Cucurbit Genetics Cooperative
Other Crop Genetics Cooperatives
Home About Membership Reports Gene Lists Conferences Links Search NCSU
Cucurbit Genetics Cooperative Report 17:1-5 (article 1) 1994

A Core Collection for Cucumber: To Be or Not To Be

J.E. Staub

Vegetable Crops Research, USDA/ARS, Department of Horticulture, University of Wisconsin-Madison, WI 53706 U.S.A.

In the recent past there have been discussions among scientists interested in germplasm management regarding the definition and construction of core collections for agronomic and horticultural crop species (Brown 1989a,b; Frankel and Brown, 1984). A core collection might be defined as a subset of a larger array of germplasm accessions (collection). A core collection attempts to maintain the majority of diversity present in a larger (whole) collection, and thus provides for the potential efficient access to the possible range of variation found in the larger collection.

A core collection should be constructed on the basis of unique descriptive determinants (characters or traits) so that the core is clearly defined in quantitative or qualitative terms. Thus, "the core concept entails identifying a range of accessions within a collection, the total of which include, with an acceptable level of probability and with minimum redundancy, most or much of the range of genetic diversity in the crop species and its relatives" (National Research Council, 1991). The U.S. Germplasm Resources Information Network (GRIN) now contains information on the evaluation of about 750 accessions for disease resistance (7 economically important diseases), cold, drought and heat tolerance, yield (as measured by F1 combining ability), various morphological characteristics (passport and evaluation data), and biochemical variation (14 isozyme loci).

There has been some debate over the applicability of the core concept to the U.S. cucumber collection. Discussions have occurred at a number of U.S. Cucurbit Crop Advisory Committee meetings and at the annual meetings of the American Society for Horticultural Science. It has been thought that large collections such as those of corn, soybeans and wheat might benefit from the use of core collections (National Research Council, 1991). Nevertheless, it has been suggested that the application of a core collection for these crops be handled cautiously. It is feared that the management of the larger collection might be compromised by exclusive maintenance of core collections. Furthermore, core collections, if not properly constructed, might misrepresent the true variability present in the collection.

Hundreds of national and international requests are made to the U.S. National Plant Introduction system each year for seed samples of cucumber accessions. Given that the U.S. cucumber collection is relatively small (1000+ accessions) and that it may be possible to devise a core collection for cucumber, the question is whether such a core should be constructed. Therefore, I conducted a survey in 1992 intended to ascertain the general feeling among U.S. cucumber researchers regarding the potential need of a cucumber core collection. If the feelings were relatively positive my intent was then to gather information which might be useful in construction a core collection. In order to reach my objectives, I surveyed 54 U.S. cucumber researchers inquiring as to: 1) their depth of knowledge regarding the Core Concept and its applicability for the management of germplasm collections; 2) their feelings regarding the construction of a core collection for cucumber; 3) whether such a core should be based on the frequency of inquiry to the GRIN system and/or regional plant introduction stations for a particular trait; and, 4) cucumber (given those found presently in GRIN) and how they would prioritize such criteria. This survey was not meant to be comprehensive or definitive, but rather to provide a springboard for further discussion.

There were 24 respondents to my questionnaire (54 surveyed). Of these one indicated that he was not knowledgeable about the Core Concept, 16 felt they were fairly knowledgeable, and seven regarded themselves as highly knowledgeable. While eight respondents felt that the Core Concept was applicable to cucumber, 12 felt it might have value. Some of their comments are provided in Table 1. Fully 23 respondents felt that the establishment of a core should not be based on the frequency to the GRIN system and/or regional introduction stations for a particular trait. The respondents differed in their perception of what criteria might form a core for cucumber (Tables 2 and 3). These results suggest that there is a desire and need for continued discussion regarding the Core Concept as it might apply to cucumber.

Table 1. Some comments from respondents to the question of whether the core concept was applicable to cucumber.

In general, my feeling is that it is not appropriate for cucumber at this time. The number of cucumber introductions is quite modest, in comparison with cereals and some other crops, and it should not be too much of a burden to maintain most, if not all, of them. In many cases, we have insufficient information to eliminate a cucumber introduction from the core. An exception is that of duplications. There are many cases known of the same cultivar assigned two or more PI numbers, and of multiple accessions at the NSSL of the same cucumber cultivars. I suggest that all but one of the duplications be eliminated from the core. Cucumber collection size is marginal, at this time for full benefit of core concept. The cantaloupe collection probably is large enough. Cucurbita species are probably large enough if all are considered together. Citrullus spp. probably is large enough. When other appropriate data are not available, geographic origin has been very useful in developing core for alfalfa, lentils, chickpeas, and Phaseolus beans.
Collections are to be limited both in number and in geographic representation. Reduction in time required to incorporate PI germplasm into commercial lines by using the most valuable material.
One of the best breeders in Egypt requested 1,000 C. Sativus accessions and, not having received them, requested my help in getting from PI Station, Ames. I cannot support such a request. A representative subset is what he should get initially. I would use the germplasm collection in helping to find traits of interest that are not in adapted germplasm, usually this would be rare or unusual traits. Limiting the number of accessions that are very active may limit the utility of the collection.
The core concept could work with large, well-documented collections. While the core concept may have great applications for agronomic crops, how often do we get requests in cucumber for more than 20 accessions from any one person?
My impression is that for some cucurbits the total number of accessions is not that high compared to wheat or corn. Is the number of cucumber accessions low enough, after deleting duplications, that it could function as a core? The core collection should be used as a genetic tool or window to view the diversity represented in a given collection. For some traits the whole collection can be screened, while for others, screening a core could provide direction and save resources at the same time. It must be noted, however, that developing a "core collection" for a crop will generate more than less germplasm maintenance problems.
Evaluation and utilization would be more efficient and more appropriate plant collection proposals could be prepared. In cases where the collection is large and represents a broad spectrum of genetic diversity for the crop and there is a high proportion of duplication, I think the core concept may have merit if carefully implemented.
Permits in-depth study of representative cultigens. I don't see the core collection so much as a management practice. I see the core collection more useful as a tool in evaluation work, seed requests for representative samples of germplasm collections, etc. Many times we receive requests for entire PI collections from people who have little knowledge of the NPGPS. Core collections would be helpful in these situations.

Table 2. Mean and standard deviation of rankings of criteria that might be considered in the establishment of a core collection for cucumber.

Respondent1
  1 2 3 4 5 6 7 8 9 10 11 12 13 143 15 16 17 18 19 20

 

(mean)

 

SD
 
1
1
1
2
1
4
1
1
1
3
1
3
2
1
3
2
3
1
1
2
1.8
1.0
Cold
5
2
3
4
3
5
5
6
5
3
32
5
5
1
4
3
4
4
2
5
3.9
1.3
Yield
6
6
5
6
5
6
3
2
6
6
6
4
1
5
1
5
5
5
6
4.7
1.7
Morphological
2
3
4
1
2
2
2
5
3
2
4
1
1
2
6
2
2
3
3
2.6
1.3
Biochemical
3
5
7
3
4
3
6
4
4
2
2
2
3
1
1
4
1
3
4
1
3.1
1.5
Suggest
5
4
2
6
1
6
5
6
4
3.5
1.9

Drought

4

Insect

4
3

Unique trait

24
Geographical origin
1
1
1

1 Some respondents did not complete question 4.
2 Respondent stated. "Considering the resources involved, this seems less useful and much more expensive."
3 Respondent felt that all parameters were equally important.
4 Trait not specified.

Table 3. Summary of comment regarding the construction of a core collection in cucumber.

Frequency of inquiry to PI stations would reflect past use, especially for PIs referred to in publications. To appropriately define a core one would need complete knowledge of all future needs for a crop. Each user of PI collections asks different questions each time they use the collection. If I'm interested in fruit size I'll scan descriptor data on fruit size and select from that basis. If I'm interested in disease resistance when the disease only occurs in Nepal, I may only evaluate accessions for Nepal and adjacent regions, etc. For a CAC or other crop experts to define one core collection would mean that the rest of the collection is less important. Unless a committee is set up to define a core for each project which would like a core, the core concept cannot work as theorized. Untill we know much more about each of our collections, probably choosing random accessions evenly distributed over a geographic range will best suffice as cores. Our experience is that PIs from similar locations are more likely to share an attribute of interest for breeding. Thus, geographical distribution of accessions, and selection within an area based on collection in different habitats, is probably the most efficient way to preserve diversity that we maynot be able to identify as yet. I'd suggest also gross morphological differences are much more significant than single gene traits such as white spine. In my opinion, the use of combining ability in this contest is useless.
Instead of defining core collections it would be more productive to a) eliminate duplications, b) obtain good data on a wide range of high priority descriptors (as determined by CAC) and c) train potential PI users on PI information and GRIN use. I suggest you redefine your quest as one to develop 'test arrays' of C. sativus which would be useful for various breeding/ genetic/research purposes. For example, a disease resistance array, an ecogeographic origin array, an American Slicer aray, etc. Maximize phenotypic diversity with each array but build around your target objecives. the 'core concept/ requires accuracy in assessing genetic diversity and I don't believe we're close to knowing that well enough.
Aim should be to maximize diversity. I can visualize disease resistance and stress tolerance as either favorable or unfavorable dependingon how they are used. As a way to come up with accessions to include in a Core, I think an advisory group consisting of the NPGS cucumber curator, members of the CCAC, CGC, and other breeders and geneticists should be selected to represent a wide array of perspectives.
  Should develop data base for cucumber PIs, then analyze diversity. Choose one PI from each group to make 40-80 top cultigens.

Literature Cited

  1. Brown, A.H.D. 1989a. The care for core collections. p. 136-156. In: A.H.D. Brown, O.H. Frankel, D.R. Marshall, and J.T. Williams (ed.). The Uses of Plant Genetic Resources. Cambridge University Press, New York.
  2. Brown, A.H.D. 1989b. Core collections: A practical approach to genetic resources management. Genome 31:818-824.
  3. Frankel, O.H. and A.H.D. Brown. 1984. Plant genetic resources today: A critical appraisal. P. 249-257. In: J.W. Holden and J.T. Williams (ed.). Crop Genetic Resources: Conservation and Evaluation, Allen and Unwin, London.
  4. National Research Council. 1991. Prescription for effectiveness. p.124-126. In: Board on Agriculture, National Academy of Sciences. Managing Global Genetic Resources: The U.S. National Plant Germplasm System. National Academy Press, Washington, D.C.
Home About Membership Reports Gene Lists Conferences Links Search NCSU
Department of Horticultural Science Box 7609North Carolina State UniversityRaleigh, NC 27695-7609919-515-5363
Page citation: Wehner, T.C., Cucurbit Genetics Cooperative;
Created by T.C. Wehner and T. Ng, 1 June 2005; design by C.T. Glenn;
send questions to T.C. Wehner; last revised on 11 December, 2009