Differences in test results on the same lot of seed often can be traced to lack of uniformity among the different bags. Seedsmen have been informed of their responsibility in this connection, and research agencies are seeking methods by which seed lots can be reliably mixed to uniformity.

Statisticians have developed minimum latitudes of variation between bags for the different quality factors, making possible homogeneity tests by which bag-to-bag variation can be determined. The principal weakness in this procedure is the large amount of time required to make tests on individual-bag samples.

THE EDUCATION and training of seed analysts depend on several factors, including the number and kinds of crop seeds the analyst is expected to test, the area from which the seeds come, the kinds of tests they are expected to make, and the supervision they have.

Analysts in large laboratories often work under supervision and are expected to perform only one phase of testing, such as tests of purity or germination. Analysts in some small, private laboratories must be prepared to make various tests on any submitted sample, without supervision. Various intermediate arrangements exist.

Analysts who must assume the responsibility for the tests should have earned at least a college degree in one of the plant sciences, preferably botany. After his college training, the prospective analyst should work under the guidance of an experienced technician. Analysts having less responsibility and working under direct supervision may not need the college background; without it, however, the chances are greatly reduced that the analyst will be able to cope with new and difficult situations.

Many a seed analyst has received his basic training by working as an apprentice under an experienced technician, completing all or part of a college education, followed by supervised laboratory experience, and studying seed testing while working toward a degree.

In large laboratories, where specialization is possible, the purity analyst should have a working knowledge of plant taxonomy and some familiarity with the plant diseases that may be detected on dry seeds. The purity analyst must know a large variety of crop and weed seeds, including the variation in characteristics of the commonly tested crop seeds and associated weed seeds. Through experience, the analyst learns the differences between seeds and inert matter in species that produce empty seeds.

The germination analyst should have training in plant physiology and a familiarity with symptoms of plant diseases. Alertness and an awareness that low germination may be the result of improper testing are prime considerations. He must recognize dormant seeds in the test, as distinguished from dead seeds, and use methods that will promote germination. Testing for variety, health of seed, seedling vigor, and moisture content requires the services of persons who have had special training in the area of work to be undertaken.

C. H. Lawshe and L. E. Albright, of Purdue University, developed tests for the selection of purity analysts by measuring adaptability, manual dexterity, and visual acuity. The tests are described in a 14-page brochure published by the Purdue University Agricultural Experiment Station, Lafayette, Ind., under the title, "A Manual for the Selection of Competent Seed Analysts."

Training programs have been instituted at Iowa State University, Mississippi State University, Oregon State College, and Purdue University. Before 1940, many supervisors in Federal, State, and commercial laboratories received their training at the Federal seed laboratory in Washington, on an apprenticeship basis. From 1940 to 1950, Federal analysts conducted regional, 1-week short courses at State and Federal laboratories. This in-service instruction was open to all analysts working in State, commercial, and private laboratories. Demands became so great that in 1951 a Federal employee was assigned full time to instruction work. Short courses have been held throughout the country.

LABORATORIES for testing seeds may be Federal, State, commercial, or private.

There are Federal laboratories at Beltsville, Md., New Brunswick, N.J., Montgomery, Ala., Kansas City, Mo., and Minneapolis, Minn. There is a Federal-State laboratory in Sacramento, Calif. These laboratories are maintained to administer the import and interstate provisions of the Federal Seed Act. They also conduct a limited amount of testing for the import and export trade and for Government agencies.

All States but three operate one or more seed-testing laboratories, which exist primarily to implement the administration of State seed laws and to test samples for farmers and seedsmen.

Commercial seed laboratories test samples received from any person who desires the service on a fee basis. Some commercial laboratories have standing contracts with seed firms. There are about 30 such laboratories in the country.

Many seed firms maintain their own private laboratories to perform testing services incidental to their operations. Most employ one analyst, but a few are large enough to permit specialization of work.

THREE organizations have helped to further testing in the United States. They are the Association of Official Seed Analysts, whose members work in Government laboratories in the United States and Canada; the International Seed Testing Association, whose secretariat is in Europe and whose members are national governments; and the Society of Commercial Seed Technologists, whose members are Americans and Canadians.

The main objectives of the three organizations are basically the same: The development and adoption of standard methods of testing seeds, promotion of research leading to the improvement of seed testing, and exchange of information through meetings and publications.

Since 1950 the Association of Official Seed Analysts and the International Seed Testing Association have worked closely together. The differences that previously existed between the American-Canadian Rules and the International Rules have been largely eliminated. A realization of the need for uniform testing procedures on an international basis and a willingness to compromise differences have made this possible.

OREN L. JUSTICE is a seed botanist and Head of the Testing Section, Seed Branch, Grain Division, Agricultural Marketing Service, U.S. Department of Agriculture. He holds advanced degrees from Ohio University and Cornell. Dr. Justice is author Of numerous articles on seed dormancy and seed testing.