Biology of Plagues PDF
Biology of Plagues PDF
Biology of Plagues PDF
Biology of Plagues:
Evidence from Historical Populations
The threat of unstoppable plagues, such as AIDS and Ebola, is always with
us. In Europe, the most devastating plagues were those from the Black
Death pandemic in the 1300s to the Great Plague of London in 1665. For
the last 100 years, it has been accepted that Yersinia pestis, the infective
agent of bubonic plague, was responsible for these epidemics. This book
combines modern concepts of epidemiology and molecular biology with
computer modelling. Applying these to the analysis of historical epidemics,
the authors show that they were not, in fact, outbreaks of bubonic plague.
Biology of Plagues oers a completely new interdisciplinary interpretation
of the plagues of Europe and establishes them within a geographical,
historical and demographic framework. This fascinating detective work
will be of interest to readers in the biological and social sciences, and
lessons learnt will underline the implications of historical plagues for
modern-day epidemiology.
is a research worker in historical demography in the School
of Biological Sciences at the University of Liverpool.
. is Emeritus Professor of Zoology also in the
School of Biological Sciences at the University of Liverpool.
They previously co-authored Human Demography and Disease (1998)
and have published many papers on population dynamics and historical
epidemiology.
Biology of Plagues:
Evidence from Historical Populations
S U SA N S C OT T AN D C HR I S TO PHE R J. DU NCAN
School of Biological Sciences
University of Liverpool
Contents
Preface
Conversion table for imperial to metric units
1 Introduction
page xiii
xiv
2
2
2
5
6
7
8
12
14
15
18
21
Transmission probability
Secondary attack rate
Basic reproductive number, R
Virulence, R and the case fatality ratio
Serial generation time: the Reed and Frost model
Contact rates
Decaying and driven epidemics
Time-series analysis of data
23
23
25
27
28
30
38
39
vi
Contents
2.9
39
41
43
44
47
47
51
53
54
55
57
58
59
61
63
65
66
68
70
72
72
74
77
78
81
82
82
86
88
91
94
Contents
4.7 The consequences of the Black Death in England
4.8 The Black Death: conclusions
4.8.1 Reservations expressed by Shrewsbury
4.8.2 The death toll of the Black Death in England
4.8.3 Epidemiological notes on the Black Death
4.8.4 Eects of malnutrition
4.9 Seasonality of the outbreaks of the Great Pestilence in
dierent localities in England
4.10 Was the Black Death an outbreak of bubonic plague?
4.11 Plagues in England following the Great Pestilence: the
14th century
4.12 Age-specic mortality of four epidemics in the 14th
century
4.13 Plagues in the 15th century
5 Case study: the plague at Penrith in 159798
5.1 Traditional account of events in the plague
5.2 Size of the population at Penrith
5.3 The three phases of the epidemic: the serial generation
time and contact rate
5.4 Spread of the epidemic at Penrith
5.5 The epidemic during the rst two phases: elucidation
of the epidemiological characteristics of plague
5.6 Explosion of the epidemic in phase three
5.7 Age- and sex-specic mortality in the plague at Penrith
5.8 Wills and testaments of those who died in the plague at
Penrith
5.9 Response of the population at Penrith after the plague
5.10 Classication of the epidemics of haemorrhagic plague
6 Pestilence and plague in the 16th century in England
6.1 The Sweating Sickness
6.2 Plagues in London in the 16th century
6.2.1 The rst half of the century
6.2.2 Did plague become endemic in London during
the second half of the 16th century?
6.2.3 Epidemics in London, 15421600
6.3 Plagues in central and southern England during the
16th century
vii
96
98
98
101
103
105
105
107
109
111
112
115
116
118
119
126
126
135
137
140
146
148
149
149
152
152
154
155
165
viii
Contents
6.4 Case study of the plague at Stratford-upon-Avon, 1564
6.5 Conclusions
170
173
175
176
178
181
182
183
184
185
186
189
190
192
193
198
199
205
206
211
211
215
217
220
221
222
222
223
225
226
226
228
234
Contents
9.2
9.3
9.4
9.5
9.6
9.7
ix
237
239
239
240
243
248
248
249
250
250
251
255
258
261
261
263
265
269
275
279
280
282
284
284
286
286
292
293
295
297
Contents
11.3 Italy
11.3.1 Occurrence and frequency of the epidemics in
Italy
11.3.2 Signs and symptoms
11.3.3 Analysis of the spread of epidemics
11.3.4 Plague epidemics in Italy
11.3.5 Conclusions
11.4 The Iberian peninsula
11.4.1 The major epidemics
11.4.2 Epidemics at Barcelona
11.5 Germany, Austria, Bohemia and Switzerland
11.6 The Benelux countries
11.7 Spread of the plague across Europe
303
304
305
306
308
315
318
318
324
328
331
331
338
344
350
350
352
352
354
356
362
364
366
367
370
371
376
378
379
381
383
Contents
xi
384
389
References
Index
396
410
389
393
Preface
xiii
Imperial unit
Metric equivalent
1 inch
1 foot
1 yard
1 mile
1 acre
1 square mile
25.4 millimetres
0.3048 metre
0.9144 metre
1.609 kilometres
0.405 hectare
259 hectares
xiv
1
Introduction
Introduction
1.1 What is a plague?
The Bible uses the word plague to describe an aiction that was regarded
as a sign of divine displeasure or as an aiction of humankind such as the
plague of locusts. Nowadays, it is a term used to describe a deadly epidemic
or pestilence and The Wordsworth Encyclopedia of Plague and Pestilence
(Kohn, 1995) lists a seemingly endless catalogue of historical epidemics
from all over the world, including smallpox, cholera, typhus and malaria.
They were all infectious and potentially lethal, caused high mortality and
were serious historic events. The inuenza pandemic of 191719, with a
nal death toll worldwide estimated at more than 20 million, is a good
example. The incubation period was often less than 2 days so that its
worldwide spread was dependent on 20th century means of rapid travel
that could move people in bulk, namely steamtrains and steamships of
which the troop ships of the First World War are a good example.
However, the basic etiology of these diseases is now usually well understood and, in spite of the terrible death toll, the percentage mortality of the
aected populations was not relatively high (Langmuir et al., 1985). In this
book, we are mainly, but not exclusively, concerned with the Black Death,
arguably the most awful epidemic ever to have struck, which raged in
Europe from 1347 to 1350, and the unremitting succession of plagues that
followed it for 300 years. These reached their peak in continental Europe
during 162531 and in England in 166566, but they then disappeared
completely after about 1670. When these plagues struck a naive population, where we have reasonably accurate data available in Italy and England, mortality could reach about 50%; we do not know what was the
causative agent in these terrible epidemics.
Longrigg, 1980; Langmuir et al., 1985; Kohn, 1995; Olson et al., 1996;
Retief & Cilliers, 1998), but the rst vivid description was given by Thucydides, himself a victim who survived the outbreak (see Page, 1953). It is
sometimes termed the Thucydides syndrome because of his evocative
narrative.
People were stricken suddenly with severe headaches, inamed eyes, and
bleeding in their mouths and throats. The next symptoms were coughing,
sneezing, and chest pains followed by stomach cramps, intensive vomiting
and diarrhoea, and unquenchable thirst. The skin was ushed, livid and
broken with small blisters and open sores. The patients burned with fever
so extreme that they could not tolerate being covered, choosing rather to
go naked. Their desire was to cast themselves into cold water, and many of
those who were unsupervised did throw themselves into public cisterns,
consumed as they were by unceasing thirst. Many became delirious and
death usually came on the seventh or eighth day of the illness, although
those who survived the rst phase often died from the weakness brought on
by constant diarrhoea. Many who recovered had lost their eyesight, their
memory, or the use of their extremities.
This plague is believed to have originated in Ethiopia and travelled
through Egypt and the eastern Mediterranean before reaching Athens. The
rst cases appeared in Piraeus, the Athenian port and base for many
travellers and merchants who probably contracted the disease in their
journeys abroad. It spread rapidly to the upper city and whole households
were left empty. Mortality among doctors, as among other attendants of
the sick, was especially high. Fearful of an attack by the Spartans, the
Athenian leader Pericles ordered the inhabitants of the surrounding countryside to move inside the city, where they could be protected by the army
and the fortied walls. Many country dwellers, coming to an already
overpopulated city, had no place to live except in poorly ventilated shacks
and tents. This mass of people, crowded together in the hot summer,
created a situation that was ideal for the rapid transmission of the disease.
Though there were many dead bodies lying unburied, there was said to be a
complete disappearance of birds of prey and dogs. Apparently it was rare
to catch the disease twice, or if someone did, the second attack was never
fatal. A peak case rate was reached during the Spartan siege, which lasted
40 days, after which the crowded refugees dispersed. The disease remained
at a low level through 429 BC (when Pericles died of it) and returned in
force in the summer of 428 BC at the time of another Spartan siege. The
disease was quiescent, or even absent, from the winter of 428 BC until the
summer of 427 BC, but broke out again in the autumn or early winter of
Introduction
427 BC. This epidemic lasted no less than a year, but there is no further
mention of the disease. The total number of Athenians who died is not
recorded but, over the 3-year period, of 13 000 enrolled hoplites (soldiers),
4400 died a mortality rate of 33%. Hagnon took the eet and sailed to
Potidaea carrying the plague there also and this made dreadful havoc
among the Athenian troops. Even those who had been there previously and
had been in good health caught the infection and so 1050 men out of 4000
were lost in about 40 days.
There have been several identications of the causative agent of the
plague at Athens, including smallpox (Littman & Littman, 1969), scarlet
fever, measles and typhus (Shrewsbury, 1950; Page, 1953) but these are all
now discredited. Langmuir et al. (1985) concluded that the clinical descriptions clearly indicated the involvement of specic organ systems and that
there was an obvious inammatory condition of the eyes and respiratory
tract; this acute respiratory infection was severe and probably necrotising;
the initiation with vomiting followed by empty retching and later by
watery diarrhoea strongly suggested a gastroenteropathy mediated by
the central nervous system rather than a local inammatory process.
Langmuir et al. (1985) believed that the skin lesions were suggestive of
bullous impetigo. They did not suggest that the Thucydides syndrome was
identical with the modern toxic shock syndrome but believed that the
same basic pathogenic mechanisms were involved, in that there was infection in predisposed hosts by a possibly non-invasive Staphylococcus sp.
that was capable of producing an exotoxin similar to toxin-1 of the toxic
shock syndrome (Rasheed et al., 1985). This toxin may have diered from
toxin-1 in that it produced predominantly enterotoxic eects and less
profound circulatory collapse, and had only moderate or no erythrogenic
potential.
Morens & Littman (1992, 1994) have approached the plague at Athens
from a dierent viewpoint and have arrived at a conclusion that is strongly
opposed to that of Langmuir et al. (1985). We describe their hypothesis
briey because they use mathematical modelling techniques that we shall
also employ to elucidate the epidemiological parameters of later plagues.
They have reduced the reliance on clinical symptoms in favour of the
epidemiology of the disease because pre-modern descriptions, which lack
detailed information on serology and accurate accounts of rashes and
other clinical features, always retain a high degree of uncertainty. Use of
the Reed and Frost mathematical model (section 2.5) led them to conclude
that, under any conditions of crowding that probably prevailed in Athens
in 430 BC, an epidemic of inuenza would have died out rapidly in a few
weeks. They excluded all common diseases and most respiratory diseases
and concluded that the cause of the Athenian epidemic could be limited to
either a reservoir disease (zoonotic or vector-borne) or one of the few
respiratory diseases that are associated with an unusual means of persistence: either environmental/fomite persistence, or adaptation to indolent
transmission among dispersed rural populations. They suggested that the
diseases in the rst category include typhus, arboviral diseases and bubonic
plague and, in the second category, smallpox. Retief & Cilliers (1998) also
reviewed the epidemiological evidence and agreed that the only possibilities are epidemic typhus, bubonic plague, arboviral disease and smallpox.
Other workers have suggested that the plague at Athens was an early
manifestation of Ebola (sections 1.3 and 13.15). Olson et al. (1996) stated
that a modern case denition of Ebola virus infection records sudden onset,
fever, headache, pharyngitis followed by cough, vomiting, diarrhoea,
maculopapular rash, and haemorrhagic diathesis, with a case fatality rate
of 50% to 90%, death typically occurring in the second week of the disease.
In a review of the 1995 Ebola outbreak in Zaire, the Centre for Disease
Control and Prevention reported that the most frequent initial symptoms
were fever (94%), diarrhoea (80%), and severe weakness (74%), with dysphagia and clinical signs of bleeding also frequently present. Symptomatic
hiccups were also reported in 15% of patients. Olson et al. (1996) concluded that the prole of the plague at Athens was remarkably similar to
that of the recent outbreaks of Ebola in Sudan and Zaire. Certainly, as we
shall see, this devastating epidemic had features in common with later
plagues in Europe (section 1.3).
Introduction
1.2.3 The Great Age of plagues: the Black Death and thereafter
The Black Death erupted in Sicily in 1347 and the pandemic spread
through Europe during the next 3 years, reaching Norway (where twothirds of the population died; Carmichael, 1997) and Sweden and crossing
to England (and thence to mainland Scotland, the Hebrides, Orkney and
the Shetland Islands) and to Ireland (Biraben, 1975) and, possibly, to
Iceland and Greenland (Kohn, 1995). Its arrival presaged a continuous
succession of epidemics in Europe for the next 300 years before it disappeared completely around 1670. The enormous mortality of the Black
Death had a major impact on the demography of Europe, and the population of England did not fully recover for 150 years. Events during 134750
are described in Chapter 4 and the demographic consequences of a major
mortality crisis on a population are discussed further in section 13.17. Were
the multiplicity of plagues throughout Europe from 1350 to 1670 all the
result of the same causative agent as that responsible for the Black Death,
albeit with some minor mutations during that time? It is not possible to
answer this question with certainty, but we believe that the most probable
explanation of the etiological and epidemiological details is that it was so.
During the second half of the 14th century, the epidemics in England and
continental Europe were less virulent but the infection gradually regained
its ferocity, reaching its peak around 1630 in France and 166566 in
England.
Introduction
Where climatic conditions were suitable and reservoir rodent species were
present locally, endemic bubonic plague could be established. Where only
the climate was suitable, as in the coasts of France, Spain and Italy,
epidemics of bubonic plague could potentially break out, having been
brought into the ports by sea, but these terminated once the local rats had
died. During the third age of plagues, 13471670, therefore, Europe probably experienced minor outbreaks of bubonic plague along the Mediterranean coasts of Italy, Spain and France in addition to the major epidemics
of haemorrhagic plague.
Finally, endemic bubonic plague erupted in a series of epidemics in India
at the very end of the 19th century and spread across southeast Asia, and so
began the fourth age of plagues. As we show in Chapter 3, the arrival of
steamships then allowed rats and their eas carrying bubonic plague to be
rapidly transported from the grain stores on the docks of China to subtropical regions wherever suitable indigenous rodents occurred.
widespread and even global spread of both new and old infectious diseases.
This new and heightened vulnerability is not mysterious; the dramatic
increases in the worldwide movement of people, goods, and ideas is the
driving force behind the globalisation of disease because not only do
people travel increasingly, they also travel much more rapidly, and go to
many more places than ever before. The lesson is clear: a health problem in
any part of the world can rapidly become a widespread health threat
(Mann, 1995).
Most emergent viruses are zoonotic, with natural animal reservoirs a
more usual source of new viruses than is the spontaneous evolution of a
new entity. Human behaviour increases the probability of the transfer of
viruses from their endogenous animal hosts to humans. The original source
of the AIDS pandemic has been traced back to a subspecies of chimpanzee
that has been used for food in West Central Africa, the hunters being
exposed to infected blood during the killing and dressing. The virus has
probably been living harmlessly in chimpanzees for hundreds of years and
may have been transferred to humans throughout history, but the socioeconomic changes in Africa provided the particular circumstances leading
to the spread of HIV and AIDS.
An outbreak of encephalitis in Malaysia in 1999, which killed 76 people
may have been caused by a more deadly version of the Hendra virus, which
was rst identied in Australia 5 years previously. The dierence is that,
whereas the virus in the earlier outbreak did not spread easily between
animals, the Malaysian version apparently did: all the Malaysian victims
were connected with pig rearing. Health ocials in Asia now fear that a
dangerous new human pathogen has emerged that has spread from fruit
bats via pigs and consequently more than 300 000 pigs were slaughtered in
southern Malaysia as an initial precautionary measure. The spinal uid
taken from ve patients contained a paramyovirus (named Nipah) and
analysis of the amino acid and RNA sequences conrmed that it is related
to the deadly Hendra virus. Why has the virus suddenly begun to kill pigs
and people when the bats may have harboured it safety for centuries?
The Ebola virus, a member of the Filoviridae, burst from obscurity with
outbreaks of severe haemorrhagic fever. It was rst associated with an
outbreak of 318 cases and a case fatality rate of 90% in Zaire and caused
150 deaths among 250 cases in Sudan. Smaller outbreaks continue to
appear periodically, particularly in East, Central and southern Africa. In
1989, a haemorrhagic disease was recognised among cynomolgus
macaques imported into the USA from the Philippines; strains of Ebola
virus were isolated and serologic studies indicated that the virus is a
10
Introduction
11
the fear that this strain may resurface, perhaps in as virulent a form as in
1918. Many pandemics originate in Asia, notably China, where enormous
numbers of ducks, pigs, and other virus-producing animals live in close
proximity to human beings (Kohn, 1995). Avian inuenza A (H5N1) virus
has recently been shown to be transmitted from patients to healthcare
workers in Hong Kong and this nding may portend a novel inuenza
virus with pandemic potential (Bridges et al., 2000).
Fragments of the virus responsible for Spanish inuenza were found in
1998 in the lungs of a woman who died in the 1918 epidemic and whose
body was preserved by huge layers of fat and the frost of Alaska and it is
hoped that it will be possible soon to map the RNA of the virus to identify
the gene that made it so deadly. Preliminary work has produced the
complete sequence of one key gene and the existing strain to which the 1918
sequences are most closely related is A/Sw/Iowa/30, the oldest classical
swine inuenza strain. More recently, inuenza 1918 RNA has been found
in respiratory tissue and the brains of Spitzbergen coal miners who died in
the epidemic and Oxford (2000) suggested that this could be a piece in the
jigsaw linking pandemic inuenza to the ensuing outbreak of the sleeping
disease encephalitis lethargica.
A virulent and drug-resistant form of typhoid caused by the pathogen
Salmonella typhi, which kills 600 000 people a year, has now emerged in
Vietnam. The study of its genome is now almost complete: the nucleus
contains three separate pieces of DNA, a massive coil some 4.5 million
bases long and two plasmids, smaller loops of genetic data. One of the
plasmids contains an array of oensive and defensive genes, which probably explain the potency of this strain of typhoid. It came as a great surprise
when it was discovered that the other plasmid contained a sequence of
5060 genes that are found in Yersinia pestis, the bacterium of bubonic
plague and thus the Vietnamese microbe appears to be fortied with the
genes of other pathogens (Farrar, 2000).
There is a seemingly endless catalogue of lethal infectious diseases that
have emerged. Some of these have been described in a very lively manner
by Garrett (1995): Lassa fever, Bolivian haemorrhagic fever, Marburg
virus, the Brazilian meningitis epidemic and the Hantaviruses. Health
ocials in New York City reported, in August 1999, an outbreak of what
appeared to be St Louis encephalitis, a disease that can spread to humans
from birds via mosquitoes. However, it has now been discovered that the
infectious agent is West Nile virus, which is normally found in Africa and
Asia and is also transmitted by mosquitoes. Helicopters sprayed entire
neighbourhoods in Queens, New York, after the disease killed horses,
12
Introduction
thousands of birds and several people; there has been a fresh outbreak in
New York City in summer 2000 and what really alarms American health
ocials is the danger of the disease establishing itself permanently in the
country. It remains an open question as to how the virus reached the USA
(Boyce, 1999). So, it should not surprise us that the classical pandemics of
historical times emerged and it is probable that they originated as viral
zoonoses. Viruses have a great capacity for mutating and are opportunistic
parasites; the worrying thought is (as suggested above) where and when
will they next strike?
13
dierent sizes. The city of London, as we shall see, was a complex population; a very large number of individuals crammed together but with subsets
delineated by class and parish, with partial intercourse. The population
was freely open, with many immigrants, travellers and merchants arriving
daily by land and sea.
The spread of an outbreak of plague may also be studied at a higher
population level, i.e. throughout a geographically dened area that might
be the size of a country or even part of a continent. Examples are island
Britain, and the Iberian peninsula, which was eectively separated from
continental Europe by the Pyrenees and, in both, plague epidemics had to
enter from the sea via the ports. These may be called metapopulations, a
term used by ecologists to describe a population of populations. The study
of metapopulation dynamics in biology is normally concerned with the
behaviour of a single species over time; there are no static populations and
likewise there is no such thing as a static metapopulation. The metapopulation concept in ecology is closely linked with the processes of population
turnover, extinction and the establishment of new populations. Ecological
metapopulation theory, with one important exception, has not been applied to human populations; indeed, as originally dened, it is not strictly
applicable because it deals with extinctions and recolonisations and makes
the simplifying assumption that each ecological site is regarded as being in
one of two alternative states, either empty or lled at their local carrying
capacity, characteristics that were rarely found in England during the age
of plagues. The exception concerns studies of spatial heterogeneity and the
epidemic spread of infectious diseases through a human metapopulation
where individuals can be either infected or uninfected, an example of the
interaction between demography and disease.
The spread of epidemics is an important part of modern Geography and
we have used such techniques as disease centroids to trace the spatial
movements of the plague in a metapopulation where it was endemic (see
section 2.12). It becomes evident that the Black Death had a dierent
pattern of spread from subsequent plague epidemics which, in turn, exhibited a range of sharply diering characteristics. The Black Death recognised no boundaries, either natural or human engendered, and spread in a
wave-like movement all across Europe and to o-shore islands in about 3
years before disappearing. We can regard its territory for this brief period
as a supermetapopulation. Bubonic plague as a disease of rodents, and
secondarily of humans, was certainly established as endemic across a huge
subtropical area by 1900, from the Levant across to China and Southeast
Asia. It has persisted for many years and we can regard this also as a
14
Introduction
Shrewsbury (1970) considered that When more than 66% of the total
annual burials occurs in the three months of July to September inclusive,
the record is almost certainly indicative of an outbreak of bubonic plague
and this led him to conclude, for example, that there were multiple plague
epidemics in northwest England in 1623. However, this area was living on
the margins of subsistence at this time and mortality was sensitive to a 5- to
6-year cycle in grain prices (Scott & Duncan, 1998) and the constituent
communities suered major mortalities not only in 1623, but in 158788
and 159697 also. We have shown that in these years the peak of wheat
prices coincided with a low in the 12-year cycle of wool prices (Scott &
Duncan, 1997, 1998). Those populations that depended on both commodities suered severely whereas those that depended on only one for their
livelihood escaped unscathed. They did not suer from plague in 1623. It is
evident that by the end of the 16th century all towns could recognise plague
15
16
Introduction
As his title suggests, Shrewsbury believed whole-heartedly that bubonic plague was responsible for the majority of the plague epidemics
in England and Scotland and yet, as a trained medical microbiologist,
he saw that the facts on many occasions, made this a biological
impossibility. He was therefore frequently forced to adapt his conclusions. When plague was reported in the months December to
February he stated that it must have been a mild winter. When other
17
18
Introduction
Morris also attacks because of Shrewsburys statement that bubonic
plague has an unvarying relationship with rodent enzootics:
Shrewsburys main contention is that the country would have had to be
constantly re-infected by fresh importations of plague-bearing rats. He has not
thought of the possibility that England might well have become an enzootic
area in which some rats at any given time are diseased. This is odd since he
knows very well that in other parts of the world plague has taken permanent
root and produced notorious enzootic or endemic centres. Indeed he argues,
mistakenly as it happens, that India has always been one such centre from
which Europe has drawn its periodic re-infections . . . Besides, if England
became, as obviously it did, a permanently enzootic area in the seventeenth
century, why should it not have done so two centuries earlier? That plague was
endemic, or at least enzootic in London, needing no imported re-infections, for
more than half a century before 1665 is abundantly clear from the annual
mortality bills.
(iv) Biraben (1975, 1976) in his two-volume work Les hommes et la peste en
France et dans les pays europeens et mediterraneens has assembled an
impressive set of data on plague epidemics in Europe after the Black
Death. He has combed the literature extensively and his bibliography
runs to over 225 pages. We have used these data-sets for analysis in
Chapters 11 and 12.
(v) Graham Twigg is a zoologist who has specialised in the biology of
rodents and who has discussed with Dr D. E. Davis the status of rats in
the Middle Ages. In 1984 he wrote The Black Death: A Biological
Reappraisal, in which he carefully develops the evidence that shows
that bubonic plague was not the cause of this great pandemic. He
summarised his seminal work in the conclusion The logistics of the
epidemic in England support the hypothesis of an air-borne organism
of high infectivity and virulence, having a short incubation period and
being spread by respiratory means (Twigg, 1989). All students of
plague should read his work.
1.7 Objectives
An epidemiologist must, by denition, be an historian, even if only in the
short term. We present a new analysis of the plagues that scourged Europe
from the 14th to the 17th centuries, approaching from biological, ecological and epidemiological viewpoints. We analyse the historical data (hopefully objectively) using modern techniques of theoretical epidemiology,
clinical molecular biology, computer-based modelling and the spatial
models of epidemic spread that have been developed by geographers.
1.7 Objectives
19
20
Introduction
2
Epidemiological concepts
22
Epidemiological concepts
Time of
infection
(A)
Incubation period
Symptoms
Time
(B)
Latent
period
Infectious period
Recovered
or dead
Fig. 2.1. The sequence of events during a simple, directly transmitted disease. The
dynamics of the disease are shown in A and the dynamics of infectiousness in B. The
victim may cease to be infectious before the end of the symptomatic period.
23
years, during which time a great many people may be infected (Halloran,
1998). We show in Chapter 5 that people in England infected with plague
were infectious for some 22 days before the symptoms appeared, a critically
important factor that ensured the establishment of an epidemic in spite of
some of the public health measures that were enforced.
24
Epidemiological concepts
SAR :
Using the data available for the plagues that we describe in detail in
subsequent chapters, estimations can be made of the household SAR,
which is concerned solely with infections within the house, although, even
with a full family reconstitution, the exact number of resident susceptibles
cannot be known; for example, some of the children may have left home.
There is another caveat to bear in mind; the record of a burial in the
registers is the only indication of an infection and we have no means of
identifying individuals who contracted the plague and recovered. However,
we have been encouraged by the consistency with which the pattern of the
spread is replicated in each parish that we have analysed. The technique
that we have used is shown diagrammatically in Fig. 2.2, where the point of
infection of the index (primary) case is shown (P ); this indicates the start of
the infection in the household. The latent period of P (see Fig. 2.1) follows
this starting point and anyone else infected during this time is designated as
a co-primary (P ) because they cannot have been infected by P . Anyone
infected during the infectious period of P (Fig. 2.2) is designated a second
ary (S ) or co-secondary (S , S etc.).
Tertiary and higher cases are those occurring after the maximum allowable time interval for the secondary cases. We assume that the infectious
period is terminated in haemorrhagic plague by the death of the victim,
who was probably displaying symptoms, on average, for the last 5 days of
his life. The picture is complicated when co-primaries are identied because
their infectious periods extend beyond the death of P . By working through
the data for a large number of households derived by family reconstitution
from the parish registers, it is possible to derive estimates of the latent,
infectious and incubation periods for haemorrhagic plague, as shown in
Fig. 2.2.
Thus, in assessing the household SAR, it is necessary to determine
whether each case in each household is a co-primary, secondary, tertiary,
or higher generation case. The estimated household SAR is the total
number of secondary cases in all households divided by the total number of
at-risk susceptibles in all households. Co-primary cases are excluded from
the denominator; tertiary or higher cases are excluded from the numerator
but included in the denominator. As we shall see, there was wide variation
in the household SAR of the plague, apparently dependent on such factors
as the season and the progress of an epidemic.
LP
(I)
P2
25
Maximum
infectious
period
of P1
LP
(I)
S1
LP
(I)
S2
LP
(I)
T1
LP
Fig. 2.2. Diagram to illustrate the time-course of the spread of an infectious disease
such as haemorrhagic plague. The time between the point of infection (I) and the
end of the infectious period (which may coincide with the death of the victim) is
shown as a scaled horizontal line which is subdivided into latent (LP) and infectious
periods. In this example, two primary cases (P and P ) are shown and the time
infectious
period) are shown by
limits when P can infect secondary cases (P s
the vertical dashed
lines. Two secondary cases (S and S ) are shown; they could
case (T
) is shown; I occurred
have been infected by either P or P . One tertiary
of both P and P and so this could not be a
after the end of the infectious periods
secondary case.
26
Epidemiological concepts
by one infective during the infectious period is the product of the number of
contacts in that time interval and the transmission probability per contact:
number of
transmission
duration of
R : contacts per ; probability ;
: cpd
infectiousness
unit time
per contact
An infectious disease does not have a specic value for R that diers
within particular host populations at particular times. For example, contact rates in rural areas will be lower than contact rates in urban areas, i.e.
R of smallpox would have been lower in rural England than in London
(Scott & Duncan, 1998). As we shall see, R for the plague in one locality
apparently diered markedly at dierent seasons.
R of Yersinia pestis, the bacterium of bubonic plague of rodents (see
Chapter 3), in infections in a rat population may be high during the season
of high ea density but low when the eas are not active. This is an example
of an indirectly transmitted disease, where the bacterium is transmitted
between two dierent host populations, the rat and the ea; it is clearly a
more complex situation than in person-to-person infection, as is illustrated
in Fig. 2.3. R for indirectly transmitted diseases depends on the product of
the two components of transmission.
R assumes that all contacts are with susceptibles and that none are
already immune, unlike the situation with the regular lethal smallpox
epidemics in rural towns in 17th century England where virtually the only
susceptibles were those born since the last epidemic (Duncan et al., 1993a;
Scott & Duncan, 1993). The plague at Penrith in 159798 is described in
detail in Chapter 5 and it was found that the mortality in the dierent age
groups was indiscriminate. Some 45% of the population died but it seems
certain that some individuals were exposed to the infection but survived;
they may have contracted the disease but not have died of it, or they may
have been resistant. We begin by assuming that all individuals in the plague
at Penrith and certainly in the Black Death (but not necessarily in London
in the 17th century) were potentially susceptible.
The parameter R allows the comparison of diseases from the viewpoint
of population biology the aim of this book. When a disease is endemic (i.e.
the disease lingers at about the same incidence for a long time) an infectious
case produces, on average, one new infectious case, i.e. R : 1. If R 1,
the disease will disappear, whereas if R 1 an epidemic will be initiated.
As an epidemic proceeds the number of susceptibles will inevitably fall and
when R 1, the epidemic will die out.
27
Fig. 2.3. Diagram to illustrate direct and indirect transmission. The quantities T
and T represent a summation of the transmission parameters for the ow of the
infectious
agent. In direct, person-to-person transmission, as in measles (A),
R : T . In indirect transmission, as in bubonic plague where Yersinia pestis cycles
rodents and eas (B), R : T ; T .
between
28
Epidemiological concepts
data but it was certainly very high indeed, although there is evidence that it
may have decreased in the later stages of some epidemics.
K
N91
(2.1)
29
and
q:19p
(2.2)
(2.3)
and the probability of any given individual having at least one adequate
contact with any of the cases will be
P : 1 9 Q : 1 9 q!R
R
R
(2.4)
R>
: S (1 9 q!R)
R
(2.5)
30
Epidemiological concepts
The results of modelling Equation 2.5 are shown in Figs. 2.4 and 2.5,
using a population of 1200 and thereby replicating a rural town in Tudor
times. The overall shapes of the graphs are similar, usually rising to a peak
of cases more slowly than they decay, but changing the parameters makes a
major dierence to the time-course of an epidemic. It can be seen that in
most epidemic proles at least 90% of the susceptible population become
infected.
Figure 2.4 illustrates the time-courses of the epidemics of dierent infections in which the mean number of contacts is standardised at 12 (i.e.
p : 0.01). Serial generation times are as follows: A : 3 days (inuenza);
B : 10 days (measles); C : 15 days (chickenpox); D : 22 days (haemorrhagic plague). All the epidemics, with p : 0.01, were completed in 5 serial
generation times and hence lasted from 15 days in inuenza (A) to 110 days
in plague (D).
Figure 2.5 illustrates the eects of changing the average number of
contacts when the serial generation time is maintained constant at 22 days,
i.e. equivalent to haemorrhagic plague. When the number of contacts was
set at the low value of 3 (Fig. 2.5A) the epidemic lasted over 200 days (9
generation times) and there was a slow build-up to the peak of cases.
Increasing the number of contacts (Fig. 2.5B and C) markedly reduced the
duration of the epidemics until, with an average of 40 contacts, the epidemic exploded dramatically and was nished in 4 generation times (Fig.
2.5D).
Clearly, Equation 2.5 depends on the random mixing of infectives and
susceptibles, which is not the case in some circumstances. Nevertheless, the
modelling shows how each disease leaves its ngerprints on the epidemic,
dependent on the serial generation time through which it can be identied
by the epidemiologist. It also reveals how the time-course and pattern of an
epidemic is modied by local factors such as the probability of an adequate
contact between two individuals, which, in turn, is dependent on such
factors as the density and social customs of the population, i.e. p will be
much higher (i) at a local summer fair than when the community is
snow-bound in the depths of winter and (ii) at the start of an epidemic when
isolation and quarantine practices were not in force. We shall see that this
is an important point because there were marked seasonal dierences in the
dynamics of plague, both in England and continental Europe.
31
(2.6)
(2.7)
dY/dt : H 9 ( ; )Y
(2.8)
dZ/dt : Y 9 Z
(2.9)
32
Epidemiological concepts
Fig. 2.4. Results of Reed and Frost modelling for a population where N : 1200
(representing a Tudor rural town) and the mean number of eective contacts : 12.
Serial generation times (days): A : 3, B : 10, C : 15 and D : 22. The duration of
the epidemic is dependent on the serial generation time. Note the dierent scales.
33
34
Epidemiological concepts
Fig. 2.5. Results of Reed and Frost modelling for a population where N : 1200 and
the serial generation time : 22 days (replicating haemorrhagic plague). Mean
number of eective contacts: A : 3, B : 6, C : 20 and D : 40.
35
36
Epidemiological concepts
(Anderson & May, 1991). The disease will maintain itself within the population provided that the reproductive rate of the infection, R, is greater than
or equal to unity; R is the expected number of secondary cases produced by
an infectious individual in a population of X susceptibles (see section 2.3).
The value of R equals R in a disease-free population, which, as will be
shown later, was the case in the Black Death and for many of the plagues in
rural England in the 16th century. For a system dened by Equations 2.6 to
2.9,
X
R :
( ; )( ; )
(2.10)
The criterion R 1 for the establishment of the disease (see section 2.3)
can be expressed as the requirement that the population of susceptibles
exceeds a threshold density, X N , where
2
N : ( ; )( ; )
2
(2.11)
(2.12)
For most infectious diseases (but not e.g. HIV), the duration of the latent
(1/) and infectious (1/) periods are of the order of days to (at most) a few
weeks, whereas life expectancy in Tudor and Stuart times (1/) was approximately 25 years (Scott & Duncan, 1998). Under these circumstances,
and
and Equations 2.10 and 2.11 may be approximated as
R : X/
(2.13)
N : /
2
(2.14)
and
37
38
Epidemiological concepts
the human life expectancy of the population, and A the average age at
infection, as above (Anderson & May, 1991). It can be shown (Scott &
Duncan, 1998) that Equation 2.15 is formally equivalent to
T:
2
([N 9 ( ; )]
(2.16)
39
40
Epidemiological concepts
Fig. 2.6. Annual smallpox deaths (ordinate) in London over some 250 years,
16471893, divided into cohorts: I : 16471707; II : 170850; III : 17511800;
IV : 180135; V : 183670; VI : 187193. The plague in 1665 and the major
smallpox epidemics of 1838 (E ) and 1871 (E ) are indicated. The dashed line
(closed triangles) gives the cumulative
number ofbaptisms in the preceding 25 years
(thousands), right-hand ordinate. Data sources: Bills of Mortality (Creighton, 1894)
and Wrigley & Schoeld (1981).
41
(I)
(II)
(III)
(IV)
42
Epidemiological concepts
Fig. 2.7. Mixing patterns of two groups where c : contact rate of set 1 with set 2,
c : contact rates within sets 1
c : contact rate of set 2 with set 1, and c and
and 2, respectively. From Halloran (1998).
having an equal chance of making contact with, and being exposed to,
infection by each other person. However, most populations do not mix
randomly but have subgroups that mix more with their own members than
with other groups. These subgroups in the community during a plague
were clearly households where the household SAR was of paramount
importance in determining the spread of the infection through the family
(see the case studies at Penrith, Chapter 5, and Eyam, Chapter 10). The
magistrates in London enforced quarantine:
The misery of those families is not to be expressed; and it was generally in such
houses that we heard the most dismal shrieks and outcries of the poor people,
terried and even frightened to death by the sight of the condition of their dearest
relations, and by the terror of being imprisoned as they were.
(Defoe, 1722)
We show later that, in rural towns, the pestilence spread rapidly within a
family once it was introduced but transmission to other households was
slower and more dicult and this is a key point if the plague were to
persist and spread in a population and not die out it must eect transmission to at least one other household.
The contact rate of individuals of group 1 with individuals of group 2 is
denoted by c (Fig. 2.7) and the contact pattern is described by a mixing
matrix that has the same number of rows and columns as the number of
mixing groups. The entries in the matrix represent the rate of contacts of
individuals within and between the groups. The mixing pattern of two
groups is represented by the matrix:
C:
c
c
c
c
On the diagonals are the rates of contacts within groups, c and c . The
o-diagonal entries, c and c , represent the rates of contacts between the
groups corresponding to that row and column (Halloran, 1998).
43
44
Epidemiological concepts
(see Fig. 2.8B), the epidemic does not explode but dies out before all the
susceptibles become infected. Examples of the plague following this pattern
are in communities of low density (e.g. in Cheshire, sections 9.1.2 and 9.6.1)
and outbreaks in winter in rural towns in England before the major
epidemic in the following summer (sections 5.5, 10.3 and 10.4).
If a population is open (e.g. London in the 17th century), with births and
immigration providing a steady supply of susceptibles, the parasite may
not die out but can persist and become endemic. Figure 2.9B illustrates an
open population with a low R ; after an epidemic the susceptibles, infec
tives and immunes remain in dynamic steady state, with the annual number of new incident cases remaining low and steady. When R is high in the
open, endemic situation, the level of infectives is higher and the number of
susceptibles is correspondingly lower (Fig. 2.9A).
45
46
Epidemiological concepts
3
The biology of bubonic plague
48
near the central Asiatic plateau and he considered this to be the home of
the disease. Stemming therefrom were the following major foci of the
disease:
(a) in the foothills of the Himalayas between India and China,
(b) in Central Africa in the region of the Great Lakes,
(c) scattered across the entire length of the Eurasian steppe from Manchuria to the Ukraine.
Hankin (1905) believed that Garhwal in India (focus (a), above) was the
area where bubonic plague was endemic. It is a mountainous and somewhat inaccessible region that had little ordinary trac with the rest of
India. However, Hankin (1905) stated that the rst recorded plague in
Garhwal in humans was only in 1822; thereafter irregular outbreaks were
recorded during the 19th century, usually conned to a few villages, some
said to be severe, and 535 deaths were recorded in the epidemic of 1877.
Only a handful of people died in most outbreaks and these are probably
authentic cases of bubonic plague that led up to the pandemic in India
which began in 1895 and continued for more than half a century.
Before the outbreak of plague in Garhwal in 1822, Hankin (1905) recorded only the following pestilences in India:
1344 Army of Sultan Mahommed Tughlak destroyed by pestilence probably near Deogiri, a town a short distance from Nassik.
1611 Plague said to have begun in Punjab. It lasted 7 years, and spread to
Delhi, Agra, Cashmere and Kandahar.
1683 Conned to Western India; lasted for 8 years in Ahmedabad.
1812 Began in Gujerat and lasted 9 years.
East Africa (focus (b), above) is believed to have been aected by plague
in the 6th century AD (Roberts, 1935) and Twigg (1984) recorded a plague
in this region at Mombasa in 1697 that lasted almost 3 years. If this were
bubonic plague, it probably spread from the caravan routes from the north
to the east coast and it was known in the northeastern Congo and at
Kisumu on Lake Victoria. Epidemics alleged to be plague occurred at
Kisumu in Kenya throughout the 18th and 19th centuries.
McNeill (1977) has argued that focus (c) originated with the trade routes
of the Mongul Empire, the Silk Road between Syria and China across the
deserts of Central Asia. He suggested that the Mongol movements brought
the causative agent of bubonic plague, the bacterium Y. pestis, to the
rodents of the Eurasian steppes. This focus could well have been the launch
pad for any epidemics of bubonic plague in the Mediterranean region after
49
1300, but the disease had probably occurred there from the 6th century
onwards. Pollitzer (1954) considered that the rst really satisfactory evidence of bubonic plague was a pandemic beginning in AD 542 and believed
that it came from Ethiopia, suggesting a central African origin.
Twigg (1984) concluded that it was likely that bubonic plague was
present in lands around the Mediterranean in some of the epidemics in the
6th century, probably in association with other diseases whose impact was
equally severe and whose dierentiation from bubonic plague was not easy
to dene at the time. McNeill (1977) considered that it may have come from
an old focus in northeast India or in central Africa. Twigg (1984) assumed
that the black rat (see section 3.5) must have existed in the Mediterranean
ports and cities in the 6th century in order to sustain any epidemics of true
bubonic plague. The black rat had not spread to northern Europe by the
time of the plague of Justinian (Shrewsbury, 1970) and, consequently, the
disease was conned to the Mediterranean coastlands (Twigg, 1984). Bubonic plague disappeared from southern Europe at the end of the 6th
century and Twigg (1984) stated that nothing that was rmly identiable as
this disease was heard of in the Mediterranean region for the next 700
years.
Molecular biology techniques were used to resolve problems of historical etiology. DNA extracts were made from the pulp of teeth extracted
from skeletons excavated from graves near Provence, southern France,
which were dated: (i) between the 13th and late 14th centuries and so did
not necessarily come from people who died during the Black Death as is
presumed by Raoult et al. (2000); (ii) about 1590 (Drancourt et al., 1998);
and (iii) during the outbreak of bubonic plague in 1722 (Drancourt et al.,
1998), see Chapter 12. No skeleton showed macroscopic signs of disease.
The incorporation of primers specic for Y. pestis rpoB (the RNA polymerase -subunit-encoding gene) and the recognised virulence-associated
pla (the plasminogen activator-encoding gene) repeatedly yielded products
that had a nucleotide sequence similar to that of modern-day isolates of the
bacterium. The specic pla sequence was obtained from six of the 12 plague
skeleton teeth but none of the seven controls. Thus a nucleic acid-based
identication of ancient bubonic plague was achieved which conrmed the
presence of the disease during the 14th century and at the end of the 16th
century on the Mediterranean coast.
Hankin (1905) described how sporadic bubonic plague rumbled on in
India in the 19th century (see above) and it also became established in
Yum-nan, China. In 1855, troops were sent to suppress a rebellion and
plague spread further, probably as a consequence of the movement of
50
51
fatal) on 54 ships arriving in England. In two instances, plague was transferred to land and on one occasion a man died. Bubonic plague probably
existed from 1900 to 1907 in a small area of Glasgow and outbreaks were
also recorded in Liverpool, Hull, Bristol and London. These foci were all in
ports and there is no evidence of any extensive spread; Twigg (1984)
described in detail the only known occasion of authenticated bubonic
plague in a rural situation in England. In summary, the outbreak occurred
in 1910 in Suolk, 4 miles from the port of Ipswich, which received grain
boats. Three of the victims who died were in the same family and a woman
who had nursed at their cottage and who lived a quarter of a mile away was
the fourth to die. At this time, brown rats and hares were found dead in the
elds and examination showed that these were infected with the bacilli of
bubonic plague over quite a wide area. After a careful analysis of the
evidence, Twigg (1984) concluded that an epizootic had become established in the brown rats and hares and that the rst victim, a 9-year-old girl
who had recently stayed on an isolated farm, had caught bubonic plague
from an infected ea; the other three victims had died of pneumonic plague
when nursing her.
52
53
54
lack of hosts; but the local gerbils and voles are more resistant to Y. pestis
and so they can serve to maintain the enzootic in the area because they do
not die from the infection. In Kurdistan, Turkey, Iraq and Syria the hosts
are gerbils: two of them, Meriones vinogradivi and M. tristrami, are highly
susceptible, but M. persicus and M. libycus, are resistant, so that the plague
exhibits cyclic epidemics in these rodents. In central Colorado, however, an
isolated colony of prairie dogs (Cynomys gunnisoni) was wiped out by
plague: they were highly susceptible and died, although the eas remained
alive in the burrows for at least a year after the last prairie dog had died.
Thus rodents are the true natural hosts of Y. pestis; other animals
(including humans) are accidental hosts only. They inhabit a wide range of
habitats (mountains, plains, steppes, deserts, cultivated elds and forests) in
temperate and tropical areas. Rats injected subcutaneously with a small
number of virulent Y. pestis die within 28 days. Necrosis and oedema are
found at the site of inoculation and the regional lymph nodes are swollen
and embedded in haemorrhagic subcutaneous tissue; the spleen is enlarged,
the liver and lungs are hyperaemic and there is often a pleural eusion.
Small necrotic foci occur in the liver and spleen of animals surviving for a
week. Rats may be infected by a trace of infective material smeared over the
conjunctiva that results in rapid death, with enlargement of cervical lymph
nodes and spleen and haemorrhages in the stomach and duodenum. Wild
rats have buboes in their necks because eas most commonly bite them
there (Christie, 1969).
55
Fig. 3.1. The sylvatic reservoir of bubonic plague: the number of rodent species
(excluding commensal species) that have been reported to have been infected with
Yersinia. Note the absence of suitable species in western Europe. After Twigg
(1989).
epizootic plague and in some epizootics dead rodents have been collected
by the barrowful; their dead bodies are often the rst indications that
plague is present in a locality. It is only when the epizootic spreads to rats
living in urban areas in loose association with humans that the real danger
of a major outbreak of bubonic plague amongst a human population
occurs. An epizootic begins to wane when the numbers of resistant rodent
hosts rises and the numbers of susceptible rodents falls.
One of the interesting things about Yersinia pestis is the ease with which
it can accommodate to new species of rodent. During the 20th century,
following the introduction by ships to the ports, plague has spread to form
sylvatic reservoirs in North America, South America and Africa. In addition, it is present in many species in Asia and India but, despite this
readiness to spread, it has not become established in any European species
in modern times (Twigg, 1989) so that it was not possible to establish an
enzootic for bubonic plague during the age of plagues after the Black
Death (Fig. 3.1).
56
epizootic, it is via the peridomestic rats and their eas that the Yersinia is
transmitted to humans.
There are only two species of rat in Europe today, namely the brown rat,
Rattus norvegicus, and the black rat, Rattus rattus (Davis, 1986). The brown
rat is not an accomplished climber and inhabits cellars and lower oors.
This species spread from Russia in the early part of the 18th century and so
did not arrive in Britain until some 400 years after the Black Death and 100
years after the disappearance of the plague. It has since spread to most parts
of the world; being the more hardy species, it is today the common rat of
temperate latitudes and lives both indoors and outdoors. In the tropics
today it is conned to ports and towns and in the countryside lives close to
human habitation (Twigg, 1984).
The black rat (or roof rat) occurs in three colour phases, one of which is
the melanistic form; it is a descendant of a rat that probably originated in
India and spread along trading routes to establish itself as the common rat
of both town and countryside in the tropics and may have arrived in
England some time in the Middle Ages (Matheson, 1939), although a
variety of dates have been suggested (see Twigg, 1989). It is widely spread
away from seaports in the tropics when there are no indigenous rodent
competitors. In temperate latitudes, however, it is conned to buildings
because, for most of the year, the outdoor temperatures are too low for its
liking. Being a good climber it has readily gained access to ships, which
have transported it over much of the world. In buildings it climbs to the
higher levels and lives in the roofs and ceilings rather than the basement; it
rarely, if ever, inhabits burrows, tunnels in the ground or aquatic habitats.
Davis (1986) suggested that the black rat in northern France and the
British Isles may have persisted in towns and the grain ports in the Middle
Ages, but their numbers were small; the population in a particular town
disappeared after a few years but may have been re-established as a result
of new introductions. Rattus rattus may have lived in small numbers in
rural areas in the much warmer Mediterranean region of France.
The black rat, therefore, is today a native of the Mediterranean area and
generally does not persist in England without recurrent introductions
(Davis, 1986): it is found in the ports (but does not spread more than
three-quarters of a mile inland; Davis, 1986) and in those inland towns that
are connected to them by canals. Modern populations of the black rat in
England have depended for their survival upon frequent topping-up by the
importation of rats in cargo from the ports and now that canal trac has
ended and containers are replacing loose cargo, not only the inland but
also the port populations of Rattus rattus have begun to die out (Twigg,
57
1989). The dynamics of the black rat in the ports of England in the 20th
century have been discussed in detail by Matheson (1939), Twigg (1984,
1989) and Davis (1986) and, for example, the black rat was found widely in
central London in the 1950s where the multistorey, centrally heated buildings with abundant food sources from the many restaurants provided ideal
conditions (Twigg, 1989).
In summary, therefore, the black rat (Rattus rattus) was the only rodent
species present that could possibly have carried bubonic plague in the
Black Death and during the subsequent plagues in the 15th, 16th and 17th
centuries. It requires the warmth of human habitations and does not
spread far from them. The Black Death acted virtually independently of
season and climate in the British Isles (Chapter 4), whereas when bubonic
plague appears in warm countries it is circumscribed by climate and breaks
out only at certain clearly dened and predictable times (Twigg, 1989). It is
inconceivable that the black rat could have transmitted bubonic plague
rapidly and widely in winter in 1348 to 1350 in northern Britain and
northern Europe, when the Black Death was raging, or that the disease
could have been propagated over mountain passes in the Alps. This
pandemic even reached the polar regions and there was said to be an
epidemic in Greenland (Kohn, 1995).
58
59
The long survival times of uninfected eas account for their persistence
in the wild in spite of such hazards as uctuating temperature and relative
humidity. At 50 C and high humidity eas survive a long time, but they die
quickly when the temperature is over 80 C or under 50 C, especially if the
atmosphere is dry. Experimentally determined survival periods, vary from
country to country: infected eas survived for 47 days in India and for 130
days in the USA. However, although eas can enter diapause and survive
for long periods in the microclimate of deserted burrows, the external
conditions of warm temperature and high humidity are essential if the ea
is to play its part in the development of the epizootic and its escalation into
an epidemic of bubonic plague in humans.
60
Fig. 3.2. Estimated temperatures in central England, AD 900 to 1900. The age of
plagues is indicated by bold arrows. The minimum temperature for the hatching of
eas and the temperature below which conditions are deleterious to all pre-adult
stages of the ea are indicated. From Twigg (1989).
essential to promote a rat epizootic. Figure 3.3 shows the mean July
temperatures over western Europe in the 20th century and this conrms
that the British Isles does not have a climate capable of sustaining regular
seasonal outbreaks of ea-borne bubonic plague in the summer months
and certainly not in winter (Twigg, 1989). Indeed, in Europe, it is only in the
southwest area and in the Mediterranean coastal region together with the
Iberian peninsula that possibly suitable conditions are to be found (Fig.
3.3); the hinterland of Marseilles experienced an outbreak of plague that
was probably bubonic in 172022 (Chapter 12) and there were a number of
outbreaks at Barcelona through the age of plagues that may also have been
bubonic (section 11.4.2).
61
Fig. 3.3. Mean temperature (C) in July in western Europe in the 20th century. Note
that it is only in Spain and the Mediterranean coast of France that summer
temperatures are suitable for sustaining seasonal outbreaks of ea-borne bubonic
plague. After Twigg (1989).
62
rodent cycle, yet some resistance on the part of the hosts is essential for the
maintenance of a plague focus. A totally susceptible population would
soon die out, as did the prairie dogs when Yersinia arrived in Colorado. It
is essential, therefore, if the enzootic is to be maintained, that there is a
balance between susceptible and resistant hosts: voles and gerbils are
susceptible to plague but mostly do not die from it; ground-squirrels are
highly susceptible, but voles and eld mice in the same area are more
resistant and they, not the ground-squirrels, maintain the enzootic. The
infection can persist even when conditions become unfavourable for the
rodent or the ea, or both; the rodent goes into hibernation and at low
body temperature becomes more resistant, the adult ea can survive without food or host for a year and the pupal stage can be prolonged until a new
host enters the burrow and vibrations from its body stimulate emergence
from the pupa (Christie, 1969).
Bubonic plague in the wild is, therefore, a disease of arid plains, rocky
escarpments, steppes, prairies and semi-deserts where Y. pestis passes
quickly from generation to generation, causing little inconvenience to its
semi-resistant mammalian hosts and none at all to Homo sapiens unless he
or she leaves an urban or village dwelling and goes out into the focus in the
countryside.
A focus may be temporary or permanent. In the former, plague is
imported accidentally from another area, but conditions are not suitable
and the outbreak is short lived; the ea and the host are perhaps not
adapted to each other; a ea may be able to survive by feeding on the blood
of a rodent but be unable to reproduce on it. The ea is essentially a
nest-dweller and if the available host is not a burrowing animal then the
infection does not become established in the area. However, an area can
become established as a permanent focus of plague if the following factors
obtain and there is nothing unfavourable in the environment: (i) there are
many rodents in the area and some (such as gerbils) keep large, permanent
burrows; (ii) some of the rodents are partially immune but nevertheless
support a prolonged bacteraemia with Y. pestis; and (iii) eas are also
present in large numbers and they are a long-lived variety and adapted to
the host. Activity in such a focus will wax and wane and there will be
changes related to season or to the breeding pattern of rodents and eas,
and weather changes such as oods or drought will aect it (Christie, 1969).
It is evident that these factors have never been present in Europe and that
persistent enzootic bubonic plague there is an impossibility.
An enzootic of a plague focus, therefore, is not static but oscillates and,
superimposed on these dynamics, are the periodic surges of the epizootics
63
64
65
These results suggest that the hms genes are required for Y. pestis to cause
blocked eas. The failure of the mutant Y. pestis to block eas could not be
attributed to rapid elimination of this form of the bacteria, because the
same percentage of eas infected with either normal or mutant Y. pestis
strains were heavily infected after 4 weeks.
Xenopsylla cheopis were then infected with either the normal or mutant
forms of Y. pestis, both of which were tagged with a uorescent dye so that
their passage through the gut of the ea could be followed. After the rst
week, it was clear that the mutant bacteria remained in the mid-gut
whereas the normal bacteria had migrated to the foregut in many eas
which eventually became packed with bacteria. Hinnebusch et al. (1996)
suggested that the mutant bacteria may fail to colonise the foregut because,
being less cohesive, they are disrupted and ushed back into the mid-gut
during feeding. These studies are being extended in an attempt to explain
the observation that the blockage of the ea foregut breaks down at
temperatures above about 27 C. Do such temperatures suppress the products of hms or other genes? Again, these studies are of considerable interest
and contribute to our understanding of bubonic plague, a major historical
and a minor present-day scourge but, we suggest, are not relevant to the
biology of haemorrhagic plague.
66
(ii) When humans invade a natural focus, for example as a hunter, bubonic plague is caught directly from the wild rodent reservoir, usually
on a small scale from trapping, skinning or eating a small animal,
although 60 000 hunters in total caught plague from marmots which
they hunted for their skins between 1910 and 1911 in Manchuria (Wu
et al., 1936; Chandler & Read, 1961).
(iii) Occasionally humans get plague from eating a domestic animal (a goat
or camel) that has roved over a natural plague focus.
(iv) However, and this is a further important point, it is movement into
new, untrodden regions, as in war or mass migration that brings
humans most often into direct contact with enzootic bubonic plague
(Christie, 1969).
67
bloody) and contacts inhale it and get pneumonic plague. From then on,
one patient infects another with pneumonic plague by direct inhalation of
the bacillus without the intervention of a ea: the onset is abrupt and severe
with rapid prostration; breathing is shallow, distressed and very rapid;
sputum is watery, teeming with yersiniae and soon mixed with blood. The
face is dusky, the temperature high, and the patient may bleed. There is
always pulmonary oedema and often a pleural eusion, but the symptoms
are those of respiratory distress and shock. The patient dies about the third,
never later than the sixth, day and without modern medical treatment,
pneumonic plague is invariably fatal. This distinction from the mortality in
untreated patients suering from bubonic plague is of importance when we
attempt to determine the possible role of Y. pestis in the plagues in earlier
centuries.
The bubo is the characteristic symptom of bubonic plague; it is a lump
formed by a swollen lymph node and is of variable size. It is found most
commonly in the groin but its location depends on where the ea bites
which, in turn, depends on how the victim is clothed. Christie (1969)
summarised this as follows:
An Indonesian peasant . . . wears only pants and a hat: the ea can bite him
anywhere, with least eort on the legs. A Libyan farmer has boots and breeches and
ouncing robes, and your ea needs all its wit to get to his skin: the arm or the neck
may be easier than the leg. When a patient gets plague from skinning some animal
the infection will be through his hands and the bubo in his axilla: if he eats the esh,
Y. pestis may settle on his tonsils and the bubo be in his neck.
The bubo appears early in the illness, on the rst or second day; in the
septicaemic type there may be no bubo at all or it may be so small as not to
be noticed and have no time to enlarge before the patient dies. Usually, the
bubo is very painful and tender and, in patients who live long enough or
survive, it breaks down and discharges pus.
There is wide variation in the onset and course of bubonic plague which
may be mild enough to be overlooked (termed pestis minor) or it may be
overwhelming. The incubation period is typically 26 days after exposure,
i.e. very short when compared with haemorraghic plague (see section 5.5).
It is important to record accurately the details of the course of the disease
so that we may compare these with such accounts as we have of the plagues
of earlier centuries. Once again, we can rely on Christie (1969), who wrote
from a wealth of personal experience of bubonic plague as follows:
Typically the onset is sudden with chills and rigors and rise of temperature to
102 F or 103 F (38.8 C to 39.4 C). The patient has a severe, splitting headache
68
and often pains in the limbs, the back and the abdomen. He may curl away from the
light or, as the painful bubo develops, take up some attitude in bed that relieves the
pressure on the painful swelling. He becomes confused, restless, irritable or apathetic, his speech slurred as if drunken, he is unable to sleep, sometimes wild or
maniacal. He may vomit. He is usually constipated, but diarrhoea can be an
ominous symptom. His eyes are suused, occasionally blood-shot, and, as the
disease advances, he may bleed into his skin, or internally into his stomach or
intestine, or from his kidney . . . Rarely he is jaundiced. Within a day or two he is
prostrate with all the symptoms of shock. His temperature may come down and he
appears better on the third day or so, but this is deceiving: he is worse the next day
and dead soon after. Most patients died between the third and sixth day: if they are
alive on the seventh day they may struggle through to recovery. In the last stages
the patient may have a cough and other signs of respiratory embarrassment, but
mostly they die without obvious signs or symptoms of pneumonia.
69
cases in the Great Pestilence of 134850 and that, in many places, the
plague rst appeared in its pneumonic form. He added that the Manchurian epidemics of 191011 and 192021 were exclusively pneumonic. These
references to plague in Manchuria in the 20th century come from the work
of Wu (1926; Wu et al., 1936), who armed that an intimate relationship
exists between rodent plague and human bubonic aections are [sic] due,
in an overwhelming majority of instances, to transmission of the virus [sic]
from the rodents through their eas. Pneumonic plagues almost always
originate in human cases with secondary lung involvement: evidence tends
to conrm experiences in areas like South-East Russia, Transbaikalia
and Manchuria that usually primary pneumonic plague is traceable to
bubonic cases with well marked secondary lung involvement (Wu et al.,
1936).
Pneumonic plague was largely absent from the southern Chinese provinces and was usually less than 4% of the total cases. It was more conspicuous in the northern provinces (12%), although Wu believes that this is
principally because of seasonal inuences, with low temperatures at the
time of the outbreaks bringing the patients into close contact and thereby
increasing the chances of respiratory infection from any secondary lung
infection. Nevertheless, in addition to the Manchurian pneumonic epidemic of 191011, repeated plague outbreaks, often pneumonic in character, were reported from the Narinsk district to the southwest of Issyk-kul
Lake and the Prjevalsk (Karakol) district to the east of the same lake but
Wu et al. (1936) say These are evidently also due to epizootics among
tarabagan-like marmots [our italics]. Finally, a pneumonic outbreak took
place in 192930 in the Alma-Atinsk (Verni) district situated in the northeast of Issyk-kul Lake, said to have been due to an epizootic among the
local hares [our italics].
According to Wu, many authors take for granted that bubonic cases
with secondary lung involvement are the source of pneumonic outbreaks
that occur if meteorological and social factors are favourable, especially
cold weather, which creates unhygienic conditions in tightly shut and
overcrowded houses. Most infections in the Manchurian epidemics occurred indoors, especially at night-time, when the workers returned to their
comparatively warm but crowded shelters to rest and sleep. But Wu
pointed out that pneumonic plague epidemics also occur in summer and in
countries with a warm climate; for example, the high incidence of lung
pestilence in Upper Egypt during the hot and dry plague season.
Where an outbreak of bubonic plague has settled into the pneumonic
form, transmission of Yersinia will be largely person-to-person and the
70
epidemic will probably follow Reed and Frost dynamics (section 2.5).
There is general agreement that the time from infection to death is short,
probably about 5 days, so that the infectious period is even shorter.
Consequently, the Reed and Frost equations predict that the outbreak will
be short-lived (see Fig. 2.4) unless it is restarted from the rats and their eas.
This is quite unlike the dynamics of haemorrhagic plague, which has a very
long incubation period and the epidemics are consequently of extended
duration. We conclude that pneumonic infection probably markedly exacerbated the mortality of many outbreaks of bubonic plague, as in Marseilles in 172022 (Chapter 12), but its main eect was probably within the
household and family and such neighbours that came to visit. It is impossible that a mortally sick person who was rapidly prostrated when infected
with the pneumonic form and who was only 3 days away from death could
have spread the disease over long distances either by land or sea as
occurred during the Black Death and in many of the plague outbreaks
thereafter.
3.14 Pathology
When Y. pestis is injected into humans by a ea most of the bacteria are
phagocytosed and killed by the polymorphonuclear leucocytes, which
enter the infection site in large numbers. However, a few bacilli are taken
up by tissue macrophages that are unable to kill them but provide a
protected environment for the organisms to resynthesize their capsular and
other virulence antigens. The re-encapsulated organisms kill the macrophage and are released into the extracellular environment, where they
resist phagocytosis by the polymorphs. The resulting infection spreads
quickly to the draining lymph nodes, which become hot, swollen and
tender, with haemorrhagic necrosis, and there is usually a gelatinous
oedema in the surrounding tissue giving rise to the bubo. The infection
spreads through the lymphatic vessels and invades the blood stream to
cause lesions in the spleen, liver, kidney and other organs of the body. In
pneumonic plague, the lymphatics of the lung are rapidly invaded by the
baccilli and the condition at autopsy is one of haemorrhagic pneumonia.
Culture of almost any organ will be positive for Y. pestis.
An autopsy of a seaman who died of authentic bubonic plague in 1900
was reported by Savage & Fitzgerald (1900) and the important points are
given below. This autopsy report can be directly compared with the few
accounts available of the examination of cadavers who died from haemorrhagic plague (see Chapter 8).
3.14 Pathology
71
72
Fig. 3.4. Cell-mediated protection against Yersinia pestis in unsensitised (top) and
sensitised (bottom) T-lymphocytes.
immune mechanism against this disease is complex and involves a combination of humoral and cellular factors. The convalescent host is solidly
immune (at least for a time) to virulent rechallenge, the inoculum being
eliminated as though the organisms were completely avirulent. Killed Y.
pestis vaccines, especially when given with a suitable adjuvant, induce some
measure of host protection, although this will be less eective than that
aorded by the live infection.
73
Fig. 3.5. An epidemic of bubonic plague at Bombay, India, from October 1905 to
September 1906. (a) Plague-infected Rattus decumanus (epizootic). (b) Plagueinfected Rattus rattus (peridomestic). (c) Human plague deaths. All results expressed as a percentage above and below the mean. The sequence of the infections
can be seen clearly. From the Plague Commission in India (1907a).
74
typically a wandering rat, but does occur in the lower oors of inhabited
buildings and is conned to Bombay City; it does not occur in the outlying
villages of the island because of the absence of gullies and drains there.
The report concluded that the persistence of the plague was associated
mainly with the R. decumanus epizootic and that this species was directly
responsible for causing the epizootic in R. rattus. The epidemic of bubonic
plague in humans, in turn, was directly attributable to the R. rattus
epizootic.
We conclude that this study illustrates well the dynamics of a bubonic
plague epidemic in humans. As always, it was completely dependent on a
pre-established epizootic in local rodents, and the maintenance of the
enzootic during non-epidemic periods was critically dependent on a balance between susceptible and resistant rats and their alternative rodent
vectors.
75
Fig. 3.6. Bubonic plague mortality in India per 100 000 population, 18961939.
From Seal (1969).
is possible that the endemic foci in the Himalayas were of long standing,
plague infection being known in the Kumaon and Gharwal districts since
1823. The infection might have persisted there as a relic of a great pestilence
of the 17th century and might have been responsible for occasional plague
outbreaks until 1877. Thereafter, it remained latent for some time, to
become active again early in the 20th century. In central India and Madhya
Pradesh, plague started after the infection of Bombay in 1896.
Plague is essentially bubonic in India; true septicaemic plague is rare,
although some bubonic types have septicaemic manifestations. Primary
pneumonic plague is also rare, and this is an important point; it generally
occurs after lung involvement in a bubonic-septicaemic case leading to
plague pneumonia, and subsequent contacts of such cases may develop
primary pneumonic plague. Such outbreaks have been known to occur in
India (Seal, 1969) but they have generally remained conned to a single
family or a few families only. The incidence of pneumonic plague in India is
generally below 1% and has never exceeded 3% in any year since 1895.
Plague is both urban and rural in India, the latter predominating. It
appears that plague has failed to gain a foothold in many of the towns in
India, perhaps because of unsuitable climatic conditions and the lack of
any ecient ea vector (as in Madras and Assam). Regular heavy annual
oods may also be responsible for keeping certain states (e.g. Bengal) free
76
Fig. 3.7. Endemic plague foci in India. Arrows indicate the directions of radiation of
the plague. After Sharif (1951).
from plague. Another factor which may play an important part is the
distribution and proportion of various types of rodent. Given a suitable
ea vector, a large proportion of R. rattus will make for easier and quicker
spread of plague among humans than a similar proportion of other rodents. On the other hand, replacement of one rodent by others, as in
Bombay, may disturb the balance and plague may recrudesce or be imported with consequent severe outbreaks. However, given a suitable climate, a suciently large rodent population, eective vectors and the
plague bacillus, Y. pestis, the infection may often become rmly entrenched
among the rats of towns and persist there for many years (Seal, 1969).
The optimal conditions for bubonic plague in India are 2025 C and a
77
78
Population
Death rate
per 1000
806 144
161 696
97 009
61 564
4431
2068
1692
20.1
31.2
24.1
35.0
104.9
102.5
360.5
of the community, with the maximum mortality in the villages rather than
towns (Hankin, 1905); this is clearly shown in Table 3.1. Furthermore,
although the plague spread readily from village to village, Hankin observed that it did not appear to be carried great distances in the epidemic
form. This is in complete contrast to the behaviour of haemorrhagic plague
in Europe (see section 13.9).
3.16 Conclusions
79
Fig. 3.9. The biology and transmission of Yersinia pestis, illustrating the central role
of the ea in maintaining the epizooticenzootic dynamics.
19056, when all the conditions are suitable, a major epidemic in humans
can erupt in a city.
The mortality from bubonic plague in India is impressive, with 12
million people dying, but it must be remembered that this occurred over 60
years over a vast area of the subcontinent with an enormous indigenous
population.
The biology of Y. pestis is summarised in Fig. 3.9; it illustrates the central
role of the ea in maintaining the epizooticenzootic dynamics by transmitting the bacterium between the dierent rodent hosts and to humans
and it emphasises the point that the disease does not spread to humans
unless it is already pre-established in the local rodent population. This is
another critical point: Y. pestis could not have been the causative agent
during the Black Death, which spread rapidly from the shores of the
Mediterranean almost to the Arctic, unless there was an indigenous
European population of rodents in which bubonic plague was already well
established. Figure 3.9 may be compared with the simple dynamics of a
normal infectious disease shown in Fig. 2.3A in which the epidemics are
short lasting and can be described by the Reed and Frost equations (see
section 2.5). Bubonic plague in humans, with its more complex biology,
does not follow Reed and Frost dynamics.
Endemic bubonic plague did not spread rapidly in the 19th century, but
80
the development of much more rapid sea travel via steamships allowed the
dispersion of infected rats to the warmer parts of the globe and, where
conditions were suitable, Y. pestis infected the local rodents. Dispersal was
assisted by the development of the railways but, even so, the spread of
bubonic plague was slow and, although it was eventually quite widespread
and grumbled on in the local rodents, there were relatively few human
deaths in South Africa and America. Again, bubonic plague did not follow
Reed and Frost dynamics in these newly emerged infestations.
The spread of bubonic plague is critically dependent on climatic conditions, particularly on temperature, which is the major environmental factor
governing the biology of both the rodents and the ea. It is for this reason
that endemic bubonic plague does not spread away from warmer regions
as, for example, the Mediterranean coast and southern Europe (see Fig.
3.3). Outbreaks in ports in more northern latitudes did not spread far and
had to be reinforced by fresh introductions of rats and eas.
During the 300-year period from the Black Death to the Great Plague of
London, the only species of rat in Britain that could have carried bubonic
plague was the black rat, an animal of warm buildings that was conned
largely to ports. Steamships in the 20th century have brought bubonic
plague to England via infected seamen and rats on several occasions and
the disease has sometimes been transmitted to the black rats of the ports
but plague has never spread and there have been few fatalities. It is
impossible that bubonic plague could have been established as an epizootic
and have spread rapidly and widely inland, sometimes in winter months,
via the agency of black rats and their eas.
Many writers, aware that the facts concerning the rapid spread of
plagues in historical times were clearly at variance with the etiology of
bubonic plague, have fallen back on invoking interhuman transmission via
pneumonic plague (see section 3.13). Christie (1969) showed clearly that
this was not so: bubonic plague always begins with an infection in rodents
and only a small percentage of the cases in humans develop the pneumonic
form in the terminal stages of the disease. Although this would invariably
have been lethal and have exacerbated the spread of the infection and the
death toll within the household, such grievously ill patients would have
been unlikely to be able to move any distance to spread plague to other
villages.
4
The Great Pestilence
The pandemic that struck Europe in 1347 was probably the most serious
outbreak of a lethal infectious disease in history in terms of the percentage
of the population that died and the speed with which it spread. This
outburst involved the whole of continental Europe, the Channel Islands,
the British Isles, Iceland and Greenland (Shrewsbury, 1970; Kohn, 1995).
This pandemic was called the Great Mortality or the Great Pestilence by
contemporary writers. It was named the Black Death in English historical
writings in 1823 and was introduced into medical literature as such in 1833
because of the black blotches, caused by subcutaneous haemorrhages, that
appeared on the skin of the diseased humans near the time of death (Kohn,
1995). Shrewsbury (1970) stated that the latter name was inappropriate
because, although the cadaver of a victim of the plague may exhibit a
purplish discoloration, the corpse did not turn black. However, since the
pseudonym The Black Death is so rmly entrenched in historical writing
we shall continue to use it.
Although contemporaneous written records are scattered and scarce, the
Black Death has been discussed exhaustively (see Creighton, 1894; Nohl,
1926; Hirshleifer, 1966; Deaux, 1969; Ziegler, 1969; Shrewsbury, 1970;
Gottfried, 1983; Twigg, 1984; Carmichael, 1986, 1997; Horrox, 1994; Kohn,
1995; Ormrod & Lindley, 1996), covering, in particular its origins, the
pattern of spread, the number of deaths from the disease and the role of the
rat as a carrier. These authors almost all assume without questioning that
the Black Death was an epidemic of bubonic plague. Many bizarre sideissues are covered in these writings: (i) the religious response and the
attitude of the Church, where the plague was considered as Divine punishment (Nohl, 1926; Horrox, 1994); (ii) pseudo-scientic contemporaneous
explanations of the plague, such as astrological causes and the dangers of
corrupted air and earthquakes (Horrox, 1994); (iii) the persecution of the
81
82
Jews, who were thought either to have deliberately spread the plague or to
have polluted society and brought on Gods vengeance (Nohl, 1926; Horrox, 1994); (iv) the erotic (Nohl, 1926) and diabolic elements in the plague,
including devil worship following the moral collapse of the Church (Nohl,
1926); (v) the eect of the Black Death on the Church and religion (Bolton,
1996); (vi) the politics of pestilence government in England after the Black
Death (Ormrod, 1996); (viii) the Black Death and English art (Lindley,
1996).
4.1 Arrival of the Black Death in Europe
Ziegler (1969) concluded that the Great Pestilence began in central Asia
and Twigg (1984) recorded that the Russian archaeologist Chwolson had
shown unusually high death rates in 1338 and 1339 near Lake Issyk-Koul,
Semiriechinsk in central Asia. He continued
Stewart (1928) in his account of the spread of Nestorian Christianity has drawn
attention to two old cemeteries, fty-ve kilometres [34 miles] apart, which contained tombstones indicating they belonged to Nestorian Christians in
Semiriechinsk. One of these graveyards contained 611 stones and for the years A.D.
13389 three inscriptions stated that the persons buried had died of plague. He
further pointed out that during those two years the number of inscriptions was
exceptionally large. He also was of the opinion that plague originated in eastern
Asia at about that time and that it spread rapidly to Asia Minor, North Africa and
Europe and reached the Crimea in A.D. 1346. Pollitzer (1954) has also referred to
this large number of burials in the Nestorian graveyards and says quite categorically that it is therefore certain that plague was in evidence in central Asia a few
years before the infection of the Crimean ports in 1346 and that from there it was
carried by ship to Europe.
Vernadsky (1953) stated that 85 000 people died in the Crimea and the
plague was probably carried thence to Constantinople, where the epidemic
raged in 1347. The Black Death was then taken to the sea-coasts and the
eastern Mediterranean, including Greece and Egypt, became infected.
Thus, although the introduction of the epidemic in Europe is said to be
from the Crimea, via the well-established sea trade routes, Twigg (1984)
suggested that there may have been parallel introductions from Syria, since
ships plied from there to the Italian and French ports. All the accounts
describe the plague spreading rapidly almost as soon as docking had taken
place.
4.2 The plague in Sicily
Nohl (1926) provided a translation of the account given by Michael of
Piazza, a Franciscan friar, of the events that followed the arrival of the
83
plague in Sicily. He wrote his history some 10 years later (for notes see
p. 85).
At the beginning of October . . . 1347, twelve Genoese galleys . . . entered the
harbour of Messina. In their bones they bore so virulent a disease that anyone who
only spoke to them was seized by a mortal illness and in no manner could evade
death. The infection spread to everyone who had any intercourse with the diseased
[note 1]. Those infected felt themselves penetrated by a pain throughout their whole
bodies and, so to say, undermined. Then there developed on their thighs or on their
upper arms a boil about the size of a lentil which the people called burn boil
(antrachi) [note 3]. This infected the whole body, and penetrated it so that the
patient violently vomited blood [note 5]. This vomiting of blood continued without
intermission for three days, there being no means of healing it, and then the patient
expired [note 6]. But not only all those who had intercourse with them died, but
also those who had touched or used any of their things [note 1]. When the
inhabitants of Messina discovered that this sudden death emanated from the
Genoese ships they hurriedly expulsed them from their harbour and town. But the
evil remained with them and caused a fearful outbreak of death. Soon men hated
each other so much that, if a son was attacked by the disease, his father would not
tend him. If, in spite of all, he dared to approach him, he was immediately infected
and could by no means escape death, but was bound to expire within three days
[note 6]. Nor was this all: all those belonging to him, dwelling in the same house
with him, even the cats and other domestic animals [note 2], followed him in death.
As the number of deaths increased in Messina many desired to confess their sins to
the priests and to draw up their last will and testament. But ecclesiastics, lawyers
and attorneys refused to enter the houses of the diseased. But if one or the other had
set foot in such a house to draw up a will or for any other purpose, he was
hopelessly abandoned to sudden death [notes 1 and 6]. Minor friars and Dominicans and members of other orders who heard the confessions of the dying were
themselves immediately overcome by death, so that some even remained in the
rooms of the dying [note 6]. Soon the corpses were lying forsaken in the houses. No
ecclesiastic, no son, no father and no relation dared to enter, but they paid hired
servants with high wages to bury the dead. But the houses of the deceased remained
open with all their valuables, with gold and jewels; anyone who chose to enter met
with no impediment, for the plague raged with such vehemence that soon there was
a shortage of servants and nally none at all. When the catastrophe had reached its
climax the Messinians resolved to emigrate. One portion of them settled in the
vineyards and elds, but a larger portion sought refuge in the town of Catania . . .
But the plague raged with greater vehemence than before. Flight was no longer of
avail . . . Many of the eeing fell down by the roadside and dragged themselves into
the elds and bushes to expire. Those who reached Catania breathed their last in
the hospitals there. The terried citizens demanded from the Patriarch prohibition
on pain of ecclesiastical ban, of burying fugitives from Messina within the town, and
so they were all thrown into deep trenches outside the walls. The population of
Catania was so godless and timid that no one among them would have intercourse
or speak to the fugitives, but each hastily ed on their approach . . . Thus the people
of Messina dispersed over the whole island of Sicily and came also to Syracuse and
84
with them the disease, so that in Syracuse innumerable people died. Sciacca,
Trapani, Girgenit and Messane, thus, were also infected by the plague, but particularly Trapani, which was completely depopulated [note 1]. The town of Catania
lost all its inhabitants [note 1] so that it ultimately sank into complete oblivion.
Here not only the burn blisters [note 3] appeared, but there developed in dierent
parts of the body gland boils in some on the sexual organs, in others on the thighs,
in others on the arms, and in others on the neck. At rst there were of the size of a
hazel-nut [note 3] and developed accompanied by violent shivering ts, which soon
rendered those attacked so weak that they could no longer stand upright, but were
forced to lie in their beds consumed by violent fever [note 4] and overcome by great
tribulation. Soon the boils grew to the size of a walnut, then to that of a hens egg or
a gooses egg, and they were exceedingly painful, and irritated the body [note 3],
causing it to vomit blood by vitiating the juices [note 5]. The blood rose from the
aected lungs to the throat, producing on the whole body a putrefying and
ultimately decomposing eect [note 5]. The sickness lasted three days, and on the
fourth, at the latest, the patient succumbed [note 6]. As soon as anyone in Catania
was seized with headache and shivering [note 4], he knew that he was bound to
pass away within the specic time [note 1] . . . But the pestilence raged from
October 1347 to April 1348 [note 7].
85
86
87
88
Fig. 4.1. The spread of the Black Death across Europe. Areas denoted by dots
escaped the plague. After Twigg (1984), Zeigler (1969) and McNeill (1977).
89
Fig. 4.2. Institutions to vacant beneces in 11 dioceses during the Black Death,
illustrating that most epidemics were long-lasting and followed Reed and Frost
dynamics. (A) Salisbury. (B) Bath & Wells. (C) Winchester. (D) Exeter. (E) Gloucester (lled area) and Worcester (open area). (F) Hereford. (G) Licheld. (H) Norwich.
(I) Lincoln. (J) Ely. Data from Shrewsbury (1970).
Shrewsbury (1970) suggested that the plague probably came to Melcombe Regis either from Calais, which was then an English possession, or
from the Channel Islands, which were in constant communication with
90
England and were severely visited by the plague in 1348 prior to its
appearance in England. We do not know whether this was the only
introduction of the pestilence into England or whether there were additional arrivals of infectives at the ports (perhaps on the east coast) during
the next 12 months.
Its arrival in 1348 is variously dated as June, July or early August; it
spread from Dorset to Bristol and thence, by way of Oxford, to London,
which it reached at the end of October or the beginning of November 1348.
It continued through the winter but the main force of the epidemic did not
begin until the spring of 1349 and Fig. 4.2 shows that in each diocese the
epidemic followed Reed and Frost dynamics (section 2.5) and persisted for
some 7 months. There is wide variation in the estimates that have been
made of the total number dying from the pestilence in London, some even
suggesting a total of 100 000 (Kohn, 1995). Ziegler (1969) concluded that a
gure between 20 000 and 30 000 deaths from a population of 60 000 to
70 000 would be a good estimate, a percentage mortality that was probably
in line with that suered in other cities.
Ziegler (1969) wrote that, although there were few impressionistic reporters and dispassionate medical records of the Black Death in England, a
number of archives contribute to the provision of the fuller picture of the
progress of the Black Death through the metapopulation than in any other
country. The most complete source available is the ecclesiastical records,
and the epidemiology of the Black Death has been traced by studying the
vacancies among the beneced clergy at that time, a practice that Shrewsbury (1970) did not think was completely satisfactory, mainly because his
ndings were not consistent with the spread of bubonic plague. Among the
lay documents, Ziegler (1969) has also studied the manorial Court Rolls
and the Account Rolls; taken together they provide a picture of life on the
medieval manor and show the incidence of the pestilence in each. However,
many fewer of them remain than is the case with ecclesiastical documents.
It was reported that the summer of 1348 was exceptionally wet, which
the populace believed to be the cause of the pestilence, although Shrewsbury pointed out that both Rattus rattus and Xenopsylla cheopis dislike
damp conditions. It is said that the Black Death always struck hardest at
seaports and coastal districts, and that stretches of marshland and fen
acted as barriers against the spread of the epidemic. Sparsely populated
hilly districts were usually only slightly aected, although villages close to
communication routes were often comparatively severely hit, all indicative
of person-to-person transmission of an infectious disease spread by travellers. According to a contemporary observation, the common people bore
91
37
37
52
33
56
58
51
47
49
48
35
Huntingdonshire
Leicestershire
Lincolnshire
Northamptonshire
Nottinghamshire
Oxfordshire
Shropshire
Somerset
Staordshire
Surrey
Warwickshire
Worcestershire
34
36
51
37
36
34
43
47
34
56
36
48
the main toll and among them it fell most heavily on the young and
vigorous (Shrewsbury, 1970).
92
epidemic began in April 1349 and was virtually over by October 1349, a
duration of 7 months. It moved from south to north and was at its peak in
Coventry in May and June, in Derby in July and August, in Shropshire in
August and reached Cheshire during September. The Great Pestilence is
believed to have begun in Derbyshire in May 1349 and Cox (1910), following an inspection of the episcopal registers, stated that the county was
pre-eminent in desolation and lost two-thirds of its beneced clergy within
12 months. He declared that in Derbyshire no class of the community
seemed to have been spared, and no place was too remote or healthily
situated to escape desolation. Lunn (1937) armed that the average plague
mortality rate of this countys beneced clergy was almost twice as high as
that of Staordshire.
In Cheshire, Middlewich was a chapelry of Sandbach, and six miles
distant from the mother church. Encumbered by the necessity of burying
their dead at Sandbach, the parishioners of Goostrey experienced serious
inconvenience during the excessive mortality of 1349. Corpses of plague
victims were left to rot at the roadside . . . so great were the perils and
hazards of the way . . ..
But elsewhere the great woodlands, for which Cheshire in the Middle
Ages was famous, impeded the progress of the plague. There are several
instances of reduced rents and of lands lying idle through lack of tenants at
Middlewich, Russheton, Chelmondeston, and Kingsley in 1350, all attributed to The Great Pestilence in the manorial records (Shrewsbury,
1970).
In 1939, R. Sharpe France presented a manuscript on the history of the
plague in Lancashire, which he had collated from many sources. The
following account is based on his paper but there is one caveat that should
be borne in mind: Sharpe France was convinced that the pandemic of
134850 was caused by bubonic plague and it is not always clear which of
his statements are based on authenticated records from Lancashire and
which have been adapted from accounts elsewhere. The pestilence in
Lancashire lasted from 8 September 1349, until 11 January 1350 and the
mortality was almost unbelievably high. The disease was swift in its
action, one day people were in high health, and the next day dead and
buried. Sometimes death occurred within 12 hours of a person being
infected, and usually within 3 days at the most. Apparently few of the upper
class died, but of the common people and the clergy a multitude known to
God only. In the aected classes it was generally the youthful and healthy
who were carried o, the old and ailing usually being spared. In Preston,
3000 people died; in Kirkham, 3000 people fell victim to the pestilence; 800
93
There is also evidence that Carlisle was involved in the pestilence because a royal grant of various liberties was made to its citizens in February
1352 in recognition of its importance as a frontier fortress against the Scots
and because it was then wasted and more than usually depressed by the
mortal pestilence lately prevalent in those parts as well as by the frequent
attacks. The conclusion that the Black Death struck severely at Carlisle is
supported by the accounts of Richard de Denton, which he presented in
1354, claiming that, because of the mortality from the pestilence lately
raging in those parts, the greater parts of the manor lands attached to the
Kings Castle at Carlisle were still lying uncultivated. For 18 months after
94
the end of the plague the entire estate had been allowed to go to waste for
lack of labourers and tenants. Mills, shing, pastures and meadow lands
could not be let during that time for want of tenants willing to take the
farms of those who had died in the plague. The jury found that Richard de
Denton had proved his facts and accepted the greatly reduced value of the
estate.
The mortality in Yorkshire was not as severe as in neighbouring Lincolnshire; nevertheless 223 beneces of the 535 parishes in the diocese of
York were vacated by death and it is believed that between 42% and 45%
of the parish clergy died. The archdeaconry of York covers a large area and
shows the usual wide variation in mortality. The deanery of Doncaster lost
59% of its clergy, whereas virtually no beneces were vacated between
Doncaster and the Humber. In Pontefract the gure was 40%, but with few
casualties in the eastern ats. The city of York, surprisingly, with 32%, was
relatively lightly aected. Sellers (1913) quoted a report by Clarkson that
Richmond was so grievously ravaged by a plague at this time that about
2000 of its inhabitants died. However, this report is not supported by any
evidence and it is almost certainly exaggerated because it is improbable
that Richmond had a population of more than 1000 persons of all ages at
any time in the 14th century (Shrewsbury, 1970).
Thompson (1914) concluded that, in the archdeaconries of Nottingham
and York, the two extremes, mountainous districts and marshland, were
comparatively immune from pestilence, while normal agricultural country
95
and the lower highlands suered most heavily. This is borne out by the
gures of the percentage mortality of the beneced clergy for the three
deaneries of Cleveland:
Bulmer
Ryedale
Cleveland
51%
28%
21%
Events at the Abbey of Meaux, 6 miles north of Hull, in August 1349 give
valuable details of the infectivity of the Black Death. Gods providence
ordained at that time, wrote the chronicler of the monastery, that in many
places the chaplains were kept alive to the very end of the pestilence in
order to bury the dead; but after this burial of the lay folk the chaplains
themselves were devoured by the plague, as the others had been before
them. The abbot and ve monks died in a single day, on 12 August 1349;
the Prior, the Cellarer, the Bursar and 17 other monks died also and, out of
50 monks and lay brethren, the chronicler claims that only 10 survived. The
infection clearly spread rapidly and the six synchronous deaths on 12
August shows that they were multiple infections from an original source.
The 80% mortality testies to the infectivity and lethality of the pestilence.
However, 10 persons in this closed community survived; most of them must
surely have been exposed to the infection and hence either recovered or
were resistant.
Bradshaw (1907) suggested that the Great Pestilence moved north from
Norfolk during the spring and early summer of 1349, with the result that
the peasants living along the Salters Track in County Durham were the
rst people in the diocese to experience it. He suggested that it may have
reached Sunderland before the middle of July, because by that time four of
the bishops tenants at Wearmouth were dead. He noted that the business
at the summer halmote (a court) at Chester-le-Street on 14 July proceeded
normally; but at the halmote at Houghton-le-Spring on 15 July the peasants were in a state of panic, and a similar panic was manifest at Easington
on the following day. There was no alarm at Middleham halmote on 17
July, however, possibly because this village lay o the main road, and the
halmotes of Stockton on the 17th, Sadberge and Darlington on the 20th,
Wolsingham on the 21st or 22nd, and Lanchester on the 23 July, all seem to
have been conducted normally. Shrewsbury (1970) quoted Bradshaw as
follows: He deduces from the contemporary records that plague erupted as
an epidemic disease at rst in the south-eastern part of the county, and he
believes that it raged with especial virulence at Billingham, where 48
tenants of the Prior died of it. Although Billingham was not a large village
96
97
He concluded that the Black Death was a stimulus towards the greater
mobility of labour (and hence towards the disintegration of the manorial
system) in spite of legislation designed to prevent it and that, in general, the
countryside was soon back to near normal conditions, although some of
the poorer villages were unable to resist the eects of the pestilence and the
temptations oered by richer neighbours. Some villages never recovered
fully, but such cases were rare in the Midlands, where a boom-town like
Leicester, which was at the centre of a prosperous agricultural area with
rapidly growing trades and industry, could quickly make up its strength by
recruiting not only peasants but free men.
The foregoing describes events and social changes that followed the
Black Death in southern and central England; no such comparable information exists for northern England but it is interesting to compare these
events with the social upheaval that occurred at Penrith in Cumbria after
98
the major mortality of the plague that devastated the town 350 years later,
as described in section 13.17. Here again, the ecological niches were quickly
lled and the community returned to pre-crisis population dynamics.
99
We have presented evidence above that shows how the Great Pestilence
did indeed spread through these northern counties. Shrewsbury continued:
In the fourteenth century England was a much wilder country than it is now. There
were large areas of uninhabited forest, waste, and fenland, and the moorlands of
Yorkshire, Somerset, Devon, and Wales were devoid of human and rat life, except
for a sprinkling of hamlets on their margins. Everywhere the population was more
thinly scattered than it is today, though it is important to remember that nearly all
the habitations were grouped in villages [and] there were next to no outlying farms.
This fact is essential for a correct understanding of the epidemiology of The Great
Pestilence in England. Outlying farms, with their relatively large colonies of
house-rats, would have served as links in the train of transmission of rat-plague
from one village concentration of house-rats to another. In the absence of such links
many probably very many villages must have completely escaped the disease.
From what is now known about the etiology of bubonic plague it was therefore a
biological impossibility for the whole or even a major part of England to have been
ravaged by The Great Pestilence of 134850.
Surely, the lack of this correction factor for non-plague deaths would have
only a small eect on Lunns statistics and Shrewsbury was attempting
here to modify the results of his own and other writers analyses to t with
his understanding of bubonic plague.
With reference to the plague in the Diocese of Lincoln, Shrewsbury said:
Thompson concludes his study with the statement that the percentages of vacant
beneces in each rural deanery work out with such unanimity that the mortality of
the beneced clergy . . . may be taken as a fair guide to the general death-rate.
Unfortunately the biological nature of bubonic plague contradicts his conclusion,
because if there is one characteristic of the disease that is absolutely constant it is
100
the erratic manner of its spread and consequently its haphazard attack in every
community. It is in fact the unanimity of these clerical statistics which precludes
their use as a guide to the national death-rate from bubonic plague; but their
unanimity might represent the activity in England of another lethal disease during
the normally cold months from December to March when the rat-eas, X. cheopis
and C. fasciatus, hibernate in temperate climates. The characteristics of this other
manifestation of The Great Pestilence for such it would have been to contemporary recorders must have been the similarity of its clinical picture to that of
bubonic plague: its winter incidence; its high infectivity; its transmission by direct
human contacts; its preference for the well-fed and, especially with regard to the
death-rate among the clergy, its high mortality in the middle-aged and elderly.
Classical epidemic typhus fever completely fulls these requirements.
Once again, he was dismissing evidence because it does not t with the
epidemiology of bubonic plague. He was swayed by the erratic spread of
this disease and appeared to be presenting evidence for two separate
causative agents operating simultaneously in the Great Pestilence, surely
an unlikely scenario.
When discussing the plague epidemics in Hull in the 1370s Shrewsbury
quoted Sellers as writing: But there is one marked dierence between these
attacks of plague of the second half of the century and the Black Death.
They were localised in towns, escape by ight was possible to the wealthy,
but in 1349 death was in the soil, town and country suered alike, the
fortunes of the wealthy and poor were equalised. Shrewsbury continued If
such a marked dierence had existed, then the Black Death was
certainly not an eruption of plague; but the statement unfortunately betrays only a sad ignorance of the immutable etiology of that disease, which
can no more change its nature than the Ethiopian his skin, or the leopard
his spots .
In Yorkshire, Shrewsbury quoted Sellers view that careful statistics
prove that more than two-thirds of the parish priests in the West Riding, and 35 of
95 in the East Riding, died during the pestilence, the inescapable inference from her
statements is that at any rate for the West Riding more than two-thirds of the
population died of bubonic plague, which, in view of the nature of that disease and
the wide dispersion of the countys population, is an impossibility.
He also stated:
If, as seems extremely probable, the bulk of the scanty population of the county of
Durham in the fourteenth century occupied the valleys of the Tees and the Wear,
the rest of the county must have been virtually desolate, with no conceivable
possibility of the spread in it of bubonic plague. In all probability the only microbial
diseases that could have spread outside those valleys under the prevailing conditions
were smallpox, typhus fever, and epidemic inuenza [our italics]. Nevertheless Gee
101
declares that the county was not spared by The Great Pestilence and he asserts
that there is documentary evidence that its death-roll was heavy. For example, he
says that in the Cursitor Roll special provision for a land title was made in the event
of the death of the assigns during the pestilence which was then raging. Moreover,
he adds, the papal registers for the next forty years give in their concessions to
monastic houses conclusive proof of the virulence of the outbreak in the north . . .
Wilson nds little contemporary information about the eect of The Great Pestilence upon the beneced clergy in this county; but he notes an ominous gap in the
diocesan register of Carlisle for the six years 1347 to 1352, which he assumes
signies a severe plague-mortality among the clergy. It is improbable that such an
assumption is justiable because the geography of this county, the sparseness of its
population, the unease of its living conditions, and the etiology of bubonic plague,
conjoin to make it unlikely that P. [ : Yersinia] pestis penetrated farther into the
county than Carlisle at this time . . . The references to the presumptive activity of
bubonic plague in Wales have most been supplied by Rees, who arms that
Snowdonia, especially Anglesey, was overrun by The Great Pestilence. He may be
correct for Anglesey, which appears to have been populous in the fourteenth
century, but it is extremely improbable that Snowdonia was suciently colonised
by the house-rat to support an epizootic of rat-plague. In any case he nullies his
armation by his succeeding statement that while certain hamlets had a heavy
mortality others were little aected and even entire commotes [Welsh manors]
escaped lightly.
102
historians, but felt that 20% to 25% may be a more likely gure. He says:
Possibly the traditional gure of a third of the population dead may be correct:
many scholars have accepted it. But modern experience of [? bubonic] plagues
suggests a fair number of spots which would remain immune and aect the total.
Perhaps 20 or 25 per cent may be nearer the mark than 33.3 . . . We hear of
depopulated but not deserted villages. When villages can be studied in groups, what
surprises us is the continuity of their life. On the vast estate of the Bishop of
Winchester in Wessex there is no sign of chaos or complete depopulation, and
following the Pestilence there is no revolution either in agriculture or in tenure.
Across the country, on the East Midland and Fenland manors of Crowland, it is
just the same: the estate accounts run on, and what is most important, there seems
always to be someone ready to take over a vacant holding.
103
that (i) the population could not have recovered so quickly and so well
from a higher level of mortality and (ii) bubonic plague could not have
spread so extensively in such a low-density population. We suspect that
this is another example of special pleading on the part of Shrewsbury in his
eorts to convince us that bubonic plague was responsible for the Black
Death.
However, equipped with the detailed records contained in the parish
registers, we are able to make a good estimate of the mortality in the plague
at Penrith that occurred 250 years later in 159798. Coupled with the
famine that immediately preceded it, over 50% of the population died in a
3-year period (section 5.2) but, although this mortality crisis had subtle and
long-lasting eects on the population dynamics (section 13.17), the community rapidly returned to steady-state conditions with the same average
number of annual births and deaths. Human populations, to the outside
observer, can adapt remarkably quickly and eectively to major mortality
crises. Epidemics in the cities of northern Italy frequently caused a 40% to
50% mortality (section 11.3). We conclude that the mortality in the more
populous parts of England during the Great Pestilence may well have been
at least 30%.
104
Shrewsbury (1970) also drew attention to the fact that certain individuals in a population, of which all the members are equally exposed to
the pestilential atmosphere, escape the disease.
(iv) There are scattered references to the death of domestic animals associated with the Great Pestilence, as in the account of Michael of Piazza
(section 4.2, note 2).
(v) It is dicult to accept unreservedly the account of the arrival of the
Great Pestilence in Sicily in the Genoese galleys (section 4.2) which
must have been at sea for a long time and which had no cases of plague
on board and yet the epidemic had spread to every part of Genoa
within 2 days. Twigg (1984) has made a careful study of shipping and
voyages between Genoa and the Crimean ports at the time of the
Black Death and concluded that the evidence is not consistent with the
causative agent being bubonic plague. Indeed, from what we have
learnt about the rapid spread, the short time between infection and
death and the high infectivity it is dicult to conceive of any disease
that could have remained extant for the 45 weeks of the voyage
(Twigg, 1984) with an apparently healthy crew who on arrival caused
widespread infection within 2 days of coming ashore. Presumably, if
the story of the Genoese galleys were true, and it is widely accepted, the
crews would have had several carriers of the disease among their
number, raising again the question of immunity to the disease.
The understanding and epidemiology of many diseases is complicated
by chronic infection in immunodecient individuals, who become silent or
inapparent carriers of the virus or bacterium, sometimes essentially for life.
Anderson & May (1991) gave the example of hepatitis B: the virus infects
people everywhere in the world, with the highest rates of infection in
sub-Saharan Africa and east Asia. Although the majority of cases are not
serious in developed countries, a minority lead to acute hepatitis with
jaundice, causing some deaths. There are estimated to be about 200 000
new hepatitis B infections in the USA each year, of which about 25%
experience acute hepatitis, leading to around 10 000 hospitalisations and
250 deaths. More generally, the outcome of the infection depends on the
immune response of the individual: an adequate immune response leads to
the production of antibodies that clear the infection and produce lifelong
immunity whereas an inadequate immune response allows continued viral
replication and, if this is maintained for 6 months or longer, such viral
production usually persists indenitely and the infected individual becomes an asymptomatic carrier of infection.
105
We agree entirely with Twigg that the Genoese galleys did not bring
bubonic plague and, on balance, we do not believe that their crews were all
silent carriers of the disease. Even if the original story is only partly true, in
all probability the arrival of the galleys was coincidental with the outbreak
of the Black Death and we discuss this point further in section 13.8.
106
Peak mortality
Nov. 1348
Dec. 1348
Jan. 1349
Jan. 1349
Mar. 1349
Apr. 1349
Apr. 1349
Apr. 1349
Apr.May 1349
May 1349
June 1349
July 1349
Mar. 1349
Jan. 1349
Apr.May 1349
Mar.June 1349
MayAug. 1349
July 1349
July 1349
JulyAug. 1349
JuneAug. 1349
JulyAug. 1349
July 1349
Sept 1349
107
reect the spread of the plague northwards and eastwards. Thus the
pestilence occurred in every month of the year and it is noteworthy that the
peak mortality in Bath and Wells was recorded in January; Shrewsbury
(1970) commented that this runs counter to the usual epidemiological
behaviour of bubonic plague in a temperate climate. It was not conned to
the summer months and in some counties the peak of the epidemic occurred in winter. It is evident that the Black Death simply struck in a diocese
when the infection arrived from the adjacent county and the epidemic then
exploded with peak mortality occurring some 23 months later and the
pestilence eectively burnt itself out after some 69 months, indicative
from Reed and Frost modelling of a long serial generation time.
108
109
4.11 Plagues in England following the Great Pestilence: the 14th century
The Black Death was followed in England by a series of epidemics in the
second half of the 14th century, although these were progressively less
violent, but Ziegler (1969) suggested that the second epidemic of 1361, by
any standards other than those of the Black Death, was catastrophic in its
dimensions. Many writers have assumed that these were fresh outbreaks of
bubonic plague, but Shrewsbury (1970) believed (and he listed his evidence)
that this assumption was not warranted.
There is no doubt that smallpox, measles, typhus fever, and dysentery were repeatedly epidemic in fourteenth-century England; pneumonia undoubtedly occurred in
epidemic form in the winter months, and whooping-cough, the enteric fevers, and
inuenza in all probability were also epidemic at times. It is certain that tuberculosis, the white scourge, was responsible for many deaths annually . . . It is probable
that diphtheria, erysipelas and poliomyelitis were locally epidemic from time to
110
time and that . . . some of the great medieval epidemics of colic that aicted the
British Isles may have been maritime importations of cholera.
112
134850
136061
1369
1375
India
1907
33
7
17
25
33
39
20
10
14
20
29
36
20
15
9
13
23
0
0
8
13
9
30
0.3
1.3
1.9
1.1
0.7
0.4
Figures quoted are the calculated percentage of people dying of plague or bubonic
plague in each age group.
Data sources: Russell (1948), taken from the Inquisitions Post Mortem; Twigg
(1984), taken from the reports of the Plague Research Commission (1907a).
4.3 (data from the Plague Research Commission, 1907a; Twigg, 1984): the
results are very dierent, not only is the overall percentage mortality of
bubonic plague very much lower, but it killed predominantly those in the
640 year age group.
113
He continued:
If a large number of people showed similar manifestations within a relatively short
period of time, then a plague was recorded; but the actual disease might be any one
of a dozen or more communicable or transmissible diseases of man. It is therefore
quite unrealistic to accept the record of a plague before the middle of the seventeenth century as synonymous with an outbreak of bubonic plague, and even after
Sydenhams day, up to the middle of the nineteenth century, bubonic plague and
epidemic typhus fever were clinically indistinguishable.
14057
Adam of Usk described a great plague that prevailed throughout England, especially among the young, that was swift in its
attack.
A major outbreak with reported high mortality in London and
in country villages where many families were said to be almost
exterminated. Shrewsbury (1970) commented:
If numerous country villages were synchronously involved . . . it must
have been a disease of high infectivity that spread rapidly by human
contacts, or indirectly by human contamination of foodstus, including
drinking water, which in the summer were largely consumed uncooked.
The most virulent epidemic of bubonic plague could not destroy much
more than one-third of the population of London which, according to
Creighton, was estimated at 44,770 in 1377, and which certainly had
not reached 90,000 by 1407. Moreover bubonic plague is essentially an
urban disease and only those villages in the near vicinity of an infected
town would be involved in the fundamental rat-epizootic, although
occasionally a village at a distance from the epizootic focus would be
infected by urban fugitives harbouring blocked eas. The gure of
30,000 deaths in London seems to have had a fascination for fteenthcentury recorders, for it is repeatedly quoted for dierent outbreaks of
disease in that city.
114
5
Case study: the plague at Penrith in 159798
116
117
1
104
2
77
3
34
4
16
5
7
6
3
7
0
8
1
June 1598 (when there were four burials) there is recorded in the published
transcript of the registers There is no sign here of a gap, but some entries
have probably been lost. Furthermore, there are wills of people who died
in the plague whose burials we have not been able to trace in the registers
and we conclude that the totals listed above are probably short by at least
50 deaths.
Of those dying during the epidemic, 485 have been identied as belonging to 242 families; a further 74 families appeared to be unaected. The
previous year (1597) was a time of extreme hardship and deprivation for the
northwest and a number of those succumbing to the pestilence of 1598 may
have been vagrants in search of food. Table 5.1 shows the number of deaths
in each family; three-quarters of the families suered one or two deaths.
Sixty-three families became extinct and 79 families had only one surviving
parent; a further eight families were left parentless.
Furness (1894) in his The History of Penrith from the Earliest Record to
the Present Time . . . gave the following account of events in the plague,
which, although it was probably anecdotal and conjectural, was of interest
because it was written at a time before the true biology of bubonic plague
had been elucidated. However, he knew all about buboes.
During this dreadful time the actual state of Penrith can scarcely be imagined. Not
a solitary marriage was registered for the whole summer. Houses supposed to
contain the infection were shunned, and their inmates suered to die unaided.
People almost feared each others looks. Evening, the time when the attacks of this
disease generally came on, had peculiar terrors during the visitation of the plague.
The rst paroxysm or period, which included from the evening to the following
night, was frequently fatal. The third and fth days were considered, on the whole,
those of greatest danger, and if they survived over the fth day and the bubo was
fully formed, then the patients were considered almost out of danger. All these
circumstances were only too well known where the plague had been for some time
raging. The wild and furious look accompanying the disease in its incipient state,
which ought to bespeak pity for the suered, dispelled all feelings of humanity from
the breast of those whom circumstances brought in their way. The staggering
occasioned by an extreme prostration of strength was a warning to his neighbour to
ee. The poor suerer sought his home perhaps he was the last of his household
and the indierence to recovery, which was considered a most unfavourable
symptom, alone relieved the horrors of his despairing and forlorn condition.
118
The farmers from outside the town would not enter while the plague
raged, all regular markets were suspended and temporary markets were set
up on the outskirts. Furness (1894) reported that the inhabitants of the
dales, who were doubtless especially timorous, came no nearer than
Pooley. Townsfolk paid for supplies by tossing coins into hollowed-out
stones containing some crude, supposedly disinfectant uid, possibly vinegar (Furness, 1894), the so-called plague stones. One such receptacle is
preserved in the grounds of Greengarth old folks home in Bridge Lane at
the southern end of the town. The other plague stone, which was located in
a small eld adjacent to Milton Street (formerly Grub Street) and served
the people coming to market from the north, has now disappeared (Irving,
1935; Watson, 1992). This partial isolation foreshadowed the successful
quarantine practised at Eyam in 1666 (see section 10.7) but was more
rudimentary than the measures introduced at Carlisle (section 7.7) and
York (sections 9.1.1 and 9.4).
119
Estimate
1554
1587
1596
1700
1500
1597
1350
1598
1599
1610
1613
1623
858
1134
1150
1642
1673
1676
1688
1233
1147
1365
1147
Source
First plague at Penrith
200 deaths, probably largely because of typhus
153 deaths of people native to Penrith, probably because
of famine
Before start of plague. Estimated from family
reconstitution study
Second plague at Penrith
485 deaths of the plague of people native to Penrith
Appearance of 65 new families during 160010
Estimated from family reconstitution study
Mortality crisis (241 deaths recorded in parish register);
probably severe famine
Protestation returns for parish (411 ; 3)
Hearth Tax for township (270 ; 4.25)
Compton census for parish
Dentons survey for parish
Data for 1642, 1673, 1676 and 1688 taken from Clark et al. (1989).
registers is collated for every family (see Wrigley, 1966; Scott, 1995; Scott &
Duncan, 1998). The calculated values are 1350 in 1597 and 1150 in 1613.
The population estimates for 1596 and 1587 are derived from the value for
1597. These results must be regarded as underestimates. Sixty-ve families
moved into Penrith in the 12 years following the plague, producing a
population estimate of 1134 in 1610, a value that agrees well with the
estimate of 1150 in 1613 derived from the family reconstitution study and
with the estimates derived from the Hearth Tax and other assessments
(Table 5.2). It is concluded that the population at Penrith was some 1700
before the famines and plague that occurred during the period 158798
and was some 1200 in the early 17th century when the community was
settling into a steady state (see section 13.17). Assuming that the population
was a minimum of 1350 at the time of the plague, approximately 45% died
during the outbreak.
5.3 The three phases of the epidemic: the serial generation time and
contact rate
The mean annual number of burials at Penrith was 49 for the period
155785; this number rose to 103 in 1586 and to 195 in 1587, when the
population experienced a rise in mortality; there was an average of 57
120
Fig. 5.1. Monthly plague burials at Penrith, August 1597 to February 1599,
illustrating the three phases of the epidemic: small autumn peak, almost complete
disappearance in winter and the explosion of deaths in the following spring
and summer.
burials per annum thereafter until the fateful years of 159698. Figure 5.1
shows the monthly plague burials at Penrith. There is a gap in the register
of burials between 13 and 24 June, see above, probably because of the
pressure of events on the vicar (whose wife and son died in May) and his
clerk, and it is almost certain that there were more victims than the 62
recorded. If so, June may have been the month of peak mortality during the
epidemic.
The monthly pattern of deaths represents the ngerprints left by the
disease upon which the historical epidemiologist may work. It is evident
that there were three phases to the pestilence: (i) the epidemic began with a
single death in late September 1597 and rose to a small autumnal peak in
NovemberDecember, (ii) during the winter there were no plague deaths in
January and only a single death in February 1598, and (iii) the pestilence
121
122
Fig. 5.2. Seasonal pattern of plague mortality in four rural towns in England. (A)
Oswestry, Shropshire, 1559. (B) Totnes, Devon, 1570. (C) Ashburton, Devon, 1625.
(D) Manchester, Lancashire, 1645. Data from Shrewsbury (1970).
123
Fig. 5.3. (A) Reed and Frost modelling of the third phase of the plague. N : 1325,
serial generation time 25 days, mean number of eective contacts : 3. (B) Plague
burials at Penrith in phase three plotted at the same intervals of the serial generation time.
124
Fig. 5.4. Pattern of deaths during the plague at Penrith 159798. The families are
arranged sequentially in order of the rst recorded death from plague. Reading the
gure horizontally shows the spread of the disease through a family; reading the
columns vertically illustrates how the disease spread from family to family. Open
circles, adult female; open squares, adult male; closed circles, child aged 115
years. Recordings begin October 1597, after Hogsons death. Note the high household R in phase one, the almost complete disappearance of the plague in the winter
months and the explosion of the epidemic in the spring of 1598 with a high
interhousehold R .
125
126
contact rates will explain minor dierences between the dynamics of the
epidemics in dierent populations. But, as we show below, reality is more
complicated than simple theory. The eective contact rates within households were much greater than between households and it is upon the latter
that the propagation of the epidemic depended. Furthermore, as we shall
see, eective contact rates diered between winter and summer.
5.4 Spread of the epidemic at Penrith
The spread of the plague is shown diagrammatically in Fig. 5.4, in which
the mortality in each family is shown horizontally (household contact rate).
The families are arranged in order of the rst recorded death. The data
presented are probably not completely accurate because sometimes, for
example, there is single record of the death of a member of a household who
was listed as a servant and the infection may have occurred within the
household of their employer. Nevertheless, Fig. 5.4 shows graphically how
the disease spread; reading the columns vertically shows how the disease
moved from household to household and the rapid spread between households in summer 1666 is clearly evident. The spread of the infection in the
rst and second stages of the epidemic (September 1597 to May 1598) is
illustrated in detail in Fig. 5.5. It has all the characteristics of a lethal
infectious disease spread person-to-person and shows a pattern identical
with the outbreak of plague at Eyam in 166566 (see Chapter 10).
5.5 The epidemic during the rst two phases: elucidation of the
epidemiological characteristics of plague
Two points are immediately apparent from Figs. 5.4 and 5.5. Firstly, only
10 families were aected in the autumn of 1597 (phase one) and the plague
then appeared to die out, with no fatalities in January and only one in
February 1598 (phase two). Secondly, once the disease was established in a
family during phase one, successive deaths within the household followed
(high household contact rate). With respect to the spread of the plague,
infections within the household are a dead-end and, for the epidemic to be
perpetuated, it must be transmitted to other households. Phases one and
two of the outbreak diered markedly from phase three in this respect,
transmission between families at Penrith was limited during the autumn
and winter (Fig. 5.5) whereas the spread was dramatic in summer 1598 (Fig.
5.4).
It is dicult to determine the epidemiological parameters of the plague
in the Black Death because there is no detailed evidence of the actual
127
Fig. 5.5. Pattern of deaths during the early stages of the plague at Penrith from the
rst death on 22 September 1597. The families are numbered sequentially in order
of the rst appearance of the epidemic (right-hand side). The rst death and the
outbreak in February 1598 are shown with large, open arrows. The break in
JanuaryFebruary 1598 can be clearly seen. The spread of the plague is shown by
solid arrows, initially within families and subsequently between families. The
steepness of the solid arrows illustrates the rapidity of spread (e.g. see May 1598).
Certain individuals are designated by letters: A, Andrew Hogson, stranger; B, C and
D, probably not infected from within the family but from individuals in other
families lying immediately to the left of the dotted lines; E, F and G, possibly also
infected from outside the family. Symbols as for Fig. 5.4.
128
events; equally, it is dicult to analyse accurately the spread of the infection in the plagues in London because of the size of the population.
However, the circumscribed epidemics in towns and villages which we
describe below (e.g. Penrith and Eyam) provide admirable material for the
critical analysis of the epidemiological characteristics.
Three factors point to a long incubation period for the disease. (i) The
duration of the summer epidemic in phase three, as explained above. (ii) In
both Penrith and Eyam (Chapter 10) the epidemic began with the death of
a stranger or visitor who had taken up lodgings and was living, albeit
temporarily, in the town. He must have been infected elsewhere, perhaps
some considerable distance away, and brought the disease into the community whilst still not showing any symptoms. Anyone showing any of the
typical signs of the plague, which were well known, would know that they
had only some 5 days to live and would not be contemplating moving a
considerable distance to communities where they were not known. (iii) The
death of the rst victim, the incoming stranger, was followed by an interval
of 15 (Eyam; see Chapter 10) or 22 (Penrith) days.
The early stages of these outbreaks, whilst the epidemic was being
established, that we have studied in detail, are much more informative than
are the confusing events in mid-summer when the infection spread with
devastating rapidity. We have analysed a number of these outbreaks (see
following chapters) and have determined from a number of sources the
following epidemiological parameters:
Latent period: 1012 days.
Infectious period before symptoms: 2022 days.
Period of symptoms: 5 days.
Total infectious period: 2527 days.
Total time from point of infection to death: 37 days, a very long time and
in complete contrast to pneumonic plague (see section 3.13).
Figure 5.6 illustrates the rst and second phases of the plague at Penrith;
the latent and infectious periods are indicated by appropriate subdivisions
of a line against each person dying. The rst line is that of Andrew Hogson,
the stranger, who died on 22 September; he was infected at a place where
the plague was raging on or about 16 August, so that he became infective
on about 28 August 1597. Dashed vertical lines are drawn on Fig. 5.6 on 28
August and 22 September, denoting the period during which Hogson, the
primary case, was infective and it is evident from this that he infected only
the family of John Railton, a cutler; in all other families the initial point of
Days
0
20
40
60
80
Andrew
J. Railton
d. Eliz. (22)
s. John (c. 20)
f. John
m. Mabell
120
S
S
S
T
s. Anthonie (14)
d. Eliz. (22)
Railton
w. Sussane
Hewer
s. Anthonie
d. Margaret (22)
f. Thomas
d. Katheren
T
T
T
T/Q
Haskew
W. Haskew
John (cleanser)
s. Thomas (13)
Watson
s. Gilbert (13)
d. Marie (17)
Cooke
m. Katheren (22)
s. Gilbert (1)
Q
Q
Walker
s. Michaell (16)
d. Alice (20)
m. (w.) Jane
Blysse
100
Hogson
A. Railton
129
d. Helen (20)
d. Annes (15)
m. (w.) Marion
s. John (13)
T/Q
T/Q
T/Q
T/Q
F
F
Q
Q
Q
Q
Fig. 5.6. The start of the outbreak of the plague at Penrith and its suggested spread
within and between families. In this and subsequent gures the duration of the
infection in each victim is shown as a standardised line that is divided into latent
and infectious periods. The duration of the infectious period of the primary case
(Hogson) is indicated by the vertical dashed lines, during which secondaries can be
infected. The vertical dotted lines are drawn between the day on which the rst
secondary case becomes infectious and the day on which the last suggested secondary dies. Tertiary infections are only possible between these dates. The families or
households are listed on the left-hand side of the gure. P, primary; S, secondary; T,
tertiary; Q, quaternary; F, fth generation infection; d., daughter; f., father; m.,
mother; s., son; w., widow. Age (years) of victim at death given in parentheses where
known. Scale: days after 16 August 1597.
Andrew Hogson
Thomas
Hewer
= ?
Isabell
Hewer
m.1571
Anthony
Railton
John
Railton
Mabell
T
Isabell
Railton
m. 1561
John
Watson
Anthony
Railton
(d. 1588)
m.
Helen (27)
T
Margaret (22)
John (23)
T
Eliz. (22)
John (20)
Frances (21)
Isabell (13)
Mabel (20)
Anthonie (2)
Katheren
Anthonie (14)
Anthonie (24)
Thomas (29)
Eliz. (22)
Janet (26)
Annes (21)
Eliz. (20)
John (20)
T/Q
Marie (17)
William (19)
John (14)
Mabel (13)
(d.pl. 16.9.1598)
Anthonie (12)
T/Q
Sussane
Nicholson
Gilbert (8)
Richard (10)
(d.pl. 4.9.1598)
Isabell (8)
(d.pl. 13.9.1598)
Lucie (7)
Fig. 5.7. Relationships between the Railton and Hewer families. Ages (years) in 1597 given in parentheses where known. Victims are
suggested as secondary (S), tertiary (T) or quaternary (Q) cases. Victims marked with a closed circle died in summer 1598 in phase three of
the epidemic. Dotted arrows indicate suggested pattern of spread of the infection between families. Dashed line indicates suggested family
relationship. d.pl., died of plague.
131
infection occurred after Hogson had died (Fig. 5.6). Likewise, vertical
dotted lines are drawn between the day on which the rst secondary case
became infectious and the day on which the last suggested secondary dies;
tertiary infections in the community are possible only between these dates.
The three co-secondaries (see section 2.2) in the family of John Railton
were, sequentially, daughter Elizabeth (aged 22 years), son John (c. 20
years) and the father John, who were infected on 7, 13 and 17 September
respectively, all probably before Hogson was showing symptoms. Three
co-secondaries infected in rapid succession demonstrate a high household
contact rate and it is possible that Hogson was lodging with the Railton
family; he probably infected nobody else. Mabell Railton, the mother, was
clearly a tertiary infection from one of her two children because she was
infected after Hogson died.
The epidemic spread next to the Anthony Railton and Hewer families by
tertiary infections; all three families were interrelated, as shown in Fig. 5.7.
Only two children of John Railton died in the rst phase of the epidemic
but three more died 12 months later in rapid succession in September 1598
in the closing stages of phase three. Two of the children of Anthony
Railton, Anthonie junior (aged 14 years) and Elizabeth (aged 22 years) were
the next tertiaries infected by the secondaries, probably their cousins (Figs.
5.6 and 5.7). They were infected after Hogson had died; the other members
of the family survived. The rst step in the spread of the epidemic to
another household, therefore, was to a closely related family and, as we
shall see in the spread of other plagues during the autumn and winter (e.g.
Eyam; section 10.3), the transmission was quite probably eected via the
youngsters. Sussane Railton, a widow, whose relationship to the Railton
families is not clear (Fig. 5.7) was the next tertiary infection.
The next group of tertiary infections were in the Hewer family; Isabell
Railton was the sister of Thomas Hewer (Fig. 5.7). Son Anthonie (probably
aged 2 years), daughter Margaret (aged 22 years) and the father Thomas
Hewer were infected on 4, 6 and 7 October, respectively, and so must have
been infected by the three original secondaries in the family of John
Railton, as shown in Fig. 5.6. We suggest that the transmission to a new
household was probably eected by Elizabeth Railton, who was the same
age as Margaret Hewer. Katheren Hewer was probably a quaternary,
infected by her brother Anthonie (Fig. 5.6).
After the death of Thomas Hewer, there was a gap of 7 days before the
plague struck again with the death of John Haskew, a cleanser, who was
infected on 14 October, the same day that Elizabeth Railton (original
co-secondary) was buried. He may have been infected at the Railton
132
T/Q
John
= ?
Haskew
[cleanser]
William
Haskew
[shoemaker]
d. 1589
John (31)
Janet (29)
Eliz.
Whiett
m. 1574
Lancelot (21)
T/Q
Thomas (13)
Ambrose (27)
Isabell (24)
Anthonie (22)
Fig. 5.8. Relationships between the Haskew families. Abbreviations as Fig. 5.7.
John
Cooke
(2)
Marion
Q
(1)
Thomas
=
Hornsbie
[Smith]
(d.pl. 6.10.1598)
133
Annes
Whiett
(d. 1586)
Margaret
(d.pl. 24.7.1598)
Katheren (22)
Q
Q
Helen (20)
Andro
(d.pl. 15.12.1598)
Sussane (17)
Q Gilbert (1)
Q
Annes (15)
John (13)
Fig. 5.9. Relationships between the Blysse, Cooke and Hornsbie families. Both
Marion Blysse and Thomas Hornsbie married twice. Abbreviations as Fig. 5.7.
where the infection was rst introduced between 27 and 30 October (Fig.
5.6). The interrelationships between the Blysse and Cooke families are
shown in Fig. 5.9: Katheren Blysse (aged 22 years) the eldest daughter, had
married John Cooke in 1595. Katheren and her son Gilbert (aged 1 year)
were both infected on 27 October and her sister, Helen Blysse (aged 20
years) on 30 October 1597 (Fig. 5.6). There is no evidence of a relationship
between the Blysse family and any of the infectives above, but the most
likely candidates would be Margaret Hewer and Elizabeth Railton, both,
like Katheren, aged 22 years. One of them may have been visiting the
Blysse family, particularly Katheren and Helen, and infected the baby at
the same time. The important point is that the infection probably occurred
indoors.
Annes Blysse (aged 15 years) and Marion Blysse (her widowed mother)
were then infected on 7 November and son John (aged 13 years) on 8
November (Figs. 5.6 and 5.9); they might have contracted the disease from
Katheren when she rst became infective. Figure 5.9 shows that John
Cooke, and Sussane Blysse survived.
There were also two separate plague deaths at this time: Janet, wife of
Robert Ladiman and John Lyvocke, a joiner, who were buried on 9 and 12
December, respectively.
Meanwhile the epidemic had also spread to the Walker family (Fig. 5.10);
the son Michaell (aged 16 years) was infected on 27 October and he then
infected his sister Alice (aged 20 years) and mother Jane (aged at least 47
years) (see Fig. 5.6). Jane Walker was buried on 26 December 1597, the last
134
Thomas
m. 1590
(probably not living
in the household)
F
Jane
(aged at least 47 years)
Alice (20)
F
Michaell (16)
Q
Fig 5.10. Plague victims in the Walker family. Abbreviations as Fig. 5.7.
plague death in phase one of the epidemic (Fig. 5.6); thereafter the outbreak
persisted through the winter only by the skin of its teeth and the inhabitants of Penrith must have thought that the epidemic had ceased.
In summary, we have shown that, by combining family reconstitution
techniques and epidemiological theory, it is possible to provide a human
story set within a scientic framework that underlay the events during the
autumn of 1597. The epidemiology summarised in Fig. 5.6 is completely
consistent with an infectious disease that was spread person-to-person and
it is possible to identify the probable infective in each case. This disease was
not bubonic plague; the disease was brought by an infective stranger who
lodged in the town; there was no colony of black rats in the cold winter of
the Eden Valley, many miles from warmer, southern ports; no eas could
have bred there; the incubation and infectious periods are completely
dierent; the inhabitants of Penrith obviously assumed that it was a
standard infectious plague; the epidemic spread within households and
between relatives during the rst phase in a predictable way, quite unlike
bubonic plague; the dynamics of phase three of the epidemic are completely
consistent with a Reed and Frost infection with a long incubation period.
The key to understanding the epidemiology of the plague lies in the
lengthy incubation period and, because of this, apparently healthy infectives could move around the country on foot or horseback covering
considerable distances before they were struck down. Another consequence was that between mid-August, when Hogson was infected somewhere else, and the end of December the epidemic had extended only to
quaternary infections (Fig. 5.6).
Only some eight or nine households were infected in the rst phase;
during the autumn there was a high household contact rate but the
135
136
FAMILY
0
Atkinson
John
Herreson
d. Annes (18)
f. Chris
John
Nicholson
d. Annes (16)
s. Richard (8 months)
m. Margaret
s. Thomas (9)
f. Richard
d. Lucie (13)
Hodgson
40
60
T
S
S
S
T
T
T
T
T
Susan
T
Q
Dobson
Mabell (13)
Newton
Isabell
Skilbecke
d. Jane (12)
s. Anthonie (16)
m. Janet
Atkinson
Eliz.
Chambers
f. Wm
m. Mabell
Q
Q
T
Q
Hodgson
w. Jane
Walker
m. Isabell
Watterson
s. Cuthebert (8)
d. Marion (13)
Alexander (36)
100
s. Thomas (5)
d. Janet
m. Anne
Castelhowe Margaret
Dobson
80
Wilson
Gibson
20
Q
T/Q
Fig. 5.11. Suggested sequence of infections in the start of phase three of the epidemic
at Penrith. The epidemic began again with John Atkinson, who is regarded as a new
primary case. Details and abbreviations as Fig. 5.6. Scale: days after 4 January 1598.
137
is interesting that the gaps between Andrew Hogson (the stranger) and
the next recorded plague victim (see Fig. 5.6) and between young John
Atkinson and the rst victim of phase three of the pestilence are 22 and 23
days, respectively, again indicative of the long incubation period of the
disease. John Atkinson junior (considered as a new primary case; probably
aged 21 years) infected four co-secondaries in three families (Fig. 5.11); of
these Annes (aged 18 years) was the daughter of Xpofer ( : Christopher)
Herreson, and Annes (aged 6 years) and Richard (aged 8 months) were the
children of Richard and Margaret Nicholson. Once more the youngsters
had a major role in the transmission of the epidemic to other families;
possibly John Atkinson and Annes Herreson were sweethearts.
Phase three of the pestilence gathered momentum in March, with at least
17 tertiary infections from the initial 4 secondaries, including Chris
Herreson, Thomas (aged 9 years), Lucie (aged 13 years) and their father,
Richard Nicholson. Burials after 1 April 1598 were probably mixed tertiaries and quaternaries and there was a gap between 7 and 14 April after
which there was a wave of infections and the epidemic exploded, illustrated
by the steepness of the solid arrows in Fig. 5.5.
It is evident from Fig. 5.4 that during phase three of the epidemic the
families diered sharply in their household contact rate; in some there was
a single burial whereas in others the infection spread through the family.
The pattern of the epidemic in phase three was quite dierent from that in
the autumn: in the spring and summer of 1598, the infection passed readily
and rapidly between families. For example, 14 people from dierent families died on 1 August.
138
Fig. 5.12. Female age-specic mortality in the plague at Penrith, 159798. (a)
Estimated percentage age-specic mortality of plague victims (closed circles); (b)
cumulative female mortality curve at Penrith, 160049 (closed squares); (c) theoretical age structure, Level 4, Women, Model West. Note the close correspondence
between lines a and c.
compared with the pattern prevailing before the crisis. The mean annual
number of burials before the plague from 1557 to 1597 was 60.2 and they
were distributed between the age groups 04, 514, 1544 and 45; in the
proportions given in Table 5.3. Subtracting column (1) from (2) gives the
observed number of excess burials because of the epidemic for each group;
see column (3). This is then compared with the expected excess burials if the
age-specic mortality corresponded with the number of people at risk in
each age group estimated from the theoretical stable conditions; Model
West, Level 4 was used to estimate the percentage of the population in each
age group. Those aged over 40 years, for which an age at death cannot be
estimated, were divided in the proportions suggested by the model population percentages. For the younger groups (i.e. under 20 years), only 34
burials were unattributed and these were again divided in accordance with
the distribution suggested by the model population statistics.
Some authors have suggested that plague mortality was more severe
among older children and adolescents rather than an equal incidence
among all classes of the population. Hirst (1952) wrote that The largest
number of cases occur in persons between the ages of 10 and 35 years, the
very young and elderly being comparatively little aected and children
aged 510 years showed the lowest mortality. In a London parish, Hollingsworth & Hollingsworth (1971) found an increased mortality only for
those aged 444 years and Pollitzer (1954) stated that adolescents and
139
Age
group
(yrs)
Mean
annual
pre-crisis
burials
155796
(1)
Corrected
annual
crisis
burials
(2)
04
514
1544
45;
All
21.0
5.2
22.8
11.2
60.2
84.8
108.6
226.4
65.0
484.8
AgeRecorded structure
excess
in model
(2) (1) population
(3)
(4)
63.8
103.4
203.6
53.8
424.6
12
20
46
22
100
Expected
excess
burials
(5)
Excess
ratio
(3)/(5)
(6)
58.2
97.0
223.0
106.6
484.8
1.1
1.1
0.9
0.5
0.9
adults aged up to 45 years displayed the greater susceptibility. It is generally agreed that those aged in their mid-40s and above were less likely to
die during a plague outbreak, which might suggest a previous exposure and
immunity to the disease. It is noteworthy that the rst plague at Penrith
was in 1554, 43 years before the great outbreak in 1597. In conclusion, no
age group was markedly aected at Penrith, although those aged over 45
years had a lower than expected mortality.
Women were said to be more susceptible to plague than men in parts of
Europe, although apparently the reverse was seen in England (Slack, 1985).
Hollingsworth & Hollingsworth (1971) and Hirst (1952) have suggested
that male mortality was more severe than female, whereas the evidence for
Eyam (Bradley, 1977) and Colyton (Schoeld, 1977) suggests that the
pattern for the sexes was similar. At Colyton, the sex ratio of burials was
93.7 males per 100 females, which was found to be close to the ratio of 92.9
for the 10 years prior to the crisis. Female victims outnumbered males in
Barnstaple, Chelmsford and Stratford-upon-Avon (Slack, 1985). Pregnant
females, who almost invariably aborted, displayed the highest mortality
and, during the 1665 plague at London, a total of 432 deaths due to
childbed, abortive or stillborn causes appeared during August and
September, when mortality was at its highest. Often the cause was more a
lack of attention during connement than any direct consequence of the
plague (Leasor, 1962). At Penrith, the sex ratio of plague victims was 137
females to 100 males.
140
To read the wills of those who died in the plague allows an insight into the
devastating rapidity of the progress of the disease and, more particularly,
reveals the tragedy in those stricken families who faced the inevitable with
dignity and resignation. They remembered the poor of the parish, often
leaving them a quantity of bigg, a poor form of barley, and several of them,
as the extracts below show, left money for the building and repair of the
bridges in the town. Several (see will (f)) bequeathed belongings that were
left to them by their father who had died in the plague but which they had
not yet inherited.
The only people who made wills during the plague were those with
substantial possessions to leave, and inspection of the extracts below shows
that these prosperous members of the town, mostly successful tradesmen,
were all interconnected and interrelated with one another and with their
witnesses. They were a separate group from the gentry, of whom only three
died. Some salient points from the wills are listed below:
(a) Michael Dobson was married on 20 July 1598 and made his will a
month later on 27 August 1598. He was buried, aged 20 years, 5 days
later, on 1 September.
He leaves to his wife Isabell (who survived the plague) all his tenements barnes and yeards with the appurtenance and my tythe estait
and tenament right during her wydowehood.
Witnesses:
John Dobson (probably brother), who, with his wife, survived the
plague but four of his children had died of plague before the will was
written. He was born before the earlier outbreak of plague in 1554.
Edward Todd, who survived with his wife, but they lost two children
after witnessing the will. He may have been born before the earlier
plague.
John Burrowe, who survived the plague with his wife and two
children.
Thomas Hornsbie, who survived the plague with his wife and two
children.
(b) John Steinson made his will on 30 August 1598 and was buried 3 days
later on 2 September. He leaves all that tenement with appurtances
which was my fathers to his eldest son Thomas (who probably survived
the plague) and all the rest of goode movable and umovable I do maike
and ordaine to my son Richard and Margaret my daughter. Richard
was buried 2 days after his father, whereas it is noteworthy that
141
Margaret died of the plague some 6 weeks later. A minor bequest was
to the poore of the parish 6 pecke of ote meale.
Witnesses:
Robert Nelson, who survived with one child whereas his wife died of
the plague.
Richard Salkeld, who lost two children in the plague but the rest of
the family survived.
Anthony Steinson (schoolmaster). All the family survived.
Robert Steinson (shoemaker), who lost one child.
Thomas Fenton and Michael Graye, both of whom survived with
their wives.
(c) Arthur Gibson lost his daughter Elizabeth on 10 July 1598 some 10
weeks before the plague struck his family again; she was buried on the
fell. A child was buried on 19 September 1598 and Arthur makes his will
2 days later on 21 September. His wife died on the next day and he and
another child (Ann aged 9 years) died on 24 September.
In his will, he wishes
my bodie to be with the bodies of my wye and children so manye of
us as shall die at this tyme be buried within the parishe church yard of
Penreth.
I give and bequest all my lande tacke and tenements to John my
eldest son [who survived the plague; he may have been living away
from home on his own land].
I give to Jane my eldest daughter hawked cowe and also if all my
children dye but her I give her twenty sheepe.
Jane is described as a servant and may have been living away from
home. She died in 1606, aged 26 years.
Witnesses:
Cuthbert Byrd who survived the plague but whose wife and three
children died. Cuthbert may have been born before the earlier
plague.
John Castelholme (? alias Castlehow) who survived but whose wife
had died of the plague earlier on 22 March 1598.
John Turner who was born before the earlier plague and survived the
plague in 1598.
(d) Robte Holme wrote his will on 25 September 1598, made additions on
the 27th and was buried on 3 October, 8 days after beginning his
testament. The following are extracts:
I give and bequeath to the buylding of Sandgate Bridge 6s 8d and 4s
3d which Willm Bowman of Kirkoswald oweth me, and to the
142
143
144
Witnesses:
John Turner, who was born before the previous plague. He and four
children survived but one child died.
Robert Wilson, a glover, who with his wife and child died in the
plague. He was born after the plague of 1554.
Thomas Suttons widow, Mabell, made her will on 20 August 1598
and was buried on 22 August 1598, 1 month after her husband. She
made the following bequests:
to my sister Jane . . . my best petticoat.
to Robt Wylsons wye my next best coat my featherbed and one shyft. [Robt
Wilson and his wife Janet were buried 4 September 1598.]
to Anthonye C. . . ell wye and her daughter workday coates.
to Reynold Lucas my husbands best fustian doublet and pairr of grene hose
and other stockinges and shorte jerkin.
These wills suggest the rapidity of the course of the disease, with death
coming in about 35 days from the appearance of symptoms. It is striking
that only 2 days before burial (not death) these dying people were still able
to think clearly and to make detailed and careful arrangements, including
the guardianship of their children. Several members of some families died
within a few days of each other (see will (h)) and in one case (will (g)) all
three members of the family died on the same day. They were obviously
well aware that the remainder of the family would probably die (will (c)).
145
Some of the older children who were left major bequests and who did not
die in the plague may have been living away, perhaps on their own
tenements (will (c)). Mabell Sutton (will (j)) was buried 1 month after her
husband Thomas; was she infected by him about the time when his symptoms showed or re-infected later? See also wills (b) and (e). Reading the wills
suggests the possibility that, once showing symptoms, children may have
died more quickly than adults (see will (h)): children were left bequests in
the will and were presumably apparently healthy but were buried on the
same day as their father.
But not everyone who must have been in contact with an infected person
died. Elizabeth Browne, a widow (see will (i)), was cared for in her last days
by the Crosbie family, all of whom survived. There are many cases where
one or more in the family, often a young child, escaped death and survived
the plague. Many of the witnesses survived the plague, in spite of being
close to a dying man when they signed his will and in spite of also having
plague in their own families. There does not appear to have been any
diculty about nding someone prepared to be exposed to the infection
when witnessing the will and some had over six witnesses. We conclude
that, within the town, the inhabitants of Penrith did not practice isolation
to any great extent.
Were these people who were clearly exposed to infection but survived the
plague immune or were they infected but recovered? If the latter, were they
subsequently immune? It was believed by the inhabitants of Eyam that it
was impossible to catch the pestilence again after recovering from an
infection (section 10.8). The parish registers of Penrith do not begin until
1556, 2 years after the plague of 1554 and so the dates of birth of the older
members of the community cannot be established. Nevertheless, judging by
the dates of marriage and of the births of their rst children, some of the
witnesses can be identied (see above) as probably having been born before
1554 and so survived the rst plague. These men seemed also to have
survived the plague of 159798 and it is possible that they were immune or
acquired immunity from the earlier outbreak of the pestilence.
When the plague in northwest England reached Warcop, the barn where
the victim died was burned (see section 7.8) because it was believed that
miasma or noxious vapours in the atmosphere were responsible for the
disease, and that miasma could be retained in clothes or bedding for long
periods and transported, possibly by domestic animals, from house to
house (Slack, 1985). In 1666, a Statutory Order was passed that a good
quantity of unslaked lime be put into the graves of plague victims, and that
the same should not be reopened within the space of a year or more for fear
146
Table 5.4. Annual baptisms and burials at Penrith before, during and after
the plague
Baptisms
Burials
158796
(mean)
1597
1598
1599
1600
16009
(mean)
68.3
75.3
46
209
27
627
56
19
67
31
59.9
43.6
147
Table 5.5. The occupations of the head of the households in Penrith aected
by plague mortality
Occupation
Shoemaker/cordwainer
Smith
Glover
Tailor
Labourer
Gentleman
Tanner
Cutler
Fiddler
Wright
Beadle
Quarryman
Maltster
Saddler
Locksmith
Waller
Gardner
Joiner
Cleanser
Vicar
Tinker
Fletcher
Capper
Postmaster
Schoolmaster
Other classications:
Householders
(no occupation given)
Servants: 7 males; 11 females
No. of families
where the head of
the household died
of the plague
No. of families
where at least 1
person died of
the plague
8
5
5
5
4
3
2
1
1
2
2
2
1
1
1
1
1
1
1
0
0
0
0
0
0
10
9
6
6
4
3
3
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
11
29
while the plague still raged, with 28 in 1598 and 59 marriages in 1599 (the
average for the previous 10 years was 19). There were 28 marriages in 1600,
after which the average for the next 10 years was 17. Family reconstitution
shows that, of the marriages taking place between 1597 and 1600, remarriages accounted for 28% in 1598, 39% in 1599 and 11% in 1600. So, of the
42 husbands and 37 wives widowed by the plague, 25 husbands remarried
(60%) but only 12 widows remarried (32%). Five of the widowers married
widows.
Table 5.5 shows the gaps left by the deaths of tradesmen in Penrith and
148
the occupations of the head of the families where at least one plague death
occurred, and it can be seen that the families of labourers and gentlemen
were equally aected; the vicar and schoolmaster also suered casualties.
6
Pestilence and plague in the 16th century in
England
It is bewildering to read Shrewsburys (1970) account of diseases in England in the 16th century and one gains the impression that plague was
rampant somewhere in almost every year of the century. In an attempt to
bring some order from confusion, to separate fact from ction and to
distinguish between the possible dierent causes of mortality, both infectious and otherwise, we have subdivided the outbreaks of disease into the
following categories:
(i) The Sweating Sickness that was a scourge for the rst half of the
century.
(ii) Epidemics apparently conned to London and the southeast corner of
England.
(iii) Years in which outbreaks of infectious disease were widespread in
England.
(iv) Plagues that were apparently conned to northern England; these are
discussed in detail in Chapter 7.
150
winter. The Sweat reappeared in 1508, 1517 and 1528, with the fth and last
outbreak in 1551, after which it disappeared from England forever. Most of
the accounts of these outbreaks in England are conned to London except
for the fth Sweat, which began in MarchApril in Shrewsbury and proceeded to London via Ludlow, Presteign, Westchester, Coventry and
Oxford with great mortality (Creighton, 1894).
What was the etiology of the Sweating Sickness? Creighton (1894)
considered that
It is only in the autumn of 1517 that the plague overlaps somewhat on the sweat,
and even then it becomes noticeable mostly in the winter following the decline of
the sweat. The two poisons had existed in English soil side by side, but had not
come out at the same seasons; also the sweat had been mostly a disease of the
greater houses, and the plague mostly of the poorer.
He continues:
Other forms of epidemic fever, in the same pestilential class as the sweat, were
coming to the front in England as well as in other parts of Europe. Thus, in 1539, a
summer of great heat and drought, divers and many honest persons died of the hot
agues, and of a great laske through the realm. The hot agues were febrile inuenzas,
and the great laske was dysentery. Again, in the autumn of 1557, there died many of
the wealthiest men all England through by a strange fever, according to one writer,
or, according to another, there prevailed divers strange and new sicknesses, taking
men and women in their heads, as strange agues and fevers, whereof many died. . . .
That epidemic corresponded to a great prevalence of inuenza on the continent,
which was probably as Protean or composite as the fevers in England. It would not
be correct to say that these new fevers or inuenzas, with more or less of a sweating
type, were the sweat somewhat modied.
151
The disease took at least 6 years to spread from England into Ireland
and Shrewsbury (1970) considered that this slow movement showed that
the Sweating Sickness was not inuenza, the identication generally favoured by medical writers. The rapid onset and haphazard manner of
spread led many to suspect that inuenza was the cause because it is well
known for its ability to mutate from a relatively mild illness to a lethal one.
However, there were no reported respiratory symptoms or secondary cases
of pneumonia, suggesting that Shrewsbury (1970) was correct.
Commenting on the Sweating Sickness, Slack (1985) said that it impressed contemporaries because it was spectacular, killing in 24 hours: It
attacked a community suddenly and then was gone. In particular, unlike
bubonic plague, it struck the prosperous: aldermen and mayors in London
and other towns . . . Yet when we have some precise evidence of its
demographic eects . . . it is not impressive. The disease swept through a
parish in the space of a very few days, a fortnight at most . . .. This is an
important point: the epidemics were of short duration and were apparently
following Reed and Frost dynamics and it is evident that the serial generation time was short, completely dierent from the epidemiology of the
Black Death and the subsequent haemorrhagic plagues.
Slack suggested that the Sweating Sickness may have been an arbovirus
infection of some kind; these viruses circulate from one vertebrate host to
another via the agency of a blood-sucking arthropod, such as the mosquito
that spreads yellow fever. Support for this view has been advanced with the
suggestion that the Sweating Sickness epidemics were always preceded by
heavy rain and sometimes by ooding, both of which would have encouraged mosquito reproduction. However, arboviruses are usually found in
the tropics and the suggestion is not convincing.
The etiology of the English Sweating Sickness has recently been explored
in detail. Dyer (1997) has analysed 680 parish registers for details of the
1551 outbreak: it was predominantly a rural disease with a limited demographic impact. He suggested that the disease was initially spread through
a zoonosis and later by person-to-person transmission because a national
epidemic was capable of very rapid transmission along the lines of communication. This hypothesis has been rened by Taviner et al. (1998), who
also believed that the preponderance in summer and scattered rural nature
suggest a viral infectious agent with a rodent reservoir, although it is not
clear what species of rodent was implicated because brown rats had not
arrived in England and black rats would have been conned to an urban
environment. They suggested that the clinical symptoms, particularly
the marked pulmonary component, are characteristic of Hantavirus
152
1531
1532
1535
1536
153
don until the autumn. This may have been an extended type (ii)
epidemic of haemorrhagic plague. Shrewsbury believed that the
epidemic was dormant in the winter because of the hibernation
of the plague eas but the conclusion and explanation are not
convincing.
An epidemic was violently active in London in autumn 1531,
causing a weekly death-roll of 300400. Henry VIII paid expenses to the poor of Greenwich, who were expelled as a precaution to prevent the spread of infection when he took refuge
there. Shrewsbury (1970) believed that this was not bubonic
plague, presumably because the little evidence available suggests an outbreak in the autumn and winter. The scale of the
mortality and the behaviour of the king may suggest an outbreak of haemorrhagic plague that broke out again after the
winter.
There was an outbreak in London in the autumn of this year
that continued into November. It caused 99 deaths from plague
and 27 from other causes. It apparently began by ravaging Kent
in the summer and then spread to London the reverse direction to be expected from an epidemic of bubonic plague among
black rats in the port of London. Contrary to Shrewsburys
(1970) conclusions, there is no evidence that this outbreak was
bubonic.
An epidemic erupted in London in midsummer 1535 and had
spread all over the city by the end of August, which was warm
and wet (regarded as a bad sign); mortality was augmented in
September but had disappeared by the end of October.
Shrewsbury (1970) claimed that bubonic plague erupted in London in spring 1536 and became epidemic during the summer,
invading Westminster where it was even in the Abbey threatening the coronation of Jane Seymour. Shrewsbury claimed that it
died out in October because frost made the eas hibernate but
gave no evidence of this. It reappeared in spring 1537 and spread
almost everywhere in London by mid-July; this pattern of a
reduced virulence of the plague over the winter months was
frequently found and is typical of type (ii) epidemics of haemorrhagic plague. There is a record of 112 plague deaths in 1 week in
the city (not excessive numbers for a major outbreak in the
metropolis); it spread along the Thames Valley (perhaps by
barge on the river) reaching Windsor and Kingston, and also
154
1538
Even Shrewsbury did not think that most of these epidemics were
bubonic. There is no evidence that they were outbreaks of typhus and we
suggest that, in the absence of additional information, most were of haemorrhagic plague.
6.2.2 Did plague become endemic in London during the second half of the
16th century?
Parish registers began in England after 1540 and a law was passed that
plague burials should be identied therein, so that much better information
is available for the second half of the century. Figure 8.1 shows the number
of plague deaths recorded in the London Bills of Mortality from 1578 to
1680 and it can be seen that the pestilence grumbled on and was virtually
endemic in the city after 1603, with some plague deaths in almost every
year. The enormous epidemics produced during the following 60 years
were not necessarily introduced from overseas because they could have
ared up from one of the isolated cases in the City or have been brought in
from the provinces by an infected traveller. This progression towards the
endemic state after 1578, with prolonged epidemics, is shown in Fig. 8.1,
and Table 6.1 shows the annual number of plague burials in 11 selected
London parishes over the period 155799; in this table we have analysed
all the outbreaks by family reconstitution and some examples are given
below. In addition, there are reported plague deaths in the parish of St
Pancras, Soper Lane, as follows: 1542, 2; 1543, 18; 1547, 6; 1548, 14; 1549, 1.
We have also examined for this period the registers of the parishes of:
St Michael Bassishaw; St Mary Somerset; St Benet, Pauls Wharf; St
Antholin, Budge Row; St Peter, Pauls Wharf; St Paul, Covent Garden;
and Our Lady in Aldermanbury. There are no recorded plague burials.
Although the total mortality in the metropolis in some years was high, as
shown in Fig. 8.1, deaths were few in the individual parishes studied.
155
Nevertheless, our analyses revealed the same pattern in each epidemic with
a latent period of 1012 days and an infectious period of 2527 days, i.e.
within households the latent period could be below the normal 12 days. As
usual, the analyses suggest (although it is impossible to prove the point)
that people were at their most infectious during the rst 10 days of the
infectious period. They were probably less infectious once the symptoms
had appeared, except, perhaps, to those who were nursing them.
With a few exceptions, these epidemics were minor aairs, indicative of a
low household and interhousehold R and hence they usually died out
quickly. And yet these people would have been living in close proximity in
London in crowded dwelling houses. We tentatively suggest that some or
many of the population in these London parishes that experienced regular,
but minor, epidemics were resistant or immune to plague, thereby reducing
the density of susceptibles and hence the eective R .
Another feature of the plague epidemics in London in both the 16th and
17th centuries (see Chapter 8) revealed by our analyses is that a high
proportion of the deaths were of apprentices and servants (both male and
female). This is probably because they were young people brought in from
the provinces who lacked any form of resistance or immunity. Rappaport
(1989) showed that between 1540 and 1589 about 15% of indentured
apprentice carpenters died before they could complete their service. Immigrant apprentices came to London in large numbers and the total number
in 1600 has been estimated as 32 000 to 40 000 (Kitch, 1986).
Table 6.1. Annual plague deaths in parishes in London in the second half of the 16th century
Parish
Year
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
4
67
49
129
14
32
9
14
7
2
1
81
6
11
4
2
29
8
3
3
36
15
20
20
7
16
(a)
8
7
17
14
20
(b)
6
10
23
1
3
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
22
164
2
65
4
11
53
4
1(c)
15(d)
Abbreviations: A, St Martin-in-the-Fields; B, St Peter, Cornhill; C, St Olave, Hart Street; D, All Hallows, Bread Street; E, St Pancras, Soper Lane; F, St
Michael, Cornhill; G, St Dionis, Backchurch; H, Our Lady in Aldermanbury; I, Kensington; J, Christ Church, Newgate Street; K, All Hallows, Honey Lane.
Notes:
(a) There is an entry in the registers between 27 March and 11 June 1563 as follows: The beginning of the plague in this pishe.
(b) Footnote at 2 August 1563 says The numerous burials during this and the two following months were owing to the prevalence of the Plague of this
year.
(c) There is an entry recording the burial of a plague victim in 1593 and there are many burials in the following months.
(d) There is a break in recording the victims of plague between 5 August and 9 October 1593, although there are a large number of burials during this
interval.
HOUSEHOLD
Days
0
Gressa
Westawe
Chapma
Spenser
Bulmer
Robyns
Beck
Bateman
Fove
Lewys
Wylson
Farpoynt
Waldeme
Pykeryng
Pady
Archer
Antony
P
John
Hwgh
Richarde
d.
Humfrey
Remens
apprentice
apprentice
Alice, apprentice
d.
Wyllya
Katheryn
Florence
apprentice
apprentice
apprentice
20
40
60
80
100
120
140
160
P
S
S
S/T
S/T
S/T
T
T
T
T
T
T
T
T/Q
Q
Q
Fig. 6.1. Suggested sequence of infections in the plague epidemic in the parish of St Pancras, Soper Lane,
London, in 1543. There were two separate primary cases (see text). Scale: days after 27 April 1543. For further
details and abbreviations, see Fig. 5.6.
159
160
Fig. 6.2. Weekly plague deaths in London from June 1563 to January 1564.
Abscissa: weeks after 12 June 1563. Data source: Creighton (1894).
Days
0
20
Lynge
apprentice P
s.
P
d.
P
d.
P
Cooke
d.
s.
s.
apprentice
Stookes
Pace
161
40
60
P
S
S
S
d.
s.
apprentice
Fig. 6.3. Suggested sequence of infections at the start of the plague epidemic in the
parish of St Dionis, Backchurch, London in 1563. Scale: days after 26 June 1563.
For further details and abbreviations, see Fig. 5.6.
162
(A)
Days
0
Apprentice
Servant
Servant
Servant
Sears
20
60
80
100
T
Q
F
F
Rychard
Sx
(B)
Days
0
Child
Apprentice
Apprentice
Servant
Servant
Servant
40
20
40
60
80
S
S
T
T/Q
T/Q
Q/F
Fig. 6.4. The epidemic in two selected households in the parish of St Dionis,
Backchurch, in 1563. (A) Mr Sears ( : Sayrs) household where the infection began
with a tertiary infection. Scale: days after 25 July 1563. (B) Mr Revelles household
where the infection began with a secondary case. Scale: days after 3 August 1563.
Note the high number of servants and apprentices dying and the slow spread of the
plague through the households. For further details and abbreviations, see Fig. 5.6.
Sx, sixth generation infection.
Fig. 6.6. Only 34 people died but the epidemic lasted from September
1582 to September 1583, indicative of a very low R within the parish.
Thus, the infection began in the autumn, typical of a type (ii) epidemic
(upper part of Fig. 6.6) with two co-primaries in two households. The
servant introduced the plague into the Poole family, infecting two
members plus another servant. The two primaries infected 56 secondaries (R : 3) but the third, fourth, fth, sixth and seventh generations
of the infection showed a fall in the contact rate, typical of winter
conditions (R successively 0.4, 2, 0.4, 1, 0.3) and hence the epidemic
died out, the last victim dying on 13 January. This type (ii) epidemic
did not survive through the winter, but 43 days later a boy in the
Cockson family was infected as a new primary, presumably from
outside the parish, so initiating a small-scale type (i) epidemic, with
deaths occurring from April to September. The lower part of Fig. 6.6
illustrates the continuing low R .
(vi) The next major outbreak in London (the last of the 16th century) was
in 1593. This epidemic began in the autumn of 1592 and is said to have
caused 2000 deaths before the end of the year. On 7 September,
Days
0
Grayes
Mowlson
Fuller
Dobson
Higham
Male
Harrolde
Davies
Francis
Adams
Smithe
Berrye
Worlington
Appres
servant
Thomas
wife
Robert
Wm
Richard
Agnes
Maria
Matthew
aunt
Eliz.
John
George
Wm
John
Stephen
Robert
Agnes
Margaret
George
Wm
Wm
163
20
40
60
80
100
120
P
S
S
S
S
T
T
S/T
S/T
S/T
S/T
T
T
T/Q
T/Q
T/Q
Q
Q
Q
Q/F
Q/F
F
F
Fig. 6.5. Suggested sequence of infections in the plague epidemic in the parish of St
Martin-in-the-Fields, London, in 1575. Scale: days after 1 April 1575. For further
details and abbreviations, see Fig. 5.6.
HOUSEHOLD
Days
0
20
40
60
80
100
120
(A)
Lether
Poole
van Oken
Knighte
Hollowaye
Fygge
Knight
Brayfeelde
Jackson
Stowton
P
d.
servant
P
m.
s.
servant
a Dutchman
S
s.
maidservant
Wm.
s.
s.
s.
maidservant
maidservant
s.
s.
d.
S
S
S
S
S/T
T/Q
T
Q
Q
Q
Q/F
F/Sx
F/Sx
Sx/Se
(B)
Cockson
Walker
Kevall
Greene
Barnatt
Hubberde
Knighte
Cowell
Johnson
Thorne
Sherington
Breese
Sherington
Wrighte
Walker
Walker
s.
P
d.
Margarett
s.
d.
John
f.
s.
s.
John
John
maidservant
Alyce, servant
Roberte
Richarde
s.
maidservant
S
S
S/T
T
Q
Q
F
Q
Q
F
F
F/Sx
Sx
Se
Se
Se
140
160
165
armed that St Katharine by the Tower was the parish where the
epidemic raged most violently; 525 burials were recorded in the registers in 1593 and Shrewsbury believed that the plague mortality rate
was at least 63%. By September, the disease was epidemic in
Greenwich and it radiated outwards through Middlesex into Essex,
Hertfordshire and Buckinghamshire. Mortality in London from the
plague of 1593 has been estimated dierently, but probably over
17 000 people died, i.e. 60% of the total of deaths from all causes.
There is little information concerning the symptoms shown by those
contracting the disease, but Creighton (1894) quoted a letter written by
Richard Stapes on 3 November 1593 commenting on the plague: my
son-in-law has buried his servant; but I cannot say his was the sickness
because the visitors reported that the tokens did not appear on him as
on the other. These tokens were probably not buboes but were the
characteristic marks on the body by which victims of the disease were
identied. Signs and tokens are discussed more fully in section 13.12;
unfortunately, we have no further information concerning the epidemic of 1593 but evidently not everyone dying at that time displayed
them.
We see from this summary that the plagues of 1563 and 1593 in London
had features in common that were also characteristic of the major plagues:
an explosive epidemic with Reed and Frost dynamics that reached its peak
in late summer, clear continuation through the winter months at a low
level, appearance of the diagnostic tokens, infection radiating outwards to
other communities and a high percentage mortality.
6.3 Plagues in central and southern England during the 16th century
During the rst half of the 16th century there are numerous anecdotal
accounts of scattered epidemics throughout England and, after 1550, many
parish registers record deaths from the pestilence or the plague. In
general, these outbreaks do not appear to have had their origins in a
London focus. In addition, many have searched the registers for years of
crisis mortality and some have ascribed these to outbreaks of plague,
although they were almost certainly the result of other factors (see section
1.5). Plagues in northern England in this century seem to have had dierent
origins and, in some cases, their rapid spread can be traced with some
accuracy. They are described in Chapter 7.
166
The universities of Oxford and Cambridge seem to have been remarkably unhealthy places, perhaps because the universities archives provide
more written evidence than in other towns, but there are repeated reports
of terms being curtailed, of the colleges being closed or even of the universities moving temporarily to another location. Sometimes these were precautionary measures.
The following is a brief, chronological summary of these various outbreaks of lethal infectious diseases in central and southern provincial
England, discounting epidemics of Sweating Sickness or of suspected typhus.
1503
1540
1544
1545
1546
1547
1551
1557
1558
1559
167
168
169
Fig. 6.7. Spread of the epidemic at Bristol in 1575. Schematic layout of streets in
which each square represents one house with the number of victims indicated. Data
from Slack (1977b).
170
J
62
F
55
M
47
A
43
M
42
J
24
J
18
A
47
S
26
O
15
N
18
mainly from a focus in the Midlands. It was grievous in Tewksbury with 560 plague burials registered; Creighton (1894) stated
that Canterbury, Nottingham and Lincoln were attacked; the
university at Cambridge was dispersed; Leicester was aected
by plague in September 1593, which continued until late in the
spring of 1594 (type (ii) epidemic which did not explode in the
second phase); plague was said to be particularly virulent in
Derby, lasting from October 1592 through the winter until
October 1593 (type (ii) epidemic), where there was not two
houses together free from it and yet it is reported that it never
entered the house of a tanner, a tobacconist or a shoemaker (see
Shrewsbury, 1970). Creighton (1894) armed that Licheld sustained a plague death-roll of 1100 and 174 of the inhabitants of
Bishops Castle, Shropshire, died of plague over a 21-month
period with a peak in August 1593. The plague also struck
savagely at Presteigne, some 16 miles south of Bishops Castle,
where 300 died, and at Gloucester where 81 plague deaths were
recorded in the registers of St Nicholas (see Shrewsbury, 1970).
171
172
FAMILY
Days
0
Degge
Perot
Walker
Bell
Lord
Ffenne
C. Smith
Tomson
Ange
Parepoynt
Sperpoynt
Lee
Bayles
Jackson
Pynsan
Bragge
Radman
Edge
Apryce
Symons
Gatly
Hatkinson
servant
wife
f.
d.
d.
wife
d.
d.
s.
s.
s.
s.
d.
wife
servant
servant
d.
d.
?
d.
f.
d.
d.
wife
d.
servant
d.
wife
f.
wife
s.
d.
d.
f.
d.
d.
?
w.
s.
wife
m.
s.
20
40
60
80
100
P
P
S
S
S
S
T
Q
S
S
S
S
S
T
Q
Q
S
T
T
T
S
T
S
S
T
T
S
S
T
S
S
S/T
S/T
S/T
S/T
S/T
T/Q
S/T
S/T
T/Q
T
T
T
Fig. 6.9. Suggested sequence of infections at the start of the plague epidemic at
Stratford-upon-Avon, 1564. Note the high R . Scale: days after 4 June 1564. For
further details and abbreviations, see Fig. 5.6. 8, female.
6.5 Conclusions
173
visited a town where the plague was just breaking out (plague was reported
at Licheld and Bristol in 1564; Creighton, 1894; Shrewsbury, 1970) and
returned to Stratford carrying the infection. Joanna Degge would not have
become infective until t : 21 (25 June), but by that time Gunne (who would
not have been showing symptoms) had already infected secondary cases in
the Perot, Walker, Bell, Lord and Ffenne families. Figure 6.9 illustrates
how the epidemic slowly got under way and spread through these families
with tertiary infections. Continued analysis of the burials in July and
August suggests that the two co-primary cases may have infected 1724
secondaries, with Oliver Gunne being responsible for the majority (R : 8
to 12).
6.5 Conclusions
We can draw few conclusions concerning these many diverse outbreaks of
lethal infectious diseases that were continually recorded in southern and
central provincial England during the 16th century. They seem to be
largely, but not entirely, conned to medium-sized and larger towns, but
this may be because we lack written evidence concerning the smaller
parishes and villages. However, if substantial outbreaks were conned to
populations above a critical size (i.e. haemorrhagic plague was density
dependent), as in the plague of northern England in 159798 (see Chapter
7), the epidemiology would be the opposite of authentic outbreaks of
bubonic plague in India, where it is essentially a rural disease of village
communities. The epidemics were generally explosive and probably followed Reed and Frost dynamics, lasting for less than 12 months, with a
peak mortality usually about late summer.
As yet, we have few details of how fast each of the epidemics spread
through the metapopulation nor which were the foci where the outbreaks
were initiated. Since, in many instances, the epidemics peaked simultaneously in late summer over a wide area (as in the Midlands in 1593), we
conclude that the disease must have spread with great rapidity and often
over substantial distances. In some instances mortality may have been
slight, perhaps some three to four times the seasonal average, but in some
towns perhaps 40% of the population died, with the bulk of the deaths
occurring within a 3-month period; this is not the epidemiology of bubonic
plague. There seem to be few studies of the age-specic mortalities in these
plagues. Were the deaths mainly in infants or in children or in the aged?
With this information it might be possible to eliminate weanling diarrhoea
or smallpox as the cause. However, we conclude that there is sucient
174
7
Plagues in the 16th century in northern
England: a metapopulation study
Plagues in northern England in the 16th and 17th centuries appear to have
dierent patterns and dynamics from the wave-like spread of the Black
Death and from the radial, or from the apparently erratic and unpredictable, movement of the epidemics in central and southern England that were
caused by the movement of infectives over substantial distances. Usually
the spread of the infection in northern England can be monitored from the
records and, frequently, the infections appear to move southwards along
well-dened corridors in the northeast from the Scottish borders and
Northumberland. The Pennines, which form the backbone of England,
eectively divided the Northern Province into eastern (Northumberland
and Yorkshire) and western (Cumberland, Westmorland and Lancashire)
halves, each with dierent terrains, and an epidemic disease was brought
across by infectives travelling on the roads through the gaps.
Thus, the Northern Territory acted as a separate metapopulation, semiisolated from Scotland and the rest of England. As we have seen, the
inhabitants of northern England had been much occupied in defending
themselves from raids and cattle stealing by the Scots, and Carlisle in the
northwest and Durham in the northeast were major defensive centres. The
terrain, particularly to the west of the Pennines, was very dierent from
that of central and southern England and this was an important determinant of the dynamics of the epidemics. The counties of Cumberland and
Westmorland have been described as backward and impoverished (Appleby, 1975) and the area remote from large industrial and trading centres;
much of it was inaccessible to travellers, and all of it regarded with
repulsion by outsiders (Thirsk, 1967); in addition, these counties were left
almost untouched by the various agricultural revolutions that spread
across the rest of the country during the 16th and 17th centuries. The
economy of the north was more like that of Scotland (where mortality
175
176
crises and harvest failures persisted up to 1690), Ireland and parts of the
Continent, rather than that of lowland England. The region suered mortality crises in 1587 and 159697 because of the synchrony of high wheat
prices and low wool prices, which led to extreme hardship and famine
(Scott & Duncan, 1998). The population in the extreme northwest, therefore, already subsisting on an inadequate diet, were further weakened and
malnourished by the famine of 159697 when the plague struck.
To the east of the Pennines, communications and conditions for farming
were better in the coastal plains of Northumberland and Durham, in spite
of the northern latitudes, cold winter climate and raids by the Scots. The
more intensive farming on the eastern side of the Northern Province was
accompanied by attempts to stabilise the distribution of land; common
elds were associated with village settlements and it appears that the land
was used fairly intensively (Thirsk, 1967), in contrast with the situation in
Cumbria. To the south lay the vast area of Yorkshire, the bulk of which lay
in the lowlands and much of the husbandry and many of the communities
resembled those of the Midlands and southern England (Thirsk, 1967).
Plague occurred much more frequently along this eastern corridor than it
did in the wilder country to the west of the Pennines.
177
while a second report, early in December, armed that it was still active at various
places in the two counties.
178
The parish registers of Penrith record that Plauge was in Penreth and
Kendall 1554; this single line is much-quoted but there appears to be no
other account nor any further details of this epidemic and Barnes (1891)
commented on the absence of local records of visitations of the plague in
Cumberland and Westmorland.
South Lancashire was badly aected by an epidemic in 1558 (Axon,
1894) that was described by Sharpe France (1939) as follows:
Liverpool in particular was badly hit, about a quarter of the population being
killed. The Town Books say that the previous year had also been a bad one. There
was a great plage in Manchester, and the authorities of Liverpool were very
concerned over the danger of its spreading to their town. Unfortunately it did so,
and an unlucky Irishman, with the Welsh name of John Hughes, was held to be
responsible. He was accused of having been ill when he arrived in the town from
Manchester and of having taken his dirty clothes to be washed at the house of a
certain Nicholas Braye. A child of Braye contracted the disease and died. Hughes
was brought before the mayor and underwent severe questioning, but the accusations could not be proved. However, whether he had carried the plague, or it had
come by some other channel, several others in the same house died shortly after,
and so after that it increased daily and daily to a great number, that died between
St. Lawrences day [10 August] and Martlemas [11 November] then next after, the
whole number of 240 and odd persons. The severity of the outbreak was such that
the St. Martins fair was cancelled and no market was held during a period of three
months.
179
the etiology and epidemiology of this outbreak is not clear. After the entry
for 17 November 1577 the Hawkshead registers recorded entered in November: In this monthe begane the pestelent sicknes in or pishe [our
parish] wch was brought in by one George Barwicke but it was not until
27 December 1577 (40 days later) that the burial of George Barwicke wch
brought in the sickness was recorded. It is dicult to reconcile these
records with the usual pattern of events in the plague where the incoming
infective is the rst to die. The rst plague burials on 19 and 24 November
1597 were of Elsabeth and Richard Barwicke, respectively, who look like
co-primaries, and on 11 and 15 December Edwine and Richard Barwicke,
respectively, were buried, apparently co-secondaries. George Barwicke
may have been a later secondary or a tertiary infection. The arrival of an
infective from outside and the temporal spread of the epidemic at Hawkshead are broadly consistent with plague epidemiology and, possibly,
George Barwicke returned to the parish with two of the members of his
family who were then already infected (Elsabeth and Richard), whereas
George was infected by his relatives and died 6 weeks later.
The plague at Hawkshead began with the two deaths in November 1577
and continued through the winter until 25 February 1578 when the epidemic nished. There was a total of 38 deaths, the majority in December
and January, when the outbreak was at its peak, with 16 burials in each
month; a type (ii) epidemic which did not are up again in the spring. Only
seven families were aected and, although family reconstitution is not
possible (only brief details are given in the registers), analysis suggests that
some of these with the same surname were related and visiting one another
but were in dierent households: 12 Tomlinsons died.
The registers also include the note Anthony Dixson buried in Langdall
the last day of December & taken up agayne & brought to Hauxhead the
xjth day of January and Shrewsbury (1970) gave a quotation that plague
was present in the valleys of the Lake District in 1577.
An epidemic erupted again in Newcastle-upon-Tyne in May 1588:
and in the succeeding months destroyed 1,727 of the townsfolk. The register of St
Johns parish records that the plague-dead were distributed among 5 burial
grounds, to wit, 340 in St Johns; 509 at the Chapel; 300 at Allhallows; 400 at St
Andrews, and 103 at St Nicholas. These gures add up to 1,652 burials, leaving 75
corpses which were presumably interred in ground that was afterwards consecrated.
The plague struck at Durham City in the following year, 1589, and,
although Creighton related that huts for the accommodation of the plaguesick were erected on Elvet Moor outside the city, the outbreak seems to
180
FAMILY
Days
0
20
40
60
80
100
120
Maysterman f. Robert
P
m. Margarete
P
s. Adam (5)
P
d. Eliz. (8)
P
d. Thomasin (6)
P
d. Jane (3)
P
S
Poor woman
Kearston
s. Richard (4)
Rolande
S
T
Caston
Margaret
Barton
Anthony
Mydfurth
Phylles
Frend
Alysse
Alyson
Ann (17)
Rutter
Myles (1)
Rowle
Jane
Whyte
Elyzabethe
Ferelesse
Thomasyn (1)
Thomas
Wm
S
T
Q
Q
Fig. 7.1. Suggested sequence of infections in the plague epidemic in the parish of St
Oswald, Durham, 1589. Scale: days after 14 August 1589. For further details and
abbreviations, see Fig. 5.6.
have been mild. The registers for St Mary-le-Bow, Durham, did not record
any plague deaths for 1589; indeed only nine burials are given for that year.
Only 21 died of the plague in St Oswald, Durham, between 20 September
1589 and 1 January 1590 so that, in this late-starting epidemic, the infection
did not continue through the winter. Only 13 families were aected and the
progress of the epidemic through the initial families and their probable
contacts are shown in Fig. 7.1. Robert Maysterman and his wife were both
buried on 20 September, with their son Adam buried on 26 September and
four daughters buried between 26 and 29 September. All the members of
the Maysterman family were co-primaries, being infected by a common,
outside source around 14 August (R : 6) so that Robert and his wife
181
became infectious around 29 August and so may have been able to carry
the infection to the poor woman and the Kearston family (Fig. 7.1), who
became the rst secondaries. Other co-secondaries were then established in
the Caston, Barton, Mydfurth, Frend and Alyson families, so that 6 coprimaries infected 7 secondaries (R : 1) who, in turn, infected 5 tertiaries
(R 1) (see Fig. 7.1) who in turn infected only 2 quaternaries (R 1) and
hence the epidemic died out. The estimated parameters for St Oswald were:
latent period : 12 days, infectious period : 25 days (including the last 5
days when the victim was showing symptoms).
From the summer of 1589, apart from the usual winter intermissions,
there was a progressive expansion of the area of activity of the disease,
culminating in the great national outburst of 1593. Thus, in the spring of
1590, Morpeth and Alnwick in Northumberland were said to be infected
with the sickness of the plague, and in 1593 Durham and York were so
sore with the plague that none of worth remained in either city because
there was the usual panic exodus of all who had the means to ee.
182
Fig. 7.2. Geographical spread of the plague from northeast to northwest England,
159798. Dashed line, land over 650 feet; dotted areas, land over 1640 feet. Circles,
market towns; triangles, smaller communities. Note movement of plague across the
Stainmore/Bowes Moor gap. The possible introduction of the plague via the port of
Newcastle is indicated. A, Appleby; C, Carlisle; Da, Darlington; Du, Durham;
Dum, Dumfries; K, Kendal; N, Newcastle; P, Penrith; R, Richmond.
dyed at Kendal . . .. London, on the other hand, was clear of plague at this
time.
On 26 May 1597 the Dean of Durham again complained that there was a
great dearth in Durham: on some days, 500 horses are in Newcastle for
foreign corn, although that town and Gateshead are dangerously infected.
On the 17 September, Lord Burghley, minister of State, is informed that the
plague increases at Newcastle, so that the Commissioners cannot yet come
thither (the Assizes were not held at all, on account of the plague at
Newcastle and Durham); foreign traders were selling corn at a high price
until some members of the town council produced a stock of corn for sale at
a shilling a bushel less. There are no gures of the plague mortality at
Newcastle in 1597, but at Darlington the total deaths by October 17 were
340.
7.4 Durham
Shrewsbury (1970) estimated that the plague burials in Durham in 1597
were as follows:
183
He regarded these as minimal gures and many of the burials were on the
moor. To this total can be added 82 burials recorded in the registers of St
Mary-le-Bow during the period July to October 1597, although none of the
entries are marked with a P, for plague; inspection of the registers conrms that, compared with the preceding and following months, this was an
enormous mortality, suggestive of an infectious disease. However, in only
two families does more than one person die, which is completely at variance with the high household contact rate often shown in plague epidemics.
In general, more adults than children seem to have died, although in the
Taylor family the sequence of burials was as follows: wife (husband survived), son aged 5 years, son aged 9 years, sister-in-law, son aged 1 year.
Creighton (1894) stated that the infection broke out again at Darlington
and Durham in September 1598.
184
Deaths nally ceased in both by December 1598 (type (ii) epidemic). The
populations of both towns suered grievously, with many victims having
to be buried on the fells.
7.7 Carlisle
185
7.7 Carlisle
Again there are no reliable records of the number of people that died of the
plague at Carlisle because none of the local registers exist for that period.
However, Hughes (1971) has examined a number of documents relating to
the period of the plague and he has presented an interesting synthesis that
gives a description of events in the town during the epidemic. A census was
taken on 20 December 1597 of the city householders with an indication of
those families that were visited by the pestilence: the number of households
was 323, of which 242 were stricken by plague, which had certainly not run
its course by that date. Hughes estimated the population of Carlisle to have
been about 1300 at that time.
It is dicult to deduce the lethality of the disease at Carlisle from the
documents that Hughes has uncovered. After saying that 242 households
were visited he added that gures against many of the names may indicate
the number of deaths in each household. If this assumption is correct it
would give a total of 149 . . .. Slack (1985) assumed from this that the
plague at Carlisle had a low mortality rate but a high morbidity rate Only
149 people died . . . but three-quarters of the towns households, 242 out of
323, were infected. If the disease did have such a low mortality rate, the
etiology would be completely dierent from that at Penrith and would be
completely at variance with the tablet in Penrith church where 1196 are
described as dying at Carlisle, although this is certainly an exaggerated
number; furthermore, this census at Carlisle was taken in December 1597
when the plague had only just begun.
Hughes (1971) described the precautions taken by the civic authorities: a
City Council meeting was apparently held on 3 November when it was
recorded necessarye observations thought meate to be kept in this Cittye,
the third day of November 1597: for the avoydinge of further infection of
the disease of the plague then suspected there to be, if so it pleace God to
blesse there carefull indeavours therein . . . Infected houses were sealed o;
the provisioning of their inhabitants arranged for and orderly arrangements made for the removal and disposal of the dead. Daily visits were to
be made by honest experienced men to discover cases of sickness. One of
the resolutions laid down was that a weekly collection must be taken in
each street for the better relief of every poor person visited; of the total of
209 9s 10d, the amount raised by the citizens themselves was only 14 4s
10d; the largest amount came from the Common Chest and donations were
received from several county gentry. The poor were attended without fee or
charge for medicine, but those who were in a position to pay were expected
186
to do so. The city gates were placed in charge of honest men whose orders
were to prevent the admission of anyone known or suspected of infection
or who came from any place where the infection was thought to be.
Foreigners and wandering beggars were expelled from the city and during
the visitations none from Rickergate, Caldewgate or Botchergate were
admitted without a permit from the city baili. Movement within the city
itself was also restricted. Arrangements were made to pay the stipends of
the ocers and ministers, of the corpse bearers and the corpse winders and
viewers, the latter apparently receiving a at rate of ten shillings per week.
A similar sum was paid to those who cleansed houses where all the
inhabitants had died, or had ed to the elds for safety. Help was also given
to those of the poor who survived though in daily contact with the sick, and
to such as had recovered from the plague.
This account supports the conclusion derived from the Penrith study
(Chapter 5) that not all those who came into intimate contact with the
infection succumbed and contracted the disease. It also suggests that
recovery from this infection was possible. The infectious nature of the
disease was fully appreciated and infected houses were marked, as usual,
with a red cross there to continue until lawfull opening of the same house.
Forty days were considered to be the period of quarantine (as usual), which
corresponds well with the estimate of 37 days from infection to death.
It is not known when the practice of removing suerers to pest-houses
began, but other properties were commandeered to deal with the emergency and several isolation hospitals were speedily built outside the city
walls. These shelters, variously termed lodges or shields, were situated at
Gosling Syke, Stanwix Bank and others directly under the city wall on the
Bitts. In each case the shields were sited near a water supply and the choice
of a site at Gosling Syke and St Lawrences Well may have been inuenced
by the fact that at both these points the limit of the cultivated elds was
reached and the moorland waste began. Strict orders were enforced regarding the burial of the dead: special biers were to be provided for carrying the
corpses, which had to be buried between 10 a.m. and 4 p.m.; no corpse was
to be lifted until the bellman gave word that the grave had been prepared
and the beadle had to walk before those carrying the body to give people
warning as he came (Hughes, 1971).
187
1598
M J J A S O N D J F M A M J J A S O N D J
Newcastle
Durham
Darlington
Brough-under-Stainmore
Warcop
Appleby
Richmond
Kendal
Edenhall
Penrith
Carlisle
Dumfries
Fig. 7.3. Timing and duration of the plague epidemics, 159798. Arrows indicate
probable spread of the disease, although Kendal and Carlisle may have received the
infection directly from Richmond rather than via Penrith. Solid lines indicate the
duration of epidemic.
aicted market centres (Fig. 7.3). Edenhall, a small parish to the east of
Penrith, appears to have suered a number of plague deaths and the
following note appears in the parish register: These 4 next following dyed
of the plaige, Itm vii . . . Pattrig Rowtlishe was buried wthn Flatts wall
neare to his own house being knowne to dye of the plaige. The death of his
wife on 8 March; his servant Elizabeth Thompson on the 11th and his
infant son John immediately followed. The rst is entered as having been
buried beside her husband near the said place, and the last was buried
beside his father and mother in said place. This seems to have been an
isolated outbreak of the disease. No further deaths from it are recorded
until the end of the following July, 1598. A baptism is recorded on 24
188
March. There was another baptism on 25 April, three burials between this
date and August, and then at the head of the next page is this entry: The 42
next following dyed of the [word wanting].
The rst plague death was on 29 July 1598. Some families suered
severely as shown by the following entries: Itm First August one child of
Andrew Atkinson of the plaige & was buried in atts cloose. Itm xv & xvi
August Andrew Atkinson wie iii other children dyed of plaige and were
buried their Lodge on Edenhall Fell at a place called Shaddowbourgh.
Twenty deaths occurred in August, and 11 in September. Some were buried
in the churchyard and others on the backside of their house, on Penrith
Fell, or Flatts cloose (see Barnes, 1891).
Dacre, to the west of Penrith, escaped unscathed, although a note in the
register for April 1598 states that they were aware of the plague in Penrith.
At Warcop, the disease appeared to be conned to one part of the parish,
and to only two families. Adam Mosse and his two children died of the
plague as it was thought on 19 October 1597 (presumably co-primaries).
On 4 November 1597 Margaret Mosse and Agnes Lancaster (presumably
co-secondaries) were buried in a garth at Blatarne. The next burials occur
on 25 May 1598 (presumably a fresh outbreak), when Richard Lancaster
and his wife died both so daynelye upon the plague as it was thought and
were buried in their own yeard at Blatarne. Another entry on 6 June 1598
stated Dyed Thomas son of Richard Lancaster of Bletarne and the barne
wherein he died burned and the corps afterwards interred. Although
plague was not given as the cause of death, the fact that purging by re was
deemed necessary would infer a highly infectious and much-feared illness.
The disease did not gain a strong foothold at Brough-under-Stainmore;
only eight deaths occurred at rst, albeit in 20 days, with seven coming
from the family of Abram Wharton. His daughter died of plague on 10
November 1597 and, after 18 days, a son died and another son 2 days later.
One day later, Abram, his mother, daughter and maid were buried. There
were no further deaths from the plague until between 6 and 30 July of the
following year (presumably a fresh infection), when seven more fell victim,
of whom ve bore the same surname.
The parish registers at Penrith record that the plague also struck at
Appleby and the inference is that it was a severe attack to merit inclusion
with Kendal and Carlisle. The outbreak apparently occurred late in the
pandemic and probably the infection was again spread from Richmond,
but might have returned eastwards from Penrith. Between 1 August 1598
and 25 March 1599, 128 persons died at Appleby, Scattergate, Colby and
Colby Leathes, and it is inferred that death was because of plague. Appleby
189
190
under my charge, were infected therewith, yet had we all of us, the cause thereof
within our sinfull hearts, as well as any others.
(see Wilson, 1975)
7.10 Conclusions
191
8
Plagues in London in the 17th century
193
Fig. 8.1. Plague deaths in London, 15781680. Data from Creighton (1894).
194
Fig. 8.2. Plague deaths in London, 15781680. Ordinate: log of the number of
deaths. Data from Creighton (1894).
The distribution and intensity of the 1603 epidemic were uneven in the
metropolitan parishes, with a wide variation in mortality rates that ranged
from 19% in St Antholin, Budge Row, to 94% in St Pancras, Soper Lane.
Shrewsbury also calculated the plague mortality rates for Stepney (95%),
Newington-Butts (90%), Islington (85%), Lambeth (97%) and Hackney
(88%) and concluded that the epidemic was much more deadly in these
overcrowded London suburbs than in the city itself. We have analysed the
epidemic of 1603 in the parish of St Helen, Bishopsgate, where the rst
plague victim was recorded in the registers on 25 June and the last on 1
February 1604 and during this time 103 people were buried. The start of
the outbreak in the rst 32 households is shown in Fig. 8.4: once again, the
servants formed a high proportion of the deaths. There were three coprimaries, with the daughter in the Marshall family being infected rst on
19 May 1603. The spread of the epidemic was slow at rst: the three
co-primaries probably infected six secondaries (R : 2) who, in turn infec
ted 13 tertiaries (R : 2). Thereafter the epidemic suddenly exploded
195
Fig. 8.3. Weekly plague mortality in London, May to December 1603. Abscissa:
weeks after 12 May 1603. Data from Creighton (1894).
196
HOUSEHOLD
Days
0
d.
P
servant
P
servant
servant
Dregghe
d.
P
s.
Ward
s.
?S
servant
?S
f.
Staniere
servant
S
Tayler
s.
f.
Tedder
servant
s.
d.
Chamber
m.
Dermer
servant
Vandeveld s.
Burges
s.
Atkinson
servant
s.
Lupton
George, skinner
Price
Gryffin
Dixon
servant
Sheppard
s.
s.
m.
Writing
wife
Curtis
d.
Winterborn servant
Letherland servant
Gray
servant
Scottyn
servant
Dod
d.
Shaw
servant
d.
Millington
servant
Maunder
servant
Thornet
Thomas
Ledills
a nurse child
Hilton
servant
servant
Christian
s.
Richardson d.
Fenner
s.
20
40
60
80
100
120
Marshall
Gramer
S
S
S
?T
T
T
T
Q
F
T
T
T
T
T
Q
T
T
T
T
+2
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
+1
+1
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Fig. 8.4. Suggested sequence of infections at the start of the plague epidemic in the
parish of St Helen, Bishopsgate, London in 1603. Scale: days after 19 May 1603. For
further details and abbreviations, see Fig. 5.6.
+1
+1
+2
+1
+1
197
198
prevent the spread of the plague in the counties of Middlesex and Surrey.
The King levied a special rate on 10 villages in Kent on 20 July 1603 to
relieve the suerers in a grievous plague; Creighton contended that such
rates were usually levied when an epidemic was nearly over and concluded
that the outbreak in Kent must have been at least as early as that in
London. Perhaps the plague in 1603 was not imported directly into the
ports of London but came from Kent.
Thus Kent became very generally infected and by October 1603 few
towns on the road from London to Dover were free from it. It spread
elsewhere in the Home Counties: Surrey and Sussex were aected and in
Essex it swept o great numbers in Colchester. The plague spread westwards along the Thames Valley to Oxford and also moved southwest to
Wiltshire where an order directed that a watch be kept in every town and
village in the county for the arrest of all vagrants who were to be summarily
ejected if they were suspected to be dangerous of infection.
Plague was very active in England during the period 16026, with major
foci in the Midlands and the northern counties, as we describe in section
9.1. It is probable that the disease exhibited changes in its epidemiology
after 1600; not only was the mortality very much greater than before, as in
London in 1603, but isolated epidemics in towns and cities changed to
pandemics in the metapopulation, as presaged by the pandemic of northern England in 159798 (section 7.3). These pandemics persisted over
winter for 3 or more years and their spread can be monitored along the
communication routes, particularly the roads (see section 13.9).
199
continue because of the low R . Only one secondary was infected (within
the Kinge family) producing an eective contact rate below 1 (0.25). Only
one tertiary case is recorded (R : 1) but she (a daughter in the Hoare
family) infected seven quaternaries (R : 7) who, in turn, infected eight fth
generation victims (R : 1), the last dying on 31 August. Thereafter, the
epidemic zzled out quite quickly because of the low R . The servant in the
Ricardsonne household was buried on 15 October but he had infected a
servant in the Nuame household who died on 31 October (not shown in
Fig. 8.6). Three months later, the maid servant in the Smithe household was
recorded as a plague burial on 1 February 1607 and must have been a
reinfection. She probably infected the servant in the Halye family who died
exactly 37 days later. Again, the plague struck particularly ercely at
servants.
There were also four plague burials in St Mary Somerset in autumn 1607
(one primary, two secondaries, one tertiary).
HOUSEHOLD
Days
0
Goodfellow
Barine
Harvey
Felps
Kinge
Wm
Patricke
servant
Wm
s.
d.
Hoare
d.
Dobbins
servant
Tinton
Elizabeth
Rutter
servant
servant
servant
servant
Obie
s.
m.
Jones
Rowland
Hayward
servant
Bray
s.
Broughton w.
Barham
d.
Oxford
servant
Griffin
d.
s.
Ware
d.
Welsh
lodger
servant
servant
Wattes
s.
Daye
d.
Hendry
Wm
Coxseter
Thomas
Jonnes
wife
Boyden
d.
Ricardsonne servant
20
40
60
80
100
120
140
160
180
200
P
P
P
P
S
F/Sx
T
Q
Q
F
Sx
Sx
Sx
Q
Q
Q
Q
Q
F
F
F
F
F
F
F
Sx
Sx
Sx/Se
Se
Se
Se
Se
E
E
Fig. 8.6. Suggested sequence of infections in the parish of St Mary Somerset, London, in 1606. Scale: days after 21 March 1606. For
further details and abbreviations, see Fig. 5.6. Sx, Se and E, sixth, seventh and eighth generation infections, respectively.
201
Fig. 8.7. Weekly plague deaths in London, March to December 1625. Abscissa:
weeks after 17 March 1625. Data from Creighton (1894).
exceptionally ne; January 1625 was warm and mild; spring was described
as wholesome but the early summer was extremely cold. June 1625 was a
month of ceaseless rain and both the hay harvest and corn harvest were
spoilt.
Figure 8.7 shows the weekly plague deaths in 1625 that Creighton (1894)
believed demonstrated how the epidemic increased after the rains in June,
but the data suggest that the epidemic followed the usual seasonal pattern:
there was a very slow build-up with 2 plague deaths in January and 12 in
February and weekly plague deaths thereafter did not exceed 30 until May
12 (Fig. 8.7) when the epidemic exploded and reached its peak in August
when over 4400 deaths were recorded in one week.
The Lord Mayor received the following reprimand from the Privy
Council in March 1625, which suggests that plague was already a serious
problem this early in the year:
202
We understand that the plague doth daylie encrease in the citty and that ther dyed
this last weeke seven of it in one parrish and although it hath beene thus encreasing
divers weekes yet wee cannot heare that any good course hath beene taken for
preventing it either by carrying the infected persons to the pesthouse or setting
watch upon them or by burning of the stue of the deceased which being of little or
noe value might easily be recompensed.
Seven deaths in one parish suggests a greater death rate than that given in
the Bills of Mortality (Fig. 8.7).
Shrewsbury (1970) has extracted from eight parish registers in London
details of the spread of the disease in 1625 in 17 selected families with a total
of 80 burials, and inspection of the data shows clearly that its progress
follows exactly the pattern that we have demonstrated in our analyses of
communities in London and the provinces, with a latent period of 1012
days.
The analysis of the start of the remarkable and severe epidemic at St
Martin-in-the-Fields in 1625, when some 191 persons died, is shown in Fig.
8.8. The epidemic began here explosively and 179 of the victims died
between 4 July and 8 August. We have identied at least 34 apparent
co-primaries (Fig. 8.8), presumably the result of multiple infections from
adjacent parishes. This epidemic at St Martin-in-the-Fields was also characterised by a low household R , since there were few secondary infections
within the families. The outbreak disappeared as quickly as it had begun
and there were no plague burials after 17 August.
Of the symptoms of the 1625 plague in London, Shrewsbury (1970)
quoted a correspondent: The physicians do in a manner agree that this
sickness is not directly the plague, as not having any sore [our italics] or any
such like accident, but only contagious in blood or kindred; he deduced
that the victims did not show buboes and suggested that their absence was
due in all probability to the extremely high virulence of the responsible
strain of Pasteurella [ : Yersinia] pestis at the start of the 1625 epidemic.
A story tells of a woman who ed to Croydon and looking back on
Streatham Hill said farewell plague but soon after was taken sick, had
these tokens on her breast [our italics] . . . (see section 13.12).
The Privy Council on the 21 October 1625 censured the undiscreet and
Fig. 8.8. (opposite) Suggested sequence of infections at the start of the plague
epidemic in the parish of St Martin-in-the-Fields, London, in 1625. Scale: days after
28 May 1625. For further details and abbreviations, see Fig. 5.6.
Days
0
A poor boy
Turner
Barrett
A poor man
Knight
Smith
Dickson
Andrewes
m.
f.
f.
d.
203
20
P
P
P
P
P
P
P
w.
d.
P
a young man
P
d.
P
s.
P
s
S
f.
S
Jackson
s.
P
Kitchen
w.
P
Latham
gravemaker
P
servant
S
Bayly
d.
P
d.
S
d.
S
d.
S
Rankins
servant
P
servant
P
Frith
f.
P
Keele
m.
P
Robbinson f.
P
d.
S
Holts
a young man
P
servant
S
f.
S
Mow
James
P
Willey
a Scottishman
P
a young man
P
Vile
one of ye bearers
P
Smith
Catherine, a stranger P
Browne
d.
P
d.
S
Makins
w.
P
Mallowgh
servant
P
Marshall
maid
P
Grantham Robert
P
Morley
m.
P
child
S
a woman
S
Parre
d.
P
Wingatt
s.
P
Saunderson s.
?S
d.
?S
infant
?S
Wright
s.
?S
Willbreed
m.
?S
Baker
f.
?S
s.
?T
d.
?T
40
60
204
Creighton concluded from this that the buboes and boils might come out
more than once and that the best chance of survival lay in their suppuration. In June 1625, Lord Russell being to go to Parliament had his
shoemaker to pull on this boots, who fell down dead of the plague in his
presence.
Again, dogs were destroyed: the parish of St Margaret, Westminster,
paid 2 17s 8d for the slaughter of 466 dogs.
By July 1625 the disease was universally distributed in London and had
spread to other parts of the kingdom, probably augmented by the ight
into the country of all those Londoners who could do so, the usual
response to a major epidemic. In mid-September the Tuscan Resident
reported from his refuge near Bedford that almost as many people were
dying of plague within the circuit of three miles from London as within the
capital itself, and that the disease was so widespread throughout the
kingdom that it was impossible to go anywhere without a danger of
contracting it (Shrewsbury, 1970).
However, the people eeing from the plague in London met with a poor
reception in the country towns and villages, as in 1603. Creighton described how They are driven back by men with bills and halberds, passing
through village after village in disgrace until they end their journey; they
sleep in stables, barns and outhouses, or even by the roadsides in ditches
and in the open elds. And that was the lot of comparatively wealthy men.
A stranger from London arrived at Southampton on 27 August and died in
the elds; he had a good store of money about him which was taken before
he was cold. The Dean of St Pauls wrote:
The citizens ed away as out of a house on re, and stued their pockets with their
best ware, and threw themselves into the highways, and were not received so much
as into barns, and perished so: some of them with more money about them than
205
would have bought the village where they died. A justice of the peace told me of one
that died so with 1400 about him.
(Creighton, 1894)
Sir John Coke sent a report to Lord Brooke on 18 October 1625 that
seems to imply that by then the disease was losing its virulence: We are full
of hope that God beginneth to stay his hand, because now in London the
tenth person dieth not of those that are sick and generally the plague seems
changed into an ague.
The plague spread throughout the Home Counties and also through the
counties bordering the English Channel, Kent, Sussex and Hampshire, by
July. Shrewsbury averred that the plague moved from London to the West
Country via Wiltshire, where it had already arrived by 5 August. It broke
out in Oxford by the end of July and in East Anglia by the end of June.
Indeed, as we describe in section 9.3, plague seems to have been widespread
in England in 1625 and this continued into the following year, although
Creighton (1894) rather belittled this provincial pandemic. The general
belief seems to be that the epidemic in London spread widely and remarkably rapidly from this focus.
8.4 Recovery of the population of London after 1625
There were 54 000 deaths registered in London in 1625, 35 000 of them
because of plague, but these gaps were rapidly lled, presumably largely
because of immigration. By 1627 baptisms were again at 8408, having been
8299 in the year before the plague. In 1629, baptisms exceeded burials by
more than 1000 and continued to be slightly in excess until the next plague
of 1636. London in 1625, therefore, is another example of a population that
recovered remarkably quickly after a mortality crisis.
The population of London in 1625 was of the order of 300 000
(Creighton, 1894; Shrewsbury, 1970) so that the 35 000 deaths represent
approximately a 12% plague mortality, although with so many eeing
from the capital and possibly dying elsewhere, the proportion of those
staying behind who died would be higher. Nevertheless, the percentage
plague mortality in London would have been much less than in some
outbreaks in the provinces. For example, in Penrith in the 159798 epidemic some 45% of the population died (section 5.2), although a smaller
proportion of the citizens of Penrith may have had the opportunity or the
means to have ed from the town. Does this suggest that the non-immigrant population of London had some degree of immunity because of
repeated exposure to the disease?
206
Fig. 8.9. Weekly plague deaths in London, April to December 1636. Abscissa: weeks
after 7 April 1636. Data from Creighton (1894).
207
HOUSEHOLD
Days
0
Cvsenes
Ewere
Fvggvs
Fellepes
Farar
Bradshae
Lowes
Clarke
Danell
Oddesle
Slege
Parie
Warine
Brane
Nowne
Harte
Mansfeld
Fell
Prise
Griffen
Dellen
Jenkines
Write
d.
P
d.
s.
d.
Anne
m.
John
Wm
s.
d.
d.
Hvee
s.
f.
John
Roger
s.
Richerd
m.
d.
d.
d.
d.
d.
f.
d.
f.
d.
s.
s.
wife
d.
Frances
20
40
60
80
100
120
140
160
180
S
S
S
S
S
S/T
T
T
Q
Q
Q
Q
F
Q
F
Sx
F
F
F
F
F
F
F
F
Sx
Se
Sx
Sx
Se
Se
Se
Se
Fig. 8.10. Suggested sequence of infections in the plague epidemic in the parish of St Mary Somerset, London, in
1636. Scale: days after 30 June 1636. For further details and abbreviations, see Fig. 5.6. Sx and Se, sixth and
seventh generation infections, respectively.
209
the sixth day after they were taken, three of them dyed in three hours, one after
another, and were all buryed in one grave, and about two days after the two
youngest both died together, and were buryed in one grave. All this while I lay sick
in another bed, and the tender looked every hour for my death; but it pleased God
most miraculously to preserve me, and without any sore breaking, only I had a
swelling in my groin, which it was long ere it sunk away, and I have been the worse
for it ever since, and when I was recovered, I was shut up with two women, one man,
and one child for three months, and neither of them had the distemper.
HOUSEHOLD
Days
0
Earith
Middleton
Gill
Stevenson
Smales
Reinolds
Perridge
Atkins
Knowles
servant
servant
d.
servant
servant
servant
m.
f.
d.
d.
s.
d.
w.
f.
d.
d.
s.
servant
d.
servant
s.
20
40
60
80
100
120
140
160
180
200
P
S
S
S
S
S
T
?Q
?Q
?Q
F
Q
Q
Sx
Se
E
E
Se
Se
Se
Se
Fig. 8.11. Suggested sequence of infections in the plague epidemic in the parish of St Michael Bassishaw, London, in 1641.
Note low R and slow spread of this small epidemic, including a 36-day gap between two burials. Scale: days after 28 April
1641. For further
details and abbreviations, see Fig. 5.6. Sx, Se and E, sixth, seventh and eighth generation infections,
respectively.
211
contact, even in summer, which extended its duration, and by an apparently short latent period for infections within the family.
212
that Y. pestis was introduced by this route. This is most unlikely; the death
was a solitary one and there were only ve other sporadic and apparently
unconnected plague burials in December 1664. Of these six deaths, three
were in Whitechapel and the other three were in separate parishes. There
was one further death in mid-February until the plague began at the end of
April. Harvey (1769) described this latent period as follows:
And being restrained to a house or two, the seeds of it conned themselves to a hard
frosty winter of near three months continuance: it lay asleep from Christmas to the
middle of February, and then broke out again in the same parish; and after another
long rest till April, put forth the malignant quality as soon as the warmth of spring
gave sucient force, and the distemper showed itself again the same place, where it
was rst: neither can it be proved that these ever met; especially after houses were
shut up.
Shrewsbury explained this account as being a description of the epidemiology of bubonic plague (although he attributed the February death
to typhus because he could not accept that the eas could be active in the
depths of winter), but it is unnecessary to take any account of these cases,
which probably did not presage the epidemic that began in 1665: plague
had been endemic in London for much of the 17th century, with a dozen or
up to 1000 deaths in most years. However, this account does illustrate how,
in non-epidemic situations, the disease could strike in winter in an apparently random and sporadic way, with nobody else dying, although some
may have contracted the illness and survived.
Weekly registered plague deaths reached a total of 43 at the beginning of
June 1665 and the rst ocial notice of the outbreak in London was a
proclamation on 14 June cancelling Barnwell fair for fear of spreading the
plague. Pepys wrote in his diary for 15 June The town grows very sickly,
and people to be afraid of it.
The plague broke out in the parish of St Giles-in-the-Fields and moved
from the western and northern suburbs towards the City, the eastern
suburbs and Southwark, the reverse direction to its usual progress.
Creighton (1894) quoted Boghurst, an apothecary, who practised in St
Giles-in-the-Fields as follows:
The plague fell rst upon the highest ground, for our parish is the highest ground
about London, and the best air, and was rst infected. Highgate, Hampstead and
Acton also all shared in it. From the west end of the town, Boghurst continues, it
gradually insinuated and crept down Holborn and the Strand, and then into the
City, and at last to the east end of the suburbs, so that it was half a year at the west
end of the city [in his experience] before the east end and Stepney was infected,
213
which was about the middle of July. Southwark, being the south suburb, was
infected almost as soon as the west end.
This account (if Defoe is to be believed) gives an insight into the epidemiology of the plague; it spread gradually and inexorably, each parish in
turn experienced a ash epidemic that quickly burnt out. Its intensity was
nearly over in one place before it had begun in another and Creighton
(1894) regarded this as the most interesting epidemiological feature of the
epidemic.
The explosive start of the major epidemic in the parish of St Michael
Bassishaw in 1665 is shown in Fig. 8.12. Plague burials were recorded in the
register from 17 June to 7 January 1666. The single primary case, the
daughter in the Chadburne household, infected four other members of her
family, some after a short latent period (Fig. 8.12), and 415 other secondaries (R : 6 to 17). During this initial period, the contact rates both within
and between households were high. The secondaries infected a very large
number (perhaps 60) of tertiary cases during the period mid-July to the end
of August but, now, the pattern of the epidemic had changed; there were
few infections within the household but a high interhousehold contact rate,
typical of outbreaks in July and August. The next wave of quaternary cases
occurred in September and the epidemic was almost completely extinguished by mid-October.
The plague spread from the metropolis to the townships and parishes
within about 25 miles (a days ride) and also along the Thames Valley (as
usual); in Deptford, only 3 miles from London Bridge, mortality was high,
but no real spatial pattern emerges because 432 died in Brentford, 20 miles
away. Shrewsbury described the events in the market town of Croydon, 10
miles to the south, where the rst plague death is believed to be that of a
fugitive from London on 12 June 1665 but the rst burial entry marked
pestis in the register is 27 July. Again, the pestilence did not explode, but 39
burials in September in Croydon compared with the monthly average of
214
Days
0
Chadburne d.
P
s.
d.
s.
m.
Morris
servant
Dawkins
s.
d.
s.
Tooke
Susan
Deare
f.
s.
s.
s.
s.
s.
s.
Field
John
Whithead
John
wife
Radford
Eliz.
Davis
d.
servant
d.
s.
Piggott
s.
s.
m.
d.
Phillips
Edward
Jenkes
Samuell
wife
Porter
Margaret
Price
d.
s.
Horderne
w.
Watts
d.
Waterworth servant
Bushby
servant
Lewis
Simon
Drinkwater Eliz., servant
Taylor
d.
d.
f.
Painsby
Bridgett
Powell
Mary
Parker
Allce
Wadbrooke Peter
Whitlock
d.
d.
Shepherd
Wm
Greene
John
Dodd
d.
20
40
S
S
S
S
S
S
T
T
S
S
T
T
T
T
T
T
S/T
S/T
T/Q
S/T
S/T
S/T
T/Q
T/Q
S/T
S/T
T/Q
T/Q
S/T
S/T
T/Q
S/T
S/T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
60
80
215
Fig. 8.12. (opposite) Suggested sequence of infections at the start of the plague
epidemic in the parish of St Michael Bassishaw, London, in 1665. Note the high
initial R and the rapid interhousehold spread. Compare with Fig. 8.11. Scale: days
after 11 May 1665. For further details and abbreviations, see Fig. 5.6.
216
Fig. 8.13. Weekly plague burials in London, May to December 1665. Abscissa:
weeks after 23 May 1665. Data from Creighton (1894).
Mortality was 68 595 but there is general agreement that this was an
underestimate: Bell (1924) averred that many plague deaths were either
deliberately hidden from the women searchers or those ocials were
bribed, or intimidated to refrain from reporting them. These views are
conrmed by the entry in Pepys diary for 30 August 1665: Abroad and
met with Hadley, our clerk, who, upon my asking how the plague goes, told
me it encreases much, and much in our parish; for, says he, there died nine
this week, though I have returned but six: which is a very ill practice, and
makes me think it is so in other places; and therefore the plague much
greater than people take it to be. In addition, because the Quakers, Jews
and Anabaptists refused to allow the plague deaths among their members
to be included in the church returns, the plague mortality among them
mostly escaped the bills. Pepys wrote on 31 August 1665, In the City died
this week 7,496, and of them 6,102 of the plague. But it is feared that the
217
true number of the dead this week is near 10,000; partly from the poor that
cannot be taken notice of, through the greatness of the number, and partly
from the Quakers and others that will not have any bell rung for them.
More people in London died of the plague in 1665 than in any other
visitation but this should be seen against the sharp rise in the population to
about 460 000, giving a plague mortality, according to the ocial gures, of
15%, which compares to a corresponding value of 13% for the earlier
epidemic of 1625.
218
circle inclining towards blue; in others a faint blue, the circle being blackish; others
again took a dusky brown tone. Often the esh was found to be spotted when no
discoloration was visible on the skin. No part was immune from these round spots,
though the neck, breast, back and thighs were the most common places for them.
The tokens sometimes were so numerous as to cover all the body.
It was the tokens, so universally dreaded, that gave to the Plague the name of
the spotted death. They customarily appeared after two to four days progress of
the disease, but might rise without any previous warning of infection. A woman, the
only one of her family left alive and thinking herself perfectly well, perceived the
pestilential spots on her breast and shortly thereafter died. A young man of good
constitution, unexpectedly nding the tokens upon him, believed them not to be
the genuine marks, he being otherwise in such vigorous health, yet within four
hours death conrmed the physicians diagnosis.
Dr. Hodges mentions as being most strange in his rst experience that many
persons came out of delirium as soon as the tokens appeared, believing that they
were in a recovering and hopeful condition. The poor suerers did not know their
fate. He recalls the case of a maid who had no idea that she was attacked by Plague,
her pulse being strong and senses perfect, and she complained of no disorder or
pain, but on examining her chest he discovered the tokens there. Within two or
three hours she was dead. The tokens sometimes rst became visible after death.
It is clear from this account that the tokens, spots upon the skin, rather
than the buboes were the most important diagnostic feature of the plague
(see section 13.12). This description of the signs and symptoms is amplied
and corroborated by Creightons precis of Boghursts writing:
Of evil omen was a white, soft, sudden, pued up tumour on the neck behind the
ears, in the armpit, or in the ank; also a large extended hard tumour under the
chin, swelling downwards upon the throat and fetching a great compass (the
brawny swelling of the submaxillary salivary glands and surround tissues). Tokens
came out after a violent sweat, which was often induced by purpose of the nurses,
who said, Cochineal is a ne thing to bring out the tokens. Nurses often killed their
patients by giving them cold drinks . . .
The botches, or buboes (swollen lymph-glands in the neck, armpits or groins),
were the most distinctive sign of the plague, having given to it the old name of the
botch. Besides these, there were the tokens (specially limited in meaning to livid
spots on the skin), carbuncles and blains. Carbuncles, sayd Boghurst, commonly
rose upon the most substantial, gross, rm esh, as the thighs, legs, backside,
buttock; they never occurred, that he saw, on the head among the hair, or on the
belly. They were not seen until the end of July, were most rife in September and
October, commonly in old people, never in children . . .
Blains are a kind of diminutive carbuncle, but are not so hard, black, and ery;
sometimes there is a little core in them. Generally they are no bigger than a
two-penny piece, or a groat at the biggest, with a bladder full of liquor on the top of
them, which, if you open but a little, will come out whitish or of a lemon or straw
colour. Besides a blain there is a thing you may call a blister, pung up the skin,
long like ones nger in gure, like a blister raised with cantharides; and such
219
Creighton continued:
Among the symptoms of a fatal issue, Boghurst mentions the following: Hiccough,
continual vomiting, sudden looseness, or two or three stools in succession, shortness of breath, stopping of urine, great inward burning and outward cold, continual
great thirst, faltering in the voice, speaking in the throat and occasionally sighing,
with a slight pulling-in one side of the mouth when they speak, sleeping with the
eyes half-open, trembling of the lips and hands and shaking of the head, staggering
in going about rooms, unwillingness to speak, hoarseness preventing speech, cramp
in the legs, stiness of one side of the neck, contraction of the jaws, the vomit
running out from the side of the mouth, prolonged bleeding at the nose, the sores
decreasing and turning black on a sudden . . . some of the infected run about
staggering like drunken men, and fall and expire in the streets; while others lie
half-dead and comatous . . . Some lie vomiting as if they had drunk poison.
(compare with the behaviour of the citizens of Chester in 1647, section 9.6.1
and with the plague at Athens, section 1.2.1).
Creighton (1894) gave the following summary of a dissection by Dr
George Thompson of a youth who died of plague:
He found what appear to have been infarcts in the lungs; the surface was stigmatised with several large ill-favoured marks, much tumied and distended, from
which, on section, their issued sanious, dreggy corruption and a pale ichor destitute
of any blood. The stomach contained a black, tenacious matter, like ink. The spleen
gave out on section an ichorish matter. The liver was pallid and the kidneys
exsanguine. There were obscure large marks on the inner surface of the intestines
and stomach. The peritoneal cavity contained a virulent ichor or thin liquor,
yellowish, or greenish. There was a decoloured clot in the right ventricle, but not
one spoonful of that ruddy liquor properly called blood could be obtained in this
pestilential body.
Bell said laconically that the dissection showed that the plague produced
far-reaching changes in the internal organs as well as aecting the skin by a
multitude of blue or black spots containing congealed blood. In fact no
organ was found to be free from changes.
Creighton (1894) summarised Boghursts account of the epidemiology:
It usually went through a whole kindred, though living in several places; which was
the cause it swept away many whole families . . . In some houses ten out of twelve
died, and sixteen out of twenty. Old people that had many sores upon them,
especially carbuncles, almost all died . . . Many people had the spotted fever and the
220
plague both together, and many the French pox and the plague both together, and
yet both sorts commonly lived. All sorts died, but more of the good than the bad,
more men than women, more of dull complexion than fair. It fell not very thick
upon old people till about the middle or slake of the disease, and most in the
decrease and declining of the disease. Cats, dogs, cattle, poultry, etc., were free from
infection.
Some died in twelve or twenty days, but more in ve or six. In summer, about
one-half that were sick died; but towards winter, three of four lived. None died
suddenly as stricken by lightning: I saw none die under twenty or twenty-four
hours. After one rising, or bubo, was broke and run, commonly another and
another would rise in several parts of the body, so that many had the disease upon
them half a year; some risings would not break under half a year [our italics], being so
deep in the esh.
The general populace believed that they caught the plague because the
infection was in the air of the place they were exposed to risk if they were
living in a plague-ridden spot. Certainly, the practice of shutting up a
family in their house when plague rst struck must have increased interhousehold contact rates.
8.6.4 Changes in virulence
The medical accounts suggest that the course and lethality of the disease
changed subtly as the epidemic progressed. Shrewsbury also drew attention to this apparent change in the virulence of the infective organism
quoting, in support, 30 recovered people emerging from the pest house. He
also quoted Hancock who said that on 10 September 1665 the disease came
to its height:
221
It now killed in two or three days, and not above one in ve recovered; or four in
ve died . . . but after this period, when the disease was on its decline, it did not kill
under eight or ten days, and not above two in ve died. So that it was calculated by
Dr Heath that there were not fewer than 60,000 people infected in the last week of
September, of whom near 40,000 recovered. For the plague being come to its crisis,
its fury began to assuage, and accordingly the Bill decreased almost 2000 that week.
For, had the mortality been in the same proportion to the numbers infected, as at
the height, 50,000 would very probably have been dead instead of 20,000, and
50,000 more would have sickened; for, in a word, the whole mass of people began to
sicken, and it looked as if none would escape, as not one house in twenty was
uninfected . . . the disease was enervated and the contagion spent. Even the Physicians themselves were surprised: wherever they visited they found their patients
better . . . so that in a few days, whole families that expected death every hour, were
revived and healed and none died at all out of them. Yet it appeared that more
people fell sick then, when not above one thousand died in a week, than when ve
or six thousand died in a week.
Boghurst corroborated these conclusions: an epidemic declined in malignity towards the end so that the buboes suppurated and some 60% of
patients recovered. Pepys wrote in his diary for 16 October 1665 Lord!
how empty the streets are, and melancholy, so many poor, sick people in
the streets full of sores . . ..
It can also be seen in section 8.6.3 that the signs and pattern of the
disease changed during the epidemic: the tokens appeared only rarely until
the middle of June and the carbuncles not until the end of July. Shrewsbury
declared that every English outbreak of bubonic plague in the past ended
in this way and he went to great lengths to explain this change in the
pathology as a progressive spontaneous decline in the virulence of Y. pestis.
222
Popular opinion is that the Great Fire, which broke out on 2 September
1666, was responsible for terminating the Great Plague and also for
eliminating bubonic plague for ever, not only from London but from the
whole of Britain, probably by destroying the rats in their burrows. This is
manifestly incorrect and Shrewsbury (who believed that Y. pestis was the
infective agent) was at pains to point this out, although he interpreted the
facts in terms of the eects on the rat population: the re was limited to the
walled city whereas the liberties and out-parishes had always been the
main foci for the epidemics. The epidemic was already fading 10 months
before the re, in the autumn of 1665, accelerated, apparently, by a reduction in virulence.
The plague did not die out in London after the Great Fire, but lingered
all through 1666 causing 1998 deaths. Indeed, the plague continued at an
endemic level, with a few deaths recorded in most years, for a further 13
years until 1679. The last major epidemic in England was a solitary
outbreak at Nottingham in 1667.
But, as we describe in section 3.1, authentic bubonic plague came to
England a number of times, 250 years later, in the 20th century, one of the
unfortunate consequences of swifter steamship travel, but no epidemic ever
developed because of the impossibility of establishing an epizootic.
8.6.7 Identity of the rst victim in households in plague
epidemics in London
We have analysed the major epidemics in London in 1593, 1603, 1625, 1636
and 1665 together with minor outbreaks in the parishes listed in Table 6.1
and have determined the identity of the rst victim in each family and the
results are shown in Table 8.1. When there were multiple infections in the
family, a child was the rst victim in 48% of the cases, in contrast with 31%
of the families where the infection began with a parent. We suggest that
most of the children were susceptible and were more likely to be infected
and to bring the disease into the household. When there was only one
person infected in the household during the epidemic (the majority of
cases), the situation was reversed, with 48% of the victims being adults and
34% children. Many of these adults may have been living in boarding
houses, or were travellers or there may have been no children in the
household. A high proportion, about 20%, of all the rst victims who
initiated the plague in their households were servants, suggesting that they
were susceptible immigrants.
223
Table 8.1. Identities of the rst victims in households in epidemics and outbreaks of plague in selected parishes in London
Multiple infections
Single infection
Adults
Year
Adults
Male Female
Child
Child
Servant
9
5
13
6
20
16
31
20
31
25
76
74
12
16
16
2
20
45
43
28
79
32
80
118
20
15
68
26
76
65
43
28
80
47
123
136
29
22
48
8
53
82
Totals
97
69
257
111
380
270
457
242
166
12
6
17
5
33
24
1593
1603
1625
1636
1665
Others
650
Note the dierence in the identity of the rst victim in multiple and single household infections. See Table 6.1 for parishes analysed.
Analysis of the intervals between the rst and second victims in the
household in the epidemics of 1603, 1625, 1636 and 1665 (N : 372) shows
that 99% of them were within 37 days, the estimated interval between the
point of infection and the end of the infectious period.
224
London, where the disease was endemic and displayed irregular, catastrophic outbreaks that decayed rapidly, possibly as described by a standard SEIR model. The dierence between the dynamics may reect the
dierent size, population densities, degree of immunity and R of the two
dierent types of population.
9
Plagues in the provinces in the 17th century
As we have seen, the plague in London in the later part of the 16th century
broke out sporadically and the epidemics steadily increased in intensity
whilst it grumbled on in the interepidemic years (section 6.2), becoming
endemic in the 17th century with outbreaks of unparalleled ferocity and
mortality (Chapter 8). Meanwhile, there were many reports of widespread
outbreaks scattered through the provinces throughout the 16th century
(section 6.3). Parish registers were required by law to record burials from
pestilence during the Elizabethan period but even though these provide a
great deal of invaluable information (much of which has still to be extracted and analysed), it is dicult to discern a pattern in the underlying
epidemiology of these outbreaks in the provinces. In general, these provincial outbreaks do not appear to have come from a focus in London. Were
there many introductions of the disease via the Kent coast (e.g. in 1532), the
south coasts (e.g. in 1544 and 1590) and the East Anglian ports (e.g. in
1585), or was the infection grumbling on through the century with low
infectivity (R O 1) but spread widely by apparently healthy carriers (be
cause of the long incubation period), with epidemics aring up when and
where the conditions were right? Improved communications and much
wider travel associated with the wool trade would have exacerbated the
spread (see discussion in section 13.9).
Inspection of the results presented in section 6.3 suggest that there might
have been two main foci (apart from London) in the metapopulation, one
in the southwest and the other in a broad band, running across the centre of
the country from East Anglia through the Midlands to the Welsh Marches.
The East Anglian ports were important centres for the importation of
goods that were then sent onwards via the river systems that they served
(see section 13.9) and plague was frequently introduced from continental
Europe by this route. In addition, the Northern Province seems to have
225
226
227
ning in July 1603 including closure of markets and the removal of suspect
cases to temporary accommodation outside the city walls. All exchange of
goods and commodities with Hull, London and Newcastle-upon-Tyne
were prohibited but, although the ocials scrupulously inspected the bales
of materials and clothing coming from these towns, they apparently ignored the travellers, seamen and merchants who brought them and probably
carried the infection.
The plague was at several towns in the wapentake (a territorial division
of Yorkshire) of Ainstie by October 1603 and, although illegal trading with
infected areas continued, the city of York was believed to be still clear of the
infection on 29 November 1603. However, the rst suspected plague victims were reported on 2 December 1603 (Sakata, 1982). The epidemic
probably grumbled on at a very low level through the winter (presumably
low R ), although trading restrictions with London were lifted on 14
February 1604, but it reappeared in mid-March 1604 (when there was an
entry in the burial registers of Holy Trinity, Micklegate) and watches were
set to ensure that vagrants did not enter the city. By early May, plague had
entered York. A viewer and clenser was sought from Newcastle, plaguelodges were erected outside the walls and a number of other public health
measures, including the killing of cats and dogs, were enacted. The Council
was now unable to control its spread through the City and reported on 5
May that there were infected houses in ve parishes, although substantial
rises in mortality did not occur until some months later, indicative of the
typical, slow build-up of the plague. In the parish of All Saints Pavement,
plague burials were specically identied; the rst occurred on 19 April,
followed by a group of four between 7 and 10 May, one on 2 June, three on
11 June and it was only from 6 July that plague burials began to occur on
an almost daily basis (Galley, 1998). This slow build-up in early summer
with the deaths widely spaced in time is typical of the start of an epidemic of
haemorrhagic plague with its long serial generation time. The epidemic
lasted until November 1604 and 82% of all burials occurred between July
and October. Total burials within the city walls numbered 2800 in 1604,
25% to 33% of the population (Galley, 1994, 1995), and the city annals
compiled in 1639 reveal that there dyed in York 3,512 persons, although it
is not possible to give such a precise gure, since the parish register of St
Olave recorded that The months followinge viz August September October november & December people dyed so fast that they could not be well
nombred (Galley, 1998).
York acted as a focus for the outward dispersion of the plague and
the Corporation House Books recorded on 18 April 1604 the plague is
228
dispersed in diverse townes and place nere vnto this Cittie and the archbishop reported that his diocese was sore visited with the sickness in
October. The parish register of St Giles, Durham, records 18 plague deaths
between September 1604 and 25 January 1605 (low winter R ); evidently
the epidemic spread northwards and continued through the following
winter but did not explode in the spring, perhaps because of a low R .
9.1.2 Chester as a focus
Plague had struck at Chester previously with a severe epidemic in 1517 and
with smaller outbreaks in 1558 and 1574 (Axon, 1894) and there were to be
repeated attacks during the rst 60 years of the 17th century. The city
seems to have acted as an entry port for the infection, which probably came
from Ireland, and then as a focus for repeated radial dissemination through
Cheshire and adjoining counties. The pestilence began in Chester in September 1602 (i.e. before the main London epidemic) in a glovers house in
Chester in St Johns Lane where 7 died (high household contact rate) and
mortality thereupon increased with 60 weekly deaths; 650 died of the
plague in 1603. In the parish of Holy Trinity, the epidemic started between
13 December 1603 and 3 January 1604 and the registers record the burial of
Alice dau to James Hand of Blacon, husbandman bur in church nere the
poore mans box: the rst that dyed of the plague in the parish but was not
known to the parish till she was buried. There were 986 plague deaths in
1604 (with 55 dying in 1 week) and 812 died between 14 October 1604 and
20 March 1605. After this, another 100 died before the epidemic ceased in
early January 1606. Cabins for the victims were erected outside the city at
the riverside, close to the New Tower and, on occasion, ve or six of the
same family died in the course of 23 weeks (high household R ) and it is
noteworthy that Creighton (1894) recorded that these multiple deaths were
particularly found in the homes of sailors. All fairs were cancelled; some of
the justices and a great number of citizens ed into the surrounding
countryside.
This plague at Chester was unusual because, although we have no
monthly totals of burials, its pattern was dierent from other provincial
epidemics that we have described previously, lasting 40 months, from
September 1602 to January 1606, presumably becoming temporarily endemic. Did the epidemic spread through the city from one focus to the next
in successive years? The rst burial in the parish of Holy and Undivided
Trinity, Chester, was followed by only three deaths in separate households
in early January and then after a break of 6 weeks the infection began again
229
Fig. 9.1. Monthly plague burials in the parish of Holy Trinity, Chester, from
January 1604 to January 1606.
on 3 March 1604 with ve deaths in three families. Plague burials were not
recorded again until a reinfection, when six people in four families died
between 18 August and 13 October 1604; these were so spaced out that
some must have been infected from an adjacent parish. Another minor
outbreak occurred during the winter with six plague burials in three
families between 29 January and 5 February 1605 and an isolated death on
24 March 1605. Plague began again 3 months later on 17 May 1605 and
grumbled on sporadically for 7 months until 9 January 1606 (the last
plague burial), during which time 64 people were buried. Most of the
victims had individual surnames and there were few families with multiple
deaths. The epidemic in Holy and Undivided Trinity did not follow Reed
and Frost dynamics, although there was an increase in deaths over the
period August to October (Fig. 9.1) and it is clear that this outbreak
followed a pattern dierent from that seen elsewhere: there were relatively
few deaths during the 2-year time period and the epidemic never exploded;
it died out several times and was reinfected from a nearby parish. Finally,
most frequently, only one person died in a household. We conclude that
this was a parish within the city and so was open to infectives entering and
that it had experienced plague previously so that many individuals were
resistant or immune.
The plague spread radially from Chester for considerable distances into
Cheshire, Shropshire and Staordshire; the infection was probably not
FAMILY
Days
0
Bennet
20
40
60
80
100
120
140
160
180
200
220
P
S
S
F
E
N
Leene
S
T
Sx
Bink
Bell
Tottie
Evans
S
S
S
T
Q
Madocke
Sx
Marton
Sx
Gile
Sx
Se
Rabonne
Se
E
E
Johnson
E
E
Fig. 9.2. Suggested sequence of infections in the plague epidemic in the parish of West Kirby, Cheshire, in 1604. Note low R . Scale: days
after 14 April 1604. For further details, see Fig. 5.6. Abbreviations: Q, F, Sx, Se, E and N, quaternary, fth, sixth, seventh, eighth and
ninth generation infections, respectively; 7, male; 8, female.
231
spread from place to place in the majority of these outbreaks, rather the
pestilence was carried out separately from the focus. Most of these secondary epidemics were in small, scattered parishes where the mortality was
slight.
(i) West Kirby, northwest Wirral, 17 miles from Chester. Plague began
here on 21 May 1604, lasted 7 months and nished on 25 December
1604 during which time 14 families were infected and 26 people died.
The spread of the epidemic through the families is shown in Fig. 9.2.
The epidemic broke out in the Bennett family on 21 May 1604 with
one female death and this was followed by two co-secondaries on 4
and 5 June. The primary case infected a further three or four contacts
in the Leene, Bink, Bell and Tottie families. The further spread of the
disease at West Kirby is shown in Fig. 9.2: the tertiary infection of a
male in the Evans family is of particular interest because he subsequently infected a quaternary (a female) in his family. If this quaternary case had not infected anyone else the epidemic would have
ceased, but she had a single contact with a female in the Bennett
family, which was then reinfected and the outbreak was maintained
and continued until the end of the year when it disappeared. The
epidemic lasted over some nine generations, although few people died
because of the low R , estimated values for which are: primary to
secondaries : 6; secondaries to tertiaries : 0.33; tertiaries to quaternaries : 1; fth to sixth generations : 4; sixth to seventh generations : 0.5; seventh to eighth generations : 2. Estimated parameters: latent period : 12 days, infectious period (including symptoms) : 25 days; a type (i) epidemic.
(ii) Malpas, Cheshire, 12 miles southeast of Chester. The plague was a
minor outbreak as the registers show:
28 June 1604: Dominick a gentlemen that died at Mrs. Maude
Breretons of Edge by the plague and was brought to the church on
a drag or sledge with a horse.
29 June 1604. Thos. Plymlose, servant to Mrs. Maude Brereton of
Edge who died of the same infection.
(iii) Whitchurch, CheshireShropshire border, 16 miles from Chester.
Plague reported here on 26 August 1604 in the Malpas registers
(Richards, 1947). It may have been the source of the infection in
Malpas, the adjoining parish, and the plague may have spread from
here to Mucklestone in Staordshire and to Alderley in Shropshire.
(iv) Nantwich, Cheshire, 17 miles southeast of Chester. Said to be
232
Days
0
Jaxson
Evans
Clarke
Waters
Lea
Smithe
Jones
Corse
Harper
Long
Raphes
Rider
Davies
Picke
Brayne
Meighen
Frankes
Pearse
Lloyd
Lewis
Jenkson
Lapington
Wayne
Powell
Moodie
Trevenor
233
d. Katherine
m.
apprentice
maid
f. Edward
maid
m. (w.) Ann
s.
d.
d. Mary
d. Joan
f. James
John
m. Judith
d. Martha
f. John
m. Marjerie
boy
d. Dorithie
d. Elizabeth
d. Elnor
w.
w. Katherine
woman
m. Jane
apprentice
maid
d. Sarah
m. Mawde
s. Richard
s. Robert
s.
d.
s. Barnabas
Francis
servant
s. Samuell
Edward
wife
s. Edward
lodger
Anne
maid
Robert
m.
20
40
60
80
100
P
S
P
S
P
S
S
T
P
S
S
S
P
P
S
S
S
S
S
T
T
S
S
?S
S
S
T
S
S/T
S/T
T/Q
S/T
S/T
T
T
T
T
T
Q
T
T
Q
T
T
T
Fig. 9.3. Suggested sequence of infections at the start of the plague epidemic in the
parish of St Mary, Shrewsbury, in 1604. Scale: days after 2 May 1604. For further
details and abbreviations, see Fig. 5.6.
234
(viii)
(ix)
(x)
(xi)
Days
0
20
Thorntune
Wyllyam
Darbishire
Ellyne
Johnsone
Joane
Cowper
Wyllyam
Ellyne
Anne
Joane
Eliz.
Isabell
S
S
Alles
Marye
?S
Isabell
Thomas
?S
Meredithe
Maddocke
Goodhare
235
40
60
T
T
Q
Q
T
F
T
Urmsone
John
Bell
Alles
Posnett
Edward
Myllner
Ellyne
Enme
Hughe
Ales
Davinsone
Enme
100
Eliz.
Wyllyam
Marye
Ap Shone
80
Q
Q
T/Q
Q
F
F
F
Fig. 9.4. Suggested sequence of infections in the plague epidemic in the parish of
Heswall in 1605. Scale: days after 19 July 1605. For further details and abbreviations, see Fig. 5.6.
120
236
Fig. 9.5. Burials recorded in Manchester parish register in 1605. After Willan (1983).
the registers suggests that this was not the result of a plague epidemic.
Plague burials were not recorded in the registers in 1605 but it is evident
from contemporary accounts that this was a major epidemic of the pestilence. There were no weddings during June to August and no baptisms in
August because of the extremitye of the sickenes. The usual preventative
measures were introduced: a tax was levied for the infected; the Court Leet
was not held; people coming into Manchester were required to produce a
certicate that they were free from infection; plague-cabins were established on Collyhurste (Willan, 1983).
The plague of 1605 began in April and followed the characteristic Reed
and Frost dynamics, gathering momentum slowly, deaths peaking in July
and probably ending in December 1605 (see Fig. 9.5), although Willan
(1983) states that sporadic cases occurred until February 1606, i.e. a persistent type (i) epidemic.
A total of 1053 burials was recorded in Manchester in 1605 but the
absence of any specic marking of plague victims in the registers makes any
analysis of the epidemic dicult. Nevertheless, the epidemiology appears
to be dierent from other outbreaks that we have described. We assume
that the pestilence began in early to mid April and by 24 April 1605 the
237
daughter in the Salle family was buried (the primary) and she was followed
16 days later (10 May) by a male Salle death and then on 14 May
(maidservant and a child), on 18 May by the widow and on 23 May by a
brother, all secondaries. This ts with approximate latent and infectious
periods of 12 and 25 days, respectively, and with the typical progress of
haemorrhagic plague in families. However, before that date, on the following days in April, the burials in the Leeze household were recorded within
the space of 5 days: 19th (female), 23rd (female), 23rd (daughter), 24th
(male), 24th (wife), 24th (infant). These must all have been household
co-primaries, i.e. infected from a common source, a most unusual pattern at
the start of an epidemic. However, the appearance of multiple primary
infections within a family was repeated many times in the plague at
Manchester; 74% of the second burials occurred within 12 days of the rst,
suggesting that within families the latent period may have been shorter than
usual. Only a further 18% were clear household secondaries, dying within
37 days of the rst burial. This high cumulative percentage dying within 12
days is similar to that recorded between the second and third burials and
between the third and fourth burials. Perhaps crowded living conditions or
the social behaviour (including going to market en famille) of the inhabitants of Manchester may have accounted for the multiple initial infections
in a household.
Manchester also had an epidemic in the late autumn of 1606 and
through the winter of 1607 that Axon (1894) recorded as plague, although
the registers do not show an excess mortality at this time. Patients were
sent to the pest-house and a tax was levied for the relief of the stricken.
The 1605 plague in Manchester may have been carried from Chester (35
miles) where the disease was raging. It then spread southwards to Stockport (5 miles) and Maccleseld (14 miles). Plague appeared in Stockport in
the autumn, on 9 October 1605, and persisted through to 14 August 1606,
during which time 51 people died (type (ii) epidemic, low mortality, low R ).
At Maccleseld, the epidemic also broke out in the autumn, on 3 September 1605 and 70 burials were recorded before 3 October 1605, spreading
particularly through families (high household contact rate), see Axon
(1894).
238
but there is no doubt that many places were aected, although most of
them did not experience a major mortality crisis. Shrewsbury (1970) has
assembled the evidence but it is not possible to determine the spatial
components of this outbreak. It also spread westwards from London along
the Thames Valley to Oxford and thence southwestwards into Wiltshire,
reaching Devizes, Marlborough and Salisbury by early autumn 1603.
Plague was already raging in Bristol by this time, having begun before July
1603 with, it is estimated, more than 3000 deaths before January 1604 (type
(i) epidemic). Plague erupted in Bath in May 1604 and peaked in July but
with only 50 deaths during 14 weeks of the summer. Bristol was said to be
reinfected in July 1604 and it may have acted as a secondary entry port for
the infection.
Plague was also reported to be active on the opposite side of the country
in East Anglia. Lowestoft and Norwich may have acted as entry ports for
this area because there was heavy mortality in both these towns in 16023.
Although a sick house was established at Ipswich in May 1603 and bailis
were appointed to supervise the infected houses and all ships were prohibited from coming to the town, only four persons were reported as dying
from plague. Only 10 plague burials were reported at Cambridge.
Thus plague could have spread to the Midlands by infectives travelling
from a number of foci: (i) directly from London, (ii) northwards from
Wiltshire, (iii) via ports on the east coast, (iv) from the port of Bristol, (v)
from the port of Chester and thence via the focus at Shrewsbury or via
Staordshire. Plague was reported scattered at this time at Northampton
(231 burials), Boston (where the market was closed), Birmingham, Congleton, Leicester (a mild epidemic), Tewkesbury (23 plague burials), Walsall
(16 plague burials) and Stamford in Lincolnshire (nearly 600 deaths).
In 1605, plague broke out in Oxford (where temporary pest-houses were
established), Godmanchester in Huntingdonshire (30 plague burials),
Northampton and Grendon in Northamptonshire (12 victims) and Cambridge (Shrewsbury, 1970).
It can be seen that the pattern of the epidemics across central England
changed in the early part of the 17th century; henceforth plague became
persistent with outbreaks scattered over the metapopulation and the pestilence was probably reinforced by infectives regularly coming in through
the ports from overseas. In general the reported mortality became less
severe, perhaps because the towns had experienced the pestilence previously. Even if better information were available it would probably be dicult
to dene the spatial component of the epidemics. Much of the movement
along the roads in the Midlands at that time would have been associated
239
with the wool trade and, since the disease had a long incubation period, the
epidemics would appear to spread almost randomly.
240
between April and December. The epidemic does not appear to have
spread southwards along the northeast corridor.
Days
0
Dawson
Clutton
241
s. Raphe
servant
s. Richard
f. Thomas
d. Eliz.
d.inl. Ann
Richard
s. John
servant
child
m. Maude
child
child
20
40
60
80
100
P
S
S
S
S
S
S
S
T
Q
Q
Q
F
Fig. 9.6. Suggested sequence of infections in the Dawson and Clutton families at
Malpas, Cheshire, in 1625. Scale: days after 18 June 1625. For further details and
abbreviations, see Fig. 5.6. d.inl., daughter-in-law.
120
242
The Clutton family were then also infected by the servant to the Dawson
family, a tertiary (Fig. 9.6) and these victims were buried by their relatives
in the churchyard. The story of the epidemic at Malpas shows how, because
of the long latent and infectious periods, the pestilence could be carried
over a great distance from London, a journey that must have taken some
time. The high household contact rate ensured the death of apparently the
whole household, who had no resistance, whereas the low population
density in the hamlet and the isolation measures that are indicated by the
relatives burying the dead around the house and not in the churchyard,
contained the outbreak.
There was also a small outbreak of plague coincident with the later
events at Malpas in the adjacent parish of Shocklack. John Handley was
buried 3 September 1625 and, with hindsight, is thought to have introduced the plague, possibly from a visit to Malpas, because his children, John
(15 years old) and Elizabeth (19 years old) both fell sick on 23 September.
John died 2 days later And by reason he died so suddenly, and having
Red-specks [the Tokens] found upon him, he was supposed to have died of
the Plague. And therefore was carried to the Church upon a Dragge by his
Mother, Elleyn Handley and Randle gylbert his half brother. And was
buried in the Church Yard, at the Steeple end out of the alley; without
Service, ringing or any other ceremonies of the Church. Elizabeth Handley
died on 27 September And because she had Red-specks found upon her
and some sore under her arm, she was likewise suspected to have died of the
Plague. And therefore she was buried in her Mothers Croft, near the
Orchard upon Wed 28 Sep. And upon Monday the 3rd day of Oct., the
aforesaid Eliz. Handley was taken up again out of her grave and brought to
the Church, upon a Dragge by her half brother Randle gylbert, aforesaid,
which buried her near to her brother John Handley the younger without
any ceremonies of the Church. Elleyn Handley (the mother) was buried 9
October 1625 And because her two children had died with Red-specks
upon them therefore she was suspected to die of the Plague. Wherefore she
was carried to the Church upon a dragge by her son, Ran Gylbert, and laid
by [the bodies of ] her two children, John and Eliz. Handley. Randle
Gylbert (aged 28 years), who assisted with these family burials and who was
supposed to have brought these troubles and sickness into the Parish of
Shocklach, was conned to keep his Mothers house, and there kept in by
watch and ward, night and day, himself alone a long time. And upon
Sunday 2 Oct. he was stripped and viewed by certain neighbours and
Parishioners, and then had no sign of any sore found upon him. He
escaped the plague and was married the following year (Richards, 1947). It
243
is noteworthy that the red spots or tokens were the characteristic symptoms of the plague.
Plague was also very active again in the northeast corridor in 162526,
appearing in various parishes: in the North Riding of Yorkshire (the city of
York escaped); in Gateshead, Barnard Castle, Whickham (a single household) and Sunderland in County Durham; and in Newcastle-upon-Tyne in
Northumberland in both 1625 and 1626 where the epidemic was probably
not severe. Sunderland was said to be dangerously infected and there were
89 plague burials at Gateshead.
244
Days
0
Browne
Grantham
Brader
d.
20
40
60
P
P
P
s.
Madison
Watson
Liell
P
d.
P
S
S
?
Gasnel
Smith
Marshall
P
s.
s.
s.
d.
P
P
P
d.
d.
Newman
S
S
P
P
Goodwin
s.
Gilpin
d.
Wattes
Dalton
P
wife
P
S
Longthorne
Hare
P
d.
S
S
Fig. 9.7. Suggested sequence of infections at the start of the plague epidemic in the
parish of Alford, Lincolnshire, in 1630. For abbreviations, see Fig. 5.6. 8, female; 7,
male. Scale: days after 15 June 1630.
245
1631. Although plague deaths are not specically recorded, probably about
43 died before the last that died of the sickness on 4 December 1631; a type
(i) epidemic. Only two families were stricken in another parish, St Peters, in
Lincoln. Many of the burials are single events in the families and there were
probably 10 co-primaries. This pattern of isolated cases in the families,
coupled with a low death rate in the city, suggests that Lincoln may have
experienced outbreaks of plague before and there is evidence of epidemics
there in 1557, 158687, 1593 (Creighton, 1894; Shrewsbury, 1970).
The plague spread from Lincolnshire to Yorkshire in 1631 where Redness and Armin were furiously infected, and 100 persons died. It broke out
at Acomb (1.5 miles west of York) and was brought into the suburbs of
York by a lewd woman from Armin [clearly an infective] and in that street
are since dead some four score persons. It has not yet got within the walls,
except in two houses, forth which all the dwellers are removed to the
pest-houses. The disease was eectively controlled and conned almost
entirely to one parish. The Council minutes for 29 August 1631 recorded
that some persons in St Lawrence churchyard (an extra-mural parish) were
visited with the infection of the plague. Six watchmen were set to keep
people away from the churchyard, all beggars and wanderers were removed from the city, a plague lodge was erected and the parish was
eectively isolated from the rest of the city, but by 31 August there were 16
dead thereof in St Lawrence and one in St Margarets parish. Rigorous
public health measures were enforced vigorously: there were restrictions on
movement, strict isolation of the sick and all those suspected of being
infected along with any contacts were immediately quarantined. Special
attention was given to ensure that St Lawrence parish remained isolated
and watch was sett at the end of St Nicholas forthwith to keepe out
strangers and to keepe in the inhabitants without Walmgate Bar but the
orders were still not suspended and individuals and families continued to
be isolated until the end of March 1632, well after any threat had disappeared (Galley, 1998). The epidemic at York in 1631 is an example of
successful public health measures against an infectious disease spread
person-to-person.
The plague also appeared in 1631 in west Yorkshire at Mireld (180
died) and Heptonstall (near forty houses infected) and then in west Lancashire at Dalton-in-Furness (360 deaths) and the adjacent Isle of Walney
(120 deaths).
Although there were only a few, small outbreaks in England in the
following year, 1632, the story of Slaidburn, Lancashire, an upland and
isolated parish in the Pennines, is of interest because all the victims were
246
FAMILY
Days
0
Eynsay
Windle
w.
P
Johannes
wife Margareta
s.
Jeneta
Alisa, jr
Alisa, sr
Johannes
Driver
Ane, spinster
Prockter
d.
s.
s.
f. Matheus
Holden
d.
f. Thomas
d.
d.
20
40
60
80
100
S
S
S
S
S
S
T
T
T
Q
Q
Q
T
T
Q
Q
carefully documented in the registers and this minor epidemic can be fully
analysed; it conforms exactly with the pattern of this infectious disease that
we have described previously. All the deaths were conned to the hamlet of
Dalehead, which is now submerged under the Stocks Reservoir. The infection may have persisted over winter from the outbreak in the Furness
peninsula; it was introduced about 30 June 1632 by widow Eynsay (primary), who died on 5 August (Fig. 9.8). There was a high household contact
rate and she directly infected three co-secondaries in her own family and
three further co-secondaries in the Windle family. Tertiary infections began
with Ane Driver and followed in the Prockter and Holden families after 15
August 1632, with a total of 17 pest deaths conned to ve families.
Analysis of the records suggests that the epidemiological parameters were:
latent period : 12 days, infectious period : 25 days. R has been cal
culated as follows: primary to secondaries : 6; secondaries to tertiaries : 0.8 (i.e. 1); tertiaries to quaternaries : 1.
After a break of some 2 years when England was apparently free, plague
was reintroduced in 1635, probably from abroad via the east coast ports of
Hull and Yarmouth and also the port of London. The epidemic continued
in Hull in 1636 after the winter, died down in the autumn and reappeared
247
again in 1638; it did not subside until late 1639 and the plague was very
persistent but there may have been reinfections that restarted the outbreak.
Shrews-bury (1970) quoted accounts that described the deaths of 2730
inhabitants of Hull during this 4-year period. Plague was also present in
Yarmouth and Lowestoft on the east coast in 1635 and it continued in East
Anglia in 163637, being recorded in Norwich, Yarmouth, Kings Lynn,
Bury St Edmonds and in Cambridgeshire and also extended to Leicester,
Horton and Melton Mowbray in the East Midlands.
The northeast corridor also had a major outbreak of plague in 1636,
probably emanating from the focus at Hull but possibly coming from a
separate introduction from overseas via the port of Newcastle-upon-Tyne,
which suered a tremendous visitation, with a death toll of over 5000. The
weekly toll of plague burials (Fig. 9.9) shows the characteristic Reed and
248
249
the following year, 1645, appearing in June and following typical Reed and
Frost dynamics over the next 6 months with a sharp rise to a clear peak in
late July with 1300 deaths (a type (i) epidemic). Outbreaks were also
recorded in 1645 in numerous towns in west Yorkshire, including Halifax,
Wakeeld, Bradford, York and Whitkirk.
250
Evidently, the appearance of the spots and not a swelling was the clinical
feature.
The monthly burials at Ottery St Mary in 1645 were: October, 15;
November, 18; December, 79; and in January 1646, 37. This plague epidemic was at its peak in winter and Shrewsbury (1970) consequently
excluded bubonic plague as the cause. It appears to be an unusual type (ii)
epidemic of haemorrhagic plague, with the typical start in autumn but an
abnormally high winter mortality and the pestilence did not break out
again in spring. It is comparable with the epidemic in 164445 in the parish
of St Oswald, Durham (section 9.5.1).
251
252
253
Fig. 9.10. Monthly plague burials at Whitchurch, Shropshire, from April 1650 to
July 1651. Data from Shrewsbury (1970).
254
FAMILY
Days
0
Bennett
Johnson
Lancelott
Coventry
Done
Hickson
Tomlinson
Ryland
Mason
Williams
Cornley
Gilman
Davye
Leather
Barlow
Fformby
Kinge
Axin
Millner
Jones
Hickson
Maddock
20
40
60
80
100
d.
s.
wife
s.
s.
w.
d.
s.
Thomas
w.
Thomas
s.
a poor woman
s.
s.
wife
d.
d.
d.
d.
s.
Robert
d.
d.
d.
d.
child
child
s.
s.
w.
wife
d.
d.
Robert
s.
d.
maid
d.
w.
s.
John
wife
Ann
d.
wife
s.
Fig. 9.11. The closing stages of the plague epidemic in the parish of Neston,
Cheshire, May 1665 to February 1666. The epidemic lingered because of the long
incubation period. Scales: days after 15 August 1665 (Part 1) and 28 August 1665
(Part 2). For further details and abbreviations, see Fig. 5.6.
Days
0
Bennett
Everton
Prout
Woode
King
Davey
Moade
Boardman
Hickson
Betson
Presscot
Woodward
Blundell
Madok
Looker
Sefton
Fforshall
Walley
Griffith
Barlow
255
20
40
60
80
100
120
140
wife
s.
s.
d.
d.
maid
d.
d.
s.
s.
David
wife
d.
s.
d.
s.
Thomas
John
wife
d.
Ann
s.
wife
w.
s.
d.
w.
s.
d.
s.
Reece
child
certainly infected by his mother, who had died on 12 October, when she
must have been in the terminal stages of her illness and showing symptoms,
again a 35-day interval (see top of Fig. 9.11, Part 1). The long latent and
infectious periods account for the very slow elimination of the plague and
the premature announcement of its cessation.
160
256
257
Fig. 9.12. Weekly plague deaths at Colchester, Essex, from September 1665 to
November 1666. Data from Creighton (1894).
258
1666 and appears to be an unusual type (ii) epidemic with a very early start
that led to a greatly extended outbreak.
This persistence of the epidemic through into 1666 was also seen at the
East Anglian ports of Ipswich, Harwich and Yarmouth. There were relatively few plague deaths in Norwich in autumn 1665, but 13 were recorded
in January 1666 and the epidemic regathered momentum in the spring and
rose to a peak of 200 deaths per week in August 1666 (type (ii) epidemic).
The second phase of the outbreak lasted from March to December 1666.
The plague moved outwards in East Anglia in 1665 and, in addition to
the outbreaks in Essex, it was reported in Cambridge (413 burials), Royston, Oundle, Ely, Peterborough and Kings Lynn and was said to be active
in 20 parishes in Buckinghamshire. Several of these outbreaks continued
into 1666: the plague broke out again in Cambridge in June 1666 and
Sturbridge fair was cancelled. The Bishop of Ely also cancelled all fairs in
his diocese because Peterborough (where it continued from summer 1665
to spring 1667) and Ely were also suering from this outbreak. Pest-houses
were erected outside Cambridge and the population apparently suered a
severe plague mortality so that the harvest could hardly be gathered. It was
not until January 1667 that the Vice Chancellor notied the scholars that
they might return to the colleges.
Ramsey in Huntingdonshire was attacked in 1666 and Shrewsbury
(1970) quoted an account that plague was imported in winter (February) in
a parcel of cloth from London. The tailor who made up the cloth and all his
family died of the disease and the subsequent epidemic destroyed more
than 400 of the townsfolk, although the rst plague burial noted in the
parish register was that of a woman who was buried in her own garden on
16 July 1666. Presumably the epidemic was started by an infective traveller
who brought the cloth from London and the high household contact rate
ensured that the infection ran through the family. The infection must have
spread to another family and persisted in the population, even with the low
contact rate in winter, but the epidemic slowly gained momentum during
the following spring.
259
outbreaks. It is not a series of the total annual mortality. After the Black
Death, the epidemics grumbled along sporadically at a low level for 150
years but after 1500 the spread gradually increased in terms of the number
of places attacked. Spectral analysis of the data-series (see section 2.8)
reveals a non-stationary short-wavelength oscillation (P : 0.05), the period of which was 57 years and a medium-wavelength oscillation that is
discussed in section 11.7.
We have previously described a persistent, short-wavelength oscillation
in English wheat prices that caused sharply changing cycles of malnutrition; these had complex and serious eects on infant mortality, the outcome of pregnancy and the epidemics of lethal infectious diseases (Scott et
al., 1995; Duncan et al., 1996a,b, 1997, 1998, 1999; Scott & Duncan, 1998,
1999a, 2000). Did this oscillation in wheat prices also drive the outbreaks of
plague?
The English wheat prices series, 14501649 (data from Bowden, 1967,
1985), is shown in Fig. 9.14. It falls into two periods: for the rst 100 years
the level was low and the trend was steady with low amplitude oscillations
superimposed thereon, but after 1550 the trend rose markedly with large-
260
Fig. 9.14. English annual national wheat price index, 14501649. Data from
Bowden (1967).
amplitude oscillations. Spectral analysis reveals a short-wavelength oscillation: period : 8 years (14501550) and 6.7 years (15501649). But ltering the two series shown in Figs. 9.13 and 9.14 revealed no cross-correlation between them. Equally, ltering for the medium-wavelength
oscillation revealed no cross-correlation between the two series.
We conclude that uctuations in the wheat prices did not drive the
plague epidemics in England. The pestilence was not truly endemic in this
separate island metapopulation (unlike France, see section 11.2) and the
outbreaks outside London were governed largely by the unpredictable
arrival of infective(s) from overseas or the metropolis. But England was
part of the enormous metapopulation of continental Europe from whence
came the infectives, so that the epidemiology of the plagues in England was
governed by the spatial components of the pestilence on the Continent,
factors that are described in Chapter 11.
10
Plague at Eyam in 166566: a case study
262
263
long distances through the countryside and starting an epidemic are quite
impossible.
The next person to die of plague, Edward Cooper, was buried 15 days
later on 22 September and if an infected ea had bitten George Viccars no
epidemic of bubonic plague could have been established among the hypothetical rodents of the village in so short a space of time; it would have
initially involved one or two rats becoming infected with Yersinia pestis,
developing the disease, infecting their eas, dying and the transference of
the eas to other rats. This cycle would need to have been repeated many
times before the epidemic spread through the supposed population of roof
rats; there would have been no possibility of establishing an epizootic and
the domestic rats would start dying in numbers. Only then would the eas
transfer to the humans.
The role in the village of George Viccars, the rst person to be infected
and die, is not clear in these traditional anecdotal accounts, but Cliord
(1989) implied that he was not a native and did not have a family with him,
presumably because there is no other mention of the surname Viccars in
the parish registers. However, it is known that he was survived by his
children, although Cliord (1989), most disappointingly, does not say
where they were living.
264
Fig. 10.1. Monthly plague burials at Eyam, August 1665 to December 1666.
bubonic plague was understood, but his account has subsequently been
greatly modied to t with accepted opinion. He clearly assumed that it
was an infectious disease and did not invoke the rats and eas of bubonic
plague of which he knew nothing. It is on his original account and the
invaluable details contained in the parish registers that we base our analysis of the origins of the plague and the events of 166566.
Wood says that Mary Cooper, the widow of a lead miner (they had
married in 1652) had, according to the registers, remarried Alexander
Hatfeild (trade not stated) on 27 March 1665, some 5 months before the
plague erupted. She was living with her two sons in the middle of three
cottages to the west of the churchyard and had taken in as a lodger George
Viccars who was a travelling tailor and the rst victim. Cliord (1989)
stated that Wood did not record whether Viccars was a regular visitor,
calling on Eyam as part of an established round, or whether he was a
stranger. Nor does he record from whence he came.
During the 17th century, there were several Viccars families living in the
parishes of Baslow, Droneld and Edensor, all close to Eyam. One family
living in Edensor, 5 miles south of Eyam, is of note: George Viccars married
Ann Allin in 1638; he may have been born to Elizabeth Viccars in 1611 at
Droneld. George and Ann had seven children at Edensor and if this were
the George Viccars, the travelling tailor who brought the plague to Eyam,
he would have been aged about 54 years in 1665 and he would still have
265
had to support some of his children. The Viccars family continued living at
Edensor and there are records of baptisms and marriages in the register
but, in any event, the problem remains: where did George Viccars become
infected in autumn 1665 in the course of his travels? We have no idea of the
area that he covered in his itinerant work. There is little evidence of plague
in Derbyshire in 1665 and no rise in mortality at Edensor in 1665; the
statements that Derby suered from an epidemic in 1665 (Batho, 1964;
Race, 1995) as well as in 1666 (Shrewsbury, 1970) were not conrmed by
Bradley (1977), who found no excessive mortality recorded in the registers.
Nevertheless, Bradley quoted an account by Hutton that in 1665 Derby
was again visited by the plague . . . The town was forsaken: the farmers
declined the market place. Shrewsbury (1970) quoted evidence that
Newark (37 miles from Eyam) lost one-third of its inhabitants from plague
in either 1665 or 1666. Alternatively, Viccars may have been to London to
collect cloth or he may have associated with an infectious carrier who came
from there. Viccars was infected about 1 August and was lodging with
Mary Cooper in Eyam when the rst secondary infection took place on 16
August 1665 (see below), probably sucient time to have made a journey
from London. Viccars lodged with Mary Cooper for at least 3 weeks before
he died.
266
FAMILY
Days
0
Viccars
George
Cooper
s. (3)
s. (12)
Hawksworth Peter
s. (15 months)
T. Thorpe
Sydall
Banes
W. Thorpe
H. Torre
20
40
80
100
P
S
T
S
T
S
d. (11)
wife
s. (9)
S
S
d. (12)
s. (7)
John
d. (23)
d. (20)
d. (9)
Mathew
wife
d. (2)
w.
S
S
wife
William
s.
S
S
wife
Humphry
T
T
T
T
T
T
T
T
T
T
Q
W. Torre
wife
s. (10 months)
T
T
Ragge
s. (10)
d. (9 months)
George
Stubbs
60
s. (18)
John
wife
T
T
T
Q
Q
Fig. 10.2. Suggested sequence of the infections during the rst phase of the epidemic
at Eyam. Scale: days after 1 August 1665. For further details and abbreviations,
see Fig. 5.6.
267
268
Presumably, either George Viccars visited these nearby houses in his work
as a tailor or they came to visit their neighbour Mary Hadfeild, nee Cooper.
The plague struck next at the families of Humphry Torre and William
Torre; they may well have been father and son, respectively, and Peter
Hawksworth (secondary, died 23 September, see above) was married to
Jane Torre, the daughter of Humphry Torre. For this reason, we believe
that the Torre families were not infected by Viccars but that the plague was
brought by Peter Hawksworth immediately he became infective. Sythe,
wife of Humphry Torre, was infected about 31 August 1665 and she then
transmitted the disease to her husband Humphry (a quaternary case).
Hawksworth also infected Amie, wife of William Torre (his sister-in-law)
and Humphry (aged 10 months) their son; William Torre died on the same
day as his wife. The Torre families were related to Rowland Mower (section
10.5) and may also have been connected with the Sydalls, because Rowland
Torre was betrothed to Emmott Sydall.
Three members of the family of George Ragge were tertiary infections
who could have been infected by any of the early secondary infectives
(Hawksworth, Thomas Thorpe, Sydall and Banes families). Tradition suggests that the Ragges lived next door to the Sydall family so that they may
have been infected from any of the little collection of cottages where the
outbreak began. Son Jonathan (aged 10 years), daughter Alice (aged 9
months) and George Ragge himself died over a space of 9 days (see Fig.
10.2). His daughter Ellen and wife Elizabeth (who may have died in 1699)
apparently survived.
Likewise, the Stubbs household was another tertiary infection that was
introduced about 25 September 1665 from any of the families listed above,
plus the families of William Thorpe and the Torres. Hugh Stubbs (aged 18
years), son of John and Ann Stubbs, was infected about 25 September and
he probably infected both his parents (quaternaries).
The contact rates, R , for the rst phase of the epidemic are: primary to
secondaries : 10; secondaries to tertiaries : 1.8; tertiaries to quaternaries : 0.2.
There are striking similarities between the initial phases of the type (ii)
epidemics at Penrith (section 5.3) and Eyam. Both were northern upland
communities where the epidemic began in the autumn. Both were started
by the arrival of a symptomless, infective stranger who took lodgings and
stayed in the town or village. They contracted the disease some distance
away and the long incubation period allowed the plague to be transmitted.
These epidemics were not begun by, say, a drover passing through. Viccars
was resident in Eyam for at least 22 days and up to 34 days; Hogson stayed
269
10.4 The second phase: maintenance of the epidemic through the winter
As in the 159798 plague at Penrith, the epidemic was only just maintained
at Eyam through the winter of 166566 (Fig. 10.3). Hugh Stubbs (aged 18
years; died 1 November 1665) probably infected both his parents (Fig. 10.2)
but Hannah Rowland (aged 15 years), who died only 4 days later on 5
November, was not infected by Hugh Stubbs. It is impossible to suggest
who may have infected her because there were some 26 victims who died
between 29 September and 24 October who could, potentially, have been
the source. These two teenagers were critically responsible for the continuation of the epidemic, not only within their families but also, more importantly, infecting other families (Fig. 10.3). Elizabeth Warrington (aged 18
FAMILY
Days
0
20
40
60
80
100
120
140
160
180
200
220
240
Rowland
d. Hannah
d. Mary (13)
s. Abel (10)
Thomas
Warrington Eliz. (18)
Rowbotham s. John (12)
Wm
s. Robert (16)
s. Samuel (8)
Rowe
infant
wife Mary
Wm
Wilson
s. Thomas
s. Isaac
s. John (9)
d. Deborah (20)
d. Alice (13)
John
Blackwall
s. Anthony (10)
Anthony
d. Anne (19)
d. Joan
m. Margaret
Alleyn
Thomas
infant
Thorpe
Alice (13)
s. Robert
s. Wm
Hadfield
Samuel (15)
Syddall
Emmott
Thornley
Isaac (13)
Fig. 10.3. Suggested sequence of infections during the winter in the second phase of the epidemic at Eyam. Scale: days after 30 September
1665. For further details and abbreviations, see Fig. 5.6.
271
272
in plague mortality) having infected her mother and sister Joan. Critically,
Anne Blackwall reinfected the family of Thomas Thorpe deceased (see
above) and also introduced the plague into the Alleyn family (Fig. 10.3).
Thomas Alleyn was infected on 1 March and died on 6 April having
infected his infant child.
Alice Thorpe (aged 13 years), both of whose parents had died in the
preceding September (section 10.3) was infected on 11 March and died on
15 April 1666. She infected her two brothers, Robert and William on 26
March and they both died on 2 May 1666. The disease was tenuously
maintained, therefore, all through the period December 1665 to May 1666
when there were few deaths.
The next slender link in the chain of events was the death of Emmott
Sydall (aged 22 years; Cliord 1989) on 29 April 1666. She had survived the
rst outbreak of plague in her family in the preceding October (see above)
but now was infected on 24 March either by Alice Thorpe (it will be
remembered that the Thorpes and Sydall families were originally near
neighbours) or by Thomas Alleyn. Emmott Sydalls mother Elizabeth had
also survived the October outbreak and she married a widower, John
Danyell (who later died of plague on 5 July), on 24 April 1666, just 5 days
before Emmott died and presumably before she showed symptoms. This
wedding was probably a critical event, because the next link in the chain
was Isaac Thornley (aged 13 years), who was infected on this day, either by
Emmott Sydall or by Robert and William Thorpe. Did Isaac Thornley
(who subsequently spread the epidemic to some 1517 people and so
caused the explosion in summer 1666; Fig. 10.4) contract the disease from
Emmott Sydall when they both attended the church for the wedding of her
mother?
Elizabeth Danyell (nee Sydall) died later from the plague intestate on 17
October (one of the last victims) but about the day of her death she made a
verbal declaration to her neighbour Rebecca Hawksworth, who testied to
that eect and a Letter of Administration was drawn up which armed
that her only surviving son Josiah (aged 3 years) was to be put in care of her
trusted friend Robert Thorpe and that her estate was to be used for his
upbringing (Cliord, 1989).
As at Penrith in 159798, we see that, when we have full details, the
plague at Eyam persisted from November to the following May only by the
skin of its teeth. Household contact rates remained high but the critical
interfamily infectivity was very low over this period. Figure 10.3 shows how
the plague, thanks to its long incubation period, persisted at Eyam over 6
Days
0
Thornley
Skydmore
Moor
Buxton
Heald
Lowe
Mellor
Townend
Archdale
Swanne
Elliott
Darbye
273
s. Isaac (13)
d. Anne
s. Jonathan (11)
m. Eliz.
s. Francis (1)
d. Jane (11)
s. Edward (11)
s. Anthony (10)
d. Mary (17)
w. Ann
s. James
d. Edytha (6)
f. Rowland
s. Rowland
d. Eliz. (18)
d. Mary (15)
d. Emmot
m. Eliz.
s. Thomas (12)
d. Sarah
Wm.
m. Ellenor
d. Eliz.
d. Anne (6)
Mary
w. Anne
d. Jane (19)
s. John (17)
s. Abel
s. John (9)
m. Eliz.
d. Deborah
George
20
40
60
80
100
P
S
S
S
S
S
S
S
S
T
S
S
Q
Q
S
S
T
T
Q
S
T
T
T
T
S
S
S
T
S
S/T
T
T
T
Fig. 10.4. Suggested sequence of infections during the start of the third phase of the
epidemic at Eyam. Isaac Thornley is regarded as a new primary case who initiated
the explosion in the late spring of 1665. Scale: days after 24 April 1665. For further
details and abbreviations, see Fig. 5.6.
months via a few crucial transmissions to new families. There must have
been many unrecorded autumnal outbreaks of plague in England in the
16th and 17th centuries that did not persist over winter.
It is interesting that so many of these interfamily infections at Eyam were
via the teenage children. They were responsible for going to each others
houses during the winter and so maintaining and spreading the epidemic.
(1)
Ellen
Skydmore
d. 1657
(2)
Francis
Thornley
d. 1664
Francis
bp. 1645
d. 1662
Isaac
bp. 1653
P
Thomas
=
Skydmore
d. 1657
Jane
bp. 1655
Thomas
Chapman
m. 1664
Jonathan
bp. 1655
m. 1638
(1)
Elizabeth
Chapman
Francis
bp. 1664
Anne
Edward
bp. 1655
Ann
P
Allyn
m. 1640
Arthur
bp. 1641
m.
Elizabeth
Walker
Richard
bp. 1642
d. 1643
Mary
bp. 1648
P
Elizabeth
bp. 1649
John
bp. 1651
Thomas
bp. 1652
Francis
bp.1653
d. 1656
Anthony
bp. 1656
P
Fig. 10.5. Interrelationships of the Thornley and Skydmore families. Plague victims marked with P in a circle. m., married;
d., died; bp., baptised.
275
276
277
Fig. 10.6. Diagram to show the deaths in the families in the nal stages of the
epidemic at Eyam, following the infections shown in Fig. 10.4. Each horizontal line
represents one family, arranged sequentially in order of the rst victim. Closed
circles, adults; open circles, children. Scales: days after 5 July 1666; LP, latent
period; IP, infectious period.
278
Finally, Abel, the son of Robert and Elizabeth Archdale, was the last
person to be infected by Isaac Thornley on 15 May and died on 20 June but
apparently no other members of the family were aected.
It is evident that the plague at Eyam entered its third phase and exploded
in June 1666 because of Isaac Thornley, whose key role corresponds to that
of George Viccars, who originally brought the plague into the village 9
months before. Once again, it was a youngster who was responsible for the
transmission of the disease between the families with a total of 17 secondary infections.
In Fig. 10.6 we illustrate how the epidemic exploded through the population, with 56 deaths in July and 78 in August 1666, but it is now impossible
to determine the lines of infection and the pattern of spread as the third
phase of the epidemic exploded.
The inhabitants of the extreme western end of the village, an area called
Shepherds Flatt, were few in number and they shut themselves up in their
houses and did not cross a small stream that divided them from the village
at Fiddlers Bridge. The story of how the plague came to two families on the
Flatt has been embroidered, but the registers suggest that again a youngster introduced the disease. Robert Kempe, aged 10 years, son of Lydia
Kempe, widow, died rst on 31 July 1666 and he infected the entire family:
Elizabeth (aged 19 years) on 11 August, Thomas (aged 23 years) on 12
August, Michael on 15 August and widow Kempe on 22 August. Their
neighbours on the Flatt, the Morten family, may have been also infected by
Robert Kempe; Sarah (aged under 4 years) died on 18 August, her mother
Margaret (nee Bagshaw) on 20 August and an infant on 24 August. The
father, Mathew Morten, was the only survivor at Shepherds Flatt (Wood,
1865). Rebecca Morten of Shepherds Flatt died of plague on 4 August 1666
but we have not been able to trace her relationship with Mathew Morten.
She could not have been infected by Robert Kempe, so that there were two
initial infections on the Flatt; she may have infected the family of Mathew
Morten.
Two features that have been recorded in our analyses of other plagues
are also evident at Eyam. Firstly, during the third phase of summer 1666
the families diered sharply; some showed the typical high household
contact rate, with secondary infections through most of the family (e.g. the
Talbotts, Bockinges and the Kempes); many families, in contrast, had only
the single initial infection; a few families, in contrast again, had multiple
initial infections in which the entire family may have died (e.g. the Hancockes and Naylours).
Secondly, when we examine the plague deaths in a family over the
279
12-month period of the epidemic, we see that, when the rst waves of the
infections in some families were completed, there were some members who
survived but were then infected from another source and died, sometimes
many months later. Some of these may have been orphans who were taken
into other peoples houses. The majority of these late reinfections must
have been exposed when the plague rst struck the family; were they
resistant or did they contract the disease and recover? If so, why did they
succumb later in the epidemic? One explanation might be that the initial
family infection was in the rst phase in autumn, when infectivity was low,
whereas the later victims succumbed in the summer of 1666 when the
infective agent was more virulent. Examples are Alexander Hadfeild (Mary
Coopers second husband) who died on 3 August, and Alice, Robert and
William Thorpe (see Fig. 10.2), who died in April and May 1666. However,
this explanation will not suce when the rst infection in the family was in
July or August (the Darbye, Taylor, Talbott and Glover families).
We see, again, that the basic unit of the epidemic was the household; the
disease usually spread readily within this unit because of a high contact
rate with a resultant high mortality to add to the plague statistics. But for
the plague to be perpetuated it had to be transmitted to other households
where, except in high summer, the eective contact rate appeared to be
lower. This dicult transmission during the autumn and winter was most
readily achieved into the households of relatives. Otherwise, the infective
agent was most frequently transmitted between households in the winter
and spring at Eyam by the youngsters. Unbeknownst to everybody they
would have been moving around, apparently completely healthy, in the
infectious state for about 20 days.
280
281
282
It is clear from the foregoing that at the time of the epidemic the people of
Eyam believed that it was caused by person-to-person infection or perhaps
by contagion. In contact with their neighbours and certainly when dealing
with strangers it was considered that the minimum safe distance was about
12 feet (Cliord, 1989), presumably out-of-doors.
The symptoms of the plague at Eyam are described by Wood (1865) as
shivering, nausea, headache and delirium. In some, these aections were so
mild as to be taken for slight indisposition
until a sudden faintness came on when the maculae, or plague-spot, the fatal token,
would soon appear on his breast, indicative of immediate death. But in most cases
the pain and delirium left no room for doubt: on the second or third day, buboes, or
carbuncles, arose about the groin and elsewhere; and if they could be made to
suppurate, recovery was probable, but if they resisted the eorts of nature, and the
skill of the physician, death was inevitable.
One of the symptoms of the plague was a sickly, sweet cloying sensation
in the nostrils. One evening, when Mompesson and his wife were returning
to the rectory, she is said to have exclaimed How sweet the air smells,
which lled the rector with alarm because he realised its import. She died a
few days later. Wood (1865) recorded that a tradition in the hamlet of
Curbar, 2 miles southeast of Eyam, during an isolated outbreak of plague
in 1632, was that a woman on leaving a house where some person was
suering from plague said to her husband Oh! my dear how sweet the air
smells. She took the distemper and died. Mompesson wrote after the
plague to his uncle My nose never smelt such noisome smells. It appears
that this sweet smell appeared just before the terminal stage when the
classic symptoms began. Could it be early evidence of necrosis of the
internal organs?
Undoubtedly some contracted the disease and recovered, as recorded by
Mompesson in his letters. Marshall Howe believed that he was immune
when he had recovered and Wood (1865) recorded that there was a general
belief that a person was never attacked twice. Others, apparently were
resistant; they were in close contact with infectives but did not contract the
disease. Mompesson is an example: throughout the epidemic he went
among his parishioners and administered to the dying; he helped them with
the writing of their wills; his wife died in his arms. He did not succumb,
although for much of the time he had a painful leg infection that his wife
feared was a symptom of the plague. Afterwards, he wrote,
the pest houses have been long empty . . . During this dreadful visitation, I have not
283
had the least symptom of disease, nor had I ever better health. My man had the
distemper, and upon the appearance of a tumour I gave him some chemical
antidotes, which operated, and after the rising broke he was very well. My maid
continued in good health, which was a blessing; for had she quailed, I should have
been ill set to have washed and gotten my provisions.
11
Continental Europe during the third age of
plagues: a study of large-scale
metapopulation dynamics
285
Fig. 11.1. Number of localities in which plague epidemics were reported in northwest Europe, 13471722. Area covered: Spain, Portugal, Italy, France, Britain,
Ireland, Benelux countries, Germany, Austria, Bohemia, Switzerland, Poland, the
Baltic, Scandinavia, Hungary, Croatia, Slovenia, Dalmatia and northwest of the
former USSR. Data from Biraben (1975).
14th and 15th centuries, and it includes mortality crises that were not the
result of infectious diseases. Nor are the data-series as valuable as the lists
of annual plague deaths in a city, as for London shown in Figs. 8.1 and 8.2.
Nevertheless the series, shown in Fig. 11.1, illustrates rstly the uctuating
spread of the plague and it can be seen that plague was present somewhere
in almost every year in Europe. The infection might be considered to be
pseudo-endemic, by which we mean that it was present in a handful of
widely scattered places every year in the metapopulation from which
infectives travelled out to start epidemics in fresh localities in the following
year. Secondly, the basal, endemic level rose steadily over the 300 years
after the Black Death, as the plague gradually established itself. Finally, the
plague ceased abruptly after 1670, as in England, with only erratic epidemics thereafter and these were not necessarily haemorrhagic plague. The
picture is complex, as would be expected, since it represents the sum of the
records from several constituent subpopulations that were widely separated and enjoyed dierent climatic conditions. Nevertheless, spectral
286
Fig. 11.2. Number of localities with plague in Europe, 13471666 (see Fig. 11.1)
ltered to reveal a medium-wavelength oscillation. Filter window : 20 to 30 years.
287
Fig. 11.3. Number of localities in France in which plague epidemics were reported,
13471722. Data from Biraben (1975).
288
Fig. 11.4. Number of localities with plague in France, 13471667 (see Fig. 11.3)
ltered to reveal a medium-wavelength oscillation. Filter window : 20 to 30 years.
It can be seen that these oscillations correspond broadly with those detected in the data-series for the whole of Europe and, indeed, we shall suggest
that the dynamics of the epidemics in France drove the oscillations in the
rest of the vast supermetapopulation of Europe.
There are few rm meteorological data for Europe over much of this
period of study but the date of the grape harvest in the vineyards of
northern and central France, Switzerland, Alsace and the Rhineland has
been taken as proxy for summer temperatures in central continental
Europe: late harvest dates are indicative of mostly cold average temperatures during the vine-growth period, April to October (Ladurie & Baulant,
1980). Figure 11.6 shows the number of days after 1 September before the
start of the grape harvest plotted annually, 14841668. Spectral analysis
reveals a 7-year oscillation therein (P : 0.05) that was strongly developed
after 1573 (P 0.005). This cross-correlates with the non-stationary, shortwavelength oscillation shown in Fig. 11.5 (ccf : 90.3) but with a 1-year
lag, i.e. warm summers were followed by widespread plague in France in
the following year. This rather weak correspondence does not necessarily
289
Fig. 11.5. Number of localities with plague in France, 13471667 (see Fig. 11.3)
ltered to reveal a short-wavelength oscillation. Filter window : 5 to 12 years.
290
Fig. 11.6. Annual date of the grape harvest in the vineyards of northern and central
France, Switzerland, Alsace and the Rhineland, 14841668. Ordinate: number of
days after 1 September before the start of the grape harvest. Data from Ladurie &
Baulant (1980).
291
Fig. 11.7. French corn price index, 14311688. Data from Baulant (1968).
travellers if they came from a town with plague; subsequently they also
expelled sick persons and required a bill of health from travellers. By 1520,
many towns had appointed plague bureaux charged with ensuring that
public health measures were enforced.
There were frequent and virulent outbreaks in France during 15201600
that were accompanied by food shortages, famines, ooding, peasant uprisings and religious wars. Epidemics occurred with greater frequency (see
Fig. 11.3) and, during this time, the health bureaux often hired armed men
to enforce the plague regulations and to maintain civic order. The epidemic
in 1564 in Lyons was particularly deadly and the town was almost paralysed after 2 months, with one-third of the houses closed; some of the
victims recovered (showing that not everyone died), only to die of hunger.
This plague spread widely through Provence and Languedoc in summer
1564, reaching Nimes in mid-July. Autumn brought a lull, although local
outbreaks were reported, most interestingly as late as mid-December until
a severe winter nally ended the epidemic (Kohn, 1995), showing that the
plague was sensitive to winter temperatures even in southern France.
292
Fig. 11.8. French corn price index, 14311688 (see Fig. 11.7) ltered to reveal a
short-wavelength oscillation. Filter window : 3 to 10 years.
293
Fig. 11.9. Cross-correlation between the French corn price index (Fig. 11.7) and the
number of localities with plague in France (Fig. 11.3), 14311531. Filter window : 3 to 10 years.
294
Fig. 11.10. Number of localities with plague in southern France (south of latitude
40 N), 13471688.
climate (south of latitude 46 N), a northern central zone and the coastal
zone of the north and northwest. The number of localities showing plague
summarised in Fig. 11.3 have been subdivided accordingly into these three
areas and the respective data-series are plotted in Figs. 11.10 to 11.12 note
the dierent scales on the ordinates.
Inspection of these plots shows that
(i) The south and central regions suered most severely in 162831 with
widespread plague, a period that was preceded and followed by times
of fewer attacks.
(ii) The south and central regions exhibited similar patterns of spread and
both showed a clear increase in frequency after 1450.
(iii) The coastal region is a smaller area and plagues there were relatively
less frequent than in the other regions of France; epidemics were
sporadic during 13501450 but became more widespread thereafter.
Spectral analysis reveals 21- and 13-year oscillations in the south, central
and coastal regions, 13471666, with the following statistical signicances:
295
Cycle
21-year
13-year
South
Central
Coastal
P : 0.005
P : 0.01
P : 0.005
P : 0.05
P : 0.05
NS
The 21-year oscillation was the stronger and both oscillations were of less
importance in the coastal regions of the west and northwest. The ltered
medium-wavelength oscillations in southern and central France are shown
in Figs. 11.13 and 11.14 respectively; they cross-correlate signicantly and
both also correlate with the coastal region (ccf : ;0.7 at zero lag), suggesting that these cycles were present throughout France.
296
Fig. 11.12. Number of localities with plague in the coastal region of north and
northwest France, 13471688.
297
tion of the outbreaks continued for the next 100 years, moving slightly to
the north and west (Fig. 11.15E), but as the 17th century progressed the
centroids became more erratic and widespread.
It can be seen that, although epidemics continued to break out in
southern France, the centroids were never located there after 1400; the
activity was concentrated around the centre of the metapopulation and
there was a considerable exchange of the infection with the Holy Roman
Empire to the northeast.
Table 11.1. Localities in France that recorded the largest number of outbreaks of plague in successive cohorts, 13471649
Cohort
134799
Strasburg (20)
Paris (16)
Avignon (12)
Burgundy (12)
Montpellier (10)
Toulouse (10)
Marseilles (7)
Provence (7)
Limoges (6)
Luxemburg (6)
Apt (5)
140099
145099
150049
155099
The number of epidemics suered by each locality in each cohort given in parentheses.
Data from Biraben (1975).
160049
Luxemburg (20)
Paris (20)
Bordeaux (19)
Amiens (18)
Angers (18)
Lille (17)
Nantes (15)
Rennes (15)
Troyes (15)
Lure (14)
Rouen (13)
299
most frequently aected must have lived almost continuously for over 100
years in fear of the pestilence.
The Black Death arrived at Strasburg in 1349 but thereafter the city was
free until 1358, after which it was repeatedly attacked for the next 150 years
(Table 11.1). The Middle Ages were Strasburgs golden period; its wealth
stemmed from the activity of its merchant class and also from its location at
the centre of numerous waterway and road communications. The Customs
House was built in 1358 to store and collect taxes on goods going through
the city: exports were textiles and grain, imports were glass, hides, fur, silk
and spices. Its location at the centre of trade routes was the reason that it
received regular visits from travelling infectives.
Bourg-en-Bresse was heavily and regularly hit by plague after 1400 and
epidemics continued to erupt there frequently for the next 200 years (Table
11.1). Probably the most important feature of the topography of Bourg-enBresse (where the rst plague hospitals were established in France in 1472)
was that it was the gateway to Geneva and to one of the passes over the
Alps. It lies on the western edge of the Jura mountains, some 37 miles
300
Fig. 11.15. Plague centroids in France. (A) 134799; (B) 140049; (C) 145099; (D)
150049; (E) 155099. The numbers on the gure are dates of years given at set
intervals.
301
302
northeast of Lyons, close to the rivers Ain, Saone and Rhone and to the
TroyesLyonsMarseilles trade route (see Fig. 11.22); all of these towns
also appear in Table 11.1. Lyons lay at the conuence of the Rhone and
Saone and attracted many foreign merchants with its four annual fairs. Its
silk weaving, printing and other industries supported a population estimated at 60 000 in the mid-16th century and it was frequently subject to
plague epidemics.
Visitations of the pestilence to Luxemburg are also included in Table
11.1. That city was strategically placed on what are now the boundaries of
modern Belgium, France and Germany and originated as a Roman crossroads. Siegfried, Count of the Ardennes, built a castle on the site in the year
963 that later developed into a formidable fortress, known as the Gibraltar
of the North, that was the origin of the town. After 1550, Luxemburg
suered from repeated and persistent epidemics.
To conclude: Table 11.1 and Fig. 11.15 show the important features of
the plague in France after the Black Death. After the progressive, wave-like
spread of the Great Pestilence, the pattern changed and the plague was
established as endemic; it was not present everywhere, rather grumbling
along, with an epidemic in a few widely scattered but important places
every year, spreading from which there may have been a clustering of local
11.3 Italy
303
11.3 Italy
Although Italy possessed a common language, local patriotism was too
strong to permit the growth of a national unity during the age of plagues
and it consisted of three main groupings of states:
(i) The City States of the North: Venice, Milan (Lombardy), Genoa and
Florence (Tuscany), immensely wealthy and jealous of one another.
(ii) The Papal States.
(iii) The Two Sicilies to the south, the most backward and poorest of
states, consisting of two very dierent regions, Naples and Sicily, and
ruled by the King of Aragon in Spain (Taylor & Morris, 1939).
It is evident from the work of Carmichael (1986) that, at least in northern
Italy, the authorities took public health measures during plague epidemics
very seriously, with physicians examining suspect cases and this point is
discussed in detail by Carmichael. They distinguished between major and
minor pests and there is the possibility that Italy suered outbreaks of
bubonic plague from an early date in addition to epidemics of what were
clearly haemorrhagic plague. The southern Mediterranean coastal climate,
proximity to seaports along the length of the land and a trading centre with
the Levant and North Africa all produce conditions that may have allowed
the development of bubonic plague epidemics that could have spread
inland but would not persist. Bubonic plague in Asia today is regarded as
an endemic disease of rural areas where an epizootic is established but
during the period after the Black Death minor epidemics of bubonic plague
in Italy may have begun in a port and then possibly have spread inland. It
is interesting that Carmichael (1986) recorded: In contrast to the devastating eects of plague on rural areas in the fourteenth century, the countryside was usually spared this scourge during the fteenth century. Thus the
304
Fig. 11.16. Number of localities in Italy in which plague epidemics were reported,
13471688. Major epidemics indicated. Data from Biraben (1975).
11.3 Italy
305
306
our studies of plagues in England. The symptoms were violet and black
spots and blotches (certain signs for which the Mantuan investigators
searched; Carmichael, 1986) or small red spots, indicative of a haemorrhagic disease, the bubo and pustules. Onset was marked with a high
temperature, vomiting, diarrhoea, cloudy urine and burning thirst (Cipolla,
1981). The acute fever in some was accompanied by madness and delirium
so much so that many hurled themselves out of windows (Carmichael,
1986).
The following reports of two autopsies of victims of the plague in
165657 in Rome and Naples respectively are of particular interest:
The exterior part of the body was found to be covered by black petechiae with a
black spot as large as a bean in the medial part of the right knee . . . The muscles of
the abdominal wall were of bad colour, the fat tissue very dry, the omentum rotten,
the guts all black, the peritoneum cyanotic, the stomach very thin, the spleen rotten,
the liver doubled in size but of bad colour and consistency, the gallbladder full of
black bile. Regarding the thorax, the pleurae were rotten, the pericardium very
hard, the mediastinum and the sagittal septum livid, the heart livid with its tip
black, both ventricles full of very dark blood. The lungs, of bad consistency and
colour, were all covered with black petechiae.
It was noticed that all the organs namely, the heart, lungs, liver, stomach, and
guts were covered with black spots. Moreover, the gallbladder was found full of
black bile which was very thick and fattish to the point that it adhered stickily to the
inner part of the gallbladder. Especially, however, the major vessels of the heart
were full of blood which was clotted and black.
(Cipolla, 1981)
The similarities between the foregoing and the case reports, description
of symptoms and autopsy reports of plagues in England, particularly in
London, are striking.
307
11.3 Italy
HOUSEHOLD
Days
0
Folchi
20
40
60
80
P
S
T
Dato
S
T
T
T
T
d'Andrea
Lorenzo
S
S
S
308
Days
0
Donato
20
40
60
P
S
S
S
S
S
Nardo Maestro
S
T
T
T
T
T
T
T
(ix)
(x)
(xi)
(xii)
The probable sequence of events shown in Fig. 11.19 has clear similarities with the epidemiology of the opening stages of the epidemics of
haemorrhagic plague in England. One primary probably infected only one
secondary (R : 1), who, in turn, infected four tertiaries (R : 4) and nine
quaternaries (R : 2).
11.3.4 Plague epidemics in Italy
Figure 11.16 reveals that, after the terrible mortality of the Black Death,
which worked its way progressively through the country, there were only a
limited number of epidemics in Italy during the following 300 years.
309
11.3 Italy
HOUSEHOLD
Days
0
A
B
C
D
E
F
G
H
I
J
K
L
20
40
60
80
100
(multiple cases)
S
S/T
T
T
T
Q
Q
Q
Q
Q
Q
Q
Q
Q
310
Fig. 11.20. Geographical distribution of the major plague epidemics in Italy. (A)
145658: open circle, Milan in 1452; closed circle, 1456; closed triangle, 1457; open
triangle, 1458. (B) 147678: closed triangle, 1476; closed diamond, 1477; closed
circle, 1478. (C) 152229: open circle, 1522; open triangle, 1523; closed triangle,
11.3 Italy
311
1524; closed circle, 1525; open square, 1526; open diamond, 1527; closed diamond,
1528; closed square, 1529. (D) 157477: open diamond, 1574; closed triangle, 1575;
closed circle, 1576; open circle, 1577. (E) (overleaf) 162931: closed triangle, 1629;
closed circle, 1630; open circle, 1631. Data from Biraben (1975).
312
11.3 Italy
313
314
11.3 Italy
315
11.3.5 Conclusions
The biology of the plague in Italy is quite dierent from the situation in
France where haemorrhagic plague was virtually endemic somewhere
throughout the 300-year period and France acted as the source for the
majority of the epidemics in the remainder of Europe. Plague was never
endemic in Italy, which we can consider as a separate metapopulation. It
probably suered from only about a dozen major epidemics in which the
mortality rates were often very high, probably on average 40% but sometimes reaching over 60%, as in Verona in 163031. Naples, with about
300 000 inhabitants, suered 150 000 deaths in 1656. The morbidity rate in
the countryside has been estimated at 66% (Benedictow, 1987). These
epidemics often lasted for 23 years in the big cities before burning out, a
clear dierence from the pattern in England, perhaps because of the
warmer climate in winter and early spring. Although the plague persisted
for longer in these cities in Italy, there was still a fall in the number of cases
over the winter, as shown by the number of patients admitted to the
pest-houses of Pistoia, Fig. 11.21 (Cipolla, 1981).
Each of these major epidemics was introduced from outside the
Fig. 11.22. Major trade routes of the Middle Ages in Europe. Ams, Amsterdam; Ant, Antwerp; Bag, Bagdad; Bas, Basel;
Br, Bruges; Col, Cologne; KL, Kings Lynn; Lo, London; Lu, Lubeck; Ma, Mainz; Mar, Marseilles; Mil, Milan; St,
Straslund.
11.3 Italy
317
Fig. 11.23. Number of localities in the Iberian peninsula in which plague epidemics
were reported, 13471688. Data from Biraben (1975).
318
successful to some extent in limiting the importations of the plague via the
northern ports on both the east and west coasts.
Apart from the major, devastating epidemics there were many years
when only a single, isolated outbreak was reported (see Fig. 11.16). Their
cause remains unknown; they may have been minor plagues that the
health authorities identied, or some may have been bubonic plague
brought in from the Levant or the ports of North Africa which infected the
local rats under favourable Mediterranean climatic conditions. Apparently, the epidemic did not spread far and soon died out because it was not
possible to form an epizootic with local resistant rodents.
15961602
319
320
Fig. 11.24. Geographical distribution of the major plague epidemics in Spain and
Portugal. (A) 15067: open circle, 1506 and 1507; closed circle, 1507. (B) 158081:
closed circle, 1580; open circle, 1581. (C) 15961602: open square, 1596; closed
diamond, 1597; closed triangle, 1598; open circle, 1599; closed circle, 1600; open
321
triangle, 1601; open diamond, 1602. (D) 164653: open square, 1646; open diamond,
1647; open triangle, 1648; open circle, 1649; closed square, 1650; closed triangle,
1651; closed circle, 1652; closed diamond, 1653.
322
164652
167882
323
Fig. 11.25. The last major epidemic in Spain, 167782. Open diamond, entry port in
1677; closed triangle, 1678; plague reported in Andalusia in 1679 but no localities
known; open circle, 1680; closed circle, 1681; open triangle, 1682.
spread in the hinterland). The pattern at Malaga is comparable to the spread of the plague at Marseilles in 1720 (Chapter
12).
In summary, the Iberian peninsula acted as a separate metapopulation
and probably experienced only four major epidemics of haemorrhagic
plague after the Black Death. Since it was never endemic, the pattern is
completely dierent from that of France. The data suggest that each
epidemic began by importation from overseas and did not spread from
France over the Pyrenees, which formed an eective isolation barrier. Each
of the four epidemics had its characteristic pattern, but Seville was implicated in all, twice acting as the point of entry. Seville shows little
temperature variation through the year, with long, dry and hot summers
and warm winters. It has acted for 2000 years as a port at the navigable
head of the Guadalquivir River, as the great market place of the Guadalquivir Valley and as the crossroads between the northeast and west of the
Iberian peninsula. The year 1492 is important in the history of Spain: it
marked the nal expulsion of the Moors, the unication of Spain under a
324
single crown and the discovery of America. For 200 years thereafter, Seville
became the gateway to the New World, the Mecca of European commerce
and the principal city of Spain. Plague epidemics were particularly
common in Andalusia and the area around Seville. Lisbon was the port of
entry for the plague that began in 1580 and Portugal was again hit in 1598.
In most of these epidemics, a radial or linear spread can be detected from
the foci, but a saltatory transmission, over distances of 200 miles can also
be seen, particularly from coastal regions to Madrid.
325
was relatively low, with an average of 10%. The greatest mortalities were in
1589 (28.8%), 1465 (20.0%) and 1530 (18.9%); if these years are excluded
the average is reduced to 8%, suggesting that these lesser outbreaks may
not have been typical haemorrhagic plague, or that the population had
developed resistance in the face of continuous attacks, although the average mortality during the rst 100 years was estimated at only 8%.
Where information is available (after 1452) the average duration of the
outbreaks was 8 months (range : 4 to 14 months), excluding the national
epidemics. The outbreak began in the autumn, was reduced in the winter
and began again in the following spring in 1457, 1466, 1490 and 1558,
reminiscent of the type (ii) epidemics recorded in England. The epidemic
that lasted from August 1475 to September 1476 was unusual: a small
number of deaths was recorded steadily over a 12-month period, including
JanuaryFebruary, with only a small peak. If the longer-lasting outbreaks
are excluded, the average duration of the seven remaining epidemics was 6
months, usually lasting from April to September, with a peak mortality
around June (see Table 11.2).
Are a mixed bag of lethal epidemics summarised in Table 11.2? Why did
they apparently not spread inland? Why did these one-o epidemics apparently cease after 1590? It will probably not be possible to answer these
questions with any degree of certainty, but some tentative suggestions can
be made. Barcelona in the 15th century was a city of merchants, navigators,
traders and professionals and it acted as a major port that had trac with
the whole of the Mediterranean and so could have received a regular
supply of infectives. The epidemics of 1507 and 165154 were probably part
of the national outbreaks of haemorrhagic plague, albeit with the slightly
dierent biology from those described in England because of the coastal
Mediterranean climate where, today, the monthly average temperature,
DecemberFebruary, is 1314 C. These outbreaks at Barcelona may have
been initiated separately from those in other parts of the Iberian peninsula
and in 1648 there was limited spread into the hinterland (Fig. 11.24D).
The other epidemics in which there was a mortality of over about 4000
may also have been haemorrhagic plague; some of these started in the
autumn, continued with a reduced mortality during the mild winter and
then broke out again in the following spring, perhaps a modied type (ii)
epidemic. However, the outbreak from October 1465 to November 1466,
with a mortality of 5000 (20%) stands out as being completely atypical
because Biraben (1975) has identied a steady build-up in mortality
through the autumn and winter that peaked in February and thereafter
declined steadily.
Start
Jan. 1362
1371
1375
1394
1408
1429
May 1439
May 1448
1452
May 1457?
Oct. 1465
Aug. 1475
Mar. 1483
Mar. 1490A
June 1494
Apr. 1501
Feb. 1507
May 1515
May 1520
Feb. 1530
Sept. 1557
Finish
1363
1396
1441
Nov. 1457
Nov. 1466
Sept. 1476
Sept. 1483
Sept. 1490
Oct. 1494
Nov. 1501
Nov. 1507B
Nov. 1515
Sept. 1520
July 1530
Aug. 1558C
Duration
(months)
Records
Peak
Estimated
total deaths
% age
mortality
Number
of other
localities
in Spain
where plague
reported
7
14
14
7
7
4
8
10
6
4
6
12
Daily
Monthly
Daily
Daily
Daily
Daily
Daily
Monthly
Daily
Daily
Daily
Daily
JulyAug.
Feb.
MayJune@
June
MayJune
July
JuneJuly
May
No clear peak
JuneJuly
April
MayJune
(5000)
(3000)
(1000)
(4000)
(2500)
(3000)
(3500)
(2000)
(1500)
3630
5000
2100
1400
3770
600
2650
3500
1160
1600
6400
4500
14.2
8.3
2.7
10.8
7.0
8.8
10.5
6.2
4.8
11.4
20.0
7.8
3.7
15.1
2.3
9.8
11.9
3.8
5.0
18.9
13.6
0
0
0
3
0
0
2
1
1
0
0
1
0
1
0
3
16
0
1
2
1
May 1589
Jan. 1651
Dec. 1589
Apr. 1654D
7
40
Daily
Aug.Sept.
12 400
20 000
28.8
45.5
Notes:
?Probably began in preceding year, but died down October to May. Type (ii) epidemic.
@Only a small peak; deaths recorded steadily over the 12-month period including the period January to February.
AVery small number of deaths recorded in the months November 1489 to February 1490. Type (ii) epidemic (?).
BThis was part of the great national epidemic of 15067, although there was a clear 3-month break, August to October 1507.
CDaily deaths through the winter. Type (ii) epidemic?
DBreaks: August to September 1652; September 1653. Part of the national epidemic.
Data from Biraben (1975).
0
16
328
The other epidemics with low mortalities may have been repeated, but
isolated, outbreaks of bubonic plague brought into the port from the
Levant or North Africa. Yersinia pestis has been identied in corpses from
the French Mediterranean coast from the 14th, 16th and 18th centuries
and, although it could apparently survive in the warmer climate, the
epidemics seem to have been short lived, persisting for up to 3 years,
probably because there were no local resistant rodent species in which a
permanent epizootic could be established.
Epidemics occurring every 810 years with a 10% mortality must have
had a severe impact on the demography of the city unless there was an
accompanying immigration to ll the ecological niches available. Biraben
(1975) has estimated that the population before the Black Death was 42 000
which had fallen progressively to 31 000 by the end of the 146566 epidemic. It fell further to a low of 25 000 by the end of the 15th century before
recovering steadily to 43 000 by 1589.
329
1450, the places recording frequent epidemics spread southwards along the
major trade route of the Middle Ages: LubeckHamburgMagdeburgNurembergAugsburg (see Fig. 11.22).
The situation changed after 1450 and continued until 1670; the plague
continued to erupt widely and epidemics persisted in the towns in the north
but the major foci, with the highest number of outbreaks, were in the
southwest, with Basel and Geneva now heading the lists, together with
Augsburg and other towns on the trade route to Milan, Innsbruck and
Venice. Augsburg experienced some 20 outbreaks of plague in the 16th
century in which a total of some 60 000 died (Kohn, 1995). Basel, situated
on the Rhine, was a large river port, Switzerlands only outlet to the sea and
the terminus of the Rhine navigation; for many centuries the Mittlere
Brucke in Basel was the only bridge on the Rhine. Basel suered from a
15-month epidemic in 161011 when, out of a population of some 15 000,
about 6000 contracted the disease, of which only approximately 3600 died.
Were some of the population showing partial immunity or resistance
because of previous exposure to the plague? The city also suered in
Table 11.3. Localities in Germany, Switzerland and Austria that recorded the largest number of outbreaks of plague in
successive cohorts, 13481670
Cohort
134899
Hamburg (10)
Lubeck (10)
Bremen (8)
Basel (8)
Mayence (6)
Cologne (4)
Magdeburg (4)
Wismar (4)
140049
145099
150049
155099
160070
Hamburg (7)
Hanover (5)
Nuremberg (5)
Bremen (5)
Lubeck (4)
Basel (4)
Vienna (3)
Nordhausen (3)
Geneva (8)
Basel (7)
Brunswick (7)
Nuremberg (6)
Hildesheim (5)
Hanover (5)
Cologne (4)
Luneburg (4)
Basel (12)
Augsburg (11)
Geneva (10)
Dresden (9)
Wittenberg (8)
Luneburg (7)
Cologne (5)
Erfurt (5)
Basel (19)
Geneva (12)
Bremen (12)
Augsburg (8)
Brunswick (8)
Dresden (7)
Nuremberg (5)
Hamburg (5)
Basel (15)
Augsburg (14)
Bremen (14)
Nuremberg (12)
Luneburg (12)
Hildesheim (12)
Dresden (11)
Geneva (10)
The number of epidemics suered by each locality in each cohort given in parentheses.
Data from Biraben (1975).
331
Switzerlands last epidemic in 166768 and a meeting was held in Bremgarten to which ocials were invited to discuss how they might prevent the
plague spreading from Basel (Kohn, 1995). There were major trade routes
from Basel to Lyons and along the Rhine to Mainz, Cologne and to the
port at Antwerp. Geneva also occupied a central position in Europe and
because of its early adherence to the principles of the Protestant Reformation became, at the beginning of the 16th century, a refuge for the persecuted and a starting-point for missionaries. It was therefore crowded with
immigrants and intending emigrants and was an ideal focus for plague
outbreaks.
332
conclude that the endemic status of the plague in France was only infrequently boosted by arrivals of infectives from overseas.
The Pyrenees to the southwest formed an eective barrier to the transmission of the plague and there is no evidence that the major epidemics in
the separate metapopulation of the Iberian peninsula came over this
mountain range; the infectives entered through the ports. There is no
signicant cross-correlation between the data-series for the Iberian peninsula (see Fig. 11.23) and the corresponding epidemics in southern or central
France.
The behaviour of the plague in Italy supercially resembles that in
Spain; it can be treated as a separate metapopulation and, although there
were many minor epidemics, major outbreaks were few. The health
authorities in the northern city states led the world in their control
measures and were probably successful in containing or preventing some of
the infections coming by sea from the Levant, North Africa, the Mediterranean islands and France. The Alps must have provided a partial barrier to
the spread of the plague from the north but, nevertheless, study of the
333
major epidemics suggests that some originated via this route, coming from
France or the Holy Roman Empire in central Europe.
Spectral analysis of the data-series of the annual number of localities
suering from plague in Italy (see Fig. 11.16) shows that a 12-year oscillation was present from 1347 to 1447 and changed thereafter to a signicant
25-year oscillation (P : 0.005). This medium-wavelength cycle cross-correlates signicantly after ltering with a comparable oscillation in southern
(ccf : ;0.65) and central (ccf : ;0.5) France. We do not suggest that
outbreaks in Italy (where plague was epidemic) were a driven system;
rather that, when plague was rampant in France, there was a greater
chance of infectives coming to Italy and starting epidemics.
The plague, obviously, did not observe political boundaries and we have
seen how epidemics broke out in Flanders and Luxemburg soon after the
Black Death and contributed to the endemic status of the disease in
France. Inspection of Fig. 11.26 shows that the plague was not endemic in
the Holy Roman Empire before 1550, although there were repeated outbreaks along the main trade routes (see section 11.5 and Fig. 11.22) and we
conclude that central Europe was repeatedly reinfected, probably usually
from France, during this time. Plague was present more continuously in the
Holy Roman Empire after 1550 and, for certain periods, approached an
endemic situation. The data-series for both France (see Figs. 11.3 and 11.4)
and Germany (see Fig. 11.26) show signicant medium-wavelength oscillations that cross-correlate signicantly. Cross-correlation studies suggest
that the medium-wavelength oscillations in the frequency of epidemics in
the metapopulation composed of Germany, Austria, Switzerland and Bohemia were initiated by comparable cycles in France, with infectives travelling long distances, usually along the dierent trade routes, and regularly
bringing the plague into central Europe.
On this basis, middle Europe (present-day France, Germany, the Benelux countries, Switzerland and Austria) can be viewed as a vast metapopulation around which the plague was carried by apparently healthy,
travelling infectives. We have, therefore, calculated the annual number of
localities where plague was reported over this huge area, using the data
given by Biraben (1975), and the results are shown in Fig. 11.28. The plague
spread progressively through the years and became more and more rmly
established, with a rising endemic level, culminating in the terrible outbreak which lasted from 1630 to 1637. The plague declined sharply thereafter and this is in marked contrast with the situation in England, where the
major outbreak was in 166566. The outbreak of 163037 was driven, as
usual, by France, where it appeared rst, and it did not rise markedly in
334
12.5 years
12.6 years
22.2 years
20.8 years
25.0 years
31.3 years
335
Fig. 11.29. The data-series of Fig. 11.28 ltered to reveal the medium-wavelength
oscillation. Filter window : 16 to 30 years.
We have identied in this chapter the major trade routes in the Middle
Ages running northsouth (see Fig. 11.22) by which the plague was frequently spread within each country but, for the eastwest transmission of
the infection between France and Germany, the cities of Basel, Geneva,
Bourg-en-Bresse, Strasburg and Luxemburg became key crossing-points
(Fig. 11.22) and suered from repeated epidemics (see Tables 11.1 and 11.3).
Inspection of the data suggests that the plague frequently circulated between these key towns.
The British Isles were separated by sea from this huge metapopulation of
northern and central Europe, so that the epidemics there were started by
infectives coming through the ports: Newcastle-upon-Tyne, Hull, York (via
the Humber and Ouse), the East Anglian river system, London and the
many ports, large and small, on the Kent and south coasts. We have
described the 14-year and 20- to 21-year, medium-wavelength oscillations
in the data-series for the frequency of epidemics in Britain (see Fig. 9.13) in
section 9.7.
The ltered medium-wavelength cycles for Britain and northern France
(lter window : 10 to 25 years) for the period 1347 to 1667 cross-correlate
336
moderately well, with a ccf of ;0.36. This correlation was poor for the
period before 1540, after which the major plagues emerged in the British
Isles. This medium-wavelength cycle in British epidemics shows better
correlation with the German series (see Fig. 11.26); overall the ccf is ;0.46
but, in contrast with the French series, the signicance is greater in earlier
years, 13471500 (ccf : ;0.67; Fig. 11.30). We conclude that, when plague
was raging in northern France or in Germany, there was a much greater
chance of apparently healthy infectives embarking on a voyage to England
across the North Sea or the English Channel.
A study of the extensive list of places where plague outbreaks were
reported in continental Europe (Biraben, 1975) shows, although it is not
complete, a dramatic change in 1500. Prior to that date, for the 150 years
after the Black Death, only very rarely were full-scale epidemics reported
from the ports along the northwestern French, Belgian and Dutch coasts,
from Brest to Amsterdam, and these would not have provided a ready
supply of infectives. Epidemics were, of course, widespread in Paris and the
337
12
The plague at Marseilles, 172022: an
outbreak of bubonic plague?
The epidemic that broke out in Marseilles in June 1720, some 50 years after
haemorrhagic plague is believed to have disappeared, has been described
and discussed in detail (Biraben, 1972, 1975; Bertrand, 1973), although the
details of its origins, arrival at the port and early infections are not
completely clear. These events are of critical importance in determining the
characteristics of the infectious agent. The diary of events in 1720 are
broadly as follows.
30 Jan.
3
5
27
28
4
6
17
April
April
April
April
May
May
May
25 May
27 May
31 May
3 June
12 June
339
14 June
340
North Africa where the Turkish passengers embarked. The voyage from
Syria to Libya had taken 9 weeks and was apparently trouble free. One of
the passengers died 2 days after joining the ship, whereas the next death
(the rst sailor) did not occur until 22 days later. Was this an epidemic of
bubonic plague? DNA sequences specic for Yersinia pestis have been
recorded from the dental pulp of plague victims buried at Marseilles in
1722 (Drancourt et al., 1998) and there is evidence from DNA studies of
bubonic plague around Provence in the 14th and 16th centuries. The
erratic and unpredictable sequence of deaths before the epidemic became
established in one area of the town, described in the diary above, might
suggest that one of the Turkish passengers who came aboard at Tripoli was
infected and that he brought with him infected eas and rats. When the rats
on board were infected in turn and an epidemic was established, bubonic
plague spread to the crew after an interval of 3 weeks. The outbreak spread
ashore via the rats and porters, guards and possibly the crew of the Grand
St Antoine and became established in an area of the town during late
Juneearly July.
Subsequent events of this important plague at Marseilles are now briey
described, with the possibility that Yersinia was responsible kept in mind.
The disagreement between the doctors when initially diagnosing the
disease suggests that the signs may have been dierent from the descriptions of haemorrhagic plague 50 years before. Most of those dying in the
early stages of the outbreak had buboes and carbuncles (Bertrand, 1973).
The faces of those dying presented dierent appearances: one pale, another
livid, another yellow, another violet. Some complained of acute pains in the
head and in all parts of the body, others were aicted with severe vomiting
or with violent swelling of the abdomen or with burning tumours.
On 1 August 1720 the doctors presented the following report of their
visit to the sick of Marseilles:
1 . . . the body of a women of sixty years of age, dead after an illness of three
days; on whom we found no marks of a pestilential disease on any part of the
body. We afterwards visited, in another house, a women of thirty-ve years of
age, who had a bubo in the groin . . . As we did not, however, perceive any
other symptom of a pestilential disease, we have reason to think that the
tumour proceeded from a very dierent malady.
2dly . . . the body of a girl of twenty years of age . . . she was seized with a violent
sickness and head-ache, accompanied with a general faintness, and died in
thirty hours after her seizure, covered with livid purple spots, having the belly
extremely distended, and of a purpule colour; and having discharged a great
quantity of blood at the nose in a very liqueed and serous state . . . attacked
with fever, pains in the head, and violent sickness . . . but in none did we nd
341
There is evidence from the accounts that the violence of the disease
ameliorated in the later stages of the epidemic. Some of those infected
suered only a few days of fever without tumours, others had an eruption
that soon regressed and thus a greater proportion of the patients in the
autumn recovered. However, many of these experienced a fresh attack in
the following spring and there is no doubt that people in this plague could
be infected a second time, even at the height of the epidemic (Bertrand,
1973).
It is noteworthy that Roberts (1966), when reviewing the literature on
plagues in England for the meeting of the Royal Society of Medicine to
commemorate the Tercentenary of the Plague of London in 1665 related in
a report of the 1720 plague of Marseilles, that the shermen netted ten
thousand dead animals in the harbour, and dragged the corpses out to sea.
Nowhere else has a reference to destruction of rats been found [our italics].
The mortality was enormous; 10 000 people ed before the cordon
sanitaire was established and some 50% of the 80 000 who remained died.
This is probably a higher mortality than would be expected with bubonic
plague, perhaps suggesting that pneumonic plague was responsible for a
high proportion of the deaths (see Wu, 1936). The pattern of the outbreak is
342
Fig. 12.1. Time-course of the epidemic at Marseilles in 1720: the number of reported
plague deaths, July to December.
shown in Fig. 12.1: the deaths were largely conned to August and September 1720, a shorter time-scale than that typically recorded in a major
outbreak of haemorrhagic plague. The mortality fell sharply during the
autumn but a few cases were reported in the following year between 18
April and 19 August 1721, when the last case was reported and severe
quarantine measures were introduced that lasted until 9 November (i.e.
twice as long as the 40-day quarantine period usually decreed in England,
France and Italy). No new cases were reported during the quarantine
period, nor during the following 3 months. However, the plague began
again on 3 February 1722; 260 persons were attacked, of which 194 died
(75%) and draconian quarantine measures were reintroduced (Biraben,
1972). In previous epidemics throughout Europe, a 40-day quarantine
period after the last death was accepted as safe because the disease did not
reappear after this time. But human quarantine periods are not applicable
to rodent and ea populations and this is good evidence for an outbreak of
bubonic plague.
Fig. 12.2. Spread of the plague from Marseilles in 172022. Dashed and dotted lines represent approximate extent of
the outbreak by 30 December 1720 and 30 December 1721, respectively. The main foci at Aix, Apt and Avignon (see
Fig. 12.3) and at Mende and Marvejols (see Fig. 12.4) are indicated. Populations: large closed circle, over 10 000;
smaller closed circles, over 1000. Open circles, localities unaected by plague. After Biraben (1975).
344
Fig. 12.3. Detailed reconstruction of the plague epidemic in the areas around Aix,
Apt and Avignon. Dates indicate the rst appearance of the plague in each of
the localities which are numbered according to Table 12.1. Populations: closed
circles, over 1000; open circles, 2001000; closed triangles, less than 200. After
Biraben (1975).
345
Table 12.1. Sequential appearance of the plague in the area around Aix, Apt
and Avignon
Locality
Date of
appearance
of plague
Estimated
population
in 1720
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
18.1720
18.1720
228.1720
258.1720
298.1720
298.1720
229.1720
229.1720
239.1720
299.1720
110.1720
610.1720
610.1720
910.1720
111.1720
111.1720
111.1720
1411.1720
1611.1720
2611.1720
212.1720
1712.1720
2912.1720
71.1721
127.1721
158.1721
158.1721
168.1721
179.1721
209.1721
259.1721
1510.1721
1510.1721
1510.1721
1510.1721
Oct 1721
Oct 1721
Oct 1721
Oct 1721
Oct 1721
Oct 1721
611.1721
?Dec 1721
28 000
4900
1700
1000
800
100
1250
750
23 041
1250
2700
3000
2000
100
410
3000
4200
700
1300
23 170
250
7000
1750
750
60
1000
1400
1228
2500
5000
Aix
Apt
Lanc on
Roussillon
Le-Puy-Sainte-Reparade
Saint-Canadet
Saint-Cannat
Caseneuve
Avignon
Saint-Martin-de-Castillon
Cucuron
Perthuis
Pelissanne
Villars-Brancas
Venelles
Saint-Remy
Salon
Rustrel
Caumont-sur-Durance
Arles
Vaugines
Tarascon
Orgon
Maillane
Boulbon
Graveson
Bedarrides
Noves
Lambesc
Orange
Saint-Genais
Chateauneuf-du-Pape
Sorgues
Monteux
Cavaillon
Caromb
Aubignan
Bonnieux
Crillon
Montfavet
Morie`res
Pernes
Robion
900
2600
5000
3000
1500
2900
650
70
1000
3500
1200
Mortality
%
23.9
5.5
4.8
13.8
5.4
34.0
27.4
2.4
31.4
26.1
30.8
12.1
11.2
12.0
8.1
33.6
21.7
1.9
44.3
13.2
3.0
6.0
14.1
31.7
1.1
22.0
14.5
8.2
346
Locality
Date of
appearance
of plague
Estimated
population
in 1720
Mortality
%
44
45
46
47
9.5.1722
9.5.1722
?May 1722
?May 1722
1750
2400
650
800
10.1
Sarrians
Le Thor
Saumane
Vede`nes
(i) on the west bank of the Rhone; (ii) along lines in an easterly direction
from the Rhone, north and south of Avignon and south of Arles; (iii) along
the River Orb; and (iv) in a ring around the Languedoc area. The outbreak
had not spread far before 20 August (Fig. 12.3; Table 12.1) but these
cordons sanitaires were completely ineective and the plague eventually
spread to over 250 other localities in Provence and neighbouring districts
(Fig. 12.2). This is in contrast with the situation at Eyam, 55 years previously, where none outside the cordon around the village was infected. We
conclude that this dierence in epidemiology is suggestive evidence of
bubonic plague spread by rodents in Provence in 172022.
The times of appearance of the plague in the hamlets, villages and towns
is listed sequentially in Table 12.1 and its initial geographical spread is
shown in Fig. 12.3. In 2 months, by the end of September 1720, the
epidemic had: (i) spread out radially in a cluster in the environs of Marseilles; (ii) appeared in Toulon and started to spread therefrom (Fig. 12.2);
(iii) spread locally in clusters from its starting points at Aix and Apt; and (iv)
appeared on 23 September at Avignon, 28 miles from Apt, where it persisted for 11 months. The epidemic then spread outwards from these foci so
that over 50 localities were aected by the end of the year: (i) a tongue
began to spread eastwards from Marseilles towards Brignoles (Fig. 12.2);
(ii) Arles was rst aected on 26 November 1720, the spread being from the
environs of either Aix or Avignon; and (iii) plague appeared at its furthest
westward point at the village of Auxillac in the Languedoc, presumably
having travelled from Avignon, and 8 days later it appeared at nearby La
Canourgue, where over 900 died (55% of the population) (see Fig. 12.4).
The plague did not spread from here at rst, but this outbreak acted later as
the centre for an epidemic in the Mende district that began in summer 1721
and aected a cluster of 45 villages and hamlets (Table 12.2; Fig. 12.4).
347
Fig. 12.4. Detailed reconstruction of the plague epidemic in the areas around
Mende and Marvejols. Each locality is numbered according to Table 12.2 and the
dates indicate the rst appearance of the plague. Populations: closed circles, over
1000; open circles, 2001000; closed triangles, less than 200. Note the small size of
the localities aected in this rural area. After Biraben (1975).
The epidemic continued to spread slowly from these foci to the east of
the Rhone, aecting neighbouring villages in January 1721, but was markedly slower during February to the end of May 1721. Thereafter it exploded again during June to September 1721, again spreading to neighbouring villages and towns and establishing another focus in the region
around Alais to the west of the Rhone.
This outbreak in the countryside around Marseilles to the east of the
Rhone, therefore, lasted about 16 months although it persisted until summer 1722 in some localities in the Languedoc area. It spread a maximum
distance from Marseilles of 135 miles. In general, its movement was gradual, travelling to nearby localities, but there were examples of apparently
saltatory behaviour. This was a rural outbreak with the epidemic clustering
in a great many localities in a relatively small area of the countryside. The
348
Table 12.2. Sequential appearance of the plague in the rural areas around
Mende and Marvejols
Locality
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Correjac
La Canourgue
Banassac
La Capelle
Brunaves
Saint-Frezal
Maleville
Marvejols
Cadoule
Saint-Georges-de-Levejac
Serres
La Calcidouze
La Masmontet
La Bastide
Tremoulis
Ispagnac
Molines
Biesse
Saint-Leger-de-Peyre
Valadou
Rechiniac
Fontjulien
Mende
Badaroux
Quezac
Le Buisson
Les Gratoux
Gabriac
Chirac
Grezes
Les Caynoux
Saint-Germain-du-Teil
Montagudet
Malbousquet
Montrodat
Vimenet
Inosses
Le Mas-Andre
Le Chambonnet
Chastelnouvel
Le Crouzet
Alteyrac
Balsie`ge
Barjac
Date of
appearance
of plague
2311.1720
812.1720
315.1721
296.1721
296.1721
17.1721
17.1721
57.1721
107.1721
127.1721
127.1721
157.1721
227.1721
18.1721
28.1721
178.1721
178.1721
178.1721
188.1721
308.1721
308.1721
29.1721
39.1721
39.1721
209.1721
209.1721
209.1721
259.1721
End Sept 1721
510.1721
510.1721
1010.1721
1010.1721
1010.1721
1010.1721
1010.1721
1010.1721
1010.1721
1010.1721
3010.1721
3010.1721
3010.1721
3110.1721
3110.1721
Estimated
population
in 1720
Mortality
%
109
1633
900
200
61.5
56.058.0
23.8
43.5
110
40.0
2756
400
40
12
15.2
20
1800
11.8
2000
19.0
87
3800
28.0
1150
7.0
130
850
400
6.9
1.3
41.5
500
32.0
200
22.0
225
349
Locality
Date of
appearance
of plague
45
46
47
48
49
50
51
3110.1721
1011.1721
2111.1721
312.1721
712.1721
712.1721
2412.1721
Lasfons
Le Masrouch
Lannuejols
Les Bories
La Serre
Boudoux
Bramonas
Estimated
population
in 1720
Mortality
%
350
351
13
Conclusions
353
al., 1996; OBrien & Dean, 1997). New mutations would have a very high
likelihood of being lost within a few dozen generations and it is highly
unlikely that a single CCR5-
32 variant as a strictly neutral mutation did
increase to modern frequencies across Europe/Asia by random genetic
drift.
A recent survey of over 4000 individuals in 38 ethnic populations revealed a cline of CCR5-
32 allele frequencies of 0% to 14% across Eurasia,
whereas the variant is absent among native African, American Indian and
East Asian ethnic groups. This might explain the rapid spread of HIV in
sub-Saharan Africa, whereas possession of the CCR5-
32 allele may have
delayed its progress in Europe. The time of the origin of the CCR5-
32
mutation has been estimated on the basis of the persistence of a common
and likely ancestral three-locus haplotype retained in linkage disequilibrium. The age of the CCR5-
32-bearing haplotype has been computed by
these methods to be approximately 700 years old (but with a wide range of
2751875 years) and it has been suggested that this haplotype was driven
upwards to the present-day frequencies of 5% to 15% by a historic, strong
selective event, probably an enormous mortality mediated by a widespread
epidemic of a pathogen that, like HIV-1, utilised CCR5 for entry into
lymphoid cells (Stephens et al., 1998). The Black Death is an excellent
candidate for such a catastrophic event and it has been suggested that the
epidemic provided the strong selective pressure that drove up the CCR5
32 mutation some 650 years ago (Stephens et al., 1998). However,
Stephens et al. (1998) assumed, of course, that the Black Death was caused
by the plague baccillus, Yersinia pestis, which carries a 70 kilobase plasmid
that encodes an eector protein, Yop1, which enters macrophages, causing
diminished immune defences. They inferred that this bacterial protein
enters via the CCR5 receptor.
Carrington et al. (1997) went further and concluded that the high predominance of codon-altering alleles among CCR5 mutants is consistent
with an adaptive accumulation of function-altering alleles for this gene,
perhaps as a consequence of historic selective pressures. Mummidi et al.
(1998) also drew attention to the complex array of polymorphisms and
genetic determinants in the HIVhost interplay. Libert et al. (1998) concluded that most, if not all,
CCR5 alleles originated from a single mutation event which probably took place a few thousand years ago in
northeastern Europe, which would pre-date the Black Death. They agreed
that the high frequency of the CCR5-
32 allele in Caucasian populations
cannot be readily explained by random genetic drift and that a selective
advantage is, or has been, associated with homo- or heterozygous carriers.
354
Conclusions
The Black Death may therefore have been a major factor in a continuum of
plague epidemics that drove up the frequency of CCR5 mutants (see also
Rottman et al., 1997; Samson et al., 1997; Kirchho et al., 1997; Simmons et
al., 1997; Madani et al., 1998; Rizzuto et al., 1998; Stephens et al., 1998;
Wang et al., 1998; Zhang et al., 1998; Hussain et al., 1998).
Thus these recent studies of the receptor for the HIV virus (Stephens et
al., 1998), surprisingly, reveal new information about the probable etiology
of the Black Death, the pandemic of 650 years ago. Since the CCR5-
32
mutation provides protection against the entry of a virus (HIV-1) into
lymphoid cells, the causative agent of the Great Pestilence was probably
also viral and targeted macrophages and monocytes. The continuation of
plague epidemics for the next 300 years probably maintained and boosted
the CCR5-
32 mutation and this continuing historic selection pressure
also accounted for the appearance, accumulation and persistence of the
additional mutations that have now been identied.
13.2 Was the same causative agent responsible for all the plagues in
England from 1348 to 1666?
This is an impossible question to answer because we have no detailed
information on the seasonal pattern of the deaths in the Great Pestilence
and can rely only on the institutions to vacant beneces in the dierent
dioceses that have been so carefully derived and collated by Shrewsbury
(1970) (see Fig. 4.2). Indeed, until parish registers began to be kept in the
mid-16th century we have little rm quantitative evidence of the pattern of
any of the epidemics. Nevertheless, the seasonal pattern of the Black Death
in 1349 in the dioceses of Hereford and Gloucester, Licheld, Norwich,
Lincoln and Ely show a remarkable similarity to the epidemics seen in
towns and cities in the provinces in the 17th century in England; each
outbreak lasted for the characteristic lengthy period of 912 months, with a
clear peak in JulyAugust.
The limited and largely anecdotal on-the-spot descriptions of the symptoms and autopsy reports of such plagues as the Black Death (section 4.2)
and the subsequent epidemics in Italy (section 11.3.2), London (section
8.6.3) and Kendal (section 7.9), are broadly similar.
Whenever Shrewsbury described an epidemic that even he saw could not
be bubonic plague because of the seasonality or other features such as a
high mortality, he resorted to averring that it must have been an outbreak
of typhus. Inspection of contemporaneous records suggests the exact reverse: it is evident that, certainly by the 16th century, they were able to
355
identify accurately the pestilence and they could distinguish it from the
English Sweating Sickness (section 6.1). They were required by law in
Elizabethan times to record plague deaths in the parish registers and they
certainly came to recognise the typical symptoms such as the tokens
(section 8.6.3). Furthermore, typhus epidemics do not follow Reed and
Frost dynamics because it is a disease with an arthropod vector.
There were probably many minor outbreaks in isolated families where
the plague went unreported. Equally, there may have been many small and
contained outbreaks that were faithfully recorded in the parish registers
between 1560 and 1660 (e.g. Slaidburn, 1632; section 9.4) and would be of
considerable interest; we should be very pleased to receive the details of
such records from people searching the registers.
In conclusion, all the major plague epidemics exhibited Reed and Frost
dynamics typical of an infectious disease with a long incubation period. All
the results are consistent with latent and infectious periods of about 1012
days and 2527 days, respectively. We believe, therefore, that the same
causative agent was probably responsible for all the plagues in England
from 1348 to 1666.
However, there are features of the Black Death that distinguish it from
the well-documented plagues after about 1560. In the diocese of Bath &
Wells, the epidemic began in autumn 1348, was at its peak in mid-winter in
January and was nished by April (see Fig. 4.2B). In the diocese of Salisbury, the epidemic was rampant through December to February and
reached its peak in March (see Fig. 4.2A). Shrewsbury (1970) was unable to
explain these mid-winter explosions as bubonic plague. This seasonal
pattern is dierent from the usual behaviour of the plague in the 15th
century in England, where the epidemic struggled to continue through the
winter and infectivity between households was low (sections 13.4 and 13.5),
although there are a few accounts of the plague peaking in winter in
England in the 17th century, for example at Durham (section 9.5.1), Ottery
St Mary (section 9.5.2) and Colchester (section 9.6.2). The Black Death
arrived in England in summer 1348 and a major epidemic occurred whenever it reached a community, irrespective of season, albeit with a slow
build-up because of the long incubation period. It arrived in the West
Country in autumn 1348 and a winter epidemic ensued. All the evidence
suggests, therefore, that the Black Death struck a completely naive metapopulation in England and infectivity was high irrespective of the time of
year. As the years went by, with the seemingly interminable succession of
plagues in the commonly infected foci, there would be changes in resistance
(see the CCR5-
32 receptor, section 13.1) and in the proportion of the
356
Conclusions
(iii)
(iv)
(v)
(vi)
357
358
Conclusions
(ix)
(x)
(xi)
(xii)
(xiii)
359
characteristics. The build-up of the epidemic was slow, the consequence of a long incubation period, but thereafter it exploded and
followed typical Reed and Frost dynamics.
Yersinia is a bacterium whereas the studies with the CCR5 receptor
suggest that the infectious agent in the plague was viral (see section
13.1). It is noteworthy that the CCR5-
32 mutation, which is supposed to provide protection from bubonic plague, is not found in
ethnic populations from eastern Asia, the area where the disease has
been endemic for many years.
Because bubonic plague is dependent on the biology and behaviour
of eas, rodents and humans, the outbreaks have complex dynamics
and appear to be haphazard. In contrast, plague epidemics were
sharply dened and predictable and all followed a standard pattern
(see our analyses in dierent parishes in dierent years). Shrewsbury
(1970) was aware of this marked discrepancy.
Mortality in outbreaks of bubonic plague is always relatively low
whereas, where we have detailed information, the death toll when
the pestilence struck a naive population in England or continental
Europe was probably about 30% to 40%. Plague mortality was
probably even higher during the Black Death, although we have no
concrete evidence. In Venice in 134748 and Genoa in 165657,
60% of the population is estimated to have died; half the population
of Milan died in an outbreak of plague in 1630, and perhaps half the
population of Padua in 1405 and of Lyons in 162829; the death toll
reached 30% in Venice in 163031 (Slack, 1988). Over the period
October 1630 to August 1631, 1198 patients were admitted to the
pest-houses in Pistoia in Tuscany and, of these, 607 (50%) died
(Cipolla, 1981). Again, Shrewsbury (1970) was aware of this major
discrepancy and was forced to the conclusion that the mortality in
the Great Pestilence was greatly overestimated.
Endemic bubonic plague is essentially a rural disease because it is an
infection of rodents. The Black Death, in contrast, struck indiscriminately in the countryside and in the towns. Major epidemics of later
plagues were mostly, but not entirely, conned to the towns. In
summary, haemorrhagic plague epidemics were density dependent
whereas bubonic plague (as in the Mende area at Marseilles) was
not.
Since bubonic plague is a disease of rodents, the arrival of an
outbreak is frequently presaged by rats dying in the streets and yet
Shrewsbury (1970), who believed that Yersinia was the infectious
360
Conclusions
361
362
Conclusions
that this was via the agency of ectoparasites, the human ea and
louse. Carmichael (1991) commented on the pestilence in 15th century Milan: Were these plagues actually contagious? If eye-witnesses insist that the identied cases of plague evidenced contagion,
is there any reason we should doubt that precious testimony?
Cipolla (1981) said, it is dicult to believe that the same blocked
ea kept jumping from one individual to the next, bringing the
plague to three separate households and he continued that the case
in question seems to support the possibility of person-to-person
transmission via the human ea.
(xix) Haemorrhagic plague disappeared in about 1670 but bubonic
plague remained to the present day in its foci in Asia, and epidemics
probably continued to break out in Mediterranean areas, including
Marseilles in 1722. Furthermore, the brown rat arrived in Europe in
the 18th century and would potentially have been a more eective
host for bubonic plague.
(xx) The characteristic signs of haemorrhagic plague were the tokens but
apparently these were not regular features of bubonic plague.
Perhaps the most telling point is that the accounts given in Chapters
411 form a consistent pattern and, taking them together and considering
them objectively, it is impossible to conceive of the plagues not being a
disease spread through contact with an infected person. Indeed, apart from
the fact that the victims of both diseases presented with enlarged glands
and subcutaneous swellings, it is dicult to suggest a more unlikely candidate than Yersinia as the infectious agent of haemorrhagic plague. In the
1720s, 60 years after the disappearance of the plague, London was swept by
a series of epidemic fevers which were characterised by buboes and carbuncles (Creighton, 1894), conrming that these clinical signs were not
exclusively diagnostic of bubonic plague.
363
with the last plague deaths in November or December. Typical duration of the epidemic: 89 months; typical mortality for a naive
population: 30% to 50%.
A variant of a type (i) epidemic was at its peak in winter; it seems to
have been recorded only during the Black Death and a few outbreaks
in the provinces thereafter. Duration, 68 months; mortality, probably about 50%. Examples: dioceses of Salisbury and Bath & Wells,
134849; Durham, 1644; Ottery St. Mary, 1645.
Also included in this category are isolated, small-scale outbreaks in
late spring or summer conned to one or two families or to a small
hamlet. Epidemic spread restricted by density-dependent factors.
Duration, about 2 months; mortality, probably under 20 persons.
Type (ii): The epidemic began with the arrival of an infective in late
August or September who had probably travelled from a locality that
was experiencing its type (i) summer peak. The outbreak developed
slowly, as usual, and there was a small autumnal peak of plague
burials but the spread of the infection was dramatically reduced by the
cold weather of December. The epidemic broke out again in spring
with the slow build-up and major peak in JulyAugust as in type (i)
epidemics. Typical duration of total outbreak: 14 months; typical total
mortality of a naive population: 30% to 40%. Examples: Penrith,
159798; Eyam, 166566; Colyton, 164546; Chestereld, 158687. A
variant of this type of epidemic was seen at Shrewsbury in 166566
and at Colchester in 1665, which began in summer and died down in
winter, as in type (i) epidemics, but reappeared in the following spring.
Another variant was an epidemic with a small autumn peak but the
infection died out completely over winter. Typical duration, 34
months; mortality, very low.
As we have shown, haemorrhagic plague was characterised by its long
incubation period, namely a latent period of 1012 days, an infectious
period before the symptoms appeared of 2022 days and about a 5-day
period of showing symptoms. It is possible that the latent period in London
was slightly shorter, with some people within the same household being
infected after 910 days and possibly only 7 days after the primary was
infected. The outbreak at Neston in 1665 (section 9.6.1; Fig. 9.11) establishes 35 days as the minimum duration of the latent and infectious periods
and there is a 36-day gap in the outbreak at the parish of St Michael
Bassishaw in London in 1641 (see Fig. 8.11). Our analyses of the London
epidemics suggest that, exceptionally, the latent and infectious periods may
364
Conclusions
have been extended beyond 37 days (up to 42 days or even longer) in cold
weather in winter.
365
366
Conclusions
cold winter months but would have emerged to work in the elds during
the labour-intensive months of spring and summer. Not only would they
establish more interhousehold contacts, but hard physical work on a
calorie-decient diet can only have exacerbated the physiological stresses
and reduced resistance to disease. The arrival of the better weather in
spring would also have promoted the movement of travellers within the
metapopulation (see section 13.9), so bringing the infection to the rural
communities. Social intercourse would be at its peak during the summer,
not only during hay-making and harvest, but also with the long-established fairs at which people from all over the country assembled. A symptomless infective could spread the infection widely at such events and this
was recognised by the authorities who cancelled the fairs, particularly
those near London, once a plague epidemic was ocially declared.
367
368
Conclusions
369
370
Conclusions
Italy: summers too hot and dry; physically smaller; partially eective
quarantine at the ports.
Holy Roman Empire: low winter temperatures.
England: low winter temperatures; infectives could come only by sea.
371
372
Conclusions
373
Fig. 13.1. The transportation system of medieval England. The major towns were
linked by road and river. Note (i) the northeast corridor where plague was readily
transmitted between Newcastle and York, (ii) the river systems of East Anglia,
which brought both trade and the plague into the country, and (iii) the paucity of
roads in Wales where there was little plague. Abbreviations: B, Bristol; Ca, Carlisle;
Ch, Chester; D, Durham; H, Hull; KL, Kings Lynn; L, London; N, Newcastle; No,
Norwich; Sh, Shrewsbury; Y, Yarmouth; Yo, York. After Hindle (1993).
appearing many miles away from the point of infection in both England
and France. This behaviour was possible because of the long incubation
period of the disease; an example is the outbreak at Malpas in 1625, which
was brought almost 200 miles by a member of the family from a plague in
London (see section 9.3).
374
Conclusions
Table 13.1. Imports and exports via the ports of Boston and Kings Lynn in
East Anglia from 1300
Trading centre
Iceland
Bergen
The Baltic
Germany
The Low Countries and
Calais
France
Imports
Exports
Fish
Fish, timber, furs
Fish, oil, furs, wax
Fish
Cloth, corn
Cloth, corn
Cloth
Cloth
Wool, cloth
Cloth, corn, sh
The great fairs which were held annually were probably a major means
of disseminating the plague in England (section 13.8) and possibly also a
source for bringing the infection into the country. This was recognised by
the authorities in the later years and fairs were cancelled once a plague was
ocially declared to have broken out. The major fairs lay south and east of
a line from Exeter to York and were situated on or near the navigable rivers
linking them to seaports or in places where there was an easy overland haul
from the nearest port.
The great fair at Sturbridge [in East Anglia], which opened on 18 September and
lasted for three weeks, was the most important of the English fairs and was of
sucient repute to attract merchants from many European trade centres. You can
be present at the great Sturbridge fair and there see Venetian glass, Bruges linen,
Spanish iron, Norwegian tar, Hanse fur, Cornish tin and Cretan wine, all for sale in
the half of a square mile which was occupied for three whole weeks . . . during
Sturbridge Fair, Blakeney, Colchester, Kings Lynn, and perhaps Norwich, were
lled with foreign vessels. The ships that brought the merchandise of the Levant
from Venice, and the other commodities from overseas . . .
(Shrewsbury, 1970)
September 18th was probably too late for the initiation of a major, type (i)
epidemic in that year but there were over a dozen other fairs and the Great
Pestilence is said to have erupted in London at the time of the St Bartholomew fair.
The open market in rural English towns catered primarily for local
demand and dealt in relatively small transactions; the private trader, by
contrast, was essentially a traveller and an individualist. He traversed alone
the southern heaths and northern moors in search of livestock, corn, or
wool; if necessary he was prepared to pluck up his roots and plant anew in
other cities and villages; he was ready to forsake home and family in search
375
of individual reward. In the inns, where he bought and sold, he met others
of his own kind, and there he discussed with them the burning issues of the
time (Thirsk, 1967) and, more importantly for our study, he must have
provided the ideal means for the transmission of the plague, talking at
length with the locals indoors.
The medieval wool trade had the greatest impact on long-range movement within the metapopulation of England, the handling and transportation of the eeces and the nished goods proceeding throughout the year
on a large scale. Sheep were usually sheared in June after washing, which
may have been done by hired workmen who, like the shearers, moved
round the country. After sale, the wool was packed and delivered to the
buyer either immediately or at some later date. The clothiers and middlemen themselves frequently came with transport to fetch wool from a
growers house, sometimes in instalments, although carriers were also
employed. Leaden Hall in the City of London, where the staplers (wool
merchants) sold much of their wool, was not only a market but was also the
largest wool warehouse in England.
The middlemen either stored their wool for a number of months or sold
them almost immediately. One large Shropshire dealer, for example, sold
wool to a Shepton Mallet clothier at Shrewsbury and later sent his servant
to deliver the wool at Bristol. In the mid-16th century, wealthy Coggeshall
clothiers regularly travelled to Shropshire every June to arrange with
middlemen for a supply of March wool.
In the 16th and 17th centuries a very large volume of wool was sold
through the weekly public markets and, in manufacturing areas and woolgrowing districts, some towns held a special market for the sale of wool and
yarn. It was estimated that 4050 horse packs of wool a week were sold at
Halifax in 1577 (Bowden, 1971).
London was the centre of commerce as well as being the epicentre of
plague mortality in England. Henry VII systematically assisted the wool
merchants in their trade with the Netherlands (Taylor & Morris, 1939) and,
during the 1540s, the volume of cloth exports through London doubled;
vast fortunes were made by the Merchant Adventurers whose convoys of
little ships sailed across to Antwerp twice yearly in May and November
(Sheppard, 1998). It is clear that several major epidemics in London came
via this route (for example in 1603; see Shrewsbury, 1970).
The possible eects of trade on the spread of the disease was tested by
multivariate analysis of deaths from the pestilence in London in 15781649
versus commodity prices, wheat, wool, straw, hides, dairy products and
hay. Only high wool prices (P : 0.009) and low prices for dairy products
376
Conclusions
377
378
Conclusions
13.12 Symptoms
379
wavelength of 2225 years (P 0.01). France was the epicentre for the
plague in Europe and this medium-wavelength oscillation correlated with
cycles in the Holy Roman Empire and Italy. We suggest that when the
epidemics were widespread in France, at the peaks of the oscillation, there
was a greater chance of infectives travelling to other metapopulations and
starting new outbreaks. This had a knock-on eect: the mediumwavelength cycle of plague deaths in England was then driven by the cycle
in Germany.
But what was the cause of the original, highly statistically signicant,
oscillation in France that acted as a driver for the spread of plague through
other parts of the supermetapopulation? It does not appear to correlate
well with external factors and was probably not exogenous. We suggest the
following explanation for this endogenous oscillation. France, throughout
the age of plagues, was composed of about 400 discrete populations (i.e.
medium- and large-sized towns) that were clearly separated from one
another and suered regularly from epidemics. Thus plague circulated
round the metapopulation and was carried by travelling infectives along
the major and minor trade routes (Fig. 11.22). Populations at the crossroads on the major routes (by road and river) were the most commonly
aected and probably, with repeated infections over the years, these communities may have built up an immunity (section 13.10). The period of the
oscillation, 2225 years, may represent a generation eect. Major epidemics were probably density dependent (section 13.6) and it may have
taken this length of time to replace the individuals in each community with
a new generation of susceptibles, so imposing an endogenous oscillation on
the plague dynamics as it circulated round the metapopulation of France,
the wavelength of which was determined by the approximate generation
time.
In summary, this may be an example of an endogenous oscillation in
France that drove (and interacted with) a corresponding exogenous oscillation in the Holy Roman Empire and Italy (for a discussion of population
oscillations, see section 13.17).
13.12 Symptoms
People in England were well able to recognise the plague, certainly by the
16th century, even far out in the Provinces; the symptoms must have been
readily recognisable, although in Italy in the mid-15th century the physicians went to considerable lengths to ensure a correct diagnosis and there
may have been more than one lethal infectious disease extant there. Over a
380
Conclusions
381
382
Conclusions
383
infection and death. Since the 40-day quarantine period was accepted
throughout the age of plagues, it is suggestive evidence that the same
causative agent was responsible for all of this time (see section 13.2).
The cordons sanitaires established at Eyam and Penrith and other
places were probably eective in containing the epidemic within the community, although the more wealthy citizens probably ed at the rst signs
of an outbreak. However, the extensive and ecient cordons sanitaires set
up around Marseilles in 172022 were completely ineective in preventing
the spread of the epidemic circumstantial evidence in favour of this being
an outbreak of bubonic plague because the movement of rats would not be
constrained by these measures (see Chapter 12).
(i) Winter conditions, even in southern France, may have constrained the
plague and eliminated its endemic status.
(ii) It has been suggested (Appleby, 1980; Carmichael, 1997) that the
disappearance of the plague corresponded with a period of low sunspot activity, globally cooler temperatures, the advance of glaciers and
the early stages of the Little Ice Age (Lamb, 1977). Pster (1980) has
analysed the most signicant eects during the Little Ice Age: the
month of March reects the climatic change most persistently and
distinctly. Cooling began in 1560 and March remained cold and
wintry together with a prevailing drought, suggesting that the month
was frequently dominated by northerly winds and blocking anticyclones. The dominant impression of the three spring months is one of
coldness and drought, with the coldest decades being the 1640s and
384
Conclusions
1690s. The second feature that has been identied by Pster is that the
month of June became wet and cool.
There is an exception to the general rule of an increasingly cold
climate during the Little Ice Age: the 1630s were clearly warmer in
marked contrast with the following decade, the 1640s, which was the
coldest period during the age of plagues (Pster, 1980). These ndings
may be compared with the severity of the plague in continental Europe
where it was at its most rampant, with widespread epidemics, in
France (see Fig. 11.3) and Germany (see Fig. 11.26) in the 1630s, which
contrast sharply with the 1640s when plague grumbled on at a lower
level in continental Europe.
It is possible, therefore, that factors contributing to the elimination
of the plague in Europe were not only the colder winters, but also the
cold weather conditions in March persisting through to June that may
have reduced the ability of the plague to re-establish itself after the
winter.
(iii) Introduction of fresh infectives into southern France, probably from
the Levant, ceased.
(iv) The causative agent mutated to a more benign form, as the haemolytic
streptococcus of scarlet fever is believed to have changed in England in
the early 20th century (Duncan et al., 1996b).
(v) An improvement in the general level of nutrition, particularly in
southern France, may have contributed to increased resistance and
hence to reduced infectivity, comparable with the changes in whooping cough lethality in England in the 19th century which were clearly
linked to improved nutrition (Scott & Duncan, 1998). Corn prices in
France were dropping sharply after 1662 (see Fig. 11.7; Baulant, 1968),
which would have improved the level of nutrition for the bulk of the
population. In contrast, 162829 were years of widespread famine in
northern Italy with unusually high prices of grain; this period immediately preceded the greatest outbreak of plague in continental Europe
(Cipolla, 1981).
The plague was not unique in suddenly appearing and disappearing.
Many zoonoses have appeared during the 20th century and have then lain
dormant; the English Sweating Sickness disappeared after the fth epidemic.
13.15 What was the causative agent of haemorrhagic plague?
It is unlikely that we shall ever be able to answer this question with
certainty, but the foregoing suggests that the disease may have been some
385
form of viral haemorrhagic fever. Sensationalised accounts of dying patients have fuelled intensive public fascination with loviruses and highlighted the global threat of emerging and re-emerging infectious diseases.
Filoviruses are the prototypical emerging pathogens: they cause a haemorrhagic disease of high case fatality associated with explosive outbreaks
caused by person-to-person transmission, have no known treatment, occur
unpredictably, and have an unknown reservoir (Khan et al., 1998). The
term viral haemorrhagic fever is a clinical and imprecise denition for
several dierent diseases (Le-Guenno, 1997) and has been described as any
of a diverse group of virus diseases that are characterised by sudden onset,
fever, aching, bleeding from internal organs, petechiae (spots resulting from
the eusion of blood under the skin) and shock. A number of distinct
haemorrhagic fever viruses are known, based on a shared ability to induce
haemorrhage by poorly understood mechanisms that typically involve the
formation of blood clots, known as disseminated intravascular coagulation
(Ramanathan & Taylor, 1997).
Filoviruses are the main causative agents of haemorrhagic fever in
humans (Feldmann et al., 1996b) but these infections can also be caused by
arenaviruses, aviviruses and bunyaviruses. Filoviruses are enveloped,
non-segmented negative-stranded RNA viruses (Palese et al., 1996) and,
once inside an infected cell, the virus transcribes its single strand of RNA
and replicates by the synthesis of an antisense positive RNA strand that
serves as a template for additional viral genome. As the infection progresses, the cytoplasm of the cell develops prominent inclusion bodies that
contain the viral nucleocapsid; the virus then assembles and buds o the
host cell, obtaining its lipoprotein coat from the plasma membrane of the
cell.
Filoviruses cause a number of serious diseases in humans today, including Ebola and Marburg. Ebola infection is arguably the most gruesome
way to die; it begins with a sudden fever and kills by the liquefaction of the
internal organs. There are four known variants of the Ebola lovirus
(Zaire, Sudan, Ivory Coast and Reston strains) together with Marbug, a
genetically related monkey virus. Since loviruses reproduce as RNA they
do not have the ability to check each copy (unlike the retrovirus HIV where
the RNA is converted into DNA inside the host cell before the virus
reproduces) so that the error rate is one million-fold greater than that of
DNA-based systems (Sanchez et al., 1996; Takada et al., 1997). Another
incredible aspect of the biology of the lovirus is the speed with which it
replicates in its host (Folks, 1998).
Marburg and Ebola viruses are believed to be evolving at similar rates
that are 100 times slower than those of retroviruses and human inuenza
386
Conclusions
virus. The divergence time between Marburg and Ebola has been estimated
to be more than several hundred years ago and most of the nucleotide
substitutions were transitions and synonymous for Marburg, suggesting
that purifying selection has operated during its evolution (Suzuki &
Gojobori, 1997). However, the genetic sequences of two isolates of EbolaZaire from epidemics 600 miles apart and separated by 18 years proved to
be virtually identical and it appears that something is restraining the
natural tendency of loviruses towards genetic divergence. Sanchez et al.
(1996) suggested that natural selection pressures have favoured the survival
of the original strain over any mutants. Filoviruses may also work by the
suppression of the immune system and Ebola victims usually die without
evidence of an eective immune response (Folks, 1998); Ebola-Zaire virus
was found to suppress the basal expression of the major histocompatability
complex class I family of proteins and inhibited the induction of multiple
genes by alpha interferon (IFN-alpha) and IFN-gamma. The events that
lead to the blockage of IFN signalling are believed to be critical for Ebola
virus-induced immunosuppression and to play a role in the pathogenesis of
the infection (Harcourt et al., 1999).
It is not yet known in which animals Ebola lies dormant during the long
stretches between human outbreaks, although hundreds of African species
have been screened in search of a reservoir. The high degree of similarity
between the 1976 and 1995 Ebola-Zaire strains suggests that the reservoir
is the same in both locations and that the animal is either widespread in
Zaire or else is a migratory species. Bats were initially regarded as likely
candidates because they could distribute Ebola over a wide area of Africa,
but this theory is now discounted. It has been noted that the geographical
distribution of dierent types of the Ebola virus coincided with the distributions of dierent small mammals that live at the forest margins. Ebola
RNA has been found in three mouse and one shrew species (MacKenzie,
1999).
Transmission of Ebola is usually by some form of direct contact with the
carrier or host species through the skin and its secretions, but an airborneroute is considered a possibility. The symptoms appear after an incubation
period of 221 days, nally resulting in fulminant shock and death (Peters
et al., 1996; Feldmann & Klenk, 1996). Schnittler & Feldmann (1998)
proposed the following hypothetical model of Ebola haemorrhagic fever.
The lovirus enters through minute lesions of the skin and mucosae and
obtains direct access to the vascular system or indirect access via the
lymphatic system (Fig. 13.2), after which virus particles are transported to
the lymph nodes where macrophages, known to be the primary sites for
387
Fig. 13.2. Hypothetical model of Ebola and Marburg haemorrhagic fever. After
Schnittler & Feldmann (1998).
388
Conclusions
virus replication (Geisbert et al., 1992; Feldmann et al., 1996a; Ryabchikova et al., 1996; Zaki & Goldsmith, 1998), are bathed by lymphatic
uid. After primary infection, replication continues in secondary and tertiary lymph nodes, resulting in the release of particles into the venous
system. The subsequent step of infection seems to be mediated by macrophages in the liver sinuses and spleen where they are in close contact with
circulating blood so that replication and activation may be initiated quickly in these organs, since macrophages can be infected without penetration
of cellular or tissue barriers (Fig. 13.2). Macrophage-derived mediators act
on multiple organs, but mainly on the endothelium, which responds by an
increase in permeability, dysregulation of vascular tone, expression of
cell-adhesion molecules and the development of a procoagulatory phenotype that together substantially contribute to the occurrence of shock.
Patients who survived showed several features that distinguished them
from non-survivors. No dierence in the viral antigen load, measured in the
serum, was detected in survivors and non-survivors, indicating that host
defence rather than inoculum size determines survival (Nabel, 1999). Most
victims of Ebola infection apparently cannot produce signicant numbers
of protective antibodies against the only protein on the surface of the virus
(called membrane-anchored glycoprotein), so that production of a vaccine
has proved dicult. However, mice have recently been injected with membrane-anchored glycoprotein and these animals produced immune cells
that expressed antibodies. These were fused with mouse cancer cells which
then produced a steady supply of antibodies that could be collected to
make vaccines. Mice immunised with these antibodies 24 hours before an
injection of Ebola managed to ght o infection.
Some of the anecdotal accounts of treatments for haemorrhagic plague
may be of relevance and interest. Lancing of the buboes (presumably when
they rst appeared) was believed by some physicians to aid survival and,
presumably, to prevent the disease progressing to the development of the
tokens. A surgeon serving in the garrison at Dunster in Somerset bled all
sick soldiers at the rst signs of the disease in 1645, taking blood until they
were like to drop down and all his patients recovered (section 9.5.2). This
bleeding would have been carried out some 30 days after the point of
infection. Was it possible that lacing swollen lymph nodes or the removal
of a major part of the infected white blood cells allowed the immune system
to overcome the infection?
Crimea-Congo haemorrhagic fever re-emerged recently in the United
Arab Emirates, with 11 cases admitted to hospital between June 1994 and
January 1995. Symptoms on admission were high fever, vomiting, diar-
389
rhoea and haemorrhagic signs. Eight patients died at a mean of 6.8 days
after admission (Schwarz et al., 1997; Hassanein et al., 1997).
We do not suggest that haemorrhagic plague in Europe during
13481670 was because of Ebola or any of the present-day viral haemorrhagic fevers that have been identied, but the close similarities with their
symptoms and pathology described above suggest that a lovirus may
have been the causative agent.
13.16 Co-existence of two plagues
Two distinct plagues existed together for hundreds of years: haemorrhagic
plague in Europe from 1347 to 1670 and bubonic plague predominantly in
central and eastern Asia but which also broken out sporadically and
erratically along the Mediterranean coastline during the age of plagues.
Traces of Yersinia pestis DNA have recently been reported in the dental
pulp of skeletons of people who died in the area around Provence in the
14th and 16th centuries as well as, of course, at Marseilles in 1720. The
frequent and irregular minor pests of Italy, and some of the epidemics at
Barcelona and along the Spanish Mediterranean coast may also have been
serious outbreaks of bubonic plague.
Bubonic plague spread in the 20th century to many subtropical regions
but the epidemics never became permanently established in Europe. We
know why this is. The climatic conditions are unsuitable over much of the
continent and there are no resident resistant rodent species.
Why, on the other hand, did haemorrhagic plague apparently not appear in central Asia or sub-Saharan Africa? We suggest that a number of
factors may have contributed to this. Firstly, the appropriate trade routes
were much longer than in Europe and infectives may not have survived the
journey. Secondly, trading was less intensive when compared with the
internal routes in Europe, so that fewer potential infectives would be
passing to and from Asia. Thirdly, the climate may have been too hot and
dry for ready droplet transmission.
13.17 Population recovery after the mortality crisis of a plague epidemic
Historians have long been puzzled by the speed with which populations
and metapopulations recovered after a major plague epidemic, particularly
the Black Death (section 4.7). As we have seen, the outbreaks were density
dependent and the major epidemics in the 16th and 17th centuries were
conned to the towns, where the mortality was frequently between 30%
and 50%.
390
Conclusions
391
Fig. 13.3. Annual baptisms at Penrith, Cumbria, England, 15571812. Three cycles
of the long-wavelength oscillation during the steady-state period and the marked
rising trend after 1750 can be seen. Abbreviations: m.c., mortality crisis resulting
from the combined eects of the famine in 1596 and the plague of 159798; b, start
of the population boom and the end of the steady-state period.
392
Conclusions
Fig. 13.4. Baptism series at Penrith, 15571757 (see Fig. 13.3) ltered to reveal the
long-wavelength oscillation. Filter window : 20 to 100 years.
the surrounding parishes that had escaped the plague so that a surge in
births and deaths followed. Once the population had reached a critical size
at the limits of the density-dependent constraints, there was a compensatory rise in emigration (Duncan et al., 1994c) and increased mortality, so
that the endogenous oscillations were maintained. These persisted at Penrith until 1750 when the economic and farming conditions changed and a
population boom began (see Fig. 13.3). Thus the demographic eects of the
mortality crisis were detectable for the following 150 years.
Similar long-wavelength oscillations in synchronous baptisms and
burials can be detected by time-series analysis after the plague of 1604 in
York and after the plague of 166566 in London (Scott & Duncan, 1998).
Major epidemics in the 16th and 17th centuries in England (and in
continental Europe) were conned to the towns and cities, whereas the
surrounding villages and smaller towns largely escaped. Immigration from
the countryside, particularly from the younger people, speedily lled the
gaps left by the mortality thereby assisting in the demographic recovery.
London took in a very large number of servants and apprentices from as
13.18 Postscript
393
Fig. 13.5. Cross-correlation between the burial and baptism series at Penrith,
16251750. Filter window : 40 to 50 years.
far aeld as York (Galley, 1998) and these would have added to the pool of
susceptibles, probably being neither resistant nor immune. Table 8.1 shows
that many of them died in the plague epidemics, leaving more gaps to be
lled by the next wave of susceptible young immigrants. We have shown in
this section, therefore, how the epidemiology of the plague interacted with
the demography, dynamics and social behaviour of the population.
13.18 Postscript
We have used two main approaches in our study of the biology of haemorrhagic plague. Firstly, we have analysed parish records in detail and have
traced the spread of the infection through and between families in identied
populations, simulating as far as is possible 400 years after the event, the
work of a modern epidemiologist. The characteristics of the disease can be
elucidated in this way. Secondly, we have examined the dierent spread of
the disease through dierent metapopulations. These two approaches
together present haemorrhagic plague in a new light.
394
Conclusions
13.18 Postscript
395
References
Alkhatib, G., Combadiere, C., Broder, C.C., Feng, Y., Kennedy, P.E., Murphy,
P.M., et al. (1996). CCCKR5: a RANTES, MIP-1, MIP-1 receptor as a
fusion cofactor for macrophage-tropic HIV-1. Science, 272, 19551958.
Anderson, R.M. & May, R.M. (1982a). Coevolution of hosts and parasites.
Parasitology, 85, 411426.
Anderson, R.M. & May, R.M. (1982b). Directly transmitted infectious diseases:
control by vaccination. Science, 215, 10531060.
Anderson, R.M. & May, R.M. (1985). Vaccination and herd immunity to
infectious diseases. Nature, 318, 323329.
Anderson, R.M. & May, R.M. (1991). Infectious Diseases of Humans. Oxford:
Oxford University Press.
Annotation (1924). Reservoir of plague in South Africa. British Medical Journal,
1, 875.
Appleby, A.B. (1973). Disease or famine? Mortality in Cumberland and
Westmorland 15801640. Economic History Review, 26, 403432.
Appleby, A.B. (1975). Agrarian capitalism or seigneurial reaction? The northwest
of England, 15001700. American Historical Review 80, 574594.
Appleby, A.B. (1980). Epidemics and famine in the Little Ice Age. Journal of
Interdisciplinary History, 10, 643663.
Axon, W.E.A. (1894). Chronological notes on the visitations of plague in
Lancashire and Cheshire. Transactions of the Lancashire and Cheshire
Antiquarian Society, 12, 5299.
Bacot, A.W. & Martin, C.J. (1914). Observations on the mechanism of the
transmission of plague by eas. Journal of Hygiene, 13, 423439.
Bailey, N.T.J. (1975). The Mathematical Theory of Infectious Diseases and its
Applications. London: Grin.
Barclay, A.J.G., Foster, A. & Sommer, A. (1987). Vitamin supplements and
mortality related to measles: a randomised clinical trial. British Medical
Journal, 294, 294296.
Barnes, H. (1891). Visitations of the plague in Cumberland and Westmorland.
Transactions of the Cumberland and Westmorland Antiquarian and
Archaeological Society, 11, 158186.
Bartlett, M.S. (1957). Measles periodicity and community size. Journal of the
Royal Statistical Society, 120, 4870.
Bartlett, M.S. (1960). The critical community size for measles in the U.S. Journal
of the Royal Statistical Society, 123, 3844.
396
References
397
398
References
Hourdsprint.
Bradshaw, F. (1907). Social and economic history. III. The Black Death. In The
Victoria History of the County of Durham, vol. II, ed. W. Page, p. 210.
London: Constable.
Bridges, C.B., Katz, J.M., Seto, W.H., Chan, P.K.S., Tsang, D., Ho, W., et al.
(2000). Risk of inuenza A (H5N1) infection among health care workers
exposed to patients with inuenza A (H5N1), Hong Kong. Journal of
Infectious Diseases, 181, 344348.
Campbell-Kease, J. (1989). A Companion to Local History Research. London: A.
& C. Black Ltd.
Carmichael, A. (1986). Plague and the Poor in Renaissance Florence. Cambridge:
Cambridge University Press.
Carmichael, A.G. (1991). Contagion theory and contagion practice in
fteenth-century Milan. Renaissance Quarterly, 44, 213256.
Carmichael, A.G. (1997). Bubonic plague: the Black Death. In Plague, Pox and
Pestilence, ed. K.F. Kiple, pp. 6067. London: Weidenfeld & Nicholson.
Carrington, M., Kissner, T., Gerrard, B., Ivanov, S., OBrien, S.J. & Dean, M.
(1997). Novel alleles of the chemokine-receptor gene CCR5. American Journal
of Human Genetics, 61, 12611267.
Chambers, J.D. (1972). Population, Economy and Society in Pre-industrial England.
Oxford: Oxford University Press.
Chandler, A.C. & Read, C.P. (1961). Introduction to Parasitology. 10th Edition.
New York: John Wiley and Sons.
Choe, H., Farzan, M., Sun, Y., Sullivan, N., Rollins, B., Ponath, P.D., et al. (1996).
The -chemokine receptors CCR3 and CCR5 facilitate infection by primary
HIV-1 isolates. Cell, 85, 11351148.
Christie, A.B. (1969). Infectious Diseases: Epidemiology and Clinical Practice. 3rd
Edition. Edinburgh: Churchill Livingstone.
Cipolla, C.M. (1981). Fighting the Plague in Seventeenth-Century Italy. Wisconsin:
University of Wisconsin Press.
Clapham, J. (1949). A Concise Economic History of Britain. Cambridge:
Cambridge University Press.
Clark, P., Gaskin, K. & Wilson, A. (1989). Population Estimates of English Small
Towns 15501851. Centre for Urban History, University of Leicester,
Working Paper No. 3.
Cli, A. (1995). Incorporating spatial components into models of epidemic spread.
In Epidemic Models: Their Structure and Relation to Data, ed. D. Mollison,
pp. 119149. Cambridge: Cambridge University Press.
Cli, A.D. & Haggett, P. (1988). Atlas of Disease Distributions: Analytic
Approaches to Epidemiological Data. Oxford: Blackwell Reference.
Cliord, J. (1989). Eyam Plague 16651666. Sheeld: The Print Centre.
Cliord, J.G. & Cliord, F. (1993). Eyam Parish Register 16301700, vol. XXI.
Chestereld: Derbyshire Record Society.
Coale, A.J. & Demeny, P. (1966). Regional Model Life Tables and Stable
Populations. Princeton, NJ: Princeton University Press.
Comford, M.E. (1907). Religious houses. In The Victoria History of the County of
Durham, vol. II, ed. W. Page, p. 99. London: Constable.
Cornelis, G.R. & Wold-Wulz, H. (1997). The Yersinia Yop virulon: a bacterial
system for subverting eurkaryotic cells. Molecular Microbiology, 23, 861867.
Cox, J.C. (1910). The Parish Registers of England. London: Methuen.
Creighton, C. (1894). History of Epidemics in Britain. Cambridge: Cambridge
References
399
University Press.
Davis, D.E. (1986). The scarcity of rats and the Black Death: an ecological history.
Journal of Interdisciplinary History, XVI, 455470.
Dean, M., Carrington, M., Winkler, C., Huttley, G.A., Smith, M.W., Allikmets, R.,
et al. (1996). Genetic restriction of HIV-1 infection and progression to AIDS
by a deletion allele of the CKR5 structural gene. Science, 273, 18561862.
Deaux, G. (1969). The Black Death 1347. New York: Weybright and Talley.
Defoe, D. (1722). Diary of a Plague Year. London: Everymans Library.
Deng, H., Liu, R., Ellmeier, W., Choe, S., Unutmaz, D., Burkhart, M., et al. (1996).
Identication of a major co-receptor for primary isolates of HIV-1. Nature,
381, 661666.
Dietz, K. (1975). In Epidemiology, ed. D. Ludwig & K.L. Cooke, pp. 104121.
Philadelphia: Society for Industrial and Applied Mathematics.
Doranz, B.J., Rucker, J., Yi, Y., Smyth, R.J., Samson, M., Peiper, S.C., et al. (1996).
A dual-tropic primary HIV-1 isolate that uses fusin and the -chemokine
receptors CKR-5, CKR-3, and CKR-2b as fusion cofactors. Cell, 85,
11491158.
Dragic, T., Litwin, V., Allaway, G.P., Martin, S.R., Huang, Y., Nagashima, K.A.,
et al. (1996). HIV-1 entry into CD4; cells is mediated by the chemokine
receptor CC-CKR5. Nature, 381, 667673.
Drancourt, M., Aboudharam, G., Signoli, M., Dutour, O. & Raoult, D. (1998).
Detection of 400-year-old Yersinia pestis DNA in human dental pulp: an
approach to the diagnosis of ancient septicemia. Proceedings of the National
Academy of Science, 95, 12 63712 640.
Drummond, J.C. & Wilbraham, A. (1991). The Englishmans Food. London:
Pimlico.
Duncan, S.R., Scott, S. & Duncan, C.J. (1992). Time series analysis of oscillations
in a model population: the eects of plague, pestilence and famine. Journal of
Theoretical Biology, 158, 293311.
Duncan, S.R., Scott, S. & Duncan, C.J. (1993a). An hypothesis for the periodicity
of smallpox epidemics as revealed by time series analysis. Journal of
Theoretical Biology, 160, 231248.
Duncan, S.R., Scott, S. & Duncan, C.J. (1993b). The dynamics of smallpox
epidemics in Britian 15501800. Demography, 30, 405423.
Duncan, S.R., Scott, S. & Duncan, C.J. (1994a). Modelling the dierent smallpox
epidemics in England. Philosophical Transactions of the Royal Society B, 346,
407419.
Duncan, S.R., Scott, S. & Duncan C.J. (1994b). Smallpox epidemics in cities in the
17th and 18th centuries. Journal of Interdisciplinary History, 25, 255271.
Duncan, S.R., Scott, S. & Duncan, C.J. (1994c). Determination of a feedback
vector that generates a non-decaying oscillation in a model population.
Journal of Theoretical Biology, 167, 6771.
Duncan, C.J., Duncan, S.R. & Scott, S. (1996a). Whooping cough epidemics in
London 17011812: infection dynamics, seasonal forcing and malnutritional
eects. Proceedings of the Royal Society B, 263, 445450.
Duncan, C.J., Duncan, S.R. & Scott, S. (1996b). The dynamics of scarlet fever
epidemics in England and Wales in the 19th Century. Epidemiology &
Infection, 117, 493499.
Duncan, C.J., Duncan, S.R. & Scott, S. (1997). The dynamics of measles
epidemics. Theoretical Population Biology, 52, 155163.
Duncan, C.J., Duncan, S.R. & Scott, S. (1998). The eects of population density
400
References
References
401
402
References
Khan, A.S., Sanchez, A. & Pieger, A.K. (1998). Filoviral haemorrhagic fevers.
British Medical Bulletin, 54, 675692.
King, E. (1988). Medieval England. London: Guild Publishing.
Kiple, K.F. & Ornelas, K.C. (1997). Sweating sickness: an English mystery. In
Plague, Pox and Pestilence, ed. K.F. Kiple, pp. 156159. London: Weidenfeld
& Nicolson.
Kirchho, F., Pohlmann, S., Hamacher, M., Means, R.E., Kraus, T., Uberla, K., et
al. (1997). Simian immunodeciency virus variants with dierential T-cell and
macrophage trophism use CCR5 and an unidentied cofactor expressed in
CEMx174 cells for ecient entry. Journal of Virology, 71, 65096516.
Kitch, M.J. (1986). Capital and kingdom: migration to later Stuart London. In
London 15001700. The Making to the Metropolis, ed. A.L. Beier & R. Finlay,
pp. 224251. London: Longman.
Kohn, G.C. (1995). The Wordsworth Encyclopedia of Plague and Pestilence. New
York: Facts on File.
Ladurie, E.L.R. & Baulant, M. (1980). Grape harvests from the fteenth through
the nineteenth centuries. Journal of Interdisciplinary History, 10, 839849.
Lamb, H.H. (1977). Climate History and the Future. Princeton, NJ: Princeton
University Press.
Landers, J. (1993). Death and the Metropolis. Cambridge: Cambridge University
Press.
Langer, W.L. (1964). The Black Death. Scientic American, 210, 114121.
Langmuir, A.D., Worthen, T.D., Solomon, J., Ray, C.G. & Petersen, E. (1985).
The Thucydides syndrome: a new hypothesis for the cause of the plague of
Athens. New England Journal of Medicine, 313, 10271030.
Last, J.M. (ed.) (1995). A Dictionary of Epidemiology, 3rd Edition. New York:
Oxford University Press.
Law, A. (1908). Social and economic history. In The Victoria History of the
County of Lancaster, vol. II, ed. W. Farrer & J. Brownbill, p. 285. London:
Constable.
Leasor, J. (1962). The Plague and The Fire. London: George Allen & Unwin Ltd.
Le-Guenno, B. (1997). Haemorrhagic fevers and ecological perturbations.
Archives of Virology Supplement, 13, 191199.
Lenski, R.E. (1988). Evolution of plague virulence. Nature, 334, 473474.
LeRiche, W.H. & Milner, J. (1971). Epidemiology as Medical Ecology. London:
Churchill Livingstone.
Levin, S. & Pimentel, D. (1981). Selection of intermediate rates of increase in
parasite host systems. American Naturalist, 117, 308315.
Libert, F., Cochaux, P., Beckman, G., Samson, M., Aksenova, M., Cao, A., et al.
(1998). The delta CCR5 mutation conferring protection against HIV-1 in
Caucasian populations has a single and recent origin in Northeastern
Europe. Human Molecular Genetics, 7, 399406.
Lilienfeld, D.E. & Stolley, P.D. (1994). Foundations of Epidemiology, 3rd Edition.
Oxford: Oxford University Press.
Lindley, P. (1996). The Black Death and English art. A debate and some
assumptions. In The Black Death in England, pp. 125146. Stamford: Paul
Watkins.
Liston, W.G. (1924). The Milroy Lectures. The Plague. British Medical Journal, 1,
900997.
Littman, R.J. & Littman, M.L. (1969). The Athenian plague: smallpox.
Transactions of the American Philological Association, 100, 261275.
References
403
Liu, R., Paxton, W.A., Choe, S., Ceradini, D., Martin, S.R., Horuk, R., et al. (1996).
Homozygous defect in HIV-1 coreceptor accounts for resistance of some
multiply-exposed individuals to HIV1 infection. Cell, 86, 367377.
Livi-Bacci (1997). A Concise History of World Population. Cambridge: Cambridge
University Press.
London, W.P. & Yorke, J.A. (1973). Recurrent outbreaks of measles, chickenpox
and mumps. I. Seasonal variation in contact rates. American Journal of
Epidemiology, 98, 453468.
Longrigg, J. (1980). The great plague of Athens. History of Science, 18, 209225.
Lunn, J. (1937). The Black Death in the Bishops Registers. Ph.D. thesis,
Cambridge University.
MacKenzie, D. (1999). Bats o the hook. New Scientist, 164, 16.
Madani, N., Kozak, S.L., Kavanaugh, M.P. & Kabat, D. (1998). gp120 envelope
glycoproteins of human immunodeciency viruses. Proceedings of the
National Academy of Sciences, USA, 95, 80058010.
Maia, J. de O.C. (1952). Some mathematical developments on the epidemic theory
formulated by Reed and Frost. Human Biology, 24, 167200.
Mann, J.M. (1995). Preface. In The Coming Plague. Newly Emerging Diseases in a
World Out of Balance, pp. xixii. London: Virago Press.
Martinson, J.J., Chapman, N.H., Rees, D.C., Lui, Y.-T. & Clegg, J.B. (1997).
Global distribution of the CCR5 gene 32-base pair deletion. Nature Genetics,
16, 100102.
Matheson, C. (1939). A survey of the status of Rattus rattus and its subspecies in
the seaports of Great Britain and Ireland. Journal of Animal Ecology, 8,
7693.
May, R.M. & Anderson, R.M. (1983). Parasitehost co-evolution. In Coevolution,
ed. D.J. Futuyma & M. Slatkin, pp. 186206. Sunderland, MA: Sinauer
Associates.
McNeill , W.H. (1977). Plagues and Peoples. Oxford: Blackwell.
Michael, N.L., Chang, G., Louie, L.G., Mascola, J.R., Dondero, D., Birx, D.L., et
al. (1997). The role of viral phenotype and CCR-5 gene defects in HIV-1
transmission and disease progression. Nature Medicine, 3, 338340.
Miller, E. (1961). Mediaeval York. The Victoria History of the City of York, vol.
III, ed. P.M. Tillott, pp. 8485. London: Constable.
Mills, S.D., Boland, A., Sory, M.P., Van De Smissen, P., Kerbourch, C., Finlay,
B.B., et al. (1997). Yersinia enterocolitica induces apoptosis in macrophages by
a process requiring functional type III secretion and translocation
mechanisms and involving YopP, presumably acting as an eector protein.
Proceedings of the National Academy of Sciences, USA, 94, 12 63812 643.
Morens, D.M. & Littman, R.J. (1972). Epidemiology of the plague of Athens.
Transactions of the American Philological Association, 122, 271304.
Morens, D.M. & Littman, R.J. (1994). The Thucydides syndrome reconsidered:
new thoughts on the plague of Athens. American Journal of Epidemiology,
140, 621627.
Morris, C. (1977). Plagues in Britain. In The Plague Reconsidered, pp. 3747.
Staord: Hourdsprint.
Morris, J.N. (1957). Use of Epidemiology, 1st Edition. London: E. & S.
Livingstone Ltd.
Mummidi, S., Ahuja, S.S., Gonzalez, E., Anderson, S.A., Santiago, E.N., Stephen,
K.T., et al. (1998). Genealogy of the CCR5 locus and chemokine system gene
variants associated with altered rates of HIV-1 disease progression. Nature
404
References
Medicine, 4, 786793.
Nabel, G.J. (1999). Surviving Ebola virus infection. Nature Medicine, 5, 373374.
Nicholson, J. & Burn, R. (1777). The History and Antiquities of Westmorland and
Cumberland. London: Strachan and Cadell.
Nohl, J. (1926). The Black Death. A Chronicle of the Plague. London: George
Allen & Unwin Ltd.
OBrien, S.J. & Dean, M. (1997). In search of AIDS-resistance genes. Scientic
American, 277, 4451.
OBrien, T., Winkler, C., Dean, M., Nelson, J.A.E., Carrington, M., Michael, N.L.,
et al. (1997). HIV-1 infection in a man homozygous for CCR5-
32. Lancet,
349, 1219.
Olshansky, S.J., Carnes, B., Rogers, R.G. & Smith, L. (1997). Infectious diseases.
New and ancient threats to world health. Population Bulletin, 52, 254.
Olsen, L.F. & Schaer, W.M. (1990). Chaos versus noisy periodicity: alternative
hypotheses for childhood epidemics. Science, 249, 499504.
Olson, P.E., Hames, C.S., Benenson, A.S. & Genovese, E.N. (1996). The
Thucydides syndrome: Ebola deja` vu (or Ebola reemergent?). Emerging
Infectious Diseases, 2, 155156.
Ormrod, M. (1996). The politics of pestilence. Government in England after the
Black Death. In The Black Death in England, pp. 147181. Stamford: Paul
Watkins.
Ormrod, W.M. & Lindley, P.G. (1996). The Black Death in England. Stamford:
Paul Watkins.
Oxford, J.S. (2000). Viral clues in frozen brain cells. The Times Higher Education
Supplement, 21 January 2000, 28.
Page, D.L. (1953). Thucydides description of the great plague. Classical
Quarterly, 47, 97119.
Palese, P., Zheng, H., Engelhardt, O.G., Pleschka, S. & Garcia Sastre, A. (1996).
Negative-strand RNA viruses: genetic engineering and applications.
Proceedings of the National Academy of Sciences, USA, 93, 11 35411 358.
Pepys, S. (1665/1906). Diary of Samuel Pepys. London: J.M. Dent & Sons Ltd.
Peters, C.J., Sanchez, A., Rollin, P.E., Ksiazek, T.G. & Murphy, F.A. (1996).
Filoviridae: Marburg and Ebola viruses. In Fields Virology, 3rd Edition, ed.
B.N. Fields, D.M. Knipe & P.M. Howley, pp. 11611176. Philadelphia:
Raven Press.
Pster, C. (1980). The Little Ice Age: thermal and wetness indices for Central
Europe. Journal of Interdisciplinary History, 10, 665696.
Plague Research Commission (1907a). The epidemiological observations made by
the Commission in Bombay City. Journal of Hygiene, 7, 724798.
Plague Research Commission (1907b). Observations made in four villages in the
neighbourhood of Bombay. Journal of Hygiene, 7, 799873.
Pollitzer, R. (1954). Plague. Geneva: World Health Organization Monography
Series No. 22.
Poole, J.C.F. & Holladay, A.J. (1979). Thucydides and the plague of Athens.
Classical Quarterly, 29, 282300.
Porteous, C. (1962). The Well-Dressing Guide. Hanley: Wood, Mitchell & Co. Ltd.
Portnoy, D.A., Wolf-Watz, H., Bolin, I., Beeder, A.B. & Falkow, S. (1984).
Characterization of common virulence plasmids in Yersinia species and their
role in the expression of outer membrane proteins. Infection and Immunity, 43,
108114.
Race, P. (1995). Some further consideration of the plague in Eyam, 1665/6. Local
References
405
406
References
References
407
408
References
University Press.
Thompson, A.H. (1914). The pestilences of the fourteenth century in the Diocese
of York. Archaeological Journal, LXXI, 97154.
Tidd, C.W., Olsen, L.F. & Schaer, W.M. (1993). The case for chaos in childhood
epidemics. II. Predicting historical epidemics from mathematical models.
Proceedings of the Royal Society B, 254, 257273.
Trevelyan, G.M. (1945). English Social History. London: Longmans.
Trout, A.P. (1973). The municipality of Parish confronts the plague of 1668.
Medical History, 17, 418423.
Twigg, G. (1978). The role of rodents in plague transmission: a worldwide review.
Mammal Review, 8, 77110.
Twigg, G. (1984). The Black Death: A Biological Reappraisal. London: Batsford
Academic.
Twigg, G. (1989). The Black Death in England: an epidemiological dilemma.
Maladie et Societe, sie`cles XIIXVII, 7598.
Twigg, G. (1993). Plague in London: spatial and temporal aspects of mortality. In
Epidemic Disease in London, ed. J.A.I. Champion, pp. 117. London:
University of London Centre for Metropolitan History.
Underwood, E.A. (1965). Charles Creighton, the man and his work. In History of
Epidemics in Britain, vol. 1. From A.D. 664 to the Great Plague, ed. C.
Creighton, pp. 43135. London: Frank Cass & Co. Ltd.
Vernadsky, G. (1953). The Mongols and Russia. New Haven, CT: Yale University
Press.
Walker, J. (1860). The History of Penrith from the Earlier Period to the Present
Time. Penrith: Hodgson Printers.
Wang, W.K., Dudek, T., Zhao, Y.J., Brumblay, H.G., Essex, M. & Lee, T.H.
(1998). CCR5 coreceptor utilization involves a highly conserved arginine
residue of HIV type 1 gp120. Proceedings of the National Academy of Sciences,
USA, 95, 57405745.
Watson, J. (1992). Lakeland Towns. Milnthorpe, Cumbria: Cicerone Press.
White, A. (ed). (1993). A History of Lancaster 11931993. Keele: Ryburn
Publishing Keele University Press.
Willan, T.S. (1983). Plague in perspective: the case of Manchester in 1605. In
Transactions of the Historic Society of Lancashire and Cheshire, ed. J.I.
Kermode & C.B. Phillips, vol. 132, pp. 2940. Gloucester: Alan Sutton
Publishing Ltd.
Wilson, E.M. (1975). Richard Leakes plague sermons, 1599. Transactions of the
Cumberland and Westmorland Antiquarian and Archaeological Society, 75,
150173.
Wood, W. (1865). The History and Antiquities of Eyam, 4th Edition. London: Bell
and Daldy.
Wrigley, E.A. (ed.) (1966). Introduction to English Historical Demography. London:
Weidenfeld and Nicholson.
Wrigley, E.A. & Schoeld, R.S. (1981). Population History of England and Wales,
15411871. A Reconstruction. London: Edward Arnold.
Wu, L.-T. (1926). A Treatise on Pneumonic Plague. Geneva: League of Nations
Health Organization.
Wu, L.-T., Chun, J.W.H., Pollitzer, R. & Wu, C.Y. (1936). Plague: A Manual for
Medical and Public Health Workers. Shanghai: Mercury Press.
Wylie, J.A.H. & Collier, L.H. (1981). The English Sweating Sickness (Sudor
Anglicus): A reappraisal. Journal of the History of Medicine and Allied
References
409
Index
410
Index
Berwick-on-Tweed, 93, 176, 177
Billingham, 95
Birmingham, 238
Bishops Castle, 170
black blisters/spots, 6, 152, 217, 218, 306
Black Death, 2, 190
arrival in Europe, 867, 370
Baltic, 87
Bath and Wells, 107
brought by Genoese galleys, 85, 104
CCR5
32, 353, 355
Channel Islands, 81, 89, 98, 108
confusion with smallpox, 102
Constantinople, 82
Crimea, 104
England, 8898
Florence, 84, 87
Genoa, 104
Greenland, 6, 81, 98, 108, 109
Iceland, 6, 81, 98, 108, 109
immunity, 104, 377
incubation period, 85, 108
infectivity, 355
Ireland, 6, 108
Italy, 87
metapopulation pandemic, 867
mortality, 87, 90, 91, 91, 106, 359, 376
North Africa, 86
Norway, 6, 87, 109
origins, 82
pattern of spread, 13, 45, 308
person-to-person contact, 85
rate of spread, 87, 88
Reed and Frost dynamics, 85, 90, 107
resistance, 355, 3778
Scotland, 6
seasonality, 89, 105, 106, 355
serial generation time, 85, 107
Sicily, 6, 826, 104
Spain, 86
Strasburg, 299
Sweden, 6, 87
symptoms, 835
black rat (Rattus rattus), 557, 76, 262, 357
Bombay, 734, 73
bubonic plague and, 567
distribution, 567, 108
habitat, 56, 90
origin, 56
plague of Justinian, 49, 284
population dynamics, 57, 73
spread, 49
blains, 21719, 380
Bohemia, 32831, 329, 333
untouched by Black Death, 103
Bologna, 312, 314, 377
Bombay
411
412
bubonic plague (cont.)
course of disease, 678
density dependence, 778, 79
dynamics in humans, 356, 359
endemic, 47
enzootic plague, 55
epidemiology, 747
epizootic plague, 55
Ethiopia, 49
eas, 7, 8
incubation period, 67, 108
India, 7, 8, 45, 478, 56, 727, 75, 76, 78
indirectly transmitted disease, 26
Levant, 13, 303
Libya, 16, 338
Marseilles, 49, 60, 284, 33850
mortality, 478, 501, 66, 75, 79, 112,
35960
not cause of haemorrhagic plague,
35662, 370, 377
pathology, 701
population dynamics, 613, 734, 356
possibly at Athens, 5
possibly in plague of Justinian, 6
rate of spread, 45
rats dying, 360
Reed and Frost modelling, 79, 80, 356
rodents, 7, 8, 360
seasonality, 14
Silk Road, 48, 389
sylvatic reservoirs, 545, 55
symptoms, 68
USA, 50, 51, 55, 62
virulence, 634
wave-like spread, 46
see also Yersinia pestis
Buckingham, 166
Buckinghamshire, 91,165, 258
Burgundy, 87, 298
Bury St Edmonds, 168, 247
Calais, 337, 374
Black Death, 87, 89
epidemics, 167, 177
Calcutta, 50, 77
Cambridge, epidemics, 111, 152, 166, 167,
168, 170, 238, 239, 243, 258
Cambridgeshire, 91, 168, 247
Canterbury, 16, 167, 168, 170, 243, 256
carbuncles, 6, 380
Eyam, 282
Florence, 110
London, 21720, 221
Marseilles, 340
Carcassone, Black Death, 87
Carlisle, 175, 181, 185, 373
Black Death, 93, 101
Index
mortality, 118, 185
pest-houses, 186
plague epidemics, 101, 111, 181, 182, 184,
1856, 187
public health measures, 118, 1856
Cartagena, 322
case-fatality ratio, 27
Catalonia, 322
CCR5 receptor, 3524, 359
CCR5
32 deletion, 3524, 359, 377
centroids, 45, 2957, 300, 301, 302, 331,
denition, 296
Channel Islands, 89, 98, 108, 168
Chelmsford, 139
Cheshire, 91, 22934, 2412, 248, 250, 251,
252
population density, 44, 98, 360
Chester, 238, 368, 371, 373
behaviour of plague victims, 219, 251
epidemics, 167, 168, 2289, 229, 239, 248,
250, 251, 252
plague houses, 251
Chester-le-Street, 95
Chestereld, 169, 363
China
bubonic plague, 7, 8, 13, 478, 49
inuenza, 1917, 10, 11
pneumonic plague, 69
chicken pox, 21, 22
Reed and Frost model, 30, 33
cholera, 2, 6, 8, 12, 23, 110, 113
climate, 61, 3834, 389
England, 59, 60
Great Plague, 211
spread of bubonic plague, 7980, 389
closed population dynamics, 434, 43
Colchester, 198, 239, 256, 257, 257, 355, 363,
374
Cologne, 316, 330, 331
Colyton, 139, 167, 249, 363
Congleton, 232, 238
Constantinople, 5, 82, 316
co-primary infectives, denition, 24
Cordoba, 318
cordon sanitaire, 261, 280, 317, 341, 344,
346, 366, 383
Cornwall, 91, 166, 167, 168, 240, 251
Corsica, 86, 312, 313
co-secondary infectives, denition, 24
County Durham, 95,176
plague, 183, 243, 248, 250
population density, 98, 100, 360
Coventry, 92, 150
Crimea, 82, 85, 104
Crimea-Congo haemorrhagic fever, 388
cross-correlation function (ccf), 39
Crowland Abbey, 97
Index
Croydon, 153, 197, 202, 213215
Cumberland, 111, 175, 178, 181
plague, 111, 115, 181, 184
population density, 98, 360
Dacre, 188
Darlington, 95, 181, 182, 182, 183, 187, 189
Dartmouth, 163, 169
decaying epidemics, 38, 369
density dependence, 3912
Black Death, 98
bubonic plague, 778, 79, 103
haemorrhagic plague, 12, 155, 173, 189,
190, 359, 366, 368, 372, 379
plague at Marseilles, 349
smallpox, 3678, 376
Deptford, 213
Derby, 92, 168, 169, 170, 250, 262, 265
Derbyshire, 91, 167, 239, 265, 378
Devonshire, 91, 99, 166, 167, 239, 240, 249
diphtheria, 6, 37
explanation of mortality crises, 106, 109
Doncaster, 94, 166
Dorsetshire, 239, 240, 249
Dover, 167, 168, 198, 256
driven epidemics, 38
Dumfries, 182, 187, 189
Dunkirk, 319
Dunster, 249, 388
Durham, 175, 176, 181, 182, 355, 373
Black Death, 96
epidemics, 17981, 180, 1823, 187, 190,
228, 248, 250, 363
mortality, 1823
dysentery, 109, 111, 113, 152
East Anglia, 168, 205, 239, 243, 247, 250,
257, 258
ports, 225, 238, 256, 258, 372, 374
river system, 250, 335, 337, 372, 373
Ebola, 1, 9, 10, 3859, 38
mortality, 5, 9
possibly at Athens, 5
Sudan, 5, 9
symptoms, 5, 385
Zaire, 5, 9, 3856
Edenhall, 187, 187
Edinburgh, 181
Egypt, 3, 82
bubonic plague, 50
origin of plague of Justinian, 5
Ely, 258
diocese, 89, 106, 353
encephalitis, 9, 11
endemic plague, 36770
in France, 27, 285303
in London, 1545, 369
413
enzootic plague
denition, 55
India, 77
population dynamics, 623
rodents, 18, 54
epidemiology
bubonic plague, 746
concepts, 2138
denition, 12
London plagues, 192
pneumonic plague, 6870
epizootic plague
Bombay, 73
denition, 55
India, 77
population dynamics, 63
Essex, 165, 168, 198, 239, 250, 2567, 258
Ethiopia, 3
bubonic plague pandemic, 49
origins of plague of Justinian, 5
Evesham, 167, 168
Exeter, 166, 168, 169, 170, 239, 374
diocese, 89, 106
Eyam, 256, 26183, 266, 266, 270, 273,
370
comparison to Penrith, 265, 268, 269, 272
contrast with Marseilles, 346
household contact rate, 269, 272, 279
immunity, 145, 281, 282, 377
mortality, 264, 27980, 378
pattern of outbreak, 126, 128, 131, 135,
277, 363
pest-houses, 280, 282
plague stone, 280
population, 27980
quarantine measures, 118, 2801, 283, 383
recovery, 146
resistance, 279, 282
seasonality, 279
sex-specic mortality, 139
symptoms, 262, 282
wills, 2756, 282
famine, 14, 114, 291, 384
in northern England, 14, 176, 1812
Faversham, 207
Filoviridae, 9, 3859
Flanders, 296, 331, 333
ea, 5760
biology, 589
blocked, 58, 65, 113
characteristics, 57
climate, 60, 77
ecology, 57, 589
epizootic, 60
human, 361
India, 757
414
Index
ea (cont.)
life cycle, 59, 105
rat, 7, 15
relative humidity, 59, 77
reproduction, 5960, 105
survival, 589
temperature, 5960, 60, 76, 153, 357
Xenopsylla cheopis, 59, 64, 90
Florence, 110, 305, 3068, 307, 308, 312,
314, 315, 377, 382
Black Death, 84, 87
France, 8, 318, 323, 331, 332, 333, 335, 336,
337, 374
Black Death, 86, 87
black rat, 56
endemic centre of plague, 14, 333, 369,
379, 383
famines, 290, 291
health bureaux, 291
plague epidemics and malnutrition, 384
plague pseudoendemic, 27, 285303, 369
plagues, 286303, 287, 288, 289, 334, 378
quarantine, 2923
regions, 2935, 294, 295, 296, 297, 299,
300
seasonality, 291, 383
wheat prices, 289, 291, 292, 293, 384
Gascony, 337
Gateshead, 182, 243, 248
Geneva, 313, 329, 330, 331, 335
Genoa, 303, 304, 312, 313, 314, 316, 317,
359, 382
Black Death, 85, 86, 104
Germany, 32831, 329, 330, 333, 334, 336,
374, 379, 384
Glasgow, bubonic plague, 51
Gloucester, 169, 170
diocese, 89, 106, 354
Gloucestershire, 91, 168, 239, 249
grape harvest, 288, 290
Great Budworth, 251
Greenland, 6, 57, 98, 108, 109, 357, 389
Greenwich, 153, 163, 165, 192
haemorrhagic plague, 14
causative agent, 3849
classication of epidemics, 148, 3623
denition, 7
density dependent, 173, 189, 359, 366, 368,
372, 379
incubation period, 24, 12635, 363
infectious period, 23, 24, 12635, 363
latent period, 24, 12635, 363
not bubonic plague, 35662, 370, 377
quarantine, 3601, 3813
reasons for disappearance, 3834
Index
India, 77
bubonic plague, 7, 8, 18, 45, 4750, 727,
73, 75, 76, 78
density dependence, 778, 103
sylvatic reservoir, 54
ea survival, 59
origin of black rat, 56
Plague Commission, 15, 727, 111, 358
infectious period, 22, 128, 134, 155, 355
Black Death, 108
denition, 22, 23, 24, 25, 28
haemorrhagic plague, 363
infectiousness, 25, 145, 355
inuenza, 4, 6, 11, 21, 23, 41, 110, 166, 385
infectious period, 46
pandemic of 191719, 2, 8, 10, 11, 394
rate of spread, 46
Reed and Frost model, 30
Spitzbergen, 11
Innsbruck, 329
interepidemic period (T ), 37, 38, 401
Ipswich, 238, 257, 258
bubonic plague, 51
pneumonic plague, 51
Ireland, 4, 108, 151, 228, 251, 252, 368, 371,
377
Italy, 8, 290, 30318, 333, 370, 376, 377, 379
autopsies, 306
bubonic plague, 303, 318
famine, 384
major and minor pests, 14, 303, 318
mortality, 103, 314, 315
plague epidemics, 14, 304, 310, 311, 312
public health measures, 14, 290, 303, 305,
314, 31718, 332, 394
quarantine, 305, 314, 317, 318, 382
seasonality, 315
signs and symptoms, 3056, 379, 380
Justinian, plague of, 56, 49, 284, 389
Kemsing, 207, 209
Kendal, 118, 178, 181, 182, 182, 184, 187
Kent, 153, 166, 198, 205, 207, 225, 239, 240,
243, 256
ports, 207, 256, 335
Keswick, 184
Kings Lynn, 247, 250, 258, 316, 372, 373,
374, 374
Kingston, 153, 167
Lambesc, 284
Lancashire, 111, 114, 175, 176, 178, 243, 245,
250, 360
Lancaster, 93, 114
Lanchester, 95
Languedoc, 291, 342, 346
415
416
Index
Index
sex-specic mortality, 139
Stratford-upon-Avon, 1701, 171
Switzerland, 329
York, 227
Nantes, 298, 337
Nantwich, 2312
Naples, 303, 304, 309, 312, 313, 314, 315, 381
Neston, 2525, 254, 363
Netherlands, 328, 331, 332, 374, 375
Newcastle-upon-Tyne, 114, 187, 226, 227,
372, 373
Black Death, 96
epidemics, 1769, 1812, 2267, 239, 243,
247, 248
port, 182, 226, 247, 256, 335
N mes, 291, 298, 343, 349
Norfolk, 95, 111, 167, 169, 240, 243, 256,
257, 360
Northampton, 168, 238, 251
Northamptonshire, 91, 167, 238, 360
Northumberland, 98, 175, 176, 177, 181,
183, 243, 360
Nottingham, 94, 166, 170, 222
Norway, 6, 87, 109, 357, 376, 389
Norwich, 146, 238, 373, 374
diocese, 89, 106, 354
epidemics, 168, 239, 240, 243, 247, 257,
258
Nuremberg, 87, 316, 329, 330
open population dynamics, 434, 44
Oporto, 319
oscillation, 1, 38, 38992
Oswestry, 122, 167
Ottery St Mary, 249, 250, 355, 363
Oundle, 167, 258
Oxford, 114, 166, 198, 205, 238, 249, 251
Black Death, 90
Sweating Sickness, 149, 150
Oxfordshire, 91, 249
Palermo, 309, 377
Padua, 359
Paris, 87, 290, 298, 336, 339, 382
Pembrokeshire, 251, 371
Penrith, 37, 97, 105, 11548, 124, 129, 178,
181, 184, 187, 378
age-specic mortality, 1378, 138
baptisms, 146, 391, 392, 393
compared to Eyam, 265, 268, 269, 272
contact rate, 11926
endogenous oscillations, 1, 38992
epidemiological characteristics, 12635,
144, 363
household contact rate, 124, 126, 127,
131, 134, 137
417
418
Index
Index
weather conditions and, 38, 41, 365
Somerset, 91, 99, 166, 169, 240, 249, 388
Southampton, 88, 163, 168, 204, 239, 256
Spain, 10, 317, 320, 321, 323, 332, 376, 377
Black Death, 87
bubonic plague, 8, 322
metapopulation, 332
plague epidemics, 14, 31824
see also Iberian peninsula
Spanish inuenza, see inuenza
spatial spread, 12, 13, 45
Black Death, 45
bubonic plague, 45
epidemics, 446, 198, 361, 3716
pattern, 213
saltatory, 361, 372
trade routes, 316
spectral analysis, 25960, 2856, 287, 294,
305, 328, 333, 334, 390
denition, 39
St Bartholomew Fair, 197, 374
Staordshire, 91, 231, 234, 238
steady state, of endemic disease, 38
Stockport, 237
Strasburg, 298, 299, 335
Stratford-upon-Avon, 139, 167, 168, 1703,
171, 172
Sturbridge Fair, 197, 258, 374
Suolk, 257, 360
Sunderland, 95, 243
supermetapopulation, denition, 134, 333
Surrey, 91, 163, 198, 239, 256
Sussex, 198, 205, 239
Sweating Sickness, 14951, 384
etiology, 1502
Reed and Frost dynamics, 151
Sweden, 6, 87
Switzerland, 32831, 329, 330, 333, 334
grape harvest, 288
sylvatic phase, 54
symptoms, 24, 835, 37981
black blisters or spots, 6, 152, 217220,
306
blains, 21719, 380
buboes, 117, 190, 197, 202, 21720, 282,
306, 307, 340, 380
bubonic plague, 678
carbuncles, 6, 110, 21720, 221, 282, 340,
380
Florence, 110
London, 165, 193, 197, 202, 204, 21720
plague at Athens, 3
plague of 1598, 18990
plague of Justinian, 6, 389
pneumonic plague, 67
Sweating Sickness, 150
tokens, 21821, 242, 243, 256, 262, 282,
419
306, 380
Syria, 338, 340
Black Death, 82
bubonic plague, 48, 54
tertiary infective, denition, 24, 25
Tewksbury, 169, 170, 238
Thames, 192, 198, 207, 213, 215, 238, 239,
256
Thucydides syndrome, 3, 4
time-series analysis, 39, 25960, 335, 336,
3789
cross-correlation function (ccf), 39, 293,
336, 393
ltering, 39
epidemics, 223
smallpox epidemics, 3941
spectral analysis, 39, 25960, 2856, 287,
294, 305, 328, 333, 334, 390
Tiverton, 169, 249
tokens, 221, 226, 243, 256, 262, 362
description, 165, 21821, 242, 282, 306,
380
London, 21718, 221
Totnes, 122, 169
Toulon, 346
identication of Yersinia pestis at, 49
Toulouse, 87, 298
toxic shock syndrome, 4
transmission, 21, 26
probability, 19, 23, 25, 43
transmission coecient ()
as a driver, 38
denition, 31
density of susceptibles, 37, 41
smallpox, 41
Troyes, 298, 302, 316
Tuscany, 14, 303, 305, 313, 359, 378
type (i) epidemics, 170, 231, 243, 245, 248,
249, 250, 251
atypical, 159, 232, 236, 240, 252
denition, 148, 190, 3623
London, 159, 162, 193
type (ii) epidemics, 170, 184, 237, 243, 248,
258, 263, 268, 327
atypical, 179, 240, 250, 252
denition, 148, 190, 363
London, 153, 162
typhoid, Salmonella typhi, 11, 113
confusion with bubonic plague, 68
typhus, 2, 4, 5, 152, 154, 355
explanation of mortality crises, 17, 106,
109, 380
confusion with bubonic plague, 68, 113,
380
United States of America, 80
420
United States of America (cont.)
inuenza of 191719, 10
bubonic plague, 50, 51, 55
ea survival, 59
prairie dogs and Yersinia, 54, 62
Valencia, 322
Venice, 312, 313, 317, 329, 359, 374, 3812
Black Death, 86, 359, 376
Verona, 315
Vienna, 87, 330
virulence, 27, 3768
haemorrhagic plague, 2201
Yersinia, 52, 634
vitamins, 105, 365
Wales, 99, 101
Warcop, 187, 188
Ware, 372
West Kirby, plague, 230, 231
West Nile Virus, 11
Westmorland, 111, 115, 175, 178, 181, 184
population density, 98, 360
wheat prices, 14, 176, 259, 260, 375
as a driver, 41
France, 289, 291, 292, 293, 384
smallpox epidemics, 41, 105
Whitchurch, 231, 252, 253
whooping cough, epidemics, 37, 109, 365
Wiltshire, 198, 205, 238, 239, 240, 249, 256
Winchester, 16, 102, 240, 249, 256
diocese, 89, 102, 106
Windsor, 153, 192
Woodchurch, 234
wool prices, 14, 176
wool trade, 375
Worcester, 16, 169, 249
diocese, 89, 106
Index
Worcestershire, 91, 239, 249, 360
Wragley, 177
Yarmouth, 239, 240, 246, 247, 256, 258, 373
yellow plague (Pestis ava), 377
Yersinia pestis, 72, 221, 2613, 317, 322, 340,
350, 389
antibodies, 71
antigenic components, 52
biology, 72, 79, 79
causative agent in plagues, 1089, 35662
characteristics, 51, 53, 359
cultures, 71
genome, 49, 52
molecular biology, 11, 634, 353
Mongul movements, 48
mutant strains, 65
R , 26
sylvatic reservoirs, 545
virulence, 52, 635
victims at Marseilles, 49, 284, 328, 340
York, 178, 368, 372, 373, 374, 392, 393
archdeaconries, 94
Black Death, 91, 94, 99
diocese, 106
epidemics, 110, 111, 167, 176, 177, 181,
190, 2268, 245, 249
pest-houses, 227, 245
port, 335, 381
quarantine, 226, 245
Yorkshire, 175, 176, 177, 240, 245, 248, 249,
250, 256, 360
Black Death, 100
conditions, 99, 176
epidemics, 183, 243
zoonotic diseases, 5, 9, 12, 21, 384
rodent enzootics, 18, 54