|
 Essay - may/june 2007
Statistics, approximations and close truths II
James C. Delouche
Professor Emeritus Mississippi State University
Seed analysis is an exact science only when a single seed is considered. A single seed can be exactly characterized as a pure seed or a seed of another variety or species. Its capacity to germinate under an optimal set of conditions can be established with considerable precision. It can be determined to be diseased or not diseased, and it can be determined to possess or not possess a specific transgenic trait, and so on.
Seeds, however, are produced, processed, packaged, marketed, planted, analyzed and/or tested not as individual particles but as populations of particles which have to be described and characterized in terms that are not exact, that are approximations and sometimes close truths. As pointed out in the previous essay seeds are different from other crop production inputs such as fertilizer, pesticides, equipment, which are manufactured and amenable to much more rigorous control in terms of variability.
The composition of a seed population, i.e., a seed lot, is affected by edaphic, climatic and biological forces and circumstances that vary among locations and times. Most seed lots, therefore, consist of a principle population of seeds with the desired inheritance and capable of germination, i.e., the pure germinable seeds, and lesser or sub-populations of the pure seeds that do not germinate, that are diseased, that are much smaller or larger than the mean seed size, and so on. And, additional lesser populations of seeds of other varieties and/or species including weeds, and debris from seeds, plants, and other field materials gathered in the harvest.
One of the principle management goals of seed production enterprises is or should be to increase the population of the pure germinable seeds in seeds lots to the highest practical and economical level by preventing, eliminating or reducing the lesser populations of undesirable seeds and contaminants. Since the mathematics of populations is statistics it is not surprising that the quality assurance and control procedures and protocols employed in the production, harvesting, processing, storage and marketing of seeds to achieve this paramount management priority are based on statistical principles and concepts.
Critical First Steps in Reducing Variability in Seed Lots
The quality assurance procedures employed to achieve the goal of the highest practical and economical level of seed quality begins with the selection of the seeds and field for production. The seeds should be from a carefully controlled and monitored class of seeds such as the foundation (basic) or other limited generation class of certified seeds or the equivalent. The production locations should be subjected to field and site inspections so that the cleanest fields in terms of weeds, the most uniform and the most fertile fields available are selected. It is critically important to know the history of the land to be used for production since producing a new variety of rice, wheat, cotton and other crops on land that was used to produce a different variety the previous production season usually results in contamination with volunteer plants from the previous crop. While planting the highest quality seeds and selection of the most desirable field for production are giant first steps in the quality assurance and control of seed production, it is important that fields be inspected at various stages of growth and maturation following well established field inspection protocols by internal (company) and external (e.g., certification) inspectors.
Soybeans pods with different maturation stages
The statistically based inspection protocols should also include some commonsense instructions. Weeds and plants of other varieties identified during inspection should be pulled up or cut and carefully removed from the field especially if they are at or near the maturity stage. For example: seeds of red rice plants can contaminate other areas in the field by shattering unless the panicles are "bagged" before they are cut and removed from the field; seeds of other plants pulled up but left in the field can be picked up by the combine harvester.
Harvesting is a critical operation in the production of high quality seeds. Portions of the production field that are non-uniform due to drought or poor drainage or weediness should be marked off and excluded from the main harvest to prevent or minimize the presence of immature and weed seeds in the population. Harvesting should be properly timed and cautiously performed to reduce the amount of mechanically damaged seeds that cannot germinate, produce abnormal seedlings or become inert material. Cleaning and inspection of the harvesting equipment is also critical because it is a frequent source of contamination with seeds of other varieties, crops and weeds. Improperly timed and controlled aeration, drying and conveying of seeds after harvest can sharply increase the portion of dead and damaged seeds in the seed lot.
These first steps in reducing the variability in harvested seeds and thus increasing the population of pure germinable seeds are very important and if conscientiously followed can produce dramatic results. There was a rather eccentric certified soybean seed producer in Mississippi who took great pride in the fact that his seeds usually met all certification standards before processing and conditioning. Although he did process his seeds to produce a "super clean" seed product, his advertisements prominently included the statement: These Seeds Met Certification Standards before Cleaning. While a few eccentric seed producers are able to produce seeds that technically do not need to be cleaned for certification, the great majority of seed producers are not as eccentric, conscientious and cautious. They have to depend heavily on processing and conditioning operations to increase the principle population of pure germinable seeds in the lot to the desired level by reducing or removing the seeds and materials in the lesser populations.
The Magic of Seed Processing and Conditioning
The results obtained from the processing and conditioning operations are sometimes almost magical. The combine harvester is omnivorous. It takes in everything in its path from soil particles to insect parts and, of course, the good seeds, other seeds, leaves, stems and so on. This mélange of materials is transformed into clean, pure seeds by subjecting it sequentially to a series of ingenious separators as described in the article on Seed Processing Units in the last (March/April) issue of SEED News. While the seed processing operations are seldom 100% efficient they are nevertheless still magical. The level of pure germinable seeds desired and feasible is most often achieved by approximations. Each separator removes contaminants in the population according to its capability so that the pure germinable seeds population in the seed lot is incrementally increased to a final approximation, i.e., an inexact but adequate result, that is usually the highest economical and practical level.
Pre-cleaners remove light and large materials by aspiration and rough sieving. The basic air and screen separator removes materials that are lighter, generally larger or smaller, thicker, and thinner than the pure seed. Precision dimensional separators more precisely remove materials that are thicker or thinner and shorter and longer than the pure seed. The density separator then removes materials that are roughly the same size as the pure seeds but are lighter or heavier. A spiral separator can remove materials that are not as round or more round than the pure seed, while the color separator remove seeds that differ in color, reflectivity or other optical properties. Examining the materials removed from the seed lot, i.e., the rejects, by each separator really can produce an overwhelming sense of the magical.
The results achieved by processing are appropriately characterized as approximations because using the separators to achieve the maximum separation possible usually results in economically prohibitive losses of the good seeds. It is neither practical nor economical to use a density separator to improve germination and vigor of seeds to the highest possible extent when most of the seeds end up in the rejects. Seed separators are magical but there are some things they cannot do. They cannot effectively and efficiently separate seeds of different varieties of wheat, soybean, cotton,and many other crops. While seeds of short, medium and long grain varieties of rice can be separated, the seeds of varieties within each class cannot be separated including those of red rice biotypes that fall within the class. And, of course, although the germination and vigor of the pure seeds can be increased within limits, the magic machine for efficiently and consistently removing the non-germinable and low vigor seeds from seed lots has not yet been invented.
While the inspection and monitoring procedures and protocols employed in the control of seed production, processing and conditioning adhere to statistical precepts, the statistics applied to seeds along with many close truths are much more prominent in the analytical procedures used to establish and verify seed quality. These will be examined in the next essay.
|
|
............
|
