of restoration in this area make a
also remains unknown. To avoid aesthetic or clear assessment of the original painting technique
visual confusion, losses to the figures’ flesh in quite difficult. Scattered hints collected from
the Kress Madonna and Child with Saint John were selected, better preserved parts of the robe can
reintegrated with comparative restraint during merely suggest the original order in which the
the recently concluded restoration. layers of pigment were applied. In several areas of
Flesh passages of the Kress painting were complete paint loss, abraded sections of a dark
retouched in roughly the same sequence that they brown, summary design drawn with a brush can
were originally painted, from shade, to middle be seen on the exposed amber-colored gesso
tone, to highlight. At points of complete loss, ground. These preparatory lines seem to sketch
the voids were filled with new, white gesso and the placement of drapery folds, but the extant
then toned to match the Michelangelo Associate’s fragments are few and far between, and it was not
signature blue-green shade of flesh underpaint. possible to link them into a meaningful drawing.
Next, these toned losses, as well as spots of the During the initial stage of painting, passages
original green preparation exposed by abrasion, intended to be deepest shade in the mantle appear
were locally glazed with brick red in passages of to have been coated with a hot, brown glaze
middle tone and shadow to integrate them with directly over the sized gesso ground.23 Areas of
the prevalent original color. The green toned semi-shade and middle tone were then laid in
losses in areas of highlighted flesh were not with a translucent, bright green that was subse-
glazed with red but were left in reserve. As the quently covered with a medium, opaque blue,
translucent, brick-red layer in the shadows and quite smoothly rendered, showing little or no
middle tones was unified by retouching, it became trace of brushwork. The margins of intended
Molly March 153

highlight along the edges of folds in these flat, ing, pure blue glaze and selected, liquid strokes
middle tone areas were then prepared with a of a hot brown transparent color similar to that
thicker layer of paler, opaque blue, slightly in forming deepest shade in the mantle of the Kress
relief of the middle tone surface. With the basic Virgin. In the Ambrosiana mantle, sections of the
locations of deep shade, middle tone and high- hatching strokes meant to show textile edges in
light already indicated, isolated sequences of brightest highlight are pure white. However, the
pure white hatching strokes were applied to con- extremities of these same white strokes lie under-
cisely articulate the tubular curves of drapery fold neath an ultra-transparent glaze of blue. The sig-
apices on top of the paler blue margins, and to nature hatching strokes remain crisp and unblended,
summarily suggest light falling on slumped planes yet the selectively placed final glaze introduces a
of interfold fabric in the flat, middle tone blue subtler transition from light into shade.26
passages.24 Finally, transparent blue glazes were
applied, presumably to soften or modify the tran- Material Aspects of Technique
sitions from shadow to middle tone to highlight, During restoration of the Kress Michelangelo
however these uppermost glazes are now extremely Associate painting, it was possible to examine
abraded in some areas and entirely lost in others. only the Kress panel and the Vienna tondo in the
It is interesting to note that the Madonna’s context of a conservation studio. Close inspection
blue robe in the Vienna tondo is also significantly confirmed many physical similarities between
damaged. Furthermore, there is a remarkable these two works that were briefly addressed by
similarity between the patterns of loss in the the Making and Meaning project. In her essay, “The
Vienna and Kress mantles. In their cleaned state, painting technique of the Manchester Madonna,”
both the Kress and Vienna paintings present blue Dunkerton discusses a profusion of tiny craters
mantles in which losses are distributed in such a that can be seen in lighter passages of the
way that they give the false impression of a patch- Manchester Madonna, the Entombment, and to some
work or brocade fabric (see fig. 7). The similarity extent, Michelangelo’s Doni Tondo. In addition, she
of the damages in the mantles of the Kress and notes that these craters—suspected to be a result
Vienna paintings may support the supposition of burst bubbles in a rapidly or vigorously applied
that these two pictures are by the same hand or gesso—are present in the preparation of the
from the same studio; the evidence is in the pattern National Gallery, London’s Virgin and Child by
of the loss. The like patterns of loss would sug- Domenico Ghirlandaio.27 In the cleaned state of
gest that the mantles in both paintings were con- the Kress Michelangelo Associate painting, craters
structed with similar, if not identical, sequences were also discovered in the thinly painted areas
of layers and mixtures of pigments/media—a such as the lower sky at the left horizon, the
“fingerprint” strata that had a specific chemical cangiante cloth under the Christ Child (see fig. 9),
vulnerability—and thus, were identically affected and the Virgin’s lilac-shaded veil. The craters
by their later, isolated cleanings.25 observed in the surface of the Kress painting are
An exceptionally well-preserved passage of blue extremely similar in size and distribution to those
mantle in the lower left corner of the Ambros- noted in the gesso grounds of Ghirlandaio and
iana’s Madonna and Child may provide an example Michelangelo. This detail, which may signify a
of how the finishing layers on the blue robes of specific workshop’s process, might eventually lend
the Kress and Vienna paintings once appeared (see support to the hypothesis that the Michelangelo
fig. 6). In this beautiful bit of eccentrically realized Associate had some in-studio contact with
drapery, hints of the bright green preparatory layer, Ghirlandaio or one of his pupils.
the opaque, medium-blue middle tone, and the Since the Virgin’s blue mantle is extremely
carefully placed white hatches of the highlights damaged in both the Kress Michelangelo Associ-
described above are ultimately unified by a pool- ate panel and the Vienna tondo, it was possible

during the cleaning of each picture to see the lines echoing the outermost contours of the
remarkably amber-yellow appearance of the gesso Christ Child’s right, inner arm, left shin and left
in areas of complete paint loss (see figs. 7 and 13). foot are actually slightly within the finally realized
At first glance, this amber tone could be attrib- perimeter of each limb (see figs. 9 and 10). During
uted merely to the presence of a sealing layer its restoration, the surface of the Kress panel was
of glue on top of a white gesso, or to staining examined via infrared reflectography (irr) with
imparted by later oils, resins and glues introduced a Hamamatsu vidicon camera and Sony monitor
during restorations. However, cross-sections of in order to look for preparatory underdrawing
samples taken from the blue robe of the Kress beneath the paint layers of the Michelangelo
painting show that the actual gesso layer appears Associate’s picture. An image of any underdraw-
quite uniformly yellow. The color apparent seems ing in the area of the extremely damaged blue
to be largely a result of an unusually high glue mantle might have assisted the restorer in recover-
content in the gesso and perhaps some trace ing some semblance of the original arrangement
inclusion of pigments, as scattered particles of of drapery folds during retouching of the robe.
red lake and black were identified in the Kress Unfortunately, virtually no underdrawing was visi-
samples analyzed.28 These findings are in accord ble in the Kress painting with the vidicon camera,
with Franz Maringer’s analysis of the amber col- save for a few dark, brush-applied shapes near the
ored preparation layer in the Vienna tondo, where upper left edge in the swag of red curtain.
he observed a very glue-rich gesso containing irr images of the Vienna tondo do show some
particles of the pigment red ochre.29 Maringer dark underdrawing, though the drawing recorded
has also been able to identify a similar ground in consists of only a scant description of drapery
a Venetian picture.30 During a discussion of the forms and a pronounced adjustment to the posi-
Vienna tondo in the Making and Meaning study, tion of the proper right foot of the Virgin in the
Dunkerton noted that this yellow-toned, glue-rich foreground. In fact, it may be interesting to note
gesso is not found in paintings produced by the that the drawing visible via irr in the Vienna
Ghirlandaio studio, though it has been identified tondo seems to be confined to corrections of the
in works of Cosimo Tura and other artists paint- picture’s design. It has been speculated that the
ing in and around Ferrara.31 Thus, the presence characteristic mint green underpaint in the flesh
of a remarkably glue-rich gesso in the Kress and of the Michelangelo Associate’s pictures is com-
Vienna pictures of the Michelangelo Associate prised of a green earth that is particularly opaque
may indicate that this artist had an early exposure to irr inspection.32 In addition to this, any draw-
to technical practices in northern Italy. ing done in a transparent red or reddish-brown
Under natural light, no significant traces of color may be invisible to infrared examination as
any preparatory underdrawing are visible on the these hues become transparent when viewed by
surface of the Kress painting. While relatively bold an irr camera while illuminated by light from
lines realized in warm brown pigment applied the infrared part of the visible light spectrum. In
with a thin brush can be seen circumscribing John theory, the Michelangelo Associate could have
the Baptist’s left arm and hand, these lines seem made a more involved preparatory design for
to be an in-process articulation or strengthening both the Kress and Vienna paintings than can
of contours within the paint layer, rather than a be imaged by irr.33 However, if this artist was
preparatory underdrawing. Furthermore, these painting in the presence of a model or finished
brown contour lines in the Kress picture may once drawing for all or parts of his picture, it might be
have been slightly covered and visually softened reasonable to assume that only a brief indication
by a translucent, final scumble of flesh coloring; of form was necessary in the underdrawing phase.
due to surface abrasion, it is possible to see in the This lack of detectible underdrawing in the Kress
cleaned state of the painting that similar brown panel is distinctly different from the profusion
Molly March 155

of preliminary designs evident in Cosimo Tura’s the group of paintings under discussion, it might
works. With Tura’s later paintings especially, be useful to study any incisions in this panel and
strongly hatched underdrawings are commonly their relationship to the final, painted forms.
quite legible when the pictures are scanned with The Kress panel and the Vienna tondo are also
an irr camera. At the very least, this would imply alike in finishing details that would have been
that the Michelangelo Associate made a selection applied in the last stages of painting: two parallel
of drawing materials and underpaint pigments that bands of mordant gilding trim the edges of both
differ from those employed by Cosimo Tura. It pictures’ blue mantles—as well as the Madonna’s
would be equally interesting to compare irr images veil in the Ambrosiana, Vienna, and Barberini
of paintings from the workshop of Ghirlandaio paintings and the drapery under the dead Christ
with those taken from the Michelangelo Associ- in the Pietà. This double line of mordant gilt trim
ate’s pictures, particularly in the context of the actually interrupts diagonally placed signature
significant body of highly finished drawings for hatching strokes at the highlighted edge of a
paintings left by Ghirlandaio. drapery fold identically in the Kress picture and
Preparatory incisions in the gesso ground of in the Barberini panel. Additional gilding deco-
the Kress, Vienna, and Barberini pictures take the rates the Vienna tondo’s cloth of honor and the
place of drawing in designing an architectural Kress Virgin’s neckline, cuffs, and the buttons
context for each painting’s figures. Linear incisions closing her red tunic at the wrists, and all of the
in the Kress panel define the straight edges of the paintings in the Michelangelo Associate group
foreground plinth and the top of the wall extend-
ing behind the Michelangelo Associate’s figures.
Incisions cut into the ground of the Vienna
tondo also indicate the intersections of architec-
tural planes as well as the placement of squares
in the picture’s checkered tile floor. Curiously, the
general locations of the reading stand at left and
the figure of John the Baptist at lower right in the
Vienna tondo are marked in the picture’s gesso by
incisions that circumscribe a vertically oriented
ellipse of pictorial surface occupied by each
painted figure. (These incised ellipses might be
later vandalism, but they are not accompanied by
any chipping or damage to the original paint.)
Incisions describing architecture in the Kress
and Vienna paintings are remarkably similar with
regard to their imprecise character; the incised
lines do not meet exactly at corners of planes
where they intersect, but over-shoot the mark in
a like manner in each picture. Furthermore, the
planes that are finally realized in paint are slightly
corrected in position or do not rigidly follow the
preliminary incisions, and the nature of these cor-
rections in the painting phase is quite comparable
in the two images. Since the conception of space
in the Ambrosiana panel has often been thought Fig. 14. The Libyan Sibyl, Michelangelo. Sistine Chapel, Vatican
to be the most advanced or Michelangelesque of Palace, Vatican State. Lunette fresco, after restoration.

have mordant gilt haloes. The presence of delicate
mordant gilt detail on the Michelangelo Associate’s
paintings at first seems oddly retardataire in the
context of the sculptural monumentality com-
monly associated with Michelangelo’s art. How-
ever, Dunkerton has noted that the use of deco-
rative mordant gilding is a persistent feature of
paintings produced by the studio of Ghirlandaio
and that it can also be found on paintings by
Granacci.34 It would be difficult to speculate
about any gilt decoration that may have been
planned for the National Gallery, London’s
Manchester Madonna as mordant gilding is typically
a final step in the creation process and the pic-
ture is unfinished. Nonetheless, there are selected
mordant gilt details enlivening the surface of
Michelangelo’s later Doni Tondo in the Gallerie
degli Uffizi, Florence.35
Another final-stage element that the Kress
panel and the Vienna tondo share is a quarter-
inch-wide, black painted border applied around
the entire perimeter of each image. In areas of
surface wear or pinpoint losses to this border, it
is possible to see that original pigment lies under-
neath. The images appear to have been finished
Fig. 15. Fanciullo Arcière, Attributed to Michelangelo, marble.
out to the edges of the support before this border
was superimposed. These black borders would be
an intriguing topic for further investigation, par-
ticularly with regard to period practice in append- During the late fifteenth- and early sixteenth-
ing frames to the paintings. It could also be useful century period in which the Michelangelo
to investigate whether the painted black border is Associate produced his images, painters were
part of a specific regional tradition. In the last particularly influenced by their study and emula-
decade of the fifteenth century and the opening tion of sculpture, both contemporary and antique.
decade of the sixteenth, similarly painted borders The most obvious manifestation of this in the
frame many of the images by the Bolognese artist Kress panel would be the similarity of the Christ
Francesco Francia and his son Giacomo; they have Child and John the Baptist figures to sculpted
also been observed on late paintings of Perugino antique cupids which were quite popular objects
and early pictures of Raphael. It may be that with artists and collectors at the beginning of the
these borders have not been studied because sixteenth century. It is more than likely that the
they seem to be routinely cropped out of photo- Michelangelo Associate copied this motif from
graphic reproductions of the paintings. In addi- Michelangelo rather than consciously working
tion, the past trimming of panel edges may have after the antique himself (figs. 14 and 15). The
caused many examples to be lost over time. Michelangelo Associate routinely translated
designs by Michelangelo in his compositions,
Concluding Remarks: and Michelangelo himself is known to have gen-
Sculptural Sources and Emulation erously shared his drawings and cartonetti with
Molly March 157

other artists.36 However, if the Michelangelo
Associate was as intimate with Michelangelo as is
currently believed, he may have had direct access
not only to drawings of the master but also to
his in-progress sculptural projects and paintings.37
In some instances, the Michelangelo Associate
may have emulated sculpture by borrowing motifs
directly from or in the presence of Michelangelo’s
reliefs.38 It is tempting to see the opalescent qual-
ity of the Michelangelo Associate’s flesh passages
as an attempt to transcribe light refracted by the
surface of carved marble. It may be equally fanci-
ful to suggest that his signature hatching strokes
in flesh and drapery reflect an aesthetic apprecia-
tion of the parallel scoring lines of a stone chisel
that cover selected surfaces of sculptures such as
Fig. 16. Virgin and Child with the Infant Saint John (Taddei Tondo),
Michelangelo’s Virgin and Child with the Infant Saint
Michelangelo, ca. 1504–05, marble. Royal Academy of Arts, John (Taddei Tondo, fig. 16), or his Pitti Tondo of the
London. same subject (fig. 17).
Parts of the figures portrayed in the Kress
Madonna and Child with Saint John are certainly
derived from specific passages in Michelangelo’s
oeuvre. The Kress Virgin’s head in profile and
the general form of her veil are most definitely a
quotation of the head and veil in Michelangelo’s
Madonna della Scala (fig. 18).39 The deportment
of the Virgin’s hands and wrists in the Kress
painting is resonantly similar to that realized
in the Madonna della Scala, the Pitti Tondo and the
Manchester Madonna (see figs. 2, 3, 17 and 18). The
Michelangelo Associate’s above-waist poses for
the Christ Child and the young Saint John in the
Kress picture correspond quite closely to the
positions of the two unfinished angels in the
upper left corner of Michelangelo’s Manchester
Madonna, and also appear notably similar to the
two painted figures behind the Libyan Sibyl in the
Sistine Chapel (see figs. 2, 3, and 14). The Kress
Virgin’s standing form with outstretched arms
holding a book is initially reminiscent of poses
in Annunciation scenes, and seems oddly incon-
gruous with the Kress image. However, the
position of the Kress Virgin has much in com-
mon with the standing angel at the far right in
Fig. 17. Virgin and Child with the Infant Saint John (Pitti Tondo),
Michelangelo, ca. 1503, marble. Museo Nazionale del Bargello, the Manchester Madonna and is quite similar to the
Florence, Italy. upper part of the seated figure in the lower left

corner of the Entombment (see figs. 2, 3, and 4), Michelangelo Associate’s figures is a result of his
and thus again may simply be a design lifted from disparate borrowings, the whole being a pastiche.
Michelangelo’s imagery. In addition, the effect of shallow space may partly
As Michelangelo’s motifs are, for the most come from his literal interpretation of a model;
part, faithfully repeated by the Michelangelo the head of the Kress Virgin may appear relatively
Associate, a direct comparison between their flat if it was directly copied from the Madonna della
treatments of a figure in space is possible. It is Scala, a stone image sculpted in low relief. The
sufficient to say here that the Michelangelo Asso- Michelangelo Associate’s figures’ disconnection
ciate does not approach Michelangelo’s mastery from their context and each other may also result
of form and seamless spatial transitions. Rather, from his emulation but incomplete realization of
the Michelangelo Associate’s pictures evoke a Michelangelo’s painting sequence. The unfinished
sense of compressed space and cut-out figures. pictures in the National Gallery, London show
The impression that the Michelangelo Associate’s that Michelangelo the painter typically brought
figures are cut-out or isolated from their pictorial individual color areas of his composition to
context may be partially a result of his picture- a high degree of finish at different times (see
making method. As noted above, only sections figs. 3 and 4). If our Associate was in a position
of each figure portrayed in the Kress painting to observe Michelangelo working, he may have
can be linked with a known Michelangelesque attempted to copy this method even though
source. It is possible that the isolation of the he seemingly lacked the painterly skill to achieve
Michelangelo’s ultimately seamless result.
A developed understanding of the unique char-
acter of the Michelangelo Associate’s adaptations
of Michelangelo’s designs will ultimately assist
scholars in discovering his identity. The Ferrarese
or extra-Florentine elongation of figures in his
pictures is affected both by selected passages of
eccentrically realized form apparently drawn
free-hand, and by his inequal distribution of
pigment in areas of drapery and flesh. The
perimeters of figures that can be directly linked
to Michelangelesque models are actually quite
artfully proportioned. However, within these
outlines, the greater density and opacity of this
Michelangelo Associate’s highlights imply atten-
uated shapes within his compositions that over-
whelm the relatively smooth, translucent planes of
middle tone and shadow, creating elongated linear
accents in his pictures that are almost visually
detached from the image portrayed.
Despite the peculiarities of our artist’s
approach to image making, his unique points
of concordance with Michelangelo’s painting
practice should not be underestimated. During
the search for the Michelangelo Associate’s iden-
Fig. 18. Madonna della Scala, Michelangelo, ca. 1490, marble. tity, there may eventually be sufficient technical
Casa Buonarroti, Florence, Italy. evidence to disqualify any artist who did not have
Molly March 159

the most intimate access to Michelangelo as he manship methods
painted. On close inspection, it is possible to see as digested by Michelangelo and passed on to
that Michelangelo himself also structured the our Associate).40 In future investigations of the
highlights in his flesh passages with hatching, Michelangelo Associate’s identity it may also be
though his strokes are infinitely finer and more important to examine the drawings of artists in
smoothly blended together than the autograph the circle of Cosimo Tura; strokes quite similar
hatches of the Michelangelo Associate (see to our artist’s signature hatches can be seen in
figs. 2 and 3). The brick-red color that notably works such as Tura’s Evangelist, and technical
separated the Michelangelo Associate’s palette analyses of the Vienna tondo and the Kress
of flesh colors from that of the Ghirlandaio panel have strengthened the hypothesis that the
studio does appear in some of Michelangelo’s Michelangelo Associate had an early exposure to
paintings; as noted previously, a brick-red hue practices in the Ferrara region.41 Further collec-
similar to that employed by the Michelangelo tion of detailed observations of these paintings
Associate is used by Michelangelo to articulate supported by directed, comparative analysis of
the middle tones and shadows of Adam’s body their materials and rigorous archival research
in the central Sistine Chapel Creation, as well as should one day crystallize an identity and link a
those of figures such as the Libyan Sibyl in the name to the unmistakably recognizable hand of
lunettes. Furthermore, painted brown contour this Michelangelo Associate.
lines such as those that circumscribe the limbs
of figures in the Kress panel can also be seen Acknowledgements
articulating exterior contours of Michelangelo’s Heartfelt thanks are due to Professor Dianne
figures in the Entombment and in the Uffizi’s Doni Dwyer Modestini for assigning me the restora-
Tondo. Michelangelo seems to employ these red- tion of the Kress Madonna and Child with Saint John
brown lines for initial design as well as for a more and for her support throughout this lengthy proj-
advanced strengthening of contours during the ect. It was a distinct privilege to have worked with
painting process, a technique also observed in the both Dianne and Mario Modestini. My deepest
works of the Michelangelo Associate. However, appreciation should also be expressed to Areli
the function of Michelangelo’s finally applied Marina, Robert La France and William Haluska
brown lines is to emphasize the illusion of his for their careful readings and helpful suggestions
figures’ volume, making their limbs seem to during the writing of this paper. Finally, I would
almost protrude from the picture plane, an effect like to gratefully acknowledge Lisa Ackerman,
not matched by the Michelangelo Associate. Executive Vice President of the Samuel H. Kress
In conclusion, it should be re-emphasized Foundation, and the Foundation itself for
that the autograph hatching strokes of the their peerless commitment to funding unique
Michelangelo Associate are fundamentally unique advanced training opportunities in the field of
in their execution and pictorial placement. As Conservation.
such, they may prove to be the most significant
material evidence we have to identify other works
and, one hopes, the artist himself. These hatchings Molly March studied the History of Art at the University
seem to have much in common with similarly of California, Berkeley (B.A.) and New York University’s
placed marks found in period drawings, particu- Institute of Fine Arts (M.A.), and holds a Certificate
larly those of Ghirlandaio, for example, his in Conservation from the Institute of Fine Arts’ Conser-
Drapery Study for a Kneeling Figure in Florence vation Center. Her advanced training was completed
(though the likeness of the Michelangelo with restorers Nancy Krieg and Marco Grassi, and as
Associate’s painted hatches to Ghirlandaio’s drawn a Samuel H. Kress Fellow in the paintings conservation
ones may simply reflect this master’s draughts- departments of the Rijksmuseum, Amsterdam and the

Gallerie degli Uffizi, Florence. She has a private studio results will be published at a later date. For Dunkerton’s
comment on morellone, see Hirst and Dunkerton (1994),
for the restoration of Old Master paintings in New York p. 102. The Vienna tondo pigment analysis is unpublished
and specializes in Italian pictures. as of this printing.
9. It is important to compare the Ambrosiana panel in its
most current state to the Kress and Vienna images. An
old photograph in the Contini-Volterra Archive at Vander-
Notes bilt University shows the Ambrosiana picture when it was
1. Zeri (1953). quite heavily restored. Any studies based on the appearance
2. Hirst and Dunkerton (1994). of this old photo of a now-absent restoration campaign
3. For illustrations of the ex-Baden bei Zurich painting and would be misleading as the figures’ flesh, the left-hand
the Barberini Pietà, see Freedberg (1972), Vol. 2. figs. 335 architectural forms and the background landscape were
and 334, respectively. A color reproduction of the Pietà once extensively repainted, changing the palette and style
appears on p. 39, plate 26 of Hirst and Dunkerton (1994). of this image entirely. Many thanks to Joseph Mella,
4. Though these paintings are often discussed together, opin- Director of the Vanderbilt Fine Arts Gallery, for facilitat-
ions differ as to whether they are all by the same artist. To ing our visit to the archive.
this viewer, the Barberini Pietà and the Madonna and Child, 10. The term “opalescent” was aptly used previously by
ex-Baden bei Zurich, seem stylistically quite similar to each Dunkerton in Hirst and Dunkerton (1994).
other, for example, in the contorted flex of the hands por- 11. Compare the coloration of flesh in the National Gallery,
trayed and the pinched features of the faces. The Kress London’s Ghirlandaio, The Virgin and Child (n63939) to
panel, the Vienna tondo, and the Ambrosiana’s Madonna and the Virgin and Child with Two Angels (n6296), attributed to
Child do not share these qualities but are extremely close Andrea del Verrocchio in the same collection. The brilliant
to each other in terms of surface character, palette, and colors characteristic of the Ghirlandaio studio may be
execution. In the past, the Ambrosiana picture has been associated with a persistent use of pure tempera technique
noted as possessing a sophisticated construction of space in this workshop during years when oil painting was
that is not present in the Kress and Vienna paintings. already quite widely practiced (see Hirst and Dunkerton
However, this might be explained by the Michelangelo (1994), p. 84). Joyce Plesters has associated Verrocchio
Associate’s use of another artist’s more advanced cartoon and his followers’ warmer, subtler technique of modeling
or model for the Ambrosiana image; the surface quality, flesh with the practice of buon fresco, where the light color
brushwork, and coloration of the cleaned Ambrosiana of the ground seen through applied color washes affects
painting are extremely similar to those in the Kress panel “all but the strongest highlights.” See Plesters in National
and the Vienna tondo. Gallery (1970), p. 27.
5. See for example: Freedberg (1972), Vol. 1, pp. 255–8. 12. Roughly concurrent with the ascendance of oil painting
6. Dr. Martina Fleischer of the Gemäldegalerie der Akademie as a popular technique, most Ghirlandaio pupils such as
der Bildenden Künste is preparing a manuscript on the Granacci seem to have moved away from the bright, pure-
Vienna tondo, and I would like to thank Dr. Franz color modeling of flesh in their later works. (For example,
Maringer for several hours of valuable discussion regard- see the beautiful Granacci, Madonna and Child recently
ing this painting, as well as Professor Norbert Baer of the given to the Metropolitan Museum of Art, New York by
Conservation Center of the Institute of Fine Arts for Mario and Dianne Dwyer Modestini.)
introducing me to Dr. Maringer. For the first part of the 13. A curious occurrence of this white-blue/green-pink mod-
technical information to be published about the Vienna eling of flesh can be seen in the work of the Abruzzese
tondo, see Maringer (1996). For Cosimo Tura’s technique, artist, Saturnino Gatti, painting in 1510. See the Metro-
see articles by Dunkerton and Marcello Toffanello in politan Museum of Art’s Gatti, The Translation of the Holy
Campbell (2002). For an earlier assessment of Tura’s House of Loreto, 1973.319. For one version of the Roman
technique, see Dunkerton (1994). dates of Pinturicchio, Granacci, and Michelangelo, see
7. During the course of the Kress painting’s restoration, a Gallwitz (1999), pp. 120, 136, 148.
heavy, “Pichetto” cradle was removed from the panel’s 14. Even so, in Milan there is an Adoration of the Child currently
reverse. This allowed the plank to acquire a very slightly attributed to the workshop of Ghirlandaio in which traces
convex, natural curve, which seems to have alleviated much of a brick-red hue are scantly employed for flesh middle
of the internal stresses that caused the past instability of tones (Pinacoteca Ambrosiana, Inv. Nr. 2. Brivio Donation,
the paint layers adjacent to the two vertical joins; see Ann 1959). As will be noted below, though this brick-red hue
Hoenigswald’s article in this volume. is not routinely employed by Ghirlandaio, if at all, it
8. In the Making and Meaning study, Dunkerton describes can be found in shaded flesh passages of paintings by
a soft lavender color typically used in frescoes of this Michelangelo, for example, in the recently cleaned Sistine
period as “morellone.” Though she rightly speculates Chapel figures of the Libyan Sibyl and Adam (see fig. 14).
that in easel painting the color is often derived from a 15. This lack of continuous opacity or consistent film thick-
mixture of lead white, red lake, and a blue pigment, ness across flesh passages of the Michelangelo Associate’s
Caput Mortuum has been identified as a coloring com- paintings may be slightly over-emphasized today due to the
ponent of the architectural planes in the Vienna tondo. apparent susceptibility of his surfaces to mechanical and
Libby Sheldon of UCL Paintings Analysis, University chemical abrasion. Nonetheless, as will be described, the
College, London is currently studying the pigments transition from dense, thickly applied strokes in highlights
used in the creation of the Kress painting, and these
Molly March 161

to a smooth, barely articulated surface in deep shadow is Analysis, London, in her study of paint samples from the
also a feature of the Michelangelo Associate’s original Kress robe.
painting technique. 25. Dr. Renate Trnek, Director of the Gemäldegalerie der
16. For discussion of Tura’s technique of heightening or Akademie der Bildenden Künste, generously granted per-
adding lights to his underdrawing, see comments through- mission for me to inspect the cleaned-state photographs
out Campbell (2002), especially p. 127. of the Vienna tondo.
17. Due to the nature of the pigments used by the Michelangelo 26. Following my departure from the Samuel H. Kress
Associate for flesh tones, the differences in thickness of Program in Paintings Conservation in May 2002, Dianne
application between highlight and shadow are fortuitously Dwyer Modestini applied final glazes to the highlights of
diagrammed by an X-radiograph of the painting (see the Kress panel’s blue mantle and flesh, and added final
fig. 12), providing a map of this technique. For future touches to the landscape and the halo of Saint John. The
investigations of attribution, comparative study of the painting’s restoration was deemed complete; the picture
X-radiograph images of pictures that may be by this was placed in a new vitrine and returned to the Kress
Michelangelo Associate, as well as those of paintings by Foundation in October 2002.
Tura and other non-Florentine artists might be useful. 27. Dunkerton in Hirst and Dunkerton (1994), pp. 90–91.
18. See Marcello Toffanello’s wonderful essay, “Cosmè Tura: Several illustrations of these craters appear on page 90 of
drawing and its pictorial complements” in Campbell (2002), this catalogue (plates 66, 67, 68, 69). It would be interest-
pp. 153–72. The painted draperies of Andrea Mantegna are ing to confirm whether these craters are actually in the
consistently heightened along fold edges with concentrated gesso layer, or rather, are within a cream-colored impri-
touches of opaque pigment or gold. Mantegna’s articula- matura selectively laid over the warm-toned gesso in areas
tion of edges in turn may be related to the art of Giovanni intended to be relatively light or pale in the final design.
Bellini. Please see the gold heightening applied in hatches 28. Again, these preliminary results are the wonderful work
that illuminates the blue robe of Christ in Bellini’s Agony in of Libby Sheldon, ucl Painting Analysis, London who is
the Garden, circa 1465, n6726, National Gallery, London, as completing the examination of cross-sections from the
well as Andrea Mantegna’s Adoration of the Shepherds, 32.130.2, Kress Michelangelo Associate panel.
Metropolitan Museum of Art, New York. 29. See Maringer (1996).
19. The only exception to this in the Kress painting would be 30. Fleischer and Maringer (1990).
the green fringe that sporadically trims the red drapery in 31. See Dunkerton in Hirst and Dunkerton (1994), pp. 94–5.
the upper register. The strokes that form this fringe are Dunkerton also mentions that a high glue content, effec-
similar in scale and appearance to those accenting the blue tively an aggressive sealing of the gesso layer, would be
mantle and creating mass in the flesh. However, in the case quite expected if the artist were preparing to paint in oil,
of the curtain, each stroke is actually a string of fringe. and that Ferrara was one of the earliest centers of oil
The function is much more direct or decoratively literal, painting development in Italy.
as opposed to the roles of hatching strokes indicating light 32. Dunkerton in Hirst and Dunkerton (1994), p. 92 and Marin-
in the blue mantle and volume in the flesh passages. ger in a personal communication during my visit to Vienna.
20. Travel to Vienna, Milan, and Rome was funded by the 33. It is interesting to note that in the Ambrosiana’s Madonna
Samuel H. Kress Foundation as part of a Fellowship for and Child, hints of a summary, blackish preparatory draw-
Advanced Training in Paintings Conservation at the Con- ing can be seen under gallery lighting conditions around
servation Center of the Institute of Fine Arts, New York. the edges of forms such as the Virgin’s hands and wrists.
21. I must extend warm thanks to Dr. Renate Trnek, Director, 34. Dunkerton in Hirst and Dunkerton (1994), p. 102.
and Professor Peter Halbgebauer, Chief Restorer, of the 35. For the location of gilt details on the Doni Tondo, ibid.
Gemäldegalerie der Akademie der Bildenden Künste 36. See Wilde (1959).
for allowing me access to the painting and for helpfully 37. For Michael Hirst’s recent suggestion that Michelangelo’s
commenting on its restoration, which was accomplished personal assistant and friend, Piero d’Argenta, may be
under great time pressure in preparation for the London the Master of the Manchester Madonna or Michelangelo
exhibition. Associate, see Hirst in Hirst and Dunkerton (1994), p. 41,
22. The existence of old retouches in between the Michelangelo and Agosti and Hirst (1996).
Associate’s strokes in the flesh tones of the Vienna tondo 38. A pentimento of the Kress Virgin’s neckline that was
prior to the most recent cleaning was noted in a report on revealed in an X-radiograph image of the panel may reflect
the picture from the Museum’s files, generously shared the Michelangelo Associate’s awareness of Michelangelo’s
with me by Dr. Franz Maringer in his Vienna offices. sculptural works. The Virgin’s red robe in the Kress
23. This hot brown glaze may be altered in color from its picture was initially designed with a square neckline that
original appearance. The pigments employed have not yet was subsequently changed during the painting process to
been analyzed. the “V”-shaped neckline. This pentimento might suggest
24. During cleaning of the Kress picture, a curious, relatively a relationship between the Kress picture and a shift in
under-bound, opaque, dull green could be seen sporadi- tasteful fashion recorded by the different necklines in
cally throughout the damaged mantle, sometimes in Michelangelo’s Taddei Tondo, ca. 1502 and his Pitti Tondo,
association with or as a preparation for another very lean ca. 1503 (see figs 12, 16, and 17). It is known that
blue. These color layers were subsequently found to con- Michelangelo was keenly aware of contemporary fashions
tain the pigment blue ochre and identified as eighteenth- of dress; his later painting, the Doni Tondo, also shows an
century restorations by Libby Sheldon of ucl Painting up-to-date V-neckline in the costume of the Madonna.

39. The head of the Madonna in Michelangelo’s Madonna della Munich/London/New York: Prestel Verlag.
Scala seems generally derived from Donatello’s famous Gregori, M., Paolucci, A. and Acidini Luchinat, C. (eds.)
relief, the Pazzi Madonna in the Staatliche Museen zu Berlin. (1992) Maestri e Botteghe. Pitture a Firenze alla Fine del
However, where Donatello fully describes the Virgin’s ear Quattrocento. Milan: Amilcare Pizzi SpA.
in his image, Michelangelo chose to cover her ear with Hirst, M. and Dunkerton, J. (1994) Making and Meaning: The
drapery. In the Kress panel, the Michelangelo Associate has Young Michelangelo. London: National Gallery Publications.
drawn from Michelangelo’s design with the ear concealed. Maringer, F. (1996) “The Ferrarese tondo of the Vienna
Thus, though Freedberg once described the Michelangelo Academy: scientific examination of the panel support and
Associate as an artist who seems to have spent many hours ground (1).” Restauratorenblätter Vol. v, No. 19. pp. 81–2.
studying Donatello, it may be that this hint of Donatello National Gallery (1970) Michelangelo’s Entombment of Christ:
in the Michelangelo Associate’s works actually was passed Some New Hypotheses and Some New Facts. London: National
to him through Michelangelo. See Freedberg (1972), Gallery Publications.
p. 256. (The Madonna della Scala has also been likened to Rossi, M. and Rovetta, A. (1997) La Pinacoteca Ambrosiana.
Donatello’s Dudley Madonna in the Victoria and Albert Milan: Electa.
Museum, London. Donatello did cover the Madonna’s ear Ruhemann, H. and Plesters, J. (1964) “The technique of
with drapery in this relief, however its underlying form painting in a ‘Madonna’ attributed to Michelangelo,”
is clearly indicated in the veil fabric, a detail mimicked The Burlington Magazine, Vol. cvi, pp. 546–54.
by Michelangelo in the Madonna della Scala but not by the Shapley, Fern Rusk (1973) Paintings from the Samuel H. Kress
Michelangelo Associate in the Kress painting.) Collection, Vol. : Italian Schools, XVI–XVIII Century. London:
40. Ghirlandaio, Drapery Study for a Kneeling Figure. Inv. 316e Phaidon.
(as Mainardi), Gabinetto dei Disegni, Gallerie degli Wallace, W (1995)“Instruction and originality in
Uffizi, Florence. Michelangelo’s drawings” in A. Ladis and C. Wood (eds.),
41. Cosmè Tura, Evangelist. Gabinetto dei Disegni, Gallerie The Craft of Art. Athens, GA: University of Georgia Press.
degli Uffizi, Florence. This drawing is reproduced in Wilde, J. (1959) “Cartonetti by Michelangelo,” The Burlington
Campbell (2002), p. 161, fig. 73. Magazine, Vol. ci, pp. 370–81.
Zeri, F. (1953) “Il Maestro della Madonna di Manchester,”
Paragone, Vol. xliii. pp. 15–27.
Bibliography
Agosti, G. and Hirst, M. (1996) “Michelangelo, Piero
d’Argenta and the Stigmatization of Saint Francis,” The Photography Credits
Burlington Magazine, Vol. 138, No. 123, pp. 683–4. Figs. 1, 2, 7, 9, 10, 12, and 13, pp. 144, 147, 149, and 151. Samuel
Brandt, K. (1996) “A marble in Manhattan: the case for H. Kress Foundation, New York, NY (k-1569).
Michelangelo,” The Burlington Magazine, Vol. 138, pp. 644–59. Figs. 3 and 4, p. 146. ©National Gallery, London.
Brandt, K., Luchinat, C., Draper, J. and Penny, N. (1999) Fig. 5, p. 146. Courtesy of Gemäldegalerie der Akademie der
Giovinezza di Michelangelo. Florence and Milan: ArtificioSkira. bildenden Künste, Vienna.
Cadogan, J. (1993) “Michelangelo in the workshop of Fig. 6, p. 146. ©Biblioteca Ambrosiana, Milan, Auth.
Domenico Ghirlandaio,” The Burlington Magazine, Vol. No. f129/05.
cxxxv, No. 1078, pp. 30–31. Fig. 8, p. 148. The Metropolitan Museum of Art, New York,
––– (2000) Domenico Ghirlandaio, Artist and Artisan. New Haven, NY, The Friedsam Collection, Bequest of Michael
CT and London: Yale University Press. Friedsam, 1931 (32.100.71). Photograph © 1990 The
Campbell, S. (2002) Cosmè Tura, Painting and Design in Renaissance Metropolitan Museum of Art.
Ferrara, ed. A. Chong. Milan: Elemond SpA and Electa. Fig. 11, p. 150. Photograph by Erich Lessing ©Art Resource,
Carpegna, N. (ed.) (1957) Catalogue of the National Gallery Barber- Inc., New York, NY.
ini Palace Rome. Rome: National Gallery Barberini Palace. Fig. 14, p. 156. Photograph by Erich Lessing ©Art Resource,
Della Pergola, P. (1954) “La Madonna di Manchester nella Inc., New York, NY.
Galleria Borghese,” Paragone, Vol. v., pp. 47–8. Fig. 15, p. 157. Photograph by Robert Lorenzson with the per-
Dunkerton, J. (1994) “Cosimo Tura as a painter and draughts- mission of the Cultural Services of the Embassy of France
man: the cleaning and examination of his Saint Jerome,” in the United States.
National Gallery Technical Bulletin, 15, pp. 42–53. Fig. 16, p. 158. ©Art Resource, Inc., New York, NY.
Dunkerton, J., Roy, A. and Smith, A. (1987) “The unmasking Fig. 17, p. 158. Photograph by Scala ©Art Resource, Inc.,
of Tura’s Allegorical Figure: a painting and its concealed New York, NY.
image,” National Gallery Technical Bulletin, xi, pp. 5–35. Fig. 18, p. 159. Photograph by Scala ©Art Resource, Inc.,
Fiocco, G. (1941) “Sul inizio di Michelangelo,” Le Arti, IV, New York, NY.
pp. 5-10. 1
Fleischer, M. and Maringer, F. (1990) “Der Amor in der
Wiener Akademiegalerie” in R.E. Straub (ed.), Die Künst
und ihre Erhaltung. Worms: Wernersche Verlagsgesellschaft,
pp. 148–68.
Freedberg, S. (1972) Painting of the High Renaissance in Rome and
Florence, 2 vols. New York: Harper and Row.
Gallwitz, K. (1999) The Handbook of Italian Renaissance Painters.
Molly March 163

Portrait of a Lady
Nicolaes Maes, 1682
Oil on canvas
47 3/4 × 39 3/4 in. (121.2 × 101 cm)
cma 1962.17 (k-1134)
Fig. 1. Portrait of a Lady, before cleaning and restoration.
Fig. 2. Portrait of a Lady (fig. 1), after cleaning and restoration.

Portrait of a Lady and Techniques in the
Late Paintings of Nicolaes Maes
Laurent Sozzani with Christopher McGlinchey
T he Kress Collection painting by Nicolaes Maes, Portrait of a Lady (figs.

1 and 2), signed and dated 1682, which is now in the Columbia Museum of
Art in Columbia, South Carolina, is examined here in the context of other
late paintings by Maes. Nicolaes Maes was a premier portrait painter in
Holland in the second half of the seventeenth century. Painted during
the period in which he devoted himself exclusively to portraiture, the Columbia
painting is typical of Nicolaes Maes’s later mature style, illustrating his facility
for capturing a likeness and rendering rich drapery and background elements
with technical economy. Portrait painting techniques and studio practices are
discussed in some detail. And an unusual glazing technique used by Maes on
numerous paintings including the Columbia Portrait of a Lady is discussed here
for the first time.
Nicolaes Maes
Nicolaes Maes was born in Dordrecht in 1634. The artist and biographer,
Arnold Houbraken (1660–1719), in the closest contemporary account of
Nicolaes Maes’s life, recorded in the second volume of his three-volume De
Groote Schouburgh der Nederlantsche Konstschilders en Schilderessen (1719)1 that “Maes’s
early training in drawing was with an unknown ordinary Dordrecht master,”
and that later, still as a young man, perhaps between 1646 and 1650, he traveled
to Amsterdam where “from Rembrandt he learned painting.”2 Exactly how long
he spent with Rembrandt is unknown; however, through Maes’s marriage it is
Laurent Sozzani with Christopher McGlinchey 165

documented that he had returned to Dordrecht by the martyred.”6 Even if the visit took place, it is
1654, where he worked as an independent painter. uncertain whether Houbraken could really have
In 1673, he moved back to Amsterdam, to live and known of the conversation. Rather, Houbraken
work until his death in 1693.3 seems to use the conversation to both acknowl-
During his early period, Maes ranked among edge Maes as a portrait painter whom he admires
the most innovative of Dutch genre painters, and as an opportune way to expound on the pre-
depicting interior scenes not only within the usual vailing theory of the hierarchy of pictorial genres.
simple, three-walled spatial arrangements, but also Though the Antwerp trip has been considered
in suites of rooms that were better suited to his a major reason for Maes’s shift in style, the
intended narratives of the intrigues of everyday change is more likely derived from the influence
domestic life. His numerous depictions of eaves- of Adrian Hanneman (1601–1671) and Jan Mijtens
droppers, painted between 1655 and 1657, best (1614–1670) among others, who were Dutch
exemplify this compositional arrangement. This proponents and followers of what was then
design innovation exercised a decisive influence on becoming an international “van Dyckian” style.
Delft painters such as Pieter de Hooch (1629–1683) Both Hanneman and Mijtens worked in The
and Johannes Vermeer (1632–1675), and had a last- Hague, not far from Dordrecht, and both worked
ing impact on seventeenth-century Dutch paintings primarily as portrait painters. Their portraits
of interiors. Rembrandt’s influence is recognizable relied heavily on Anthony van Dyck (1599–1641),
in Maes’s early genre paintings in the use of iso- and Hanneman especially is credited with playing
lated areas of opaque color and a rich chiaroscuro a major role in disseminating van Dyck’s influ-
to render forms and their spatial relationships.4 ence throughout Holland.7 During his stay in
The use of a restricted palette, rich in browns and Amsterdam, Maes would also have had the
reds, is also characteristic of these early paintings. opportunity to become familiar with the work of
His early portraits, those from the 1650s, also Jan Lievens (1607–1674), Bartholomeus van der
show a degree of indebtedness to Rembrandt. Helst (1613–1670), Ferdinand Bol (1616–1680), and
However, by the 1660s Maes had moved away Govert Flink (1615–1660), who were also incor-
from Rembrandt’s style and what he may have porating the Flemish style into their work.
considered a more staid type of portraiture and By the time he returned to Dordrecht around
began to develop a more “van Dyckian” style. 1654, Nicolaes Maes had begun to devote himself
In Houbraken we read that Maes “learned … to portrait painting. It is not difficult to imagine
painting from Rembrandt but soon gave up this that the younger Maes, in pursuing his career and
way of painting, particularly when he devoted looking to a more classicist style, would have
himself to portraiture and saw that young girls, come under the influence of other successful por-
especially, take more pleasure being shown in traitists and begun painting in a style that has
white than in brown.”5 since been considered a reflection of the national
Houbraken also records that Maes had a good spirit of the times.8 By the 1660s Maes devoted
character, worked hard and was serious in devel- himself exclusively to portrait painting. These
oping himself as an artist; he adds that Maes vis- portraits depict the well-to-do, usually dressed
ited Antwerp where he was able to see the works in the formal costume of the day and set in
of Rubens, van Dyck, and other masters. During sumptuous, often brightly colored backgrounds.
that visit he also met with artists including Jacob They began to become standardized, depicting
Jordaens (1593–1678). This visit is enlivened with different sitters in similar poses, with similar
a conversation in which Jordaens asks Maes, details in clothing, attributes, and surroundings.
“What do you make?” and upon Maes’s reply An overview of the body of Maes’s early genre
of “I am a portrait painter,” Jordaens comments, paintings reveals a limited number of subjects and
“Brother, I have pity on you, for you are one of their frequent repetition.9 One can imagine the

ease with which Maes would later develop as a years earlier more correctly credits van Dyck as
portraitist content with repeating similar pictorial the “First Painter that e’er put Ladies dresse into
elements over and over. Such repetition would a careless Romance.”12 This sort of dress is seen
have been a necessary requirement for a prolific in the Maes portraits of Simon van Alphen (fig. 3)
production. It is just this sensibility that con- and another Portrait of a Lady, possibly Mary
tributed to the lower status given to portraiture Stuart (fig. 4).
in the theory of painting hierarchy alluded to by In 1673 Nicolaes Maes returned to Amsterdam
Houbraken when he wrote about Maes’s visit with where his success as a portrait painter continued
Jordaens. In comparison to the creative freedom a unabated until his death in 1693. As Houbraken
history painter had in depicting noble and learned further recorded:
subjects, the portrait painter’s subordination of
Having settled in Amsterdam, so much work
the imagination in capturing the likeness of the came his way that it was deemed a favor if one
sitter and the constant repetition of settings is person was granted the opportunity to sit for his
one reason for the differences in status. portrait before another, and so it remained until
Using a system described here as a fill-in meth- the end of his life, which is why he left a large
od, a painter might have a few different standard- number of portraits incomplete [at his death].13
size pre-prepared supports, any number of stock
poses, and a variety of standard drapery types and In conjunction with this success, Maes utilized a
backgrounds that could be adapted according to variety of techniques that allowed him to rapidly
the wishes and budget of the client. In a limited achieve the desired luminosity and rich finish of
number of sittings, organized both for the com- his pictures. His working method may also have
fort of the sitter and the demands of the painting contributed to the number of incomplete works
technique (such as the need for a paint layer to in his studio after he died.
dry before subsequent paint can be applied), the
face and possibly the hands would be completed.
Then the painting could be finished through the
addition of the costume and background without
the sitter’s presence.10 Though his style continued
to change, it appears that Maes worked in this
way for the rest of his career.
By the 1670s Maes had developed a new por-
trait style in which very elegantly and colorfully
dressed sitters were placed against architectural
elements often leading to a view of a pastoral
landscape. Changes in costume types typify this
shift. Under the influence of English and Flemish
fashions, Maes begins to dress his sitters in what
was later described by Gerard de Lairesse in his
Het Groot Schilderboeck (Amsterdam, 1707) as “the
Painter-like or antique manner, but by the igno-
rant Commonalty, the Roman Manner,” which
“signifies, a loose, Airy Undress, somewhat favor-
ing the Mode, but in no wise way agreeing with
the ancient Roman Habit.”11 Lairesse credits Lely
Fig. 3. Portrait of Simon van Alphen, Nicolaes Maes, ca. 1677,
as the originator of the “antique manner” of oil on canvas, 28 1/8 × 22 1/2 in. (71.5 × 57 cm). Rijksmuseum,
dress, whereas Sir William Sanderson writing fifty Amsterdam.

composition. A table at her left and a chair, par-
tially visible on her right, are also painted in red.
All architectural elements are in muted browns.
The subdued tone of the compositional elements
and background contribute to the possibility of
her being a widow.
The Portrait of a Lady was examined at the
Columbia Museum of Art, while the majority
of the other Maes paintings discussed here were
examined at the Rijksmuseum, Amsterdam.
The Painter’s Technique

Support
The Portrait of a Lady is painted on a single piece
of plain weave linen canvas now measuring
121.2 × 101 cm. Though he occasionally painted
on wood supports, Maes favored canvas for his
paintings, both large and small. The large format
of this painting—one of the larger portraits in
Fig. 4. Portrait of a Lady (possibly Mary Stuart), Nicolaes Maes, his oeuvre—implies an important commission.
ca. 1677, oil on canvas, 27 × 22 1/4 in. (68.7 × 56.5 cm). Timken The size corresponds closely to a contemporary
Museum of Art, San Diego, CA. English standard size for three-quarter standing
or sitting portraits, 127 × 101 cm.16
Comparing the sizes of approximately 300
The Kress Collection PORTRAIT OF A LADY portraits by Maes from after 1655, estimated as
The Kress/Columbia picture, painted in Maes’s one-third of his total output,17 we see that only
mature style, is signed and dated “N Maes 1682.” thirty-two paintings are larger. Of these, only six
The identity of the sitter, an older woman of are significantly larger, and only two of the six
some means, is unknown.14 It is likely that she portray single sitters. Another thirty paintings are
is a widow, recognized as such by the mourning only slightly smaller. This group of approximately
clothes she wears, which include a widow’s peak sixty paintings is relatively close in size to the
of black lace (tipmuts in Dutch) partially covering Columbia painting. Together they create a cluster
her hair and perhaps the single golden band on of pictures with one dimension measuring near
the index finger of her right hand. A white blouse 107 cm and the other being proportional. In
and a gray scarf accent a black silk or satin dress. seventeenth-century Holland, cloth was woven
The absence of lace in the white blouse could on loom widths measured in units of ells and/or
indicate that she is in a later stage of mourning.15 half ells; 107 cm approximates the seventeenth-
It is also possible that, as an older woman, she is century measuring unit for cloth 1.5 ells wide. This
dressing in the fashion that was prevalent when was the width of a standard bed sheet in Holland
she was young. and one of the most common loom sizes in use
To her left, an open view of the sky indicates in the seventeenth century.18 Using the entire
that she is either in front of a window or on width of cloth as one of the dimensions for a
an open balcony. Muted reds and browns domi- painting would minimize waste, and standardizing
nate the background. A dark red curtain billows sizes could further reduce the time and material
across the top and down the right side of the needed to prepare the canvas for painting.19

Preparation
The Columbia picture was prepared with a single
ground layer. An admixture of earth colors and
black and white pigments resulting in a warm buff
color was used. This color is common not only
in other Maes paintings, but also in the period.
It is often found alone or over a first ground that
would have been applied to fill and reduce the
texture of the weave of the fabric support.20 The
tone of the ground plays an important part in
the final appearance of the painting where thinly
applied upper layers rely on the luminosity of the Fig. 6. Portrait of a Lady (fig. 1), detail showing opaquely painted
ground color. This is often the case in quickly or shadow under the sleeve of the blouse on the proper left arm.
loosely rendered backgrounds and is true of the
architectural elements in the Columbia picture.
Where visible, as in shadows or where a reserved light modeling and glazing with a semi-transpar-
area was left unpainted, the ground may be only ent dark color over the ground. The shadow of
lightly or partially scumbled or glazed over. This her proper right arm under the sleeve of the
slight addition of an overlaid color integrates these blouse is also only lightly covered ground color,
areas into the composition. The shadows of the but the same shadow on her proper left arm is
neck and nose of the sitter are created with only opaquely painted in a warm pink (figs. 5 and 6).
Underpainting
Infrared reflectograms of other Maes paintings
show what appears to be a broadly applied under-
painting for the portrait.21 Though not read as
a detailed sketch (detailed linear underdrawing
in a Maes portrait has yet to be observed), the
image visible in infrared reflectography (irr)
may have served as the primary rendering for the
position, basic form and proportions of the com-
position. Some broad underpaint lines can also
be seen in costumes. Maes could have made a
more detailed sketch in chalk or another material,
such as a thin umber paint, that is neither visible
in the finished picture nor with irr. The lack
of detailed preparatory drawings for portraits
by Maes suggests that he sketched his designs
directly onto the prepared canvas. Although no
examination with irr was carried out on the
Columbia painting, in the face of the Portrait of a
Lady, a dark tone is slightly visible through thin-
ner areas of the flesh paint and through open
cracks giving the impression of the presence of
Fig. 5. Portrait of a Lady (fig. 1), detail of shadow under the
sleeve of the blouse on the proper right arm created by light an underpainting. Similar underpaint can be seen
semi-transparent scumbling over ground. in the face, neck, and chest in a smaller painting

by Maes, the portrait of Belchje Hulft (fig. 7). An
irr image of the Hulft portrait (fig. 8) gives a
clearer idea of what is partially visible to the
naked eye in both paintings. A large paint loss
in the same painting clearly reveals the underpaint
used to indicate the oval surround of that portrait
(fig. 9). Similar underpaint can also be seen in
an unpainted area of the iris of Hulft’s proper
right eye.
Portraiture
Maes’s portraits are carefully rendered and give
the impression of having been painted quickly,
wet-in-wet, with careful blending of every brush-
stroke. Though many wet-in-wet passages can be
found, especially in the costumes and background,
it is probable that the build-up in the face is the
result of a patient paint application, in which
delicate hatching was used to blend together pre-
viously applied patches of color. When they were
dry, or at least partially dry, Maes could repeat
that process using similar colors, again hatched Fig. 7. Portrait of Belchje Hulft, Nicolaes Maes, ca. 1680, oil
on canvas, 17 1/2 × 13 in. (44.5 × 33.1 cm). Rijksmuseum,
together, and then further integrated with final Amsterdam.
glazes and scumbles.
Portrait sittings at the time were shorter than
one might imagine: a portrait could be painted in
three to four sittings of one to two hours. The
sittings would be spaced to allow time for paint
to dry sufficiently before the next sitting.22
Though scant, period descriptions of portrait
painting do provide insight into the techniques
used to achieve the final appearance in a portrait.
Kirby Talley, in his thesis, Portrait Painting in England:
Studies in the Technical Literature Before ,23 has
distilled and collated the notes of many early
writers into a useful reference. Many of the
portrait painters working in England during the
seventeenth century were from the Continent,
and their techniques would be those they learned
at home. In general, Dutch painting techniques
such as those used by Nicolaes Maes are similar
to those of artists who traveled to work in
England.24
By correlating contemporary sources with
Fig. 8. Portrait of Belchje Hulft (fig. 7), infrared reflectogram
observations of related portrait paintings, a gen- composite showing broad underdrawing of the face and
eral description of the portrait painting process chest as well as finer sketched lines for the costume.

can be pieced together. Over the broad but accu-
rate guide of the underpainting (or underdraw-
ing), the first application of the flesh tones was
made. Often referred to as the dead coloring, this
paint would be applied in broad patches of light
and shadow placed adjacent to one another, in
colors closely approximating the intended mod-
eled final tones. These brushstrokes were laid
next to, not over, each other just as Sir Peter Lely
(1618–1680) advised a Mr. Fever, “lay on your
patches of colouring one by another & not colour
upon colour, & only hack them together & keep
them beautiful and clear.”25 Careful hatching was
an important step addressed by many authors.
Daniel King in his Secrets manuscript in a section
on oil painting, states that after laying in the sep-
arate patches of flesh tones for the basic shadows
and highlights, the areas would be carefully
blended “by which means the colors incorporate
& the faintness of the colors is taken away,”26 that
is, the differences between these patches of color
resulting from the quick application would be
reduced. This process is called sweetening and is
done to blend the colors, creating an effect as if
“they were all laid on at once and not at several
times.”27 Marshall Smith, a gentleman writer, in Fig. 10. Self Portrait, Casper Netscher, ca. 1670, oil on oak
The Art of Painting,28 printed in 1692 in London, panel, 11 × 8 5/8 in. (28 × 22 cm). Rijksmuseum, Amsterdam.
Detail showing the palette and many brushes.
warns that in hatching the colors into one another
you must use “a light Hand, taking great care
that you strike not the Shaddows so far, to foul
the Lights, nor the Lights so as to Injure the Shaddows.”29 Sweetening was facilitated by the
use of many paint brushes, one for each different
color—each one kept clean for the use of a
different color. Indeed, one sees that numerous
brushes are routinely represented in addition to
the artist’s palette when these are depicted in self-
portraits or portraits of artists (fig. 10).
Once dried, the dead coloring could be
smoothed and rubbed “very thin over with a
mixture of Nut-Oyle and varnish, for too much
will change the Colour”30 and the painting
process repeated, with even greater care, to pro-
duce the desired final effects. The initial broad
paint application can often be discerned by
Fig. 9. Portrait of Belchje Hulft (fig. 7), detail of underpaint identifying bolder brushstrokes where the colors
indicating the oval surround seen through a paint loss. differ more distinctly from one another. The

brushstrokes of paint. This differs significantly
from the fine blending of the colors visible in the
finished painting. In the X-radiograph, the appli-
cation of lead-rich flesh color is seen as less dense
in shadow areas where the ground and/or under-
paint played a greater role in the modeling.
In Maes’s pendant portraits of Elizabeth van
der Meer and Maarten Pauw (figs. 13 and 14),
the initial broad patches of flesh-color paint are
visible on the surface. The colors are not as fully
blended together. This is especially so at the con-
Fig. 11. Portrait of Belchje Hulft (fig. 7), detail, photomicrograph tours of the shadows.
of the proper right eyebrow showing the fine hatching used to In Maes’s paintings, finished details such as
blend the flesh tone into the shadow of the eyebrow. eyes, nose, and lips that may have been partially
indicated as areas of reserve as a portrait pro-
gressed, were finished only shortly before the sur-
finer brushstrokes used in the sweetening in most face completely dried. As the face was finished the
paintings are difficult to see but occasionally fine hair was laid in. In the X-radiograph of the face
hatching can be found leading into and out of of the Columbia portrait, a slight outline of the
shadows. Over time the fine brushstrokes may be outer form of the hair and an extension of the
even more difficult to observe if the impasto has forehead to the right are visible. This outlining
been softened and flattened through aggressive
restoration interventions. In a photomicrograph
from the Portrait of Belchje Hulft, it is possible to see
some brushstrokes of fine hatching as well as the
scattered pigments of final glazes (fig. 11).
An X-radiograph can be helpful in revealing
the brushwork and deciphering patterns of paint
application. Contemporary descriptions of pal-
ettes prepared for portrait painting often refer to
the tempering of a primary white mixture with
other colors to create the different tones needed.31
Willem Beurs, the author of De Groote Waereld in
‘t kleen geschildert, published in Amsterdam in 1692,
devotes a short chapter to the painting of “living
persons.” 32 In describing a simplified palette for
this, Beurs lists thirteen color mixtures,33 nine of
which contain white—most likely lead white. The
presence and relative thickness of lead white is
primarily what is read in an X-radiograph.
In the X-radiograph of the face of the Portrait
of a Lady it is possible to decipher broader, thicker
brushstrokes that have been blended together by
finer brushstrokes (fig. 12). The rougher appear-
Fig. 12. Portrait of a Lady (fig. 1), X-radiograph, detail of the
ance of the paint application in the X-radiograph face showing the finer brushwork found in the structure of
is accounted for by the numerous individual the painting.

Fig. 15. Portrait of a Lady (fig. 1), detail of the widow’s peak
showing the reintroduction of flesh-colored paint that creates
the lace pattern on the forehead.
may indicate that some fundamental drawing was

done with a lead-based, flesh-colored paint. The
Fig. 13. Portrait of Elizabeth van der Meer (pendant to
lighter paint on the right extends, for example,
Maarten Pauw), Nicolaes Maes, oil on oak panel, under the edge of the hair and indicates the shape
17 3/8 × 12 1/4 in. (44 × 31.2 cm). Collection of Pauw of the head. Over this light underpaint, the hair
van Wieldrecht, The Netherlands.
was painted in; this created more contrast and
visibility of the strands of hair than if they were
painted directly onto the darker ground or a dark
underpaint. When modeling the face, a reserve,
visible in the X-radiograph, was left for the point
of the lace widow’s peak. The point of the reserve
is rounder than the point created by the subse-
quently applied black lace. After painting in the
black lace it was necessary to recreate its transpar-
ency. To achieve this, flesh color was reintroduced
over the widow’s peak (fig. 15). This paint was wetter
and slightly darker than the first flesh paint appli-
cations and, as can be seen in the X-radiograph,
contained less of the white lead pigment.
While Maes painted the face in the Columbia
picture, he darkened the background to the left
and above the sitter with an arc of dense brown
paint. This corrected the contour of the cheek
and gave more contrast to the left side of the face
against the background. This type of adjusting
or outlining is not unusual and is often seen in
both finished and unfinished portraits by many
Fig. 14. Portrait of Maarten Pauw (pendant to Elizabeth painters. In this portrait, similar paint may also
van der Meer), Nicolaes Maes, oil on oak panel,
17 3/8 × 12 1/4 in. (44 × 31.2 cm). Collection of Pauw have been used at the same stage to indicate the
van Wieldrecht, The Netherlands. architectural forms and the shapes of the curtain,

table, and chair. Further analysis might reveal
that this paint is directly related to the previously
described dark underpaint of the figure. Darker
contour lines that are clearly visible within the
architecture and along the edges of the curtain are
partially painted over. Some of these lines con-
tinue under the costume, indicating their role in
the initial laying-in of the composition.
The costume and background were often sec-
ondary to the portrait. It was not uncommon for
a busy portraitist to employ assistants or even
professional drapery painters to complete compo- Fig. 16. Portrait of a Lady (fig. 1), detail of the stylized paint
sitions after the portraits were laid in. Little is application used to render the dress.
known of Maes’s studio, and there is no record of
how it operated. Only a few possible students or
assistants have been identified.34 This sequence is evident in the Columbia painting
Given the number of paintings currently where it appears that the costume was completed
attributed to Maes, he could have done all the before the background, though some indications
portrait work himself. It has been estimated that for the background were in place before the cos-
during his most productive years he would have tume was painted. Confusion as to the order of
had to complete only one portrait every two painting can arise when transitions are subse-
weeks.35 The account by Houbraken of many quently carefully covered or sharpened with addi-
unfinished paintings being left in his studio when tional paint after two adjacent areas have been
Maes died seems unfounded, unless they were completed.
later finished, if not by other artists, then by one It is clear in the Columbia painting, however,
or more assistants. Though it has also been argued that after completing the face and hands and
that the simplification and repetition of Maes’s before painting the black dress, Maes painted the
later paintings indicates the probable use of assis- sleeves and neck of the white blouse and then the
tants, no evidence for this exists either.36 On the scarf. To paint the scarf, somewhat dry gray paint
contrary, Maes’s display of skill in portraiture and was applied in long brushstrokes, which skipped
his ability—in his later works—to integrate the along, leaving the paint and even the ground
costumes and backgrounds so well with the por- below slightly exposed. This gives the scarf a light
trait indicates that even the strong stylization was transparent appearance. The highlights and final
a personal choice. The simplicity of the tech- touches of the shawl, which extend over and along
niques employed seems quite intentional, making the transition of the black dress, were added after
it unlikely that assistants were needed to complete he completed the dress.
his works. The painting of the dress is not literally
With great economy Maes could create daz- descriptive but schematic with an abstract quality
zling effects in both simple and very fanciful dress (fig. 16).37 This is especially true in the area to
with little effort. The costume and background the right, below the hand, where the folds of the
of the Columbia Portrait of a Lady are painted with cloth are difficult to interpret. This daring show
just such skillful simplicity. of bravura must be by Maes himself. It is difficult
As a general practice, the face and hands were to imagine that such a schematic design on an
finished first, after which the costume and back- important commission would be acceptable to
ground would be painted in. In most cases this is the patron if either an assistant or a professional
seen in the overlapping of paint at the transitions. drapery painter had worked on it.

In painting the larger areas of the dress, the white, dominates the bluer areas of the sky. In the
gray and black must have been laid in simultane- Columbia painting the smalt, which is prone to
ously, in no specific order, leaving reserves for one color change in an oil medium, appears to have
color next to the other. In the largest expanse of faded.38 The increased translucency of the pig-
gray, that of the bodice and the proper right ment and yellowing of the oil medium give the
sleeve, the black touches indicating folds and sky a somewhat muddied appearance. However,
seams were applied over the gray. Elsewhere the despite the discoloration, the light tone of sky
black and gray cover more equal areas, and the creates a balanced composition with other high-
two colors were laid on in alternating bands that lighted background elements that surround the
were then blended together wet-in-wet, sometimes sitter.
with the black over and into the gray and some- The curtain, chair, and table are quickly exe-
times vice versa. Small areas of uncovered ground cuted with broad shadows, abstracted highlights,
can occasionally be found between the two colors. and monochromatic mid-tones. A minimum of
Black lines were then used to indicate the seams, detail is found in the highlights of the chair and
to clarify the forms of the folds, and to sharpen in the folds of the cloths. What may be a book
the outer edges of the dress. The white highlights on the table under the sitter’s elbow is hardly rec-
were added last. Finally, black was also used to ognizable as such. The broad mid-tones of the
sharpen the final contours of the white blouse, curtain and the back of the chair are accentuated
the scarf, and the hands and wrists. with a free application of a light brown paint
The much smaller portrait of Elizabeth van containing the same discrete, large, white particles
der Meer (see fig. 13), who also wears a black cos- found in the architecture. In all its applications,
tume, was painted in the same general sequence. that brown paint extends slightly over the previ-
The dress is simpler in design with a plain black ously rendered contours including the edges of
bodice and only minimal gray highlights to create the hair and the costume.
volume. In this small scale the execution does not Over the entire curtain, the chair, and the
have the extreme abstraction found in the Colum- table, Maes applied an even layer of monochro-
bia dress, but it too has a schematic quality. matic red glaze. What appears as almost amateur
The background of the Columbia painting is or perhaps incomplete was, however, calculated
created with similar economy. Immediately behind to achieve a particular finished effect. The red
the sitter’s head, the area was evidently quickly increases the depth of the darkest shadows and is
covered, filling in the architectural elements with- most vivid over the highlights. It is subtler over
out fully resolving the forms. Only scant shading the brown mid-tones, but there the overall effect
and a few darker, drawn lines create any illusion of the red color is increased by the reflectance of
of depth. The paint is lighter than the dark light off the large white pigment particles and
brown used to correct the contour of the face back through the transparent red glaze.
and extends over it. This brown is lighter because
of the inclusion of clearly visible large particles The Unusual Technique of
of coarsely ground lead white. Had he used more Overall Glazing with Red
finely ground white pigment the effect would have This technique would be an anomaly if found
been different. only on this picture; or it could be a quick trick
The sky is also schematically rendered and by someone other than the artist to finish a paint-
painted like many of Maes’s skies. The colors ing or to cover damage from a harsh cleaning.
have a dark appearance that contributes to the But this technique of complete glazing has been
overall subdued tone of the background. The pig- observed on other Maes paintings where it creates
ment smalt, ground from glass colored with an even more dramatic effect.39 Maes uses this
cobalt oxide, here mixed with lamp black and lead technique for its economy, quickly bringing bril-

liant red coloring over previously modeled forms. Maes is unusual in that he boldly applied the
In rendering such colored fabrics he not only red glaze uniformly over the entire underpainting
follows traditional techniques but takes glazing with no variation in either color or density of the glaze
a step further. in relation to the shadows or highlights below.
Could this be a unique innovation by Nicolaes The initially modeled drapery thus becomes a
Maes? In the Columbia portrait, as in other Maes complete underpainting. Future research may
paintings, all the modeling for this particular red reveal that Maes also used the technique for
glazing technique is done in the underpainting. colors other than red and that other artists also
The red glaze is freely applied over the modeling used similar techniques.41
without any consideration for the forms below.
It is used only to give the final overall color. Examples of Glazing from Early Literature
The form of the drapery relies entirely on the The transparent red pigments used during this
underpaint. period were made from various natural dyestuffs,
Generally speaking, in easel painting, color artificially extracted and precipitated onto a
glazes are used either to delineate design motifs substrate.42 Analysis of samples of the trans-
(as in brocades), to locally add color or shadow parent red from Maes’s paintings, including the
to underpainted modeling, or to actually model Columbia Portrait of a Lady indicate that his red
final forms. When intensifying the color or shad- contains carminic acid derived from the cochineal
ow of previously applied undermodeling, the insect.43 After 1600, cochineal and brazilwood
glaze is usually applied in a painterly manner fol- imported from the New World became the pre-
lowing the underlying modeling, denser (usually dominant source materials for red dyes and the
thicker) over shadows, lighter (usually thinner) manufacture of red lake pigments used in western
over mid-tones and highlights. The same is gener- Europe. Cochineal, however, was the red colorant
ally true when using a translucent color in con- of choice for painters, as brazilwood was unstable
junction with other colors to initially build a and faded rapidly when used in painting. Numer-
form or when the glaze alone is used to model a ous early literary references to glazing correspond
form. Maes also used translucent paints in these to what is found in the glazing technique used
other ways. For example, the coat in the portrait by Maes.44
of Maarten Pauw (see fig. 14) is modeled only In 1649, for example, Francisco Pacheco refers
with red lake paint mixtures. In that painting, to painting various red draperies and comments
subtle mixtures of a red lake, black, and other on the quality of cochineal red. He gives instruc-
light-colored pigments are applied directly over tions for different bosquexo (a more or less fin-
the ground, to model the sitter’s coat. No under- ished underpainting) that can be glazed or left
painting or undermodeling was used. In other unglazed. For a red glaze in oil, he suggests the
instances Maes would complete a drapery and use of carmine from Florence or maybe Hon-
then with a similar transparent color, glaze over duras, both of which he states are better than that
the highlights only. from the Indies. One suggestion for painting a
Maes’s application of the red glaze over a fully crimson cloth reads, “model with vermilion and
developed and modeled drapery is not unusual. carmine, adding lead white or a little black as
Overall applications of glazes coloring under- desired … when dry, glaze once or twice with
modeled drapery are often noted in sixteenth- good Florentine carmine.” Further he discusses
century Venetian paintings40 as well as in the the importance of using good quality colors and
works of painters directly influenced by them. “to wash the painting before glazing so the glaze
However, these artists invariably used the glaze will spread over it.”45
to accentuate the form with thicker and thinner A much earlier reference to glazing from Gian
applications following shadows and highlights. Paolo Lomazzo’s Scritti sulle arti, 1584 (translated

in 1598 as A Tracte Containing the Artes of Curious tion for glazing drapery, though without going
Paintinge, Carvinge & Buildinge), gives a description into detail as to color build-up: “to paint beauti-
close to what we observe in Maes though it refers ful drapery, place your first layers in a suitable
only to the painting of gems and other transpar- color and only then glaze, and if it works, you
ent bodies. Lomazzo states that the form of the can bring about a glowing transparency of velvet
object is first underpainted in dead colors and and beautiful satins.”51
“afterwards laxeered over with simple, pure and Many of the authors seem to borrow from
clear lake, which doth most artificially represent previous ones. The material in many treatises on
those lights and shadows, which in thruth are not painting methods and materials was often copied
there …”46 from one to the next with slight variations,
Another early reference, found in the Arts of additions, and deletions. Marshall Smith, in the
Poetry, and Painting and Symetry, with Principles of Per- previously mentioned Art of Painting, was aware
spective (1615) by Filipe Nunes, also gives general of this as a potential problem when he wrote in
glazing instructions in a section on “How to his preface:
Glaze.” Referring first to using green verdigris,47
I Expect a full cry of Critticks, a Plagiary! A
he instructs: Plagiary! but first hear my Confession. I have
taken several things from Lomazzo, Vincent,
first paint evenly in white and “preto” (a dark Teftling and others … I have taken all that is
brown or black tone) that which you wish to Necessary, Corrected divers Errours, and added
glaze, and see that you make the highlights quite many things, not (to my Knowledge) Publish’d
white and the darks quite dark. After it is good before.
and dry … you can glaze it this way: Wrap a bit
of cotton with a bit of very soft linen cloth, In writing on glazing he repeats and elaborates
making a kind of brush … As you spread the on his predecessors. Detailing underpaint colors
verdigris you will see the lights appear as greens
he writes:
and the darks appear as dark greens.
for Scarlet, your drapery must be in the heights,
Though he is describing verdigris, he follows Vermillion, and brought down first with Indian
with, “The same can be done with lacra.”48 Red, then with Bone-Black: when Finish’d, Glaze
In the well-known De Mayerne tract, Pictoria it twice or thrice thin, according to the Body of
Sculptoria & quae subalternarum artium (1620), a rec- your Lake … always remembering, that you
Glaze not the Heightenings where there is any
ognized compilation of techniques from various
white, by reason it will render it Purpleish and
sources, glazing technique is noted in a section on take away the Beauty.52
the working of various colors. This includes a red
“Laque” that can be glazed in one of two man- The warning regarding purple, though important
ners with a “clean brush” or, as described by for producing a good red, was also exploited by
Nunes, “… with a shredded clothe with cotton painters who wanted to produce purple drapery.
inside …” Over a dead color rendering done with Reglazing or repainting to intensify the color
a mixture of lac, white, and brown, “… make a over a glazed area is often mentioned. This makes
glaze of beautiful lac … and after on the glaze it more difficult to recognize an overall glaze
you can further lighten and darken.”49 application. This could be especially true when
The Harley treatise (1664) also refers to using a shadows are added as mentioned by Jose Garcia
cloth to glaze overall, “when it [the undermodel- Hidalgo in 1693, “if a carmesi is desired glaze
ing] is dry glaze it over with faire lake, that is, it over once or twice with good fine carmine
strike it thereon allover & rub it all over with a and reinforce the darkest shadows with lamp-
little stuft with cotton, this is make ye lake even.”50 black …”53 In fact, Maes does use this technique
Karel van Mander in 1618 gave a direct descrip- of reinforcement of a shadow over the initial

overall glaze application.
Comparative Glazing in
Paintings by Nicolaes Maes
Clearer examples of how Maes’s technique accords
with these descriptions can be found in two
unrelated portraits, those of Simon van Alphen
(see fig. 3) and Belchje Hulft (see fig. 7). Red is
the dominant color in the costumes in both of
these portraits. The glaze application can clearly
be seen where distinct brushstrokes of the red
glaze are visible in normal light and by the overall
appearance of the glaze in ultraviolet (uv)
fluorescence (figs. 17 and 18).
Under uv light the outline of the glaze cover-
ing Simon van Alphen’s mantle does not exactly
follow the outline of the underpainted garment.
The darker color at the edge of the glaze is the
Fig. 17. Portrait of Simon van Alphen (fig. 3), uv photo- underpaint without the telltale fluorescence of
graph showing the fluorescence of the red glazed the glaze over it. The darker diagonal strip in the
mantle; note the black underpaint at the edges of the middle right is a reinforcement of the underlying
mantle, reinforcement to a shadow below the hand,
and the painterly application of a brown glaze over the shadow added, as noted by Hidalgo, over the red
background, notably where brushed out onto the sky. glazing. In a detail from the mantle (fig. 19), the
pattern of the brushstroke of the red is clearly
visible crossing over the undermodeling and
perpendicular to it. Cross-sections of paint sam-
ples from the Alphen portrait show the paint
build-up and the location of evenly applied red
glaze (figs. 20a–b and 21a–b). Here the red glaze
is clearly visible as a distinct uniform layer. It
can also be seen that Maes often used the same
translucent red pigment either alone or mixed
with other colors, predominantly vermilion and
red earth pigments, to build the undermodeling
of the drapery. Both the high concentration of
medium and nature of the pigment used cause
this glaze to be strikingly visible in uv light. How-
ever, the green fluorescence seen in the painting in
uv light is similar to an oxidized natural resin or
oil varnish and is, therefore, easily masked by the
thinnest oxidized varnish layer. Once this tech-
nique was identified in a painting being restored
(and from which the old oxidized varnish layers
had been removed), it was not difficult to recog-
Fig. 18. Portrait of Belchje Hulft (fig. 7), uv photograph
showing the fluorescence of the red glazed mantle; its
nize this technique in many of Maes’s paintings.
imprecise application can be noted along the edges. In the Portrait of Belchje Hulft (see fig. 7), the red

Fig. 20a. Portrait of Simon van Alphen (fig. 3), cross-
section (200 ×) from a middle tone area in the red
mantle.
Fig. 20b. Portrait of Simon van Alphen (fig. 3), cross-

section (200 ×) in uv light from a middle tone area
in the red mantle.
Fig. 19. Portrait of Simon van Alphen (fig. 3), detail of the free
brushstroke of red glaze crossing perpendicular over the folds
of the underpainted mantle.
mantle crossing her bodice is also fully modeled

in the underpaint and then glazed overall. This
and the area of red to the right in the background Fig. 21a. Portrait of Simon van Alphen (fig. 3), cross-
both have the characteristic uv fluorescence of the section (200 ×) from the reinforced shadow in the
red mantle.
red lake. However, in the red in the background,
which is meant to be in shadow, there is no obvi-
ous undermodeling. For a more subdued effect
the red was applied directly over a thin under-
paint, uniformly, with little or no painterly effect,
as it was in the background of the Columbia
painting. In normal light the background red
is not as brilliant as the foreground red. The
brilliance of Belchje Hulft’s mantle, on the other
hand, is the result of the high key of the fully
modeled underpaint colors. The differences in Fig. 21b. Portrait of Simon van Alphen (fig. 3), cross-
section (200 ×) in uv light from the reinforced
the paint layers creating these effects can be seen shadow in the red mantle; red glaze is visible
in paint cross-sections taken from both the fore- between layers, under the black.

Fig. 22a. Portrait of Belchje Hulft (fig. 7), cross-section (200 ×) from a highlight of the red mantle.
Fig. 22b. Portrait of Belchje Hulft (fig. 7), cross-section (200 ×) in uv light from a highlight of the
red mantle.
Fig. 23a. Portrait of Belchje Hulft (fig. 7), cross-section Fig. 24a. Portrait of a Lady (fig. 1), cross-section
(200 ×) of the background with red glaze directly (200 ×) of the red curtain taken from the upper
over only a thin paint layer and the ground. tacking edge of the painting.
Fig. 23b. Portrait of Belchje Hulft (fig. 7), cross-section Fig. 24b. Portrait of a Lady (fig. 1), cross-section
(200 ×) in uv light of the background with red (200 ×) in uv light of the red curtain taken from
glaze directly over a thin paint layer and the ground. the upper tacking edge of the painting.

ground and background areas of the red mantle the orange-colored drapery in the portrait of
of the Hulft portrait (figs. 22a–b and 23a–b). Groenendijk became an underpaint, glazed over
Further, the build-up of the red background with a layer of undifferentiated red (fig. 27). This
paint in the Hulft painting and the red in a resulted in an entirely different final finish of the
sample from the background of the Columbia two costumes. There is no doubt that both paint-
picture (figs. 24a–b) are similar. Marshall Smith, ings are finished. Cross-sections of paint from
whose 1692 treatise has already been cited, noted each costume reveal the similarity of the two
the obvious in a section on glazing in his book, underpaints and the respective absence and pres-
in that “the same Ground by the diversity of ence of the red glaze layer (figs. 28 and 29).
Glazing Colours, produceth divers Colours in These examples illustrate how the brilliant red
Drapery; likewise the Glazing Colours by diversity glaze was often used for costumes, while in the
of Grounds.”54 Columbia painting it is used only in the back-
An apt illustration of the simplicity and ground. In the Portrait of a Lady in the Timken
economy of this glazing technique is found in a Museum of Art, San Diego, a billowing back-
pair of pendant portraits by Maes, the portraits ground cloth has been glazed brilliant red with
of Petronella Dunois and Pieter Groenendijk the undifferentiated technique. The sitter’s red
(figs. 25 and 26). The drapery covering both shawl in the same painting differs from the cur-
figures was first fully modeled in orange colors. tain in that Maes modeled the shawl more tradi-
This modeling in the portrait of Dunois was tionally, first with different mixtures of red lake
left to represent an orange drapery, whereas and then by only selective glazing. The high key
Fig. 25. Portrait of Petronella Dunois, Nicolaes Maes, ca. 1680, Fig. 26. Portrait of Pieter Groenendijk, Nicolaes Maes, ca. 1680,
oil on canvas, 27 1/4 × 22 3/4 in. (69.2 × 57.8 cm). Rijksmuseum, oil on canvas, 27 1/4 × 22 3/4 in. (69.2 × 57.8 cm). Rijksmuseum,
Amsterdam. The final orange mantle is similar to the under- Amsterdam. The red glaze of his mantle covers an under-
paint of the red mantle her husband, Pieter Groendijk, wears modeling similar to the final paint in his wife’s portrait.
in the pendant painting.

of the red in the background drapery is similar
to that in the previous uniformly glazed costume
examples, again contrasting with the subdued tone
seen in the Columbia painting. As in all the exam-
ples, the final color differences depend on the
color of the underpaint rather than on the glaze
alone. These differences illustrate that Maes was
well aware of his options. Though there may be
slight fading or discoloration in the red in any of
the paintings,55 it should be assumed that the
more subtle color in the Columbia painting, as in Fig. 28. Portrait of Petronella Dunois (fig. 25), cross-section (200 ×)
the background of the portrait of Belchje Hulft, from the orange mantle; note the absence of the red glaze.
was intentional. As previously mentioned, the
large discrete particles of white lead included in
the brown paint used in the mid-tones of the
Columbia painting not only lighten the brown
color but also allow the red glaze to be more
visible than if it were applied over a finely pig-
Fig. 29. Portrait of Pieter Groenendijk (fig. 26), cross-section (200 ×)

from the mid-tone of the red mantle; the red glaze is applied
over an underpaint of orange similar to the final layer of the
mantle in the pendant portrait of Petronella Dunois.
Fig. 27. Portrait of Pieter Groenendijk (fig. 26), detail of the Fig. 30. Portrait of a Lady (fig. 1), detail of the red glaze over
free brushstrokes of the red glaze, applied without direct the chair and the curtain, with added strokes of red glaze
correspondence to the forms in the underpainted mantle. emphasizing the highlights of the chair back.

mented uniform brown color. This was calculated
to create the right balance between the reds of
the mid-tones and those of the shadows and
highlights. To achieve a darker effect in the
shadow of the table covering, Maes used a brown
underpaint without the large white pigment parti-
cles found elsewhere. As a final touch intended to
further increase the contrasts, Maes accentuated
the red highlights of the chair back by adding
quick dashes of extra red color (fig. 30).
Nicolaes Maes was a very accomplished and
adept portraitist. His technique was based on a
skillful economy of paint application and on
the use of a limited palette. The success of the
Columbia painting is the result of the careful ren-
dering of the portrait, which captures the power-
ful presence of the sitter without over-working,
placed in a subdued yet colored background that
perfectly frames the sitter. Arnold Houbraken
recorded that:
He [Maes] had an adroit and flattering brush

which served him very well in painting portraits,
to which he devoted himself exclusively and [to
him a very great share had fallen], I do not know
if there was any painter before or after him who
was so successful at capturing a likeness.56
Fig. 31. Portrait of Nicolaes Maes, Jacobus Houbraken, mixed
engraving and etching, 5 3/4 × 4 in. (14.5 × 10 cm).
As much as Maes’s prosperity was based on paint-
ing skill and his ability to please his clients, his
economic success was also undoubtedly linked to
the ease and speed of his execution. In his later Acknowledgements
works he rendered the costumes and backgrounds My sincere thanks go to the many people,
with such economy that upon close analysis the especially those colleagues at the Rijksmuseum,
form of a drapery or a background design can who gave input into the discussions regarding
appear illegible or lost. Nonetheless, in normal the glazing technique of Nicolaes Maes. For
viewing, the rich appearance of the silks and their encouragement, my gratitude also goes to
satins seem to shimmer successfully as a lasting the late Dr. Hubert von Sonnenburg, as well as
expression of the intended effect. The red glazing to Andrea Rothe, Dorothy Mahon and Dianne
technique described here, found in many Maes Dwyer Modestini, all of whom recognized the
paintings, has a special place in his oeuvre, value of publishing a description of this tech-
enabling him to achieve this goal. If a painter’s nique; to Eneida Parreira and Gwen Tauber for
skill can be measured in his ability to achieve a their patient readings of the text.
desired effect with the most economy, Maes was
an indisputable master.

Laurent Sozzani graduated from the Winterthur Museum– from 1656 to 1659 and again from 1663 to 1666. Mijtens was
a governor from 1667 to 1668 and then dean from 1669
University of Delaware Program in Art Conservation in until 1670, the year of his death.
1984. From 1984 to 1986 he worked with Perry Huston 8. Differing opinions as to the influences on Maes are put
and Associates in Fort Worth, Texas, and from 1986 to forward in the writings of Thoré and Hofstede de Groote.
Wilhelm Bürger writing in 1860 under the name Et.-Théo.-
1989 he was a Fellow in the Paintings Conservation Jos. Thoré in Musées de la Hollande II … (p. 26) recalling
Department of the Metropolitan Museum of Art. He Houbraken’s account of Maes’s Antwerp visit, concluded
has been a Senior Paintings Restorer at the Rijksmuseum that it had been a major catalyst for his change in style.
This view was countered by Cornelis Hofstede de Groot
in Amsterdam since . in A Catalogue Raisonné of the Works of the Most Eminent Dutch
Painters of the Seventeenth Century,  (London, MacMillan &
Christopher McGlinchey has worked as a conservation Co. Ltd., 1916, p. 474). Hofstede de Groote wrote that the
sources of Maes’s mature style, “are to be found in the
scientist at the Museum of Modern Art since 1999 and pictures of Hanneman, Flinck, Van der Helst, Eekhout
as an Adjunct Professor of Conservation Science for and Lievens. The later portraits of Nicolaes Maes brought
New York University’s conservation program since 1985. no new Flemish element into the evolution of Dutch
painting; the artist developed in the spirit of the times and
He majored in Chemistry and minored in Art History of the national Dutch school.” Both from Robinson 1996
at Boston University and has a Masters in Polymer (cited in note 3). See Robinson 1996 for discussion on the
Science and Engineering from Polytechnic University subject (pp. 41–95).
9. L. Krempel, Studien zu den Daterierten Gemälden der Nicolaes
of Brooklyn. His work at MoMA has focused on con- Maes (–), Phil. Diss., Bonn, Rheinisch Friedrich-
structing a scientific research lab capable of assisting Wilhelm Universität (Peter Mountain, Michael Imhof
with the technical examination of contemporary art and Verlag, 2000); and Werner Sumowski, Gemälde der
Rembrandt-Schüler, 6 vols. (Landau/Pfalz, Edition PVA,
the development of new conservation techniques specific Pfälzische Verlagsanstaldt Gmbt 1983), both for a good
to the preservation of modern art. pictorial overview. Also Robinson 1996 (cited in note 3),
for further discussion.
10. M. K. Talley, Portrait Painting in England: Studies in the Technical
Notes Literature Before  (London, Paul Mellon Centre for
1. Arnold Houbraken, De Groote Schouburgh der Nederlantsche Studies in British Art, 1981), provides an overview of
Konstschilders en Schilderessen, 3 vols., 1718–21; Vol. ii, 1719, seventeenth-century English portrait painting practice.
pp. 273–7. “Fill-in technique” is meant to denote studio procedures
2. Houbraken 1719 (cited in note 1), Vol. ii, pp. 273–4. whereby paintings were finished in the absence of the sitter
3. W. W. Robinson, The Early Works of Nicolaes Maes,  to using such practices as: Anthony van Dyck’s use of stan-
. Thesis presented to the Department of Fine Arts, dard drawings to complete the hands of sitters (Jo Kirby,
Harvard University, Cambridge, Massachusetts, May 1996, “The painter’s trade in the seventeenth century: theory and
UMI Dissertation Services, pp. 4–5. practice,” National Gallery Technical Bulletin, Vol. 20 (1999),
4. Ibid., for a discussion of this topic, pp. 203–4. p. 29); Sir Peter Lely’s use of numerous stock poses,
5. Houbraken 1719 (cited in note 1), Vol. ii, pp. 273–4. numbered so that a client could easily choose the most
Translation from Robinson 1996 (cited in note 3), p. 47. suitable one. Lely was not alone in using assistants to paint
6. Houbraken 1719 (cited in note 1), Vol. ii, pp. 275–6. draperies and backgrounds, doing only the portrait him-
7. Mijtens and Hanneman were undoubtedly closely self. An inventory of Lely’s studio made after his death
associated after circa 1640. Adrian Hanneman, the older included forty-five different pieces of drapery and numer-
of the two, spent approximately twelve years in London ous unfinished poses including fourteen half-lengths and
(1626–38). Van Dyck settled in London in 1632, and forty-two without heads (Talley 1981, pp. 335 and 367–70).
Hanneman may have worked as an assistant in his studio. At the time of Sir Godfrey Kneller’s death approximately
Hanneman returned to The Hague in 1638 and joined the 500 unfinished paintings were left in his studio with
painters’ guild there in 1640. Jan Mijtens lived all his life in instructions in his will for an assistant to finish them
The Hague and joined the same guild one year earlier. Jan (Talley 1981, p. 356). Though it can be argued that Maes
Mijtens may have trained with his uncle Daniel Mijtens probably did not have a large production studio with many
after Daniel’s return from London to The Hague around assistants, his repetitive style does imply the use of similar
1634. As early as 1618, Daniel was in London where he was procedures and devices.
a court painter, first to James i and later to Charles i, only 11. Gerard de Lairesse, The Art of Painting, in All its Branches from
to be overshadowed by van Dyck’s popularity in London translation by J.F. Fritsch, London, 1738, pp. 345 and 347;
after 1632, when Daniel returned to Holland. In 1656 when taken from Talley 1981 (cited in note 10), pp. 369–70.
the Hague painters and sculptors disassociated themselves 12. Sir William Sanderson, Graphice, or, The use of the “Pen”
from the guild and set up a new artists’ organization, and “Pensill” Or, The Most Excellent Art of Painting. London:
Pictura, both Hanneman and Jan Mijtens were active as Robert Crofts, 1658, p. 39; taken from Talley 1981 (cited
founders. Hanneman became Pictura’s first dean, serving in note 10), p. 234.

13. Houbraken 1719 (cited in note 1), Vol. ii, pp. 273–4; trans- been cut at that edge from a larger piece of pre-primed
lation from Robinson 1996 (cited in note 3), p. 13. canvas, or the original stretching to this size may have
14. A provenance search was carried out by Maria L. Gilbert, begun evenly with the top edge; either process could
Director, Provenance Index, the Getty Art History result in little cusping. The dimensions indicate that the
Information Program, for the Columbia Museum of Art. warp direction of the fabric is probably vertical. See van
From the file: “The provenance of the picture has been de Wetering 1997 and van de Wetering, “The canvas sup-
traced back to 1835 /1844 as having come from the collec- port” in J. Bruyn, et. al., Corpus of Rembrandt (The Hague:
tion of Joseph Strutt, Esquire of Derby, then through Martinus Nijoff, 1986), Vol. ii, pp. 15–43, for further
family inheritance to public auction … , and through discussion on seventeenth-century canvas sizes and stretch-
various private owners to Samuel H. Kress in 1938.” ing procedures.
15. Phyllis Cunnington and Catherine Lucas, Costume for Births, 20. In the latter half of the seventeenth century in Holland,
Marriages & Deaths. London: Adam and Charles Black, 1972, the double ground is often associated with commercially
pp. 261–9. prepared canvases. These were supplied by a specialist in
16. This comes from an English standard size of 50 × 40 in. priming, a premuerder. The commercially prepared canvas
called a half-length or three-quarters. Talley 1981 (cited in would sometimes be grounded or toned again by the artist
note 10), p. 250. with the color he wanted as an underlayer. That this large
17. This data compiled from Krempel 2000 (cited in note 9); canvas has only a single ground layer indicates the possi-
Krempel estimates the total oeuvre of Maes would include bility that it may have been prepared only in the studio
approximately 900 portraits. of the artist.
18. E. van de Wetering, Rembrandt: The Painter at Work. 21. Similar irr images have been seen in many Maes paintings.
Amsterdam: Amsterdam University Press, 1997, p. 123. An underpaint or underdrawing layer in irr reflectography
19. Width measurements could be 2/4, 4/4, 6/4, etc. referring to usually indicates the presence of carbon in an underpaint
1/2 ell, 1 ell, 1.5 ell, etc. In the seventeenth century every layer. This poses a question as an underpaint layer has yet
city had its own units of measure. Both Amsterdam and to be clearly identified in the analysis of a cross-section of
Dordrecht used ell measures equal to 68.8 cm and 68.5 cm a paint sample from the portrait in a Maes painting.
respectively. Amsterdam also used the Brabrants ell of 22. Van Dyck is said to have had occasional sittings lasting the
69.4 cm and the Brugge ell of 70 cm for cloth measures. whole day; whether this is true is not known (Kirby 1999
In addition, cloth woven in Ghent, which was regularly (cited in note 10), p. 15). Kneller, whose productivity was
imported into Holland, could be measured in the Ghent very high, could have up to fourteen different clients sit
ell of 76.5 cm. Therefore, cloth considered to be 1.5 ell, in one day for short periods; then he required between ten
could measure between 103 and 114 cm wide depending on and twelve sittings to complete each portrait (Talley 1981
its origin. These are not exact measures, only minimum (cited in note 10), p. 350). Lely would have three to four
measures. By using the entire width (or close to it) of a 1.5 sittings of one to three hours (Talley 1981, p. 330). The
ell cloth for one of the dimensions, a variety of painting daybooks of the artist Mary Beales indicate a similar num-
sizes was produced creating the clusters. Similar other clus- ber of sittings each of the same length as Lely’s. This was
ters of Maes paintings correspond to cloth of either probably the norm for many artists (Talley 1981, pp. 293–6).
1/2 ell (+ / - 35 cm) or 1 ell (+ / - 70 cm) used either in its 23. Talley 1981 and also Kirby 1999 (both cited in note 10).
full width or in half-widths. Though a range of sizes is 24. The English school of the seventeenth century was domi-
seen within the clusters, they appear to represent Maes’s nated by continental artists, the most famous among them
standard or preferred sizes. Van de Wetering (1997 (cited being Anthony van Dyck from Antwerp. Like van Dyck,
in note 18), p. 125), also noted a similar size cluster related Sir Peter Lely and Sir Godfrey Kneller both learned paint-
to 1.5 ell when he grouped Rembrandt’s paintings. All the ing in Holland. They would have needed only to adapt
above ell size conversions are from J.M. Verhoeff, De oude stylistically to English tastes. See Talley 1981 and Kirby
Nederlandse maten en gewichten (Amsterdam: P. J. Meertens- 1999 (both cited in note 10).
Instituut, 1983). All four tacking edges of the Columbia 25. William Gandy, Notes on Painting, –. Transcribed in
painting have been cut away. However, there is little reason the Memorandum Book of Osias Humphrey, –, b.m. ms.
to believe that the size of the painting has been greatly Additional 22, 950, f. 24v; taken from Kirby 1999 (cited in
altered. Cusping that occurs as a deformation of the lines note 10), p. 335.
of the canvas weave as it was originally pulled and tied or 26. David King, Secrets in the Noble Arte of Miniature or Limning,
tacked to a stretching frame can give clues to the original ca –, ms. Add 12, 461, f. 47r.; taken from Talley
size of a painting. The pattern of the cusping is locked 1981 (cited in note 10), p. 210.
into the weave by the dried ground and/or paint and is 27. From Anonymous, The Excellency of the Pen and Pencil,
best read in an overall X-radiograph of the painting. An London, 1668, p. 102; taken from Talley 1981 (cited in note
overall X-radiograph of the Columbia portrait was not 10), p. 258.
available; nonetheless, the cusping from the stretching is 28. Marshall Smith, The Art Of Painting According to the Theory
visible to the naked eye on both sides, at the bottom edge, and Practice of the Best “Italian, French and Germane” Masters,
and slightly at the top edge. The absence of strong cusping London, 1692.
at the top edge is in itself not reason enough to postulate 29. Ibid., p. 80.
that the painting is trimmed, as this sort of minimal 30. Ibid., pp. 80–81.
cusping has been observed in other Maes paintings where 31. Talley 1981 (cited in note 10), for examples and a summary,
the tacking edges are intact. The present canvas may have p. 399.

32. Willem Beurs, De Groote Waereld in ‘t kleen geschildert, of 41. Other paintings, when fully examined, may also reveal sim-
Schilderagtig Tafereel van s’Weerlds Schilderyen, kortelijk vervat in ilar glaze applications. Future examination of paintings by
Ses Boeken. Verlarende de Hoof verwen, haar verscheide mengelingen artists such as Daniel Mytens, Sir Peter Lely, and Herman
in Oly, en der zelver gebruik, Amsterdam, 1692. Sevende Verhelst, to name only a few, whose pictures have red glaze
Hoofdeel, p. 184, Van ‘t Kolorijt van een Leevendig Mensche; passages reminiscent of the overall glazing of Maes, may
Agtste Hoofdeel, p. 187, Van t’kolorijt van een Dood Mensche. provide more examples. Recognizing a glaze as an overall
The previous chapters deal with aspects of still-life paint- undifferentiated layer can be difficult if a painting has not
ing and the painting of insects and animals of all kinds, been completely cleared of old oxidized varnish layers. An
landscapes and meats—raw, cooked, unsalted and salted. oxidized varnish can mask the appearance when viewed in
33. Ibid., p. 186. This is actually a very minimal palette. Talley uv light. An example of an overall undifferentiated glaze
(1981 (cited in note 10), p. 389), mentions a palette of of verdigris over completed undermodeling has been
fifteen colors that then gives twenty-seven tints and reported on a painting by Jan Baptist Weenix (1621–1660),
another of forty colors. A Dog and Cat Near a Disemboweled Deer (Rijksmuseum
34. Krempel extrapolates from Houbraken the following sk-a 591): Arie Wallert (ed.), Still Lifes: Techniques and Style,
possible Maes students: Jan de Haen, Jacob Moelart, An Examination of Paintings from the Rijksmuseum (Amsterdam,
Johannes Vollevens, Justus de Gelder and Margaretha Rijksmuseum, 1999), p. 91 (published to accompany an
van Godewijck (who may have also painted two or three exhibition). For more on verdigris glazes see Margriet
garden backgrounds for Maes) (Krempel 2000 (cited in van Eikema Hommes, “Verdigris glazes in historical
note 9), pp. 39–41). oil paintings: recipes and techniques,” ZKK-Zeitschrift für
35. Krempel estimates that during his most productive period Kunsttechnologie und Konservierung, 15.2001, Vol. 1, pp. 163–95.
(1675–78) Maes would have completed on average thirty- 42. Red lake pigments made from dyestuffs precipitated on
two paintings a year, and in 1675, his most productive year, aluminum hydroxide (or less commonly onto calcium car-
a minimum of fifty-five (Krempel 2000 (cited in note 9), bonate) have been identified in easel paintings from the
pp. 40–41). early fifteenth century to modern times. Numerous com-
36. Alan Chong and M.E. Wiesman, “De figuurschilderkunst mon names have been used in texts for these reds. These
in Dordrecht, 1620–1719” in the exhibition catalogue, De derive from either the morphology of the source: grana
Zichtbaere Werelt (Dordrechts Museum, 1992, p. 28), argue (grains), cimatura de grana (shearings of grain-dyed textiles),
that the stylization of Maes’s paintings after 1680 indicates coccus (berries), vermiculum (worms); the color: carmine,
the regular use of assistants. This argument is countered sanguine (blood), topias (topaz); or the place of manufac-
by Krempel, who though he assumes that Maes would have ture: Florentine lake, Venetian lake, Paris red (generally
had assistants to prepare his palettes and do other non- cities with large trade monopolies). Recipes from 1400 to
painting tasks, he did not have a large production studio 1900 for lake dyes cite brazilwood, the madder plant and
or school (Krempel 2000 (cited in note 9), p. 40). the scale insects—lac, kermes and cochineal—as the pri-
37. The Kress painting Portrait of a Lady was cleaned and mary source material. All, with the exception of brazil-
restored by Dianne Dwyer Modestini in 1997–98 at the wood, which is only rarely found in paintings, have been
Conservation Center of the Institute of Fine Arts, New identified in numerous easel paintings analyzed by research-
York University. A condition report written at that time ers. In the Netherlands up to the early seventeenth century
notes: “The black silk of the dress is painted in an almost kermes, madder, and occasionally Old World (Polish)
abstract manner, using only middle tone and shadow, so cochineal were the primary sources for the red lakes. After
that the folds of the drapery are flat and two dimensional.” 1600 New World cochineal becomes the predominant
From the Columbia Museum of Art files, undated. source material for red lake used in the Netherlands.
38. In many paintings, Maes applied other color glazes, often Cochineal is the source of the colorant found in the glazes
less visible in uv light, either locally or in a similar overall, of Maes. See also J. Kirby, “A spectrophotometric method
undifferentiated method. Some of these, altered by aging, for the identification of the lake pigment dyestuffs”
may have also contributed to the present darkened appear- (National Gallery Technical Bulletin (Sept. 1977), pp. 35–45) for
ance of the skies and backgrounds of many of his later a detailed description of various red lake colorants.
paintings. Less pigmented than the red glaze, these other 43. Analysis was carried out by Christopher McGlinchey,
colored glazes are easily confused with old varnish layers Museum of Modern Art, New York, NY. Samples of
and have sometimes been abraded or even lost in past red were first examined by microscopic methods where it
restorations. For a recent discussion on the color changes was determined that the red pigment was an amorphous,
of smalt (and the fading of red lakes) see Marika Spring, low refractive index pigment with no visually observable
N. Penny, R. White, and M. Wyld, “Color change in The auto-fluorescence. For paintings of this date, these charac-
Conversion of the Magdalena attributed to Pedro Campaña,” teristics suggest the use of carmine. The samples were sub-
National Gallery Technical Bulletin, Vol. 22 (2001), pp. 54–63. sequently extracted according to the procedure described
39. Laurent Sozzani, “An unanticipated use of a transparent in a chapter on carmine by Helmut Schweppe and Heinz
red glaze in the later paintings of Nicolaas Maes” in Roosen-Runge (Artists’ Pigments, Robert L. Feller (ed.),
summaries of posters at the iic Dublin Congress, 1998, Oxford: Oxford University Press, 1986, Vol. 1, p. 276).
Painting Techniques, History, Materials and Studio Practice. The red extract was characterized by ftir microscopy and
40. A good example can be seen in many paintings of Jacopo indicates absorption spectra similar to free carminic acid.
Tintoretto where transparent glazes give color to forms This indicates a colorant derived from the dried bodies
freely undermodeled in white. of cochineal insects, assumed to be New World Dactylopius

coccus Costa. This confirmed a previous analysis by Arie Photography Credits
Wallert of the Rijksmuseum that also identified cochineal Figs. 1, 2, 5, 6, 12, 15, 16, 24a, 24b, and 30, pp. 164, 169, 172, 173,
lake in a transparent red sample from a Maes painting. 174, 180, and 182. Columbia Museum of Art, Columbia,
He used direct temperature resolved mass spectrometry SC, Samuel H. Kress Collection (cma 1962.17).
(dtms), photo luminescence spectrometry (pls), thin-layer Figs. 3, 17, 19, 20a, 20b, 21a, and 21b, pp. 167, 178, and 179.
chromatography (tlc), and microchemical tests; Sozzani Rijksmuseum, Amsterdam (sk-a 4881).
1998 (cited in note 39). For further discussion on analytical Fig. 4, p. 168. Courtesy of the Timken Museum of Art,
methods see J. Kirby and R. White, “The identification of San Diego, CA.
red lake pigments and a discussion of their use,” National Figs. 7, 9, 11, 18, 22a, 22b, 23a, and 23b, pp. 170, 171, 172, 178,
Gallery Technical Bulletin, 17 (1996), pp. 56–80. and 180. Rijksmuseum, Amsterdam (sk-a 1269).
44. We would like to thank Margriet van Eikema Hommes, Fig. 8, p. 170. Rijksmuseum, Amsterdam (sk-a 1269). irr
Ph.D., Kunsthistorisch Instituut der Universiteit van montage assemblage courtesy of Monique Berends.
Amsterdam, for supplying many of the following source Fig. 10, p. 171. Rijksmuseum, Amsterdam (sk-a 2666).
references. These were gleaned from her research for Figs. 13 and 14, p. 173. Courtesy of the collection of Pauw van
her doctoral thesis: Discoloration in Renaissance and Baroque Wieldrecht, The Netherlands. Photograph ©Peter
Oil Paintings, Studies into Painting Recipes, Theoretical Concepts Mokook.
and Scientific Data (published by Archetype Publications, Figs. 25 and 28, pp. 181 and 182. Rijksmuseum, Amsterdam
London, in 2004 as Changing Pictures: Discoloration in th (sk-a 4889).
to th Century Oil Paintings). Figs. 26, 27, and 29, pp. 181 and 182. Rijksmuseum, Amsterdam
45. Francisco Pacheco, Arte de la Pintura, 1649, from Artists’ (sk-a 4888).
Techniques in Golden Age Spain: Six Treatises in Translation, edited Fig. 31, p. 183. Houbraken 1719 (cited in note 1), vol. ii, p. 277.
and translated by Zahira Veliz. Cambridge: Cambridge 1
University Press, 1986, p. 73.
46. G.P. Lomazzo, A Tracte Containing the Artes of Curious
Paintinge, Carvinge & Buildinge, English translation by
Richard Haydocke. Oxford: Joseph Barnes, 1598 (Third
book, Chapter ix), p. 107.
47. For more on the use of verdigris glazes see van Eikema
Hommes 2001 (cited in note 41).
48. Filipe Nunes, Arte poetica, e da pintura e symetria, com principios
da perspectiva, 1615, from Veliz 1986 (cited in note 45), p. 5.
49. T. de Mayerne, Pictoria Sculptoria et quae subalternarum artium,
1620 (pp. 221–2), from Quellen Für Maltechnik … De
Mayyerne Manuskript, Ernst Berger, Vaduz, Sändig Reprint
Verlag, 1984, p. 282.
50. Harley, ms 6376 British Library.
51. Hessel Miedema, “Den grondt der edel vry schilder-const /Karel
van Mander,” edited, translated and annotated by Hessel
Miedema. Utrecht: Haentjens Dekker & Gumbert, 1973,
fol. 44.
52. Smith 1692 (cited in note 28), p. 84.
53. José Garcia Hidalgo, Principios Para Estudiar el Nobilisísimo
y Real Arte de la Pintura, 1693, from Veliz 1986 (cited in
note 45), p. 135.
54. Smith 1692 (cited in note 28), p. 84.
55. See D. Saunders and J. Kirby, “Light-induced color
changes in red and yellow lake pigments,” National Gallery
Technical Bulletin, 15 (1994), pp. 79–97. From the data pre-
sented here it can be assumed that the lake reds in all
Nicolaes Maes paintings have faded slightly. Regarding the
Columbia painting, a restoration report written in 1996 by
Mario and Dianne Dwyer Modestini state that back-
ground elements are “… glazed broadly with a red lake.
The final glazes have either faded or have been partly
removed in past cleanings.” From the Columbia Museum
of Art files, dated April 14, 1996.
56. Houbraken 1719 (cited in note 1), Vol. ii, p. 274.
Translation from Robinson 1996 (cited in note 3), p. 51.

View of the Molo
Canaletto (Giovanni Antonio Canal), ca. 1725
Oil on canvas
26 1/2 × 32 3/4 in. (67.3 × 83.2 cm)
cma 1954.44 (k-252)
Fig. 1. View of the Molo, before treatment.
Fig. 2. View of the Molo (fig. 1), after treatment.

View of the Molo: A Canaletto
Attribution Reinstated
Elise Effmann
S hadowed by the doubt of its attribution for decades, the View of the Molo
was sent to New York in 1997 from the Columbia Museum of Art in
South Carolina for conservation. This occasion was a valuable oppor-
tunity for close technical and stylistic examination. Shortly after Samuel
H. Kress bought it in 1933, the painting was accepted unanimously as a
genuine Canaletto by scholars evaluating the collection (figs. 1 and 2).1 How-
ever, by the time the Kress Collection catalogue was published four decades
later, enough suspicion had been raised to reduce the Venetian scene to
“attributed
to Canaletto” in the entry.2 This downward revision of the work was largely a
result of skepticism expressed by the foremost Canaletto authority of the day,
W.G. Constable. Constable had such serious doubts about “the character of
the brush work and the drawing” that he believed not only that the scene “is
not by Canaletto,” but is “probably a work of the earlier 19th Century.”3 It was
this view that was published in his two-volume catalogue raisonné in 1962; the
revised edition remains the definitive study of the artist’s oeuvre to date.4
As research into the numerous paintings by and attributed to Canaletto
and his studio continues, art historians have come to depend more frequently
on technical study. Since 1980, published studies focusing both on individual
paintings and on the broader development of the artist’s technique have been
important in establishing a clearer chronology and also in resolving issues of
attribution.5 These studies, as well as the opportunity to view a large number
Elise Effmann 189

of his early paintings assembled for a recent exhi- Museum of Art painting shares numerous other
bition in Venice, helped to strengthen the case for similarities in its structure and paint handling
the Kress View of the Molo being an early work with several works by Canaletto datable from 1724
entirely by the hand of Canaletto.6 to 1726, most notably the series of four Venetian
Although he never wavered from painting views commissioned by the Lucchese textile mer-
veduti and capricci, Canaletto’s style changed chant, Stefano Conti (fig. 3). Through payment
markedly over the five decades of his activity. receipts, these paintings can be securely dated
The sleek scenes in his later works, filled with to 1725–26 and are critical to establishing the
bright airy skies and formulaic brushstrokes, painter’s early chronology.7 These four scenes are
differ vastly in feeling from the immediacy and high points of Canaletto’s early painting style,
intimacy of his paintings from the 1720s. These showing the great care he took in rendering tex-
early views are characterized by exuberant and ture and light through a masterful control of his
highly textured brushwork, dramatic contrasts medium. Many idiosyncratic aspects of technique,
of deep shadow and radiant sunlight, and a low such as the body of the paint, the sequence of
overall tone absent in his paintings from the fol- its application, the treatment of pictorial details
lowing decades. However, as Canaletto strove to such as figures, stormy skies, and still water, as
establish his mature style, his work was also well as the distinctive appearance of the architec-
marked by an unrestrained experimentation in tural details, all share striking similarities with
composition, level of finish, and technique. The the painting in Columbia, South Carolina.
works are so different from his later work that Like the majority of Canaletto’s paintings
some paintings have only recently been recognized from the 1720s, View of the Molo is painted on a
as by his hand. With the exception of archival evi- plain-weave canvas that has been primed with
dence, the examination of characteristic technical a red ground.8 On this preparation layer, the
features, such as the extent and color of under- general form of the architecture was blocked in
painting and the application of architectural with broad brushstrokes of thick paint, quickly
elements, is perhaps the most tangible method establishing the basic planes of the composition
for placing a painting within Canaletto’s vast and with short horizontal marks for the quay and
varied oeuvre. sweeping diagonal ones in the Palazzo Ducale.9
In addition to stylistic parallels, the Columbia The texture of these first rapidly applied strokes
is clearly visible through the successive thinner
paint layers and enlivens the finished surface
(fig. 4). The principal composition thus estab-
lished, the sky and water were underpainted in
gray. The selective use of a gray underpaint is
typical of Canaletto’s technique in the mid-1720s;
later in the decade he simplified it to a uniform
gray layer applied overall.10 It is interesting to
note in the Kress painting that the initial roofline
of the Palazzo Ducale was lowered with the appli-
cation of this gray layer to the sky. The artist
then scored a perspectival line into the wet gray
paint to mark the upper limit of the ornamental
crenellations crowning the façade.11
Fig. 3. Grand Canal: Looking North from near the Rialto Bridge, The energetic brushwork that is so clear on
Canaletto, 1725, oil on canvas, 35 1/4 × 51 3/4 in. (89.5 × 131.4 cm).
Courtesy of the Pinacoteca Giovanni e Marella Agnelli al the surface of the painting is even more noticeable
Lingotto, Turin. in the X-radiograph (fig. 5). The sheer exuberance

dimension to the folds in fabrics, animate the
stone surface of the quay with flickers of light
created by pastose dabs of paint, and create a
stillness in the water through long horizontal
strokes of his brush.
Although Canaletto used materials that were
all commonplace in eighteenth-century Venetian
painting, it is of particular interest that he was
one of the first artists to introduce the newly dis-
covered Prussian blue into his palette. The earliest
modern synthetic pigment, it was first made in
Germany around 1704 and soon became a valued
alternative to other more costly blues. Canaletto,
who relied heavily on blue because of his subject
matter, used this new color almost exclusively
throughout his career.14 When viewed under high
magnification, the large blue flakes visible in a
sample taken from the water of View of the Molo
display optical characteristics identical to samples
taken from other several other eighteenth-century
Venetian paintings that have been identified as the
early form of Prussian blue.15 Unlike the color
achieved by later manufacture, which resulted in a
very finely divided product, the typical eighteenth-
century process was far less consistent and often
Fig. 4. View of the Molo (fig. 1), detail. resulted in pigment particles substantially larger
than those of modern Prussian blue.16 The pres-
ence of this early form is further evidence of an
eighteenth-century origin for the painting—not
and texture of the paint application, often worked
wet-into-wet, is visible in the two gondolas in the
left foreground. This produces an X-radiograph
strikingly different from the more restrained and
controlled depiction of a similar scene from
the Kress Collection at the El Paso Museum of
Art that dates to the early 1730s.12 In fact, the
“unusual appearance … under X-ray” of the
South Carolina painting was cited in the 1973
collection catalogue as supporting Constable’s
opinion that the painting was not by the Venetian
artist.13 However, it is precisely this mastery of
paint used to achieve a variety of textural effects
visible throughout the composition that is so
characteristic of his work during the mid-1720s.
The great control he wielded over the consistency
of paint allowed Canaletto to give physical Fig. 5. View of the Molo (fig. 1), X-radiograph.
Elise Effmann 191

the nineteenth century, as Constable conjectured.
At some point after work had progressed well
beyond the initial lay-in stage, the painting was
apparently abandoned for a period of time. When
it was taken up again, the artist made several
significant modifications to the composition that
show the reworking of an already dry paint layer.
Detectable on the surface of the painting, and
clearly visible in the X-radiograph, a large boat
with a mast that reached nearly to the center bal-
cony of the Palazzo Ducale once dominated the
lower right corner of the canvas (fig. 6). In the Fig. 7. View of the Molo (fig. 1), detail of the black material used
reconsideration, the composition was extended in architectural details.
outwards at the right by almost an inch. The mast
and spar of the boat were scraped down, resulting
in jagged losses of a type that could only occur Major adjustments like these to Canaletto’s early
after the paint had had enough time to dry fully. work are not uncommon and attest to his com-
These losses and the expanded right edge of positional creativity and willingness to manipulate
the scene were repainted with the same quickly the topography and architecture as he painted.17
applied and textured paint dabs. The remainder Canaletto added the glazed shadows, figures,
of the boat was painted out and the bridge was and architectural details in the final stages of
shifted to the right, with the addition of the painting. The handling of figures and architecture
small and somewhat awkward figure standing in is germane when considering dating and attribu-
the gondola. It also appears that the left edge of tion. The behavior of the material used to render
the Palazzo Ducale was extended and the center features in the architecture, such as the lines fram-
balcony and crowning statue of Justice were ing the windows and the shadowed areas under
shifted to the left, although it is not clear whether the arches of the Palazzo Ducale, is of particular
this was concurrent with other changes or earlier. interest. The medium used for the Kress painting
appears to be identical to one identified by Viola
Pemberton-Pigott in her examination of two of
the dated paintings Canaletto did for Conti. She
noted that in a technique peculiar to Canaletto:
“the architectural detailing has been painted in a
black substance that has reticulated on drying into
broken lines of microscopic black beading which
has the instant effect of softening and blending
the lines into their surroundings.”18 She suggests
that the material might be ink, and it certainly
has this appearance in the Kress painting. Areas
painted with this medium appear to have resisted
its application to a certain degree, creating an
effect that is significantly different from the other
oil glazes (fig. 7). Canaletto used this black mate-
rial in other (but not all) paintings datable to the
Fig. 6. View of the Molo (fig. 1), detail of X-radiograph showing middle years of the 1720s, later abandoning it
pentimenti in the lower right corner. altogether in favor of lines painted in a fluid gray

and exuberantly painted primary paint layer,
and the particular handling of final details and
glazes—is completely consistent with that of
other paintings by Canaletto dating between 1724
and 1726. This was a period in his career in which
he was actively experimenting with style, creating
expressively and carefully painted compositions,
and developing techniques that he would later
simplify to meet the high demand for his work.
Although there has been speculation about the
participation of assistants in Canaletto’s studio,
Fig. 8. View of the Molo (fig. 1), detail of gondolier. particularly from the mid-1730s onwards, it seems
highly unlikely that his paintings from the 1720s,
when he was developing his own style and estab-
lishing his reputation among collectors, would
or black paint.19 The depiction of the figures is involve any other hand than his own. The penti-
typical of other early works by Canaletto. In a mento in the lower right corner attests to the
technique very characteristic of his paintings from
the 1720s, he first underpainted the figures in
View of the Molo with pure black before colorfully
clothing them with fabric practically sculpted out
of soft pastose paint. Canaletto used this method
for painting figures until about 1727, when he
began to simplify his technique and omit the step
of underpainting altogether.20 Although not con-
clusive evidence on its own, the close similarity to
other early works in both appearance and paint
handling of the gondolier and the robed magis-
trates in red and gray in the foreground, as well as
the colorful little dots for figures in the distance,
provides additional support for the attribution of
the Kress painting to Canaletto (figs. 8 and 9).21
When the technique and process of View of
the Molo are considered as a whole, it is unlikely
that anyone but Canaletto could have painted
this scene. The presence of the early form of
Prussian blue in the water dates the painting to
the early to mid-eighteenth century, undermining
W.G. Constable’s ascription of the work to a
nineteenth-century imitator. However, the clearest
evidence for the authorship of the painting lies in
specific idiosyncratic aspects of the technique—
construction and details that imitators would not
notice and could not know. Every aspect of the
technique of View of the Molo—the red-grounded Fig. 9. View of the Molo (fig. 1), detail of Santa Maria
canvas, the selective gray underpainting, a textured della Salute.
Elise Effmann 193

creativity of the mind at work, a mind actively Clarendon Press).
5. See: Pamela England, “An account of Canaletto’s painting
reconsidering the composition as a whole by technique” in Charlotte Millar and Oliver Millar (eds.),
choosing to eliminate the boat crowding the right Canaletto (exhib. cat.). London: The Queen’s Gallery,
foreground. This type of major compositional Buckingham Palace, 1980, pp. 127–8; David Bomford and
Ashok Roy, “Canaletto’s Venice: The Feastday of S. Roch,”
change indicates the hand of a master, not a National Gallery Technical Bulletin, 6 (1982) pp. 40–44; Keith
follower. It is also unlikely that a follower or Laing, “Canaletto: Campo SS. Giovanni e Paolo,” Bulletin of the
copyist would emulate the more idiosyncratic Hamilton Kerr Institute No. 1 (1988) pp. 96–8; Viola
Pemberton-Pigott, “The development of Canaletto’s
elements of Canaletto’s early technique, such as painting technique” in Katharine Baetjer and J.G. Links
the ink-like material used for the architectural (eds.), Canaletto (exhib. cat.). New York: Metropolitan
details found in both the Kress and the Conti Museum of Art, 1989, pp. 53–63; David Bomford and
Ashok Roy, “Canaletto’s Stonemason’s Yard and San Simeone
paintings.22 From the Palazzo Ducale’s pink Piccolo,” National Gallery Technical Bulletin, No. 14 (1993),
marble and white stone façade stained by algae pp. 35–41; David Bomford and Gabriele Finaldi,
growth to the minute rigging of the boats in the “Technique and style in Canaletto’s paintings” in Venice
through Canaletto’s Eyes (exhib. cat.). London: National
distance, all aspects of View of the Molo are infused Gallery of London and York, City Art Gallery, 1998,
with a care, calculation, and painterly skill that pp. 54–63; Viola Pemberton-Pigott, “Canaletto ‘prima
comfortably places the Kress painting with other maniera’: tradizione e innovazione nelle techniche pit-
toriche degli esorti di Canaletto” in A. Bettagno and
accepted early works by Canaletto. B.A. Kowalczyk (eds.), Canaletto: Prima Maniera (exhib.
cat., Milan, 2001), pp. 185–9.
Acknowledgements 6. Canaletto, Prima Maniera, exhibition, Fondazione Giorgio
Cini, Venice, Italy, March 18 – June 10, 2001.
I would like to thank Katharine Baetjer, Curator 7. The four paintings, now in the Pinacoteca Giovanni e
of European Paintings at the Metropolitan Mus- Maria Agnelli in Turin, are: Grand Canal: The Rialto
eum of Art, and Mark Tucker, Senior Paintings Bridge from the North (completed by November 25, 1725),
Constable/Links 1989 (hereafter referred to as c/l) 234;
Conservator at the Philadelphia Museum of Art, Grand Canal: Looking North from near the Rialto Bridge (by
for their helpful suggestions and support, and November 25, 1725), c/l 230; Santi Giovanni e Paolo and
Viola Pemberton-Pigott, former Senior Paintings the Scuola di San Marco (by May 1726), c/l 304; Grand Canal:
From Santa Maria della Carità to the Bacino di San Marco (by
Conservator of the Royal Collection, for sharing June 15, 1726), c/l 194.
her years of research and opinions about the artist. 8. It is quite likely that Canaletto used canvas already primed
by the supplier. References to pre-primed canvas (telle
imprimite) purchased for the artist appear in letters from
Elise Effmann is an Assistant Conservator of Paintings Anton Maria Zanetti il Vecchio to Arthur Pond dated
at the Kimbell Art Museum. She received a Diploma in December 24, 1728 and March 5, 1729. See Bettagno and
Conservation and an M.A. in Art History from the Kowalczyk 2001 (cited in note 5), pp. 210–11.
9. This is a technique noted in all the Conti paintings at
Conservation Center, Institute of Fine Arts, New York the exhibition in Venice in 2001. The diagonal strokes in
University. the building to the left in c/l 234 are visible in a detail
reproduced in Canaletto, Metropolitan Museum of Art,
1989 (cited in note 5), p. 92.
Notes 10. Pemberton-Pigott 2001 (cited in note 5), p. 186; Bomford
1. The positive opinions were given by Giuseppe Fiocco, and Finaldi 1998 (cited in note 5), p. 55; Pemberton-Pigott
Roberto Longhi (1933), Raimond van Marle, F. Mason 1989 (cited in note 5), p. 63. Pemberton-Pigott (ibid.)
Perkins, and William Suida (1935). Photocopies of notes that Canaletto used gray under the sky and water
documents, Samuel H. Kress Foundation Archives. and red under the buildings until about 1727–28. By
2. Fern Rusk Shapley, Paintings from the Samuel H. Kress 1729–30, he began to lighten the tone of his paintings
Collection, Vol. : Italian Schools, XVI–XVIII Century. London: and simplify his process covering the entire canvas with
Phaidon, 1973, pp. 163–4. a uniform beige or pale gray layer.
3. W.G. Constable to W. Suida, March 17, 1954, Samuel H. 11. Pemberton-Pigott 1989 (cited in note 5) notes that
Kress Foundation Archives. Canaletto used mechanical aids such as rulers and com-
4. W.G. Constable, Canaletto: Giovanni Antonio Canal, passes throughout his career.
–, 2 vols. Oxford: Clarendon Press, 1962. The 12. View of the Molo, El Paso Museum of Art, El Paso, Texas,
second edition was revised by J.G. Links in 1976 (Oxford: Kress No. k-2174, c/l 88.
Clarendon Press), and the volumes were reissued with 13. Shapley 1973 (cited in note 2), p. 163. The origin of this
supplemental and additional plates in 1989 (Oxford: opinion is unclear. Alan Burroughs, the conservation

scientist who documented much of the collection by the comparison of poor quality reproductions. In fact, the
X-radiograph, had, in contrast, noted in an undated but best published image and detail of the “Turin” version is
earlier report that the image exhibited a “skill and free- actually a picture of the South Carolina painting (see
dom, sense of space and independence of technique” that André Corboz, Canaletto: Una Venezia Immaginaria, 2 vols.,
supported its attribution to Canaletto. Photocopy of doc- Milan: Alfieri Electa, 1985, Vol. i, fig. 35, p. 53; and Vol. ii, p.
ument, Kress Foundation Archives. 51, p. 579). In addition to the topographical inaccuracies
14. Prussian blue has been found in every analyzed painting discussed by Katharine Baetjer in her paper in this volume,
except for Piazza San Marco: Looking East along the Central Line several technical elements are visible even in a color trans-
c/l 1 (circa 1723) in which ultramarine was used. parency of the Turin painting that seem to indicate that
Pemberton-Pigott 1989 (cited in note 5), pp. 62–3. it is a copy of the Kress painting by a follower. The con-
15. Dispersed pigment samples were analyzed by polarized spicuous absence of black under the figures and gray
light microscopy. A gemologist’s Chelsea filter was under the blue water differs from Canaletto’s deliberate
particularly helpful in making this identification, distin- practice of underpainting these areas to achieve specific
guishing pigment particles that transmit red light, like color and tonal effects.
ultramarine, from those that do not, like Prussian blue.
The samples from the Kress painting were identical in
appearance to Prussian blue particles identified by Mark
Photography Credits
Tucker in The Ducal Palace by Francesco Guardi in the Figs. 1, 2, 4, 5, 6, 7, 8, and 9, pp. 188, 191, 192, and 193.
collection of the Philadelphia Museum of Art. Further Columbia Museum of Art, Columbia, SC, Samuel H.
microchemical tests on the Kress samples were inconclusive Kress Collection (cma 1954.44).
due to contamination by the mounting medium. For more Fig. 3, p. 190. Courtesy of the Pinacoteca Giovanni e Marella
on the use of a Chelsea filter see M.R. Schilling and D.A. Agnelli al Lingotto, Turin.
Scott, “Letter to the Editor,” Journal of the American Institute 1
for Conservation, Vol. 28, No. 2 (1989), p. 137.
16. For a discussion of the early manufacture of Prussian blue,
see Jo Kirby, “Fading and colour change of Prussian blue:
occurrences and early reports,” National Gallery Technical
Bulletin, Vol. 14 (1993), pp. 62–71; and Frank S. Welsh,
“Particle characteristics of Prussian blue in an historical
oil paint,” Journal of the American Institute for Conservation, 27
(1988), pp. 55–63.
17. Pemberton-Pigott has noted extensive reworkings of other
paintings from the 1720s. See Pemberton-Pigott 2001
(cited in note 5), p. 188.
18. Ibid. I wish to thank Viola Pemberton-Pigott for her
generosity in sharing her original text in English.
19. Other paintings seen at the recent exhibition in Venice
that show evidence of this technique are: Grand Canal:
Looking North-East from near the Palazzo Corner-Spinelli to the
Rialto Bridge, Dresden Gemäldegalerie, c/l 208; The Bacino
of San Marco looking North, Upton House, The National
Trust, c/l 144; Grand Canal: The Salute and the Dogana, from
near the Palazzo Corner, Berlin Staatliche Gemäldegalerie,
c/l 181; also Grand Canal: Looking North-East from the Palazzo
Balbi to the Rialto Bridge, Ferens Art Gallery, Hull City
Museums, c/l 214.
20. Pemberton-Pigott 2001 (cited in note 5), p. 187.
21. For comparison see the gondoliers in the foreground of
the Ferens Art Gallery painting (c/l 214) and Grand
Canal: Looking North-East from the Palazzo Balbi to the Rialto
Bridge, Ca’Rezzonico, c/l 210 and also the red-robed
magistrate in the Conti painting Il Campo Santi Giovanni
e Paolo. In addition, a painting on copper of The Molo
Looking West (c/l 89) has an almost identical grouping of
magistrate figures.
22. It is revealing to compare the Kress painting with the
nearly identical version of the image in the Galleria
Sabauda in Turin, Italy. Both Constable and André Corboz
name the Turin painting as the primary version in their
books, and there has been much confusion over the years
about the relationship of the two paintings stemming from
Elise Effmann 195

Fig. 1. The Molo, Venice, Looking West, Luca Carlevaris, 1709, oil on canvas, 19 7/8 × 47 1/8 in. (50.5 × 119.7 cm).
The Metropolitan Museum of Art, New York, NY.
Fig. 2. View of the Molo, Canaletto, ca. 1725, oil on canvas, 26 1/2 × 32 3/4 in. (67.3 × 83.2 cm).
Columbia Museum of Art, Columbia, SC. After treatment (see also fig. 2, p. 188).

Canaletto Paints the Molo from the
Ponte della Paglia
Katharine Baetjer
V iew painting flourished in eighteenth-century Venice. The period may

be said to have begun in 1703 with the publication of Luca Carlevaris’s
compendium of engravings, Le Fabriche, e Vedute di Venetia, and to have ended
in 1797 with the fall of the Venetian Republic.1 The principal view painters
belonged to three succeeding generations: the Udinese Carlevaris (1663–1730)
was the eldest, while his principal successors and competitors were the Venetians
Canaletto (Giovanni Antonio Canal, 1697–1768), Canaletto’s nephew Bernardo
Bellotto (1722–1780), and Francesco Guardi (1712–1793). All painted topograph-
ical and festival scenes and capricci—real and imaginary buildings and ruins,
unrealistically combined. As Venice’s own citizens had limited interest in paint-
ings of their native city, success in this genre depended upon the patronage
of Italians living outside the Venetian Republic and of visitors from north of
the Alps, notably ambassadors to the Serenissima and also English gentlemen
who completed their education with travel on the European continent, on the
so-called Grand Tour.
Foreign patrons preferred easily recognizable subjects: the church and piazza
of San Marco; the quays and harbor basin and the island church of San Giorgio
Maggiore; the Grand Canal from the Dogana da Mar, or customs house, to
San Simeone Piccolo (opposite the modern train station). Such views were
painted as independent single canvases, pairs, groups of four, and larger sets.
A popular motif was a bird’s-eye view of the quay, or Molo, to the west with
the arcaded pink-and-white patterned façade of the Palazzo Ducale, the seat
of Venetian government, in steep perspective to the right. The subject was a
favorite with both Carlevaris (fig. 1) and Canaletto (fig. 2). In both canvases the
Katharine Baetjer 197

Fonteghetto della Farina stands at the end of the
Molo with, in succession, the Granai, or Public
Granaries; the Zecca, or Mint; the Biblioteca Mar-
ciana, Sansovino’s famous library; and the columns
of Saints Theodore and Mark. On the other side
of the opening of the Grand Canal is the Dogana,
crowned by a statue of Fortune standing on a
golden globe supported by Atlases. To the left lies
the Giudecca Canal with the island of the Giudecca
and Palladio’s church of the Redentore, while to
the right is Longhena’s Santa Maria della Salute on
the Grand Canal. In readiness for the doge, a galley
with a striped awning is anchored at the Molo. Fig. 3. View of the Molo, Canaletto, ca. 1730–35, oil on canvas,
Canaletto, born in the parish of San Leo, came 44 1/2 × 63 1/4 in. (113 × 160.6 cm). El Paso Museum of Art,
El Paso, TX.
to public notice in the early 1720s when he was
in his mid-twenties. The son of a theatrical scene
painter, he had worked with his father in Rome
before renouncing the theater to return to Venice, In 1933, Samuel H. Kress bought a painting by
where he was registered in the local painters’ guild Canaletto representing View of the Molo (figs. 2, 5,
in 1720.2 His reputation was established by 1723, and 6) from Alessandro Contini-Bonacossi, the
when the Giovanelli brothers commissioned from Florentine count who was his principal dealer.5
him two enormous capricci to decorate their villa at Mr. Kress lent the painting to the National Gal-
Noventa Padovana, on the Brenta Canal near Padua.3 lery of Art, Washington from 1941 until 1952, and
Thereafter he was recommended—in preference in 1954 the Kress Foundation presented it to the
to Carlevaris—to the Lucchese textile merchant Columbia Museum of Art, Columbia, South
Stefano Conti, for whom between August 1725 Carolina. The Columbia view of the Molo is
and June 1726 he painted four canvases representing probably to be identified with a picture consigned
the Rialto Bridge, the church of Santi Giovanni e anonymously by the London firm Gurr Johns to
Paolo, and the Grand Canal seen from the Rialto Christie’s, where it was sold as the property of a
Bridge and from Santa Maria della Carità.4 gentleman on June 12, 1931, as lot 59, “The Doge’s
Reporting what he described as a generally held Palace, Venice,” measuring 26 × 32 1/4 in., for £ 483
opinion, Conti’s agent in Venice, Alessandro to “Holland.”6 Contini-Bonacossi often used an
Marchesini, famously remarked that the differ- agent to buy Italian works on the London art
ence between the two artists lay in the fact that market before World War ii, and although the sale
in Canaletto’s works “si vede lucer entro il sole”: this catalogue description is inadequate for purposes
remark, variously translated, seems to suggest that of identification, the correspondence in size (the
Carlevaris’s paintings present a more uniform sun- Kress canvas measures 26 1/2 × 32 3/4 in.) is close
light, while Canaletto was better able to suggest and the proximate dates of the transactions are
the transitory nature of light effects. The 1720s— suggestive.
which encompassed Canaletto’s maturation and Roberto Longhi’s opinion, inscribed on the
his development of an early style characterized by reverse of a photograph of the picture in the
vigorous handling and acute sensitivity to light and Kress Archives, is dated “Roma 1933” and identifies
atmosphere—concern us here. Modern scholar- “questa magnifica veduta” as a work of Canaletto’s
ship holds that by 1730 at the latest, his work had best (i.e. his early) period.7 The Kress Foundation
become brighter, tighter, and in general more in then followed standard practice in soliciting
accord with visitors’ perceptions of the city (fig. 3). additional written endorsements from William

Suida, in 1935, as well as from Giuseppe Fiocco, for any information that could either make cer-
Raimond van Marle, and Adolfo Venturi.8 The tain that the two paintings are the same or settle
picture was first published in the National Gallery that they are in fact different.10
of Art’s 1941 Preliminary Catalogue of Paintings and
Sculpture, which included both gifts and loans to The various editions of the Canaletto oeuvre
the Museum. The brief comment reflected the catalogue were prepared using annotated photo-
opinion of Canaletto expert and Boston Museum graphs and notes, information from which has
curator W.G. Constable: “Painted c. 1730. May be been kindly supplied by Charles Beddington.
related in style to [the Conti pictures] … A ver- What are apparently Constable’s earliest com-
sion is in the Gallery of Turin.”9 ments on the Kress picture, made from a Witt
In a letter of March 17, 1954, responding to a Library photo mount on Courtauld Institute of
request for further information from Suida, who Art stationery, indicate that at first he accepted
by then was Kress Foundation curator of research, the attribution in full. He also recorded the Ex
Constable reversed his earlier opinion: (collection) as that of Sir Francis Ewan, London.
Later, studying an image of the painting belong-
This of all the Kress pictures is the one with ing to the Photo Library of the Museum of Fine
which I am most familiar. After careful study Arts, Boston, he again concluded that it was an
and much hesitation, I’ve reluctantly come to the
view that it is not by Canaletto and is probably a
autograph early work. In January 1953, however,
work of the earlier nineteenth century. This view when he examined it with John Walker of the
is based mainly on the character of the brush National Gallery of Art in Washington, he found
work and the drawing … I have not been able both the architecture (“scratchy and thin”) and
to clear up one or two obscurities in the history. the figures (“impressionist and sketchy”) wanting.
The picture is said to have come from Sir Francis Walker agreed. Meanwhile, Constable seems always
Swan [typescript corrected in ink to read Ewan], to have accepted the attribution to Canaletto of
London, but so far I have not managed to find the Galleria Sabauda picture on the basis of the
anything about this collection. Moreover, there
Anderson photograph, simply adding, without
was a picture (which I have only seen in the
photograph) sold in the Ashurst sale, Christie’s, comment, the number 86 of the catalogue raisonné
June 12, 1931, No. 59, which is apparently identi- on the reverse.
cal with the Kress picture. I’d be most grateful Suida published a catalogue of the Kress
Collection at Columbia, South Carolina in 1954,
but Constable’s letter of March 17 evidently
came too late for inclusion.11 However when,
in 1962, Suida’s successor Alessandro Contini-
Bonacossi (the dealer’s nephew) published an
expanded volume, he had to take into account
Constable’s 1962 monograph. This he did in a
note: “W. G. Constable in his recent book rejects
the attribution to Canaletto and considers this
painting ‘perhaps by an early nineteenth-century
imitator.’ (!)”12 He did not mention that it was
listed as a variant of the painting in Turin. There
the matter rested until 1973, when Fern Rusk
Shapley, writing also for the Kress Foundation,
opted for “Attributed to Canaletto.”13 Shapley
Fig. 4. View of the Molo, after Canaletto, oil on canvas, 26 × emphasized the connection with the version at
33 7/8 in. (66 × 86 cm). Galleria Sabauda, Turin, 1871. the Galleria Sabauda (figs. 4, 7, and 8):

Fig. 5. View of the Molo (fig. 2), detail.


[In the Turin picture] the viewpoint, from the While Canaletto made numerous versions of
Ponte della Paglia, at the extreme right, is the his most popular compositions, he generally did
same as in [the Columbia picture], and figures, not paint exact replicas. He showed the buildings
boats, and even size of canvas correspond very of Venice more or less as he saw them, recording
closely. It would seem that one of the two
paintings is a copy of the other, or that both
changes in and repairs to various structures as
are copies of a third, now lost. The unusual these occurred.18 Nearly 250 years later the city
appearance of [the Columbia picture] under is largely the same, and Canaletto’s audience still
X-ray tends to support the opinion that this assumes that his topography is accurate, while the
painting may be by a follower rather than by most persistent among us go out and have a look
Canaletto himself. from time to time, to be sure of the details and
to check for discrepancies. By contrast to the
Mario Modestini points out that Shapley buildings, the boats and the people—individuals
was under some pressure from the Foundation involved in the business of daily life, washing,
not to follow Constable.14 She chose not to building, pushing, posing—quite naturally differ
offer a definitive opinion as to the relative merits from one variant of a cityscape to another.
of the two canvases. It is curious, however, Canaletto’s figures are distinctively dressed in
that she should have cited the evidence of the accordance with their place of birth and their role
X-radiograph (see fig. 5, p. 191) of the Columbia in society; the nearer they are, the more highly
picture as negative: what this shows is a major they are individualized.
pentiment in the lower right, including a catenary There is one noticeable topographical error in
curve which could be read as the edge of a par- the Columbia painting: the subsidiary dome of
tially lowered sail. The pentiment, together with Santa Maria della Salute is flanked by the two
the extension of the composition at the right slender towers, whereas from Canaletto’s chosen
edge, seems to me to support its primacy over point of view, as indeed from the Grand Canal,
that at the Galleria Sabauda. The Turin painting one of the towers should be on the near side
was bought in 1871 through Baron Sallier de la and the other should be largely hidden.19 At the
Tour, having belonged previously to a lawyer of Palazzo Ducale, additionally, there are twenty-one
the Martelli family of that city.15 It also lacks Gothic arches where there should be seventeen,
an early history, while neither the Turin nor the at ground level, and thirty-eight where there
Kress picture has ever been included in a major should be thirty-four, on the balcony. (When
exhibition.16 seen straight on, the arcades of the building are
The Columbia view of the Molo is in a good symmetrical, with two arches above corresponding
state, as Elise Effmann explains in her paper to one below. However for Canaletto, the number
(in this volume, pp. 189–95) but by 1997 the var- of arches depicted seems to have depended to
nish had yellowed and the retouches discolored. some extent on the scale of the picture as well as
Treatment at the Conservation Center of the on the degree to which he wished to emphasize
Institute of Fine Arts afforded an opportunity to the effect of recession. He did not bother to
re-evaluate the picture in the light of Constable’s count and neither did we until recently.) In this
long-standing opinion. But the “character of the painting the asymmetry is in fact invisible because
brush work and the drawing” did not suggest the angle is so steep. The Columbia picture shows
inauthenticity. On the contrary, the handling and the entire façade of the Palazzo Ducale including
techniques seemed typical of Canaletto’s practice the half quatrefoil at the end of the balcony,
in the second half of the 1720s. What then was omitting only the sculpture marking the near
the relationship to the Galleria Sabauda picture?17 corner of the building. The columns at ground
And were there any topographical or other anom- level are partially whitewashed, apparently as a
alies in either canvas or in both? surface on which to inscribe notices intended for

the public. The last five arches, behind the Ponte in blue trousers, a gray vest, white shirt, and red
della Paglia, are boarded up. The last three arches turban; the same knife grinder talking to a woman
of the balcony above are partitioned off; one with an impatient child. These identical details, or
pane of one of the three windows is open. In the in the case of the sculpture its absence, seem to
eighteenth century there were fruit and vegetable me to support the view that the Columbia picture
stands at the ends of the bridge. To the left of the is the autograph version of this subject.
vendor’s hut shown here is a knife grinder’s stall, In the 1720s Canaletto made various sketches
while to the right and below, set into the masonry as well as other paintings of the Molo looking
of the bridge, is a niche containing the tabernacle from east to west. While none can be dated
of the gondoliers’ guild. with any degree of certainty, there is one festival
The towers of Santa Maria della Salute and picture with which the Columbia canvas can be
the various details described above are all the same usefully compared: among the young painter’s
in the Sabauda picture, but there are some addi- most important early commissions, it represents
tional inaccuracies and uncharacteristic omissions. the Reception of the French Ambassador at the Doge’s
Of the six windows on the façade of the Palazzo Palace.21 Louis xv’s ambassador, the Comte de
Ducale, the two at each end of the Sabauda pic- Gery, had been in Venice since December 1723,
ture are aligned while the middle two, flanking but he did not present his credentials to the
the balcony, appear to be higher. In fact, as in the doge until November 5, 1726. He was recalled in
Columbia painting and most other Canalettos October 1731. Canaletto’s canvas must date earlier
of this subject, only the two original Gothic win- in these years rather than later, and shows him to
dows at the extreme right are lower.20 In the Turin have been greatly influenced by Carlevaris, who
painting, the statue of Justice at the summit of had already won acclaim as the inventor of this
the façade of the Palazzo Ducale is missing, as are sub-genre, painting the receptions of the Abate
all the sculptures on the volutes at the Salute, and de Pompone in 1706, of the Duke of Manchester
Fortune upon her golden ball has disappeared in 1707, and of the Conte di Colloredo in 1726.
from the customs house. There were two or three Canaletto would have advanced his career when,
principal points of arrival for visitors reaching displacing Carlevaris, he secured Gery’s important
Venice by water, one of which was the customs order, as had been the case with Conti’s commis-
house. Ships of substantial size, and thus most sion. In style Canaletto’s figures are unusually
foreign vessels, could anchor only in the basin or similar to those of Carlevaris. Canaletto’s Reception
in the Giudecca Canal. Their merchandise passed of the French Ambassador suggests that he understood
through the customs house. The elegant building the appeal to foreigners of his rival’s style. To
with the statue of Fortune would therefore have accommodate himself to their taste, he apparently
been the first thing many people saw upon disem- introduced a blonder palette and smoother finish
barking. Canaletto rejoiced in the sharp irregular earlier than had once been thought.
contours created by the fretwork of statuary When arriving for his official reception by
against the sky and would not have omitted such the doge, a foreign ambassador disembarked from
a quantity of important architectural sculpture. It his boat at the corner of the Molo. Carlevaris
is difficult to imagine any Venetian view painter and Canaletto both adopted a low viewpoint for
having done so. But what is most disconcerting is reception pictures on account of the primacy
that all of the figures of any size in the Columbia of the ceremony playing out in the foreground,
and Sabauda pictures match, not only in their but otherwise Canaletto preferred a higher and
number, scale, and disposition, but also in the necessarily imaginary bird’s-eye view. The original
colors of all of their garments. There are the same impetus for the disposition of the subject absent
officials, in red or blue, or in black robes with full- the ceremonial aspect may have come from
bottom wigs; the same oriental, seen from behind, Carlevaris. While Canaletto’s Reception of the French

Ambassador must date from 1726–27, there is no offered no additional commentary, apparently
reason why the first of his bird’s-eye views of the never saw either. The Kress View of the Molo appears
Molo could not be earlier. A painting similar in to be entirely autograph and a date for it of about
subject and style to that in Columbia appeared 1725 is supported by comparison with the Conti
with its pendant on the London art market in views of 1725–26, and with the painting that was
summer 2002 (fig. 9).22 Topographically, it differs probably commissioned or bought by Edward
in that the corner of Palazzo Ducale and more Southwell in the latter year.
of the span of the bridge are visible at the right.
Santa Maria della Salute is inaccurate in some Katharine Baetjer, a graduate of Radcliffe College and
details: the subsidiary dome is smaller than in New York University, has been since 1981 a curator in
fact, the towers are inconsistent in size, and both the Department of European Paintings at the Metropolitan
are too tall. The handling seems to be much the Museum of Art. She specializes in eighteenth-century
same, with proportionately less detail than in the painting and with the late J.G. Links was organizer and
Columbia picture on account of the smaller size co-author of the catalogue for Canaletto, an exhibition
of the canvas. This new view of the Molo with its held at the Metropolitan Museum in –.
pendant, showing Piazza San Marco, is believed to
have been owned by Edward Southwell (1705–1755), Notes
who visited Venice in 1726. 1. The first Carlevaris exhibition in America was held at the
According to the late J.G. Links, Constable Timken Museum of Art, San Diego, April 27– August 31,
2001. Charles Beddington in Luca Carlevarijs: Views of Venice
devoted no less than forty years to the prepara- (exhib. cat., San Diego, 2001) provides an overview of the
tion of the first edition of his Canaletto cata- artist’s career.
logue raisonné and identified primary versions 2. The standard catalogue is W.G. Constable, Canaletto:
Giovanni Antonio Canal, –, 2 vols. (Oxford,
of roughly 500 compositions, with secondary Clarendon, 1962); a second edition was revised by J.G.
versions numbering as many as twenty in some Links, 2 vols. (Oxford, Clarendon, 1976), and reissued
cases.23 For the most part, Canaletto’s patrons with a supplement and additional plates, 2 vols. (Oxford,
Clarendon, 1989). This was followed by J.G. Links, A
were not Italian and until recently there were few Supplement to W.G. Constable’s Canaletto: Giovanni Antonio Canal:
paintings by Canaletto in public or private collec- 1697–1768 (London, Pallas Athene Arts, 1998), a posthu-
tions in Italy. Constable knew the Kress work but mous publication with amendments and additions by
Links and Charles Beddington. The artist’s early work
it seems improbable that he ever saw the canvas in has been the subject of much valuable recent research, for
the Galleria Sabauda, which should be identified which see especially the catalogue of the exhibition shown
as a somewhat damaged early copy. Links, who March 18 – June 10, 2001 at the Fondazione Giorgio Cini in
Venice: A. Bettagno and B.A. Kowalczyk (eds.), Canaletto:
Prima Maniera (exhib. cat., Milan, 2001).
3. See Constable 1962 and Links 1989 (both cited in note 2),
Vol. 1, figs. 479** and 479*** and Vol. 2, p. 449, nos.
479** and 479***; see also Charles Beddington in Bettagno
and Kowalczyk 2001 (cited in note 2), pp. 58–61, no. 30,
ills. (color), and pp. 68–9, no. 34, ill. (color).
4. Most recently Kowalczyk in Bettagno and Kowalczyk 2001
(cited in note 2), pp. 144–53, nos. 62–5, ills. (color). For a
complete transcription of the Conti documents see Marina
Magrini, “Canaletto e dintori: i primi anni di Canaletto
attraverso le lettere dei contemporanei” in Bettagno and
Kowalczyk 2001, pp. 191–5 and 197–208, documents 1–24
and 26–33.
5. Constable 1962 (cited in note 2), Vol. 2, p. 218, no. 86(a).
Later editions do not differ, nor is there an illustration.
See also: Fern Rusk Shapley, Paintings from the Samuel H.
Fig. 9. View of the Molo, Canaletto, ca. 1726, oil on canvas, Kress Collection, Vol. : Italian Schools, XVI–XVIII Century
18 1/4 × 29 1/4 in. (46.4 × 74.3 cm). Private collection. (London, Phaidon, 1973), pp. 163–4, no. k-252, fig. 315.

6. The name of the consignor has been supplied by Christie’s 21. For this painting belonging to the State Hermitage Mus-
through the good offices of Francis Russell and Jane eum, St. Petersburg, see A. Bettagno and B.A. Kowalczyk
Vernon, whose help I gratefully acknowledge. The cata- (eds.), Venezia da Stato a Mito (exhib. cat., Venice, 1997),
logue description is generic; the size is 26 × 32 1/4 in. The pp. 357–9, no. 39, and pp. 218–19, color plates.
modern firm of Gurr Johns holds no files dating to 1931 22. The pair was offered by Sotheby’s, London, on July 10,
that could be checked. 2002, as no. 79, described and illustrated in the catalogue,
7. The photograph bearing Longhi’s expertise must have pp. 218–23, and sold privately.
been solicited by Count Contini-Bonacossi. Unlike the 23. Constable 1962 and Links 1989 (both cited in note 2),
other opinions in the Kress Foundation Archives, it is Vol. 1, p. xiii.
not stamped with the name of New York photographer
Murray R. Keyes.
8. Only the opinion of Suida is dated.
Photography Credits
9. Preliminary Catalogue of Paintings and Sculpture, National Fig. 1, p. 196. The Metropolitan Museum of Art, New York,
Gallery of Art (Washington, D.C., 1941), pp. 31–2, NY, Robert Lehman Collection, 1975 (1975.1.90). Photo-
no. 234, and Book of Illustrations, National Gallery of Art graph © 2000 The Metropolitan Museum of Art.
(Washington, D.C., 1941), p. 77, no. 234. The date of Figs. 2, 5, and 6, pp. 196 and 200. Columbia Museum of Art,
acquisition is given in error as 1939, while Sir Francis Ewan Columbia, SC, Samuel H. Kress Collection (cma 1954.44).
is identified as the former owner. One other Kress painting Fig. 3, p. 198. El Paso Museum of Art, El Paso, TX, Samuel
bought from Contini-Bonacossi is said to have belonged H. Kress Collection (1961-6 /41).
to Sir Francis Ewan, Fungai’s Martyrdom of Saint Lucy, for Figs. 4, 7, and 8, pp. 199 and 201. Galleria Sabauda, Turin, 1871.
which see Fern Rusk Shapley, Paintings from the Samuel H. Fig. 9, p. 204. Photograph courtesy of Sotheby’s, New
Kress Collection, Vol. : Italian Schools, XV–XVI Century (London, York, NY.
Phaidon, 1968), p. 109, no. k-248, fig. 271. 1
10. The letter, the balance of which contains supporting
information about the Conti pictures, is among Kress
Foundation archival materials.
11. William E. Suida, Art of the Italian Renaissance from the
Samuel H. Kress Collection, The Columbia Museum of Art
(Columbia, SC, 1954), p. 59, no. 25, ill. p. 58, cites the
additional opinion of F. Mason Perkins.
12. Alessandro Contini-Bonacossi, Art of the Renaissance from
the Samuel H. Kress Collection. Columbia, SC: The Columbia
Museum of Art, 1962, pp. 101–2, no. 35, fig. 35.
13. Shapley 1973 (cited in note 5), pp. 163–4. The opinion of
the Tietzes has not been preserved in manuscript.
14. Conversation with Mario Modestini, September 2001.
15. Constable 1962 (cited in note 2), Vol. 2, p. 218, no. 86,
noted an exhibition, Venezia, at the Petit Palais, Paris, in
1919. The painting was shown at the Palazzo Pitti in
Florence in 1922: U. Ojetti, L. Dami, and N. Tarchiani,
La Pittura Italiana del Seicento e del Settecento alla mostra di Palazzo
Pitti (exhib. cat., Milan and Rome, 1922), pl. 43. See also
Noemi Gabrielli, Galleria Sabauda: Maestri Italiani (Turin(?),
Ilte, 1971), p. 91, pl. 162, fig. 435.
16. The exhibition history of the Kress picture: Berkshire
Museum, Pittsfield, MA, Canaletto and Bellotto, August 2–31,
1960, No. 14; Allentown Art Museum, Allentown, PA,
The Circle of Canaletto, February 21 – March 21, 1971, No. 8.
17. For the transparency of the Sabauda picture we thank the
Director of the Museum, Dr. Michaela Di Macho. It is
evident that there is significant wear and damage, notably
along the bottom edge and at the lower left corner.
18. On this subject see André Corboz, Canaletto: Una Venezia
Immaginaria (Milan, Alfieri Electa, 1985), 2 vols.
19. Ibid., Vol. 1, p. 52 and figs. 30–35. Figure 35 is evidently
taken from a photograph of the Columbia picture and not
that in Turin.
20. Ibid., Vol. 1, pp. 46–7 and fig. 18, for changes Canaletto
made when painting the bacino façade of the Palazzo
Ducale.

View of the Grand Canal with Dogana
Francesco Guardi, 1775–80
Oil on canvas
16 1/2 × 26 1/4 in. (41.9 × 66.7 cm)
cma 1954.46 (k-1947)
Fig. 1. View of the Grand Canal with Dogana, before cleaning and restoration.
Fig. 2. View of the Grand Canal with Dogana (fig. 1), after cleaning and restoration.

View of the Grand Canal with Dogana and
Guardi Studio Practices
Helen Spande
T he view across the Grand Canal of this sunny Venetian promontory

is one of approximately twenty closely related compositions attributed to
the eighteenth-century painter Francesco Guardi (1712–1793). Each scene
shows the activity along the water and on the quay near the customs house
or Dogana di Mare. In this version, the church of Santa Maria della
Salute occupies a noble position on the right of the canvas. Its slender bell
tower and seminary buildings extend behind, partially obscured by the dark
brick, crenellated warehouse of the Dogana. Miniature characters depicted
with the briefest gesture of the brush pose alone or in pairs against the
architecture. Spaced elegantly across the foreground, three gondolas ferry
passengers across the canal. The gondoliers’ poles leave delicate white ruffles
on the surface of the water.
With its generous portion of clear sky, this painting is a classic Venetian
view painting or veduta. English travelers on the “Grand Tour” found these
appealing, affordable1 views to be the perfect souvenir. This version of the
popular scene, View of the Grand Canal with Dogana, was purchased by Samuel
H. Kress in 1953 (figs. 1 and 2). Now in the Columbia Museum of Art, South
Carolina, this work has close counterparts in the Wallace Collection, London,
the Fine Arts Museums of San Francisco, the Nelson-Atkins Museum in
Kansas City, and elsewhere. Examination of the painting shows it to be in
good condition and structurally sound. The scene is painted on a lightweight,
coarsely woven canvas. The painting has been glue-lined, and the canvas visible
on the reverse is of modern origin.
Helen Spande 207

The painting was brought to the Conservation cusping, the faint scallop pattern along the edges
Center of the Institute of Fine Arts, New York formed by slightly uneven tensions on the weave
University for cleaning; it had a thick natural of the canvas as it is stretched for the first time
resin varnish layer that had become severely dis- and nailed to the stretcher. Once the canvas is
colored over time (see fig. 1). Removal of the old sized and painted, the cusping pattern becomes a
yellow varnish not only uncovered the intended fixed record. The presence or absence of cusping
color relationships and illusion of depth in the usually provides sufficient information to deter-
painting, but also afforded an opportunity for mine whether the image still retains its original
a thorough examination. X-radiography is one dimensions, for a canvas will be missing cusping
technique that often reveals aspects of the artist’s along the sides that have been cut down. In this
process such as which components were laid case, the cusping pattern was established when the
in on the canvas first, or whether their shapes floral composition was painted and cannot yield
were modified during painting. In this case, the anything conclusive about the Venetian scene.
X-radiograph provided the surprising discovery There are several precedents for the discovery
of an entirely unrelated composition beneath the of a decorative design underneath a Guardi
surface image. Guardi painted this Venetian scene view painting. As discussed in the catalogue of
on top of a decorative floral composition showing seventeenth- and eighteenth-century Italian paint-
symmetrical scrollwork flourishes framing a round ings at the National Gallery of Art, Washington,2
vase with flowers (fig. 3). both View on the Cannaregio Canal, Venice (fig. 4) and
While the surface scene appears not to have Temporary Tribune in the Campo San Zanipolo, Venice
been cropped, the floral image underneath was (fig. 6) have been painted over fragments of earlier
cropped along the left side and top. Only the works also revealed by X-radiography (figs. 5 and 7).
canvas weave along the lower edge shows clear The former appears to have been made from the
Fig. 3. View of the Grand Canal with Dogana (fig. 1), X-radiograph.

left end of a decorative painting with scrollwork
(overall a stronger and clearer design than the one
beneath View of the Grand Canal with Dogana) while
the latter appears to have been painted over a still
life of a vase of flowers. This recycling of canvas
illustrates the thrifty practices of the Guardi
studio. The National Gallery of Art finds the
decorative underpainting with scrollwork to
resemble fairly closely the design on a number of
altar frontals that were made by Francesco Guardi
and his workshop. Guardi also painted a number
of floral still lifes, though the symmetrical nature
of the large flourishes in this painting make it
unlikely that this is one of them. Unfortunately
neither of the National Gallery of Art under-
paintings matches the floral composition under
this Venetian landscape.
Fig. 6. Temporary Tribune in the Campo San Zanipolo, Venice,

Francesco Guardi, 1782 or after, oil on canvas, 14 3/4 × 12 3/8 in.
(37.5 x31.5 cm). National Gallery of Art, Washington, D.C.
Fig. 4. View on the Cannaregio Canal, Venice, Francesco Guardi,

ca. 1775-1780, oil on canvas, 19 3/4 × 30 1/4 in. (50 × 76.8 cm).
National Gallery of Art, Washington, D.C.
Fig. 5. View on the Cannaregio Canal, Venice (fig. 4), X-radiograph. Fig. 7. Temporary Tribune in the Campo San Zanipolo, Venice (fig. 6),
X-radiograph.
Helen Spande 209

By examining some areas of the X-radiograph
image where the lead white highlights of the
underlying decoration show most clearly, it is
possible to see that the painting underneath
shows extensive wear and damage. There are no
regular scrape marks or other traces to suggest
the paint was sanded down or intentionally
abraded in preparation for repainting, but this is
a possibility. The only other conclusion is that
the surface had become damaged from wear and
use during its original incarnation as a decorative
panel. Whereas some artists simply change their
minds about a composition and paint over it
immediately, scraping off paint that may not even
have dried fully, in this case the original canvas
appears to be considerably older than Guardi’s
view painting.
There is a warm brown oil ground visible
beneath the water in the foreground. This dark
ground shows through especially in areas of
abrasion or where the paint is thinly applied.
Around the upper edges of the buildings, a pale
violet layer appears. This may have been a ground
layer for the sky. Along the right margin, one
structure was painted over the blue of the sky,
which gives it a different tonality than the rest Fig. 8. View of the Grand Canal with Dogana (fig. 1), detail
of dome.
of the architecture that has preparation layers in
another color. The painting of the sky was done
in two campaigns: a glossy sky blue beneath a of the details have a slightly mechanical quality,
leaner, grayer light blue, separated by a layer of such as the white prow of the craft in the lower
varnish. It is possible that the second layer is a right corner, but this may be a result of the over-
later overpaint, but it is also possible that the all wear of the surface. The architecture is hardly
modifications were made while the work was Guardi’s most exacting rendition, but it is pleasing
still in Guardi’s studio. The craquelure penetrates in its detail and for the play of light and shadow.
through the ground layer, and in no place is there It appears that the fine dark lines of the architec-
obvious overpaint that crosses over it, demon- ture have been rendered not with a brush, but
strating that the top sky layer is quite old. When with a quill pen, for the even lines show the tell-
one is aware of the existence of the original floral tale split where the pressure of the stroke forces
design, this makes more sense. The design from the nib apart. This drawing also leaves tiny
the underlying decoration probably began to show gouged trails in the wet paint beneath the dome
through not long after the work was painted and of Santa Maria della Salute (fig. 8).
varnished and required masking by the second Some research into Guardi’s materials has
sky layer. already been conducted; these analyses used a
Though the brushwork is quick and gestural, variety of techniques including scanning electron
it is also fairly precise, and the gondoliers are microscopy coupled with electron dispersive
described with brief but expressive marks. Some spectroscopy, polarized light microscopy, and

X-radiograph diffraction.3 These findings provide painting technique—even the underlying deco-
a means of comparing the pigments used in this rative image—are consistent with the materials
painting with Guardi’s known palette. available at the time and match phenomena found
Guardi’s standard palette consists of vermilion, in other Francesco Guardi paintings.
lead white, bone black, green earth, van Dyck
brown, Naples yellow, iron oxide, red lake, and Helen Spande received a B.A. in Chemistry and Studio
earth pigments such as raw sienna and yellow Art from Williams College, and completed her training at
ochre. For a blue pigment, Guardi has been shown the Conservation Center, Institute of Fine Arts, New York
to have used both Prussian blue and ultramarine University, where she received an M.A. in Art History.
as well as ultramarine ash. It is likely that he chose Her specialization is in paintings conservation. She is
among these based on availability and expense. A currently the Assistant Collection Care Coordinator at
ubiquitous finding both in other Guardis and in New York University’s Villa La Pietra in Florence, Italy.
this painting is that the paint contains both tiny
particles and large coarse agglomerates. Even
Notes
Prussian blue, a pigment characterized by particles 1. Francesco Guardi’s clients were probably “middle-class
of submicron size, occurs in aggregates large Venetians and English visitors of modest means” as
enough to see clearly without magnification. The discussed in De Grazia et al. (1996), pp. 120–21.
2. De Grazia et al. (1996).
visual effect of mixing coarse and fine particles 3. Discussed in Albertson and Coddington (1981), pp. 101–19.
is both characteristic of Guardi and an easily 4. Feller (1986), p. 226.
recognized and replicated technique. This texture
alone cannot be used to prove that a painting is Bibliography
an authentic Francesco Guardi. However, this data Albertson, Rita and Coddington, James (1981) “A preliminary
in conjunction with the pigment analysis of View investigation of Francesco Guardi’s painting technique,”
Papers Presented at the Art Conservation Training Programs
of the Grand Canal with Dogana does allow us to Conference, New York University, New York, pp. 101–19.
draw some conclusions. Several different colors Beal, M. (1988) “An ambassador’s reception at the sublime
were analyzed by polarized light microscopy, and porte, rediscovering paintings by Antonio Guardi and his
studio,” Apollo, Vol. 127, No 313, pp. 173–9.
essentially all of the pigments found, including Binion, Alice (1976) Antonio and Francesco Guardi: Their Life and
Prussian blue and Naples yellow, are consistent Milieu with a Catalogue of their Figure Drawings. New York:
with the expected palette of Francesco Guardi. Garland Publishing.
Brown, Beverly Louise (1993) “Guardi exhibitions,” The
For example, in terms of the chronological devel- Burlington Magazine, Vol. 135, No. 1089, pp. 850–51.
opment of pigments, Naples yellow “enjoyed its Butler, Marigene H. (1970) “Technical note,” Painting in Italy
greatest popularity between roughly 1750 and 1850 in the Eighteenth Century: Rococo to Romanticism. Chicago: Art
Institute of Chicago.
after which it gradually became replaced by lead De Grazia, D., Garberson, E. with Bowron, E.P., Lukehart,
chromate and cadmium sulfide yellows.”4 How- P.M. and Merling, M. (1996) Italian Paintings of the Seventeenth
ever, Naples yellow is still available today. Its pre- and Eighteenth Centuries. The Collections of the National
Gallery of Art Systematic Catalogue. Washington, D.C.:
cursor, lead-tin yellow, was commonly used until National Gallery of Art.
about 1626 and disappeared completely after the Feller, R. (ed.) Artists’ Pigments: A Handbook of their History and
middle of the eighteenth century. Prussian blue, Characteristics, Vol. 1. Oxford University Press, 1986.
Fleming, Stuart (1975) Authenticity in Art: The Scientific Detection
first made in 1704 and also used by Canaletto, was of Forgery. London: Institute of Physics, p. 31.
an inexpensive alternative to ultramarine blue and Mahon, Denis (1968) “When did Francesco Guardi become
well suited to the hues of the sea and the sky. a ‘Vedutista’?” The Burlington Magazine, Vol. 110, pp. 69–73.
Mijnlieff, Ewoud (1997) “Some material aspects of the
Although no single piece of evidence defin- paintings by Francesco Guardi and his circle in Dutch
itively proves that this painting is a work by the collections,” Italian Paintings From the Seventeenth and Eighteenth
master’s hand, the cumulative effect of this Centuries in Dutch Public Collections. Florence: Centro Di,
pp. 191–4.
examination’s findings add support to a Guardi Morassi, Antonio (1963) “Francesco Guardi as a pupil of
attribution. The support, the pigments, and the Canaletto,” Connoisseur, Vol. 152, No. 613.
Helen Spande 211

Mucchi, Ludovico and Bertuzzi, Alberto (1983) Nella Profondità
dei Dipinti: La Radiografia nell’Indagine Pittorica. Milan.
Rowlands, Eliot W. (1996) “Francesco Guardi” in The
Collections of the Nelson-Atkins Museum of Art: Italian Paintings
–, Kansas City, Missouri, pp. 457–60.
Russell, Francis (1996) “Guardi and the English tourist,” The
Burlington Magazine, 138, pp. 4–11.
Photography Credits
Figs. 1, 2, 3, and 8, pp. 206, 208, and 210. Columbia Museum
of Art, Columbia, SC, Samuel H. Kress Collection
(cma 1954.46).
Figs. 4 and 5, p. 209. National Gallery of Art, Washington,
D.C., Samuel H. Kress Collection (1939.1.113). Images
© 2005 Board of Trustees, National Gallery of Art,
Washington, D.C.
Figs. 6 and 7, p. 209. National Gallery of Art, Washington,
D.C., Samuel H. Kress Collection (1939.1.129). Images
© 2005 Board of Trustees, National Gallery of Art,
Washington, D.C.
1

Appendices & Index
Appendix i
Kress Collection Paintings Treated
in the Samuel H. Kress Program in Paintings Conservation
at the Conservation Center of the Institute of Fine Arts,
New York University
1988–89 (pilot program)

k-29 Riminese School, Crucifixion, Allentown, Pennsylvania; Jennifer Sherman
k-239 Anton Raphael Mengs, Portrait of Pope Clement XIII, New Orleans, Louisiana; Claire Gerhardt
k-420 Andrea di Giusto, Nativity, Allentown, Pennsylvania; Diane Harvey
k-1025 Alunno di Benozzo, Holy Trinity with Saints, Denver, Colorado; Gwendolyn Jones
k-1544 Salvatore Rosa, Landscape with Soldiers in a Ravine, New Orleans, Louisiana; Ellen Pratt
1989–90 (no kress class offered)

k-541 Pesellino and Studio, The Seven Virtues, Birmingham, Alabama; Jennifer Sherman (in
Dianne Dwyer Modestini’s studio)

k-540 Pesellino and Studio, The Seven Liberal Arts, Birmingham, Alabama; Gwendolyn Jones (in
Dianne Dwyer Modestini’s studio)

k-1431 Leandro Bassano, The Last Judgement, Birmingham, Alabama; Annette Rupprecht
k-1814 Sebastiano Ricci, Christ Resurrected Surrounded by Angels, Columbia, South Carolina;
Annette Rupprecht
k-1952 Alessandro Magnasco, Pulcinella Singing to his Many Children, Columbia, South Carolina;
Annette Rupprecht
1992–93
k-17 Niccolò di Pietro Gerini, Four Crowned Martyrs Before Diocletian, Denver, Colorado;
Annette Rupprecht
k-60 Garofalo, Madonna in Glory, Coral Gables, Florida; Annette Rupprecht
k-74 Jacopo di Cione, The Eucharistic Ecce Homo, Denver, Colorado; Annette Rupprecht
k-367 Josse Lieferinxe, Abraham Visited by Three Angels, Denver, Colorado; Dianne Dwyer Modestini
k-1240 Studio of Botticelli, Madonna and Child, El Paso, Texas; Isabelle Tokumaru
k-1276 Attributed to Alessandro Tiarini, The Warrior, Tulsa, Oklahama; Ann Baldwin
k-1286 Guidoccio Cozzarelli, Scenes from the Life of the Virgin, Coral Gables, Florida;
k-1691 Bellotto, Entrance to a Palace, El Paso, Texas; Dianne Dwyer Modestini
k-1711 Elizabeth Vigée-Lebrun, The Grafin van Schönfeld with her Daughter, Tucson, Arizona;
Molly March
k-1749 Battista Dossi, The Hunt of the Caledonian Boar, El Paso, Texas; Jean Dommermuth
Appendices & Index 215

1993–94
k-521 Master of the Louvre Life of the Virgin, Annunciation Diptych, Columbia, South Carolina;
Molly March
k-1035 Horace Vernet, Marchesa Cunegonda Misciattelli, Tucson, Arizona; Annette Rupprecht and
k-1202 Girolamo di Carpi, Ruggiero Saving Angelica, El Paso, Texas; Suzanne Siano
k-1557 School of Perugino, Saint Sebastian, Princeton University Art Museum, Princeton, New Jersey
(formerly at the New Jersey State Museum, Trenton, New Jersey); Annette Rupprecht
k-1589 Bernardo Bellotto, Imaginary Landscape, Columbia, South Carolina; Jean Dommermuth
k-1637 Carlo Dolci, Vision of St. Louis of Toulouse, New Orleans, Louisiana; Annette Rupprecht
k-1665 Coccorante, Port in Calm Weather, Coral Gables, Florida; Dianne Dwyer Modestini and
Annette Rupprecht
k-1779 Battista da Vicenza, The Crucifixion, New Orleans, Louisiana; Elma O’Donohue
k-1986 Van Beyeren, Banquet Still Life, Seattle, Washington; Annette Rupprecht and
k-1997 Peter Paul Rubens, The Last Supper, Seattle, Washington; Dianne Dwyer Modestini
Annette Rupprecht treated 16 additional paintings on site at the El Paso Museum of Art in June 1993.
1994–95
k-18 Attributed to Niccolò da Varallo, St. John Preaching, Columbia, South Carolina;
Kirsten Younger
k-113 Girolamo Genga, St. Augustine Giving the Habit of his Order to Three Catechumens, Columbia,
South Carolina; Friederike Steckling
k-592 Master of the Goodhart Madonna, Madonna and Child with Four Saints (five panel polyptych)
Birmingham, Alabama; Jennifer Sherman
k-1084 Master of the Blessed Clare, Nativity, Coral Gables, Florida; Jennifer Sherman
k-1275 Ambrogio Borgognone, Madonna and Child, Coral Gables, Florida; Lucrezia Kargere
k-1402 Lavinia Fontana, Christ with the Symbols of the Passion, El Paso, Texas; Annette Rupprecht
k-1629 Rocco Marconi, The Adultress before Christ, Coral Gables, Florida; Suzanne Siano
k-1666 Coccorante, Port in a Tempest, Coral Gables, Florida; Dianne Dwyer Modestini et al.
k-2148 Francois Boucher, Joseph Presenting his Father and Brothers to Pharaoh, Columbia, South Carolina;
Annette Rupprecht
1995–96
k-11 Follower of Mantegna, The Triumph of Time, Denver, Colorado; Kirsten Younger
k-222 Girolamo di Benvenuto, Venus and Cupid, Denver, Colorado; Mika Koyano
k-484 Antolinez y Sarabia, Jacob and Rachel at the Well, El Paso, Texas; Friederike Steckling
k-1049 Follower of Cosimo Rosselli, Adoration of the Child, Seattle, Washington; Kirsten Younger
k-1640 Bernardo Strozzi, Saint Catherine, Columbia, South Carolina; Jennifer Sherman
k-1729 Bacchiacca, Lute Player, New Orleans, Louisiana; Rikke Foulke
k-1787 Luca Giordano, The Deposition of Christ, Tulsa, Oklahama; Dianne Dwyer Modestini
k-1949 Gentileschi, Portrait of a Young Woman as a Sibyl, Houston, Texas; technical examination,
Jennifer Sherman and Dianne Dwyer Modestini
k-2071 Vincenzo Catena, Niccolò Fabbri, Columbia, South Carolina; Dianne Dwyer Modestini
216 Appendices & Index

Jennifer Sherman treated 13 additional paintings on site at Columbia, Missouri in 1995–96, as follows:
k-337 Bramantino, Madonna and Child
k-344 Giuseppe Bazzani, The Imbecile (A Laughing Man)
k-372 Alunno di Benozzo, Processional Cross
k-1033 Girolamo Romanino, Christ Blessing
k-1097 Altobello Melone, Madonna and Child
k-1112 Paris Bordone, Thetis and Hephaestus
k-1180 Attributed to Carlo Francesco Nuvolone, Portrait of Giovanni Battista Silva
k-1182 Attributed to Gian Francesco de’Maineri, Ex Voto
k-1195a Michele di Matteo, Mater Dolorosa
k-1195b Michele di Matteo, St. John the Evangelist
k-1590 Pietro Rotari, Portrait of a Young Woman
k-1633 Workshop of Rembrandt, Abraham Sacrificing Isaac
k-1638 Bartolomeo Montagna, Temptation of St. Anthony
1996–97
k-12 Follower of Mantegna, Triumph of Chastity, Denver, Colorado; Wendy Partridge
k-70 Defendente Ferrari, Christ in the House of Mary and Martha, Denver, Colorado;
k-300 Giovanni dal Ponte, Saint George, Columbia, South Carolina; Sue Ann Chui
k-493 Vicenza Foppa, Saint Christopher, Denver, Colorado; Dianne Dwyer Modestini
k-595 Lorenzo Lotto, Saint Jerome, Allentown, Pennsylvania; Jennifer Sherman
k-1018 Alvise Vivarini, St. John the Baptist, Denver, Colorado; Annette Rupprecht
k-1223 Tanzio da Varallo, Saint John in the Wilderness, Tulsa, Oklahama; Dianne Dwyer Modestini
1997–98
k-10 Follower of Mantegna, Triumph of Divinity, Denver, Colorado; Sue Ann Chui
k-13 Follower of Mantegna, Triumph of Love, Denver, Colorado; Jennifer Sherman
k-14 Follower of Mantegna, Triumph of Fame, Denver, Colorado; Friederike Steckling
k-15 Follower of Mantegna, Triumph of Death, Denver, Colorado; Helen Im
k-252 Attributed to Canaletto, View of the Molo, Columbia, South Carolina; Elise Effmann
k-300 Giovanni dal Ponte, Madonna and Child, Columbia, South Carolina; Dianne Dwyer Modestini
k-300 Giovanni dal Ponte, Archangel Michael, Columbia, South Carolina; Jennifer Sherman
k-378 Fungai, Saint Sigismund, Columbia, Missouri; Friederike Steckling
k-596 Follower of Pontormo, Portrait of a Lady, Staten Island, New York; class project
k-1134 Nicolaes Maes, Portrait of a Lady, Columbia, South Carolina; Dianne Dwyer Modestini
k-1138 Follower of Jacopo del Casentino, Saint Prosper, Staten Island, New York; class project
k-1142 Follower of Giovanni di Paolo, Madonna Adoring the Child, Staten Island, New York; class project
k-1236 Attributed to Michele di Ridolfo, Archangel Raphael with Tobias, Staten Island, New York; class
project
k-1788 Italian School, The Crucifixion, Staten Island, New York; class project
k-1866 Murillo, Virgin of the Annunciation, Houston, Texas; Friederike Steckling
k-1932 Attributed to Beccafumi, Venus and Cupid with Vulcan, New Orleans, Louisiana; Wendy Partridge
and Dianne Dwyer Modestini
k-2189 Sebastiano del Piombo, Anton Francesco degli Albizzi, Houston, Texas; Dianne Dwyer Modestini

1998–99
k-81 Follower of Pinturicchio, Madonna and Child in a Landscape, Alexander City, Alabama;
Winstone Wells
k-127 Attributed to Paris Bordone, Diana as Huntress, Alexander City, Alabama; class project
k-525 Master of the Bambino Vispo (Starnina), Saint Nicholas of Bari, El Paso, Texas; Helen Spande
k-1196 Giovanni Battista da Udine, Madonna and Child Enthroned, Alexander City, Alabama;
Mario Modestini
k-1197 Allegretto Nuzi, Crucifixion, Birmingham, Alabama; Helen Spande
k-1263 Francesco Zaganelli, Pieta, Tucson, Arizona; Friederike Steckling
k-1401 Andrea de Bellis (formerly given to Cavallino), The Sacrifice of Noah, Houston, Texas; Helen Im
k-1662 Juan Pantoja de la Cruz, Margaret of Austria, Queen of Spain, Houston, Texas; Elise Effmann
k-1705 Francesco Castiglione, The Sacrifice of Noah, El Paso, Texas; Molly March
k-1801 Jacopo Tintoretto and Assistant, Gabriele Emo, Seattle, Washington; Jennifer Sherman
1999–2000
k-391 Pietro Rotari, A Girl in a Red Dress, El Paso, Texas; Winstone Wells
k-392 Pietro Rotari, A Girl in a Blue Dress, El Paso, Texas; Meghan Goldmann
k-464 Giuseppe Maria Crespi, Cupids Frolicking, El Paso, Texas; Jennifer Sherman
k-465 Giuseppe Maria Crespi, Cupids Frolicking, El Paso, Texas; Jennifer Sherman
k-1067 Girolamo Romanino, Madonna and Child, Savannah, Georgia; Winstone Wells
k-1146 Giovanni Larciani, Madonna and Child, Lewisburg, Pennsylvania; Meghan Goldmann
k-1274 Marco Basaiti, St. Anthony of Padua, Lewisburg, Pennsylvania; Helen Spande
k-1404 Giuseppe Maria Crespi, The Visitation, Tucson, Arizona; Jennifer Sherman and
k-1569 Michelangelo Associate, Madonna and Child with Saint John, New York, New York; Molly March
k-1947 Francesco Guardi, View of Grand Canal with Dogana, Columbia, South Carolina; Helen Spande
2000–2001
k-201 Tuscan School, Crucifixion, Coral Gables, Florida; Sandhja Jain
k-543 Francesco d’Antonio, Madonna and Child, Denver, Colorado; Matthew Hayes
k-1033 Girolamo Romanino, Salvator Mundi, Columbia, Missouri; Sue Ann Chui
k-1179 Taddeo di Bartolo, Madonna and Child, Tulsa, Oklahama; Jennifer Sherman
k-1530 De Ferrari, Joseph’s Coat Brought to Jacob, El Paso, Texas; Winstone Wells
k-1716 Follower of Cimabue, Madonna and Child, Columbia, South Carolina; Sue Ann Chui
k-1726 Domenico Ghirlandaio, Coronation of the Virgin, Denver, Colorado; Dianne Dwyer Modestini
k-1764 Bernardino Luini, Pieta, Houston, Texas; Dianne Dwyer Modestini
k-1776 Macrino d’Alba, Adoration of the Shepherds, El Paso, Texas; Isabelle Duvernois
2001–02
k-122 Jan van der Straet (previously Italian school mid-16th century), The Charity of St. Nicholas,
Columbia, South Carolina; Monica Griesbach and Karen Thomas
k-256 Spinello Aretino, Madonna and Child, Lewisburg, Pennsylvania; Monica Griesbach
k-318 Antoniazzo Romano, Crucifixion with St. Francis, Lewisburg, Pennsylvania; Karen Thomas
k-428 Giovanni da Bologna, Madonna and Child, Denver, Colorado; Dianne Dwyer Modestini
k-544 Perugino, Saint Bartholomew, Birmingham, Alabama; Molly March

k-1108b Giovanni di Ser Giovanni, St. Bernardino of Siena Preaching, Birmingham, Alabama;
Monica Griesbach
k-1434 Osservanza Master (formerly Sienese School), The Adoration of the Shepherds with St. John the Baptist
and St. Bartholomew, El Paso, Texas; Dianne Dwyer Modestini
k-1541 Agostino Tassi, Landscape, Lewisburg, Pennsylvania; Matthew Hayes
k-1569 Michelangelo Associate, Madonna and Child with Saint John, New York, New York; Molly March
and Dianne Dwyer Modestini
k-1569 Frame of Michelangelo Follower, New York, New York; Sue Ann Chui
k-1625 Cesare da Sesto, Madonna and Child with Sts. John and George, San Francisco, California;
2002–03
k-337 Bramantino, Madonna and Child, Columbia, Missouri; Molly March and Dianne Dwyer Modestini
k-347 Giampietrino, Cleopatra, Lewisburg, Pennsylvania; Nica Gutman
k-427 Correggio, Portrait of a Young Woman, Coral Gables, Florida; Molly March, Monica Griesbach and
k-1626 Bernardo Zenale, Madonna and Child with Saints, Denver, Colorado; Dianne Dwyer Modestini and
Karen Thomas
k-1636 Francois Perrier, Galatea, Lewisburg, Pennsylvania; Monica Griesbach and Wanji Seo
k-1639 Giulio Carpioni, Bacchanal, Columbia, South Carolina; Joanne Klaar
2003–04
k-430 Leonardo Studio, Madonna and Child, Denver, Colorado; Nica Gutman
k-544 Pietro Perugino, St. Bartholomew, Birmingham, Alabama; Nica Gutman
k-551 Taddeo di Bartolo, St. James Major, Memphis, Tennessee; Wanji Seo
k-552 Taddeo di Bartolo, St. John the Baptist, Memphis, Tennessee; Joanne Klaar
k-553 Taddeo di Bartolo, St. Catherine of Alexandria, New Orleans, Louisiana; Corey D’Augustine
k-554 Taddeo di Bartolo, Bishop Saint Blessing, New Orleans, Louisiana; Lauren Fly
2004–05
k-1048 Attributed to L’Ortolano, The Presentation in the Temple, Tempe, Arizona; Kelly Keegan
k-1228 Bicci di Lorenzo, The Nativity, Tempe, Arizona; Amanda Frisosky
k-1816a Giuseppe Zais, Landscape, Tempe, Arizona; Lauren Fly
Additional Kress paintings treated prior to, outside the context of the program, or on site
k-148 Albertinelli (formerly attributed to Fra Bartolommeo), Madonna and Child with Saints and Angels,
Columbia, South Carolina; Dianne Dwyer Modestini
k-232 Tiepolo, Portrait of a Boy Holding a Book, New Orleans, Louisiana (on site);
k-255 Hals, The Young Fisherman, Allentown, Pennsylvania; Dianne Dwyer Modestini
k-316 Sellaio, Adoration of the Magi, Memphis, Tennessee; Rob Sawchuck
k-357 Attributed to Palma Vecchio, Portrait of a Man, Coral Gables, Florida; Dianne Dwyer Modestini
k-1088 Biagio d’Antonio, The Adoration of the Child with Saints and Angels, Tulsa, Oklahama (on site);
k-1103 Girolamo da Santa Croce, The Annunciation, Columbia, South Carolina; Dianne Dwyer Modestini

k-1108a Giovanni di Ser Giovanni, St. Anthony Tempted by Gold, Birmingham, Alabama;
k-1111 Garofalo, The Meditation of St. Jerome, New Orleans, Louisiana; Dianne Dwyer Modestini
k-1114 Magnasco, Landscape with Figures, Birmingham, Alabama; Dianne Dwyer Modestini
k-1203 Beccafumi, Vision of Saint Catherine, Tulsa, Oklahama; Dianne Dwyer Modestini
k-1220 Vincenzo Foppa, St. Paul, New Orleans, Louisiana; Dianne Dwyer Modestini
k-1274 Canaletto, View of the Molo, El Paso, Texas; Dianne Dwyer Modestini
k-1410 Botticelli, Nativity, Columbia, South Carolina; Mario Modestini
k-1623 Domenichino, Madonna of Loreto Appearing to Three Saints, Raleigh, North Carolina (on site);
k-1653 Follower of Mantegna, Madonna and Child, Tulsa, Oklahama; Dianne Dwyer Modestini
k-1695 Licinio, Allegory, Coral Gables, Florida; Dianne Dwyer Modestini
k-1710 Attributed to Bellini, Portrait of a Bearded Man, Tulsa, Oklahama; Dianne Dwyer Modestini
k-1723 Florentine School ca. 1475, Madonna and Child with an Angel, Memphis, Tennessee; Rob Sawchuck
k-1773 Tanzio da Varallo, Rest on the Flight to Egypt, Houston, Texas; Dianne Dwyer Modestini
k-1774a Gaulli, The Sacrifice of Isaac, Atlanta, Georgia (on site); Dianne Dwyer Modestini
k-1774b Gaulli, The Thanksgiving of Noah, Atlanta, Georgia (on site); Dianne Dwyer Modestini
k-1805 Canaletto, The Piazzetta in Venice, Allentown, Pennsylvania; Dianne Dwyer Modestini
k-1806 Canaletto, View of the Grand Canal, Birmingham, Alabama (on site); Dianne Dwyer Modestini
k-1808 Guardi, Santa Maria della Salute, Coral Gables, Florida; Dianne Dwyer Modestini
k-1874 Titian, Adoration of the Child, Raleigh, North Carolina; Mario Modestini
k-1914 Bellotto, Market Place at Pirna, Houston, Texas; Dianne Dwyer Modestini
k-1949 Gentileschi, Portrait of a Young Woman as a Sibyl, Houston, Texas (on site);
k-1953 Magnasco, The Supper of Pulcinella and Columbina, Raleigh, North Carolina (on site);
k-2090 Attributed to Jacob van Ruisdael, Westphalian Landscape, Allentown, Pennsylvania;
k-2112 Stanzione, Assumption of the Virgin, Raleigh, North Carolina (on site); Dianne Dwyer Modestini
k-2120 de Hooch, Nursing Mother and Children in an Interior, San Francisco, California; Mario Modestini
k-2121 Mastelletta, Landscape with Figures, Coral Gables, Florida; Dianne Dwyer Modestini
k-2185 Steen, ‘Soo de Ouden Songen,’ Allentown, Pennsylvania; Dianne Dwyer Modestini

Appendix ii
The Distribution of the Kress Collection
National Gallery of Art

1,815 works of art
Kress Special Collections
 works of art
Metropolitan Museum of Art
Morgan Library
Philadelphia Museum of Art
Kress Regional Collections

785 works of art
Allentown Art Museum
Birmingham Museum of Art
Columbia Museum of Art
Denver Art Museum
El Paso Museum of Art
Fine Arts Museums of San Francisco
High Museum of Art
Honolulu Academy of Arts
Lowe Art Museum, University of Miami
Memphis Brooks Museum of Art
Museum of Fine Arts, Houston
Nelson-Atkins Museum of Art
New Orleans Museum of Art
North Carolina Museum of Art
Philbrook Museum of Art
Portland Art Museum
Seattle Art Museum
University of Arizona Museum of Art

Kress Study Collections
285 works of art
Allen Memorial Art Museum, Oberlin College
Arizona State University Art Museum, Tempe
Armstrong-Browning Library, Baylor University
Austin Arts Center, Trinity College
Bowdoin College Museum of Art
Samek Art Gallery, Bucknell University
David and Alfred Smart Museum of Art, University of Chicago
Doris Ulmann Galleries, Berea College
Elvehjem Museum of Art, University of Wisconsin-Madison
Fairfield University
Georgia Museum of Art, University of Georgia
Howard University Gallery of Art
Indiana University Art Museum
Mead Art Museum, Amherst College
Museo de Arte de Ponce
Museum of Art and Archaeology,
University of Missouri, Columbia
Pomona College Museum of Art
Sheldon Memorial Art Gallery and Sculpture Garden,
University of Nebraska-Lincoln
Snite Museum of Art, University of Notre Dame
Spencer Museum of Art, University of Kansas
Staten Island Institute of Arts and Sciences
Vanderbilt University Fine Arts Gallery
Williams College Museum of Art
Kress Gifts
 institutions received  works of art

Index of Names and Places
A
Andrepont, Paul 39
Apollonio di Giovanni 99, 100, 107, 110
Arena Chapel 20
B
Baetjer, Katharine 14, 194, 195, 197, 204
Baldovinetti, Alesso 52, 53
Barbara of Brandenburg 105, 106, 107
Barberini 21, 145, 151, 152, 156, 161, 163
Bardi, Pietro Maria 45
Bedotti 24
Bellini, Giovanni 24, 55, 162, 219
Bellotto, Bernardo 197, 205, 214, 215, 219
Berenson, Bernard 38, 41, 44, 52, 58, 83, 91, 96, 115, 118, 125, 126, 137, 140, 141
Berger, Gustav 50, 62
Birmingham Museum of Art 13, 64, 65, 66, 74, 75, 76, 214, 215, 217, 218, 219, 220
Bol, Ferdinand 166
Borenius 137, 140, 141
Botticelli 13, 49, 78, 79, 80, 81, 83, 84, 88, 89, 90, 91, 92, 100, 148, 214, 219
Bracaglia, Bartolo 42, 49
Bramantino 22, 24, 216, 218
Bronzino 54
Bugiardini, Giuliano 18
Bullaty, Sonja 50
Burroughs, Alan 37, 38, 41, 75, 95, 96, 194
C
Canaletto (Giovanni Antonio Canal) iv, 13, 56, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 202,
203, 204, 205, 211, 216, 219
Carlevaris, Luca 196, 197, 198, 203, 204
Cavalcaselle, Giovanni Battista 12, 20, 22, 26, 27, 140, 141
Cavenaghi, Luigi 24, 26, 27
Cennini, Cennino 48
Chigi-Saracini 116, 117, 119, 120, 121, 122, 124, 126
Christiansen, Keith 28, 39, 109, 110, 115, 117, 118, 124, 126, 131
Chui, Sue Ann 9, 28, 62, 109, 216, 217, 218
Cini, Count Vittorio 49, 62
Colloredo 100, 104, 109, 203
Columbia Museum of Art iv, 13, 14, 78, 79, 80, 81, 83, 84, 85, 86, 87, 89, 90, 91, 92, 126, 164, 165, 168, 169,
172, 173, 174, 175, 176, 179, 181, 182, 183, 185, 186, 187, 188, 189, 190, 195, 196, 198, 199, 202, 203, 204, 205,
206, 207, 212, 214, 215, 216, 217, 218, 219, 220

Contini-Bonacossi, Count Alessandro 32, 37, 38, 39, 40, 41, 44, 45, 46, 49, 50, 54, 57, 58, 66, 90, 91, 95, 126,
131, 133, 134, 161, 198, 199, 205
Cook Collection 49, 58, 133
Costabili Collection 24, 25
Cozzarelli, Guidoccio 13, 128, 129, 130, 131, 214
Crivelli 21
D
da San Giorgio, Eusebio 137, 138, 139, 140
da Vinci, Leonardo 32, 52, 218
Dale, Chester 32, 40
de Hooch, Pieter 60, 166, 219
de Wild, Louis 37, 39, 40, 41
della Francesca, Piero 20, 21
Denver Art Museum 13, 43, 94, 95, 96, 97, 98, 99, 100, 101, 103, 106, 111, 214, 215, 216, 217, 218, 220
di Benvenuto, Girolamo 94, 95, 100, 106, 107, 109, 124, 215, 216, 217, 219
di Giovanni, Benvenuto 95, 97
di Paolo, Giannicola 113, 116, 117, 125, 126, 136, 137, 140, 141, 216
Doge Andrea Gritti 59
Douglas, R. Langton 79, 83, 90, 91, 129, 131
Duveen 32, 40, 49, 52, 53, 54, 81, 91
E
Edward Solly Collection 133, 141
Effmann, Elise 13, 14, 189, 194, 202, 216, 217
El Paso Museum of Art 13, 112, 113, 114, 115, 116, 117, 118, 121, 122, 124, 125, 126, 127, 191, 194, 198, 205, 214, 215,
217, 218, 219, 220
Ellis, Margaret Holben 7, 28
Emerson, Guy 44, 45, 49, 50, 52, 53, 56, 57, 58, 59, 61, 62
F
Fatta, Angelo 36, 42, 46, 50, 57, 59, 96
Finley, David 33, 40, 41, 43, 50, 52, 57
Fiorentino, Pier Francesco 53, 100
Fredericksen, Burton B. 83, 91, 141
G
Geiger, Fred 45, 49, 57
Gherardesca Collection 95
Ghirlandaio, Domenico 147, 148, 150, 154, 155, 156, 157, 160, 161, 163, 217
Ghisolfi, Bernardo 108
Giudecca Canal 198, 203
Gnoli, Umberto 137, 139, 140, 141
Gonzaga 13, 100, 102, 104, 105, 106, 108, 109, 110

Goodhart Ducciesque Master (Goodhart Master) 13, 64, 65, 66, 67, 69, 70, 75, 76
Granacci 148, 157, 161
Grünewald 53, 60
Guardi, Francesco 12, 14, 195, 197, 206, 207, 208, 209, 210, 211, 212, 217, 219
Guggenheim 9, 32
H
Hamilton, Carl 58
Hanneman, Adrian 166, 184
Hecht, Henry 36, 46, 52
Heinemann 32
Hoenigswald, Ann 12, 31, 40, 62, 161
Huckleberry Hill 50, 51, 56, 58, 59
I
Ioni, Icilio Federico 13, 72, 75
J
Joni, Icilio Federico see Ioni, Icilio Federico
Jordaens, Jacob 166, 167
K
Kiehart, Paul 36, 37, 39, 40, 41, 46, 57
King, Daniel 171
Kneisel, Wolfgang 48
Knoedler 49, 59, 60
Köster, Christian 19, 24
Kress Collection iv, v, 3, 4, 5, 8, 9, 11, 12, 14, 15, 29, 31, 33, 34, 37, 39, 40, 43, 44, 48, 49, 50, 52, 54, 56, 58, 60,
61, 62, 75, 76, 79, 81, 85, 90, 91, 92, 95, 97, 99, 100, 110, 117, 126, 127, 131, 133, 138, 140, 141, 142, 163, 165,
168, 187, 189, 191, 194, 195, 204, 205, 212, 214, 220
Kress Foundation 3, 4, 5, 7, 8, 9, 12, 13, 15, 33, 34, 38, 39, 41, 43, 44, 45, 46, 49, 50, 52, 54, 57, 58, 59, 60, 61, 62,
66, 75, 76, 81, 83, 87, 88, 91, 97, 111, 113, 116, 125, 126, 131, 141, 144, 160, 162, 163, 194, 195, 198, 199, 202, 205
Kress, Rush 12, 31, 32, 33, 36, 38, 39, 40, 41, 43, 44, 45, 46, 49, 50, 52, 53, 54, 57, 58, 59, 60, 61, 62
Kress, Samuel H. 12, 31, 32, 33, 36, 39, 40, 66, 81, 85, 96, 116, 125, 126, 127, 131, 133, 139, 141, 142, 144, 160, 162,
163, 185, 187, 194, 195, 198, 204, 205, 212, 214
L
Lasinio, Carlo 23
Lehman Collection 65, 75, 117, 121, 123, 124, 125, 127, 205
Lehman, Robert 32, 65
Leonardo see da Vinci, Leonardo
Lewisohn 32
Lievens, Jan 166, 184

Lippi, Filippino 80, 83, 84, 88
Lomeo, Angelo 50
Longhi, Roberto 114, 115, 116, 118, 126, 127, 137, 140, 194, 198, 205
Lowe Art Museum 13, 128, 131, 220
M
Mack, Charles R. 13, 79, 90
Maes, Nicolaes 12, 14, 164, 165, 166, 167, 168, 169, 170, 172, 173, 174, 175, 176, 177, 178, 181, 182, 183, 184, 185,
186, 187, 216
Mahon, Dorothy 8, 39, 183
Manfrin Collection 24
Manning, Robert 42, 50, 113, 125
Mantegna, Andrea 22, 26, 55, 98, 100, 101, 106, 107, 108, 109, 110, 150, 162, 215, 216, 219
March, Molly 9, 13, 145, 160, 214, 215, 217, 218
Marchese Francesco  , 
Marincola, Michele 11, 15
Marini, Antonio 22
Master of the Greenville Tondo 136, 137, 138, 139, 140, 141
Master of the Manchester Madonna 145, 162
Master of the Osservanza 113, 114, 115, 116, 117, 118, 119, 121, 122, 123, 124, 126, 218
McCarthy, Joe 36, 39
McGlinchey, Christopher 14, 75, 165, 184, 186
Mellon, Andrew 32, 43, 49
Mellon Collection 52
Mellon Foundation 55
Mellon, Paul 61
Metropolitan Museum of Art 8, 9, 14, 21, 24, 32, 33, 34, 36, 38, 39, 40, 41, 44, 55, 65, 73, 75, 76, 103, 106, 110,
117, 125, 126, 127, 129, 130, 131, 141, 148, 150, 161, 162, 163, 184, 194, 196, 204, 205, 220
Michelangelo 13, 145, 146, 147, 148, 152, 154, 156, 157, 158, 159, 160, 161, 162, 163
Michelangelo Associate 13, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162,
163, 217, 218
Middeldorf, Ulrich 60
Modestini, Dianne Dwyer 4, 8, 12, 13, 15, 28, 39, 43, 61, 75, 76, 81, 87, 90, 91, 95, 97, 109, 113, 125, 129, 131, 160,
161, 162, 183, 186, 187, 214, 215, 216, 217, 218, 219
Modestini, Mario v, 4, 9, 12, 13, 36, 39, 42, 43, 50, 57, 61, 62, 71, 74, 75, 76, 81, 87, 126, 134, 160, 161, 187, 202,
205, 217, 219
Molteni, Giuseppe 20, 21, 22, 24, 25
Morelli, Giovanni 20, 21, 22, 26, 27, 103
Murphy, Franklin 60, 61
N
National Gallery of Art 9, 30, 33, 34, 35, 36, 38, 39, 40, 41, 43, 44, 45, 49, 50, 52, 53, 55, 56, 57, 58, 59, 60, 61,
62, 113, 114, 122, 123, 126, 134, 140, 141, 142, 198, 199, 205, 208, 209, 211, 212, 220
Northwick, Lord 137

O
Offner, Richard 65, 75
Okawa, Mika 13, 95, 97
Osservanza Master see Master of the Osservanza
P
Palazzo Ducale 190, 192, 194, 197, 202, 203, 204, 205
Partridge, Wendy 11, 12, 13, 14, 19, 28, 99, 109, 216
Perry, Marilyn 3, 5, 8, 39, 61, 62
Perugino, Pietro 13, 55, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 157, 215, 217, 218
Peruzzi 22
Philadelphia Museum of Art 49, 125, 142, 194, 195, 220
Pichetto, Stephen 12, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 43, 44, 45, 46, 47, 52, 56, 60, 61, 62, 96, 161
Pinturicchio, Bernardino 84, 137, 148, 161, 217
Pisanello 21
Ponte della Paglia 197, 202, 203
Princeton University Art Museum 132, 138, 142, 215
Q
Quarantelli 42, 49, 50, 55, 57, 59
R
Rembrandt 165, 166, 184, 185, 216
Rigosi, Claudio 49
Roggeman, Girard 36
Rospigliosi 47, 62
Rothe, Andrea 183
Rucellai, Giovanni 103
Rupprecht, Annette 9, 13, 133, 139, 214, 215, 216
S
Saint Catherine of Alexandria 118, 119, 120, 121, 218
Santa Maria della Salute 193, 195, 198, 202, 203, 204, 207, 210, 219
Santi, Raphael 137
Sassetta 76, 113, 114, 115, 116, 117, 118, 121, 122, 123, 125, 126, 127
Secco-Suardo, Count Giovanni 44, 47, 61
Sforza 13, 104, 105, 106, 108, 109
Shaneyfelt, Sheri Francis 13, 133, 139, 140, 141
Shapley, Fern Rusk 52, 60, 75, 76, 83, 91, 95, 96, 97, 100, 109, 110, 125, 126, 127, 129, 131, 137, 140, 141, 142, 163,
194, 199, 202, 204, 205
Sherman, Jennifer 9, 13, 28, 65, 75, 109, 214, 215, 216, 217
Smith, Marshall 171, 177, 181, 185, 187

Sozzani, Laurent 14, 165, 184, 186
Spande, Helen 14, 207, 211, 217
Steckling, Friederike 9, 109, 215, 216, 217
Suida, Dr. William E. 40, 44, 45, 49, 50, 52, 59, 60, 75, 118, 125, 126, 137, 140, 142, 194, 199, 205
Sullivan, Frank 36, 56, 61
T
Timken Museum of Art 168, 181, 187, 204
Triumphs of Petrarch 13, 98, 99, 100, 101, 106, 107, 108
Tucker, Mark 194, 195
Tura, Cosimo (also Cosmè) 23, 25, 147, 150, 155, 156, 160, 161, 162, 163
V
van der Helst, Bartholomeus 166, 184
Vannoni, Ferruccio 52, 54, 81, 82, 83, 91
Vannucci, Cristoforo 133
Vasari, Giorgio 83, 84, 139, 140, 141, 142
Venturi, Lionello 83, 118, 126
Vermeer, Johannes 166
Villa Vittoria 54, 58
Volterra, Gualtiero 44, 45, 54, 57, 161
von Sonnenburg, Dr. Hubert 55, 183
W
Walker, John 32, 33, 34, 35, 38, 40, 41, 44, 46, 49, 50, 52, 53, 54, 55, 56, 58, 59, 60, 61, 199
Walters 9, 32, 142
Warburg 32, 41
Wildenstein 49, 60, 114, 126
Z
Zeri, Federico 75, 76, 83, 91, 96, 115, 125, 127, 140, 141, 142, 161, 163

Studying and Conserving Paintings

Uploaded by

Copyright:

Available Formats

Studying and Conserving Paintings

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Studying and Conserving Paintings

Uploaded by

Copyright:

Available Formats

Studying and Conserving Paintings

Occasional Papers on the Samuel H. Kress Collection

©2006 Archetype Publications Ltd., The Conservation Center of the

Design, typography, and production by Gail A. Cimino

On the cover: detail of View of the Molo, Canaletto, ca. 1725.

Technical Studies & Treatment

Appendices & Index

222 Appendix ii: The Distribution of the Kress Collection

O ld Master paintings survive to us as fragile remnants of former

I n order to take full advantage of its location in New York City,

Margaret Holben Ellis 7

Margaret Holben Ellis 9

T his collection of fifteen papers will, we hope, be the ﬁrst in a series

I n nineteenth-century Europe many of the diﬃcult philosophical

Conveniently, a pigmented varnish also helped

If there was not enough harmony, Eastlake

There is no indication that this was ever done.

of all American museums where owners and responsible keepers of

T he sudden death of Stephen Pichetto on January 20, 1949 was a

Dianne Dwyer Modestini with Mario Modestini 43

Dianne Dwyer Modestini with Mario Modestini 45

Dianne Dwyer Modestini with Mario Modestini 47

Dianne Dwyer Modestini with Mario Modestini 49

Fig. 6. The studio at Huckleberry Hill in the 1950s.

Dianne Dwyer Modestini with Mario Modestini 51

Dianne Dwyer Modestini with Mario Modestini 53

Dianne Dwyer Modestini with Mario Modestini 55

Dianne Dwyer Modestini with Mario Modestini 57

Dianne Dwyer Modestini with Mario Modestini 59

Dianne Dwyer Modestini with Mario Modestini 61

64 Technical Studies & Treatment

T his study focuses on a Trecento ﬁve-panel polyptych, Madonna and Child

66 Technical Studies & Treatment

68 Technical Studies & Treatment

70 Technical Studies & Treatment

72 Technical Studies & Treatment

74 Technical Studies & Treatment

Appendix: Panel Paintings with Silver Grounds

76 Technical Studies & Treatment

Fig. 1. Nativity, during 1994–95 conservation.

Fig. 2. Nativity (ﬁg. 1), after cleaning and restoration.

78 Technical Studies & Treatment

I n 1943, Renaissance art historian, R. Langton Douglas concluded

80 Technical Studies & Treatment

82 Technical Studies & Treatment

84 Technical Studies & Treatment

86 Technical Studies & Treatment

88 Technical Studies & Treatment

Lightbown also noted that “conversely, there

90 Technical Studies & Treatment

92 Technical Studies & Treatment

Fig. 1. Venus and Cupid, before cleaning and restoration.

94 Technical Studies & Treatment

Mika Okawa & Dianne Dwyer Modestini 95

96 Technical Studies & Treatment

Dianne Dwyer Modestini is a paintings conservator,

Mika Okawa is a paintings conservator at the Art Conser-

Mika Okawa & Dianne Dwyer Modestini 97

98 Technical Studies & Treatment

100 Technical Studies & Treatment