PNL Volume 17 1985 GENE BANKS 89
Marx, G. A. and P. McInnis
The collection of genetic stocks of peas maintained in the
Department of Horticultural Sciences at Cornell University's Experiment
Station in Geneva has been growing steadily since it was started in
1959. Comprehensive and diverse, the collection contains lines from
various other collections, e.g. the Weibullsholm and USDA Plant
Introduction collections, and from individual workers throughout the
world. It also contains wild and primitive forms, some varieties,
breeding lines, and material from other miscellaneous sources. But by
far the largest part of the collection comprises 1ines derived from over
5,000 crosses made over the last 25 years or more. Some 75,000 seed
packets are now on hand. Included are multi-marker and multi-resistant
stocks, isolines, groups of linked genes, etc. Over the years, many
hundreds of seed samples have been distributed to serve the varied needs
of users from many countries. This collection, hereinafter referred to
as the Geneva collection, is separate and distinct from the collection
of USDA Plant Introduction (P.I.) accessions, maintained nearby at the
Northeast Regional Plant Introduction Station, Geneva, New York.
The importance of and need for maintaining the Geneva collection is
generally recognized, but the difficulty of organizing the collection to
make it readily accessible is not well appreciated. The Geneva
collection is a working collection as opposed to a museum collection.
It is therefore dynamic, with new lines being added regularly as a
result of crosses made among lines already present and with new lines
from outside sources. This dynamicism complicates record keeping and
data retrieval. All data have been assiduously collected and manually
recorded in books over a 25-year period. It is essential of course to
link the identifying number of a given lino with the appropriate genetic
and descriptive data for that line. Since, alter a cross is made, the
identifying number for each progeny which descends from a given cross
changes with each succeeding generat ion, and since the data collected in
each generation may be slightly different from the generation before
(until the line is genetically fixed), each new cross generates a wealth
of information which must be evaluated and collated.
Heretofore, all record keeping was done manually. For example,
each generation (3 per year) requires a new planting plan. Recorded in
that plan, in addition to the current Identification number, is all the
information pertaining to the history ofeach entry, i.e. previous
identification number, cross number, identity of parents, pedigree, and
relevant genetic or descriptive data. To trace specific lineages a
whole series of data books had to be individually consulted for the
identification numbers used in successively earlier generations.
Moreover, it was only possible to trace backward from a given point, not
forward. A particular cross may produce dozens, even hundreds, of
entries after many generations. Searching for genetic or descriptive
data was still more difficult.
PNL Volume 17 1985
Although gratifying progress has been made, much more work is
required before the collection can be considered to be organized and
ready for routine use. Considerably more data need to be entered into
the computer. When a core collection of the most useful entries has
been established, a continuing effort must be made to add to or refine
the information already at hand. Stocks must be propagated to maintain
their viability and, as each accession is grown for that purpose,
genetic and descriptive data not previously noted will be added. A
catalog of available Pisum stocks in the collection is not presently
available. Such a catalog is needed if we and others are to benefit
fully from the resources which have been diligently and systematically
developed over nearly three decades.
Since October, 1983, the USDA-ARS has provided critical financial
support for organizing and maintaining the Geneva collection, and the
task is now underway. The collection has become a part of the National
Plant Germplasm System (NPGS), but control and direction remain with the
curator (GAM). Our goal is to develop a cross-indexed, multi-referenced
"core" collection of stocks consisting of 4000 to 5000 genetically
defined lines. To this end, more than thirty computer programs have
been written to accomplish various retrieval objectives.
Hosted by uCoz