CROP SCIENCE, VOL.

52, JANUARYFEBRUARY 2012 231

RESEARCH
G
enetic resources have been utilized through programs of
domestication and improvement (Berthaud, 1997), promot-
ing the use and distribution of plants under diferent ecological,
cultural, and technological conditions in relation to the centers of
origin (Harlan, 1992; Casas and Barbera, 2002). Artifcial selec-
tion has produced morphological, physiological, and genetic
changes that modify mating systems and the genetic structure
of the populations of genetic resources (Hawkes, 1983; Doebley,
1989). Domestication is a relatively recent occurrence (the past
10,000 yr), and for this reason, it is expected that crops will exhibit
less genetic variation and little neutral genetic diferentiation with
respect to their progenitors, despite the large morphological difer-
ences among crops and their wild relatives (Doebley, 1989, 1992).
The genus Capsicum originated in South America and includes 29
Genetic Diversity and Structure of Pepper
(Capsicum Annuum L.) from Northwestern Mexico
Analyzed by Microsatellite Markers
Antonio Pacheco-Olvera,* Sergio Hernndez-Verdugo, Vctor Rocha-Ramrez,
Antonio Gonzlez-Rodrguez, and Ken Oyama
ABSTRACT
The analysis of the variability and genetic struc-
ture of wild and landrace populations of pepper
(Capsicum annuum L.) is important for the man-
agement and conservation of valuable genetic
resources and to understand the consequences
of domestication on the patterns of neutral
genetic variation. For this purpose, 12 popula-
tions of wild peppers, 3 landrace populations
and 7 hybrid populations from northwestern
Mexico were studied using microsatellites. On
average, 3.62 alleles per locus were detected
in the wild relatives, 3.37 in the landraces, and
3.08 in the hybrids. According to the average
values of expected heterozygosity (He), slightly
greater genetic diversity was found among the
wild relatives (He = 0.466) than in the hybrids
(He = 0.440) or the landraces (He = 0.422). In
terms of the average number of alleles per
locus and the average expected heterozygos-
ity, reductions of 8.18 and 10.25% were found
in the genetic diversity of the landraces and
hybrids, respectively, with respect to the wild
populations. The genetic differentiation among
the populations was the highest among hybrids
(G
ST
= 0.324), followed by landraces (0.309) and
wild relatives (the lowest, at 0.208). Cluster anal-
ysis clearly demarcated the wild relatives and
domesticated populations into different groups.
The high levels of genetic diversity found among
C. annuum in northwestern Mexico suggest that
the wild and landrace populations are a valuable
resource that should be conserved.
A. Pacheco-Olvera and S. Hernndez-Verdugo, Facultad de Agronoma,
Univ. Autnoma de Sinaloa, Culiacn, Sinaloa, Mxico, Carret-
era Culiacn-Eldorado km 17.5; V. Rocha-Ramrez, A. Gonzlez-
Rodrguez, and K. Oyama, Centro de Investigaciones en Ecosistemas,
Univ. Nacional Autnoma de Mxico (UNAM), Antigua Carretera a
Ptzcuaro No 8701, Colonia San Jos de la Huerta, 58190, Morelia
Michoacn, Mxico. Received 14 June 2011. *Corresponding author
(apo@uas.uasnet.mx).
Abbreviations: A, average number of alleles per locus; F
IS
, inbreed-
ing coef cient; F
IT
, total inbreeding; F
ST
, genetic diferentiation; GD,
genetic distance; G
ST
, coef cient of genetic diferentiation; H
E
H
O

P, percentage of polymorphic loci; PCR, polymerase chain reaction;
RAPD, random amplifed polymorphic DNA; SSR, simple sequence
repeat, or microsatellite; UPGMA, unweighted pair group method
with arithmetic averaging.
Published in Crop Sci. 52:231241 (2012).
doi: 10.2135/cropsci2011.06.0319
Published online 11 Nov. 2011.
Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA
All rights reserved. No part of this periodical may be reproduced or transmitted in any
form or by any means, electronic or mechanical, including photocopying, recording,
or any information storage and retrieval system, without permission in writing from
the publisher. Permission for printing and for reprinting the material contained herein
has been obtained by the publisher.
232 WWW.CROPS.ORG CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012
species (Hunziker, 1979; Eshbaugh 1980; Pickersgill, 1984;
Hernndez-Verdugo et al., 2001), of which 5 (C. annuum,
C. chinense, C. frutense, C. baccatum, and C. pubescens) have
been domesticated (Pickersgill, 1997). Peppers were one of
the frst plants to be domesticated in the Americas, and they
are used worldwide as a spice, condiment, and a vegetable
(MacNeish, 1964). Capsicum annuum is the most economi-
cally important species and includes the majority of peppers
in the Americas, Africa, and Asia as well as several non-
pungent sweet materials that grow in temperate regions of
Europe and North America (Pickersgill, 1997). The wild
populations of this species are known as Capsicum annuum
var. glabriusculum (Heiser and Pickersgill, 1975; Hernndez-
Verdugo et al., 1999), which is considered the progenitor of
domesticated peppers. These populations, which have the
common name chiltepin or piquines, are widely dis-
tributed throughout Mexico. Individual plants are generally
found under trees in tropical deciduous forests, orchards,
and grazing pastures (Hernndez-Verdugo et al., 1999).
All the wild forms are perennials, herbs, or climbers.
Their fruits are small, red, and spicy, and they are eaten by
birds, which spread their seeds (Vzquez-Dvila, 1996). Cap-
sicum annuum var. annuum has been collected, cultivated, and
consumed in Mexico for hundreds of years. Plantings began
approximately 8000 yr ago, followed by the cultivation and
defnitive domestication of the spicy fruits approximately
6000 yr ago (Perry and Kent, 2007). Since pre-Columbian
times indigenous communities have selected the character-
istics with the greatest agricultural importance, such as the
shape and size of the fruit (Pickersgill et al., 1979), which has
resulted in high morphological variation and a large num-
ber of landrace varieties that are adapted to local conditions.
The landraces are native varieties that have been cultivated
for long periods of time with intermediate but stable perfor-
mance under a system of agricultural management. These
varieties constitute an important genetic resource that is in
danger of being lost because these crops are being displaced
by improved hybrids with higher yields (Fri, 1986; Zeven,
1998), whose demand in the market has increased over the
last 30 yr. However, these hybrid varieties have the disad-
vantage of being susceptible to plagues and diseases inherent
to the crop. Therefore, a study of the variability and genetic
structures of wild and landrace populations is important for
the management and conservation of this valuable genetic
resource. Such variability constitutes an important genetic
reservoir that can help solve agricultural problems, for exam-
ple, by providing resistance to attack from plagues and dis-
eases (Hernndez-Verdugo et al., 1998; Votava et al., 2005).
Species of the genus Capsicum have been studied using
morphological, cytogenetic, and molecular markers with
diferent results depending on the origin of the materials
(germplasm bank), the level of domestication of the species
(wild and cultivated), and the genetic markers used (McLeod
et al., 1983; Loaiza-Figueroa et al., 1989; Conicella et al.,
1990; Lefebvr et al., 1993; 2001; Prince et al., 1992, 1995;
Hernndez-Verdugo et al., 2001; Oyama et al., 2006).
Recently, Aguilar-Melndez et al. (2009) examined
the nucleotide diversity of the sequences of the Dhn (dehy-
drin, an osmotic stress-related gene), G3pdh (gene encod-
ing glyceraldehyde 3-phosphate dehydrogenase), and Waxy
(gene sequence encoding granule-bound starch synthase)
genes in 58 semi-wild populations and 22 domesticated
populations. The genetic diversity was found to be similar
among semi-wild populations (populations with morpho-
logical characteristics that are similar to those of wild popu-
lations and that are grown with minimal management) and
domesticated populations of Capsicum.
The objective of this study was to compare the popu-
lation genetic variation and structures of wild relatives,
landraces, and hybrid populations of C. annuum to under-
stand the consequences of the domestication process on
the patterns of neutral genetic variation. For this purpose,
microsatellites (also called simple sequence repeats, or
SSRs) were used, as these tools have been applied to a
great variety of plants owing to the high degree of poly-
morphism, codominance, and reproducibility (Hagh-
nazari et al., 2005). In addition, the identifcation and
characterization of numerous microsatellite loci for the
Capsicum genus have recently been reported (Lee et al.,
2004; Minamiyama et al., 2006; Portis et al., 2007; Huang
et al., 2000; Ince et al., 2010).
MATERIALS AND METHODS
Plant Materials
The plant material was sampled in northwestern Mexico, in
the states of Nayarit, Sinaloa, and Sonora. In total, 12 wild
populations of C. annuum were collected, covering a latitudi-
nal interval of approximately 767 km from 2124 to 2711 N
(Table 1, Fig. 1). Eleven of the wild populations were sampled
in deciduous tropical forests in the states of Sonora and Sinaloa,
and the other population was sampled within a commercial
orchard at the location of El Llano in the state of Nayarit. The
domesticated populations of C. annuum were sampled in the
crop felds of Sinaloa and correspond to the landrace and hybrid
groups. The landrace peppers belong to the Tequila, Cascabel,
and Cola de Rata types. The hybrids correspond to the com-
mercial types, Hngaro, Guajillo, Anaheim, Cambray, Serrano,
Poblano, and Chilaca. Leaves were sampled from 20 plants that
were randomly selected from each population. The leaves were
immediately frozen in liquid nitrogen and fnally transported
to the laboratory and stored at 80C until DNA extraction.
DNA Extraction and Amplication
of Microsatellites
The total DNA was extracted from young leaves using the
method of Lefort and Douglas (1999). Initially, 14 microsatel-
lite primer pairs developed for the Capsicum genus were tested,
but only 7 were selected to be scored in all sampled individuals
on the basis of the reproducibility of the results and the levels of
CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012 WWW.CROPS.ORG 233
and Horvitz, 1953). To determine the signifcance of F
ST
, the test
X
2
= 2NF
ST
(k1) was performed with [(k1)(s-1)] degrees of free-
dom, where s is the number of subpopulations (Workman and
Niswander, 1970). The genetic diversity within each subpopu-
lation (H
S
) was calculated, along with the genetic diversity of
variation revealed. These seven were the following: EPMS342,
EPMS417, EPMS440, EPMS480, EPMS501 (Nagy et al., 2007),
EPMS643, and EPMS919 (Portis et al., 2007). Polymerase chain
reactions (PCR) were performed using a PCR multiplex kit
(QIAGEN, Hombrechtikon, Switzerland) with 5-L reactions as
follows: 1X PCR multiplex Master Mix, 2 M of each primer,
distilled water, and 20 ng of DNA. The temperature profle for
PCR amplifcation consisted of a cycle of 95C for 5 min, fol-
lowed by 35 cycles of 94C for 1 min, 60C for 1.5 min, and
72C for 1 min. After cycling, there was a fnal elongation step at
72C for 5 min. The PCR-amplifed products were analyzed by
capillary electrophoresis using an ABI 3100-Avant (Applied Bio-
systems, Hitachi, Japan) automated sequencer. Each of the sam-
ples analyzed in the sequencer contained 10 L of formamide, 0.4
L of Gene Scan 500 LIZ, and 1.0 L of the PCR product. For
visualization of the alleles, the Peak Scanner program version 1.0
(Applied Biosystems) was used.
Statistical Analyses
The genetic diversity of the populations was quantifed accord-
ing to the average number of alleles per locus (A), the percentage
of polymorphic loci (P), the observed heterozygosity (H
O
), and
the expected heterozygosity (H
E
) using the GDA program ver-
sion 1.0 (Lewis and Zaykin, 2001). To assess whether the seven
loci analyzed provide independent information, a test of pairwise
linkage disequilibrium was conducted using Arlequin software
version 3.0 (Excof er et al., 2005). The genetic structure of the
populations was calculated with Wrights F statistics (F
IS,
F
ST
, and
F
IT
) using GenAlex software versions 6.4 (Peakall and Smouse,
2006). To determine whether the F
IS
and F
IT
values were signif-
cantly diferent from zero with respect to the Hardy-Weinberg
equilibrium, a X
2
test was performed using the following equa-
tion: X
2
= F
2
N(k1) with [k(k1)]/2 degrees of freedom, where
N is the size of the sample and k is the number of loci analyzed (Li
Table 1. Collection sites of Capsicum annuum popula-
tionsstudied.
Population Abbreviation Status Lat Long
Mocuzary MCU Wild 27 11 48.60 109 07 25.62
Estacin
Cuaternaria
EST Wild 27 05 00.00 109 33 00.00
Potrero de
Alcantar
POA Wild 26 54 00.00 108 53 00.00
Rancho el
Coyote
RCY Wild 26 53 55.50 108 46 28.80
Tehueco TEH Wild 26 20 00.00 108 45 00.00
Lode Vega LVE Wild 26 22 00.00 108 40 00.00
Pilares PIL Wild 25 04 00.00 107 21 00.00
Chapeteado CHA Wild 24 29 00.00 107 26 00.00
Alcoyonqui ALC Wild 24 54 00.00 107 12 00.00
Tabal TAB Wild 24 24 00.00 107 05 00.00
Otates OTA Wild 23 02 00.00 105 55 00.00
Llano LLA Wild 21 24 00.00 105 10 00.00
Tequila TEQ Landrace 22 43 07.50 105 10 00.00
Cascabel CAS Landrace 22 43 07.50 105 51 29.58
Cola de Rata CRA Landrace 22 41 01.80 105 47 48.84
Hungaro HUN Hybrid 22 45 19.68 105 52 43.02
Guajillo GUA Hybrid 22 41 01.80 105 47 48.84
Anahim ANA Hybrid 23 01 25.20 106 10 16.62
Cambray CAM Hybrid 22 53 47.22 105 58 16.56
Serrano SER Hybrid 23 12 00.18 106 13 49.50
Poblano POB Hybrid 23 01 25.20 106 10 16.62
Chilaca CHI Hybrid 22 43 53.16 105 50 50.04
Figure 1. Map showing the distribution of populations of Capsicum annuum studied in Northwestern Mexico.
234 WWW.CROPS.ORG CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012
the total population (H
T
), the total genetic diversity distributed
among populations (D
ST
), and the measurement of population
diferentiation (G
ST
) for the seven loci (Nei, 1978), where G
ST

= D
ST
/H
T
and D
ST
= H
T
H
S
. These statistics were estimated
using FSTAT software version 2.9.3.2 (Goudet, 1995). In addi-
tion, a nonparametric analysis of molecular variance (AMOVA)
was performed as described by Stewart and Excof er (1996)
using four hierarchical levels: (1) wild vs. landrace vs. hybrid
populations, (2) wild populations, (3) landrace populations, and
(4) hybrid populations. The analysis was performed using Arle-
quin software version 3.0 (Excof er et al., 2005). Finally, Bayes-
ian clustering was performed using the STRUCTURE program
(Pritchard et al., 2000), where all individuals of the three groups
of peppers were analyzed together under a mixed model with
correlated allelic frequencies. To obtain the most probable K
value (number of genetic groups), values of K from 1 to 5 were
tested, with 10 independent runs for each K. The longitude of
the run was 500,000 steps followed by 10
6
iterations. The K value
with the greatest probability was calculated estimating the maxi-
mum value of the K statistic, according to Evanno et al. (2005).
The genetic similarity relationships among the populations
under study were estimated by the unweighted pair group method
with arithmetic averaging (UPGMA) using Neis genetic dis-
tances (1972), calculated with the Tools for Population Genetics
Analyses (TFPGA) program version 1.3 (Miller, 1997). Similarly,
an Analysis of Principal Coordinates was performed with Neis
genetic distances (1972) using the GenAlex software version 6.4
(Peakall and Smouse, 2006). With the goal of testing the isola-
tion by distance model in the wild populations, the correlation
between pairwise genetic distances and pairwise geographic dis-
tances among populations was analyzed with a Mantel test using
TFPGA software version 1.3 (Miller,1997).
To determine the possible genetic barriers in the distribu-
tion of the wild populations of C. annuum from the northwest
of Mexico, the Monmoniers maximum diference algorithm
was applied (Monmonier, 1973) with the BARRIERS pro-
gram version 2.2 (Manni et al., 2004). For the analysis of the
barriers, Neis genetic distances (1983) were employed and esti-
mated for the 12 wild populations.
RESULTS
Population Genetic Variation
In wild populations, an average number of alleles per locus
(A) of 3.6 was detected with the seven SSR markers, with
a range from 2.6 (Mocuzari) to 5.0 (Estacin Cuaternaria).
In the landrace populations, A was 3.4, with a range from
3.0 (Tequila) to 3.8 (Cola de rata), and in the hybrid popu-
lations A was 3.1, with a range from 2.0 (Poblano) to 4.0
(Cambray) (Table 2). The percentage of polymorphic loci
(P) was 89.3% among the wild populations, with a range
from 71.4% (Alcoyonqui and Tabal) to 100% (various
populations). Among landrace populations, the average
was 80.2% and the minimum and maximum values were
71.4 (Cascabel) and 85.7% (Tequila), respectively. In the
hybrid populations, the average was 93.2%, with a mini-
mum of 66.7 (Poblano) and a maximum of 100% (various
populations). The average H
O
and H
E
were, respectively,
0.225 and 0.466 for the wild populations, 0.201 and 0.422
for the landrace populations, and 0.237 and 0.440 for the
hybrid populations (Table 2). The diferences in the aver-
age values of A, P, H
O
, and H
E
among the three groups of
pepper populations were not signifcant. Among the three
groups of pepper populations that were analyzed, the
average H
E
value exceeded the H
O
values, which indicates
that there is an excess of homozygous individuals. The
test for pairwise linkage disequilibrium after a Bonferroni
correction was nonsignifcant in all cases, indicating that
the seven analyzed loci provide independent information.
Genetic Differentiation among Populations
The average H
T
obtained with the seven microsatellite loci
for the wild populations was 0.658, with a range from 0.216
(EPMS480 locus) to 0.885 (EPMS417). For the landrace
populations, the average H
T
was 0.648, with a range from
0.461 (EPMS342) to 0.821 (EPMS417). For the hybrid pep-
pers, the average H
T
was 0.725, with a range from 0.625
(EPMS440) to 0.868 (EPMS919) (Table 3). The average
intrapopulation genetic diversity for each group was highest
for the wild peppers (H
S
= 0.501), followed by the hybrid
populations (H
S
= 0.493) and then by the landraces (H
S
=
0.460). The average coef cient of genetic diferentiation
(G
ST
) was 0.208 for the wild populations, with a range from
0.037 (EPMS440) to 0.295 (EPMS919). For the landraces,
the average was 0.309, with a range from 0.033 (EPMS501)
to 0.710 (EPMS480). For the hybrid populations, the average
G
ST
value was 0.324, with a range from 0.192 (EPMS440)
to 0.474 (EPMS342) (Table 3). The average inbreeding
coef cient (F
IS
) within the wild populations was positive
and statistically diferent from zero (0.412) (Table 4). The
only value that was not diferent from zero was found for
the EPMS440 locus, which indicates Hardy-Weinberg
equilibrium. The landrace populations presented positive
values that were diferent from zero at all of their loci, with
an average F
IS
value of 0.542. The hybrid populations pre-
sented positive values that were diferent from zero at fve
loci. The total inbreeding (F
IT
) among the three groups of
peppers varied from 0.709 among the landrace peppers to
0.560 for the wild peppers. These results indicate that there
is a high level of genetic diferentiation among the popula-
tions of C. annuum for the three levels of domestication.
The genetic diferentiation was greatest among the hybrid
populations (F
ST
= 0.424), followed by the landraces (F
ST
=
0.396), and fnally the wild populations (F
ST
=0.297). The
analysis of molecular variation at the three hierarchical lev-
els (Table 5) showed that most of the variation occurred
within the populations (74.2%). However, a considerable
proportion of the variation was found between populations
(15.7%). The remaining variation (10.1%) was distributed
among the three population groups (wild, landraces, and
hybrids), indicating a clear distinction among these groups.
CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012 WWW.CROPS.ORG 235
Relationships between Populations
The cluster analysis performed with the UPGMA algo-
rithm generated a dendrogram that divided the popula-
tions into two genetic groups: group I, which was formed
by two landrace populations mixed with the hybrids; and
group II, which was formed by the wild populations, the
Tequila landrace population, and the Chilaca hybrid (Fig.
2). The parental lines of commercial hybrids are unknown.
It is possible that the Chilaca variety was obtained from
parental lines that were introgressed with genes from wild
populations. For the Neis (1972) genetic distance (GD), all
the pepper populations presented an average value of 0.703,
with a minimum value of 0.112 and a maximum value of
2.975. Within the wild populations, the average GD value
was 0.455, with a minimum distance of 0.112 (EST and
PIL) and a maximum distance of 1.121 (LLA and LVE). The
landrace populations presented an average distance of 0.745,
with a minimum distance of 0.566 (CAS and CRA) and a
maximum distance of 0.955 (CAS and TEQ). The hybrid
populations presented an average distance of 0.908, with a
minimum value of 0.402 (HUN and CHI) and a maximum
value of 2.201 (SER and ANA). The landrace populations
were separated from the wild ones by a distance of 0.290,
and the hybrid populations were separated from the wild
ones by a distance of 0.453. Finally, the hybrid populations
were interlaced with the landrace populations.
In the principal coordinate analysis, coordinate 1
explained 45.0% and coordinate 2 explained 16.3% (Fig. 3) of
the total variation. The greatest discrimination was observed
only for the axes of the two frst main coordinates. The axis
of coordinate 1 clearly separated the wild populations from
the landrace and hybrid populations, yet the landrace popu-
lations were grouped with the hybrid populations, refecting
similarity among these two groups of domesticated peppers.
The analysis with the STRUCTURE program and the
subsequent evaluation of the K statistic indicated that K = 2
is the most probable number of clearly diferentiated genetic
groups of wild and domesticated C. annuum in northwestern
Mexico. One group is formed by the wild populations and
the other by the landrace and hybrid populations (Fig. 4).
However, in this test, the Tequila landrace and the Chilaca
hybrid population presented evidence of introgression by the
wild relatives. The test of isolation by distance performed
only with the wild populations indicated a signifcant and
positive correlation between the geographic distance matrix
(Km) and the genetic distance (r = 0.476; P = 0.03) (Fig. 5).
Using the BARRIERS program, fve genetic barriers in the
distribution of the wild populations of C. annuum in north-
western Mexico were found with bootstrap-support greater
than 59% (Fig. 6). The frst barrier (Barrier 1) separated the
Table 2. Genetic variation parameters for wild, landrace, and
hybrid populations of Capsicum annuum studied: A, number
of alleles per locus; P, percentage of polymorphic loci; H
O
,
observed heterozygosity; and H
E
, expected heterozygosity.
Population A P H
O
H
E
Wild
Mocuzary 2.57 85.71 0.253 0.384
Estacin Cuaternaria 5.00 100.00 0.315 0.592
Potrero de Alcantar 3.86 85.71 0.197 0.533
Rancho el Coyote 3.14 100.00 0.255 0.446
Tehueco 3.29 85.71 0.238 0.461
Lode Vega 2.71 100.00 0.215 0.447
Pilares 4.00 100.00 0.196 0.509
Chapeteado 3.86 85.71 0.209 0.473
Alcoyonqui 4.71 71.43 0.246 0.452
Tabal 3.86 71.43 0.167 0.516
Otates 3.29 100.00 0.214 0.448
Llano 3.14 85.71 0.196 0.327
Average 3.62 89.29 0.225 0.466
Landraces
Tequila 3.00 85.71 0.263 0.449
Cascabel 3.29 71.42 0.168 0.358
Cola de Rata 3.83 83.33 0.172 0.459
Average 3.37 80.16 0.201 0.422
Hybrids
Hungaro 2.86 85.71 0.232 0.466
Guajillo 3.42 100.00 0.271 0.519
Anaheim 2.86 100.00 0.302 0.468
Cambray 4.00 100.00 0.179 0.473
Serrano 3.43 100.00 0.277 0.461
Poblano 2.00 66.67 0.111 0.191
Chilaca 3.00 100.00 0.289 0.504
Average 3.08 93.20 0.237 0.440
Table 3. Total (H
T
), within (H
S
), and between (D
ST
) population genetic diversity and the genetic differentiation coefcient (G
ST
)
for wild, landrace, and hybrid populations of Capsicum annuum.
Wild Landraces Hybrids
Locus H
T
H
S
D
ST
G
ST
H
T
H
S
D
ST
G
ST
H
T
H
S
D
ST
G
ST
EPMS480 0.216 0.189 0.027 0.123 0.574 0.166 0.408 0.710 0.606 0.391 0.215 0.355
EPMS501 0.878 0.681 0.197 0.224 0.649 0.628 0.022 0.033 0.769 0.583 0.185 0.241
EPMS342 0.771 0.549 0.223 0.289 0.461 0.284 0.177 0.383 0.644 0.339 0.305 0.474
EPMS417 0.885 0.714 0.171 0.194 0.821 0.685 0.136 0.166 0.861 0.650 0.210 0.244
EPMS643 0.843 0.595 0.248 0.294 0.798 0.627 0.171 0.214 0.700 0.386 0.314 0.448
EPMS440 0.237 0.229 0.009 0.037 0.650 0.412 0.238 0.366 0.625 0.504 0.120 0.192
EPMS919 0.776 0.548 0.229 0.295 0.582 0.415 0.168 0.288 0.868 0.595 0.272 0.314
Mean
SD
0.658
0.298
0.501
0.209
0.158
0.099
0.208
0.099
0.648
0.127
0.460
0.195
0.188
0.117
0.309
0.214
0.725
0.110
0.493
0.122
0.232
0.070
0.324
0.108
236 WWW.CROPS.ORG CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012
LLA population from the Sinaloa and Sonora populations.
The second barrier (Barrier 2) is located in the central part of
the state of Sinaloa, separating the TAB populations from the
CHA, PIL, and ALC populations. The third barrier (Barrier
3) is located between the OTA populations and the rest of
the populations present in Sinaloa and Sonora. The fourth
barrier (Barrier 4) separated the populations of the north
of Sinaloa and Sonora. Finally, the ffth barrier (Barrier 5)
is located between the CHA population and the PIL and
ALCpopulations.
Table 4. Wrights F statistics of wild, landrace, and hybrid Capsicum annuum populations.
Wild Landraces Hybrids
Locus F
IS
F
IT
F
ST
F
IS
F
IT
F
ST
F
IS
F
IT
F
ST
EPMS480 0.172*** 0.312*** 0.169*** 0.856*** 0.961*** 0.728*** 0.476*** 0.678*** 0.385***
EPMS501 0.604*** 0.713*** 0.274*** 0.522*** 0.556*** 0.071* 0.687*** 0.776*** 0.282***
EPMS342 0.651*** 0.809*** 0.454*** 0.500*** 0.712*** 0.425*** 0.064 0.549*** 0.518***
EPMS417 0.547*** 0.693*** 0.323*** 0.515*** 0.642*** 0.263*** 0.134 0.558*** 0.489***
EPMS643 0.479*** 0.696*** 0.416*** 0.466*** 0.767*** 0.564*** 0.402*** 0.689*** 0.480***
EPMS440 0.064 0.027 0.070 0.408*** 0.641*** 0.394*** 0.288*** 0.599*** 0.437***
EPMS919 0.475*** 0.673*** 0.377*** 0.525*** 0.682*** 0.331*** 0.698*** 0.812*** 0.378***
Mean
SD
0.412***
0.260
0.560***
0.283***
0.297***
0.138
0.542***
0.144
0.709***
0.129
0.396***
0.210
0.393***
0.249
0.666***
0.103
0.424***
0.082
*Signicant at the 0.05 probability level.
** Signicant at the 0.01 probability level.
***Signicant at the 0.001 probability level.
Table 5. Analysis of molecular variance (AMOVA) conducted on seven simple sequence repeat loci of wild, landrace, and hybrid
Capsicum annuum populations.
Sources of variation Df Sum of squares Variance components % Total P
Wilds vs. Landraces vs. Hybrids 2 17123.863 35.45589 10.07 P < 0.0001
Among populations 19 35880.797 55.41416 15.75 P < 0.0001
Within populations 624 162911.60 261.07628 74.18 P < 0.0001
Wilds
Among populations 11 19356.601 52.96179 19.75 P < 0.0001
Within populations 338 72752.001 215.24261 80.25 P < 0.0001
Landraces
Among populations 2 2131.91 21.2162 4.42 P < 0.0001
Within populations 83 38103.067 459.07309 95.58 P < 0.0001
Hybrids
Among populations 6 14392.286 71.40927 21.78 P < 0.0001
Within populations 203 52056.533 256.43612 78.22 P < 0.0001
Figure 2. Dendrogram based on Neis (1972) genetic distance, applying the unweighted pair-group method with arithmetic averaging
(UPGMA) clustering algorithm, between wild (W), landrace (L), and hybrid (H) populations of Capsicum annuum.
CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012 WWW.CROPS.ORG 237
DISCUSSION
In this study, we analyzed the variation and genetic struc-
ture of wild, landrace, and hybrid populations of C. ann-
uum of northwestern Mexico. The results indicated high
levels of genetic diversity both within and among the pop-
ulations of C. annuum. The genetic variation in our study
was greater than that found in previous studies of wild and
domesticated populations of the Capsicum genus in Mexico
that were performed with samples from germplasm banks
with isozyme markers (Loaiza-Figueroa et al., 1989).
The results of the current study more closely coincide
with previous studies performed on natural populations
with isozymes and random amplifed polymorphic DNA
(RAPD) (Hernndez-Verdugo et al., 2001; Oyama et al.,
2006). The genetic diversity of the domesticated peppers
as compared with the wild ones has slightly decreased at
the seven loci studied, both in terms of allelic diversity
and in terms of expected heterozygosity. However, hybrid
populations showed higher expected heterozygosity than
landraces. This could be a result of heterozygote forma-
tion during the breeding process of hybrid varieties. In
contrast, landraces are maintained through generations of
open pollination, including self-pollination. Among the
landrace populations, we found an average reduction of
8.2% in genetic diversity, and in the hybrid populations
we found a reduction of 10.3%. A similar pattern was
found with RAPDs (Oyama et al., 2006), with a genetic
erosion of only ~8%. Similarly, for three functional genes
(Dhn, G3ph, and Waxy), the domesticated peppers have
Figure 3. Associations between 22 Capsicum annuum populations revealed by principal coordinate analysis of Neis (1972)
geneticdistances.
Figure 4. (A) Mean and standard deviation in InP (D) for ve
independent runs of STRUCTURE plotted against the number of
genetic groups (K) used in the analysis. (B) Values of K plotted
against K. In both cases the peak indicates the most probable
number of genetic groups given the data.
Figure 5. A Mantels test for correlation between genetic distance
(Nei, 1972) and geographic distance (km) showed a moderate
correlation, r = 0.476, P < 0.030 from 10,000 randomizations.
238 WWW.CROPS.ORG CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012
retained 91% of the diversity found in wild C. annuum
peppers in Mexico (Aguilar-Melndez et al., 2009). This
reduction in genetic diversity among domesticated popu-
lations may be associated with the domestication process.
Our results show a positive coef cient of inbreeding for
all the populations, with the landrace populations presenting
the highest average value (0.542) and the hybrid populations
presenting the lowest average value (0.393). These levels of
inbreeding may be due to the reproduction system of this
self-pollinating species, which shows only 7.8 to 38.6% cross
pollination (Ballester and de Vicente, 1998). However, these
high to moderate levels of inbreeding may also be the product
of the limited movement of genes among populations (Elam,
1998), which by itself may cause signifcant substructuring
and an increase in inbreeding (Wright, 1943; Rohlf and Sch-
nell, 1971; Turner et al., 1982; Soakal and Wartenberg, 1983;
Hamrick and Loveless, 1986). In contrast to previous studies
of isoenzymes of wild and domesticated C. annuum in north-
western Mexico (Hernndez-Verdugo et al., 2001), where
the average values of genetic diferentiation were 0.056 for
the wild populations and 0.167 for the domesticated popula-
tions, in this study we found that the domestication process
has changed the distribution of the genetic variation of the
domesticated populations in comparison with the wild popu-
lations. In the wild populations, we found average G
ST
values
of 0.208 and F
ST
values of 0.287, while in the landraces pep-
pers we found an average G
ST
value of 0.309 and an average
F
ST
value of 0.396. Finally, we found greater diferentiation
among the modern pepper varieties (hybrids), with an aver-
age G
ST
value of 0.324 and an average F
ST
value of 0.424. It is
possible that this genetic diferentiation among domesticated
populations is increasingly more intense than that observed
among the wild populations. The average G
ST
values in this
study are similar to the average values obtained with RAPDs
(Oyama et al., 2006) among wild and domesticated popu-
lations of C. annuum in northwestern Mexico. Normally a
genetic diferentiation among populations (F
ST
) greater than
0.15 is considered to indicate considerable genetic diferentia-
tion (Frankham et al., 2002).
The diferences in the G
ST
averages among the isoen-
zymes and the DNA markers (RAPDs and microsatellites)
may be due to the variation of some enzymatic loci that are
not selectively neutral and may determine deviations in the
estimates of the parameters that describe the genetic structure
of the populations (Ferguson et al., 1998). On the basis of a
matrix of genetic distances (Nei, 1972), a dendrogram was
constructed with the UPGMA method. This dendrogram
separated the pepper populations into two groups, one group-
ing the wild populations and the other the cultivated pepper
populations, with the exception of the Tequila (landrace) and
Chilaca (hybrid) populations, which were grouped with the
wild populations. The principal coordinate analysis showed
tight clustering among the wild populations, while among
the landrace and hybrid populations there is greater separa-
tion, refecting greater genetic diferentiation among these
populations. An analysis was performed with the Bayesian
Figure 6. Map showing the distribution of Capsicum annuum used in the present study (abbreviations indicate the collection sites of
populations). Each pie chart represents the proportions in each population of the two genetic groups as assigned by the program
STRUCTURE. Red and green represent the genetic groups corresponding to wild and domesticated populations, respectively, and blue
lines represent calculated barriers based on genetic distances with the BARRIERS software.
CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012 WWW.CROPS.ORG 239
algorithm implemented by the STRUCTURE program,
with the goal of determining the most probable number
of genetic groups. This analysis produced a most probable
value of K = 2. However, this analysis also suggested various
degrees of mixing among the two main genetic groups (wild
and domesticated). The signifcance of Mantels test refected
isolation by distance in the structure of the wild populations
being studied, where the populations that are the most distant
from each other are genetically more diferent than popula-
tions that are geographically closer. The natural isolation by
distance results in limited gene fow, where the probability of
gene fow between two populations is a function of the geo-
graphic distance between them. In the selfng populations
of wild peppers, where little or no pollen fow is present,
the gene fow must be produced by the movement of seeds,
particularly through fruit consumption, with the seeds later
disseminated by birds (Laborde and Pozo-Campodnico
1982; Pozo-Campodnico et al., 1991; Vzquez-Dvila,
1996). The results obtained from the BARRIERS program
identifed important barriers that separate the majority of the
wild populations through the gradient of their distribution
with bootstrap support of >59%. These barriers may possibly
be due to the course of some riverbeds as well as to certain
changes in vegetation altitude and climate that may result in
some process of diferentiation among the populations.
CONCLUSIONS
The populations of C. annuum of northwest Mexico may
be classifed into two genetic groups. The frst group is
formed by wild populations and the second group is formed
by landraces and hybrids. The genetic variation within
the wild, landrace, and hybrid groups is high, although
when comparing the landrace and hybrid population with
the wild pepper populations, there is a slight reduction in
A (alleles per locus) and in H
E
(expected heterozygosity).
The genetic diferentiation among the domesticated pep-
per populations (C. annuum) is greater than that among
the wild populations.
Acknowledgments
We thank the National Council of Science and Technology,
CONACYT, Mxico, for fnancial support (research project
106129) as well as Universidad Autnoma de Sinaloa, project
PROFAPI 2008/084. We are grateful to Juan Pealoza Ramirez,
Ana Luisa Albarrn L, and Maria L. Herrera Arroyo for com-
puter support and suggested improvements to the manuscript.
References
Aguilar-Melndez, A., L. Peter, L. Mikeal, and K. Seung-Chul.
2009. Genetic diversity and structure in semiwild and domes-
ticated chiles (Capsicum annuum; Solanaceae) from Mexico.
Am. J. Bot. 96:11901202. doi:10.3732/ajb.0800155
Ballester, J., and M.C. de Vicente. 1998. Determination of F1
hybrid seed purity in pepper using PCR-based markers.
Euphytica 103:223226. doi:10.1023/A:1018372523343
Berthaud, J. 1997. Strategies for conservation of genetic resources
in relation with their utilization. Euphytica 96:112.
doi:10.1023/A:1002922220521
Casas, A., and G. Barbera. 2002. Mesoamerican domestication and
difusion. p. 143162. In P.S. Nobel (ed.) Cacti: Biology and
uses. The Univ. of California Press, Los Angeles, CA.
Conicella, C., A. Errico, and F. Saccardo. 1990. Cytogenetic and
isozyme studies of wild and cultivated Capsicum annuum.
Genome 33:279282.
Doebley, J. 1989. Isozymic evidence and evolution of crop plants.
p. 165191. In E.D. Soltis and P.M. Soltis (ed.) Isozymes in
Plant Biology. Discorides, Portland, OR.
Doebley, J. 1992. Molecular systematic and crop evolution. p. 202
222. In P.S. Soltis, D.E. Soltis, and J.J. Doyle (ed.) Molecular
systematic of plants. Chapman & Hall, New York, London.
Elam, D. 1998. Population genetics of vernal pool plants: The-
ory, data and conservation implications. p. 180189. In C.W.
Witham, E. Baunder, D. Belk, W. Ferren, and R. Ornduf (ed.)
Ecology, conservation and management of vernal pool ecosys-
tems. California Native Plants Society, Sacramento, CA.
Eshbaugh, W.H. 1980. The taxonomy of the genus Capsicum
(Solanaceae). Phytologia 47:153166.
Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the num-
ber of clusters of individuals using the software Structure:
A simulation study. Mol. Ecol. 14:26112620. doi:10.1111/
j.1365-294X.2005.02553.x
Excof er, L., G. Laval, and S. Schneider. 2005. Arlequin ver. 3.0:
An integrated software package for population genetics data
analysis. Evol. Bioinform. Online 1:4750.
Fri, M. 1986. Pepper (Capsicum annuum L.). p. 344362. In Y.P.S.
Bajaj (ed.) Biotechnology in Agriculture and Forestry 2.
Springer-Verlag, Berlin Heidelberg.
Ferguson, M.E., H.J. Newbury, N. Maxted, B.V. Ford-Lloyd, and
L.D. Robertson. 1998. Population genetic structure in Lens
taxa revealed by isozyme and RAPD analysis. Genet. Resour.
Crop Evol. 44:257269.
Frankham, R., J.D. Ballou, and D.A. Briscou. 2002. Introduction
to conservation genetics. Cambridge, UK.
Goudet, J. 1995. Fstat version 1.2: A computer program to calcu-
late F-statistics. J. Hered. 86:485486.
Haghnazari, A., R. Samimifard, J. Najaf, and M. Mardi. 2005.
Genetic diversity in pea (Pisum sativum L.) accessions detected
by sequence tagged microsatellite markers. J. Genet. Breed.
59:145152.
Hamrick, J.L., and M.D. Loveless. 1986. The infuence of seed dis-
persal mechanisms on the genetic structure of plant populations.
p. 211223. In A. Estrada and T.H. Fleming (ed.) Frugivores
and seed dispersal. Junk Publ., The Hague, the Netherlands.
Harlan, J.C. 1992. Crops and man, second ed. ASA and CSSA,
Madison, WI.
Hawkes, J.G. 1983. The diversity of crop plants. Harvard Univ.
Press, Cambridge, MA.
Heiser, C.B., and B. Pickersgill. 1975. Names for the bird peppers
(Capsicum-Solanaceae). Baileya 19:151153.
Hernndez-Verdugo, S., R.G. Guevara-Gonzlez, R.F. Rivera-
Bustamante, C. Vzquez-Yaes, and K. Oyama. 1998. Los
parientes silvestres del chile (Capsicum spp.) como recursos
genticos. Bol. Soc. Bot. Mex. 62:171181 (In Spanish).
Hernndez-Verdugo, S., P. Dvila, and K. Oyama. 1999. Snte-
sis del conocimiento taxonmico, origen y domesticacin del
gnero Capsicum. Bol. Soc. Bot. Mex. 64:6584 (In Spanish).
Hernndez-Verdugo, S., R. Luna-Reyes, and K. Oyama. 2001.
240 WWW.CROPS.ORG CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012
Genetic structure and diferentiation of wild and domesti-
cated populations of Capsicum annuum from Mxico. Plant
Syst. Evol. 226:129142. doi:10.1007/s006060170061
Huang, S., B. Zhang, D. Milbourne, L. Cardle, G. Yang, and
J. Guo. 2000. Development of pepper SSR markers from
sequence databases. Euphytica 117:163168.
Hunziker, A.T. 1979. South American Solanaceae: A synoptic sur-
vey. p. 4985. In J.K. Hawkes, R.L. Lester, and A.D. Skelding
(ed.) Biology and taxonomy of Solanaceae. Linnean Society
Symposium, Series, no. 7. Academic Press, New York.
Ince, A.G., M. Karaca, and A. Naci. 2010. Polymorphic microsat-
ellite markers transferable across Capsicum species. Plant Mol.
Biol. Rep. 28:285291. doi:10.1007/s11105-009-0151-y
Laborde, C.J., and O. Pozo-Campodnico. 1982. Presente y
pasado del chile en Mxico. Publicacin Especial Num. 85.
SARH-INIA. Mxico.
Lee, J.M., S.H. Nahm, Y.M. Kim, and B.D. Kim. 2004. Char-
acterization and molecular genetic mapping of microsatellite
loci in pepper. Theor. Appl. Genet. 108:619627. doi:10.1007/
s00122-003-1467-x
Lefebvr, V., A. Palloix, and M. Rives. 1993. Nuclear RFLP
between pepper cultivars (Capsicum annuum L.). Euphytica
71:189199. doi:10.1007/BF00040408
Lefebvr, V., B. Gof net, J.C. Chauvet, B. Caromel, P. Signoret,
R. Brand, and A. Palloix. 2001. Evaluation of genetic distances
between pepper inbred lines for cultivar protection purposes:
Comparison of AFLP, RAPD and phenotypic data. Theor.
Appl. Genet. 102:741750. doi:10.1007/s001220051705
Lefort, F., and G.C. Douglas. 1999. An ef cient micro-method of
DNA isolation from mature leaves of four hardwood tree spe-
cies Acer, Fraxinus, Prunus and Quercus. Ann. For. Sci. 56:259
263. doi:10.1051/forest:19990308
Lewis, P.O., and D. Zaykin. 2001. Genetic data analysis: Computer
program for the analysis of allelic data. Version 1.0 (d16c).
Free program distributed by the authors over the Internet
from http://lewis.eeb.uconn.edu/lewishome/software.html
(verifed 27 Oct. 2011).
Li, C.C., and D.G. Horvitz. 1953. Some methods of estimating
inbreeding coef cient. Am. J. Hum. Genet. 5:107117.
Loaiza-Figueroa, F., K. Ritland, C.J. Laborde, and S.D. Tanks-
ley. 1989. Patterns of genetic variation of the genus Capsi-
cum (Solanaceae) in Mexico. Plant Syst. Evol. 165:159188.
doi:10.1007/BF00936000
MacNeish, R.S. 1964. Ancient mesoamerican civilization. Science
143:531537. doi:10.1126/science.143.3606.531
Manni, F., E. Gurard, and E. Heyer. 2004. Geographic patterns
of (genetic, morphologic, linguistic) variation: How barriers
can be detected by Monmoniers algorithm. Hum. Biol.
76:173190. doi:10.1353/hub.2004.0034
McLeod, M.J., S.I. Guttman, W.H. Eshbaugh, and R.E. Rayle.
1983. An electrophoretic study of evolution in Capsicum
(Solanaceae). Evolution 37:562574. doi:10.2307/2408269
Minamiyama, Y., M. Tsuro, and M. Hirai. 2006. An SSR-based
linkage map of Capsicum annuum. Mol. Breed. 18:157169.
doi:10.1007/s11032-006-9024-3
Miller, M.P. 1997. Tool for Population Genetic Analyses (TFPGA)
version 1.3. A Windows program for the analysis of allozyme
and molecular population genetic data.
Monmonier, M. 1973. Maximum-diference barriers: An alternative
numerical regionalization method. Geogr. Anal. 3:245261.
Nagy, I., A. Stgel, Z. Sasvri, M. Rder, and M. Ganal. 2007.
Development, characterization and transferability to other
Solanaceae of microsatellite markers in pepper (Capsicum ann-
uum L.). Genome 50:668688. doi:10.1139/G07-047
Nei, M. 1972. Genetic distance between populations. Am. Nat.
106:283292. doi:10.1086/282771
Nei, M. 1978. Estimation of average heterozygosity and genetic dis-
tance from a small number of individuals. Genetics 89:583590.
Oyama, K., S. Hernndez-Verdugo, C. Snchez, A. Gonzlez-
Rodrguez, P. Snchez-Pea, J.A. Garzn-Tiznado, and A.
Casas. 2006. Genetic structure of wild and domesticated pop-
ulations of Capsicum annuum (Solanaceae) from northwestern
Mexico analyzed by RAPDs. Genet. Resour. Crop Evol.
53:553562. doi:10.1007/s10722-004-2363-1
Peakall, R., and P.E. Smouse. 2006. GENALEX 6: Genetic anal-
ysis in Excel. Population genetic software for teaching and
research. Mol. Ecol. Notes 6:288295. doi:10.1111/j.1471-
8286.2005.01155.x
Perry, L., and V. Kent. 2007. Precolumbian use of chili peppers
in the valley of Oaxaca, Mxico. Proc. Natl. Acad. Sci. USA
104:1190511909. doi:10.1073/pnas.0704936104
Pickersgill, B., C.B. Heiser, and J. McNeill. 1979. Numerical tax-
onomic studies on variation and domestication in some spe-
cies of Capsicum. p. 679700. In J.G. Hawkes, R.N. Lester,
and A.D. Skelding (ed.) The biology and taxonomy of the
Solanaceae. Academic Press, London.
Pickersgill, B. 1984. Migration of chili peppers, Capsicum spp., in
the Americas. p. 105123. In D. Stone (ed.) Papers of the Pea-
body Museum of Archeology and Ethnology, Vol. 76. Har-
vard Univ. Press.
Pickersgill, B. 1997. Genetic resources and breeding of Capsicum
spp. Euphytica 96:129133. doi:10.1023/A:1002913228101
Portis, E., I. Nagy, Z. Sasvri, A. Stgel, L. Barchi, and S. Lantery.
2007. The design of Capsicum spp. SSR assays via analy-
sis of in silico DNA sequence, and their potential utility for
genetic mapping. Plant Sci. 172:640648. doi:10.1016/j.
plantsci.2006.11.016
Pozo-Campodnico, O., H. Montes, and J. Redondo. 1991. Chile
(Capsicum spp.). p. 217238. In Avances en el estudio de los
recursos ftogenticos de Mxico, Sociedad Mexicana de
Fitogentica A.C. Mexico. (In Spanish).
Prince, J.P., E. Pochard, and S.D. Tanksley. 1992. Restriction frag-
ment length polymorphism and genetic distance among Mex-
ican accessions of pepper. Genome 36:404417. doi:10.1139/
g93-056
Prince, J.P., V.K. Lackney, C. Angeles, J.R. Blauth, and M.M.
Kyle. 1995. A survey of DNA polymorphism within the genus
Capsicum and the fngerprinting of pepper cultivars. Genome
38:224231. doi:10.1139/g95-027
Pritchard, J.K., M. Stephens, and P. Donnelly. 2000. Inference of
population structure using multilocus genotype data. Genet-
ics 155:945959.
Rohlf, F.J., and G.D. Schnell. 1971. An investigation of the
isolation-by-distance model. Am. Nat. 105:295324.
doi:10.1086/282727
Soakal, R.R., and D.E. Wartenberg. 1983. A test of spatial auto-
correlation analysis using an isolation-by-distance model.
Genetics 105:219237.
Stewart, C.N., and L. Excof er. 1996. Assessing population
genetic structure and variability with RAPD data: Applica-
tion to Vaccinium macrocarpon (American Cranberry). J. Evol.
Biol. 9:153171. doi:10.1046/j.1420-9101.1996.9020153.x
Turner, M.E., J.C. Stephens, and W.W. Anderson. 1982. Homo-
zygosity and patch structure in plant populations as a result
CROP SCIENCE, VOL. 52, JANUARYFEBRUARY 2012 WWW.CROPS.ORG 241
of nearest-neighbor pollination. Proc. Natl. Acad. Sci. USA
79:203207. doi:10.1073/pnas.79.1.203
Vzquez- Dvila. M.A. 1996. El amash y el pistoqu: Un ejemplo
de la etnoecologa de los chontales de Tabasco, Mxico. Etno-
ecolgica 3:5963.
Votava, E., B. Jit, and Bosland. 2005. Genetic diversity of chile
(Capsicum annuum var. annuum L.) landraces from northern
New Mexico, Colorado, and Mexico. Econ. Bot. 59:817.
doi:10.1663/0013-0001(2005)059[0008:GDOCCA]2.0.CO;2
Workman, P., and J. Niswander. 1970. Populations studies on
southwestern Indians tribes. Local genetic diferentiation in
the Papago. Am. J. Hum. Genet. 22:2449.
Wright, S. 1943. Isolation by distance. Genetics 28:114138.
Zeven, A.C. 1998. Landraces: A review of defnitions and classifca-
tions. Euphytica 104:127139. doi:10.1023/A:1018683119237