U.S. Department of Agriculture: Information on the Condition of the
National Plant Germplasm System (Chapter Report, 10/16/97,
GAO/RCED-98-20).

Pursuant to a congressional request, GAO surveyed the 680 members of the
40 crop germplasm committees (CGC) for their views on the sufficiency of
the National Plant Germplasm System's (NPGS) principal activities
focusing on: (1) acquiring germplasm to ensure the diversity of the
collections in order to reduce crop vulnerability; (2) developing and
documenting information on germplasm; and (3) preserving germplasm.

GAO noted that: (1) just over half of the CGSs reported that the genetic
diversity contained in NPGS' collections is sufficient to reduce the
vulnerability of their crops; (2) considering both this collection and
all other freely available collections, almost three-quarters of the
committees said that the diversity in these collections is sufficient
for reducing their crops' vulnerability; (3) at the same time, the
committees identified several concerns affecting the diversity of their
collections, and they ranked the acquisition of germplasm as the highest
priority for the germplasm system if more funding becomes available; (4)
current acquisition efforts are hindered by problems in obtaining
germplasm from some countries and by the Department of Agriculture's
(USDA) management of the quarantine system, which has contributed to the
loss of germplasm and delays in its release for certain plants; (5)
according to the crop committees, many of the system's collections lack
sufficient information on germplasm traits to facilitate the germplasm's
use in crop breeding; (6) officials of the germplasm system acknowledged
that some information on plant traits, such as resistance to disease or
plant structure, has not been developed because it is considered to be a
lower priority than preserving germplasm; in other instances, the
information has been developed by scientists outside of the system and
has not been provided for entry into the database; (7) preservation
activities--viability testing, regeneration, and the long-term backup
storage of germplasm--have not kept pace with the preservation needs of
the collections; (8) only minimal viability testing--testing the seeds
in a sample to determine how many are alive in order to prevent the loss
of the sample--has occurred at two of four major locations; (9) in
addition, the system has significant backlogs for regenerating (that is,
replenishing) germplasm at the four major locations; and (10) over
one-third of the system's germplasm is not stored in the system's
secure, long-term storage facility, thereby increasing the risk that
samples located around the nation could be lost through environmental
damage or other catastrophes.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-98-20
     TITLE:  U.S. Department of Agriculture: Information on the 
             Condition of the National Plant Germplasm System
      DATE:  10/16/97
   SUBJECT:  Agricultural research
             Agricultural programs
             Surveys
             Plants (organisms)
             Biological research
             Agricultural production
             Genetics
IDENTIFIER:  National Plant Germplasm System
             USDA Germplasm Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Congressional Committees

October 1997

U.S.  DEPARTMENT OF AGRICULTURE -
INFORMATION ON THE CONDITION OF
THE NATIONAL PLANT GERMPLASM
SYSTEM

GAO/RCED-98-20

Information on Germplasm System

(150713)


Abbreviations
=============================================================== ABBREV

  APHIS - Animal and Plant Health Inspection Service
  ARS - Agricultural Research Service
  CGC - Crop Germplasm Committee
  GRIN - Germplasm Resources Information Network
  NGRL - National Germplasm Resources Laboratory
  NPGS - National Plant Germplasm System
  NSSL - National Seed Storage Laboratory
  USDA - U.S.  Department of Agriculture

Letter
=============================================================== LETTER


B-277794

October 16, 1997

Congressional Committees

This report provides information on the U.S.  Department of
Agriculture's National Plant Germplasm System.  This system provides
germplasm that is used by plant breeders and researchers to develop
new and improved plant varieties for crop production.  To conduct our
work, we surveyed the members of the 40 Crop Germplasm Committees
that advise the Department on the acquisition, preservation, and
information needs of the germplasm collections. 

We are sending copies of this report to interested congressional
committees; the Secretary of Agriculture; and the chairmen of the 40
germplasm committees surveyed.  We will also make copies available
upon request. 

If you or your staff have any questions, please call me at (202)
512-5138.  Major contributors to this report are listed in appendix
V. 

Robert A.  Robinson
Director, Food and
 Agriculture Issues

Congressional Committees

The Honorable Richard G.  Lugar
Chairman
The Honorable Tom Harkin
Ranking Minority Member
Committee on Agriculture, Nutrition,
 and Forestry
United States Senate

The Honorable Mitch McConnell
Chairman
The Honorable Patrick J.  Leahy
Ranking Minority Member
Subcommittee on Research, Nutrition,
 and General Legislation
Committee on Agriculture, Nutrition,
 and Forestry
United States Senate

The Honorable Robert F.  (Bob) Smith
Chairman
The Honorable Charles W.  Stenholm
Ranking Minority Member
Committee on Agriculture
House of Representatives

The Honorable Larry Combest
Chairman
The Honorable Calvin M.  Dooley
Ranking Minority Member
Subcommittee on Forestry, Resource
 Conservation, and Research
Committee on Agriculture
House of Representatives


EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

The U.S.  agricultural sector--renowned for its productivity--owes
much of its success to a continuing flow of improved crop varieties
that produce higher yields and better withstand pests, diseases, and
extreme climates.  The genes necessary for these crops are contained
in plant germplasm--the material in seeds or other plant parts that
controls heredity.  To maintain high levels of agricultural
productivity, plant breeders need access to an ample supply of
germplasm with diverse genetic characteristics. 

The U.S.  Department of Agriculture's (USDA) National Plant Germplasm
System (NPGS) maintains germplasm collections for over 85 crops at
sites nationwide.  Forty Crop Germplasm Committees (CGC) provide
technical advice and guidance to NPGS on germplasm activities.  The
CGCs are composed of crop experts, including the NPGS curators who
are responsible for maintaining and preserving the collections. 

Because of the importance of germplasm to U.S.  agricultural
productivity and food security, GAO surveyed the 680 members of the
40 CGCs for their views on the sufficiency of NPGS' principal
activities--(1) acquiring germplasm to ensure the diversity of the
collections in order to reduce crop vulnerability, (2) developing and
documenting information on germplasm, and (3) preserving germplasm. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

NPGS is primarily a federally and state-supported effort aimed at
maintaining supplies of plant germplasm with diverse genetic traits
for use in breeding and scientific research.  The diversity in
germplasm collections enables breeders to develop improved crops that
are more productive and often less vulnerable to pests and diseases. 
These collections are particularly important because the diversity of
germplasm worldwide has been reduced by several factors, such as the
widespread use of genetically uniform crops in commercial agriculture
and the destruction of natural habitats that have been important
sources of germplasm. 

The Agricultural Marketing Act of 1946 established the main
components of NPGS as well as a legal basis for federal and state
cooperation in managing plant genetic resources.  NPGS' current
organizational structure--a geographically dispersed network of
germplasm collections administered primarily by USDA's Agricultural
Research Service (ARS)--emerged in the early 1970s.  NPGS maintains
about 440,000 germplasm samples for over 85 crops at 22 sites
throughout the country and in Puerto Rico; almost half of these
samples are maintained at four regional plant introduction stations. 
Germplasm samples are held in crop collections, each of which
generally includes four types of germplasm (for example, germplasm
from cultivated plants and germplasm from wild relatives of
cultivated plants).  Each type of germplasm contains genetic material
that plays an important role in the collections' overall diversity. 

Most of NPGS' germplasm is imported from other countries and must
comply with U.S.  quarantine regulations, which are intended to
prevent the introduction of pests and pathogens into the United
States.  Germplasm collections are also maintained by other countries
and international organizations, as well as by U.S.  and foreign
universities and private companies.  These collections vary
considerably in terms of the quality of preservation, and only some
are freely available to breeders. 

Although ARS provides the lion's share of support for NPGS, the
system is also supported by the states.  Private industry also funds
selected NPGS projects and transfers germplasm to the public in the
form of new plant varieties and hybrids.  In fiscal year 1996, NPGS'
total funding was $23.3 million, $19.5 million of which was provided
by ARS.  From fiscal years 1992 through 1996, ARS' funding of NPGS
has declined by 14 percent, in constant dollars, while the total size
of the collections has increased by 10 percent. 


   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

Just over half of the Crop Germplasm Committees reported that the
genetic diversity contained in the National Plant Germplasm System's
collections is sufficient to reduce the vulnerability of their crops. 
Considering both this collection and all other freely available
collections, almost three-quarters of the committees said that the
diversity in these collections is sufficient for reducing their
crops' vulnerability.  At the same time, the committees identified
several concerns affecting the diversity of their collections, and
they ranked the acquisition of germplasm as the highest priority for
the germplasm system if more funding becomes available.  Current
acquisition efforts are hindered by problems in obtaining germplasm
from some countries and by USDA's management of the quarantine
system, which has contributed to the loss of germplasm and delays in
its release for certain plants. 

According to the crop committees, many of the system's collections
lack sufficient information on germplasm traits to facilitate the
germplasm's use in crop breeding.  Officials of the germplasm system
acknowledged that some information on plant traits, such as
resistance to disease or plant structure, either has not been
developed or has not always been entered into the system's database. 
In some cases, the information has not been developed because it is
considered to be a lower priority than preserving germplasm; in other
instances, the information has been developed by scientists outside
of the system and has not been provided for entry into the database. 

Preservation activities--viability testing, regeneration, and the
long-term backup storage of germplasm--have not kept pace with the
preservation needs of the collections.  Only minimal viability
testing--testing the seeds in a sample to determine how many are
alive in order to prevent the loss of the sample--has occurred at two
of four major locations.  In addition, the system has significant
backlogs for regenerating (that is, replenishing) germplasm at the
four major locations.  Finally, over one-third of the system's
germplasm is not stored in the system's secure, long-term storage
facility, thereby increasing the risk that samples located around the
nation could be lost through environmental damage or other
catastrophes. 


   PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4


      IMPORTANCE OF INCREASING
      DIVERSITY UNDERSCORED, BUT
      SOME OBSTACLES HINDER
      ACQUISITION
-------------------------------------------------------- Chapter 0:4.1

Over half of the CGCs reported that the genetic diversity of NPGS'
collections for their crops is sufficient to reduce crop
vulnerability.  Moreover, when all freely available collections
(including NPGS') were considered, almost three-fourths of the CGCs
reported that the collections--including those for many major
crops--are sufficiently diverse.  Nonetheless, the acquisition of
germplasm was viewed as NPGS' top priority--out of 14
germplasm-related activities--in the event of additional funding. 
Many CGCs identified concerns affecting the diversity of their
collections that may contribute to the importance they place on
increased acquisition.  These include inadequate diversity in one or
more of the four types of germplasm making up the collections and the
potential loss of germplasm that is at risk in nature. 

Although CGCs want to acquire more germplasm, difficulties with some
countries prevent such acquisition.  Furthermore, the Convention on
Biological Diversity has the potential to restrict NPGS' acquisition
of germplasm if its signatories make the germplasm subject to certain
restrictions that are inconsistent with NPGS' policy. 

The acquisition of germplasm has also been hampered by USDA's
management of the quarantine process, which has contributed to the
loss or delayed release of certain germplasm.  Thirteen CGCs, most of
whose germplasm often undergoes more intensive scrutiny in
quarantine, reported that quarantine regulations and processes have
resulted in delays in the timely release of germplasm; 5 CGCs
reported problems with the release of viable germplasm.  CGCs for
crops such as prunus (e.g., cherry and peach trees), apples, pears,
potatoes, and corn were among those reporting quarantine-related
problems. 


      GERMPLASM INFORMATION IS
      REPORTED TO BE INSUFFICIENT
-------------------------------------------------------- Chapter 0:4.2

Most CGCs reported that NPGS' germplasm collections for their crops
lack important information on germplasm traits needed for crop
breeding.  Breeders need such information to select germplasm with
the traits they are seeking from the myriad of germplasm samples. 
Specifically, three-quarters of the CGCs reported insufficiencies
with evaluation information, which describes traits, such as
resistance to disease and yield, that are of particular interest to
plant breeders.  Furthermore, almost half found insufficiencies in
characterization information, which describes traits, such as color
and plant structure, that are little influenced by the environment. 
On the other hand, most CGCs reported that passport information is
sufficient for crop-breeding purposes.  Passport information
describes, among other things, the site of origin of the germplasm. 

Some evaluation and characterization information has not been
developed and entered into NPGS' database for a number of reasons. 
These reasons include the large amount of germplasm that needs to be
evaluated and characterized, the resource-intensive nature of these
activities, and limited resources.  In addition, most evaluations of
NPGS' germplasm are conducted by scientists outside of NPGS--often
university and other ARS scientists--who do not always provide NPGS
with the resulting information for entry into the database.  CGCs
estimated that, on average, 50 percent of the useful evaluation
information relating to their NPGS collections is not in the
database.  Characterization information, on the other hand, is
primarily developed by NPGS' curators.  However, NPGS officials said
that characterizing germplasm is generally a lower priority than
preserving it.  Unlike evaluation and characterization information,
passport information should be provided when a sample is donated to
NPGS.  However, many samples lack some passport information, largely
because donors do not always have or provide the information. 


      PRESERVATION ACTIVITIES HAVE
      NOT KEPT PACE WITH THE
      COLLECTIONS' NEEDS
-------------------------------------------------------- Chapter 0:4.3

Preservation activities--including viability testing, regeneration,
and secure, long-term backup storage of germplasm--have not kept pace
with the preservation needs of NPGS' collections.  Two major NPGS
sites, accounting for over one-quarter of the active collections, do
not conduct sufficient viability testing to determine the quantity of
viable seeds, according to NPGS data and officials.  Viability
testing should generally be conducted every 5 to 10 years at these
sites, depending on the type of plant and storage conditions,
according to the site managers.  However, in 10 years, the two sites
have tested less than one-fourth of their germplasm. 

Furthermore, NPGS has significant backlogs of germplasm requiring
regeneration--growing the seeds in order to produce a sufficient
supply of viable germplasm.  For example, at one site that
distributes a wide variety of germplasm, about half of its over
60,000 samples required regeneration, and one collection could take
as much as 75 years to regenerate, given the current level of
resources.  NPGS officials said that limited staff resources were the
biggest problem contributing to these backlogs. 

Finally, only 61 percent of NPGS' approximately 440,000 seed samples
are backed up in the system's secure, long-term storage facility,
designed to minimize the loss of germplasm viability.  A primary
reason for the lack of backup is that sites do not provide germplasm
to this facility when a germplasm sample has too few seeds.  In such
instances, the sample must be regenerated before it can be backed up. 
Furthermore, as of August 1997, NPGS' secure, long-term facility had
a 16-month backlog of about 27,000 samples that have to be tested for
viability before being placed in permanent, long-term storage. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

GAO is making no recommendations in this report. 


   AGENCY COMMENTS
---------------------------------------------------------- Chapter 0:6

GAO provided a draft of this report to USDA for review and comment. 
USDA did not take issue with any of the information in the report. 
USDA noted that while NPGS has made large strides since earlier
reviews conducted by GAO and the National Research Council, its
successes have been dwarfed by its increasing responsibilities in the
face of declining resources.  USDA stated that unless NPGS' funding
is augmented, the system will need to juggle its multiple, sometimes
divergent, priorities by making incremental progress in addressing an
exceptionally broad range of user demands.  In addition, USDA said
that the Department would continue to work with other agencies and
the private sector to ensure that NPGS is managed effectively.  USDA
included an attachment to its comments highlighting the progress made
since 1990 in addressing NPGS' managerial goals. 

GAO appreciates the challenges that NPGS faces in juggling its
multiple priorities and managing its increasing collections in the
face of declining resources.  In that regard, GAO supports USDA's
efforts to improve the management of NPGS to make the most effective
use of its limited resources.  GAO believes that the information
provided in this report will assist congressional and other
decisionmakers in future deliberations on the role of NPGS and the
resources available to NPGS for carrying out its role.  Appendix IV
contains the complete text of USDA's comments and GAO's response. 


INTRODUCTION
============================================================ Chapter 1

The U.S.  agricultural sector--renowned for its productivity--owes
much of its success to a continuing flow of improved crop varieties
that produce higher yields and better withstand pests, diseases, and
climate extremes.  The genes necessary for these improved crops are
contained in plant germplasm--the material in seeds or other plant
parts that controls heredity.  To maintain a high level of
agricultural productivity, plant breeders need access to an ample
supply of germplasm with diverse genetic characteristics so that they
can continue to develop plant varieties that will provide increased
yields and better resist pests, diseases, and environmental stresses. 
However, the diversity of germplasm available to present and future
generations of breeders has been reduced by several factors,
including the widespread use of genetically uniform crops in
commercial agriculture and the destruction of natural habitats, such
as forests, that have been important sources of germplasm. 

In the United States, the National Plant Germplasm System (NPGS),
primarily administered by the U.S.  Department of Agriculture (USDA),
maintains germplasm collections for over 85 crops at 22 sites
nationwide and in Puerto Rico.  These collections contain numerous
germplasm samples\1 and provide breeders with access to germplasm
with a broad range of genetic traits.  In addition to maintaining the
collections, NPGS is responsible for acquiring germplasm, developing
and documenting information that describes the germplasm in the
collections, and distributing germplasm to plant breeders and other
users in the United States and worldwide. 


--------------------
\1 A germplasm sample (sometimes referred to as an accession) is a
distinct, uniquely identified sample of seeds or plants that is part
of a germplasm collection. 


   GERMPLASM COLLECTIONS ARE
   CRITICAL TO AGRICULTURAL
   PRODUCTIVITY, FOOD SECURITY,
   AND BIODIVERSITY
---------------------------------------------------------- Chapter 1:1

Germplasm collections are an important source of genetic material for
plant breeders targeting specific traits, such as higher yield,
increased resistance to disease and pests, good taste, improved
nutritional quality, and environmental and climatic hardiness.  To be
of greatest use, these collections need to be genetically diverse,
thereby giving breeders more possibilities to find the traits they
need to develop improved crop varieties.  In addition, information on
germplasm traits and other related information (e.g., site of origin
of the germplasm) should be obtained and documented, and the
germplasm must be adequately preserved to be of optimal use to
potential users. 

Diverse germplasm has played a key role in increasing food security
through enhanced crop productivity and reduced crop vulnerability to
pests and diseases.  For example: 

  -- According to a survey on the use of germplasm in 18 crops grown
     in the United States from 1976 to 1980,\2 from 1 percent (sweet
     clover) to 90 percent (sunflower and tomato) of the crop
     varieties had been improved in part by the use of germplasm from
     wild relatives of the cultivated crops. 

  -- The high productivity of modern wheat--resistant to many pests,
     diseases, and other stresses--results from combining germplasm
     from various varieties of wheat grown around the world to create
     improved wheat varieties.  For example, one well-known germplasm
     sample from Turkey has been a source of resistance for three
     different types of disease--common bunt, stripe rust, and snow
     mold.  This germplasm also has the ability to establish vigorous
     seedlings in hot, dry soils that deter the emergence of many
     other varieties.\3

  -- Most of the genes for insect and disease resistance in tomatoes
     come from a related wild species\4 that originated outside of
     the United States.  Germplasm from wild species is also a source
     of tolerance to environmental stress, such as drought.  In
     particular, the discovery of resistance to a soil-borne organism
     known as the root-knot nematode has made the difference between
     growing or not growing tomatoes in many subtropical areas of the
     United States (such as southern California and Florida). 

In addition to providing a source of genetic diversity for plant
breeders, germplasm collections serve as an archive for rare and
endangered crop species.  The loss of biodiversity worldwide has made
the need for these collections all the more compelling.  Expanding
human populations, urbanization, deforestation, destruction of the
environment, and other factors threaten many of the world's plant
genetic resources.  These resources are vital to the future of
agricultural productivity and the world's food security.  Many
national and international collections have been established to
rescue and conserve these resources for future use. 

In breeding plant germplasm into a narrowing genetic base of highly
productive crop varieties, breeders have also reduced the genetic
diversity of these crops, making them more uniform.  Genetic
uniformity in breeding also results when breeders inadvertently
eliminate certain traits (such as resistance to disease and pests)
that do not contribute directly to the desired characteristic (such
as high yield) for which they were searching.  While the resulting
genetic uniformity can offer substantial advantages in both the
quantity and quality of a commercial crop, it can also make crops
more vulnerable to pests, diseases, and environmental hazards.\5 A
narrow genetic base presents the potential danger of substantial crop
loss if a crop's genetically uniform characteristics are suddenly and
adversely affected by disease, insects, or poor weather.  The risk of
loss through the genetic vulnerability of uniform, common-origin
planted crops is a serious concern. 

Such losses have occurred in the past.  The Irish potato famine of
the 1840s was a major factor in the death, impoverishment, and
emigration of millions of Irish people.  A single variety of the
potato became Ireland's staple food after its arrival from South
America in the eighteenth century.  The widespread use of this single
variety increased the potato crop's vulnerability to a previously
unknown blight, which devastated a number of successive potato
harvests.  While the United States has not experienced such a
widespread loss, several sizable crop failures have occurred as a
result of a crop's vulnerability to a particular disease.  For
example, in the late 1950s and early 1960s, about 70 percent of the
wheat crop in the Pacific Northwest was wiped out by a disease known
as stripe rust.  In 1970, a disease known as the southern corn leaf
blight swept from the southeastern United States to the Great Plains,
costing farmers 15 percent of their corn crop that year. 


--------------------
\2 Managing Global Genetic Resources:  Agricultural Crop Issues and
Policies, National Research Council (Washington, D.C.:  National
Academy Press, 1993). 

\3 Cox, T.S., "The Contribution of Introduced Germplasm to the
Development of U.S.  Wheat Cultivars," Use of Plant Introductions in
Cultivar Development, Part 1, CSSA Special Publication No.  17, 1991. 

\4 A wild species is one that has not been subject to breeding to
alter it from its state. 

\5 Increased vulnerability can occur because genetically similar
varieties or hybrids of a crop create a dependence on a single
genetic source of resistance.  Insects and pathogens are continually
evolving, and in genetically uniform crops, the pest may need to
overcome only one set of resistance genes--as opposed to numerous
sets of resistance genes in a genetically diverse farm landscape. 


   MOST GERMPLASM FOR U.S.  CROPS
   COMES FROM OTHER COUNTRIES
---------------------------------------------------------- Chapter 1:2

U.S.  agriculture is based on crops that originated from areas
outside of the United States.  For example, as shown in figure 1.1,
corn originated in Mexico and Guatemala, wheat in the Near East (in
such countries as Iran), and soybeans in China.  Crops of economic
importance that are native to the United States are limited and
include sunflowers, cranberries, blueberries, strawberries, and
pecans.  Thus, almost all the germplasm needed to increase the
genetic diversity of U.S.  agriculture comes from foreign locations. 

   Figure 1.1:  Centers of Origin
   of Selected Crops

   (See figure in printed
   edition.)

Note:  The pointer locations indicate general regions where crops are
believed to have first been domesticated.  In some cases, the center
of origin is uncertain.  Other geographic regions also harbor
important genetic diversity for these crops. 

Source:  This map was developed by GAO using data provided by NPGS'
Plant Exchange Office. 

While immigrants to the United States, including the first colonists
from Europe, brought seeds with them, native North Americans had
already introduced corn, beans, and other crops from Central and
South America.  Today, to obtain new germplasm for U.S.  collections,
plant breeders and researchers often rely on collections located in
foreign countries or on plant exploration trips to the centers of
origin for their crops.  Between 1986 and 1996, an estimated 75
percent of the germplasm samples added to NPGS' collections were
obtained from foreign countries. 

Although plant exploration trips are an important source of
germplasm, most of the germplasm in NPGS has been obtained from
existing collections both in the United States and in foreign
national and international collections.  Some of the U.S.  and
foreign collections belong to universities and private companies. 
Other foreign collections include (1) an international collection
based in 16 international agricultural research centers that is
administered by the Consultative Group on International Agricultural
Research\6 and (2) foreign national collections. 

The international agricultural research centers, located primarily in
developing countries, specialize in research intended to enhance the
nutrition and well-being of poor people through sustainable
improvements in the productivity of agriculture, forestry, and
fisheries.  These centers, according to the International Plant
Genetic Resources Institute,\7

have together assembled the world's largest international collection
of plant genetic resources for food and agriculture.  They account
for a significant proportion, possibly over 30 percent, of the
world's unique germplasm samples maintained in collections away from
their native environment.  The international research centers are
funded by voluntary contributions, and their plant germplasm has
historically been freely available to any user.  Moreover, users have
not applied intellectual property protection to the material.  The
United States works cooperatively with these centers to support
international activities to preserve germplasm.  For example, U.S. 
germplasm facilities maintain duplicate collections for some of the
international centers to provide for secure backup.  In addition,
U.S.  scientists help various centers screen germplasm for resistance
to pests and pathogens and serve in scientific liaison roles between
the centers and the U.S.  Agency for International Development. 

Finally, many countries, including most European nations, maintain
germplasm collections.  These national collections vary considerably
in terms of the quality of preservation, organizational structure,
the number of crops preserved, and the access provided to requesters. 
One of the largest collections of plant germplasm in the world is
maintained at Russia's Vavilov Institute of Plant Industry, named for
the Russian scientist who was a pioneer in the study of plants. 


--------------------
\6 The purpose of the consultative group is to promote sustainable
agriculture for food security in developing countries.  The
consultative group is jointly sponsored by the World Bank, the Food
and Agriculture Organization of the United Nations, the United
Nations Development Program, and the United Nations Environment
Program.  Fifty-three members, including the United States, provide
funds that support the consultative group. 

\7 The International Plant Genetic Resources Institute is an
autonomous, international scientific organization sponsored by the
consultative group. 


   PROFILE OF USDA'S NATIONAL
   PLANT GERMPLASM SYSTEM
---------------------------------------------------------- Chapter 1:3

The National Plant Germplasm System is primarily a federally and
state-supported effort aimed at maintaining supplies of germplasm
with diverse genetic traits for use in breeding and scientific
research.  While NPGS has been evolving since USDA established its
plant-collecting program in 1898, the main components of NPGS were
not established until the passage of the Agricultural Marketing Act
of 1946.  The act also provided a legal basis for state and federal
cooperation in managing crop genetic resources.  The current
organizational structure of NPGS--a geographically dispersed network
of germplasm collections administered primarily by USDA's
Agricultural Research Service (ARS)--emerged in the early 1970s. 
Although ARS provides the lion's share of support for NPGS, the
system is also supported by the agricultural experiment stations at
the state level.\8

In addition, private industry provides some support for selected
projects and develops and transfers germplasm in the form of plant
hybrids and varieties to farmers and other consumers. 

NPGS' major activities are (1) acquiring germplasm, (2) developing
and documenting information on the germplasm in its collections, and
(3) preserving the germplasm.  (See table 1.1.) NPGS also distributes
samples, free of charge, on request to plant breeders and other
scientists.  NPGS maintains about 440,000 germplasm samples for over
85 crops.  In 1996, NPGS distributed about 106,000 germplasm samples
to requesters in the United States and in 94 countries; it received
about 7,800 germplasm samples, about 5,000 of which originated in
foreign countries. 



                               Table 1.1
                
                         NPGS' Major Activities

Activity                        Description
------------------------------  --------------------------------------
Acquisition                     Collecting plant germplasm from
                                natural habitats and through exchange
                                with other scientists or collections.

Development and documentation   Development--characterizing some of
of information                  the germplasm's genetic traits, such
                                as height and color. Documentation--
                                entering these and other data in NPGS'
                                database, called the Germplasm
                                Resources Information Network.

Preservation                    Storing and maintaining germplasm to
                                ensure a diverse supply of germplasm.
                                In addition, NPGS distributes
                                germplasm to breeders and other
                                researchers.
----------------------------------------------------------------------
NPGS is responsible for developing characterization information--data
on traits such as plant structure and color that are little
influenced by the environment.  However, other information critical
to the use of NPGS germplasm and documented in the Germplasm
Resources Information Network (GRIN) is generally developed outside
of NPGS.  (GRIN, a database of NPGS' holdings, is available to
scientists and researchers worldwide.) For example, most evaluation
data, which document traits typically affected by environmental
conditions (e.g., plant yield and disease resistance), are developed
outside of NPGS.\9 These data are particularly important in providing
plant breeders with the information they need to select the specific
germplasm samples they seek from the sometimes thousands of possible
choices offered by NPGS.  Passport data, often provided by the person
or organization that collected or supplied the germplasm, document
the geographic origin and ecological conditions of its site of
origin. 

Other germplasm collections in the United States--beyond NPGS'--are
maintained by private companies, institutions such as universities
and state agricultural experiment stations, and nonprofit
organizations such as the Seed Savers Exchange.  Some of these
collections, as well as some foreign collections, are not freely
available to users of germplasm.  Although NPGS could not provide
information on the number, size, and condition of all of these
collections, they represent a substantial germplasm pool. 


--------------------
\8 Agricultural experiment stations are supported primarily by the
states but also receive support from USDA's Cooperative State
Research, Education, and Extension Service. 

\9 Up until 1992, NPGS received funding for germplasm evaluations. 
Since then, funding for these evaluations has been transferred from
NPGS to other ARS research programs. 


      NPGS MAINTAINS GERMPLASM
      COLLECTIONS AT SITES
      THROUGHOUT THE UNITED STATES
-------------------------------------------------------- Chapter 1:3.1

NPGS maintains collections at 22 sites throughout the United States
and in Puerto Rico.  In addition, staff at 10 other sites work
cooperatively with NPGS but do not receive NPGS funding.  NPGS also
maintains the National Seed Storage Laboratory (NSSL) and the
National Germplasm Resources Laboratory (NGRL).  Figure 1.2 shows the
locations of these sites and laboratories. 

   Figure 1.2:  NPGS Sites of
   Major Importance

   (See figure in printed
   edition.)

Source:  NPGS. 

While most NPGS collections are maintained at sites that house
germplasm for numerous crops, NPGS also has five sites that
specialize in crop-specific collections, such as potatoes or
soybeans.  In addition, NPGS has nine sites that are national clonal
germplasm repositories and four that maintain genetic stock
collections.\10 The four regional plant introduction stations\11 are
responsible for maintaining many of the major seed-reproducing
species held by NPGS.  In total, as of June 1997, they accounted for
almost half of the germplasm samples maintained in NPGS collections. 

NPGS sites generally contain either "backup" or "active" collections,
depending on the storage objectives.\12 Backup collections maintain
germplasm for long-term conservation, and active collections maintain
germplasm for short- to medium-term conservation and distribution. 
Germplasm is maintained either as seeds or as living plants.  The
latter category is generally referred to as "clonal" germplasm and
includes fruit trees, sugarcane, and strawberries.  Clonal germplasm
is likely to lose some of its distinct genetic characteristics when
reproduced from seed; therefore, it is reproduced asexually from its
own plant parts.  Clonal germplasm can be costly to preserve.  Some
fruit trees, for example, may require isolation to prevent loss from
pests as well as screened protection and other measures to ensure the
normal development of plants or to keep the fruit free of pests. 

At each site, crop curators and other staff are responsible for
maintaining the germplasm collections.  Curators regenerate (or
replenish) germplasm samples by growing additional plants from seed
or other plant parts to ensure that an adequate number of samples are
available for (1) distribution to plant breeders, research
scientists, and institutions and (2) storage in long-term
collections.  In the process of regeneration, curators must ensure
that each plant generation is as genetically similar to its
predecessor as possible.  During regeneration, curators also document
certain plant characteristics (such as plant height and color) if
this information is not already available.  Curators and other staff
are responsible for entering information about each germplasm sample
into GRIN. 

The National Seed Storage Laboratory (NSSL) at Fort Collins,
Colorado, maintains the long-term backup collection of seeds for NPGS
and some non-NPGS collections located in the United States and
foreign countries and conducts research on preserving plant
germplasm.  NSSL's storage facilities were modernized and expanded
fourfold in 1992, with high-security vaults to protect the germplasm
against natural disasters.  The collection duplicates (or backs up)
many of the germplasm samples in NPGS' active collections in the
event that the germplasm kept in active collections is lost. 
Germplasm can be lost for a variety of reasons, including natural
disasters or degeneration through inadequate storage.  Seeds
preserved at NSSL are kept in colder, more secure conditions (i.e.,
sealed, moisture-proof containers in vaults at -18 degrees Celsius or
containers over liquid nitrogen at -160 degrees Celsius) that
preserve them longer than seeds preserved at many active sites.\13
With few exceptions, such as apple buds that can be preserved in
liquid nitrogen, NSSL does not back up clonal germplasm.  Clonal
collections may be backed up--in greenhouses, as tissue culture, or
through cryopreservation--\14 at the same sites as their active
collections. 

The National Germplasm Resources Laboratory, located in Beltsville,
Maryland, contains several units that support NPGS.  The Plant
Exchange Office--with extensive input from the CGCs and NPGS' crop
curators--is responsible for setting priorities for the germplasm
needs of NPGS' collections.  Furthermore, the Office coordinates
plant exploration trips, facilitates germplasm exchanges with other
collections, and documents the entry of germplasm into NPGS,
including its passport data.  In addition, the Germplasm Resources
Information Network/Database Management Unit manages GRIN, NPGS'
database, which provides information for users and managers, such as
passport information on NPGS samples. 

ARS' Plant Germplasm Quarantine Office works with USDA's Animal and
Plant Health Inspection Service (APHIS) in administering the National
Plant Germplasm Quarantine Center in Beltsville, Maryland.\15 These
sites test specific types of imported germplasm for pests and
pathogens before the germplasm is introduced into the United States. 
All plant germplasm coming into the United States must comply with
federal quarantine regulations intended to prevent the introduction
of pests and pathogens that are not widespread in the United States. 
APHIS writes, interprets, and enforces quarantine regulations, while
ARS is generally responsible for providing research support,
including the development of tests for pests and pathogens.  In
addition, ARS, through a 1986 Memorandum of Understanding with APHIS,
maintains and tests germplasm that falls into the "prohibited"
quarantine category.\16


--------------------
\10 Clonal repositories hold germplasm (e.g., fruit trees) that are
maintained as living plants or plant parts.  Genetic stock
collections contain germplasm with one or more special genetic traits
that make them of interest to researchers. 

\11 The four regional plant introduction stations--located at Ames,
Iowa; Pullman, Washington; Geneva, New York; and Griffin,
Georgia--are jointly operated by ARS and the state agricultural
experiment stations of the region. 

\12 According to ARS, genetic stock collections are classified
separately because specialized care and trained personnel are needed
to maintain them. 

\13 In contrast, seeds in many active collections are generally
stored at 5 degrees Celsius, although active collection sites are
increasing the use of storage at -18 degree Celsius to reduce losses. 

\14 Tissue culture is a technique for cultivating cells, tissues, or
plant parts in a sterile, synthetic medium.  Cryopreservation
involves maintaining tissues or seeds in long-term storage at
ultralow temperatures, typically between -150 degrees and -196
degrees Celsius. 

\15 In addition, ARS administers a quarantine facility in St.  Croix,
U.S.  Virgin Islands, that tests corn, sorghum, and millet. 

\16 The prohibited category is the most stringent quarantine
category, requiring that germplasm be sent to a quarantine facility
for testing or observation before it is introduced into the United
States. 


      SUPPORT FOR NPGS COMES FROM
      SEVERAL SOURCES
-------------------------------------------------------- Chapter 1:3.2

NPGS' activities are supported at the federal level primarily by ARS,
with additional support provided by states' land grant universities
through their agricultural experiment stations.  Many of NPGS'
collections have been jointly developed and maintained by federal and
state scientists at states' agricultural experiment stations, and
most NPGS sites are located on experiment station properties.  State
universities provide in-kind support in the form of services,
personnel, and facilities.  In addition, private industry provides
limited support, such as regenerating germplasm at company sites or
funding special projects. 

In fiscal year 1996, NPGS funding was $23.3 million.  Of this amount,
$19.5 million was provided by ARS; $1.4 million by USDA's Cooperative
State Research, Education, and Extension Service; $1.3 million by
APHIS; $0.8 million (in-kind support) from the states' agricultural
experiment stations; and $0.3 million from other nonfederal sources. 
Included in the ARS funding was $3.9 million for plant collection
activities--germplasm acquisition, quarantine, and
classification--and $15.6 million for such activities as
preservation, documentation, and distribution.  From fiscal years
1992 through 1996, ARS' funding for NPGS has been essentially level;
however, if calculated in constant dollars, funding declined by 14
percent during this period.  During this period, NPGS' germplasm
collections increased by 10 percent. 


      MANAGEMENT OF NPGS IS HIGHLY
      DECENTRALIZED
-------------------------------------------------------- Chapter 1:3.3

While ARS has the primary responsibility for managing NPGS, no single
individual or entity has overall authority for managing the entire
system.  Within ARS, numerous officials and committees have different
levels of authority and responsibility for components of the system. 

ARS' National Program Leader for Plant Genetic Resources has a broad
range of leadership responsibilities for the system, including
developing budget proposals, planning resource allocations among the
NPGS sites, and addressing international issues affecting
germplasm.\17 The program leader also participates in and is advised
by various groups that make recommendations concerning NPGS'
operations and policies.  The program leader, however, has limited
authority for the budgets, projects, or management of each NPGS site. 
Responsibility for these activities rests with (1) ARS' area
directors, who have direct oversight responsibility and authority for
the NPGS sites located within their areas of jurisdiction, (2) NPGS'
site leaders, and (3) ARS' national program staff.  In particular,
the area directors coordinate some site program reviews, conduct
performance ratings for key administrative staff, hire personnel, and
manage discretionary funding for NPGS sites located in their
jurisdiction. 

Because of the broad array of crops represented in NPGS'
collections--each requiring specific scientific and technical
expertise--NPGS relies on 40 Crop Germplasm Committees (CGC) to
provide expert advice on technical matters relating to germplasm
activities.  Among other things, the CGCs are expected to provide
recommendations on the management of the germplasm collections for
their crops, including setting priorities for acquisition and
evaluation research.  CGC members--representing ARS, universities,
and the private sector--include plant breeders, NPGS curators,
pathologists, and other scientists who are experts on specific crops. 
A crop committee can represent one crop group or several.  For
example, the soybean CGC provides advice on soybeans, while the leafy
vegetable CGC is responsible for lettuce, spinach, chicory, and
celery.  (See app.  III for a listing of the CGCs and the crops for
which they are responsible.) These committees generally meet about
once a year and issue reports on the status of their respective
collections.  However, they receive no funding for their work or
related expenses. 


--------------------
\17 The position of national program leader for plant genetic
resources has been vacant for the past 5 years.  There have been five
acting program leaders during this period. 


   PAST GAO AND NATIONAL RESEARCH
   COUNCIL REPORTS HAVE CITED MANY
   NPGS SHORTCOMINGS
---------------------------------------------------------- Chapter 1:4

GAO and National Research Council reports, dating as far back as
1981, have cited management and organizational shortcomings and needs
that have hindered NPGS' overall effectiveness.  In 1981, for
example, GAO concluded that insufficient management attention by USDA
to germplasm collection, storage, and maintenance had endangered the
preservation of germplasm in the United States.\18 Another GAO
report, issued earlier that year, recommended that USDA centralize
control over the Department's genetic resources and develop a
comprehensive plan for their use.\19 In 1990, GAO reported that ARS
had difficulty in setting priorities and allocating funding among the
various plant germplasm management activities.\20

In a comprehensive evaluation of NPGS issued in 1991,\21 the National
Research Council concluded that NPGS had no discernible structure and
organization for managing and setting priorities for its activities,
formulating national policies, or developing budgets to act on
emerging priorities.  The Council made many recommendations,
including that USDA strengthen NPGS by centralizing its management
and budgeting functions and by establishing clear goals and policies
for NPGS' leadership to use in developing long-range plans.  Other
recommendations included expanding the capacity of NSSL and providing
financial support to the CGCs.\22

During the 1990s, USDA has made several changes to address some of
the operational shortcomings discussed above.  In particular, it has
expanded NSSL's long-term, secure storage facility fourfold. 
Furthermore, NPGS' sites with active collections are making greater
use of -18 degree Celsius storage to improve germplasm preservation. 
In addition, NPGS' GRIN database has been substantially improved by
the addition of such features as a new search function and access to
users through the Internet. 


--------------------
\18 Better Collection and Maintenance Procedures Needed to Help
Protect Agriculture's Germplasm Resources (CED-82-7, Dec.  4, 1981). 

\19 The Department of Agriculture Can Minimize the Risk of Potential
Failures (CED-81-75, Apr.  10, 1981). 

\20 Plant Germplasm:  Improving Data for Management Decisions
(PEMD-91-5A, Oct.  10, 1990). 

\21 Managing Global Genetic Resources:  The U.S.  National Plant
Germplasm System, National Research Council (Washington, D.C.: 
National Academy Press, 1991). 

\22 NSSL was expanded in 1992.  The CGCs were formerly called the
Crop Advisory Committees. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:5

We surveyed the members of the 40 CGCs for their views on the
sufficiency of NPGS' principal activities--acquiring germplasm to
ensure the diversity of the collections in order to reduce crop
vulnerability, developing and documenting information on germplasm,
and preserving germplasm.  Specifically, we surveyed the 680 members
of the CGCs--including 38 additional experts identified by USDA.  The
median CGC response rate was 86 percent, and all NPGS curators
participated in the survey.  We conducted this survey from November
1996 through March 1997. 

In addition, we obtained information about NPGS' major
activities--acquisition, development and documentation of
information, and preservation--from interviews with the following: 
two acting National Program Leaders for Plant Genetic Resources;
several NGRL officials responsible for plant exploration, quarantine,
and GRIN; the Director, National Plant Germplasm Quarantine Center,
APHIS; the Director and research leaders, NSSL; the site leaders of
the four regional plant introduction stations and the Davis,
California clonal repository; a number of curators and breeders at
various NPGS sites; and ARS budget staff.  We visited NGRL and APHIS
officials in Beltsville, Maryland; two of the four regional plant
introduction stations (Ames, Iowa, and Griffin, Georgia); the
National Soybean Collection, Urbana, Illinois; and NSSL in Fort
Collins, Colorado.  We also interviewed officials from USDA's
Economic Research Service; Pioneer Hi-Bred International, Inc., a
large seed producer; the Department of State; and the Agency for
International Development. 

In addition, we reviewed (1) NPGS program documents, including
planning and budget documents; (2) acquisition and preservation data
(based on GRIN data) provided to us by NGRL officials, as well as
preservation data provided by officials from the four plant
introduction stations; (3) CGC reports; (4) site and program reviews;
and (5) documents from the Food and Agriculture Organization of the
United Nations and from international sources related to germplasm
access.  We did not verify the accuracy and reliability of the data
provided by NPGS. 

We conducted our review from July 1996 through September 1997 in
accordance with generally accepted government auditing standards.  We
provided USDA with a draft of our report for review and comment. 
These comments and our response to them are in appendix IV. 


CGCS UNDERSCORED IMPORTANCE OF
ACQUIRING GERMPLASM TO INCREASE
GENETIC DIVERSITY, BUT SOME
OBSTACLES HINDER ACQUISITION
============================================================ Chapter 2

Most CGCs reported that the overall diversity in freely available
germplasm collections\1 --including NPGS'--is sufficient for reducing
their crops' vulnerability.  Nonetheless, they ranked the acquisition
of additional germplasm as a top priority for NPGS, thereby
underscoring the importance they place on having maximum genetic
diversity in NPGS' collections.  A number of issues may be
contributing to the CGCs' emphasis on acquiring germplasm for the
NPGS collection.  For example, most CGCs said that at least one of
the four types of germplasm that generally constitute their
collections is inadequate; each type contains genetic material that
plays an important role in a collection's overall diversity. 

Most CGCs considered acquiring more germplasm to be a top priority;
however, problems with some countries have hindered access to
potential sources of new germplasm in those areas.  In addition,
certain provisions in the Convention on Biological Diversity, which
entered into force in 1993, may place constraints on the use of and
access to some foreign germplasm in the future. 

Even when NPGS acquires new germplasm, its release to breeders and
research scientists has sometimes been delayed as a result of
problems in USDA's management of the quarantine process.  According
to many CGCs whose germplasm generally undergoes the most intensive
quarantine testing, the process has resulted in the delayed release
and, to a lesser extent, the loss of some germplasm. 


--------------------
\1 Freely available (i.e., without restrictions) collections include
NPGS' and international collections as well as some university and
private collections and many foreign national collections.  It is
always possible that a collection that is currently freely available
may, in the future, become restricted or unavailable. 


   MOST CGCS REPORTED THAT
   GERMPLASM COLLECTIONS ARE
   SUFFICIENTLY DIVERSE, BUT THEY
   STILL WANT TO INCREASE
   GERMPLASM ACQUISITION
---------------------------------------------------------- Chapter 2:1

When all freely available collections were taken into account, almost
three-quarters of the CGCs reported that these collections are
sufficiently diverse for reducing the vulnerability of their crops. 
For the NPGS collections alone, just over half the CGCs reported that
the genetic diversity of their NPGS collections is sufficient to
reduce crop vulnerability.  Nonetheless, the CGCs overall viewed the
acquisition of additional germplasm as a top NPGS priority--out of 14
germplasm-related activities--in the event of additional funding. 
Several concerns highlighted by the CGCs in our survey may contribute
to the importance they place on increased acquisition.  These
concerns include the lack of diversity within specific parts of their
collections and the potential loss of germplasm that is endangered in
nature or in at-risk collections (e.g., collections of scientists who
are retiring). 


      MOST CGCS BELIEVED THAT
      GERMPLASM COLLECTIONS HAVE
      SUFFICIENT DIVERSITY
-------------------------------------------------------- Chapter 2:1.1

When all freely available collections (including NPGS') were
considered, 29 of the 40 CGCs reported that the genetic diversity in
the collections for their crops is sufficient for reducing crop
vulnerability.  Major crops--such as corn, wheat, and soybeans--are
in this category.  The sufficiency of the collections declined
somewhat when only NPGS collections were considered:  Twenty-two, or
just over half of the CGCs reported that the NPGS collections for
their crops have sufficient genetic diversity overall to reduce crop
vulnerability.  (See fig.  2.1.)

   Figure 2.1:  CGCs' Perceptions
   of the Diversity of All Freely
   Available Collections and of
   the Diversity of the NPGS
   Collections for Their Crops

   (See figure in printed
   edition.)

Nine CGCs said that the genetic diversity of the NPGS collection for
their crops is insufficient for reducing crop vulnerability:  grapes,
cool season food legumes, sweet potatoes, cucurbits (e.g., squash and
melons), tropical fruit and nut, walnuts, herbaceous ornamentals,
prunus (e.g., peach and cherry trees), and woody landscape.  In
addition, nine CGCs said that their collections have neither
sufficient nor insufficient diversity. 


      CGCS BELIEVED THAT GERMPLASM
      ACQUISITION SHOULD BE A TOP
      PRIORITY FOR NPGS
-------------------------------------------------------- Chapter 2:1.2

While over half the CGCs believed that the genetic diversity of their
NPGS germplasm collections for their crops is sufficient, they all
reported that it is moderately to extremely important to increase the
diversity of their NPGS collections.\2 The importance the CGCs placed
on increasing diversity is underscored by the high priority given to
germplasm acquisition in the event of additional funding--of 14
germplasm-related activities, the CGCs, on average, gave acquisition
the highest ranking.  (Fig.  2.2 shows the average ranking that CGCs
gave to each activity, with 1 being the highest possible ranking.)

   Figure 2.2:  CGCs' Ranking of
   the Priority to Be Given to 14
   Germplasm-Related Activities
   in the Event of Additional
   Funding

   (See figure in printed
   edition.)

Note:  If all 40 CGCs ranked one activity as their first priority,
then its average ranking would be 1.  Enhancement and breeding are
ancillary NPGS activities and are primarily funded by other ARS
programs and by universities and the private sector. 

All 40 CGCs stated that they knew of germplasm samples that would
increase the genetic diversity of the NPGS collections and that
should be added to them.  For example, the Wheat CGC's 1996 report to
NPGS cited three critical collection needs for the NPGS wheat
collection and specified where much of this germplasm could be
obtained, including landraces (seeds passed down by farmers from one
generation to another to produce desired plant characteristics) from
Guatemala, where they have not been collected before, and wild wheat
relatives from Albania, Greece, and the former Yugoslavia. 
Similarly, the Sweet Potato CGC wanted to enhance the limited genetic
diversity of the NPGS sweet potato collection by obtaining a
representative sample of germplasm from the International Potato
Center in Peru.  This collection contains about 6,500 germplasm
samples of sweet potato, compared with about 1,170 in the NPGS
collection. 


--------------------
\2 Of the 40 CGCs, 14 reported that it is moderately important, 25
reported that it is very important, and 1 reported that it is
extremely important to increase the genetic diversity of their
collections. 


      SEVERAL PROBLEMS ASSOCIATED
      WITH THE COLLECTIONS MAY
      CONTRIBUTE TO PRIORITY GIVEN
      TO GERMPLASM ACQUISITION
-------------------------------------------------------- Chapter 2:1.3

Although most CGCs reported that their NPGS collections overall are
sufficiently diverse at this time, they cited several concerns with
the collections that may account for the importance they place on
increased acquisition.  First, most CGCs reported that at least one
of the following types of germplasm in their collections is
insufficiently diverse for reducing crop vulnerability:  wild and
weedy relatives of cultivated crops, landraces, and genetic stocks. 
Only obsolete and current cultivars, the fourth type of germplasm
samples in a collection, are considered to be sufficient by most
CGCs.  Specifically: 

  -- Wild and weedy relatives of crops were reported to be
     insufficient by almost half the CGCs, including those for major
     crops such as corn and soybeans.  Wild relatives have often been
     used to improve crops, such as tomatoes, and sometimes to
     develop new ones. 

  -- Landraces--many of which are grown from selected quality seed
     passed down by farmers from one generation to another--were
     reported to be insufficient by 12 of the 40 CGCs.  Landraces are
     rich sources of genes for traits such as resistance to pests and
     pathogens. 

  -- Genetic stocks are insufficiently diverse, according to over
     half the CGCs, including those for major crops such as alfalfa,
     peanuts, and grapes.  While genetic stock material is essential
     to genetic research, according to NPGS officials, it has
     generally played a minor role in commercial breeding programs. 
     However, it is expected to become increasingly important in
     breeding programs that use molecular genetic tools to manipulate
     and transfer genes to create new products, according to the
     National Research Council.\3

  -- Obsolete and current cultivars are sufficient for reducing the
     vulnerability of their crops, according to most CGCs.  Only five
     CGCs cited insufficiencies in this area. 

Furthermore, 39 CGCs said that NPGS should place increased emphasis
on acquiring germplasm endangered in nature or acquiring germplasm
from collections at risk, such as the Vavilov collection in Russia or
the collections of scientists who are retiring.  If such collections
are not obtained and preserved, their germplasm may be lost. 
Finally, 37 CGCs reported that certain plants are becoming extinct or
hard to find.\4

NPGS' acquisition policy is to rely heavily on the 40 CGCs and the
NPGS curators to assess the adequacy of their respective germplasm
collections and recommend areas where additional acquisition may be
needed.  However, NPGS has not developed a comprehensive, long-term
plan to establish critical acquisition needs for its germplasm
collections and priorities for collection trips to fill those needs. 
Currently, NPGS' collection trips are based primarily on proposals
that are submitted to NPGS' Plant Exchange Office by federal and
university scientists and endorsed by the appropriate CGCs.  In
addition, staff from the Plant Exchange Office occasionally make or
participate in collection trips.  However, some exploration trips are
funded by other USDA or non-USDA federal agencies.\5

According to NPGS officials in the Plant Exchange Office, some
germplasm collections are more frequently targeted for collection
trips than others because (1) the gaps in some collections are better
known and (2) some collections have more assertive champions--e.g., a
germplasm curator, CGC, or other interested scientist who
aggressively seeks out collection opportunities.  This approach may
overlook the needs of some crops.  For example, according to the head
of the Plant Exchange Office, 16 of the CGCs' reports state
acquisition needs only in a general fashion and therefore are of
limited value for planning or setting acquisition priorities. 

The exchange officer acknowledged the need to develop a long-term
plan that would reflect collection priorities for each crop.  He
noted that such a plan would use existing funds more efficiently and
help ensure that the needs of all crops are being addressed.  NPGS
has been working to develop such a plan for several years, but
progress has been slow because the office has lacked the resources to
adequately staff the project and provide needed scientific expertise. 
The initial plan, which is intended to be flexible to accommodate
changing needs and conditions, is expected to be completed by Spring
1998. 

Concerns about NPGS' acquisition planning process are long-standing. 
For example, over 15 years ago, GAO recommended that a long-range
plan be developed to address gaps in germplasm collections and
objectives for collecting or otherwise acquiring needed germplasm.\6
In 1991, the National Research Council recommended, among other
things, that NPGS develop a comprehensive plan for plant exploration. 
The Council noted that in the past, the lack of an exploration plan
resulted in some crops receiving attention, while others went
unserved.\7


--------------------
\3 Managing Global Genetic Resources:  Agricultural Crop Issues and
Policies, Board on Agriculture, National Research Council
(Washington, D.C.:  National Academy Press, 1993). 

\4 As discussed in ch.  1, there are also concerns about the
vulnerability of crops to pests and pathogens.  All 40 CGCs reported
that such risk is a serious problem for their crops:  Six said
genetic vulnerability is a very serious problem, 30 said it is
moderately serious, and 4 said it is somewhat serious.  The six CGCs
reporting very serious problems represented oats, cool season food
legumes, tropical fruit and nut, grapes, walnuts, and prunus. 

\5 The Plant Exchange Office often works with other agencies within
USDA and other agencies, such as the U.S.  Agency for International
Development, to obtain funding. 

\6 The Department of Agriculture Can Minimize the Risk of Potential
Crop Failures (CED-81-75, Apr.  10, 1981). 

\7 The U.S.  National Plant Germplasm System:  Managing Global
Genetic Resources, National Research Council (Washington, D.C.: 
National Academy Press, 1991). 


   CGCS REPORT PROBLEMS IN
   ACQUIRING FOREIGN GERMPLASM
---------------------------------------------------------- Chapter 2:2

Although CGCs want to acquire more germplasm, most reported that
difficulties between the United States and some foreign countries
have hindered NPGS' efforts to obtain the germplasm needed to
increase the diversity of its collections.\8 For example, the Soybean
CGC report indicated that relations between the United States and
North Korea have hindered the CGC from obtaining germplasm from North
Korea.  The report stated that the few soybean germplasm samples from
North Korea in NPGS' collection were either obtained more than 60
years ago or have been received since then through third parties. 
Several other CGC reports--including those for sugarbeets, peas, and
wheat--cited difficulties in obtaining germplasm from the Middle
East.  The Wheat CGC, for example, noted that Iran, a country with
which the United States does not have diplomatic relations, holds
potentially valuable wheat germplasm. 

In addition, issues relating to the ownership and use of foreign
germplasm have become more problematic as a result of the entry into
force of the Convention on Biological Diversity in 1993.\9 Prior to
the Convention, germplasm from most countries, other than those where
access was restricted, has been generally available to requesters. 
However, the Convention recognizes the sovereign rights of nations
over their natural resources and their rights to exchange these
resources under terms mutually agreeable to the nation and the
germplasm recipient.  Officials from NPGS, the State Department, the
Agency for International Development, and the World Bank observed
that access to plant germplasm could be reduced as a result of these
provisions but that the full impact of the Convention may be unknown
for a number of years. 

However, one likely result of the Convention will be the increased
use of material transfer agreements--contracts that require germplasm
users to agree to certain conditions in exchange for the use of the
germplasm.  These agreements may require, for example, that the
requester not seek intellectual property rights or claim ownership
over the germplasm.  USDA officials will sign material transfer
agreements only if their terms are consistent with NPGS' policy to
provide users with free and open access to germplasm. 


--------------------
\8 Of the 40 CGCs, 13 reported that long-standing political
difficulties had hindered the acquisition of germplasm from foreign
countries to some extent, 22 to a moderate extent, and 4 to a great
extent.  One CGC said that such difficulties created little or no
hindrance in NPGS' ability to increase diversity for its crop
collection. 

\9 The Convention on Biological Diversity is a legally binding
framework--for countries that have consented to it--for conserving
and utilizing global diversity.  The U.  S.  Congress has not yet
consented to it. 


   USDA'S MANAGEMENT OF QUARANTINE
   PROGRAM HAS HAMPERED
   ACQUISITION OF SOME GERMPLASM
---------------------------------------------------------- Chapter 2:3

A number of problems related primarily to USDA's overall management
of the germplasm quarantine program have hampered the program's
effectiveness and resulted in delays in the release of some
germplasm.  While most CGCs reported that U.S.  quarantine
regulations and processes have been effective in reducing the
introduction of pests and pathogens into the United States, 13 CGCs,
most of whose germplasm often undergoes more intensive scrutiny in
quarantine, reported problems with the timeliness of the quarantine
process, and 5 reported problems with the release of viable
germplasm.  While the CGC for prunus (e.g., cherry and peach trees)
reported that USDA's regulations and processing have been very
ineffective in both of the above areas, CGCs for crops such as
apples, pears, potatoes, and corn also reported problems. 


      POOR CROP PRODUCTION
      PRACTICES, INADEQUATE
      FACILITIES, AND OUTMODED
      TESTING PROCEDURES HAVE
      CREATED PROBLEMS FOR
      QUARANTINED GERMPLASM
-------------------------------------------------------- Chapter 2:3.1

All plant germplasm coming into the United States must comply with
federal quarantine regulations intended to prevent the introduction
of pests and pathogens not widespread in the United States.  These
regulations range from a category requiring only visual inspection at
the port of entry for germplasm such as the seeds of most vegetables
and flowers, to a category--known as "prohibited"--requiring that the
germplasm be sent to a quarantine facility for testing or observation
before release.\10 Although less than 3 percent of the world's plant
species are in this latter category, it includes a wide range of
crops:  all or most clonally propagated prunus, apples, pears,
potatoes, sugarcane, strawberries, sweet potatoes, grapes, certain
woody landscape plants, and grasses as well as the seeds of wheat,
corn, and rice from some regions where there are serious diseases not
already in the United States. 

Thirteen CGCs--most of whose germplasm is often in the prohibited
category--reported that USDA's management of the quarantine process
hinders the timely acquisition of viable germplasm.  In addition, ARS
officials told us that some germplasm has died while in quarantine
because it was poorly maintained.\11 The specific types of problems
identified by the CGCs, ARS and APHIS officials, and ARS reviews
included (1) poor production practices during quarantine, (2)
inadequate facilities or sites, and (3) the types of testing
procedures that are currently in use. 


--------------------
\10 APHIS gives certain qualified importers of germplasm for some
crops a permit that enables them to test and observe the germplasm in
their own facilities to ensure it meets USDA regulations. 

\11 According to ARS officials, some germplasm dies in quarantine
because it is in poor condition when it arrives at the quarantine
facility. 


         POOR PRODUCTION PRACTICES
         HAVE RESULTED IN DEAD
         GERMPLASM AND DELAYS IN
         RELEASE
------------------------------------------------------ Chapter 2:3.1.1

Eleven CGCs, representing such germplasm collections as prunus,
apples, pears, potatoes, and sweet potatoes, reported that poor crop
production practices--such as inadequate watering, soil preparation,
and weeding--during quarantine hinder the timely acquisition of
viable germplasm.  Furthermore, an internal review of tree-growing
practices at the Maryland quarantine facility, conducted in 1996 by a
horticultural scientist at the request of ARS, noted the death of
several thousand fruit trees planted between 1993 and 1995.\12

The review cited improper horticultural practices as a major cause of
many of the deaths and recommended improved practices. 

When trees in quarantine are not properly maintained, they may die
and their germplasm will need to be imported again.  For example, an
ARS scientist at the quarantine office estimated that about 20
percent of all prunus germplasm samples brought into the country in
the past 10 years had died because they did not receive proper
horticultural care. 

In addition, poor production practices have kept trees from maturing
sufficiently to permit testing, thereby delaying the release of
germplasm.  Such delays have occurred with the germplasm of prunus,
apple, pear, and quince trees.  For example, since 1991, the release
of hundreds of germplasm samples for apple, pear, quince, and prunus
trees has been delayed as a result of inadequate horticultural
practices, according to the ARS scientists at the quarantine office
who test and monitor these trees.  Delays for most of the clonal
apple, pear, and quince germplasm have been about 8 to 10 years.\13

Furthermore, the average time for the unconditional release of prunus
germplasm at the Maryland quarantine facility has been about 10
years; however, generally no more than 4 years should be required,
according to APHIS officials.  ARS officials expect that it will not
unconditionally release apple, pear, quince, or prunus clonal
material until the year 2000 or later because of horticultural
practices that have resulted in the lack of mature trees needed for
testing. 


--------------------
\12 According to the Research Leader of the National Germplasm
Resources Laboratory, the trees that died were primarily trees to be
used for testing purposes and generally did not include imported
plant material. 

\13 This refers to any release that is not conditional on any federal
restrictions, e.g., requiring further observation or limiting the use
of the germplasm. 


         INADEQUATE FACILITIES
         HAVE HINDERED HEALTH OF
         QUARANTINED PLANTS
------------------------------------------------------ Chapter 2:3.1.2

Thirteen CGCs--including those for prunus, pears, corn, and
rice--reported that conditions at the quarantine facilities used to
grow their plants hinder the timely release of viable germplasm. 
Problems with quarantine facilities were also reported in ARS reviews
in 1994 and 1996.\14 The 1996 review stated that conditions at the
quarantine facilities in Maryland were not conducive to promoting
plant health.  For example, it noted that the Maryland site's soil
was unsuitable for growing trees and recommended the installation of
space heaters in the screenhouses to keep the temperature slightly
above freezing.  In addition, a plant breeder on the pear CGC said
that the Beltsville facility is not ideal for pears or apples because
the climate of the mid-Atlantic region is conducive to the
development of fire blight, a serious bacterial disease that is
difficult to control once trees are infected. 


--------------------
\14 The 1994 review was an in-depth review of the National Germplasm
Resources Laboratory, including the Plant Germplasm Quarantine
Office.  The 1996 review addressed tree-growing practices at the
plant quarantine fields and greenhouses. 


         OUTMODED TESTING
         PROCEDURES HAVE
         CONTRIBUTED TO DELAYS IN
         RELEASE OF GERMPLASM
------------------------------------------------------ Chapter 2:3.1.3

Sixteen CGCs--including the prunus, apple, pear, corn, wheat, rice,
and potato CGCs-- reported that required testing procedures hinder
the timely acquisition (e.g., introduction and distribution) of
viable germplasm for their crops.\15 While ARS is responsible for
developing new tests, APHIS must approve the tests that are used as
well as the release of germplasm from quarantine.  Nearly all of the
quarantine testing procedures currently in use date back to the early
1980s or before.  These procedures involve testing for pathogens such
as viruses and other infectious agents.  For some crops, testing
begins by closely observing the quarantined plants for symptoms of
disease during plant growth and subjecting the plants to a battery of
tests for latent pathogens.  Some tests for trees can take
considerable time because the tree must first bear fruit before tests
can be completed.  For example, tests on prunus trees generally
require a minimum of about 3, and no more than 4, years to complete,
according to APHIS officials. 

More sophisticated testing methods using molecular techniques to
identify pathogens are being developed, and some are already
available.  These tests could save considerable time in quarantine as
well as the costs associated with caring for the plants during that
time.  Such tests could also curtail the loss of germplasm that is
associated with longer quarantine periods, according to APHIS and ARS
officials.  ARS has developed, and APHIS has approved, molecular
tests for potato viruses; these tests have cut quarantine testing
from 2 years to 1, according to an ARS scientist.  In addition, APHIS
is currently reviewing newly developed molecular tests for detecting
certain diseases in prunus that would allow the conditional release
of prunus in about 18 months, on average.  In addition, ARS is
working on the development of molecular tests for certain sweet
potato pathogens. 

However, some plant breeders are concerned that the development and
approval of new testing methods has been unduly slow.  A 1994 review
of the germplasm quarantine office, conducted by ARS and university
scientists at the request of ARS, noted that virtually all popular
new apple and pear trees clones of foreign origin enter the United
States illegally, without pathogen testing.  It stated that both ARS
and APHIS needed to adopt policies that would make pathogen testing
more responsive to the needs of the deciduous fruit industry and its
associated germplasm collections and CGCs. 


--------------------
\15 The remainder of the crops were grapes, small fruits (e.g.,
berries), peanuts, sweet potatoes, cucurbits (e.g., melons), grass,
sunflower, herbaceous ornamental plants, and woody landscape. 


MANY NPGS GERMPLASM COLLECTIONS
LACK INFORMATION NEEDED FOR CROP
BREEDING
============================================================ Chapter 3

According to most CGCs, NPGS collections for their crops lack
sufficient information on germplasm traits to facilitate the
germplasm's use in crop breeding.  Specifically, these CGCs raised
concerns about two types of information--evaluation and
characterization.  Evaluation information describes traits (such as
yield and resistance to disease) of particular interest to plant
breeders, while characterization information describes traits (such
as plant structure, seed type, and color) that are little influenced
by environmental conditions.  Most CGCs reported that passport
data--a third type of information that describes, among other things,
the site of origin of the germplasm--are sufficient for breeding
crops. 

NPGS officials acknowledged that gaps exist in needed information, in
part because the information has not been developed and in part
because the information that has been developed has not always been
entered into NPGS' centralized database--the Germplasm Resources
Information Network (GRIN).  They noted, however, that given their
limited resources, the day-to-day tasks of preserving germplasm to
maintain its viability take precedence over developing and
documenting information. 


   MANY CGCS REPORTED THAT
   EVALUATION AND CHARACTERIZATION
   INFORMATION ARE INSUFFICIENT
---------------------------------------------------------- Chapter 3:1

Three-quarters of the CGCs reported insufficiencies with evaluation
information, and almost half found characterization information
insufficient for crop-breeding purposes.  On the other hand, most
CGCs reported that passport information is sufficient for crop-
breeding purposes.  Several NPGS managers told us, however, that
passport information--particularly for older samples--is not adequate
for NPGS' internal planning and management. 


      MOST CGCS BELIEVED THAT
      EVALUATION DATA ARE
      INSUFFICIENT
-------------------------------------------------------- Chapter 3:1.1

Breeders need comprehensive evaluation information to select
germplasm with the traits they are seeking from the myriad of
germplasm samples.  According to the National Research Council,
evaluation is a prerequisite for the use of germplasm--germplasm
samples that are not evaluated remain mostly curiosities.\1 In
developing evaluation data, scientists test germplasm samples for
various traits under a wide range of conditions.  Although the
preliminary evaluation of traits is generally considered an NPGS
activity, most evaluations are part of the research that accompanies
breeding programs and are conducted and funded primarily through
other ARS programs and universities.  In addition, industry conducts
and funds a small amount of germplasm evaluation for NPGS. 

Thirty of the 40 CGCs reported that the evaluation information on
their NPGS collections is somewhat or very insufficient for crop
breeding, and only 3 reported that it is somewhat sufficient--the
alfalfa, sugarbeets, and tropical fruit and nut CGCs.  Figure 3.1
shows the sufficiency of evaluation information, as reported by the
40 CGCs. 

   Figure 3.1:  CGCs' Perceptions
   of the Sufficiency of
   Evaluation Information for Crop
   Breeding

   (See figure in printed
   edition.)

The CGCs reported that the trait most likely to have been
evaluated--of the five traits we asked for their views on--is
"resistance to pests and pathogens considered to be a serious risk."
Even so, less than half the CGCs reported that their germplasm has
been evaluated to a moderate extent for this trait and only one to a
great extent.  For the remaining four evaluation traits, 35 to 38
CGCs reported their germplasm had been evaluated only to some,
little, or no extent.  These traits include tolerance to abiotic
stresses, such as salt or drought, considered a serious risk; quality
characteristics, such as flavor or appearance; production
characteristics, such as yield; and root stock traits.\2 (See fig. 
3.2.) While identifying shortcomings in the evaluation information,
almost half of the CGCs said that NPGS' management of evaluation data
has improved since about 1990.  (In addition, 20 CGCs said that there
has been no change, and 1 said it has worsened.)

   Figure 3.2:  CGCs' Assessment
   of the Extent to Which Major
   Traits Have Been Evaluated

   (See figure in printed
   edition.)

Note:  One CGC did not respond to the question regarding root stock
traits. 


--------------------
\1 Managing Global Genetic Resources:  Agricultural Crop Issues and
Policies, National Research Council (Washington, D.C.:  National
Academy Press, 1993). 

\2 Root stocks are used in grafting clonal crops. 


      ALMOST HALF THE CGCS
      REPORTED THAT
      CHARACTERIZATION INFORMATION
      IS INSUFFICIENT
-------------------------------------------------------- Chapter 3:1.2

Characterization data provide information on highly inheritable
traits that are little influenced by varying environmental
conditions.  These data help distinguish germplasm samples of the
same type of plant from one another and provide a baseline for
ensuring that the genetic integrity of a germplasm sample is
maintained.\3 It is generally the responsibility of NPGS curators to
develop characterization information when they regenerate germplasm
samples. 

Nineteen of the 40 CGCs reported that characterization information on
their NPGS germplasm is somewhat or very insufficient for crop
breeding.  These 19 CGCs included some economically important crops,
such as cotton, grapes, and peanuts.  Only nine CGCs reported that
characterization information for their crops' germplasm is somewhat
sufficient for breeding.  Figure 3.3 shows the sufficiency of
characterization information, as reported by the CGCs.  In addition,
over half the CGCs said that NPGS' management of characterization
data has improved since 1990. 

   Figure 3.3:  CGCs' Perceptions
   of the Sufficiency of
   Characterization Information
   for Crop Breeding

   (See figure in printed
   edition.)


--------------------
\3 During germplasm regeneration, considerable care must be taken to
minimize genetic shifts to the resulting seeds, or offspring. 
Genetic markers measured in characterization can be used to determine
whether shifts have occurred. 


      MOST CGCS REPORTED THAT
      PASSPORT INFORMATION IS
      SUFFICIENT FOR CROP
      BREEDING, BUT NPGS MANAGERS
      SAID IT IS INADEQUATE FOR
      THEIR PURPOSES
-------------------------------------------------------- Chapter 3:1.3

Passport information includes the data on the plant's classification,
the location of the germplasm sample's origin, and the ecology of
that location.  This information is essential for assessing the
quality of the collections and for using and managing these
collections.\4

NPGS uses the data to ensure, for example, that it does not
unnecessarily collect samples that have previously been collected
from the same location.\5 Passport data are generally the first data
obtained on a new germplasm sample and are often provided by the
donor when the germplasm is given to NPGS.  However, much germplasm
is donated to NPGS without complete passport information.\6

Although NPGS' passport information may be incomplete, the CGCs were
considerably more positive about the passport information than about
either evaluation or characterization information.  As shown in
figure 3.4, almost three-quarters of the CGCs reported that passport
information for their crops is somewhat or very sufficient for
crop-breeding purposes.  Only five CGCs reported passport information
to be somewhat insufficient for breeding. 

   Figure 3.4:  CGCs' Perceptions
   of the Sufficiency of Passport
   Information for Crop Breeding

   (See figure in printed
   edition.)

Furthermore, three-quarters of the CGCs said that NPGS' management of
passport data has improved since about 1990. 

Although most CGCs found passport information to be somewhat or very
sufficient for crop-breeding purposes, NPGS officials told us that it
is not sufficient for their internal planning for germplasm
acquisition.  About two-thirds of NPGS' samples lack passport data on
the location of origin, according to the GRIN data provided by NPGS
officials.  This information is key to pinpointing areas where
germplasm has already been collected, thereby minimizing the
possibility of unnecessarily collecting material already in the NPGS
collection.  Origin information also assists in targeting sites for
future collection trips.  Furthermore, according to NPGS officials,
even when location information is available, it is sometimes
inaccurate or incomplete.  GRIN data, for example, show that 90
percent of NPGS' samples have no information on the latitude and
longitude of the site of origin. 

Incomplete passport information also makes it more difficult for
curators to determine which samples are unique and which are
duplicates.\7 Identification of duplicate samples is necessary to
avoid needless duplication of costly germplasm-related activities,
such as preservation, characterization, and evaluation.  Curators for
about half of the crop collections reported that it is moderately to
extremely important to decrease the duplication of samples in their
NPGS collection.  For example, the sorghum curator estimated that
about 10 to 25 percent or more of the samples in the sorghum
collection are duplicates.  He added that the elimination of these
duplicates would be expensive and time-consuming because many samples
lack complete passport data. 


--------------------
\4 Managing Global Genetic Resources:  Agricultural Crop Issues and
Policies, National Research Council (Washington, D.C.:  National
Academy Press, 1993). 

\5 On the other hand, NPGS may use passport data to resample rich
areas or to recover lost samples from the same location. 

\6 For older samples, this information will likely be unobtainable
for various reasons--e.g., the original collector did not provide it
or no other relevant records are available. 

\7 Curators responding to GAO's survey were more negative regarding
passport information than the CGCs.  Curators on 15 CGCs found
passport information insufficient for their crops; curators on 18
CGCs found it sufficient.  On the CGCs that had strong differences of
opinion among members, curators may have focused on a different
aspect of the information (e.g., taxonomy versus site of origin) than
other CGC members. 


   NEEDED INFORMATION IS NOT
   AVAILABLE FOR SEVERAL REASONS
---------------------------------------------------------- Chapter 3:2

While some information has not been developed because of resource
constraints, even data that have been developed have not always been
entered into GRIN.  NPGS officials told us that developing,
obtaining, and documenting information in GRIN are lower priorities
than preserving the germplasm collections, and in some cases, these
activities are outside the system's control. 


      SOME INFORMATION HAS NOT
      BEEN DEVELOPED OR ENTERED
      INTO THE DATABASE
-------------------------------------------------------- Chapter 3:2.1

Thirty-nine CGCs estimated that, on average, 50 percent of existing,
useful evaluation data on their collections are not in GRIN.\8
According to the NPGS managers of several sites and ARS officials who
oversee crop-specific research programs, gaps in evaluation data for
NPGS germplasm result from a variety of factors, including the large
amount of germplasm that needs to be evaluated, the
resource-intensive nature of evaluations, and limited resources.  In
addition, most germplasm evaluations are conducted outside of NPGS,
primarily by ARS and university scientists who do not always provide
NPGS with the resulting information for entry into GRIN.  Thus, even
when evaluation data exist, they are not always available through
GRIN.  Some scientists who conduct germplasm evaluations are funded
by ARS and are required to submit their evaluation results to NPGS. 
However, other scientists, not funded by ARS, conduct evaluations as
part of their larger research objectives.  According to a former
National Program Leader for Plant Genetic Resources, some of these
evaluations merit inclusion in GRIN; however, he said that NPGS does
not have a clear policy on the curators' responsibility in obtaining
this information. 

Several CGC reports developed for NPGS have identified the need to
enter additional evaluation information into GRIN.  For example, the
1996 corn CGC report stated that much evaluation data had accumulated
without being entered into GRIN or otherwise disseminated. 
Furthermore, according to the 1996 CGC report for cucurbits (e.g.,
squash, watermelon, cucumbers), NPGS has had relatively few requests
for watermelon germplasm, in part because of the lack of relevant
evaluation data in GRIN. 

In addition, NPGS does not have a process for tracking whether
scientists under agreement with ARS to evaluate NPGS germplasm have
submitted evaluation data for entry into GRIN.  As a result, NPGS has
little assurance that the results of these ARS-supported evaluations
are entered into GRIN.  While several NPGS managers said they believe
that most of this information is in GRIN, NPGS is nonetheless
developing a system to track the information.  The system is expected
to be completed by early 1998. 

Finally, some passport information--for example, the location of
origin--cannot be developed because the germplasm samples were
provided many years ago, and it would be very difficult or impossible
to reconstruct the missing data.  In addition, some passport
information may be available but has not been added to GRIN. 
Although GRIN may not have complete data, 36 CGCs reported that it
effectively provides information about their NPGS germplasm
collections.  Thirty-seven CGCs reported that NPGS' management of
GRIN had improved since about 1990, making it the NPGS activity that
was cited most frequently as having improved.\9


--------------------
\8 Members of one CGC provided no estimate. 

\9 CGCs were asked how much NPGS management of 13 activities had
improved or worsened since about 1990. 


      SEVERAL NPGS MANAGERS STATED
      THAT MAINTAINING GERMPLASM
      VIABILITY IS A HIGHER
      PRIORITY THAN
      INFORMATION-RELATED
      ACTIVITIES
-------------------------------------------------------- Chapter 3:2.2

According to several NPGS officials responsible for managing
germplasm activities, preserving germplasm to keep it viable is of
more fundamental importance than developing information and making it
available.  In addition, the total number of germplasm samples in
NPGS' collections has increased about 29 percent from 1986 through
1996, according to the GRIN data provided by an NPGS official.  With
larger collections come greater demands on curators' time and
resources.  Therefore, the development and documentation of
characterization information, which is done primarily by NPGS
curators, occurs only as time permits.  A case in point is the
cucurbit collection.  The CGC for cucurbits reported that
characterization and evaluation information is insufficient for
breeding of its crops.  However, the curators for these crops
reported that some cucurbit regeneration backlogs had increased and
that between 5 and 40 years would be required to regenerate various
parts of this collection given current resources. 


PRESERVATION ACTIVITIES HAVE NOT
KEPT PACE WITH NEEDS OF THE
COLLECTION
============================================================ Chapter 4

Preservation activities--including viability testing, germplasm
regeneration, and secure, long-term backup storage of germplasm--have
not kept pace with the preservation needs of the collections.  First,
only minimal viability testing--testing that determines the amount of
live germplasm in a sample--has been conducted at some sites,
including two plant introduction stations that account for over
one-fourth of NPGS' germplasm samples.  Viability testing is needed
to determine when germplasm should be reproduced to prevent the loss
of the sample.  Second, NPGS has significant backlogs for
regenerating germplasm at all four plant introduction stations. 
Regeneration--reproducing germplasm to obtain sufficient numbers of
viable seeds--is essential, particularly when viability is known to
be low or has not been tested.  Third, over one-third of NPGS'
germplasm is not backed up in NPGS' National Seed Storage Laboratory
(NSSL), which provides secure, long-term storage for the system. 
Germplasm that is not backed up at NSSL is at greater risk of being
lost. 


   MUCH GERMPLASM AT TWO MAJOR
   LOCATIONS HAS NOT BEEN TESTED
   FOR VIABILITY
---------------------------------------------------------- Chapter 4:1

NPGS' standards require that viability testing be conducted as often
as is needed for each species.  Managers of three plant introduction
stations stated that the germplasm in their collections should be
tested every 5 to 10 years, depending on the species and the storage
conditions for the germplasm.\1 Viability testing is important to
determine when the sample is at risk of being lost. 

According to NPGS data and NPGS officials, the amount of testing at
some locations--including two of the four plant introduction
stations--is insufficient.  These two stations account for more than
one-quarter of NPGS' active collection.  The stations--in Griffin,
Georgia, and Pullman, Washington--had tested less than one-fourth of
their germplasm from 1986 through 1996.\2 A curator at the Griffin
station cited a specific consequence of the failure to test for
viability on a regular basis--all 10 samples of recently tested
butternut squash were dead.  The collection had previously not been
tested for many years.  As a result, he feared that much or all of
this collection of about 500 samples--the only one of its kind in
NPGS--may be dead. 

While agreeing that viability testing is important, the Griffin and
Pullman station managers told us that, given their large regeneration
backlogs, focusing their limited resources on regeneration to
maintain germplasm viability is more likely to save diversity in the
germplasm collections than testing the germplasm.  Other obstacles
cited as reasons for infrequent testing include the large numbers of
different species to test and the lack of testing methods for some of
them. 

NSSL also conducts viability tests on the germplasm it maintains in
long-term storage.  At NSSL, 82 percent of its samples have been
tested, 69 percent from 1985 through 1996.  Of the 18 percent never
tested, 61 percent do not have enough seeds for testing,\3 and 39
percent are part of a backlog that has not yet been processed because
of the lack of resources, according to NSSL data and NPGS officials. 

While NPGS' data indicate that viability testing is not conducted as
often as it should be, responses to our survey on the sufficiency of
viability testing were mixed.  Only 4 of the 40 CGCs we surveyed
reported that NPGS' viability testing activities are insufficient for
their crops, although 29 indicated that the current staff levels for
testing (as well as for regeneration) have hindered the preservation
of their collections.  However, when we examined the responses of the
curators alone--who are responsible for maintaining and preserving
the collections and are most knowledgeable about their
condition--curators for part or all of 16 of 38 crop collections
(including major crops such as corn, alfalfa, and cotton) reported
that viability testing for their crop collections is insufficient.\4
For example, the curator responsible for over 80 percent of the corn
collection reported that regeneration and viability testing are
somewhat insufficient and should be the first priority in case of
additional funding. 


--------------------
\1 Viability testing is conducted primarily on seeds because the
viability of clonal material can generally be determined by
observation.  The leader of the Geneva Plant Introduction Station
stated that in the future this site's germplasm will need to be
tested only every 10 to 30 years because the collection is now stored
at about -18 degrees Celsius. 

\2 In contrast, at the two other stations--Ames, Iowa, and Geneva,
New York--about 60 percent or more of the germplasm had been tested
for viability in the past 10 years. 

\3 Some of these are seeds of special genetic stocks that will be
used in research and should not be sacrificed for germination tests. 

\4 Curators for two CGCs reported having no basis to judge.  In
addition, 15 CGCs have multiple curators on their committees, each of
whom is responsible for parts of the collection. 


   NPGS HAS SIGNIFICANT BACKLOGS
   OF GERMPLASM REQUIRING
   REGENERATION
---------------------------------------------------------- Chapter 4:2

Regeneration is necessary to ensure that NPGS has an adequate supply
of viable seeds.  NPGS generally schedules a sample for regeneration
when the viability of the sample is low--i.e., more than 35 percent
of the sample's seeds are dead--or the quantity of seeds is too low
for distribution.  NPGS has significant backlogs of germplasm
requiring regeneration.  According to NPGS officials, large backlogs
may cause the loss of diversity in collections or prevent
distribution to users and to NSSL for secure backup. 

NPGS officials from two plant introduction stations told us that,
generally, their sites' germplasm that is low in viability or
quantity should be regenerated within 2 to 5 years in order to
minimize the loss of diversity in their collections over the long
term.  However, it may take as much as 75 to 100 years for the
samples at these two locations that need regenerating to be
regenerated, according to NPGS curators.  Table 4.1 shows the
estimated number of years required to regenerate samples, at current
resource levels, for various crops at the four plant introduction
stations, as of Spring 1997.  Some of these years are underestimated
because they do not include the regeneration that would be required
to provide germplasm for secure backup to NSSL and material to users
that has been correctly regenerated.\5



                               Table 4.1
                
                 Estimated Years Required to Regenerate
                 the Samples of Major Seed Crops at the
                 Plant Introduction Stations at Current
                            Resource Levels

                                         Percent                Median
                                 Total  of these    Range of     years
                       Total    number   samples       years  required
                   number of        of  requirin    required        to
Plant                  major   samples         g          to  regenera
introduction            seed  in these  regenera  regenerate        te
station                crops     crops      tion     samples   samples
----------------  ----------  --------  --------  ----------  --------
Ames, Iowa                10    35,300        35        5-23        10
Geneva,                    6     8,900        35        3-20         5
 New York
Griffin, Georgia           9    63,690        16       5-100        10
Pullman,                  17    63,932        51        3-75         7
 Washington
----------------------------------------------------------------------
Notes:  Major seed crops are those representing the station's largest
collections.  Although these data are primarily for seed crops, a
small number of clonally propagated samples are included.  In
addition, sites did not provide estimates for the years required to
regenerate the samples for a few crops. 

Source:  Estimates were provided by each of the four plant
introduction stations. 

As table 4.1 shows, of the four plant introduction stations, the
Pullman, Washington, location has the biggest backlog in terms of the
percentage of samples requiring regeneration.  Such regeneration is
important not only for preservation of diversity but also for
supplying seed to NSSL for long-term, secure backup. 

Several factors contribute to these backlogs.  The biggest single
factor is the limited number of permanent employees and seasonal
laborers available to manage and carry out the necessary field and
greenhouse activities, according to NPGS officials.  Furthermore, at
some locations, facilities for regeneration are inadequate, and at
others the growing conditions for germplasm are less than ideal for
producing good yields of high-quality seed.\6 For some collections,
these regional climatic conditions also contribute to the development
of pests and pathogens, which can hinder the preservation and use of
germplasm.\7 To overcome these problems and increase its capacity to
regenerate quality seed, NPGS recently established a new site--at
Parlier, California--that is in an arid region with a long growing
season.  The Department has requested increased funding for genetic
resources research in the fiscal year 1998 budget, part of which is
to increase regeneration capability, according to an NPGS official. 

CGC responses to our survey regarding the sufficiency of regeneration
activities were similar to those on viability testing.  Only 7 of the
40 CGCs we surveyed reported that NPGS' regeneration activities are
insufficient for their collections, although 29 CGCs reported that
the lack of staff for regeneration and viability testing had hindered
the preservation of their collections.  When we examined the
responses of the curators (those most knowledgeable about the
collections' conditions), curators for part or all of 15 of 39 crop
collections reported that regeneration is insufficient for part or
all of their crop collections.\8 The curator responsible for most of
NPGS' corn collection reported that regeneration is insufficient and
that the 15-year regeneration backlog for corn placed an important
part of this collection at the risk of losing diversity. 


--------------------
\5 According to NPGS officials, in past decades germplasm in some
collections was regenerated incorrectly because of inadequate
curatorial knowledge, adverse environmental conditions (e.g., hail,
windstorm), or lack of resources.  For example, some germplasm was
regenerated with an insufficient plant population and some without
controlling pollination to prevent contamination from other plants. 
NPGS officials believe that practices involving human error have
largely been eliminated.  However, some of this germplasm still needs
to be replaced through regeneration using correct methods. 

\6 Curators for part or all of 16 of 40 crop collections--including
corn and tomato-- reported that the ability to produce high-quality
seed or maintain clonal crops at present sites hindered the
preservation of their collections. 

\7 Curators for part or all of the 14 crop collections reported that
the ability to test for and maintain pathogen-free collections
hindered the preservation of their collections. 

\8 Curators for one CGC reported having no basis to judge in response
to this question. 


   MUCH GERMPLASM IS NOT IN
   LONG-TERM BACKUP STORAGE
---------------------------------------------------------- Chapter 4:3

Although NPGS' policy requires that all seed samples in active
collections be backed up at NSSL, over one-third are not. 
Furthermore, methods to ensure the secure backup of most clonal
germplasm have not yet been developed.  Backup is needed to provide
protection against losses at the active sites resulting from (1)
deterioration, which generally occurs more rapidly in seeds stored at
active sites, or (2) human error, extreme weather, equipment failure,
flood, fire, vandalism, or other catastrophes. 

Sixty-one percent of the approximately 440,000 seed samples at NPGS'
active sites are backed up at NSSL, where they are stored at -18
degrees Celsius or in containers over liquid nitrogen to slow
deterioration.\9 Of these backed-up samples, 44 percent do not meet
NPGS' standards and goals for the quantity of seeds and the
percentage that should be viable--65 percent.  The seed samples not
stored at NSSL are at increased risk of deterioration because seeds
generally deteriorate much more rapidly at active sites, which
generally store germplasm at warmer temperatures--5 degrees
Celsius.\10

According to NPGS officials, seeds have not been adequately backed up
primarily because of the large regeneration backlogs at active sites. 
That is, until the sites regenerate germplasm, they often do not have
a sufficient number or quality of seeds to send to NSSL for backup
storage.  In addition, even when they have sufficient quantities of
seeds, some sites have not sent the seeds to NSSL because before they
can be sent, the sites must reinventory the germplasm samples and
repackage the seeds.  According to NPGS officials, these activities
use resources that are in short supply.  In addition, NSSL has its
own 16-month backlog of about 27,000 samples that must be processed
(which includes viability testing) before being placed in secure,
long-term storage. 

The backup of clonal samples is even more limited, with only 4
percent of the approximately 30,000 samples at the active sites
backed up at NSSL.  This limited backup occurs because the methods
for providing secure, long-term storage for most clonal germplasm
have not yet been developed.\11 Clonal germplasm may be backed up--in
greenhouses as living plants, as tissue culture, or through
cryopreservation--at the active site where the primary collection is
maintained.  Thus, in case of a natural disaster, disease, or other
catastrophe, both the active and backup samples could be destroyed. 
For example, in 1992, over 2,000 germplasm samples were lost at NPGS'
Miami facility following Hurricane Andrew.  These samples were not
backed up at another NPGS site or at NSSL.  Included in this group
were about 30 percent of the mango and avocado collections and about
50 percent of the site's ornamental collection (e.g., palm trees). 
The storm uprooted the trees, and they could not be successfully
replanted.  The curator for these crops stated that most of this
material will not be replaced because of resource constraints,
difficulties in locating the material, and difficulties in getting
foreign collections to provide replacement samples. 

CGC responses to our survey regarding the sufficiency of backup
storage of germplasm varied.  Only 6 of the 40 CGCs surveyed reported
that NPGS' activity in the area of backup storage/preservation is
insufficient for their crop collections.  In contrast, the curators
for part or all of 15 of 40 crop collections reported that NPGS'
activity in the area of backup storage/preservation of their crop
collection is insufficient.\12 The curators for the collections of
six major crops--corn, soybeans, wheat, alfalfa, potato, and
cotton--reported no insufficiencies in this area. 


--------------------
\9 According to the director of NSSL, a higher percentage of the
germplasm of the 50 most important crops is backed up. 

\10 While plant introduction stations have recently acquired some -18
degrees Celsius storage capacity, most of their germplasm is still
stored at 5 degrees Celsius. 

\11 According to ARS' Assistant Administrator for Genetic Resources,
research on methodologies for clonal crop cryopreservation will be
NSSL's highest research priority if new funding is made available. 

\12 Curators for nine collections--including citrus fruits, peanuts,
and sugarcane--indicated that backup was insufficient for their
collections overall. 


SURVEY METHODOLOGY
=========================================================== Appendix I

We surveyed crop germplasm experts identified by NPGS.  These experts
included the 542 members of 40 Crop Germplasm Committees (CGCs),
including all NPGS curators and CGC chairs; 27 recently retired CGC
members; and 38 experts who were not serving on a CGC.  Forty-five of
those surveyed served on more than one CGC and were asked to complete
one survey for each CGC on which they served.  For the purposes of
our survey, experts not currently serving on a CGC were assigned
membership on the CGC that represented their area of expertise.  In
all, we mailed questionnaires to 680 CGC "members"--one questionnaire
to each of the 562 members serving on one CGC and 118 questionnaires
to the 45 experts serving on more than one CGC.  We followed up this
initial mailing with additional mailings and telephone calls to
encourage response.  We conducted our survey from November 1996
through March 1997. 

We received a total of 576 usable questionnaires, including responses
from all the NPGS curators, for a response rate of 85 percent.  Only
two CGC chairs did not participate in the survey (alfalfa and small
fruits).  Response rates varied across CGCs, from a low of 57 percent
for the vigna and pepper CGCs to a high of 100 percent for three CGCs
(corn, sugarbeets, and tobacco).  Response rates were above 70
percent for all but four CGCs (cotton, new crops, peppers, and
vigna).  The median response rate for CGCs was 86 percent. 

We analyzed the survey results by CGC.  To obtain a single CGC
response for each question, we aggregated the responses of the CGC
members on that committee.  We performed this aggregation by first
selecting only those CGC members who had a substantive opinion on a
particular question (that is, the member did not select "no basis to
judge" as his or her response).  We did not use the opinion if the
question asked about the entire NPGS collection but the respondent
answered for only a minor portion of the collection, unless the
respondent was an NPGS curator.  The selected members' responses were
aggregated by using one of three statistics, depending on the type of
question.  The mean response was used for questions requiring a
numeric response.  (See, for example, app.  II, questions 12 and 44.)
The median response was used for questions requiring an evaluation of
the NPGS collection or NPGS management.  (See, for example, app.  II,
questions 7 and 42.) When the median was between two rating
categories, we reported the results in the category with the lower
intensity.  For questions that required the respondent to sort
information into nonnumeric, nonrating categories, we used the
percentage of CGC members who selected each category to represent the
CGC response.  (See, for example, app.  II, question 11.)

Appendix II contains a copy of our survey with the results aggregated
by CGC.  In order to report the data completely and show instances in
which the median was between two rating categories, we altered the
original format of the questionnaire by deleting the response option
"no basis to judge" from questions 17 and 18 and changing the size of
the response boxes for these and several other questions.  We used
the letter "t" to indicate the number of medians that were between a
given category and the next most intense category for that question. 




(See figure in printed edition.)Appendix II
RESULTS OF GAO'S SURVEY OF CROP
GERMPLASM COMMITTEES
=========================================================== Appendix I



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


CROP GERMPLASM COMMITTEES AND THE
CROPS FOR WHICH THEY ARE
RESPONSIBLE
========================================================= Appendix III

CGC             Crop                  Subcrop                      Total samples
--------------  --------------------  ----------------------  ------------------
Alfalfa         Alfalfa               Alfalfa                              3,003

                                      Wild relatives of                    4,515
                                      alfalfa

                                                                           7,518

Apple           Apple                 Apple                                2,563

                                      Wild relatives of                    2,246
                                      apple

                                                                           4,809

Barley          Barley                Barley                              28,338

                                      Wild relatives of                    2,074
                                      barley

                                                                          30,412

Carya           Chestnut              Chestnut                                18

                Pecan                 Pecan                                  563

                                      Wild relatives of                      318
                                      pecan

                                                                             899

Citrus          Citrus                Grapefruit                              59

                                      Lemon                                   69

                                      Lime                                    21

                                      Orange                                 236

                                      Orange, sour                            45

                                      Pummelo                                 93

                                      Wild relatives of                      453
                                      citrus

                Date Palm             Date Palm                               98

                Kumquat               Kumquat                                 13

                                                                           1,087

Clover          Astragalus            Astragalus                             852

                Clover                Clover, crimson                         40

                                      Clover, red                          1,284

                                      Clover, sweet                          896

                                      Clover, white                          822

                                      Wild relatives of                    3,781
                                      clover

                Lespedeza             Lespedeza                              152

                Trefoil               Trefoil                                930

                                                                           8,757

Cool season     Chickpea              Chickpea                             4,434
food legume

                                      Wild relatives of                      174
                                      chickpea

                Faba bean             Faba bean                              538

                                      Wild relatives of faba               1,381
                                      bean

                Lentil                Lentil                               2,724

                                      Wild relatives of                      149
                                      lentil

                Lupins                Lupins                               1,287

                                                                          10,687

Cotton          Cotton                Cotton                               4,810

                                      Wild relatives of                    2,099
                                      cotton

                                                                           6,909

Crucifer        Crucifers             Broccoli                                88
                (Brassicas)

                                      Brussel sprouts                         84

                                      Cabbage                              1,032

                                      Canola                                 422

                                      Cauliflower                            504

                                      Mustard                              1,100

                                      Oil Brassica                           544

                                      Rapeseed                               655

                                      Rutabaga                                24

                                      Turnip                                 139

                                      Wild relatives of                    1,250
                                      crucifers

                Radish                Radish                                 748

                                      Wild relatives of                       10
                                      radish

                                                                           6,600

Cucurbit        Cucumber              Cucumber                             1,551

                Melon                 Melons (honeydew,
                                      cantaloupe)                          3,069

                Melon/cucumber        Wild relatives of                      580
                                      melon/cucumbers

                Squash                Pumpkin                                891

                                      Squash                                 831

                                      Zucchini squash                      1,127

                                      Wild relatives of                      531
                                      squash

                Watermelon            Watermelon                           1,862

                                      Wild relatives of                       34
                                      watermelon

                                                                          10,476

Grape           Grape                 Grape                                1,183

                                      Wild relatives of                    2,726
                                      grapes

                                                                           3,909

Grass           Andropogon            Andropogon                           1,100

                Bentgrass             Bentgrass                              254

                Bermudagrass          Bermudagrass                           524

                Bluegrass             Bluegrass                              837

                Bothriochloa          Bothriochloa                           672

                Bouteloua             Bouteloua                              110

                Bromegrass            Bromegrass                           1,071

                Buchloe               Buchloe                                 13

                Canarygrass           Canarygrass                            759

                Cenchrus              Cenchrus                               857

                Digitaria             Digitaria                              652

                Elytrigia             Elytrigia                              835

                Fescue                Fescue                               2,050

                Gammagrass            Gammagrass                              93

                                      Wild relatives of                      105
                                      gammagrass

                Millet, Italian       Millet, Italian                        759

                                      Wild relatives of                      248
                                      Italian millet

                Millet, pearl         Millet, pearl                        1,137

                                      Wild relatives of                      266
                                      pearl millet

                Oatgrass              Oatgrass                               228

                Orchardgrass          Orchardgrass                         1,464

                Panicum               Millet                                 724

                                      Wild relatives of                    1,128
                                      panicum

                Paspalum              Paspalum                             1,501

                Ryegrass              Ryegrass                             1,335

                Timothy               Timothy                                626

                Wheatgrasses          Wheatgrasses                         1,679

                Wild ryegrass         Wild ryegrass                          555

                Zoysia                Zoysia                                 119

                                                                          21,701

Herbaceous      Aster                 Aster                                   10
Ornamental

                Begonia               Begonia                                  4

                Chrysanthemum         Chrysanthemum                           23

                Day Lily              Day Lily                                 8

                Dianthus              Dianthus                                90

                Euphorbs              Poinsettia                               3

                Gentian               Gentian                                  1

                Geranium              Geranium                                 3

                Impatiens             Impatiens                               18

                Liatris               Liatris                                 12

                Lily                  Lily                                    28

                Petunia               Petunia                                 96

                Zinnia                Zinnia                                  80

                                                                             376

Juglans         Walnut                Walnut                                 266

                                      Walnut, black                           35

                                      Wild relatives of                      162
                                      walnut

                                                                             463

Leafy           Celery                Celery                                  86
vegetable

                                      Wild relatives of                      129
                                      celery

                Chicory               Chicory                                250

                Lettuce               Lettuce                              1,282

                                      Wild relatives of                      222
                                      lettuce

                Parsnip               Parsnip                                 63

                Spinach               Spinach                                379

                                                                           2,411

Maize           Corn                  Corn                                23,414

                                      Wild relatives of corn                 251

                                                                          23,665

New Crops A     maranth A             maranth 1                             ,818

                                      Wild relatives of                    1,482
                                      amaranth

                Apios                 Apios                                    3

                Calendula             Calendula                               87

                Castor bean           Castor bean                          1,032

                Crambe                Crambe                                 304

                Crotalaria            Crotalaria                             260

                Cuphea                Cuphea                                 808

                Euphorbs              Wild relatives of                       87
                                      euphorbia

                Evening primrose      Evening primrose                       614

                Guar                  Guar                                 1,303

                Guayule               Guayule                                187

                Jojoba                Jojoba                                 155

                Kenaf                 Kenaf                                  306

                                      Roselle                                144

                                      Wild relatives of                      350
                                      kenaf

                Lesquerella           Lesquerella                            136

                Leucaena              Leucaena                               573

                Lunaria               Lunaria                                  6

                Meadowfoam            Meadowfoam                              56

                Mesquite              Mesquite                                73

                Perilla               Perilla                                 22

                Quinoa                Quinoa                                 169

                                      Wild relatives of                       52
                                      quinoa

                Safflower             Safflower                            2,321

                                      Wild relatives of                      120
                                      safflower

                Sesame                Sesame                               1,221

                                      Wild relatives of                        9
                                      sesame

                Stokes Aster          Stokes Aster                            39

                Vernonia              Vernonia                               267

                Yucca                 Yucca                                   15

                                                                          14,019

Oat             Oat                   Oat                                 10,269

                                      Wild relatives of oat               11,597

                                                                          21,866

Pea             Pea                   Pea                                  4,245

                                      Wild relatives of pea                  222

                                                                           4,467

Peanut          Peanut                Peanut                               8,434

                                      Wild relatives of                    1,115
                                      peanut

                                                                           9,549

Peppers         Peppers               Peppers                              2,594

                                      Wild relatives of                    1,399
                                      pepper

                                                                           3,993

Phaseolus       Bean                  Bean                                11,560

                                      Bean, lima                           1,063

                                      Wild relatives of bean               1,192

                                                                          13,815

Potato          Potato                Potato                               1,312

                                      Wild relatives of                    5,778
                                      potato

                                                                           7,090

Prunus          Stone fruits          Almond                                 117

                                      Apricot                                325

                                      Cherry                                 395

                                      Nectarine                                9

                                      Peach                                  436

                                      Plum                                   237

                                      Wild relatives of                    1,224
                                      stone fruits

                                                                           2,743

Pyrus           Pear                  Pear                                   939

                                      Wild relatives of pear               1,368

                                                                           2,307

Rice            Rice                  Rice                                18,332

                                      Wild relatives of rice                 241

                                                                          18,573

Root and Bulb   Carrot                Carrot                                  55

                                      Wild relatives of                      824
                                      carrot

                Onion/Garlic          Garlic                                 122

                                      Leek                                     2

                                      Onion                                1,081

                                      Wild relatives of                      901
                                      onion/garlic

                                                                           2,985

Small Fruit     Blueberry             Blueberry                              205

                Cranberry             Cranberry                              121

                Blueberry/cranberry   Wild relatives of                      864
                                      blueberry/cranberry

                Currant/Gooseberry    Currant/gooseberry                   1,084

                Raspberry             Raspberry                              336

                                      Wild relatives of                    1,384
                                      raspberry

                Strawberry            Strawberry                             504

                                      Wild relatives of                    1,018
                                      strawberry

                                                                           5,516

Sorghum         Sorghum               Sorghum                             39,931

                                      Wild relatives of                      684
                                      sorghum

                                                                          40,615

Soybean         Soybean               Soybean                             17,420

                                      Wild relatives of                    1,833
                                      soybean

                                                                          19,253

Sugarbeet       Beet                  Beet                                 1,567

                                      Wild relatives of beet                 715

                                                                           2,282

Sugarcane       Sugarcane             Sugarcane                              919

                                      Wild relatives of                    2,360
                                      sugarcane

                                                                           3,279

Sunflower       Sunflower             Sunflower                            2,673

                                      Wild relatives of                    1,202
                                      sunflower

                                                                           3,875

Sweet Potato    Sweet potato          Sweet potato                           720

                                      Wild relatives of                      452
                                      sweet potato

                                                                           1,172

Tobacco         Tobacco               Tobacco                              1,841

                                      Wild relatives of                      305
                                      tobacco

                                                                           2,146

Tomato          Tomato                Tomato                               8,123

                                      Wild relatives of                    1,983
                                      tomato

                                                                          10,106

Tropical Fruit  Avocado               Avocado                                474
and Nut

                                      Wild relatives of                       14
                                      avocado

                Banana                Banana                                 184

                Brazil nut            Brazil nut                               1

                Breadfruit            Breadfruit                              66

                Cashew                Cashew                                   1

                Cherimoya             Cherimoya                               86

                Coffee                Coffee                                   1

                Guava                 Guava                                   83

                Kiwi                  Kiwi                                    12

                                      Wild relatives of kiwi                  63

                Litchi nut            Litchi nut                             135

                Macadamia             Macadamia                               27

                Mango                 Mango                                  295

                Papaya                Papaya                                 154

                                      Wild relatives of                       23
                                      papaya

                Passion fruit         Passion fruit                           36

                Pineapple             Pineapple                              137

                                      Wild relatives of                       25
                                      pineapple

                Rambutan              Rambutan                                39

                Star fruit            Star fruit                              70

                                                                           1,926

Vigna           Cowpea                Cowpea (blackeyed pea)               7,783
                (blackeyed pea)

                                      Adzuki bean                            302

                                      Black gram                             303

                                      Mung bean                            3,919

                                      Wild relatives of                      503
                                      Vigna

                                                                          12,810

Wheat           Rye                   Rye                                  1,815

                                      Wild relatives of rye                  106

                Triticale             Triticale                            1,411

                Wheat                 Wheat                               34,618

                                      Wheat, durum                         6,901

                                      Wild relatives of                    7,685
                                      wheat

                                                                          52,536

Woody           Arborvitae            Arborvitae                               9
Landscape

                Barberry              Barberry                                35

                Cedar                 Cedar                                    3

                Cypress               Cypress                                 12

                Dogwood               Dogwood                                170

                Elm                   Elm                                     59

                Fir                   Fir                                     22

                Hemlock               Hemlock                                 16

                Holly                 Holly                                  130

                Juniper               Juniper                                 71

                Larch                 Larch                                    5

                Lilac                 Lilac                                   35

                Magnolia              Magnolia                                44

                Maple                 Maple                                  225

                Oak                   Oak                                     57

                Pine                  Pine                                    81

                Privet                Privet                                  37

                Redbud                Redbud                                  66

                Rhododendron          Rhododendron                           100

                Rose                  Rose                                   150

                Silverbell            Silverbell                             106

                Sourwood              Sourwood                                 6

                Spiraea               Spiraea                                 50

                Spruce                Spruce                                  20

                Viburnum              Viburnum                               105

                Yew                   Yew                                     20

                                                                           1,634

================================================================================
All CGCs                                                                 399,236
--------------------------------------------------------------------------------
Notes:  The information in this appendix was provided by NPGS
officials from the GRIN database as of February 28, 1997.  In
addition to the 399,236 germplasm samples shown above, NPGS maintains
more than 35,000 other samples that are not listed here because they
have no CGCs providing advice and guidance on them. 




(See figure in printed edition.)Appendix IV
COMMENTS FROM THE U.S.  DEPARTMENT
OF AGRICULTURE
========================================================= Appendix III



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


The following are GAO's comments on USDA's September 17, 1997,
letter. 


   GAO'S COMMENTS
------------------------------------------------------- Appendix III:1

1.  We agree that NPGS has made improvements in a number of areas
over the past 6 years, as USDA discusses in the attachment to its
letter.  However, the purpose of our review was to obtain the views
of the CGCs--crop experts who advise NPGS--on the sufficiency of
NPGS' principal activities:  acquisition, development and
documentation of information, and preservation of germplasm.  Thus,
the report focuses on the current status of NPGS' activities and not
on improvements made to the system.  However, chapter 1 discusses
actions taken during the 1990s to address identified shortcomings--in
particular, the expansion of NSSL's long-term storage capacity, the
increased use of -18 degree Celsius storage by NPGS sites, and
improvements made to the GRIN database.  In addition, other chapters
discuss areas where most CGCs reported that aspects of NPGS
collections or activities were sufficient.  Therefore, given the
purpose of our review and the language already incorporated into the
report, we did not add information on other improvements. 

2.  While our report cites curators and CGCs as having different
views on the sufficiency of some NPGS activities--e.g., preservation
and passport information--they do not, for the most part, have
different views on NPGS' top priorities.  According to survey
responses, both curators and CGCs, on average, viewed acquisition as
their top priority if additional funding becomes available. 
Development and documentation of characterization information is also
ranked highly by curators and CGCs (they ranked it second and third,
respectively), as is development and documentation of evaluation
information, which is ranked fifth by curators and second by CGCs. 
On the other hand, there were greater differences in the CGCs' and
curators' ranking of regeneration and viability testing, with
curators ranking it third and CGCs, eighth.  (See app.  II, question
44.)

3.  We wish to clarify USDA's interpretation of our survey results. 
While chapter 3 notes that almost all CGCs reported that the
management of GRIN has improved since about 1990 and three-quarters
said that the management of passport data had improved, the survey
results are less clear-cut with regard to the management of
characterization and evaluation data.  Specifically, over half (22)
the CGCs said that the management of characterization data has
improved, 17 said that there is no change, and 1 CGC said that it has
worsened.  For the management of evaluation data, just under half
(19) said that the management of evaluation data has improved, half
(20) said that there is no change, and 1 said that it has worsened. 
(See question 42, app.  II.)

In chapter 4, we state that relatively few CGCs reported that
regeneration, viability testing, and backup storage are insufficient
for their crop collections.  However, we also report that almost
three-quarters of the CGCs stated that the lack of staff for
regeneration and viability testing has hindered preservation of their
crop collections.  In response to question 43 on the amount of
funding NPGS provides for these activities, given current resources,
19 CGCs reported that for regeneration and viability testing it is
about the right amount and 21 reported that it is probably too
little.  For backup storage/preservation, 26 CGCs reported that it is
probably the right amount and 14 that it is probably too little. 
(See question 43, app.  II.)

4.  We appreciate the challenges NPGS faces in having to juggle
multiple priorities and manage continually increasing collections in
the face of declining resources.  We hope that our report will
provide useful information to congressional and other decisonmakers
in future deliberations on the role of NPGS and the resources
available to NPGS for carrying out its role. 

5.  We support USDA's efforts to optimize the management of NPGS to
make most effective use of its limited resources. 


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V

RESOURCES, COMMUNITY, AND ECONOMIC
DEVELOPMENT DIVISION,
WASHINGTON, D.C. 

Jerilynn B.  Hoy, Assistant Director
Beverly A.  Peterson, Evaluator-in-Charge
Sonja J.  Bensen
Nancy S.  Bowser
Carolyn M.  Boyce
Carol Herrnstadt Shulman

*** End of document. ***