It All Starts with the Seed
by Steve Peters

The success of any seed company that provides agricultural seed for gardeners and farmers depends upon the quality of the seed. Superior-quality seed and the resulting plants must have most, if not all, of the following characteristics: 1) a high germination rate, 2) early seedling vigor, 3) resistance to pests and diseases, 4) a reasonably high degree of plant uniformity (depending on the use), 5) a high yield, or sometimes more importantly a reliable yield every year, 6) excellent flavor (if it's an edible crop), and 7) high nutritive value. The initial seed a farmer plants to produce a crop is known as "stock seed"; the genetic makeup of this stock seed should contain all these desirable agronomic and horticultural traits. Growers of stock seed must adhere to specific protocols to achieve the desired results. Producing good stock seed is time-consuming; however, when done correctly, it is the best insurance for good seed quality for many years to come. Most experts in the seed industry believe that producing good stock seed is the single most important investment a company can make to achieve long-term economic stability and to provide the best product to its customers.

Over the past few seasons at our Research Farm in New Mexico, we have observed that most of our radish varieties, while certainly high germinating and vigorous, produce a fairly high percentage of off-types—i.e., roots that deviate in shape and/or color from the archetypal form. Therefore, we decided to improve some of our current varieties as well as begin stock seed development on other promising, colorful varieties that we do not yet offer. We are carrying out this project in conjunction with the Organic Seed Alliance, a non-profit organization dedicated to supporting the ethical development and stewardship of agricultural seed. They accomplish their goals through collaborative educational and research programs with organic farmers and various agricultural organizations; consequently, a major objective of our project is to train existing and potential seed farmers to produce the highest-quality stock seed.

To insure the greatest chance of significantly improving a particular variety, the first step in our project was to identify the best strain of the variety. We began by procuring seed from two to four reputable sources (including Seeds of Change when the strain was available) of seven radish varieties: French Breakfast, D'Avignon, Hailstone, Cherry Belle, Champion, Pink Beauty, and Plum Purple. In early spring, each seed lot was planted in California in ten-foot rows replicated twice. Approximately thirty days later, the radishes were harvested and evaluated for vigor and uniformity. We then choose the best strain of each of the seven varieties to be planted immediately on other Seeds of Change growers' farms. The goal was to produce 10,000 radish roots of each variety so that we could choose the best 2,000 roots for replanting. This process would enable us to strictly adhere to the ideal archetype and still produce enough stock seed for the next few years. A farmer in Oregon grew five of the seven radish varieties; when they were at the prime eating stage, each variety was harvested (all 10,000 roots of each). The radish tops were removed from the 2,000 selected roots and placed in boxes in a cooler until they were ready to be replanted. Next, we replanted the five varieties in five locations separated by at least one mile, to prevent any cross-pollination among the varieties; thus, the farmer who grew the five varieties could replant only one of them on his own farm. The other four were taken to neighboring farms. Of the original seven varieties mentioned initially, the remaining two were grown in Idaho; the selected roots were replanted on two separate farms.

All varieties have since sent up flower stalks, with seed harvest to come in early to mid-fall. Once the mature seed is harvested, a small amount of each variety will be replanted in California, where the growing season is longer. It is also essential to plant the original seed from which these selections were made, because it is the only way to determine whether the selection we produced 1) improved the variety, 2) made it worse (which is possible), or 3) didn't change it at all. If there was a significant improvement, we may decide that the stock seed phase of development is complete and that we are ready to begin production of larger quantities for resale. If we have lowered the quality, we need to go back to the original generation and start over, or find other sources of that variety that may be better. If little or no progress has been made, we might continue the process and perhaps make a stricter selection the next time.

Beans have perfect flowers (male and female reproductive organs on the same flower) and hence are self-pollinating; therefore, the likelihood of cross-pollinating between varieties of beans is much less than with, for example, radishes, which outcross profusely. Nevertheless, some outcrossing does occur (usually less than 1-2%) due to varying environmental conditions. As a result, varietal purity can deteriorate over time, which means that stock seed production is just as necessary for beans as it is with radishes. When beans are grown in large enough quantities, any off-types present in the population can be observed. The four top-selling bean varieties at Seeds Of Change—Bush Blue Lake, Provider, Haricot Vert, Maxibel, and Roc D'Or—all have a very small percentage of off-types. Although these may be unnoticeable to the gardener planting only 50 to 100 seeds, many market growers also grow these varieties, so it was necessary to run them through a stock seed program. If the selections are done well it should take only one season to achieve 100% (or nearly so) varietal purity.

Bean pod off-typing can be readily detected; however, other plant characteristics are more difficult to discern unless the plants are observed almost daily. Thus, we chose two professional bean seed growers in Idaho—who are intimately familiar with beans at all growth stages—to carry out the stock seed production. The first off-type trait they detected was vining plants, because all these varieties are bush types. All the vining plants that appear will be rogued (eliminated) before any mature seed is harvested. Different-colored flowers were rogued from the Providers, although the grower determined that it was a physical mixing with another variety rather than a genetic crossing. Another selection will involve eliminating plants that are susceptible to curly top virus, a disease carried by the beet leafhopper. Curly top can reduce yields significantly; young plants affected by the virus will die before they produce seed. Plants that produce pods may get the disease late in the season; these must be removed manually so that they won't be included in the seed harvest. The final roguing will occur when the beans are in the fresh eating stage. The Roc D'Or variety, for example, is a round, yellow wax bean; any flat pods or green pods will be removed before they are allowed to develop mature seed.

Our tomato stock seed projects involve the Crimson Sprinter, Martian Giant, and Three Sisters varieties. We have concentrated primarily on developing populations of plants that are resistant (or at least tolerant) to two of the most important tomato fungal diseases, Early Blight (Alternaria solani) and Septoria Leaf Spot (Septoria lycopersici). These diseases are particularly prevalent in the northeastern and midwestern states, where hot, humid summers are the norm. Identifying and selecting disease-resistant plants can, of course, occur only if the disease is present in significant amounts. The Crimson Sprinter tomato, for example, is being grown in a location in eastern North Dakota that has had high levels of septoria leaf spot disease in previous years, as well as some early blight. The grower planted 150 plants to make sure that he had enough surviving plants at the end of the season from which to harvest fruit. It is recommended to harvest fruits from at least twenty-five plants to maintain sufficient genetic diversity within the varietal population. The grower allowed all of the diseased plant residue from previous years to remain on the soil surface to assure that the disease was present. He also placed the plants one foot apart, which is too crowded for normal production, but created more favorable conditions for disease occurrence due to the restricted air circulation among the plants. In addition, the summer of 2005 was abnormally wet in North Dakota (about 17 inches of rain from early June to late August), which helped spread disease. Plant tissue samples are being sent to a pathology lab to positively identify the existence of the diseases in question.

In late July, the grower carried out an early harvest of Crimson Sprinter fruit from a few of the plants. The seed from this harvest will be kept separate from all other subsequent harvests with the hope that this may be a significantly earlier-maturing selection. Although the septoria existed on all the plants, it was, as of mid-August, not severe enough to eliminate any of the plants; therefore, the grower harvested fruit from all the plants at this time. Before his final harvest sometime in late September, the grower will eliminate all the plants with the most severe incidence of disease (perhaps 50% of the plants). This final seed harvest will be kept separate from the earlier harvest. In 2006, seed from the initial harvest and the final harvest will be planted side by side, in disease-prone fields, to determine if a truly superior, disease-resistant strain has been developed. This same process will be performed on the Three Sisters and Martian Giant in 2005-2006, as well as on several more varieties over the next few years.

Beets are the fourth major crop for which we are producing stock seed. The varieties being improved are Detroit Dark Red, Chioggia, Pronto, and a Crosby-type variety with exceptional leaf and root characteristics that will be released in 2007. The primary selection criteria include root shape and color, leaf color and size, and surface smoothness. As with the radishes, we began by evaluating and choosing the best seed lot of each variety. Then we selected for the archetypal root shape and replanted only the very best roots. The primary grower for this stock seed is located in southern Oregon on a farm that has an abundance of rhizoctonia dry rot, a fungal disease that causes the surface of the beet to become rough, scaly, and corky. In the industry, this roughening and darkening is known as "scurf." Almost all beets found in the marketplace have at least a little bit of scurf. It is probably not always rhizoctonia, but it nevertheless detracts from the overall appearance. The Oregon beet grower has been ruthlessly selecting only the smoothest roots, i.e., those with the least amount of rhizoctonia, for three generations. I recently saw his 2005 crop of the Crosby-type beet, and I can honestly say that these were the smoothest, most beautiful roots I have ever seen; they will surely make any market grower of beets very happy.

While these stock seed initiatives may take many years and much expense to complete to satisfaction, they are an important and integral part of our overall commitment to producing the highest-quality organic seed available.

Steve Peters
Commercial Seed Manager