| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ecology |
University of Colorado Museum, University of Colorado, 265 UCB—Bruce Curtis Building, Boulder, Colorado 80309-0265 USA; and 3Department of Ecology and Evolutionary Biology, University of Colorado, Boulder, Colorado 80309 USA
Received for publication February 20, 2006. Accepted for publication October 5, 2006.
ABSTRACT
Humans are having a profound impact on the geographic distributions of plant populations. In crop species, domestication has been accompanied by the geographic expansion of cultivated populations relative to their wild ancestors. We used a geographical information system (GIS)-based approach to investigate differences in the environmental factors characterizing the geographic distributions of cultivated and wild populations of the Mesoamerican fruit tree Spondias purpurea. Locality data for 86 cultivated and 28 wild S. purpurea populations were used in conjunction with environmental data layers and Maxent, a maximum entropy application for predicting species distributions. Interpredictivity analyses and principal components analysis revealed that the predicted distribution of wild S. purpurea is nested within the cultivated distribution and that the ecological niche (defined by environmental characteristics) of cultivated S. purpurea has expanded relative to that of wild populations. Significant differences between wild and cultivated populations were detected for five environmental variables, corresponding to the expansion of S. purpurea during the domestication process from its native habitat in the Mesoamerican tropical dry forests into less seasonal habitats. These data suggest that humans have altered the range of habitats occupied by cultivated S. purpurea populations relative to their wild progenitors.
Key Words: Anacardiaceae • domestication • Mesoamerica • niche conservatism • species distribution model • Spondias purpurea • tropical dry forest
Humans are influencing the geographic distributions of plants through displacement resulting from habitat destruction, the deliberate spread of economically valuable taxa, and through accidental introductions. In many crop species, the geographic area occupied by cultivated populations has expanded dramatically in recent times relative to the areas occupied by their wild progenitors. This expansion is evidenced by the contemporary distributions of many of the world's most economically important crops (e.g., chilis, cucumbers, oranges, peanuts, pineapples, potatoes, rice, soybeans, tomatoes, wheat), with the bulk of their production on continents other than those where the crop originated (Simpson and Ogorzaly, 1995
). Although the expanded geographic distribution of cultivated populations relative to their wild progenitors is well known, specific differences in the environmental characteristics of regions occupied by cultivated vs. wild populations have not been documented. Have humans simply transported cultivated individuals into regions that resemble the habitat of their wild progenitors? Or under human influence, have cultivated populations been able to expand into regions that differ significantly from regions occupied by their wild progenitors?
Plant domestication occurs as humans selectively maintain and/or cultivate in agricultural habitats a subset of wild individuals. During the course of domestication, evolutionary processes such as selection and drift result in morphological and genetic changes in the cultivated populations making them distinct from their wild progenitors (Clegg et al., 1984
; Gepts and Clegg, 1989
; Ennos, 1997
; Eyre-Walker et al., 1998
; Saunders et al., 2001
; Anthony et al., 2002
; Hancock, 2004
). Differences in the environmental characteristics of the regions occupied by cultivated and wild populations could reflect human influences (e.g., transportation, watering, fertilization, protection, clearing of competing plants) that facilitate the persistence of cultivated genotypes in regions where, in the absence of the human contributions, the species does not occur. Alternatively, the expanded geographic range of cultivated populations relative to their wild ancestors could be a result of artificial selection for characteristics that allowed populations to inhabit a wider diversity of habitats.
Differences in the environmental characteristics of areas occupied by organisms can be examined by modeling species distributions, a technique that integrates locality data, GIS data, and modeling algorithms (e.g., Anderson et al., 2002
; Anderson and Martinez-Meyer, 2004
; Elith et al., 2006
; Phillips et al., 2006
). The resulting distribution model describes the common environmental and climatological characteristics of the known range of a given species or group of populations (Peterson, 2003
; Soberón and Peterson, 2004
). This approach has been used to predict species distributions (Illoldi-Rangel et al., 2004
); to predict the potential geographic range of invading species (Peterson, 2003
; Mau-Crimmins et al., 2006
), to examine the evolution of ecological niches (Peterson and Holt, 2003
; Rice et al., 2003
; Martínez-Meyer et al., 2004a
, b
; Hoffman, 2005
), to investigate speciation mechanisms (Graham et al., 2004
), and to predict changes in the distributions of fauna and flora associated with projected models of climate change (Peterson et al., 2002
; Siqueira and Peterson, 2003
; Oberhauser and Peterson, 2003
; Thomas et al., 2004
). In crop species, GIS-based analyses have been used to predict yields of different cultivars in various geographic areas (Jeutong et al., 2000
; Caldiz et al., 2002
), to explore the distributions of wild relatives of crop species (Greene et al., 1999a
, b
; Hijmans and Spooner, 2001
; Jarvis et al., 2004
), and to model future distributions of crop pests and diseases (Bernardi, 2001
).
In this study, we used GIS data sets and predictive modeling to investigate the environmental and climatological factors characterizing the geographic distributions of cultivated populations and the wild populations from which they were derived. We focused on the Mesoamerican fruit tree Spondias purpurea L. (known locally as ciruela Mexicana, jocote, purple mombin, or hog plum), a species cultivated throughout the neotropics and subtropics for its plumlike fruits, which are eaten fresh, sold in local markets, and made into jams and beverages (Avitia García, 1997
; Baraona Cockrell, 2000
). Although some are intensively cultivated in orchards, the majority of S. purpurea trees are planted in informal agricultural habitats such as backyard gardens, living fences, and small multicrop farms (Cuevas, 1994
) and have not been subjected to extensive breeding. Cultivated S. purpurea trees were derived from wild populations in at least two distinct geographic regions within Mesoamerica (Miller and Schaal, 2005
, 2006
). Today, the wild (undomesticated) populations of S. purpurea can be found in the tropical dry forests of Mexico and Central America (Mandujano et al., 1994
; Mooney et al., 1995
; Miller and Schaal, 2005
, 2006
). There are clear morphological differences between cultivated and wild S. purpurea populations, indicating that selection and domestication has occurred in this species. Fruits of the wild jocotes are bright red or yellow in some regions (cultivated fruits can be red, orange, yellow, green, or purple) and are smaller and more acidic than the cultivated fruits, with considerably less flesh surrounding the seed. Wild S. purpurea trees reproduce from seed and native populations are age-structured with a variety of juvenile and mature individuals present; cultivated S. purpurea trees are propagated vegetatively (Miller, 2004
).
To quantify differences in the environmental characteristics of regions occupied by cultivated S. purpurea populations and wild S. purpurea populations, we use field-collected locality data, GIS databases, and the distribution modeling method Maxent to (1) model the predicted area of occurrence of wild S. purpurea populations and compare it to the predicted area of occurrence of cultivated S. purpurea populations, and (2) characterize the mean and variance of several environmental parameters for wild and cultivated populations of S. purpurea and examine the null hypothesis that wild and cultivated populations occur in the same types of habitats.
MATERIALS AND METHODS
Sampling
One hundred and fourteen distinct localities with S. purpurea were sampled, and each population included at least one and as many as 80 S. purpurea trees. In total, 86 cultivated populations and 28 wild populations from Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, and Panama were included in the study (Table 1, Fig. 1). All populations were visited by the first author (A.J.M.) at least once during field studies that took place in 2000, 2001, 2002, and 2005. Cultivated populations were distinguished from wild populations by (1) habitat: cultivated populations were found in agricultural environments including backyards, living fences, small farms, and orchards; wild populations were found in primary or secondary forests, (2) reproduction: cultivated populations are propagated exclusively vegetatively from large cuttings (the physical form of the tree trunk often reflects this method of propagation); wild populations grow from seeds and have obvious age-structured populations, and (3) fruit morphology: cultivated fruits are much larger and sweeter than wild fruits and have a wide range of colors; wild fruits have very little "meat" (fleshy mesocarp) relative to cultivated fruits they taste very acidic, and are usually red or yellow in color. Herbarium specimens were collected for 105 of the 114 populations and were deposited at the Missouri Botanical Garden (St. Louis, Missouri, USA) and in regional herbaria. Collection numbers 2005-1 through 2005-9 were vouchered digitally and are available upon request from the authors.
|
|
), and geographic information system (GIS) environmental data layers, we predicted the geographic distribution of wild and cultivated populations of S. purpurea. Maxent is a general-purpose machine-learning approach to the modeling of species distributions using presence-only data (Phillips et al., 2006
). It predicts the potential distribution of a species by estimating the probability distribution of maximum entropy across a specified region, subject to a set of constraints that represent the missing information (lack of absence data) about the target distribution (Phillips et al., 2006
). Like other niche-based models constructed from presence-only data, the predicted distribution describes suitability in ecological (environmental and climatological) space, which is then projected onto geographic space revealing a prediction of the geographic distribution of the taxon of interest (Phillips et al., 2006
). For our analyses, we used 18 GIS data layers from the WorldClim Global Climate GIS database (30-s resolution) (Hijmans et al., 2004
, 2005
). These included topographic and bioclimatic variables representing elevation (m), annual mean temperature (°C), mean diurnal temperature range (°C), isothermality, temperature seasonality (°C), maximum temperature of warmest month (°C), minimum temperature of coldest month (°C), temperature annual range (°C), mean temperature of wettest quarter (°C), mean temperature of driest quarter (°C), mean temperature of warmest quarter (°C), mean temperature of coldest quarter (°C), annual precipitation (mm), precipitation seasonality (mm), precipitation of wettest quarter (mm), precipitation of driest quarter (mm), precipitation of warmest quarter (mm), and precipitation of coldest quarter (mm).
Species distributions were predicted for both wild and cultivated populations using the locality data in Table 1. For each group (wild or cultivated), the Maxent algorithm was run using the default parameters including a maximum of 500 iterations with a convergence threshold of 0.00001. During model development, 50% of the localities were used for model training, while 50% of the localities were held back to test model accuracy. Cumulative probability distributions ranging from 0 to 100 were generated for both wild and cultivated populations that represent a relative measure of the probability of occurrence for the modeled group. A binomial probability distribution was applied to the localities that were held back for model testing to assess the accuracy of each predicted distribution (Phillips et al., 2006
).
To assess the interpredictivity of the cultivated and wild model predictions, we used a binomial probability distribution to determine whether the number of times that the occurrence data points from wild populations overlapped a threshold-based predicted distribution of cultivated populations was different than random and vice versa (Peterson et al., 1999
; Rice et al., 2003
; Knouft et al., 2006
). The distribution threshold was set at the minimum probability area containing all of the training localities (Phillips et al., 2006
). Although Maxent produces a modeled species distribution with relative probabilities of occurrence between 0 and 100, the minimum probability threshold allows for identification of a standardized percentage of the distribution area, thus allowing for comparisons of different models (e.g., wild and cultivated) (Phillips et al., 2006
). The interpredictivity analysis tests whether the percentage of actual occurrence data points for cultivated populations that falls within the modeled distribution of the wild populations corresponds to the proportion of land area in Mexico and Central America that is covered by the wild S. purpurea predicted distribution. Greater than expected overlap is consistent with the idea that domestication has not been accompanied by a significant shift or expansion in the ecological factors characterizing the distribution of cultivated S. purpurea populations; alternatively, less than expected overlap or overlap that does not deviate from a random frequency may indicate that the distribution of cultivated S. purpurea populations has shifted and/or expanded relative to the distribution of the wild ancestors.
As an additional test of the reciprocal quality of each prediction, the interpredictive effectiveness of each model was directly compared. For this analysis, values from the cumulative probability distribution for cultivated populations were extracted at actual wild localities. Similarly, values from the cumulative probability distribution for wild populations were extracted at actual cultivated localities. The extracted values were compared using a Mann–Whitney U test.
In addition to investigating distribution similarity using species locality data and the predictions generated by the Maxent algorithms, we qualitatively examined the overlap of the "environmental envelopes" of wild and cultivated populations using GIS-derived topographic and environmental data extracted from localities for each group (wild and cultivated). We generated the environmental envelope for each group based on data extracted from the 18 WorldClim Global Climate GIS variables used in the Maxent analyses (30-s resolution) (Hijmans et al., 2004
, 2005
). Environmental data for each group were compiled by importing population locality points (Table 1) into DIVA-GIS (Hijmans et al., 2001
). Environmental data were then extracted from each GIS layer to provide 18 topographic and climatic measures for each locality point. All topographic and climatic data were log10-transformed to standardize data for statistical analyses. A principal components analysis (PCA) was performed on the correlation matrix of transformed data to generate data needed to construct an environmental envelope based on information from the wild as well as the cultivated data sets. To generate and compare the environmental envelope of each group, principal component scores from the first three axes of the PCA were plotted in x, y space for the wild and cultivated populations (similar to Knouft et al., 2006
).
Comparisons of wild and cultivated population environmental variables
Using DIVA-GIS we compiled topographic and environmental data for wild and cultivated populations from GIS data sets based on the localities in Table 1. To avoid redundancy among variables (e.g., mean temperature warmest quarter, mean temperature warmest month), we selected 10 layers (Tables 2, 3) from the 30-s resolution 30-yr WorldClim data sets (Hijmans et al., 2004
) to use in the comparison. Each topographic and environmental measure for wild and cultivated populations was compared using a Mann–Whitney U test. The variances of environmental variables were compared between wild and cultivated populations by calculating an F statistic to determine if these two groups occurred in regions with differing ranges of environmental characteristics. Because multiple tests were performed for each set of analyses, we applied a sequential Bonferroni correction to our tests (
= 0.05) (Holm, 1979
).
|
|
RESULTS
Species distribution modeling
Based on known occurrences of cultivated S. purpurea populations and their wild progenitors, we generated distribution maps predicting the possible areas where cultivated and wild S. purpurea populations might occur (Fig. 2a, b). Predictions for both wild and cultivated populations were highly significant based on a binomial probability distribution test calculated from the held-back test localities (Wild AUC of ROC: training data = 0.975, test data = 0.914, P < 0.0001; Cultivated AUC of ROC: training data = 0.929, test data = 0.889, P < 0.0001).
|
The first three principal components explained 81.36% of the overall variance in the data (PC1 = 38.14%, PC2 = 29.08%, PC3 = 14.14%; Appendix 1). Comparisons of the principal component scores between wild and cultivated populations in two-dimensional space indicates that the environmental envelope of wild populations is nested within the environmental envelope of cultivated populations in all cases (Fig. 3).
|
DISCUSSION
During the domestication of S. purpurea, humans preferentially cultivated trees with an abundance of large, juicy, sweet fruits resulting in increased variation in the color, size, and taste of S. purpurea fruits in cultivated populations (Miller, 2004
). Further, domestication of S. purpurea resulted in reduced levels of genetic variation in cultivated S. purpurea populations as compared with their wild progenitors (Miller and Schaal, 2006
). Here, we have identified another fundamental difference between wild and cultivated S. purpurea populations. Cultivated populations occupy an expanded geographic distribution relative to their wild progenitors, and there are measurable differences in the environmental factors that characterize the distributions of cultivated and wild S. purpurea populations.
Species distribution models and the evolution of the "ecological niche" in a domesticated species
Species distribution models based on environmental and climatological factors have been referred to as a representation of the "ecological niche" of a species, or the range of biotic and abiotic characteristics in which a species is able to persist (Peterson, 2003
and references therein; Martínez-Meyer et al., 2004a
); however, there is debate in the literature about what exactly the modeled ecological niche represents. Some authors assume that the ecological niche model represents the fundamental ecological niche, which is the range of all theoretical possibilities where a given species could live, defined in coarse-scale climatic dimensions (the "bioclimatic envelope" or the "climatic niche") (Pearson and Dawson, 2003
; Soberón and Peterson, 2005
). The assumption is that by examining species across their entire geographic distributions, a view of the fundamental ecological niche can be assembled (Peterson et al., 1999
; Peterson, 2001
; Wiens and Graham, 2005
). Others, however, have suggested that the ecological niche, which is modeled from known localities, represents an approximation of the species' realized niche, the subset of the fundamental niche that it actually occupies in the study area and environmental dimensions being considered (Phillips et al., 2006
). In this study, distribution models were produced based on known localities of extant S. purpurea populations in Mesoamerica; therefore, in this discussion we assume that the distribution models produced for cultivated and wild S. purpurea populations approximate the realized ecological niches of the two groups examined in this region.
The constancy of ecological niches within evolutionary lineages is an important topic in evolutionary ecology: the use of ecological niche models to predict unsampled localities, areas of potential range expansion, and future distributions based on global climate change models depends fundamentally upon the assumption that niches are relatively stable over time (e.g., Peterson et al., 1999
; Peterson, 2001
, 2003
; Peterson and Holt, 2003
; Illoldi-Rangel et al., 2004
, Mau-Crimmins et al., 2006
). Results from a variety of taxa provide evidence for phylogenetic niche conservation (the tendency of species to retain similar ecological niches over evolutionary time scales) at the interspecific level and above (Peterson et al., 1999
; Martínez-Meyer et al., 2004a
, b
; Wiens, 2004
). Studies of intraspecific changes in the ecological niche, however, are relatively rare, with previous investigations focusing on insular passerine birds (Scott et al., 2003
), monarch butterflies (Oberhauser and Peterson, 2003
), diurnal raptors (Galeotti and Rubolini, 2004
), and Mexican birds (Peterson and Holt, 2003
). Domesticated species present an excellent opportunity to investigate intraspecific niche differentiation under intense artificial selection that takes place on a relatively short time scale (<10 000 years), facilitating an understanding of some potential impacts of humans on species distributions.
Distribution models constructed here indicate that the ecological niche of S. purpurea populations has been conserved during the evolution of the S. purpurea lineage. Interpredictivity analyses reveal that 100% of the actual sampled localities of the wild ancestors are contained within the predicted distribution of the cultivated S. purpurea populations. Of broader relevance to crop biologists is our finding that the predicted geographic distribution of the wild S. purpurea populations is not a good indicator of sites of cultivated S. purpurea populations (the predicted distribution of the wild S. purpurea populations encompassed just 34% of actual sampled localities of cultivated S. purpurea). Rather, it is the predicted distribution of cultivated S. purpurea populations that functions as a good indicator of the locations of wild populations. PCA analyses provide further support for conservation of the niche of wild populations within the niche of the cultivated populations: the portion of the PCA space occupied by wild localities is contained within the portion of PCA space occupied by the cultivated populations (Fig. 3). During the domestication of S. purpurea, the ecological and environmental characteristics of the regions occupied by wild populations have been retained and represent a subset of the regions occupied by cultivated populations.
In addition, data presented here indicate that the niche of cultivated S. purpurea has expanded significantly relative to the niche of the wild populations during the course of domestication. Approximately 66% of the cultivated populations fall outside of the predicted distribution of wild S. purpurea populations. Further, in the PCA analyses the cultivated populations occupy a much broader portion of the PCA space than the wild populations (Fig. 3). Finally, cultivated populations have significantly higher variances in five environmental variables (Table 3). Humans have facilitated the expansion of cultivated S. purpurea populations into regions where, in nature, wild S. purpurea populations are not found.
The impact of cultivation on specific aspects of habitat occupied by S. purpurea
In addition to providing evidence for differences in the geographic distributions of cultivated and wild S. purpurea populations, our data reveal that cultivated and wild S. purpurea populations do not occur in exactly the same varieties of habitats. We have identified specific climatological factors that differ in mean and variance between cultivated and wild populations (Table 2). The geographic regions occupied by cultivated populations are wetter throughout the year and less seasonal than the geographic regions occupied by wild S. purpurea populations. Wild S. purpurea populations are found in the Mesoamerican dry forests, which have a wide temperature range and marked seasonality characterized by distinct wet and dry seasons (Murphy and Lugo, 1986
). Cultivated populations are found in these areas as well as regions with less pronounced seasonality and more rainfall. In previous studies, researchers have documented vegetation changes in the tropical dry forests following intensive anthropogenic disturbances (Burgos and Maas, 2004
) and have tracked succession history following agriculture and grazing on lands previously occupied by tropical dry forests (Ruiz et al., 2005
). There are, however, no known studies documenting the expansion of a dry forest species into other habitats. In the case of S. purpurea, selection during the domestication process produced a measurable change in habitat in this dry forest native.
Differences in the environmental and climatological factors characterizing the geographic distributions of cultivated and wild populations could be the result of selection during the domestication process for trees that can survive in a wide variety of habitats. The relatively expanded geographic distribution of cultivated populations could reflect the various contributions of humans toward the survival of trees in agricultural habitats, including the facilitation of transport and reproduction, elimination of competition, and supply of water and additional resources. Alternatively, the relatively limited distribution of wild S. purpurea populations in nature could be the product of competition and prehuman biogeographical history. One of the reasons S. purpurea was chosen for this study is that the majority of cultivated S. purpurea populations have yet to undergo the intensive selection, breeding, and care associated with modern agriculture. None of the cultivated populations included in this study were fertilized, watered, or protected to increase the ability of individual trees to survive in a particular region. The only form of deliberate care we observed was the clearing of debris from underneath some cultivated trees. Therefore, it is our interpretation that differences observed in the environmental and climatological characteristics of S. purpurea populations reflect real differences in their distributions. These data set up the testable hypothesis that these differences reflect artificial selection during domestication; however, reciprocal transplant experiments are required to determine if there is a heritable basis for the habitat differences.
Implications of ecological and climatological data for cultivation and conservation of S. purpurea
Spondias purpurea produces juicy, plumlike fruits that are high in vitamin C; it has been identified as a very promising tree crop because it is highly drought-resistant and it grows on poor soil (Cuevas, 1994
). Our data corroborate observations that cultivated S. purpurea populations can (and do) grow in a wide range of habitats, highlighting its importance as a regional cash crop.
In addition, these data reveal that the distribution of wild S. purpurea populations is remarkably narrow relative to the cultivated descendents. The native habitat of the wild progenitors of cultivated S. purpurea, the Mesoamerican dry forests, is characterized by several months of severe drought (Mooney et al., 1995
; Trejo and Dirzo, 2002
); the floristic composition includes primarily small deciduous trees, lianas, and shrubs (Trejo and Dirzo, 2002
). It has been estimated that less than 2% of the tropical dry forests remain (Janzen, 1988
). The results of this study emphasize the uniqueness of wild S. purpurea populations and the habitats in which they evolved, and underscore the importance of their conservation.
|
1 The authors thank three anonymous reviewers and members of the Missouri Botanical Garden Analysis Unit for helpful comments on an earlier version of the manuscript and R. Aguilar F., A. Anzueto, G. Borjas, G. Carnevali, J. Castilla Canales, L. Chavez V., I. Diaz, A.-C. Gomez, A. Herrera, M. Diaz, A. and C. MacVean, M. Merello, A. Molina R., P. Moreño, A. Muehlenbachs, M. Olson, E. O'Mahoney-Cubbison, A. Paschke, E. Pimienta Barrios, B. Ramirez Hernandez, R. Rueda, R. Ruenes, N. Ventura, and B. Wong for assistance in the field. Funding was provided by the National Science Foundation (PD 0105134), Botanical Society of America Karling Graduate Award, Organization for Tropical Studies, Washington University Division of Biology and Biomedical Sciences, Mellon Foundation Grant to the Missouri Botanical Garden, and the University of Colorado Museum. 
2 Author for correspondence (e-mail: amille75{at}slu.edu
)
101 5 Present address: Department of Biology, Saint Louis University, 3507 Laclede Avenue, St. Louis, MO 63103-2010 USA
LITERATURE CITED
Anderson R. P. Martínez-Meyer E.. 2004. Modeling species' geographic distributions for preliminary conservation assessments: an implementation with the spiny pocket mice (Heteromys) of Ecuador. Biological Conservation 116: 167-179.[CrossRef][ISI]
Anderson R. P. Peterson A. T. Gomez-LaVerde M.. 2002. Using niche-based GIS modeling to test geographic predictions of competitive exclusion and competitive release in South American pocket mice. Oikos 98: 3-16.[CrossRef][ISI]
Anthony F. Combes M. C. Astorga C. Bertrand B. Graziosi G. Lashermes P.. 2002. The origin of cultivated Coffea arabica L. varieties revealed by AFLP and SSR markers. Theoretical and Applied Genetics 104: 894-900.[CrossRef][ISI][Medline]
Avitia-Garcia E.. 1997. Estructura floral y anatomía del aborto de ovulos y semillas en ciruela mexicana (Spondias purpurea L). Horticultura Mexicana 5: 282-288.
Baraona Cockrell M.. 2000. Jocote, anona, y cas. Tres frutas campesinas de America Editorial Universidad Nacional Heredia, San Jose, Costa Rica.
Bernardi M.. 2001. Linkages between FAO agroclimatic data resources and the development of GIS models for control of vector-borne diseases. Acta Tropical 79: 21-34.[CrossRef]
Burgos A. Maass J. M.. 2004. Vegetation change associated with land-use in tropical dry forest areas of western Mexico. Agriculture, Ecosystems, and Environment 104: 475-481.[CrossRef]
Caldiz D. O. Haverkort A. J. Struik P. C.. 2002. Analysis of a complex crop production system in interdependent agro-ecological zones: a methodological approach for potatoes in Argentina. Agricultural Systems 73: 297-311.[CrossRef][ISI]
Clegg M. T. Brown A. H. D. Whitfield P. R.. 1984. Chloroplast DNA diversity in wild and cultivated barley: implications for genetic conservation. Genetical Research 43: 339-343.[ISI]
Cuevas J. A.. 1994. Spanish plum, red mombin (Spondias purpurea). In J. E. Hernándo Bermejo, J. León [eds.], Neglected crops: 1492 from a different perspective, 111–115. Plant Production and Protection, series no. 26 Food and Agriculture Organization, Rome, Italy.
Elith J. Graham C. H. Anderson R. P. Dudík M. Ferrier S. Guisand A. Hijmans R. J. Huettmann F. Leathwick J. R. Lehmann A. Li J. Lohmann L. G. Loiselle B. A. Manion G. Mortiz C. Nakamura M. Nakazawa Y. Overton J. McC. Peterson A. T. Phillips S. J. Richardson K. Schachetti-Pereira R. Schapire R. E. Soberón J. Williams S. Wisz M. S. Zimmerman N. E.. 2006. Novel methods improve prediction of species' distributions from occurrence data. Ecography 29: 129-151.[CrossRef][ISI]
Ennos R. A.. 1997. The influence of agriculture on genetic biodiversity. Biodiversity and Conservation in Agriculture, British Crop Protection Council Symposium Proceedings no. 69 British Crop Protection Council, Surrey, UK.
Eyre-Walker A. Gaut R. L. Hilton H. Feldman D. L. Gaut B. S.. 1998. Investigation of the bottleneck leading to the domestication of maize. Proceedings of the National Academy of Sciences, USA 95: 4441-4446.
Galeotti P. Rubolini D.. 2004. The niche variation hypothesis and the evolution of colour polymorphism in birds: a comparative study of owls, nightjars, and raptors. Biological Journal of the Linnean Society 82: 237-248.[CrossRef]
Gepts P. Clegg M. T.. 1989. Genetic diversity in pearl millet (Pennisetum glaucum [L.] R. Br.) at the DNA sequence level. Heredity 80: 203-208.
Graham C. H. Ron S. R. Santos J. C. Schneider C. J. Moritz C.. 2004. Integrating phylogenetics and environmental niche models to explore speciation mechanisms in dendrobatid frogs. Evolution 58: 1781-1793.[ISI][Medline]
Greene S. L. Hart T. C. Afonin A.. 1999a. Using geographic information to acquire wild crop germplasm for ex situ collections. I. Map development and field use. Crop Science 39: 836-842.
Greene S. L. Hart T. C. Afonin A.. 1999b. Using geographic information to acquire wild crop germplasm for ex situ collections. II. Post-collection analysis. Crop Science 39: 843-849.
Hancock J. F.. 2004. Plant evolution and the origin of crop species CABI Publishing, Cambridge, UK.
Hijmans R. J. Cameron S. E. Parra J. L. Jones P. G. Jarvis A.. 2004. The WorldClim interpolated global terrestrial climate surfaces, version 1.3 Computer program available at website http://biogeo.berkeley.edu/ [accessed April, 2006].
Hijmans R. J. Cameron S. E. Parra J. L. Jones P. G. Jarvis A.. 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25: 1965-1978.[CrossRef][ISI]
Hijmans R. J. Guarino L. Cruz M. Rojas E.. 2001. Computer tools for spatial analysis of genetic resources data: 1. DIVA-GIS. Plant Genetic Resources Newsletter 127: 15-19.
Hijmans R. J. Spooner D. M.. 2001. Geographic distribution of wild potato species. American Journal of Botany 88: 2101-2112.
Hoffman M. H.. 2005. Evolution of the realized climatic niche in the genus Arabidopsis (Brassicaceae). Evolution 59: 145-1436.
Holm S.. 1979. A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics 6: 65-70.[ISI]
Illoldi-Rangel P. Sánchez-Cordero V. Peterson A. T.. 2004. Predicting distributions of Mexican mammals using ecological niche modeling. Journal of Mammology 85: 658-662.[CrossRef]
Janzen D. H.. 1988. Tropical dry forests. The most endangered major tropical ecosystem. In E. O. Wilson [ed.], Biodiversity National Academy Press, Washington, D.C., USA.
Jarvis A. Ferguson M. E. Williams D. E. Guarino L. Jones P. G. Stalker H. T. Valls J. F. M. Pittman R. N. Simpson C. E. Bramel P.. 2004. Biogeography of wild Arachis: assessing conservation status and setting future priorities. Crop Science 43: 1100-1108.[ISI]
Jeutong F. Eskridge K. M. Waltman W. J. Smith O. S.. 2000. Comparison of bioclimatic indices for prediction of maize yields. Crop Science 40: 1612-1617.
Knouft J. H. Losos J. B. Glor R. E. Kolbe J. J.. 2006. Phylogenetic analysis of the evolution of the niche in lizards of the Anolis sagrei group. Ecology 87: S29-S38.[CrossRef][ISI][Medline]
Mandujano S. Gallina D. Bullock S. H.. 1994. Frugivory and dispersal of Spondias purpurea (Anacardiaceae) in tropical deciduous forest in Mexico. Revista de Biología Tropical 42: 107-114.
Martínez-Meyer E. Peterson A. T. Hargrove W. W.. 2004a. Ecological niche as stable distributional constraints on mammal species, with implications for Pleistocene extinctions and climate change projections for biodiversity. Global Ecology and Biogeography 13: 305-314.[CrossRef][ISI]
Martínez-Meyer E. Peterson A. T. Navarro-Sigüenza A. G.. 2004b. Evolution of seasonal ecological niches in the Passerina buntings (Aves: Cardinalidae). Proceedings of the Royal Society of London, series B Biological Sciences 271: 1151-1157.[Medline]
Mau-Crimmins T. M. Schussman H. R. Geiger E. L.. 2006. Can the invaded range of a species be predicted sufficiently using only native-range data? Lehmann lovegrass (Eragrostis lehmanniana) in the southwestern United States. Ecological Modelling 193: 736-746.[CrossRef]
Miller A. J.. 2004. Origin and domestication of a Mesoamerican fruit tree, Spondias purpurea L. (Anacardiaceae) Ph.D. dissertation, Washington University, St. Louis, Missouri, USA.
Miller A. J. Schaal B. A.. 2005. Domestication of a Mesoamerican cultivated fruit tree. Proceedings of the National Academy of Sciences, USA 102: 12801-1206.
Miller A. J. Schaal B. A.. 2006. Domestication and the distribution of genetic variation in wild and cultivated populations of the Mesoamerican fruit tree, Spondias purpurea L. (Anacardiaceae). Molecular Ecology 15: 1467-1480.[CrossRef][Medline]
Mooney H. A. Bullock S. H. Medina E.. 1995. Introduction. In S. H. Bullock, H. A. Mooney, E. Medina [eds.], Seasonally dry tropical forests Cambridge University Press, Cambridge, UK.
Murphy P. G. Lugo A. E.. 1986. Ecology of tropical dry forest. Annual Review of Ecology and Systematics 17: 67-88.[CrossRef][ISI]
Oberhauser K. Peterson A. T.. 2003. Modeling current and future potential wintering distributions of eastern North American monarch butterflies. Proceedings of the National Academy of Sciences, USA 100: 14063-14068.
Pearson R. G. Dawson T. P.. 2003. Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful?. Global Ecology and Biogeography 12: 361-371.[CrossRef][ISI]
Peterson A. T.. 2001. Predicting species' geographic distributions based on ecological niche modeling. Condor 103: 599-605.[CrossRef][ISI]
Peterson A. T.. 2003. Predicting the geography of species' invasions via ecological niche modeling. Quarterly Review of Biology 787: 419-432.
Peterson A. T. Holt R. D.. 2003. Niche differentiation in Mexican birds: using point occurrences to detect ecological innovation. Ecology Letters 6: 774-782.[CrossRef][ISI]
Peterson A. T. Ortega-Huerta M. A. Bartley J. Sanchez-Cordero V. Soberón J. Buddemeir R. H. Stockwell D. R. B.. 2002. Future projections for Mexican faunas under global climate change scenarios. Nature 416: 626-629.[CrossRef][Medline]
Peterson A. T. Soberón J. Sanchez-Cordero V.. 1999. Conservatism of ecological niches in evolutionary time. Science 285: 1265-1267.
Phillips S. J. Anderson R. P. Schapire R. E.. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling 190: 231-259.[CrossRef]
Rice N. H. Martínez-Meyer E. Peterson A. T.. 2003. Ecological niche differentiation in the Aphelocoma jays: a phylogenetic perspective. Biological Journal of the Linnean Society 80: 369-383.[CrossRef]
Ruiz J. Fandiño M. C. Chazdon R. L.. 2005. Vegetation structure, composition, and species richness across a 56-year chronosequence of dry tropical forest on Providencia Island, Colombia. Biotropica 37: 520-530.[CrossRef][ISI]
Saunders J. A. Pedroni M. J. Penrose L. D. J. Fish A. J.. 2001. AFLP analysis of opium poppy. Crop Science 41: 1596-1601.
Scott S. N. Clegg S. M. Blomberg S. P. Kikkawa J. Owens I. P. F.. 2003. Morphological shifts in island-dwelling birds: the roles of generalist foraging and niche expansion. Evolution 57: 2147-2156.[ISI][Medline]
Simpson B. B. Ogorzaly M. C.. 1995. Economic botany. Plants in our world, 3rd ed McGraw-Hill Science/Engineering/Math, Columbus, Ohio, USA.
Siqueira M. F. D. Peterson A. T.. 2003. Consequences of global climate change for geographic distributions of cerrado tree species. Biota Neotropica 3: 1-14.
Soberón J. Peterson A. T.. 2004. Biodiversity informatics: managing and applying primary biodiversity data. Philosophical Transactions of the Royal Society of London B, 359: 689-698.[CrossRef]
Soberón J. Peterson A. T.. 2005. Interpretation of models of fundamental ecological niches and species' distributional areas. Biodiversity Informatics 2: 1-10.
Thomas C. D. Cameron A. Green R. E. Bakkenes M. Beaumont L. J. Collinham Y. C. Erasmus B. F. N. Ferreira De Siqueria M. Grainger A. Hannah L. Hughes L. Huntley B. Van Jaarsveld A. S. Midgley G. E. Miles L. Ortego-Huerta M. A. Peterson A. T. Phillips O. L. Williams S. E.. 2004. Extinction risk from climate change. Nature 427: 145-148.[CrossRef][Medline]
Trejo I. Dirzo R.. 2002. Floristic diversity of Mexican seasonally dry tropical forests. Biodiversity and Conservation 11: 2063-2048.[CrossRef][ISI]
Wiens J. J.. 2004. Speciation and ecology revisited: phylogenetic niche conservatism and the origin of species. Evolution 58: 193-197.[ISI][Medline]
Wiens J. J. Graham C. H.. 2005. Niche conservatism: integrating evolution, ecology, and conservation biology. Annual Review of Ecology, Evolution, and Systematics 36: 519-539.[CrossRef][ISI]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
