Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 24 January 2006
Published in Crop Sci 46:428-436 (2006)
© 2006 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rice, E. B.
Right arrow Articles by Kresovich, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Rice, E. B.
Right arrow Articles by Kresovich, S.
Agricola
Right arrow Articles by Rice, E. B.
Right arrow Articles by Kresovich, S.
Related Collections
Right arrow Maize
Right arrow Plant Genetic Resources
Right arrow Crop Genetics

PLANT GENETIC RESOURCES

Conservation and Change: A Comparison of In situ and Ex situ Conservation of Jala Maize Germplasm

Elizabeth B. Ricea,*, Margaret E. Smitha, Sharon E. Mitchellb and Stephen Kresovichc

a Dep. of Plant Breeding and Genetics, Cornell Univ., Ithaca, NY 14853
c Dep. of Plant Breeding and Genetics and Institute for Genomic Diversity, 130 Biotechnology Building, Cornell Univ., Ithaca, NY 14853
b Institute for Genomic Diversity, 151 Biotechnology Building, Cornell Univ., Ithaca, NY 14853

* Corresponding author (ebr6{at}cornell.edu)


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Conservation of agricultural genetic resources, whether in situ (in farmers' fields) or ex situ (in a germplasm repository), provides variation for breeding and selection efforts. In this study, we assayed levels of genetic diversity from in situ and ex situ populations of Jala, a very tall, large-eared race of maize (Zea mays L. subsp. mays), using 22 simple sequence repeat (SSR) loci. For comparison, we also analyzed diversity in reference populations of other maize races and wild relatives of maize, the teosintes (both Zea spp. and Z. mays subsp.). As expected, both the in situ and ex situ Jala populations were not as diverse as the teosintes (gene diversity, He, {approx} 0.60 and 0.7, respectively), but they were surprisingly more diverse than the populations representing other maize races (He = 0.55). The older ex situ Jala populations were less diverse and more differentiated than more recently collected accessions. Despite this fact, we observed similar levels of overall genetic diversity in the ex situ and in situ populations (He = 0.62 and 0.61, respectively), and little differentiation between these groups (Fst = 0.01). Therefore, the ex situ Jala collections, as a whole, contained the diversity found today in the field, even though diversity was under-represented in individual repository populations.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
IN A WORLD of increasing population, changing cultures, globalization, and emigration of indigenous communities, the loss of genetic diversity from farmers' fields in centers of origin is a subject of major concern (Brush, 1999; FAO, 1997). To preserve the genetic resources found in agricultural systems, two strategies are employed. Ex situ conservation captures genetic material, usually seed, and protects it in a germplasm repository. In contrast, in situ conservation allows evolutionary processes to continue by preserving varieties in farmers' fields, under farmer management. These two strategies are usually seen as complementary (Altieri and Merrick, 1987; Brush, 1991).

Ex situ conservation is frequently perceived as a static process, but genetic change can occur (Wood and Lenne, 1997). Cold storage of accessions is stable, with little, if any, genetic change occurring (Roberts, 1975). However, stored seed is not viable indefinitely and must be regenerated. To produce fresh seed true to its parental genetic composition and as genetically diverse as possible, a population is planted and seed is produced using controlled pollination where necessary. During this process, genetic change can occur because of both genetic drift and selection in the regeneration environment, a phenomenon commonly referred to as "genetic shift." Population genetic models establish clear guidelines for maintaining maximum diversity during regeneration (Crossa, 1989). Similar models (Crossa et al., 1993, 1994) also guide collection methods. It is important to note that a population conserved ex situ can never be more diverse than the genetic material originally collected in the field. However, in the early years of ex situ conservation, the emphasis of collection focused on phenotypic diversity—conserving varieties, and perhaps less critically, genetic diversity within them. As knowledge of the molecular genetics of individuals and populations has grown, the goals and methods of ex situ conservation of cross-pollinated plant species have shifted subtly to focus on capturing and maintaining maximum allelic diversity within varieties and populations (Wang et al., 2004).

In situ, or on-farm, conservation allows crops to evolve dynamically with farmers' needs and the changing environment. However, it is relatively risky, as crops may be lost to a host of environmental and economic influences. Increasingly, studies use interdisciplinary approaches to understand the complex socioeconomic, agronomic, and genetic decisions involved in farmers' seed management [for a detailed example, see Bellon et al. (2003) and Pressoir and Berthaud (2004)]. Though few studies have compared the effects of in situ and ex situ conservation, Parzies et al. (2000) found a decline in genetic diversity within barley (Hordeum vulgare L. subsp. vulgare) accessions related to length of ex situ storage and regeneration, while Soleri and Smith (1995) found phenotypic differences between varieties of Hopi maize conserved in situ and ex situ.

In the study presented here, we compared in situ and ex situ conservation of the Mexican maize race Jala as assessed by simple sequence repeat (SSR) loci. Specifically, we (i) determined the amount of genetic diversity in Jala populations compared with reference populations of other maize races and teosintes, (ii) investigated the partitioning of diversity in maize (Zea mays L.) and its implications for conservation of genetic diversity, (iii) characterized genetic dynamics within and between farmers' fields, and (iv) compared the genetic outcomes of in situ and ex situ conservation of Jala. As such, we highlight a specific case study of conservation of a maize race, both in farmers' fields and the germplasm repository.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Plant Material
The populations studied (Table 1) comprised eight populations each of Jala, a maize race, collected both from farmers' fields and germplasm repositories, and a diverse sampling of other maize races and teosintes. Field populations were collected in November 1999 from the Jala valley (Nayarit, Mexico) by taking one kernel from each of 20 to 60 ears. The Jala variety, with its very tall plants (up to 5 m) and very large ears (up to 45 cm of grain) is distinctive enough to ensure comparison of the same variety, past and present. The ex situ Jala populations were collected between 1944 and 1988 from the same location as the in situ populations and held at the International Maize and Wheat Improvement Center (CIMMYT) in Mexico City, Mexico. For comparisons of genetic diversity and genetic differentiation, we used populations that represented the breadth of diversity (as assessed by SSRs) found both in other maize races and close wild relatives, the teosintes (Matsuoka et al., 2002b).


View this table:
[in this window]
[in a new window]
  		Table 1. In situ Jala, ex situ Jala, accessions representing maize races, and teosinte populations studied, with seed source and catalog numbers.

 
Sampling
Twenty-four individuals were sampled from each population except for Zea mays subsp. mexicana c, where only 22 individuals were assayed. Crossa et al. (1993) showed that the sample size (n) required to retain at least one copy of k alleles present at frequency p at m loci with probability P should be larger than:

Formula

With our sampling strategy (n = 24, m = 22 loci, and k = 5 alleles), alleles with frequency p > 0.144 can be detected with P = 95% confidence. On a per locus basis (m = 1 and k = 5), alleles with frequency p > 0.087 can be detected with P = 95% confidence and rare alleles (p = 0.05) can be detected with P = 65.8% confidence.

DNA Extraction and SSR Data
For each population in Table 1, we extracted DNA from 24 individual plants, 10 to 20 d post-germination, using a CTAB method (Doyle and Doyle, 1987). We used 22 fluorescently labeled primer pairs to amplify SSR loci (Table 2). SSR primer sequences are available from the Maize Genetics and Genomics Database (MaizeGDB; http://www.maizegdb.org/ssr.php; verified 20 September 2005). The 22 SSR loci were widely distributed throughout the ten maize chromosomes (Table 2) with approximately one locus per chromosome arm.


View this table:
[in this window]
[in a new window]
  		Table 2. SSR markers used to evaluate Jala, maize races, and teosintes, with associated map bin, core repeat, allele size range, number of alleles, and percent missing data.

 
Collection and Analysis of SSR Data
Polymerase chain reactions (PCRs) were performed as described by Matsuoka et al. (2002a), with one modification. To facilitate allele scoring, addition of adenine deoxyriboside to the 3' end of PCR products by Taq polymerase was promoted by increasing the final extension time at 72° from 10 min to 1 h. Genotyping was performed on automated DNA sequencers (models 377XL or 3730XL, Applied Biosystems, Foster City, CA) and data were analyzed with either Genotyper or GeneMapper softwares (Applied Biosystems), depending on the collection platform. Loci that failed to amplify or showed persistent three- or four-banded patterns were not scored (i.e., treated as missing data).

Population Parameters
Unbiased Fst (Weir and Cockerham, 1984) measures divergence between populations on the basis of differences in allele frequencies. Values for Fst range from 0 (completely undifferentiated) to 1 (completely differentiated). Populations with little divergence have Fst values less than 0.05 while moderately differentiated populations have Fst values between 0.05 and 0.15, greatly differentiated populations have Fst values between 0.15 and 0.25, and very greatly differentiated populations have Fst values greater than 0.25 (Hartl and Clark, 1997). We used Genepop 3.2 (Raymond and Rousset, 1995) to calculate pair-wise Fst values between populations and population allele frequencies. Confidence intervals for Fst values (1000 bootstraps) were estimated by PowerMarker 3.23 (Liu and Muse, 2004). GDA 1.1 (Lewis and Zaykin, 2002; Ni et al., 2002) was used to estimate allele number (An), and expected heterozygosity (He), also called gene diversity. Because population sample sizes were comparable, allele number (An) was a fairly unbiased measure of diversity. Gene diversity (He), a commonly used measure of diversity, reflects both allele number and allele frequency. The analysis of molecular variation (AMOVA) was done by Arlequin 2.0 (Schneider et al., 2000) and linear regressions were performed in Excel (Microsoft).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY
 REFERENCES
 
The range of diversity in maize can be organized hierarchically into maize races, varieties, and seed lots. A race will contain many varieties with similar traits. The seed a farmer plants is called a seed lot and is one representation of a variety (Louette, 1999). A maize variety has a set of distinguishable characteristics: e.g., tall stalk, white grain, floury texture, red silks. "Traditional" varieties, also known as "landrace" varieties, are those grown by farmers for many generations, usually from saved seed. "Improved" varieties are usually produced by a seed company for commercial release. To ensure comparison of the same variety, past to present, this study focused on Jala, a unique variety of maize from the town of Jala in the Mexican state of Nayarit. Jala maize is extremely tall (up to 5 m) and bears very long ears (up to 45 cm). It appeared in the literature as early as 1924 (Kempton, 1924) and has been the target of a campaign to promote in situ conservation (Listman and Estrada, 1992). Jala is both a variety, planted by farmers and a unique maize race.

SSR Diversity in All Populations
For the 22 SSR loci surveyed, 274 alleles were present in 790 individuals sampled from all populations (Jala, other maize races, and teosintes) (Table 2). All SSR loci were polymorphic, with the number of alleles per locus ranging from 2 to 24 (average 12.5 alleles per locus) (Table 2). The average proportion of missing data across all loci and individuals was 3.7% (Table 2), with smaller proportions of missing data in maize populations (average 1.5% for Jala, both in situ and ex situ, and other maize races) and larger proportions in the teosintes (average 10.5%)(Table 3). Average allele number per locus ranged from 2.5 to 6.5, with the teosintes generally having more alleles per locus than maize (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 3. Descriptive information across 22 SSR loci organized by population category, with category averages and totals.

 
Jala Diversity Compared with Populations Representing Teosinte and Other Maize Races
Maize was most likely domesticated from teosinte, probably Z. mays subsp. parviglumis, about 9000 yr ago (Doebley, 1990; Matsuoka et al., 2002b). The strong selection pressures experienced during domestication should lead to a reduction in genetic diversity in the domesticate (maize) compared with the wild relative (teosinte). Therefore, it was not surprising that alleles found in the teosinte populations comprised 90% of the total alleles identified. On the other hand, all the maize populations, including Jala and other races, had only 54 to 58% of the total alleles. Consistent with a domestication bottleneck, nearly all unique alleles (52 of 60, or 87%) were observed in the teosintes (Table 3). The teosinte populations were highly differentiated (Fst = 0.25) (Table 4) and more diverse (overall He = 0.73) than the maize populations (He ranged from 0.55 to 0.62) (Table 3). As the teosintes included different Zea mays subspecies and Zea species, this result was not surprising.


View this table:
[in this window]
[in a new window]
  		Table 4. Fst measures of population differentiation{dagger} within and among the teosintes, maize races, ex situ Jala, and Jala from farmers' fields based on 22 SSR loci.

 
Like the teosintes, the populations representing maize races (excluding Jala) were highly differentiated (Fst = 0.23) (Table 4). This result probably reflected the fact that we intentionally sampled races that represented a broad spectrum of genetic diversity in maize. Although highly differentiated, the populations representing maize races, themselves maintained ex situ, were significantly less diverse than the Jala ex situ populations (He = 0.55 and 0.62, respectively; p ≤ 0.0001) (Table 3). The lower diversity in populations representing maize races might be due to a small founder population and/or ex situ management history. If the populations were often regenerated to maintain seed stocks, for example, they may have experienced more genetic drift than the less-regenerated ex situ Jala populations. Unfortunately, regeneration histories were not available. On the other hand, the lower variation in populations representing maize races might reflect natural levels of variation, and ex situ Jala populations may be unexpectedly diverse. Additional analyses of several accessions per race and comparison with additional races would be required to better understand levels of diversity.

In Situ Jala Populations
Genetic diversity in populations of Jala from farmers' fields was consistent with the high levels of diversity found in ex situ Jala accessions (He = 0.61 and 0.62, respectively) (Table 3). The in situ Jala populations showed very little, if any, differentiation from one another; 17 of 28 pair-wise comparisons between farmers' fields (61%) included zero in their 95% confidence interval (Table 5). An AMOVA analysis showed the vast majority (94%) of the genetic variation in Jala accessions (both in situ and ex situ) can be explained by differences among individuals, rather than by same-year differences among fields (4%) or by year of collection (2%). These results were consistent with recent studies showing high levels of genetic diversity at the field-level for cross-pollinated species like pearl millet [Pennisetum glaucum (L.) R. Br.] (Busso et al., 2000) and maize (Pressoir and Berthaud, 2004), as well as predominantly inbreeding species like sorghum [Sorghum bicolor (L.) Moench] (Dje et al., 1999).


View this table:
[in this window]
[in a new window]
  		Table 5. Pair-wise Fst values for Jala populations,{dagger} ex situ and in situ, with history of collection and regeneration, based on 22 SSR markers.

 
Current similarity among populations in situ Jala could be attributed to (i) a recent common seed source, (ii) genetic similarity due to a historic event causing a small effective population size (Ne) for Jala (e.g., reduced seed supply or small planted area), or (iii) pollen or seed mixing among farms. It is doubtful that a common seed source or small Ne were responsible for the observed similarity among populations. Farmers' accounts described long periods of time managing the same Jala seed lot without replacement or supplementation [average of 26 yr, though some younger farmers purchased Jala seed each year from older farmers (Rice, 2004)]. Also, Jala itself was highly diverse (Table 3), even though the variety accounted for only 5% of the maize harvested and was planted in widely scattered, small areas (average 1.14 ha per farmer, range 0.06–3.00 ha) (Rice, 2004). The effective population size (Ne) for Jala, therefore, is apparently large. Pollen and seed flows, on the other hand, have been important for low levels of maize varietal differentiation elsewhere in Mexico (Pressoir and Berthaud, 2004). Although several recent studies have concluded that distances of 200 m were sufficient to isolate maize plots from pollen flow (Luna et al., 2001; Ma et al., 2004; Stevens et al., 2004), the average Jala plot was long, narrow, and small (1.14 ha) with a high proportion of its area within a few rows of a different plot of maize. Therefore, it is likely that geneflow occurred among plots. Geneflow into maize plots shows exponential decline with distance from the pollen source with rates as high as 62% cross-pollination in border rows and up to 16% in the thirteenth row. Gene transfer into advanced generations of improved varieties from the Jala valley also indicated strong pollen flow (Rice, 2004). Given these accounts, the low incidence of farmers replacing their Jala seed, and the current genetic data, it seems likely that pollen mixing was more important than seed mixing for the lack of genetic differentiation among Jala farmers' seed lots.

Ex Situ Jala Populations
As a group, the ex situ Jala populations were diverse (Table 3) and showed moderate differentiation (Table 4). As a whole, these accessions had as many alleles as the in situ Jala populations (about 7.2 alleles) (Table 3). Differentiation was more pronounced among the older ex situ populations (particularly 1944 populations) and less differentiation was observed among more recently collected populations (Table 5).

On the basis of the lack of genetic differentiation observed among the in situ populations, one might hypothesize that the Jala valley functions as a united genepool and predict that ex situ Jala populations collected in the same year would be more like one another than those collected in different years. Indeed, pair-wise Fst comparisons showed that the three ex situ populations collected in 1988 were most similar (Fst = 0.03–0.04) (Table 5). A 1988 collection of a different traditional variety from the Jala valley, Tampiqueño, also showed similar, low levels of differentiation from the Jala 1988 populations (Fst ranged from 0.02–0.05). Notably, populations FF5 and 1988a (Fst = 0.03) came from the same farmer who stated he had not replaced, supplemented, or lost his seed in the intervening years. The FF5 population showed less differentiation from other Jala fields in the same year than from the same seed lot sampled in 1988. This accession was no more different from the farmer's own 1988 population than from other 1988 populations. These data support the hypothesis that year of collection is a stronger predictor of genetic similarity than seed source for these Jala populations.

The same premise does not hold true for the older ex situ populations. The two populations collected in 1951 were moderately differentiated (Fst = 0.07) and the two populations collected in 1944 were the two most highly differentiated Jala populations (Fst = 0.17). The 1944 populations also had fewer alleles and lower gene diversity, while the more recently collected ex situ populations were more diverse, both in terms of gene diversity (He) and number of alleles (An), as shown by regression analysis (for all loci, p value for He ≤ 0.000, R2 = 0.62; p value for An ≤ 0.000, R2 = 0.72). The differentiation observed between the 1944 accessions could either reflect true differences among populations in the field in 1944 or could result from collection bottlenecks or genetic drift during ex situ management.

Though we have no way of measuring differentiation of Jala fields in the past, historical patterns of Jala cultivation make high levels of differentiation unlikely. In the 1940s and 1950s, Jala was planted on the majority of the Jala valley, while today it represents only 5% of the valley's maize area (Rice, 2004). Given these facts, in the 1940s and 1950s, there would have been more Jala pollen and more Jala–Jala cross-plot pollination, thus leading to genetic similarity in a given year. Therefore, it seems likely that genetic drift has affected these ex situ populations, a problem predicted by models (Lacy, 1987) and seen in other studies (Parzies et al., 2000). Today the maize regeneration process is carefully managed to maintain genetic diversity, maximize effective population size and minimize genetic drift (Crossa, 1989; Crossa et al., 1993, 1994). However, very little is known about how these accessions were regenerated in the past. The 1988, 1951, and 1944 accessions were each regenerated two or three times (Table 5). The number of regenerations did not appear to be a good predictor of allele number, gene diversity, or year of collection [regressions all not significant (p values > 0.10, data not shown)], suggesting that regeneration method, rather than the number of regenerations, may be the critical factor affecting these populations. A single regeneration that sampled too few gametes could have powerful long-term effects by reducing the effective population size for the repository population. Additionally, we know very little about how the ex situ populations were collected.

Comparison of In Situ and Ex Situ Jala
The patterns of diversity and differentiation of in situ and ex situ Jala populations reinforce the possibility of genetic drift during repository maintenance, especially for the older accessions, but are also consistent with the changing historical patterns of Jala cultivation. As a group, the ex situ populations were nearly as diverse as and little differentiated from Jala in farmers' fields (Tables 3 and 4). These results demonstrate that the ex situ Jala collections, as a whole, contained the diversity found today in the field, even though diversity is under-represented in individual ex situ populations. The Jala populations from 1944 are more genetically similar to in situ Jala and to other ex situ Jala accessions than they were to each other (Table 5)—relationships consistent with genetic drift in the repository and suggestive that the 1944 differences are not due to misclassification. The populations of in situ Jala showed little differentiation from the most recently collected ex situ populations from 1988 (Fst ranged from 0.03–0.04; Table 5). The genetic similarity between the 1988 accessions and in situ Jala probably reflects low genetic drift, likely attributable to good ex situ sampling and regeneration procedures as well as sampling of a genetically similar maize environment.

The low genetic diversity of older ex situ populations and higher diversity of recent ones were also consistent with historical patterns of Jala cultivation (assuming that cross-pollination of the Jala variety with non-Jala varieties increased the number of alleles and gene diversity found in Jala). The 1944 accessions show the least diversity of the ex situ populations. In 1944, Jala was the predominant variety grown in the valley and would have had few other varieties with which to cross-pollinate. Populations from the 1950s and 1960s, when traditional varieties began to arrive from other parts of Mexico, showed intermediate levels of diversity (Table 3). The 1988 Jala ex situ populations were the most diverse. By 1988, the combination of cultivated maize varieties was similar to today, with many improved varieties (albeit different ones from those planted today), many traditional varieties from other parts of Mexico, and reduced areas of the traditional Jala variety (Rice, 2004). With the data at hand, it is impossible to discriminate changes that occurred in the germplasm repository from changes that occurred in the field. The extent to which Jala has changed from introgression of new alleles (whether from traditional or improved varieties) is obscured by the potential for genetic change because of collection method or drift in the ex situ accessions.

The increased differentiation in the Jala ex situ populations with time (Table 5) raises two important concerns: (i) loss of alleles that were present in the 1940s and 1950s and (ii) "contamination" by new alleles from new varieties in the valley. However, our data do not show strong supporting evidence for either concern. The common alleles [frequency ≥0.05; Marshall and Brown (1975)] are invariant over time; 92% (77 of 83) of the common alleles before 1953 were also common in farmers' fields. The same was true for the high frequency alleles (≥0.35), where 86% (15 of 18) of the high frequency alleles before 1953 were also the high frequency alleles in farmers' fields. There are a few "rare" alleles (<5%) sampled in the older populations that are not detected in more modern populations. Also there are "rare" alleles sampled in modern populations that were not detected in older populations. As these are low frequency alleles, however, we do not know if their appearance or disappearance reflects sampling limitations or true changes in the population.

Conservation Target
To capture maximum genetic diversity with minimum effort, collection efforts should target alleles common in local areas but rare elsewhere (Marshall and Brown, 1975). By using the populations found in farmers' fields in Jala as an example of locally distributed alleles and the alleles found in the maize race populations as a proxy for widespread alleles, we can roughly quantify the target for conservation (Table 6). These results are limited by the precision of our sampling method: rare alleles, with the given sampling, are only captured with 66% confidence, while common alleles, the target of interest, will be captured with higher confidence (95% if allele frequency, p > 0.144). Therefore, this estimate is likely to represent an upper limit of the local common alleles of interest. The target for conservation—the local, common alleles—constituted less than 10% of the alleles in this study. As a by-product of conservation targeted to common locally distributed alleles, many of the rare, local alleles would be also captured, as well as widespread alleles, whether common or rare.


View this table:
[in this window]
[in a new window]
  		Table 6. Evaluation of the target for conservation: common and rare allele frequencies compared with widespread and local allele distribution for Jala data.

 

   SUMMARY
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
REFERENCES
 
Conservation Implications
Our SSR data indicated that Jala has been effectively preserved in both the field and in the germplasm repository. There was strong identity between the in situ and ex situ collections with little differentiation between these groups (Fst = 0.01), and the most common alleles were the same in both groups. Genetic differences were observed, however, among the ex situ populations. Older accessions were less diverse and more differentiated than newer acquisitions, and the most recent collections showed the most similarity to the in situ populations. This result suggests that plant breeders utilizing repository collections should assay many accessions to capture the diversity present.

For both the in situ and ex situ material, we observed strong genetic similarity and little differentiation between Jala populations collected in the same year. This observation implies that the Jala valley comprises a single genetic population. Therefore, by extension, when collecting a germplasm repository population, the critical sampling issue is to collect many individuals, rather than collecting a few individuals from many fields.


   ACKNOWLEDGMENTS
 
We thank Joanne A. Labate, Julie C. Ho, Susan R. McCouch, the USDA-PGRU group at Geneva, NY, and two anonymous reviewers for their comments and suggestions. Suketoshi Taba, at CIMMYT provided invaluable support of this project. John Doebley and Major Goodman helped us select diverse maize and teosinte populations. Thanks also to Anna Herforth, J. Arahón Hernández Guzmán, Francisco Rivas, Israel Ruvalcalva, Aquiles Carballo Carballo, and Charolotte Achayara for help with data collection.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Supported by Cornell International Institute for Food, Agriculture, and Development (CIIFAD), International Maize and Wheat Improvement Center (CIMMYT), and NSF-RTG in Conservation and Sustainable Development. This work was submitted to Cornell Univ. by E.B. Rice in partial fulfillment of the requirements for the Ph.D. degree.

Received for publication June 13, 2005.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rice, E. B.
Right arrow Articles by Kresovich, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Rice, E. B.
Right arrow Articles by Kresovich, S.
Agricola
Right arrow Articles by Rice, E. B.
Right arrow Articles by Kresovich, S.
Related Collections
Right arrow Maize
Right arrow Plant Genetic Resources
Right arrow Crop Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome