| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
What's this? |
Systematics and Phytogeography |
2International Potato Center, Apartado 1558, Lima 12, Peru; and 3USDA, Agricultural Research Service, Department of Horticulture, University of Wisconsin, 1575 Linden Drive, Madison, Wisconsin 53706-1590 USA
Received for publication December 12, 2000. Accepted for publication April 19, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: GIS • geographic distribution • geographic information systems • potato • section Petota • Solanum • Solanaceae • species richness
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Contreras and Spooner, 1999
). The cultivated potato (Solanum tuberosum L. and six other cultivated species that are only grown in the Andes) is one of the world's principal food crops (Walker, Schmiediche, and Hijmans, 1999
).
In addition to these seven cultivated potato species, there are 199 wild potato species (Spooner and Hijmans, in press
). These wild species all grow in the Americas, from the southwestern United States to central Argentina and Chile. They have been collected intensively to contribute to potato genebanks, and they have been used in breeding programs to improve the cultivated potato (Ross, 1986
; Hawkes, 1990
). There is no evidence that wild potatoes are currently being domesticated or traded or that their geographic distribution is deliberately altered by man, except for the case of S. chacoense, a species that is grown as an ornamental plant in Lima, Peru (Ochoa, 1962
). However, in some parts of the Andes there is ongoing gene flow between wild and cultivated potato species (Hawkes, 1979
; Watanabe and Peloquin, 1989
; Rabinowitz et al., 1990
).
Despite a purported great variability in wild potatoes (Hawkes, 1990
), many of the species have a general appearance similar to the cultivated potato (Spooner and Van den Berg, 1992
). A typical wild potato has pinnately dissected leaves (a few species have entire leaves), corollas in various shades of purple (colors vary from white to blue to purple and pink), corollas pentagonal to rotate (some are stellate), and fruits spherical to ovoid (some are conical). This similarity has led to widely conflicting taxonomic treatments of wild potatoes. Hawkes (1990)
and Spooner and Hijmans (2001)
provide the most recent taxonomic overviews of the entire group.
Wild potatoes have been the object of intensive study. Spooner and Hijmans (2001)
recently reviewed wild potato systematics and germplasm collection. Jansky (2000)
reviewed the value of wild and cultivated potatoes in breeding for disease resistance. However, there has been no comprehensive analysis of the geographic distribution of wild potatoes, which we provide here for the first time. We used geographic information systems (GIS) to analyze a large georeferenced database of locations where wild potatoes were observed.
We computed country- and species-level statistics. For each species, we estimated the geographic area over which it occurs and mapped the number of observations and species richness, using grid cells. We determined the minimum number of grid cells needed to include all species and related species richness to latitude and altitude. Species richness is used because it is a simple, widely used, well-understood, and useful measure of taxonomic diversity (Gaston, 1996
) and because it is less sensitive than diversity indices to the problems of unsystematic sampling intensities and procedures (Hijmans et al., 2000)
.
This study explores the use of geographic information systems (GIS) to describe the geographic distribution of wild crop relatives. This type of study can provide baseline data for further GIS analysis for exploration, conservation, and use of germplasm of wild crop relatives (Guarino et al., in press
), as well as for studies of the factors that explain the geographic distribution of these species.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). (2) Data from 16 collecting expeditions in 12 countries by D. M. Spooner and coworkers (Spooner and Hijmans, 2001
). These included records that are not in genebanks because they were collected as herbarium species or were lost as living specimens after collection. (3) A database of herbarium records developed by J. G. Hawkes (Hawkes, 1997
). (4) Hawkes and Hjerting (1969
; Argentina, Brazil, Paraguay and Uruguay). (5) Hawkes and Hjerting (1989
; Bolivia). (6) Ochoa (1990
; Bolivia). (7) Ochoa (1999
, Peru). (8) Spooner et al. (1998
; Guatemala). (9) Spooner et al. (1999
; Peru). (10) Spooner, Hoekstra, and Vilchez (2001
; Costa Rica). (11) Spooner et al. (in press
; Mexico).
We used all records from sources 1–3 that included a species name and passport data. Passport data include a description of the location of origin (such as locality name), administrative units (such as departments and districts), and geographic coordinates. However, coordinate data were absent for many records. For the genebank databases, coordinates were assigned using the locality description where possible. For Hawkes' (1997)
database this was only attempted for species for which we had fewer than five observations with coordinate data. Sources 5–7 were used to verify and improve the geographic coverage of our species distribution data (see below). Additional herbarium records were taken from sources 8–11.
The presence of coordinate data allowed the analysis of the database with GIS; we used ArcView-GIS (Environmental Systems Research Institute, 1999
) and DIVA-GIS software (Hijmans et al., 2001)
. Coordinate data in genebank databases often lack precision and were checked and modified following procedures described by Hijmans et al. (1999)
. First, we checked for gross errors, such as accessions located in the oceans. Then, we made overlays (simultaneous spatial queries) of the collection sites and administrative boundary databases (first level subdivision for Mexico and Central America; first and second level for the United States and South America; and first, second, and third level subdivision for Peru). We compared the names of the administrative units according to the wild potato distribution database with those of the administrative boundary database. In case of discrepancies between the two databases, the coordinates were checked against the locality description and new coordinates were assigned where needed.
Dot maps of the distribution of all species were compared with published species distribution maps from the floristic sources 5–7. When general areas of occurrence were already represented on our maps, we did not include additional points, because of a possible lack of precision of many of these maps, and the risk of duplicating records. However, if it appeared that our distribution maps did not include all major areas where a species was reported to occur, we did copy additional observations for these areas. Species names follow Spooner and Hijmans (in press)
, who list 196 wild species in sect. Petota and three outgroup species in sect. Etuberosum. Taxonomic groups below the species level (subspecies, varieties, and forms) are all treated within their component species.
Country- and species-level distribution
The number of observations and species in the database were tabulated by country. This was done separately for rare species, here defined as species for which we had fewer than five observations. The number of observations per species was calculated and plotted. The average number of observations per species was calculated to assess intensity of collection by country, given the species richness it harbors.
Area of distribution
For each species we estimated the area over which it occurs, using two statistics: (1) Maximum distance (MaxD) between two observations of a single species was calculated as the largest distance (in meters) between all possible pairs of observations of one species. (2) We assigned a circular area (CAr) with a radius r to each observation and calculated the total area of all circles per species. Areas where circles of a species overlap are only included once. Area is expressed as the area relative to the area of one circle, i.e., the number of circular areas covered. We decided to use a radius of 50 km (i.e., CA50).
The assumption is that each point observation represents a group of plants that covers a circular area with a 50-km radius. Expressing CAr as the number of circles instead of the absolute area makes it more easily comparable across different studies and scales (when a radius other than 50 km is chosen).
The CA50 statistic was plotted against the number of observations to explore differences in abundance between species. A species with a relatively high number of observations per CA50 would be abundant within its area of distribution, whereas a low number would indicate that a species was more scattered over the range in which it occurs.
To describe species distributions we use the terms "endemic" and "rare." We use "endemic" for species that occur in relatively small areas (have a small range size) (Rabinowitz, 1981
; Gaston and Williams, 1996
) and "rare" for species that have been observed in relatively few cases.
Grid based distribution
We compared the number of observations and species using a grid with 50 x 50 km cells and summarized the results by country. We used ArcView-GIS to transform the coordinate data to the Lambert equal-area azimuthal projection, with 80° W as the central meridian and the equator as the reference latitude (Environmental Systems Research Institute, 1999
). Because the origin of a grid is arbitrary but can influence the results, it may not be accurate to assign a point to one cell only, if the point is located near one or three other cells. Therefore, the data were assigned to grid cells using a circular neighborhood (Bonham-Carter, 1994
; Cressie, 1991
) with a radius of 50 km, using the DIVA-GIS software. All the observations within that neighborhood were assigned to its respective grid cell, and an observation can, therefore, be assigned more than once. The result is a smoother grid, which is less biased by the origin of the grid and also less sensitive to small changes (errors) in the coordinate data. When we discuss grid cells in this paper, these refer to circles with an area of 
r2 = 7854 km2 with their center in the middle of the grid cells with an area of 2500 km2.
To assess the distribution of species over grid cells, we plotted the number of species per grid cell against the number of grid cells. Data on plant distributions can be biased due to spatial differences in recorder (a person who takes data) effort (Rich and Woodruff, 1992
; Gaston, 1996
; Hijmans et al., 2000
). In extreme cases, differences in number of species between areas would reflect the amount of time spent there by recorders and not reflect actual differences in distribution. We assessed the extent to which the number of observations predicts the number of species by plotting the number of species vs. the number of observations per grid cell.
Complementarity analysis
To further analyze aspects of species distribution and endemicity we identified the smallest area (number of grid cells) needed to capture all wild potato species. This type of complementarity analysis is typically used in studies for optimal reserve selection (Csuti et al., 1997
). We used the algorithm described by Rebelo (1994
; see also Rebelo and Siegfried, 1992
) and implemented in the DIVA-GIS software that selects grid cells so as to identify the minimum set of cells that captures a maximum amount of species. The algorithm selects the cell with most species in it and then, step by step, selects cells that contain the highest number of additional (not previously included) species. In the case of cells having the same number of additional species, a random cell is selected from such cells. Selecting these complementary cells is a nonlinear optimization problem for which Rebelo's (1994)
algorithm finds a near-optimal solution. We determined the minimum number of grid cells needed to include all species and mapped the location of these grid cells.
Distribution by latitude and altitude
To summarize the species distribution data, the number of species was tabulated by latitude and altitude. First, the number of species that occur in strips of 1° latitude was determined. Then, to obtain a smoothed line, for each 1° latitude zone the moving average was calculated, using five adjacent zones (two at each side). The GTOPO30, a 30'' grid (each cell is 
0.8 km2) with altitude data (United States Geological Survey, 1998
) was used to estimate altitude for all wild potato localities. This estimate was only used for the records for which the passport data did not include altitude. Observations were then grouped in classes of 250 m altitude and the number of species per class was plotted. The five-observation moving average was calculated and plotted.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
and for the interpotato genebank by Huamán, Hoekstra, and Bamberg (2001)
.
|
Area of distribution
Most species are narrow endemics and only 39 species occur in two or more countries. Most of these country co-occurrences are from Bolivia and Argentina (13 species in common) and from Bolivia and Peru (10 species in common). Solanum chacoense is the only species that occurs in five countries, followed by S. acaule and S. commersonii which occur in four countries. The average greatest distance between two observations of the same species (MaxD) is 411 km. For 77 species, MaxD is <100 km, and for 100 species (50% of the total), MaxD is <200 km (Fig. 2). The greatest MaxD observed was for S. acaule. At the time of this research, this number had recently increased by 732 km to a total of 3253 km, due to the recent discovery of this species in Ecuador (Spooner, Castillo T., and López J. [1992
] who identified it as S. albicans; but Kardolus [1998
] recognized it as an anomalous new hexaploid variety of S. acaule). Average circular area (CA50) over all species is 6.7, but its distribution is strongly skewed. Seventy-two species have a CA50 of <2, and 124 have a CA50 of <5 (Fig. 2).
|
|
|
|
|
|
There are few cells with many species (Fig. 6). Cells with more than 15 species are only found in Peru, Bolivia, and Argentina; Ecuador and Mexico are the only two additional countries that have cells with nine or more species (Table 2). Only 5% of the cells have more than ten species, while 52% of the cells only have one species (Fig. 6).
|
Complementarity analysis
Although nine grid cells are enough to capture 51% of all wild potato species, the minimum number of grid cells needed to capture all species at least once is 59 (out of 1317 total cells) (Figs. 8, 9). Twenty-three cells contribute only one additional species each (Fig. 9). The locations of the first 15 grid cells that get selected by Rebelo's (1994)
algorithm follow a pattern that can only partly be inferred from Fig. 5. The early appearance of cells from Mexico and Ecuador (grid cells number 4 and 7 in Fig. 8) may seem surprising because they have only an intermediate level of species richness (Fig. 5), but the species in these countries are all different from those observed in Peru and Bolivia (the single observation of S. acaule mentioned above is the only exception). Areas in Argentina are selected later (numbers 6 and 13 in Fig. 8) than might be expected on the basis of its high number of species (Fig. 5). This is because some of the species in these cells were already included in area number 3 (Fig. 8) in southern Bolivia.
|
|
Distribution by latitude and altitude
Wild potatoes occur between 38° N and 41° S. The highest number of species per degree latitude (>20) occurs between 8° S and 20° S, i.e., from north-central Peru to central Bolivia, and around 20° N, in the central Mexican highlands (Fig. 10). The distribution of the number of species by latitude follows a bimodal distribution. There is a remarkably similar pattern between 20° and 40° in both hemispheres. However, in the zone between 20° N and 20° S, and particularly the zones between 8° N and 15° N and 8° S and 15° S, the number of species is rather different, with a conspicuously higher number of species in the southern hemisphere.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), this is the first study in which a group of closely related wild crop relatives is systematically analyzed using GIS. This study is also unique in its use of a very large number of georeferenced observations for a single group of closely related wild species. Wild potatoes occur between 38° N and 41° S. Species richness of wild potatoes is particularly high in the Central and South American tropical highlands, with clear peaks between 8° S and 20° S and around 20° N, i.e., areas in the Andes of northern Argentina, Bolivia, Ecuador, and Peru, and in central Mexico. Peru stands out for the high number of wild potato species as well as for the high number of rare wild potato species.
Many wild potatoes species are narrowly endemic, and yet a selection of only nine grid cells was needed to include 51% of the species, which emphasizes the presence of areas of high species richness. In a number of countries, the most species-rich grid cell has a high percentage all species. This might facilitate the design of in situ conservation reserves to protect these species (as called for by Huamán, Hoekstra, and Bamberg, 2000
). However, the clustering of species on the scale used in this study is not directly meaningful for conservation programs, which typically would operate in considerably smaller areas.
The lower species richness around the equator, particularly in the northern hemisphere, as compared with higher tropical latitudes, contrasts with the general pattern of increasing species richness (of all flora and fauna) towards the equator (e.g., Blackburn and Gaston, 1996
; Gaston and Williams, 1996
). The absence of cool tropical highlands appears to be an important factor that explains the paucity of wild potato species around the equator, particularly in the northern hemisphere. The climate in these equatorial areas is also more humid and less seasonal. The absence of a clear dry (or cold) season could diminish the relative fitness of tuber-bearing perennials such as wild potatoes. At higher latitudes, where the data are more similar for both hemispheres, there is a considerable stretch of high mountains in central Mexico (the Mexican transvolcanic belt). At even higher latitudes, wild potatoes mainly occur below 1500 m altitude.
Our data were gathered from several sources (expeditions) and this may have led to some redundancies. Particularly, type localities of rare species were visited by different expeditions, as these species may not be found elsewhere. Hence, some of these species are even more rare and endemic than appears from our data. Overall, however, it may make our data more reliable given the timing dependency of the results of wild potato exploration: there are differences within and among years in the likelihood of finding certain species in certain locations.
Some of our records are recent, but many date back many years. In some cases, the habitat in which the species occurred has now disappeared, and the species may no longer occur there. For example, Spooner et al. (1998)
describe a rapid rate of loss of wild potato habitat in upland forests in Guatemala. However, our recent experience in Peru, and elsewhere, indicates that this is not always the case. For example, Spooner et al. (1999)
and Salas et al. (2001)
collected many wild potato species in Peru in the exact location, often at the type locality, where they had been collected many years before. In some cases, it was not possible to collect at documented localities, but this was often attributed to phenology, as wild potatoes often have a short growing period. In other cases, incomplete locality data hindered collections.
Recorder effort (bias) (Rich and Woodruff, 1992
; Gaston, 1996
; Hijmans et al., 2000
) also influences the results. Nevertheless, the consistency of the results (there are no sudden gaps in the distributions) leads us to believe that we have presented a good representation of overall wild potato distribution, and this large database is one of the most comprehensive for any group of plants. Although the number of observations per grid cell is a reasonably good predictor of the number of species in that cell, we do not think that a high number of species follows causally from a high number of observations.
Wild potatoes have been the object of intensive exploration over many years (Spooner and Hijmans, 2001
). While there will still be some areas where further exploration would discover additional species, a low number of observations in a cell is more likely to be the result than the cause of low species richness. Hence, the logic might be reversed: much collection has taken place in areas with a known high species diversity, referred to by Hijmans et al. (2000)
as "hotspot bias." Attempts could be made to correct for recorder effort through rarefaction (estimating how many species would have been observed given a constant sample size per grid cell; Sanders, 1968
; Prendergast et al., 1993
; Gaston, 1996
). However, this would also lead to a great loss of information, because the observed number of species would be replaced by an estimate.
Distributions of wild relatives of crop plants have previously been summarized by country (e.g., Huamán, Hoekstra, and Bamberg, 2001
). However, countries (or their subsidiary administrative units) have different shapes and sizes. Hence, they have only limited value in comparing geographic distributions of wild plants, despite the advantage of being familiar entities. Equal-area grid cells as used in this study are clearly to be preferred. Nevertheless, there are a number of methodological issues that need to be considered when using grids.
Resolution (cell size) of the grid affects the results. Number of species per grid cell will increase with the size of the grid cell, but this increase will be different among areas. We used a 50 x 50 km grid (and a circular neighborhood with a radius of 50 km) to strike a balance between the desire for high resolution and geographic sampling bias (Hijmans et al., 2000)
. This bias becomes less important when grid cell size increases. We used a neighborhood of 7854 km2; most other studies of species distribution on a continental (or global) scale have used much larger grid cells. For example, Gaston and Williams (1996)
review various global studies using grids of 611 000 km2. The same was also used by Blackburn and Gaston (1996)
for a study on birds in the Americas. Given the high density of observations, it was not necessary to use such large grid cells in our study. This would have amounted to serious information loss, and even smaller grid cells will be more appropriate for design of in situ reserves or for planning collecting expeditions to specific areas.
We used a circular neighborhood to assign values to grid cells because this gives a smoother result, particularly for areas with few observations (Cressie, 1991
; Bonham-Carter, 1994
) and is less sensitive to the origin of the grid and small errors in the locality data. However, this does include yet other factors (size, shape, and method used to define the neighborhood) that need to be considered when interpreting the results, in addition to the scale effect (effect of grid cell size). Research is needed to better understand the effects of different gridding methods and scales, in relation to data density and quality and the objectives of the study.
A complication of using species richness is the existence of conflicting taxonomic classifications (Gaston, 1996
). Wild potatoes are a classic case in this respect (Harlan and De Wet, 1971
; Spooner and Van den Berg, 1992
). We use the list of species provided by Spooner and Hijmans (2001)
, which is a compilation of taxonomic names that updates Hawkes (1990)
. Nevertheless, future changes in species circumscription will likely change the results presented here (Spooner and Hijmans, 2001
). For example, the results of ongoing research described by Van den Berg et al. (1998)
and Miller and Spooner (1999)
on the 30 taxa of the S. brevicaule complex suggest a need for reduction of the number of species in this group. These species occur in southern Peru and Bolivia and taxonomic revision would reduce species richness here.
Although Peru seems to be reasonably well explored (number of species over observations), it has an extraordinarily high number of apparently rare species. This indicates that Peru may still harbor unknown species, as illustrated by the ten new Peruvian wild potato species described by C. M. Ochoa between 1998 and 1999 (Ochoa, 1999
; Spooner and Hijmans, 2001
). Peruvian species are also underrepresented in genebanks, and a collecting program is currently under way to fill this important gap (Spooner et al., 1999
; Salas et al., 2001
).
|
|
FOOTNOTES |
|---|
4 Author for reprint requests.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Bonham-Carter G. 1994 Geographic information systems for geoscientists. Modelling with GIS. Computer Methods in the Geosciences 13. Pergamon/Elsevier, London, UK
Contreras M. A. D. M. Spooner 1999 Revision of Solanum section Etuberosum (subgenus Potatoe). In M. Nee, D. E. Symon, R. N. Lester, and J. P. Jessop [eds.], Solanaceae IV, 227–245. Royal Botanic Gardens, Kew, UK
Cressie N. A. C. 1991 Statistics for spatial data. John Wiley & Sons, New York, New York, USA
Csuti B. S. Polasky P. H. Williams R. L. Pressey J. D. Camm M. Kershaw A. R. Kiester B. Downs R. Hamilton M. Hurst K. Sahr 1997 A comparison of reserve selection algorithms using data on terrestrial vertebrates in Oregon. Biological Conservation 80: 83-97[CrossRef][Web of Science]
Environmental Systems Research Institute. 1999 ArcView-GIS 3.1. Environmental Systems Research Institute, Redlands, California, USA
Gaston K. J. 1996 Species richness: measure and measurement. In K. J. Gaston [ed.], Biodiversity, a biology of numbers and difference, 77–113. Blackwell Science, London, UK
Gaston K. J. P. H. Williams 1996 Spatial patterns in taxonomic diversity. In K. J. Gaston [ed.], Biodiversity, a biology of numbers and difference, 202–229. Blackwell Science, London, UK
Guarino L. A. Jarvis R. J. Hijmans N. Maxted In press Geographic information systems (GIS) and the conservation and use of plant genetic resources. In: Proceedings of the international conference on science and technology for managing plant genetic diversity in the 21st century (SAT21), Kuala Lumpur, Malaysia, 12–16 June 2000. International Plant Genetic Resources Institute, Rome, Italy
Harlan J. R. J. M. J. De Wet 1971 Towards a rational classification of cultivated plants. Taxon 20: 509-517[CrossRef]
Hawkes J. G. 1979 Evolution and polyploidy in potato species. In J. G. Hawkes, R. N. Lester, and A. D. Skelding [eds.], The biology and taxonomy of the Solanaceae, 637–645. Academic Press, London, UK
Hawkes J. G. 1990 The potato: evolution, biodiversity and genetic resources. Belhaven Press, London, UK
Hawkes J. G. 1997 A database for wild and cultivated potatoes. Euphytica 93: 155-161[CrossRef][Web of Science]
Hawkes J. G. J. P. Hjerting 1969 The potatoes of Argentina, Brazil, Paraguay and Uruguay: a biosystematic study. Clarendon Press, Oxford, UK
Hawkes J. G. J. P. Hjerting 1989 The potatoes of Bolivia, their breeding value and evolutionary relationships. Oxford University Press, Oxford, UK
Hijmans R. J. M. Cruz E. Rojas L. Guarino 2001 DIVA-GIS, version 1.4. A geographic information system for the management and analysis of genetic resources data. Manual. International Potato Center and International Plant Genetic Resources Institute, Lima, Peru
Hijmans R. J. K. A. Garrett Z. HuamÁn D. P. Zhang M. Schreuder M. Bonierbale 2000 Assessing the geographic representativeness of genebank collections: the case of Bolivian wild potatoes. Conservation Biology 14: 1755-1765[CrossRef][Web of Science]
Hijmans R. J. M. Schreuder J. De La Cruz L. Guarino 1999 Using GIS to check co-ordinates of germplasm accessions. Genetic Resources and Crop Evolution 46: 291-296[CrossRef][Web of Science]
Huamán Z. R. Hoekstra J. B. Bamberg 2000 The Inter-genebank potato database and the dimensions of wild potato germplasm. American Journal of Potato Research 77: 353–362 [Web of Science]
Jansky S. H. 2000 Breeding for disease resistance in potato. Plant Breeding Reviews 19: 69-155
Kardolus J. P. 1998 A biosystematic analysis of Solanum acaule. Ph.D. dissertation, Wageningen Agricultural University, Wageningen, The Netherlands
Miller J. T. D. M. Spooner 1999 Collapse of species boundaries in the Solanum brevicaule complex (Solanaceae, Sect. Petota): molecular data. Plant Systematics and Evolution 214: 103-130[CrossRef][Web of Science]
Ochoa C. M. 1962 Los Los Solanum tuberíferos silvestres del Perú (Tuberarium subsecc. Hyperbasarthrum). Privately published
Ochoa C. M. 1990 The potatoes of South America: Bolivia. Cambridge University Press, Cambridge, UK
Ochoa C. M. 1999 Las papas de sudamerica: Peru. International Potato Center, Lima, Peru
Prendergast J. R. S. N. Wood J. H. Lawton B. C. Eversham 1993 Correcting for variation in recording effort in analyses of diversity hotspots. Biodiversity Letters 1: 39-53[CrossRef]
Rabinowitz D. 1981 Seven forms of rarity. In H. Syne [ed.], The biological aspects of rare plant conservation, 205–217. John Wiley & Sons, New York, New York, USA
Rabinowitz D. C. R. Linder R. Ortega D. Begazo H. Murguia D. S. Douches C. F. Quiros 1990 High levels of interspecific hybridization between Solanum sparsipilum and S. stenotomum in experimental plots in the Andes. American Potato Journal 67: 73-81[Web of Science]
Rebelo A. G. 1994 Iterative selection procedures: centres of endemism and optimal placement of reserves. Strelitzia 1: 231-257
Rebelo A. G. W. R. Siegfried 1992 Where should nature reserves be located in the Cape Floristic Region, South Africa? Models for the spatial configuration of a reserve network aimed at maximizing the protection of diversity. Conservation Biology 6: 243-252
Rich T. C. G. E. R. Woodruff 1992 Recording bias in botanical surveys. Watsonia 19: 73-75
Ross H. 1986 Potato breeding: problems and perspectives. Advances in Plant Breeding 13. Paul Parey Verlag, Berlin, Germany
Salas A. R. D. M. Spooner Z. Huamán R. Vinci Torres M. R. Hoekstra K. Schüler R. J. Hijmans 2001 Taxonomy and collections of wild potato species in Central and Southern Peru in 1999. American Journal of Potato Research 78: 197–207 [Web of Science]
Sanders H. L. 1968 Marine benthic diversity: a comparative study. American Naturalist 102: 243-282[CrossRef][Web of Science]
Spooner D. M. R. Castillo T. L. López J 1992 Ecuador, 1991 potato germplasm collecting expedition: taxonomy and new germplasm resources. Euphytica 60: 159-169[Web of Science]
Spooner D. M. R. J. Hijmans 2001 Potato systematics and germplasm collecting, 1989–2000. American Journal of Potato Research 78: 237–268 [Web of Science]
Spooner D. M. R. Hoekstra R. G. Van den Berg V. Martnez 1998 Solanum sect. Petota in Guatemala: taxonomy and genetic resources. American Journal of Potato Research 75: 3-17
Spooner D. M. R. Hoekstra B. Vilchez 2001 Solanum section Petota in Costa Rica: taxonomy and genetic resources. American Journal of Potato Research 78: 91–98 [Web of Science]
Spooner D. M. A. Salas Z. Huamán R. J. Hijmans 1999 Wild potato collecting expedition in southern Peru (Departments of Apurímac, Arequipa, Cusco, Moquegua, Puno, Tacna) in 1998: taxonomy and new genetic resources. American Journal of Potato Research 76: 103-119[Web of Science]
Spooner D. M. R. G. Van den Berg 1992 An analysis of recent taxonomic concepts in wild potatoes (Solanum sect. Petota). Genetic Resources and Crop Evolution 39: 23-37[CrossRef]
Spooner D. M. R. G. Van den Berg A. Rivera-Peña P. Velguth A. Del Rio A. Salas In press Taxonomy of Mexican and Central American members of Solanum Series Conicibaccata (sect. Petota). Systematic Botany.
United States Geological Survey. 1998 . Database available at: http://edcwww.cr.usgs.gov/landdaac/gtopo30/gtopo30.html
Van den Berg R. G. J. T. Miller M. L. Ugarte J. P. Kardolus J. Villand J. Nienhuis D. M. Spooner 1998 Collapse of morphological species in the wild potato Solanum brevicaule complex (Solanaceae: sect. Petota). American Journal of Botany 85: 92-109[Abstract]
Walker T. S. P. E. Schmiediche R. J. Hijmans 1999 World trends and patterns in the potato crop: an economic and geographic survey. Potato Research 42: 241-264
Watanabe K. S. J. Peloquin 1989 Occurrence of 2n pollen and ps gene frequencies in cultivated groups and their related wild species in tuber-bearing solanums. Theoretical and Applied Genetics 78: 329-336[Web of Science]
Zeven A. C. P. M. Zhukovsky 1975 Dictionary of cultivated plants and their centres of diversity. Excluding ornamentals, forest trees and lower plants. Centre for Agricultural Publishing and Documentation (PUDOC), Wageningen, The Netherlands
CiteULike Complore Connotea Del.icio.us Digg Facebook Reddit Technorati Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  J. A. Ruiz Corral, N. Duran Puga, J. d. J. Sanchez Gonzalez, J. Ron Parra, D. R. Gonzalez Eguiarte, J.B. Holland, and G. Medina Garcia Climatic Adaptation and Ecological Descriptors of 42 Mexican Maize Races Crop Sci., July 1, 2008; 48(4): 1502 - 1512. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Miller and J. H. Knouft GIS-based characterization of the geographic distributions of wild and cultivated populations of the Mesoamerican fruit tree Spondias purpurea (Anacardiaceae) Am. J. Botany, December 1, 2006; 93(12): 1757 - 1767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Jansky, R. Simon, and D. M. Spooner A Test of Taxonomic Predictivity: Resistance to White Mold in Wild Relatives of Cultivated Potato Crop Sci., November 21, 2006; 46(6): 2561 - 2570. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. del Rio and J. B. Bamberg Geographical Parameters and Proximity to Related Species Predict Genetic Variation in the Inbred Potato Species Solanum verrucosum Schlechtd Crop Sci., July 1, 2004; 44(4): 1170 - 1177. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
5), and the ratio of observations to species
