DEMOGRAPHY IN THE CACTACEAE 173 The Botanical Review 69(2): 173–203 Demographic Trends in the Cactaceae HÉCTOR GODÍNEZ–ÁLVAREZ Unidad de Biología, Tecnología y Prototipos Facultad de Estudios Superiores Iztacala Universidad Nacional Autónoma de México Avenida de los Barrios 1, Los Reyes Iztacala Tlalnepantla 54090, Ap. Postal 314, Edo. de México, Mexico TERESA VALVERDE Departamento de Ecología y Recursos Naturales Facultad de Ciencias Universidad Nacional Autónoma de México Ciudad Universitaria Mexico City, D.F. 04510, Mexico AND PABLO ORTEGA–BAES Laboratorio de Investigaciones Botánicas Facultad de Ciencias Naturales Universidad Nacional de Salta Buenos Aires 177, Salta 4400, Argentina I. II. III. IV. V. VI. VII. VIII. IX. X. XI. Abstract/Resumen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Distribution and Abundance Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Seed Germination and Seedling Establishment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Growth and Reproduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Size Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Population Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Elasticity Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Matrix Models and Matrix Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix 1: Demographic Information, Endangerment, and Geographical Distribution of Cactus Species with Different Life-Forms . . . . . . . . . . . . . . . . Copies of this issue [69(2)] may be purchased from the NYBG Press, The New York Botanical Garden, Bronx, NY 10458–5128, U.S.A.; nybgpress@nybg.org. Please inquire as to prices. Issued 16 December 2003 © 2003 The New York Botanical Garden 173 174 175 177 178 183 187 189 191 194 194 195 195 202 174 THE BOTANICAL REVIEW I. Abstract Although our biological knowledge regarding cactus species is thorough in many areas, only in recent years have ecologists addressed their demographic behavior. Here we attempt a first review of the present knowledge on cactus demography, including an analysis of the published information on species with different growth forms and life-history traits. Our review shows that cactus distribution ranges are determined by environmental heterogeneity and by species-specific physiological requirements. Temperature extremes may pose latitudinal and altitudinal distribution limits. At a more local scale, soil properties dramatically affect cactus distribution. Most cacti show a clumped spatial distribution pattern, which may be the reflection of a patchy resource distribution within their heterogeneous environments. The association of cacti with nurse plants is another factor that may account for this aggregated distribution. Many cacti grow in association with these perennial nurse plants, particularly during early lifecycle phases. The shade provided by nurse plants results in reduced evapotranspiration and buffered temperatures, which enhance cactus germination and establishment. In some cases a certain degree of specificity has been detected between particular cactus species and certain nurse plants. Yet some globose cacti may establish in the absence of nurse plants. In these cases, rocks and other soil irregularities may facilitate germination and establishment. Cacti are slow-growing species. Several abiotic factors, such as water and nutrient availability, may affect their growth rate. Competition and positive associations (i.e., mycorrhizae and nurse–cacti association) may also affect growth rate. Age at first reproduction varies greatly in relation to plant longevity. In general, cactus reproductive capacity increases with plant size. Populations are often composed of an uneven number of individuals distributed in the different size categories. This type of population structure reflects massive but infrequent recruitment events, apparently associated with benign periods of abundant rainfall. A few cactus species have been analyzed through the use of population-projection matrices. A total of 17 matrices were compiled and compared. Most of them reflect populations that are close to the numerical equilibrium (␭ = close to unity). Elasticity analyses revealed that the persistence of individuals in their current size category (“stasis”) is the demographic process that contributes the most to population growth rate. Also, adult categories (rather than juveniles or seedlings) show the largest contributions to ␭. No differences were apparent regarding this matter between cacti with different life-forms. This review shows that our knowledge of cactus population ecology is still incipient and rather unevenly distributed: some topics are well developed; for others the available information is still very limited. Our ability to preserve the great number of cactus species that are now endangered depends on our capacity to deepen our ecological understanding of their population processes. Resumen A pesar de que nuestro conocimiento biológico sobre las cactáceas es basto en muchas áreas, ha sido sólo hasta fechas recientes que los ecólogos han abordado el estudio de su comportamiento demográfico. En este artículo presentamos una primera revisión del conocimiento actual sobre demografía de cactáceas, incluyendo un análisis de la información publicada sobre especies con diferentes formas de crecimiento y características de historia de vida. Nuestra revisión muestra que el área de distribución de las cactáceas se ve afectada por la heterogeneidad del ambiente y por los requerimientos fisiológicos de cada especie. Los valores extremos de temperatura fijan límites latitudinales y altitudinales de distribución a muchas cactáceas. A una escala más local, las propiedades del suelo juegan un papel fundamental. La mayoría de los cactus muestran una distribución espacial agregada, lo cual es un reflejo de la distribución DEMOGRAPHY IN THE CACTACEAE 175 aparchonada de los recursos en los ambientes altamente heterogéneos que habitan. La asociación de los cactus con plantas nodriza es otro de los factores que explica su distribución agregada. Muchos cactus crecen en asociación con estas plantas nodriza, particularmente durante los primeros estadios. La sombra de estas plantas perennes reduce la evapotranspiración y amortigua la temperatura, lo cual incrementa la germinación y el establecimiento de los cactus. En algunos casos se ha detectado especificidad entre especies particulares de cactus y ciertas plantas nodriza. Aun así, algunos cactos globosos pueden establecerse en ausencia de plantas nodrizas. En estos casos, las rocas y otras irregularidades del terreno podrían facilitar la germinación y el establecimiento. Los cactus son especies de lento crecimiento. Varios factores abióticos, como el agua y la disponibilidad de nutrientes, pueden afectar su tasa de crecimiento. La competencia y las asociaciones positivas (i.e., formación de micorrizas y asociación con plantas nodriza) también pueden afectar su tasa de crecimiento. La edad a la primera reproducción varía mucho con relación a la longevidad de las plantas. En general, la capacidad reproductiva de los cactus aumenta conforme aumenta su tamaño. Las poblaciones frecuentemente están compuestas de individuos distribuidos de manera irregular entre las diferentes clases de tamaño. Este tipo de estructura poblacional refleja eventos masivos pero poco frecuentes de reclutamiento, aparentemente asociados a períodos benignos de abundantes lluvias. Sólo unas cuantas especies de cactus se han estudiado a través de matrices de proyección poblacional. En este estudio se compilaron y se compararon 17 matrices. La mayoría de ellas reflejan poblaciones que se encuentran cerca del equilibrio numérico (␭ = cerca de la unidad). Los análisis de elasticidad revelaron que la persistencia de los individuos en su misma categoría de tamaño (“estasis”) es el proceso demográfico que mayormente contribuye a la tasa de crecimiento poblacional. También se vio que las categorías de adultos (y no las de juveniles o plántulas) fueron las que mostraron un mayor aporte a ␭. No se encontraron diferencias a este respecto entre especies de cactus con diferente forma de crecimiento. Esta revisión muestra que nuestro conocimiento sobre la ecología poblacional de los cactus es aún incipiente y está distribuido de manera dispareja: algunos temas están bien desarrollados mientras que para otros la información disponible es aún muy limitada. La posibilidad de conservar el gran número de especies de cactáceas que se encuentran amenazadas depende de nuestra capacidad para profundizar en el entendimiento ecológico de sus procesos poblacionales. II. Introduction Cacti are a very diverse and complex plant family comprising more than 2000 species, most of which are native to arid and semi-arid environments in the Americas (Bravo-Hollis & SánchezMejorada, 1978, 1991). The extraordinary adaptations of these plants to low water availability and other stressful conditions have made them the subject of a number of morphological and physiological studies, including aspects related to their photosynthetic pathways, their waterstorage ability, and the various other ways in which they optimize their water-use efficiency (among others, Bravo-Hollis & Sánchez-Mejorada, 1978, 1991; Gibson & Nobel, 1986; Nobel, 1988; Altesor et al., 1992). This information, along with a host of botanical records and taxonomic analyses, has been tremendously valuable in aiding our understanding of the biology and evolution of this fascinating plant family. Yet the lack of ecological information on cacti is still notorious in the biological literature. In recent years a number of authors have published ecological studies on cacti, covering aspect such as reproductive biology (Tinoco-Ojanguren & Molina-Freaner, 2000; Fleming et al., 2001), plant–pollinator and plant–disperser interactions (Fleming & Valiente-Banuet, 2002; Godínez-Álvarez et al., 2002), the ecophysiology of ger- 176 THE BOTANICAL REVIEW mination and early growth (Rojas-Aréchiga & Vázquez-Yanes, 2000; Ruedas et al., 2000; Flores & Briones, 2001; Rojas-Aréchiga et al., 2001), and cactus–nurse plant interactions (McAuliffe, 1984a, 1984b; Valiente-Banuet & Ezcurra, 1991; Valiente-Banuet et al., 1991a, 1991b; FloresMartínez et al., 1994, 1998). Additionally, recent articles have addressed the demography of cactus populations, including species with contrasting distribution ranges, growth forms, and life-history traits. It is this kind of ecological information that we are concerned with in the present article. We believe that our present knowledge of the demography of cactus species deserves a first attempt to search for patterns and trends that may influence the future development of this particular area of plant ecology. A detailed evaluation of the present status of our knowledge will allow us to establish priorities and encourage researchers to develop ideas about particular issues that may eventually fill the gaps on this subject. Part of the need to study the demographic trends emerging in the Cactaceae is related to the fact that there is a high number of threatened species in this plant family (Anderson & Taylor, 1994; Hunt, 1999). In fact, the whole family is included in appendix 2 of the CITES book of endangered species (Hunt, 1999). Some of the reasons that may account for this are: 1) Many cacti are highly restricted in their distribution and occupy very specific habitats, which makes them prone to extinction by habitat destruction and land-use change (Esparza-Olguín et al., 2002); 2) They seem to be particularly sensitive to disturbances due to their low individual growth rates and highly vulnerable early stages of development (Hernández & Godínez-Álvarez, 1994); 3) Illegal collection and trade are an increasing pressure on many cactus populations, especially since they have become more fashionable in recent years; and 4) Their habitats are frequently associated with poor areas in developing countries, where the pressure of increasing human populations on land-use change toward farming and cattle ranching are very strong. Thus the increasing demographic information on cactus species will certainly aid in the design of conservation and management plans and will offer a tool with which to evaluate the conservation status of poorly known species in this plant family. In addition to conservation and management interests, demographic information is the basis on which our understanding of population dynamics and life-history traits is supported (Silvertown et al., 1993). Recent demographic information on cactus species is adding significantly to our understanding of plant population dynamics in nature, since they include longlived species that inhabit dry tropical areas, two particularities that are poorly represented in the demographic literature. In this article we attempt a first review of the demographic features of cactus species, according to the literature published to date, which is summarized in Appendix 1. A decade ago Silvertown et al. (1993) published their first literature review on plant demographic trends. In that review, in which interesting trends between demography and lifeform emerged, only one cactus, Carnegiea gigantea, was included. In this article we use the approach developed by Silvertown et al. (1993) to analyze the information on cactus demography published mainly in the last decade. We use a comparative approach that focuses on cactus species with different life-forms, since a correlation between life-form and some demographic properties appears to be emerging among desert plants (Flores & Briones, 2001). We consider the following life-forms, according to the classification used by other authors (Gibson & Nobel, 1986): columnar cacti (species with long, column-like, ribbed stems measuring >2.0 m in height); barrel cacti (plants with ribbed stems, with a maximum height of 0.5–2 m); globose cacti (species with single or multiple hemispherical stems); cylindropuntias (opuntias with cylindrical stems); and platyopuntias (opuntias with flattened stems). In addition to data on population dynamics, we include valuable information on other aspects of cactus population ecology, such as distribution and abundance patterns, data on growth, reproduction, and population structure, and survival and establishment during early life-cycle phases. Our aim is to search for DEMOGRAPHY IN THE CACTACEAE 177 trends and patterns that may contribute to our understanding of the factors that determine longterm population dynamics in nature, which still remain a central issue in ecology (Horvitz & Schemske, 1995). III. Distribution and Abundance Patterns Studies on the distribution and abundance patterns of cacti have shown that these ecological attributes may vary both temporally and spatially (Bowers et al., 1995; de Viana, 1996–1997; Bowers, 1997b; Mourelle & Ezcurra, 1997; Parker, 1988a, 1993). With regard to the distribution of cacti at a regional scale, environmental heterogeneity as well as the particular physiological requirements of each species appear to determine their distribution range. For instance, the columnar cacti Carnegiea gigantea, Lophocereus schottii, and Stenocereus thurberi occupy different habitats along the topographic and edaphic gradients of the Sonoran Desert (Parker, 1988a). In addition, the northern distribution limits of these species appear to be determined by the occurrence of low temperatures, which severely damage their stem apices (Nobel, 1980). Mourelle and Ezcurra (1997) also mention that variation in environmental factors such as temperature, topography, and rainfall may be associated with the occurrence of different Argentine cacti. At a local scale, the three spatial distribution patterns described in the ecological literature (random, uniform, and clumped) have been found in cactus populations (Gulmon et al., 1979; de Viana et al., 1990; Martínez et al., 1993; Huerta & Escobar, 1998; Valverde et al., 1999; Cody, 2000). Yet clumped distributions appear to be far more common than random or uniform distributions. It has been suggested that the aggregation of individuals that has been found in cacti with different life-forms (i.e., Capiapoa cinerea, Ferocactus hystrix, Ariocarpus trigonus, Mammillaria gaumeri, M. magnimamma, M. oteroi, Turbinicarpus pseudopectinatus, Trichocereus pasacana, and Neobuxbaumia macrocephala) may be related to the patchy distribution of conditions that enhance seedling establishment and plant growth (de Viana et al., 1990; Valverde et al., 1999; Martínez et al., 2001; Esparza-Olguín et al., 2002), which is bound to exert a particularly strong effect on those species that require the shade of a nurse plant. Other potential mechanisms that may also account for a clumped distribution pattern are restricted seed dispersal (Martínez et al., 1993, 1994) and patchy distribution of seeds in the soil (Hutto et al., 1986; de Viana, 1996–1997; de Viana et al., 2001). Additionally, some studies have shown that young and adult plants within a population may have different distribution patterns. Intraspecific interactions, such as competition leading to self-thinning processes, may be responsible for this variation (de Viana et al., 1990). In regard to abundance patterns, it is difficult to find a general trend among cactus species because population densities may vary greatly, from very few individuals per hectare to thousands per hectare (Valiente-Banuet & Ezcurra, 1991; Martínez et al., 1993, 1994; Schmalzel et al., 1995; de Viana, 1996–1997; Huerta & Escobar, 1998; Mandujano et al., 1998; Valverde et al., 1999; Esparza-Olguín et al., 2002). The density of a population is the result of its birth and death rates, which may be affected by a host of biotic and biotic factors at regional and local scales. Apparently, massive but infrequent recruitment events are associated with favorable rainfall and temperature periods. In particular, infrequent events such as “El Niño” may increase the number of new recruits in some populations. This has been suggested for the columnar cacti Carnegiea gigantea and Stenocereus thurberi (Parker, 1993; Pierson & Turner, 1998), the barrel cacti Echinocactus polycephalus, Ferocactus acanthodes, and F. cylindraceus (Jordan & Nobel, 1981; Goldberg & Turner, 1986; Bowers et al., 1995; Bowers, 1997b), and the platyopuntias Opuntia acanthocarpa, O. basilaris, and O. erinacea (Bowers et al., 1995). 178 THE BOTANICAL REVIEW Contrastingly, Haman (2001) reports that El Niño events have an adverse effect on two species of Opuntia (O. echios and O. galapageia) in the Galapagos Islands, by increasing seedling, juvenile, and adult mortality due to water logging following heavy rainfall. In addition to rainfall patterns, other factors such as the length of the dry period, the availability of safe sites for establishment and growth, the presence of trees and shrubs in the habitat, soil erosion, and the effect of livestock grazing are important in determining cactus density at a particular site (Turner et al., 1966; Jordan & Nobel, 1981, 1982; Nobel, 1989; Valiente-Banuet & Ezcurra, 1991; Parker, 1993; Bowers et al., 1995; Bowers, 1997b; Pierson & Turner, 1998; Esparza-Olguín et al., 2002). Soil characteristics appear to play an important role in the distribution and abundance of cactus species in arid and semi-arid landscapes (Parker, 1991). The edaphic conditions in which some species live may be highly specialized. It has been observed that columnar cacti such as Cephalocereus columna-trajani, Escontria chiotilla, Neobuxbaumia tetetzo, Pachycereus fulviceps, P. pringlei, Stenocereus gummosus, and S. thurberi, as well as globose cacti such as Mammillaria crucigera, are associated with particular soil types (Meyrán, 1980; Valiente-Banuet et al., 1995; Ortega-Baes, 2001; Contreras & Valverde, 2002; Valiente-Banuet & GodínezÁlvarez, 2002). Certain edaphic properties, such as soil texture, may dramatically affect water availability, thus preventing the establishment of certain species while favoring others, depending on their particular water requirements and tolerances for germination and early growth. IV. Seed Germination and Seedling Establishment A great number of studies have addressed the germination behavior of cactus species under both experimental and natural conditions. Since Rojas-Aréchiga and Vázquez-Yanes (2000) recently published a thorough review of the germination ecology of cacti, in this section only some general aspects will be discussed. The main factors that affect seed germination are water, temperature, and light. However, embryo immaturity, salinity, seed age, plant hormones, inhibitory compounds in the testa, plant domestication, and other factors may also affect this process (Rojas-Aréchiga & VázquezYanes, 2000; Rojas-Aréchiga et al., 2001). Of all these factors, water availability appears to be the most important for germination of cactus seeds, since it is the most limiting factor under the conditions that prevail in deserts. Dubrovsky (1996, 1998) found that seeds of the columnar cacti Carnegiea gigantea, Pachycereus pecten-aboriginum, Stenocereus gummosus, and S. thurberi, as well as seeds of the barrel cactus Ferocactus peninsulae, germinated more rapidly and accumulated higher biomass when they were subjected to hydration–dehydration cycles. This phenomenon is known as “seed hydration memory” and may facilitate the germination process, leading to an increase in the probability that seedlings will survive (Dubrovsky, 1996). Most studies of seed germination of cactus species have been conducted under controlled conditions, and only a few have considered their germination behavior in the field. Further field research is necessary in this area, since there are important aspects of seed ecology of which the demographic implications are ignored. One of these aspects is the potential of cactus seeds to remain viable in the soil for long periods of time, forming long-term buried seed banks (Rojas-Aréchiga & Vázquez-Yanes, 2000; Rojas-Aréchiga & Batis, 2001). The only available information on this subject suggests that seeds of the barrel cactus Ferocactus wislizeni may remain viable in the soil for at least 18 months; postdispersal seed predation in this species is high, although seeds may escape from predators when hidden among rocks (Bowers, 2000). Also, Ortega-Baes (2001) found viable seeds of Escontria chiotilla of unknown age (at least 12 months old) buried in the soil, and field experiments with Mammillaria magnimamma showed DEMOGRAPHY IN THE CACTACEAE 179 that its seeds may remain viable in the soil for more than a year (Valverde et al., in press). Additionally, several Mammillaria species have been reported to retain fruits in their stems, which may play a similar ecological role as a buried seed bank (Rodríguez-Ortega & Franco, 2001). Studies of seed germination and seedling survival of cactus species under natural conditions have shown that the presence of perennial plants is necessary for the successful recruitment of new individuals (Niering et al., 1963; Turner et al., 1966; Steenbergh & Lowe, 1969, 1977; Jordan & Nobel, 1981, 1982; McAuliffe, 1984a; Valiente-Banuet & Ezcurra, 1991; Mandujano et al., 1998). Under the harsh environmental conditions that prevail in deserts, these perennial nurse plants function as islands within which physical and/or biotic conditions are beneficial for the establishment and growth of young cacti (Callaway, 1995). The nurse plants decrease the contrast between maximum and minimum soil temperatures (Shreve, 1931; Jordan & Nobel, 1981; Hutto et al., 1986; Valiente-Banuet et al., 1991a; Arriaga et al., 1993; Suzán et al., 1996), provide protection against predators (Steenbergh & Lowe, 1977; Hutto et al., 1986; Valiente-Banuet & Ezcurra, 1991), increase soil-nutrient availability under their canopies (García-Moya & McKell, 1970; Nobel, 1989), and provide a shady environment in which evapotranspiration is substantially reduced (Turner et al., 1966; Steenbergh & Lowe, 1977; Nolasco et al., 1997; Godínez-Álvarez & Valiente-Banuet, 1998). Additionally, they constitute roosting sites for dispersers, thereby attracting important amounts of seed dispersal (Steenbergh & Lowe, 1977; Yeaton, 1978; Hutto et al., 1986; Godínez-Álvarez et al., 2002). The literature on the nurse-plant phenomenon in cacti is summarized in Table I. The first point we may note from this table is that the association with perennial plants occurs in all of the life-forms that are represented in the cactus family. For instance, Suzán et al. (1996) reported that 17 cactus species from the Sonoran Desert with different life-forms grow in association with the long-lived tree Olneya tesota. Also, Arriaga et al. (1993) reported that the number of individuals of Stenocereus thurberi and Ferocactus peninsulae was significantly higher under the canopy of perennial plants than in open spaces in the tropical dry forests of northwestern Mexico. In the Tehuacán Valley of central Mexico, five cactus species with different life-forms were found to be associated with different perennial plants (Valiente-Banuet et al., 1991a). In South America, the columnar cactus Trichocereus atacamensis shows a positive association with trees and shrubs (de Viana, 1996–1997; de Viana et al., 2001). In most of these cases it has been suggested that the shade provided by nurse plant is the main factor that enhances seedling establishment by buffering extreme temperatures and reducing evapotranspiration. The identity of nurse plants so far reported is varied (Table I). Different plant species may play this role, including shrubs, trees, grasses, agaves, and even cacti (McAuliffe, 1984b; Hutto et al., 1986; Nobel, 1989; Valiente-Banuet et al., 1991a; Suzán et al., 1996; Mandujano et al., 1998; de Viana et al., 2001; Ortega-Baes, 2001). The number of cactus seedlings and juveniles established beneath the canopies of some nurse-plant species is higher than expected by chance, according to their relative abundance in their plant communities (Hutto et al., 1986; ValienteBanuet et al., 1991a, 1991b; Mandujano et al., 1998; de Viana et al., 2001; Ortega-Baes, 2001). This pattern suggests that, in some cases, the nurse–cactus association may have reached a certain degree of specificity, which may be related to particular traits (i.e., morphology, plant architecture, phenology) that determine differences in the quality of each perennial plant as nurse for different cactus species (Callaway, 1998). Certain traits, such as the presence of spines, deep roots, long-lived leaves, secondary metabolites, and nitrogen-fixation ability, among others, may favor the establishment of new individuals under the canopy of specific perennial plants (McAuliffe, 1984a; Suzán et al., 1996; Callaway, 1998). Species Columnar cacti Carnegiea gigantea* Lophocereus schottii* Neobuxbaumia macrocephala N. tetetzo* Pachycereus pringlei Stenocereus thurberi Ambrosia deltoidea Cercidium microphylum Encelia farinosa Larrea tridentata Olneya tesota Prosopis juliflora Caesalpinia melanadenia Acacia cochliacantha Fouquieria formosa Mimosa luisana M. polyantha Zizyphus pedunculata Celtis pallida Lycium andersonii L. berlandieri Olneya tesota Zizyphus obtusifolia Lippia graveolens? Caesalpinia melanadenia Castela tortuosa Eupatorium odoratum Mimosa luisana Olneya tesota Prosopis articulata Haematoxylon brasiletto Jatropha vernicosa Tecoma stans Life-cycle stage Mechanism Source S–J–A Shade Frost protection Predation avoidance Turner et al., 1966 Steenbergh & Lowe, 1969; Hutto et al., 1986 Franco & Nobel, 1989 S–J S–J–A Shade Shade Valiente-Banuet et al., 1991a Ortega-Baes, 2001 S–J Shade Frost protection Suzán et al., 1996 Parker, 1989 S–J? S–J Shade? Shade Esparza-Olguín et al., 2002 Valiente-Banuet et al., 1991a Valiente-Banuet & Ezcurra, 1991 Flores-Martínez et al., 1994 S–J Shade Increased soil nutrients Shade Valiente-Banuet et al., 1995 Carrillo-García et al., 2000 Parker, 1987 Arriaga et al., 1993 S–J–A THE BOTANICAL REVIEW Cephalocereus hoppenstedtii* Escontria chiotilla* Nurse plant 180 Table I Association between nurse plants and different cacti species (* = species with a higher number of recruits growing in association with nurse plants than expected by chance; S = seedlings; J = juveniles; A = adults; Mechanism = advantages offered by the nurse plant to the growing cacti; ? = uncertainty in the proposed aspect) Trichocereus atacamensis* Barrel cacti Echinocereus conglomeratus E. engelmannii* Echinomastus erectrocentrus F. cylindraceus F. peninsulae Hamatocactus hamatacanthus Peniocereus greggii P. striatus Globose cacti Ariocarpus trigonus Coryphantha pallida* – Castela tortuosa Eupatorium odoratum Caesalpinia melanadenia Castela tortuosa Castela tortuosa Eupatorium odoratum Shade De Viana, 1996–1997 De Viana et al., 2001 – Silvertown & Wilson, 1994 Predation avoidance Shade McAuliffe, 1984b Suzán et al., 1996 S Shade Increased soil nutrients Franco & Nobel, 1989 S–J–A S–J–A – Increased soil nutrients Shade – Nobel, 1989 Arriaga et al., 1993 Silvertown & Wilson, 1994 S–J–A S–J–A Shade Shade Suzán et al., 1996 Suzán et al., 1996 S–J–A S–J–A Shade? Shade Martínez et al., 1993 Valiente-Banuet et al., 1991a S–J–A – S–J–A Shade – Shade Valiente-Banuet et al., 1991a Rodríguez-Ortega & Ezcurra, 2001 Valiente-Banuet et al., 1991a – S–J–A S–J–A 181 Mammillaria casoi* M. carnea M. colina* Euphorbia antisyphilitica Hamatocactus hamatacanthus Opuntia leptocaulis Opuntia fulgida Celtis pallida Lycium andersonii L. berlandieri Olneya tesota Zizyphus obtusifolia Ambrosia dumosa Ephedra aspera Hilaria rigida Hilaria rigida Haematoxylon brasiletto Euphorbia antisyphilitica Opuntia leptocaulis Celtis pallida Lycium andersonii L. berlandieri Olneya tesota Zizyphus obtusifolia S–J–A DEMOGRAPHY IN THE CACTACEAE Ferocactus acanthodes Aphyllocladus spartoides Larrea divaricata Prosopis ferox 182 Table I, continued Species Globose cacti, continued M. gaumeri M. microcarpa M. oteroi M. thornberi O. echinocarpa O. leptocaulis O. ramosissima Platyopuntias O. rastrera* O. streptacantha – Opuntia fulgida Quercus castanea Celtis pallida Lycium andersonii L. berlandieri Olneya tesota Zizyphus obtusifolia Hilaria rigida Krameria sp. Hilaria rigida Krameria sp. Larrea tridentata Hilaria rigida Krameria sp. Cordia parvifolia Hilaria mutica Jatrpoha dioica Prosopis glandulosa Acacia schaffneri Life-cycle stage Mechanism Source S–J Shade? – S–J–A – Shade Leirana-Alcocer & Parra-Tabla, 1999 McAuliffe, 1984b Martínez et al., 2001 Suzán et al., 1996 – – Cody, 1993 – – Cody, 1993 S–J – Seed dispersal site? – Yeaton, 1978 Cody, 1993 S–J Shade Predation avoidance Mandujano et al., 1998 Shade Flores & Yeaton, 2000 S–J–A THE BOTANICAL REVIEW Cylindropuntias Opuntia acanthocarpa Nurse plant DEMOGRAPHY IN THE CACTACEAE 183 Despite the fact that nurse plants may offer cacti some protection during the initial life-cycle phases, mortality is tremendously high at this stage. Survivorship curves of cactus seedlings growing under the canopies of different trees and shrubs are mainly type III (sensu Pearl, 1928; see Appendix 1; Turner et al., 1966; Steenbergh & Lowe, 1969; Valiente-Banuet & Ezcurra, 1991; Mandujano et al., 1998; Esparza-Olguín et al., 2002; Valverde et al., in press). Yet the intensity of seedling mortality is lower beneath the canopies of nurse plants than in open spaces. For instance, seedlings of Mammillaria gaumeri associated with nurse plants showed a constant mortality rate through time (i.e., type II), whereas those growing in bare soil showed a clear type III curve (Leirana-Alcocer & Parra-Tabla, 1999). Although the evidence discussed so far may suggest that the nurse plant–cactus association is prevalent throughout the Cactaceae family (Table I), we must bear in mind that there are a number of examples in which new cactus seedlings have been reported to establish in the absence of nurse plants. Curiously, these species are mainly globose cacti, such as Ariocarpus fissuratus, Epithelantha bokei, Mammillaria magnimamma, M. lasiacantha, M. pectinifera, and Turbinicarpus pseudopectinatus, which may grow in open spaces (Nobel et al., 1986; Arriaga et al., 1993; Martínez et al., 1994; Valverde et al., 1999; Rodríguez-Ortega & Ezcurra, 2001). For other desert plants the presence of rocks on the ground, as well as other irregularities, appear to be of vital importance for seed germination and/or seedling survival. Rocks may reduce solar radiation and thermal loadings, as well as prolong the presence of soil moisture in their immediate vicinity (Larmuth & Harvey, 1978). Thus the importance of rocks and other surface irregularities (cavities, holes, etc.) as potential facilitators of germination and seedling survival in different cactus species is one of the areas to which further research should be directed. V. Growth and Reproduction After seed germination and seedling establishment, growth and reproduction of cacti may be affected by extrinsic factors such as biological interactions and environmental conditions, as well as by intrinsic factors such as genetic variation, allometry, and metabolic rates. Biological interactions with other plants, animals, or microorganisms may have a positive or negative impact on cacti. During early growth, positive interactions such as the association with vesicular-arbuscular mycorrhizal fungi can enhance biomass production by increasing nutrient absorption efficiency. This appears to be the case of the columnar cactus Pachycereus pectenaboriginum in a tropical dry forest in Mexico (Rincón et al., 1993). The presence of mycorrhizae has also been reported in Mammillaria magnimamma (Ruedas et al., 2000), although their effect on the growth of this species has not been analyzed. Other biotic interactions may negatively affect the growth of cacti; such is the case of intra- and interspecific competition. Paradoxically, these interactions are enhanced under the canopies of trees and shrubs where seedling establishment is most successful. McAuliffe and Janzen (1986) documented the effect of intraspecific competition for limited water in dense aggregations of young Carnegiea gigantea plants associated with the canopy of perennial shrubs; in this case, the effects of competition were decreases in water uptake, water storage, apical growth, and reproductive potential (McAuliffe & Janzen, 1986). It has also been reported that some columnar cacti (C. gigantea and Neobuxbaumia tetetzo) may establish intense interspecific competition for water with their nurse plants, producing a decline in leaf, flower, and fruit production in the nurse trees and shrubs (McAuliffe, 1984a; Flores-Martínez et al., 1994, 1998). This type of interaction has also been documented between the barrel cactus Ferocactus acanthodes and the perennial bunchgrass Hilaria rigida (Franco & Nobel, 1989), between Opuntia leptocaulis and the shrub Larrea tridentata (Yeaton, 1978), and between O. rastrera, the perennial tussock grass Hilaria mutica, and the shrub L. tridentata (Briones et al., 1998). Further research is needed to deepen 184 THE BOTANICAL REVIEW our understanding of the way in which both positive and negative interactions affect cactus growth. In addition to biotic interactions, cactus growth may be affected by environmental conditions. The literature on this subject indicates that winter precipitation and frost decrease the growth rate of Lophocereus schottii and Stenocereus thurberi (Parker, 1988b). In contrast, high summer and low winter temperatures, soil-moisture availability, and pre- and postsummer dry periods are among the main environmental factors that affect the growth rate of Carnegiea gigantea (Steenbergh & Lowe, 1977). Experimental analyses of the early growth of cacti also show that it is affected by environmental conditions. For instance, a high concentration of nutrients significantly increased seedling growth in Mammillaria magnimamma, Pachycereus hollianus, and P. pringlei (Godínez-Álvarez & Valiente-Banuet, 1998; Carrillo-García et al., 2000; Ruedas et al., 2000). Effects of direct solar radiation were contradictory, for it increased seedling growth in M. magnimamma (Ruedas et al., 2000) but decreased the growth of Neobuxbaumia tetetzo, Opuntia rastrera, P. hollianus, P. pringlei, and S. thurberi seedlings (Nolasco et al., 1997; Godínez-Álvarez & Valiente-Banuet, 1998; Mandujano et al., 1998; Carrillo-García et al., 2000). Genetic variation among individuals, as well as allometry, are also among the factors that affect growth rates. With regard to allometry, there is some variation in the annual growth rate of Carnegiea gigantea related to changes in plant form associated with ontogeny (Hastings & Alcorn, 1961; Steenbergh & Lowe, 1977). Thus, after seedling establishment, individuals grow rapidly until they reach a height of about 2 m. This fast growth is the result of an increase in the photosynthetic surface area and water-storage capacity of individuals. However, growth rate decreases when plants are between 2 and 6 m tall, since they start using some of the products of photosynthesis to generate reproductive structures. In larger plants growth occurs mainly through the proliferation of branches, while apparently the main stem reaches a maximum height (Shreve, 1910; Hastings & Alcorn, 1961; Steenbergh & Lowe, 1977). The growth pattern is different for other species of columnar cacti, such as Lophocereus schottii and Stenocereus thurberi. In these species the annual growth rate increases with increasing plant size (Parker, 1988b). Differences in growth rate among species of columnar cacti may be related to the particular branching pattern of each species. For instance, L. schottii and S. thurberi grow through the production of new stems, whereas Carnegiea gigantea produces a single stem with several branches. Since all cactus species are slow-growing, long-lived plants, there is little information about the relationship between size and age (Table II), and therefore age-related life-history traits have been difficult to investigate. The available information shows that age at first reproduction is variable, depending on the species, life-form, and plant longevity. In the long-lived columnar cacti, plants reproduce for the first time when individuals are 33 years old in Carnegiea gigantea (Steenbergh & Lowe, 1977), 70 years old in Cephalocereus columna-trajani (ZavalaHurtado & Díaz-Solís, 1995), and more than 90 years old in Neobuxbaumia macrocephala (Esparza-Olguín et al., 2002). For a shorter-lived globose cactus (Mammillaria magnimamma) and a platyopuntia (Opuntia engelmannii), first reproduction occurs when individuals are 4 and 9–11 years old, respectively (Bowers, 1996a; Valverde et al., in press). Once plants have reached reproductive size, fecundity appears to increase with increasing size (Fig. 1; Appendix 1). This pattern was observed in species with different life-forms, such as the globose cacti Coryphantha robbinsorum, Mammillaria crucigera, M. magnimamma (Schmalzel et al., 1995; Contreras & Valverde, 2002; Valverde et al., in press), the platyopuntia O. rastrera (Mandujano et al., 2001), the barrel cactus Echinomastus erectrocentrus (Johnson, 1992; Johnson et al., 1992), and the columnar cacti Carnegiea gigantea, Escontria chiotilla, DEMOGRAPHY IN THE CACTACEAE 185 Table II Age or size at which first reproduction occurs in cacti with different life-forms Age (years) Size (meters) Columnar cacti Carnegiea gigantea Cephalocereus columna-trajani Lophocereus schottii Stenocereus thurberi 33 70 – – 2.2 3.3 >2.0 2.0–2.5 4–10a Barrel cacti Echinomastus erectrocentrus – 0.024 Johnson et al., 1992 Globose cacti Coryphantha robbinsorum Mammillaria magnimamma – 4 0.013 – Schmalzel et al., 1995 Valverde et al., in press 9–11 6–13b Bowers, 1996a Species Opuntioid cacti Opuntia englemannii Source Steenbergh & Lowe, 1977 Zavala-Hurtado & Díaz-Solís, 1995 Parker, 1989 Parker, 1987 a Plant size measured as the number of branches. b Plant size measured as the number of cladodes. Lophocereus schottii, Neobuxbaumia macrocephala, N. tetetzo, and Stenocereus thurberi (Steenbergh & Lowe, 1977; Parker, 1987, 1989; Godínez-Álvarez et al., 1999; Ortega-Baes, 2001; Esparza-Olguín et al., 2002). It has been suggested that this pattern is related to an increase in the number of branches as plants age, at least in columnar cacti (Steenbergh & Lowe, 1977). Reproductive phenology varies between species with different life-forms (see Appendix 1). Species in the different genera of globose cacti, such as Mammillaria, Sclerocactus, and Turbinicarpus (Martínez et al., 1994; Lockwood, 1995; Contreras & Valverde, 2002; Valverde et al., in press), begin to reproduce in early winter, and their reproductive periods last for more than 5 months. In contrast, species in the different genera of columnar cacti may reproduce in autumn/winter (Cereus, Myrtillocactus, Pachycereus, Pilosocereus, Selenicereus, Stenocereus) or spring/summer (Carnegiea, Cephalocereus, Escontria, Hylocereus, Lophocereus, Neobuxbaumia, Pachycereus, Pilosocereus, Polaskia, Stenocereus, Subpilocereus), and their reproductive periods last for 2 to 4 months (Steenbergh & Lowe, 1977; Weiss et al., 1994; Valiente-Banuet et al., 1996, 1997a, 1997b; Nassar et al., 1997; Casas et al., 1999; Rojas-Martínez et al., 1999; Ruíz et al., 2000; Fleming et al., 2001; Esparza-Olguín et al., 2002). Other columnar cacti, however, may produce flowers and fruits throughout the year (Praecereus, Stenocereus, Weberbaurocereus, Weberocereus; Sahley, 1996; Nassar et al., 1997; Tschapka et al., 1999; Ruíz et al., 2000). Species in the genus Opuntia produce flowers and fruits in spring/summer (O. brunneogemmia, O. compressa, O. discata, O. imbricata, O. lindheimeri, O. phaecantha, O. polyacantha, O. rastrera, O. stricta, O. viridirubra; Grant et al., 1979; Spears, 1987; Osborn et al., 1988; McFarland et al., 1989; Mandujano et al., 1996; Schlindwein & Wittmann, 1997) or in winter (O. spinosissima; NegrónOrtíz, 1998). Finally, barrel cacti in the genera Echinocereus, Echinomastus, and Ferocactus reproduce mainly during spring/summer (Breckenridge & Miller, 1982; Johnson, 1992; McIntosh, 2002). Field observations regarding flower and fruit production in some cactus species (Carnegiea gigantea, Echinomastus erectrocentrus, Lophocereus schottii, Mammillaria crucigera, M. mag- 186 THE BOTANICAL REVIEW Fig. 1. Examples of the relationship between fecundity and size for cacti with different life-forms. a. The globose cactus Mammillaria crucigera. b. The columnar cactus Neobuxbaumia tetetzo. c. The platyopuntia Opuntia rastrera. DEMOGRAPHY IN THE CACTACEAE 187 nimamma, Neobuxbaumia macrocephala, N. tetetzo, Pachycereus pringlei, Stenocereus queretaroensis, S. thurberi; Steenbergh & Lowe, 1977; Johnson, 1992; Johnson et al., 1992; Pimienta et al., 1998; Godínez-Álvarez et al., 1999; Fleming et al., 2001; Contreras & Valverde, 2002; Esparza-Olguín et al., 2002; Valverde et al., in press) have shown that there is interannual variation in the timing and intensity of reproductive events. These variations may be related to particular environmental factors, such as the presence of winter rainfall, the early start of the rainy period, and/or a high total annual precipitation (Bowers, 1996b; Valverde et al., in press). The number of fruits produced per plant and the number of seeds per fruit vary according to species and life-form (Table III). For instance, fruits of columnar cacti may produce more than 1000 seeds, whereas fruits of globose cacti generally produce fewer than 100 seeds. The number of seeds per fruit in barrel cacti, cylindropuntias, and platyopuntias varies between 10 and 200. In addition to sexual reproduction, many cacti are capable of spreading vegetatively (Parker & Hamrick, 1992; Mandujano et al., 1998). This ability is common in the genus Opuntia, although plants in other genera, such as Lophocereus, Myrtillocactus, Pachycereus, and Stenocereus, may also spread through ramet production (pers. obs.). The relative importance of vegetative spread compared with sexual reproduction is related to different factors such as soil type, the presence of nurse plants, herbivory, and flooding (Mandujano et al., 1998). Opuntia engelmanni plants apparently adjust the relative proportion of new cladodes and flowers in response to the previous year’s resource expenditure (Bowers, 1996c). VI. Size Structure Population structure is the result of size- (or age-) specific birth and mortality rates (Silvertown, 1987). The particular size- (or age-) structure found in a population may be determined by several extrinsic and intrinsic factors. Among these, seed production, germination rates, soil seed banks, vegetative spread, herbivory, and weather patterns may play an important role (Silvertown, 1987). Histograms representing population structure for different cactus species often show an uneven distribution of individuals among size classes (Fig. 2; Appendix 1), which suggests that recruitment occurs in pulses, apparently associated with infrequent favorable conditions for germination and establishment (Mandujano et al., 2001; EsparzaOlguín et al., 2002). These favorable conditions appear to be related to particular precipitation and temperature combinations for species such as Carnegiea gigantea, Echinocactus horizonthalonius, Ferocactus cylindraceus, Neobuxbaumia macrocephala, and Opuntia echios (Reid et al., 1983; Bowers, 1997b; Pierson & Turner, 1998; Hicks & Mauchamp, 2000; Esparza-Olguín et al., 2002). As noted above, recruitment pulses of some species may occur in rainy years associated with particular climatic events such as El Niño (Hicks & Mauchamp, 2000). In the case of Opuntia rastrera, germination and establishment are enhanced when nurse plants are readily available and herbivory is low (Mandujano et al., 1998). Other cactus species show a type of population structure in which the number of individuals decreases monotonically with size or age (i.e., Ferocactus acanthodes, Neobuxbaumia tetetzo, Opuntia echios; Jordan & Nobel, 1981; Godínez-Álvarez et al., 1999; Hicks & Mauchamp, 2000; Fig. 2; Appendix 1). This type of population structure suggests growing populations that are regenerating constantly, in which mortality rate decreases with size or age (Martínez-Ramos & Álvarez-Buylla, 1995). Populations that spread throughout a variety of environmental conditions, such as those occupying altitudinal gradients, may show distinct population structures in different habitat types. This is the case in Opuntia echios, in which populations in a low arid habitat show a 188 THE BOTANICAL REVIEW Table III Mean number of seeds per fruit and standard errors for species of cacti with different life-forms Species Columnar cacti Carnegiea gigantea Cereus peruvianus Lophocereus schottii Neobuxbaumia macrocephala N. mezcalaensis N. tetetzo Pachycereus pringlei Pilosocereus moritzianus Stenocereus griseus S. gummosus S. stellatus S. thurberi Subpilocereus horrispinus S. repandus Weberbaurocereus weberbaueri Barrel cacti Echinomastus erectrocentrus Melocactus violaceus Globose cacti Mammillaria crucigera M. magnimamma Cylindropuntias Opuntia imbricata Platyopuntias Opuntia brunneogemmia O. engelmannii Opuntia phaeacantha O. polyacantha O. rastrera O. stricta O. viridirubra Standard error Number of fruits sampled 2263 1358 1241.8 1349.8 1946.7 1952 126 465 552 496 933 1288.1 1330.4 1329.1 2496 1121 1581 674 750 934 1969 536.9 351.3 1123 1483 1067 108 140 112.7 99.3 131.8 – 6.3 209 30 32.9 51.1 91.7 162.2 141.1 165.3 91 62.2 99.8 15 35 161.3 101.5 77.3 51.6 93 – 18 15 25 51 44 10 168 25 10 10 35 23 15 18 15–27 27 15–27 25 20 20 19 28 8 15–27 15–27 14 86.6 21.8 4.4 1.6 – 16 Johnson, 1992 Figueira et al., 1994 20 93.48 93.5 12.6 36.8 42 112 69 69 Contreras & Valverde, 2002 Valverde et al., in press 62 8.4 36 McFarland et al., 1989 55.2 143.9 171.6 5.8 21.5 16.8 20 20 20 Schlindwein & Wittmann, 1997 Bowers, 1997a 30 30 31 30 36 – – – 20 Osborn et al., 1988 Seeds per fruit 50.2 72.7 25.1 9.0 208 60.6 65.3 67.5 63.7 6.9 5.7 3.6 2.5 0.13 3.7 4.8 4.5 3.7 Source Steenbergh & Lowe, 1977 Fleming et al., 2001 Silva, 1988 Parker, 1989 Esparza-Olguín et al., 2002 Valiente-Banuet et al., 1997a Valiente-Banuet et al., 1997a Godínez-Alvarez et al., 2002 Fleming et al., 2001 Nassar et al., 1997 Silvius, 1995 Nassar et al., 1997 León de la Luz & Cadena, 1991 Casas et al., 1999 Parker, 1987 Fleming et al., 2001 Nassar et al., 1997 Nassar et al., 1997 Sahley, 1996 Osborn et al., 1988 Mandujano et al., 1996 Spears, 1987 Schlindwein & Wittmann, 1997 DEMOGRAPHY IN THE CACTACEAE 189 Fig. 2. Population size structures. a. The barrel cactus Echinocactus horizonthalonius. b. The globose cactus Mammillaria crucigera. c. The columnar cactus Neobuxbaumia macrocephala. d. The columnar cactus Neobuxbaumia tetetzo. smooth decline in the number of individuals from small to large size classes, whereas populations at higher altitudes show a bimodal structure with few plants of intermediate sizes (Hicks & Mauchamp, 2000). VII. Population Dynamics With the increasing popularity of projection matrix models, plant population ecology has grown significantly in recent decades. However, most of the populations for which projection matrices have been constructed correspond to herbaceous species from temperate habitats (e.g., Bierzychudek, 1982; Valverde & Silvertown, 1998). There are only a few references for cactus populations that include the use of matrix models. Yet the information available is quite valuable, for it includes cactus species with different growth forms for which demographic information has been gathered during several years and/or at various locations (Table IV). For each species, we consider the matrices of different years and/or sites separately, instead of constructing one average matrix per species. We believe that this approach allows us to analyze the variation in demographic behavior both within species and between species. If average matrices were built, the within-species variation would be totally obscured, and flawed conclusions could arise regarding the differences between species. From the information presented in Table IV, it may be noted that population growth rate (␭) is close to unity for most records; only in two populations is it significantly lower than unity (Carnegiea gigantea and Mammillaria crucigera 2), and only in one population is it significantly larger than unity (Mammillaria magnimamma 2). This suggest that most cactus populations studied are close to the numerical equilibrium. Yet annual projection matrices may be a limited tool for detecting long-term increases or declines in population numbers, especially in 190 Table IV Summary of population dynamics data for 17 cactus populations of eight species. Population growth rates (␭) were calculated from projection matrices from n individuals each. Elasticity values e1 to e6 correspond to matrix entries representing the seed bank (e1), sexual reproduction via seed and/or seedling production (e2), clonal spread (e3), retrogression (e4), stasis (e6), and growth (e6). Elasticity values per demographic process are reproduction (R), survival (S) and growth (G); and per life-cycle stage are seedlings (S), juveniles (J), and adults (A). Elasticity values Per demographic process Species ␭ n e1 e2 e3 e4 0.540 15 1.000 1900 0 0.009 0.979 0.994 206 206 0.006 0.006 1.011 1.04 400 400 0.020 0.025 0.956 1.333 0.967 0.945 203 203 206 206 0.003 0.207 2E–05 0.006 0.015 0.019 7E–06 0.125 0.977 0.896 230 230 0.001 0.002 0.002 0.006 1.093 1.006 1.042 63 39 63 0.059 0.018 0.058 1.04 3000 0.997 250 0.001 1E–04 0.014 0.001 0.052 0.021 0 0.05 0.100 0.050 e5 e6 R S G S J A Source 0.99 0.914 0 0.078 0 0.000 0.99 0.914 0 0.078 0.13 0.080 0 0.387 0.865 0.534 1 2 0.837 0.885 0.105 0.088 0.006 0.006 0.889 0.906 0.105 0.088 0.033 0.093 0.182 0.163 0.785 0.744 3 0.791 0.763 0.189 0.212 0.020 0.025 0.791 0.763 0.189 0.212 0.040 0.045 0.760 0.755 0.2 0.2 4 0.934 0.044 0.334 0.436 0.999 7E–04 0.713 0.156 0.003 0.207 2E–05 0.006 0.948 0.353 0.999 0.838 0.044 0.436 0.001 0.156 0.003 0.010 0.208 0.323 1E–04 2E–04 0.006 0.047 0.987 0.469 0.999 0.947 5 0.984 0.971 0.013 0.021 0.001 0.002 0.986 0.977 0.013 0.021 0.002 1E–04 0.004 0.003 0.994 0.997 6 0.762 0.927 0.769 0.179 0.054 0.173 0.059 0.018 0.058 0.762 0.927 0.769 0.178 0.054 0.173 0.215 0.053 0.157 0.265 0.067 0.140 0.521 0.879 0.703 7 0.570 0.780 0.280 0.150 0.001 0.020 0.670 0.830 0.330 0.150 0.015 0.010 0.200 0.045 0.785 0.94 8 Sources: 1 = Steenbergh & Lowe, 1977, in Silvertown et al., 1993; 2 = Godínez-Álvarez et al., 1999; 3 = Esparza-Olguín et al., 2002; 4 = Ortega-Baes, 2001; 5 = Valverde et al., in press; 6 = Contreras & Valverde, 2002; 7 = Schmalzel et al., 1995; 8 = Mandujano et al., 2001. a One population in each case, sampled in different years. b Two different sites sampled within the same time period. c Two sites sampled in two different years. d Each record correspond to a different site, and in each case the average results of several years is given. THE BOTANICAL REVIEW Carnegia gigantea Neobuxbaumia tetetzo Neobuxbaumia macrocephala a 1 2 Escontria chiotilla b 1 2 Mammillaria magnimamma c 1 2 3 4 Mammillaria crucigera a 1 2 Coryphanta robbinsorum d 1 2 3 Opuntia rastrera d 1 2 Per life-cycle stage DEMOGRAPHY IN THE CACTACEAE 191 long-lived species in which relevant population processes may take place at the scale of decades (Pierson & Turner, 1998; de Kroon et al., 2000). As noted in previous sections, the analysis of transition matrices revealed that mortality decreases and reproduction increases with increasing plant size in all species. Additionally, plant growth rates are slow, which may be noted by consistently low transition and high “stasis” probabilities in most population matrices. Only the matrices for Mammillaria species show rapid growth transitions, in which plants may grow more than one size category from one year to the next. In some species “regressions” to smaller size categories are possible due to branch death, particularly in globose species. Among the species analyzed, vegetative spread occurs only in Opuntia rastrera, although, as noted above, other species may also show this ability. Another aspect that may be noticed from the analyzed projection matrices is the almost total absence of a seed bank in the models, with the exception of O. rastrera. Although it appears that the seeds of various cactus species have the ability to remain viable in the soil for more than a year (see section IV), it is believed that most seeds either germinate (given their high germinability) or die shortly after they are shed. A. ELASTICITY ANALYSES The relative contribution of different transitions to the value of ␭ has traditionally been evaluated through elasticity analyses (de Kroon et al., 1986). In addition to evaluating the elasticity of individual matrix entries, it is possible to add up elasticity values of several matrix entries corresponding to the same demographic process in order to analyze the relative contribution of each process to population growth rate (Silvertown et al., 1993). Here we subdivided the life cycle of cactus species into six different types of transitions, following Silvertown et al. (1993): persistence of seeds in the seed bank (e1), sexual reproduction (e2), clonal propagation (e3), retrogression to smaller size classes (e4), persistence or “stasis” in the current size class (e5), and growth to larger size classes (e6). Table IV shows the values of e1–e6 for the 17 matrices (corresponding to eight species) available in the literature for cactus species. Several trends are apparent from these data: as noted above, seed banks and vegetative propagation are absent in most cases (with the exception of Opuntia rastrera 2); sexual reproduction contributes very little to population growth rate; and the most important contributions are in e5, which correspond to the persistence of individuals in their current size class. These trends are also apparent when the different elasticity values were added up to account for the three main demographic processes considered by Silvertown et al. (1993): fecundity (e1 + e2), survival (e4 + e5), and growth (e3 + e6) (Table IV). Thus, plotting the different populations in the demographic triangle, we found that most of them fell toward the survival end (Fig. 3, top), which coincides with the pattern observed for long-lived perennials (Silvertown et al., 1993). Only one population, Mammillaria magnimamma 2, fell closer to the center of the triangle, which implies a balance between the contribution of growth and survival and a relatively high contribution of fecundity to population growth rate. This population is the one that exhibits the largest ␭. Thus, as observed by other authors (Valverde & Silvertown, 1998; de Kroon et al., 2000), those populations with larger ␭s tend to be located closer to the center of the triangle, whereas lower ␭s generally occupy the far-right corner of the triangle (Fig. 4, top). The position of the different populations in the demographic triangle does not seem to be related to either life-form or longevity, as was suggested by Rosas-Barrera and Mandujano (2002). For instance, the Carnegiea gigantea and the Mammillaria magnimamma 3 populations (a columnar long-lived species and a globose shorter-lived species, respectively) are located at the far-right corner of the triangle, almost in the vortex, while those populations located closer to the center of the triangle include Escontria chiotilla 2, Coryphantha robbinsorum 1, Opuntia rastrera 1, and Mammillaria magnimamma 2, representing three different life-forms. 192 THE BOTANICAL REVIEW Fig. 3 (above). Triangular ordination of elasticity values for the 17 matrices reported in Table IV. The three axes correspond to fecundity, growth, and survival (top) and seedlings, juveniles, and adults (bottom). See the text for further explanation. Fig. 4 (facing page). Elasticity values for the 17 matrices reported in Table IV ordained on a triangular plot representing different demographic processes (top) and different life-cycle stages (bottom), along with the ␭ value for each matrix represented in the y axis. 194 THE BOTANICAL REVIEW In addition to evaluating the contribution of particular demographic processes, as proposed by Silvertown et al. (1993), it is possible to use elasticity values to analyze the contribution of different life-cycle phases to population growth rate. Here we have considered three life-cycle phases: seedlings (newly established individuals, generally corresponding to the first and, occasionally, also the second and the third size classes); juveniles (categories of well-established individuals, with mortality probabilities lower than those of seedlings but no sexual reproduction); and adults (reproductive individuals belonging to larger size classes) (Table IV). In most of the cactus populations we detected a relatively larger contribution of adults to population growth rate, compared with those of seedlings and juveniles (Fig. 3, bottom). Only the two Escontria chiotilla populations (located toward the top of the triangle in Fig. 3, bottom) show a high contribution of juveniles to ␭, whereas Mammillaria magnimamma 2 and Coryphantha robbinsorum 1 show also a substantial contribution from seedlings to ␭. The latter high elasticity values for the seedling phase coincide with the largest ␭s (Fig. 4, bottom). This pattern is not particularly clear, however, since some populations with very low ␭ values have relatively high seedling-elasticity values (i.e., Carnegiea gigantea) and some populations with a relatively high ␭ show very low elasticity values for the seedling phase (Table IV; Fig. 4, bottom). B. MATRIX MODELS AND MATRIX SIMULATIONS In some of the demographic studies compiled in Table IV, authors also report the results of matrix simulation exercises in which particular matrix entries were modified to explore the potential effect of these changes on population growth rate. A common result of these exercises is that a substantial increase in fecundity does result in an increase in population growth rate, despite the fact that fecundity entries generally show low elasticities, whereas changes in other matrix entries generally have a low impact on ␭ (Schmalzel et al., 1995; Ortega-Baes, 2001; Contreras & Valverde, 2002; Esparza-Olguín et al., 2002). Mandujano et al. (2001) combined the demographic information for Opuntia rastrera for eight years and at two sites in order to generate different projection models; that is, the time invariant (most generally used) yearly matrices, average matrices for several years, periodic matrices, and stochastic projections. They conclude that the ␭ values obtained using these different methods are quite similar. However, they assert that average matrices are not an adequate way of exploring environmental heterogeneity, because they do not take into account variability in demographic behavior across years. In this case the authors recommend the use of periodic matrices because they integrate interannual variation and give more consistent elasticity values (Mandujano et al., 2001). Stochastic simulations have also been used in Mammillaria magnimamma (Valverde et al., in press). The authors suggest that this type of approach may be useful in projecting long-term population dynamics of cactus species, especially as an aid to understanding not only long-term population growth rates but also transient dynamics. This approach may be especially fruitful when the probability of occurrence of environmental phenomena that most dramatically affect mortality and recruitment are known and can be incorporated in the stochastic projection (Valverde et al., in press). VIII. Conclusions and Perspectives The review presented here leaves us with the certainty that cactus population ecology is definitely a consolidating subject within plant ecology. The number of publications and the amount of ecological information on the subject is already substantial and is growing day by day. Yet certain topics are clearly more developed than others. The ecology of seed germination is perhaps one of the subjects to which greater attention has been directed, although we still DEMOGRAPHY IN THE CACTACEAE 195 lack information about the potential role of seed banks. In recent years the nurse–cactus association has also been the subject of a substantial amount of work; however, we need further research in order to evaluate the importance of this phenomenon for cacti with different lifeforms. On the other hand, analysis of cactus population dynamics through matrix models is rather incipient. On this same topic, the relative importance of sexual reproduction versus vegetative propagation in the maintenance of those populations that may show clonal growth is still unknown. In regard to the different life-forms represented in the Cactaceae, columnar cacti have received the greatest attention; ecological research on the globose and the barrel cacti, as well as the cylindropuntias and platyopuntias, has been less intense. In addition, most of the literature refers to North American and Central American cacti, and little is known about South American species. In exploring the literature on cactus population ecology, it has been interesting to discover that biotic interactions such as nurse–cactus associations, pollination and seed dispersal by animals, seed predation, herbivory, and competition, appear to be important organizing forces in desert and semi-desert communities (Valiente-Banuet & Godínez-Álvarez, 2002; ValienteBanuet et al., 2002). Previous work on desert ecology had failed to recognize this aspect, since it was believed that the main organizing forces of communities in stressful habitats were limiting abiotic factors (Noy-Meir, 1973, 1974, 1979/1980; Grime, 1979). However, the evidence compiled in this review suggests that biotic forces may exert dramatic effects on desert populations, therefore functioning as key factors determining community composition and structure. We conclude with a reflection on cactus conservation. As we mentioned before, it is well known that many cactus species are vulnerable or in danger of extinction. Yet, of all the cactus species for which any kind of population ecology information is available, only a few have been classified as rare, vulnerable, or endangered (see Appendix 1). Only through a better understanding of population processes will we be able to account for ecological phenomena such as rarity and population decline, which are widespread among endangered plants. Thus we must intensify our efforts to deepen our understanding of cactus population dynamics if we are to offer any tools with which to preserve the biodiversity of this fascinating plant family. IX. Acknowledgments We are grateful to M. C. Mandujano for supplying us with a copy of Rosas-Barrera & Mandujano (2002) even before it was published. We thank M. Franco and J. Silvertown for allowing us to use the matrix they compiled for Carnegiea gigantea (used in Silvertown et al., 1993) from previously published work. We are also grateful to A. Valiente-Banuet, who supervised the work of P. Ortega-Baes on Escontria chiotilla. X. Literature Cited Altesor, A., E. Ezcurra & C. Silva. 1992. Changes in the photosynthetic metabolism during the early ontogeny of four cactus species. Acta Oecol. 13: 777–785. Anderson, E. F., S. Arias Montes & N. P. Taylor. 1994. Threatened cacti of Mexico. Royal Botanical Gardens, Kew, England. Arriaga, L., Y. Maya, S. Díaz & J. Cancino. 1993. Association between cacti and nurse perennials in a heterogeneous tropical dry forest in northwestern Mexico. J. Veg. Sci. 4: 349–356. Bierzychudek, P. 1982. Life histories and demography of shade-tolerant temperate forest herbs: A review. New Phytol. 90: 757–776. Bowers, J. E. 1996a. Growth rate and life span of a prickly pear cactus, Opuntia engelmannii, in the northern Sonoran desert. Southw. Naturalist 41: 315–318. 196 THE BOTANICAL REVIEW ———. 1996b. Environmental determinants of flowering date in the columnar cactus Carnegiea gigantea in the Sonoran Desert. Madroño 43: 69–84. ———. 1996c. More flowers or new cladodes? Environmental correlates and biological consequences of sexual reproduction in a Sonoran Desert prickly pear cactus, Opuntia engelmannii. Bull. Torrey Bot. Club 123: 34–40. ———. 1997a. The effect of drought on Engelmann prickly pear (Cactaceae: Opuntia engelmannii) fruit and seed production. Southw. Naturalist 42: 240–242. ———. 1997b. Demographic patterns of Ferocactus cylindraceus in relation to substrate age and grazing history. Pl. Ecol. 133: 37–48. ———. 2000. Does Ferocactus wislizeni (Cactaceae) have a between-year seed bank? J. Arid Environm. 45: 197–205. ———, R. H. Webb & R. J. Rondeau. 1995. Longevity, recruitment and mortality of desert plants in Grand Canyon, Arizona, USA. J. Veg. Sci. 6: 551–564. Bravo-Hollis, H. & H. Sánchez-Mejorada. 1978. Las Cactáceas de México. Vol. 1. Universidad Nacional Autónoma de México, Mexico City. ——— & ———. 1991. Las Cactáceas de México. Vols. 2–3. Universidad Nacional Autónoma de México, Mexico City. Breckenridge, F. G., III & J. M. Miller. 1982. Pollination biology, distribution, and chemeotaxonomy of the Echinocereus enneacanthus complex (Cactaceae). Syst. Bot. 7: 365–378. Briones, O., C. Montaña & E. Ezcurra. 1998. Competition intensity as a function of resource availability in a semiarid ecosystem. Oecologia 116: 365–372. Callaway, R. M. 1995. Positive interactions among plants. Bot. Rev. (Lancaster) 61: 306–349. ———. 1998. Are positive interactions species-specific? Oikos 82: 202–207. Carrillo-García, A., Y. Bashan & G. J. Bethlenfalvay. 2000. Resource-island soils and the survival of the giant cactus, cardon, of Baja California Sur. Pl. & Soil 218: 207–214. Casas, A., A. Valiente-Banuet, A. Rojas-Martínez & P. Dávila. 1999. Reproductive biology and the process of domestication of the columnar cactus Stenocereus stellatus in central Mexico. Amer. J. Bot. 86: 534–542. Cody, M. L. 1993. Do cholla (Opuntia spp., Subgenus Cylindropuntia) use or need nurse plants in the Mojave Desert? J. Arid Environm. 24: 139–154. ———. 2000. Slow-motion population dynamics in Mojave Desert perennial plants. J. Veg. Sci. 11: 351–358. Contreras, C. & T. Valverde. 2002. Evaluation of the conservation status of a rare cactus (Mammillaria crucigera) through the analysis of its population dynamics. J. Arid Environm. 51: 89–102. De Kroon, H., H. Plaiser, J. van Groenendael & H. Caswell. 1986. Elasticity: The relative contribution of demographic parameters to population growth rate. Ecology 67: 1427–1431. ———, ——— & J. Ehrlen. 2000. Elasticities: A review of methods and model limitations. Ecology 81: 607–618. De Viana, M. L. 1996–1997. Distribución espacial de Trichocereus pasacana (Cactaceae) en relación al espacio disponible y al banco de semillas. Revista Biol. Trop. 44/45: 95–103. ———, N. Acreche, R. Acosta & L. Moraña. 1990. Población y asociaciones de Trichocereus pasacana (Cactaceae) en Los Cardones, Argentina. Revista Biol. Trop. 38: 383–386. ———, S. Sühring & B. F. J. Manly. 2001. Application of randomization methods to study the association of Trichocereus pasacana (Cactaceae) with potential nurse plants. Pl. Ecol. 156: 193–197. Dubrovsky, J. G. 1996. Seed hydration memory in Sonoran Desert cacti and its ecological implication. Amer. J. Bot. 83: 624–632. ———. 1998. Discontinuous hydration as a facultative requirement for seed germination in two cactus species of the Sonoran Desert. J. Torrey Bot. Soc. 125: 33–39. Esparza-Olguín, L., T. Valverde & E. Vilchis-Anaya. 2002. Demographic analysis of a rare columnar cactus (Neobuxbaumia macrocephala) in the Tehuacan Valley, Mexico. Biol. Con. 103: 349–359. Figueira, J. E. C., J. Vasconcellos-Neto, M. A. Garcia & A. L. Teixeira de Souza. 1994. Saurocory in Melocactus violaceus (Cactaceae). Biotropica 26: 295–301. Fleming, T. H. & A. Valiente-Banuet (eds.). 2002. Columnar cacti and their mutualists: Evolution, ecology, and conservation. University of Arizona Press, Tucson. ———, C. T. Sahley, J. N. Holland, J. D. Nason & J. L. Hamrick. 2001. Sonoran Desert columnar cacti and the evolution of generalized pollination systems. Ecol. Monogr. 71: 511–530. DEMOGRAPHY IN THE CACTACEAE 197 Flores, F. J. L. & R. I. Yeaton. 2000. La importancia de la competencia en la organización de las comunidades vegetales en el altiplano mexicano. Interciencia (Caracas) 25: 365–371. Flores, J. & O. Briones. 2001. Plant life-form and germination in a Mexican inter-tropical desert: Effects of soli water potential and temperature. J. Arid Environm. 47: 485–497. Flores-Martínez, A., E. Ezcurra & S. Sánchez-Colón. 1994. Effect of Neobuxbaumia tetetzo on growth and fecundity of its nurse plant Mimosa luisana. J. Ecol. 82: 325–330. ———, ——— & ———. 1998. Water availability and the competitive effect of a columnar cactus on its nurse plant. Acta Oecol. 19: 1–8. Franco, A. C. & P. S. Nobel. 1989. Effect of nurse plants on the microhabitat and growth of cacti. J. Ecol. 77: 870–886. García-Moya, E. & M. C. McKell. 1970. Contribution of shrubs in the economy of a desert-wash plant community. Ecology 51: 81–88. Gibson, A. C. & P. S. Nobel. 1986. The cactus primer. Harvard Univ. Press. Cambridge, MA. Godínez-Álvarez, H. & A. Valiente-Banuet. 1998. Germination and early seedling growth of Tehuacan Valley cacti species: The role of soils and seed ingestion by dispersers on seedling growth. J. Arid Environm. 39: 21–31. ———, ——— & B. L. Valiente. 1999. Biotic interactions and the population dynamics of the longlived columnar cactus Neobuxbaumia tetetzo in the Tehuacán Valley, Mexico. Canad. J. Bot. 77: 203–208. ———, ——— & A. Rojas-Martínez. 2002. The role of seed dispersers in the population dynamics of the columnar cactus Neobuxbaumia tetetzo. Ecology 83: 2617–2629. Goldberg, D. E. & R. M. Turner. 1986. Vegetation change and plant demography in permanent plots in the Sonoran Desert. Ecology 67: 695–712. Grant, V., K. A. Grant & P. D. Hurd Jr. 1979. Pollination of Opuntia lindheimeri and related species. Pl. Syst. Evol. 132: 313–320. Grime, J. P. 1979. Plant strategies and vegetation processes. Wiley, Chichester. Gulmon, S. L., P. W. Rundel, J. R. Ehleringer & H. A. Mooney. 1979. Spatial relationships and competition in a Chilean desert cactus. Oecologia 44: 40–43. Haman, O. 2001. Demographic studies of three indigenous stand-forming plant taxa (Scalesia, Opuntia, and Bursera) in the Galapagos Islands, Ecuador. Biodiv. & Conserv. 10: 223–250. Hastings, J. R. & S. M. Alcorn. 1961. Physical determinations of growth and age in the giant cactus. J. Ariz. Acad. Sci. 2: 32–39. Hernández, H. M. & H. Godínez-Álvarez. 1994. Contribución al conocimiento de las Cactáceas mexicanas amenazadas. Acta Bot. Mex. 26: 33–52. Hicks, D. J. & A. Mauchamp. 2000. Population structure and growth patterns of Opuntia echios var. gigantea along an elevational gradient in the Galapagos Islands. Biotropica 32: 235–243. Horvitz, C. C. & D. W. Schemske. 1995. Spatiotemporal variation in demographic transitions of a tropical understorey herb: Projection matrix analysis. Ecol. Monogr. 65: 155–192. Huerta, F. M. & V. E. Escobar. 1998. Estatus ecológico actual de Ferocactus hystrix (DC) Lindsay en los Llanos de Ojuelos, Jalisco-Zacatecas. Cact. Suc. Mex. 43: 57–64. Hunt, D. (comp.). 1999. C.I.T.E.S. Cactaceae checklist. Ed. 2. Royal Botanic Gardens & International Organization for Succulent Plant Study, Kew, England. Hutto, R. L., J. R. McAuliffe & L. Hogan. 1986. Distributional associates of the saguaro (Carnegiea gigantea). Southw. Naturalist 31: 469–476. Johnson, R. A. 1992. Pollination and reproductive ecology of Acuña cactus, Echinomastus erectrocentrus var. acunensis (Cactaceae). Int. J. Pl. Sci. 153: 400–408. ———, M. A. Baker, D. J. Pinkava & G. A. Ruffner. 1992. Seedling establishment, mortality and flower production of the Acuña cactus, Echinomastus eretrocentrus var. acunensis (Cactaceae). Pp. 170–180 in R. Sivinski & K. Lightfoot (eds.), Southwestern rare and endangered plants: Proceedings of the Southwestern Rare and Endangered Plant Conference. New Mexico Forestry and Resources Conservation Division, Energy, Minerals and Natural Resources Department, Santa Fe, NM. Jordan, P. W. & P. S. Nobel. 1981. Seedling establishment of Ferocactus acanthodes in relation to drought. Ecology 62: 901–906. ——— & ———. 1982. Height distribution of two species of cacti in relation to rainfall, seedling establishment, and growth. Bot. Gaz. 143: 511–517. 198 THE BOTANICAL REVIEW Larmuth, J. & H. J. Harvey. 1978. Aspects of the occurrence of desert plants. J. Arid Environm. 1: 129–133. Leirana-Alcocer, J. & V. Parra-Tabla. 1999. Factors affecting the distribution, abundance and seedling survival of Mammillaria gaumeri, an endemic cactus of coastal Yucatán, Mexico. J. Arid Environm. 41: 421–428. León de la Luz, J. L. & R. D. Cadena. 1991. Evaluación de la reproducción por semilla de la pitaya agria (Stenocereus gummosus) en Baja California Sur, México. Acta Bot. Mex. 14: 75–87. Lockwood, M. W. 1995. Notes on life history of Ancistrocactus tobuschii (Cactaceae) in Kinney County, Texas. Southw. Naturalist 40: 428–430. Mandujano, M. C., C. Montaña & L. Eguiarte. 1996. Reproductive ecology and inbreeding depression in Opuntia rastrera (Cactaceae) in the Chihuahuan Desert: Why are sexually derived recruitments so rare?. Amer. J. Bot. 83: 63–70. ———, ———, I. Méndez & J. Golubov. 1998. The relative contributions of sexual reproduction and clonal propagation in Opuntia rastrera from two habitats in the Chihuahuan Desert. J. Ecol. 86: 911–921. ———, ———, M. Franco, J. Golubov & A. Flores-Martínez. 2001. Integration of demographic annual variability in a clonal desert cactus. Ecology 82: 344–359. Martínez, D., A. Flores-Martínez, F. López & G. Manzanero. 2001. Aspectos ecológicos de Mammillaria oteroi Glass & R. Foster en la región mixteca de Oaxaca, México. Cact. Suc. Mex. 46: 32–39. Martínez, J. G., H. Suzán & C. A. Salazar. 1993. Aspectos ecológicos y demográficos de Ariocarpus trigonus (Weber) Schumann. Cact. Suc. Mex. 38: 30–38. ———, ——— & ———. 1994. Aspectos ecológicos y demográficos de Neolloydia pseudopectinata (Backeberg) E. F. Anderson. Cact. Suc. Mex. 39: 27–33. Martínez-Ramos, M. & E. Álvarez-Buylla. 1995. Ecología de poblaciones de plantas de una selva húmeda de México. Bol. Soc. Bot. México 56: 121–153. McAuliffe, J. R. 1984a. Sahuaro–nurse tree associations in the Sonoran Desert: Competitive effects of sahuaros. Oecologia 64: 319–321. ———. 1984b. Prey refugia and the distributions of two sonoran desert cacti. Oecologia 65: 82–85. ——— & F. J. Janzen. 1986. Effects of intraspecific crowding on water uptake, water storage, apical growth, and reproductive potential in the sahuaro cactus, Carnegiea gigantea. Bot. Gaz. 147: 334–341. McFarland, J. D., P. G. Kevan & M. A. Lane. 1989. Pollination biology of Opuntia imbricata (Cactaceae) in southern Colorado. Canad. J. Bot. 67: 24–28. McIntosh, M. E. 2002. Flowering phenology and reproductive output in two sister species of Ferocactus (Cactaceae). Pl. Ecol. 159: 1–13. Meyrán, J. 1980. Guía botánica de Cactáceas y otras suculentas del Valle de Tehuacán. Sociedad Mexicana de Cactología A.C., Mexico City. Mourelle, C. & E. Ezcurra. 1997. Differentiation diversity of Argentine cacti and its relationship to environmental factors. J. Veg. Sci. 8: 547–558. Nassar, J. M., N. Ramírez & O. Linares. 1997. Comparative pollination biology of Venezuelan columnar cacti and the role of nectar-feeding bats in their sexual reproduction. Amer. J. Bot. 84: 918–927. Negrón-Ortiz, V. 1998. Reproductive biology of a rare cactus, Opuntia spinosissima (Cactaceae), in the Florida Keys: Why is seed set very low?. Sexual Pl. Reprod. 11: 208–212. Niering, W. A., R. H. Whittaker & C. H. Lowe. 1963. The saguaro: A population in relation to environment. Science 142: 15–23. Nobel, P. S. 1980. Morphology, nurse plant and minimum apical temperatures for young Carnegiea gigantea. Bot. Gaz. 141: 188–191. ———. 1988. Environmental biology of agaves and cacti. Cambridge University Press. Cambridge, MA. ———. 1989. Temperature, water availability, and nutrient levels at various soil depths: Consequences for shallow-rooted desert succulents, including nurse plant effects. Amer. J. Bot. 76: 1486–1489. ———, G. N. Geller, S. C. Kee & A. D. Zimmerman. 1986. Temperatures and thermal tolerances for cacti exposed to high temperatures near the soil surface. Pl. Cell Environ. 9: 279–287. Nolasco, H., F. Vega-Villasante & A. Díaz-Rondero. 1997. Seed germination of Stenocereus thurberi (Cactaceae) under different solar irradiation levels. J. Arid Environm. 36: 123–132. Noy-Meir, I. 1973. Desert ecosystems: Environment and producers. Annual Rev. Ecol. Syst. 4: 25–51. ———. 1974. Desert ecosystems: higher trophic levels. Annual Rev. Ecol. Syst. 5: 195–214. DEMOGRAPHY IN THE CACTACEAE 199 ———. 1979/1980. Structure and function of desert ecosystems. Israel J. Bot. 28: 1–19. Ortega-Baes, P. F. 2001. Demografía de la cactácea columnar Escontria chiotilla. M.S. thesis, Universidad Nacional Autónoma de México, Mexico City. Osborn, M. M., P. G. Kevan & M. A. Lane. 1988. Pollination biology of Opuntia polyacantha and Opuntia phaeacantha (Cactaceae) in southern Colorado. Pl. Syst. Evol. 159: 85–94. Parker, K. C. 1987. Seedcrop characteristics and minimum reproductive size of organ pipe cactus (Stenocereus thurberi) in southern Arizona. Madroño 34: 294–303. ———. 1988a. Environmental relationships and vegetation associates of columnar cacti in the northern Sonoran Desert. Vegetatio 78: 125–140. ———. 1988b. Growth rates of Stenocereus thurberi and Lophocereus schottii in southern Arizona. Bot. Gaz. 149: 335–346. ———. 1989. Height structure and reproductive characteristics of Senita, Lophocereus schottii (Cactaceae), in southern Arizona. Southw. Naturalist 34: 392–401. ———. 1991. Topography, substrate, and vegetation patterns in the northern Sonoran Desert. J. Biogeogr. 18: 151–163. ———. 1993. Climatic effects on regeneration trends for two columnar cacti in the northern Sonoran Desert. Ann. Assoc. Amer. Geogr. 83: 452–474. ——— & J. L. Hamrick. 1992. Genetic diversity and clonal structure in a columnar cactus, Lophocereus schottii. Amer. J. Bot. 79: 86–96. Pearl, R. 1928. The rate of living, being an account of some experimental studies on the biology of life duration. Knopf, New York.. Pierson, E. A. & R. M. Turner. 1998. An 85-year study of saguaro (Carnegiea gigantea) demography. Ecology 79: 2676–2693. Pimienta, E., G. Hernández, A. Domingues & P. S. Nobel. 1998. Growth and development of the arborescent cactus Stenocereus queretaroensis in a subtropical semiarid environment, including effects of gibbberellic acid. Tree Physiol. 18: 59–64. Reid, W., R. Lozano & R. Odom. 1983. Non-equilibrium population structure in three Chihuahuan Desert cacti. Southw. Naturalist 28: 115–117. Rincón, E., P. Huante & Y. Ramírez. 1993. Influence of vesicular-arbuscular mycorrhizae on biomass production by the cactus Pachycereus pecten-aboriginum. Mycorrhiza 3: 79–81. Rodríguez-Ortega, C. E. & E. Ezcurra. 2001. Distribución espacial en el hábitat de Mammillaria pectinifera y M. carnea en el Valle de Zapotitlán Salinas, Puebla, México. Cact. Suc. Mex. 45: 4–14. ——— & M. Franco. 2001. La retención de semillas en el género Mammillaria. Cact. Suc. Mex. 46: 63–67. Rojas-Aréchiga, M. & A. Batis. 2001. Las semillas de Cactáceas ¿forman bancos en el suelo?. Cact. Suc. Mex. 46: 76–82. ——— & C. Vázquez-Yanes. 2000. Cactus seed germination: A review. J. Arid Environm. 44: 85–104. ———, A. Casas & C. Vázquez-Yanes. 2001. Seed germination of wild and cultivated Stenocereus stellatus (Cactaceae) from the Tehuacán-Cuicatlán Valley, Central Mexico. J. Arid Environm. 49: 279–287. Rojas-Martínez, A., A. Valiente-Banuet, M. C. Arizmendi, A. Alcántara-Eguren & H. T. Arita. 1999. Seasonal distribution of the long-nosed bat (Leptonycteris curasoae) in North America: Does a generalized migration pattern really exist?. J. Biogeogr. 26: 1065–1077. Rosas-Barrera, M. L. & M. C. Mandujano. 2002. La diversidad de historias de vida de Cactáceas, aproximación por el triángulo demográfico. Cact. Suc. Mex. 47: 33–41. Ruedas, M., T. Valverde & S. C. Argüero. 2000. Respuesta germinativa y crecimiento temprano de plántulas de Mammillaria magnimamma (Cactaceae) bajo diferentes condiciones ambientales. Bol. Soc. Bot. México 66: 25–35. Ruíz, A., M. Santos, J. Cavalier & P. J. Soriano. 2000. Estudio fenológico de Cactáceas en el enclave seco de la Tatacoa, Colombia. Biotropica 32: 397–407. Sahley, C. T. 1996. Bat and hummingbird pollination of an autotetraploid columnar cactus, Werberbaurocereus werberbaueri (Cactaceae). Amer. J. Bot. 83: 1329–1336. Schlindwein, C. & D. Wittmann. 1997. Stamen movements in flowers of Opuntia (Cactaceae) favour oligolectic pollinators. Pl. Syst. Evol. 204: 179–193. Schmalzel, R. J., F. W. Reichenbacher & S. Rutman. 1995. Demographic study of the rare Coryphantha robbinsorum (Cactaceae) in southeastern Arizona. Madroño 42: 332–348. 200 THE BOTANICAL REVIEW Shreve, F. 1910. The rate of establishment of the giant cactus. Pl. World 10: 235–240. ———. 1931. Physical conditions in sun and shade. Ecology 12: 96–104. Silva, W. R. 1988. Ornitocoria em Cereus peruvianus (Cactaceae) na Serra do Japi, Estado de Sao Paulo. Revista Brasil. Biol. 48: 381–389. Silvertown, J. W. 1987. Introduction to plant population ecology. Ed. 2. Longman Scientific & Technical, Essex, England. ——— & J. B. Wilson. 1994. Community structure in a desert perennial community. Ecology 75: 409–417. ———, M. Franco, I. Pisanty & A. Mendoza. 1993. Comparative plant demography: Relative importance of life-cycle components to the finite rate of increase in woody and herbaceous perennials. J. Ecol. 81: 465–476. Silvius, K. M. 1995. Avian consumers of cardón fruits (Stenocereus griseus: Cactaceae) on Margarita Island, Venezuela. Biotropica 27: 96–105. Spears, E. E., Jr. 1987. Island and mainland pollination ecology of Centrosema virginianum and Opuntia stricta. J. Ecol. 75: 351–362. Steenbergh, W. F. & C. H. Lowe. 1969. Critical factors during the first years of life of the saguaro (Cereus giganteus) at Saguaro National Monument, Arizona. Ecology 50: 823–834. ——— & ———. 1977. Ecology of the saguaro: II, Reproduction, germination, establishment, growth, and survival of the young plant. Department of the Interior, National Park Service, Washington, DC. Suzán, H., G. P. Nabhan & D. T. Patten. 1996. The importance of Olneya tesota as a nurse plant in the Sonoran Desert. J. Veg. Sci. 7: 635–644. Tinoco-Ojanguren, C. & F. Molina-Freaner. 2000. Flower orientation in Pachycereus pringlei. Canad. J. Bot. 78: 1489–1494. Tschapka, M., O. von Helversen & W. Barthlott. 1999. Bat pollination of Weberocereus tunilla, an epiphytic rain forest cactus with functional flagelliflory. Pl. Biol. (Germany) 1: 554–559. Turner, R. M., S. M. Alcorn, G. Olin & J. A. Booth. 1966. The influence of shade, soil, and water on saguaro seedling establishment. Bot. Gaz. 127: 95–102. Valiente-Banuet, A. & E. Ezcurra. 1991. Shade as a cause of the association between the cactus Neobuxbaumia tetetzo and the nurse plant Mimosa luisana in the Tehuacán Valley, Mexico. J. Ecol. 79: 961–971. ——— & H. Godínez-Álvarez. 2002. Population and community ecology. Pp. 91–108 in P. S. Nobel (ed.), Cacti: Biology and uses. University of California Press, Berkeley. ———, A. Bolongaro-Crevenna, O. Briones, E. Ezcurra, M. Rosas, H. Núñez, G. Barnard & E. Vázquez. 1991a. Spatial relationships between cacti and nurse shrubs in a semi-arid environment in central Mexico. J. Veg. Sci. 2: 15–20. ———, F. Vite & J. A. Zavala-Hurtado. 1991b. Interaction between the cactus Neobuxbaumia tetetzo and the nurse shrub Mimosa luisana. J. Veg. Sci 2: 11–14. ———, P. Dávila, R. J. Ortega, M. C. Arizmendi, J. L. León, A. Breceda & J. Cancino. 1995. Influencia de la evolución de una pendiente de piedemonte en una vegetación de cardonal de Pachycereus pringlei en Baja California Sur, México. Investigación Geogr. Bol. 3: 101–113. ———, M. C. Arizmendi, A. Rojas-Martínez & L. Domínguez-Canseco. 1996. Ecological relationships between columnar cacti and nectar-feeding bats in Mexico. J. Trop. Ecol. 12: 103–119. ———, A. Rojas-Martínez, M. C. Arizmendi & P. Dávila. 1997a. Pollination biology of two columnar cacti (Neobuxbaumia mezcalaensis and Neobuxbaumia macrocephala) in the Tehuacan Valley, central Mexico. Amer. J. Bot. 84: 452–455. ———, ———, A. Casas, M. C. Arizmendi & P. Dávila. 1997b. Pollination biology of two winterblooming giant columnar cacti in the Tehuacán Valley, central Mexico. J. Arid Environm. 37: 331–341. ———, M. C. Arizmendi, A. Rojas-Martínez, A. Casas, H. Godínez-Álvarez, C. Silva & P. Dávila. 2002. Biotic interactions and population dynamics of columnar cacti. Pp. 225–240 in T. H. Fleming & A. Valiente-Banuet (eds.), Columnar cacti and their mutualists: Evolution, ecology, and conservation. University of Arizona Press, Tucson. Valverde, T. & J. Silvertown. 1998. Spatial variation in the demography of a forest understory herb (Primula vulgaris): Projection matrix models. J. Ecol. 86: 545–562. ———, M. L. Trejo & S. Castillo. 1999. Patrón de distribución y abundancia de Mammillaria magnimamma en la reserva del pedregal de San Angel, México D. F. Cact. Suc. Mex. 44: 64–74. DEMOGRAPHY IN THE CACTACEAE 201 ———, S. Quijas, M. López-Villavicencio & S. Castillo. In press. Population dynamics of Mammillaria magnimamma Haworth. (Cactaceae) in a Mexican lava-field. Pl. Ecol. Weiss, J., A. Nerd & Y. Mizrahi. 1994. Flowering behavior and pollination requirements in climbing cacti with fruit crop potential. HortScience 29: 1487–1492. Yeaton, R. I. 1978. A cyclical relationship between Larrea tridentata and Opuntia leptocaulis in the northern Chihuahuan Desert. J. Ecol. 66: 651–656. Zavala-Hurtado, J. A. & A. Díaz-Solís. 1995. Repair, growth, age and reproduction in the giant columnar cactus Cephalocereus columna-trajani (Karwinski ex. Pfeiffer) Schumann (Cactaceae). J. Arid Environm. 31: 21–31. Globose cacti Ancistrocactus tobuschii Ariocarpus trigonus Coryphantha robbinsorum Mammillaria crucigera M. gaumeri 1 3 1 2/3 M. magnimamma 3 1 3 M. microcarpa Neolloydia pseudopectinata 3 3 1 2 Barrel cacti Copiapoa cinerea Echinocactus horizonthalonius Echinocereus engelmanni E. triglochidiatus 1/3 3 CITES IUCN I I 3 1 Endangered species 1 v e r 1 3 1 1 3 1 Geographical distribution Source N* N* N Lockwood, 1995 Martínez et al., 1993 Schmalzel et al., 1995 N* N* N* N* Contreras & Valverde, 2002 Leirana-Alcocer & Parra-Tabla, 1999 Valverde et al., 1999; Ruedas et al., 2000; Valverde et al., in press McAuliffe, 1984b Martínez et al., 1994 N* I 1 3 v v S N Gulmon et al., 1979 Reid et al., 1983 3 e i N N McAuliffe, 1984b Reid et al., 1983 1 THE BOTANICAL REVIEW Species Pattern of Recruit- SurvivorSize Population distribution ment ship Fecundity structure dynamics 202 XI. Appendix 1: Demographic information, endangerment, and geographical distribution of cactus species with different life-forms. Pattern of distribution: 1= random, 2 = regular, 3 = clumped; Recruitment: 1 = under the canopy of trees and shrubs, 2 = in spaces deprived of vegetation; Survivorship: 1 = mortality is higher in the last size/age categories than in the first ones, 2 = mortality remains constant with age/size, 3 = mortality is higher in the first age/size categories than in later ones; Fecundity: 1 = fecundity increases with size/age, 2 = fecundity increases until a certain size/age and then remains constant, 3 = fecundity increases with size/age, reaching a maximum, then decreases; Size structure: 1 = number of individuals decreases with size/age, 2 = number of individuals increases with size/age, 3 = number of individuals varies with size/age; Population dynamics: 1 = species analyzed with matrix models; CITES: I = species listed in appendix I; IUCN: e = endangered, v = vulnerable, r = rare, i = indeterminate;Geographical distribution: N = North America, S = South America, * = endemic to one country of the geographical region. 1 Echinomastus erectrocentrus Ferocactus acanthodes 1 F. cylindraceus F. histrix F. wislizeni Columnar catci Carnegiea gigantea Pachycereus pringlei Stenocereus thurberi Trichocereus pasacana 1 1 3 1 3 1 1 1 1 3 1 N 3 1 3 N N* N 3 N 1 N* 3 3 3 1 1 1 1 i 3 e N* N S 1/3 i S* 1 1 3 1 3 1 N* N N* N* 1 3 N 1 3 Opuntioid cacti Opuntia echios O. rastrera 3 e 3 1 N* Johnson, 1992; Johnson et al., 1992 Jordan & Nobel, 1981; Franco & Nobel, 1989 Bowers, 1997b Huerta & Escobar, 1998 Reid et al., 1983 Niering et al., 1963; Turner et al., 1966; Steenbergh & Lowe, 1969; Hutto et al., 1986; Franco & Nobel, 1989; Parker, 1993; Silvertown et al., 1993; Pierson & Turner, 1998 Zavala-Hurtado & DíazSolís, 1995 Ortega-Baes, 2001 Parker, 1989 Esparza-Olguín et al., 2002 Valiente-Banuet & Ezcurra, 1991; Valiente-Banuet et al., 1991a, 1991b; Godínez-Álvarez e al., 1999 Carrillo-García et al., 2000 Parker, 1987, 1993 De Viana, 1996–1997 DEMOGRAPHY IN THE CACTACEAE Cephalocereus columnatrajani Escontria chiotilla Lophocereus schottii Neobuxbaumia macrocephala N. tetetzo 3 I Hicks & Mauchamp, 2000; Hamann, 2001 Mandujano et al., 1998, 2001 203