Plants of a species bearing the sexes separately usually occur in approximately equal numbers. Thus, unless the seedsman can identify the male plants early in their life and remove some of them, a great surplus of them may take space that he would prefer to have occupied by seed-bearing individuals. In general, 10 to 20 percent of males in a planting will furnish effective pollination.

Excess males in a seed spinach field can be chopped out as they become recognizable to allow a higher proportion of females and more space for them.

The sex of asparagus plants can be identified the first year, and a suitable ratio of the two sexes can be transplanted for a perennial seed plot, if desired.

Many fruit and other trees are dioecious. Some have complete flowers but are self-sterile. In either case, the female parent will require the presence of a pollinator tree within a reasonable distance of bee flight or for wind pollination. If no pollinator is nearby and bees are plentiful, small branches of flowers that bear compatible pollen may be cut from distant trees and placed in buckets of water in the trees to be pollinated.

American holly (Ilex opaca) is a common example of a dioecious species. Male trees bloom but never produce fruits and seeds. Their flowers contain pollen-bearing anthers that are visible to the naked eye, but no pistils. Female trees produce flowers having pistils but only rudimentary or shriveled-up anthers without pollen.

THE COMMERCIAL PRODUCTION of hybrid seed depends on suitable flowering habits of the plants to be hybridized.

Hybrid seed of corn is relatively simple to produce because the male and female inflorescences are large and well separated on the plant. To control pollination in a seed field, one needs only to pull the tassels from the rows of female parent plants as they appear and leave the rows of pollinator plants undisturbed. The pollinator must, of course, be one that discharges its pollen at the time the seed plants are in silk.

Hybrid seed of tomato is expensive because it must be produced by hand pollination, flower by flower, usually after emasculation of the flower of the seed parent, also by hand. The pollen is not dispersed by wind and very little by any insect. The flowers are naturally self-pollinated almost entirely.

Hybrid onion seeds became commercially feasible about 1940 with the development of seed-parent inbred lines possessing what is commonly called cytoplasmic male sterility. Such a line produces no pollen but is perpetuated by recurrent fertilization by an appropriate self-fertile line that possesses cytoplasmic male fertility. This second line must be bred and maintained solely for perpetuating the male-sterile line. The 100-percent male-sterile progeny of these two is then planted to produce mother bulbs.

They, in the next year, produce flowers that are field pollinated by a third inbred to produce commercial hybrid seed. Every fifth or sixth row in the seed field usually is a pollinator line. Bees do the pollinating.

The principle of male sterility has been applied commercially to the producing of hybrid seed of other crops such as corn, sorghum, and sugarbeets.

Hybrid seed of spinach is produced commercially by the use of newly developed inbred parent lines, which consist of purely staminate (male) and purely pistillate (female) plants. The male plants are chopped out of the seed-parent rows as they become recognizable. Thus the complete cross-pollination by nearby rows of an inbred pollinator is insured.

Japanese seedsmen produce hybrid seed of cabbage and related plants. They use selected parent lines that are rather highly (but not completely) self-sterile and highly cross-fertile. When plants of two such lines are set alternately in the seed field, about 98 percent of the harvested seed is hybrid. The seedlings that arise from seeds resulting from self-pollination in such a seed lot are smaller than the hybrid seedlings (from cross-pollinated flowers). These nonhybrids are largely left behind in the plant bed at transplanting time so that the transplanted cabbage field consists of practically 100 percent hybrids.

The Japanese also first developed seedless watermelon seed and produced it commercially. This involves a relatively difficult and expensive procedure that may never become popular in the United States, but it illustrates a remarkable application of an unusual flower behavior.

Seedless watermelons are seedless because they develop from flowers that are both self-sterile and cross-sterile. Although these flowers cannot produce seeds, they do respond to the stimulus of pollination sufficiently to form fruits that are normal except for seedlessness. Then where do the seeds come from, for growing the seedless melons? They are produced by hybridizing two parent lines, the seed line of which is a special kind.

The seed-parent line is a tetraploid, produced by doubling the normal number of chromosomes of a watermelon variety through the use of colchicine. Flowers of selected tetraploid plants are then crossed with a selected pollen parent that has the normal (diploid) number of chromosomes. This crossing is done either by hand pollination of artificially protected flowers or by insect pollination of flowers on isolated plants from which all male flowers are removed before they open. The resulting hybrid seed is abnormal. It is triploid it has 1.5 times the normal number of chromosomes. It can produce a plant capable of fruiting when pollinated by suitable diploid pollen, but it is too abnormal to reproduce itself by seeds.

I HAVE mentioned only examples of why and how the seed producer must take into account the flowering behavior of the plants from which he wants to obtain seed. There are almost endless variations in detail of behavior among species and varieties that seeds-men handle. Furthermore, flowering behavior is but one part of the complex subject of seed production, although it is a large and vitally controlling part.

The more we can learn about the details of the flowering habit, including sex expression, and the factors that affect flowering, the more effective the work of producing seed can be.

IN CONCLUSION, then, the incidental or casual saving of seeds can lead to disappointing results with the crops grown from them. The flowering habit of the species underlies a large share of the disappointments. Indeed, flowering habit is a major factor in determining whether any seeds will form to be saved. Unless the gardener and farmer know something of the flowering habit of plants, they often are in no position to get good seeds from them. Sometimes, even when they know a plant's flowering habits, the requirements for getting good seeds may be beyond their control in their particular situation especially with varieties having flowers that are readily cross-pollinated and that are not well isolated from other plantings of the same species. We look with little or no favor on the casual saving of seed because of its hazards.

VICTOR R. BOSWELL joined the Department of Agriculture in 1928. His own research has dealt primarily with the growth, development, behavior, responses, and inherent qualities of a wide range of vegetable crops and seeds thereof. He has written research, review, and popular articles for many publications.