A second type of variation which is of value to the breeder is those known as "sports," or muta tions (Fig. SO). These differ from individual vari ations only in degree. They are what may be termed large-type and ordinarily reproduce true to seed. A very large number of our new races and varieties of cultivated plants are the results of such mutations or seedling sports. All vegetable-growers know that far the larger number of their new varie ties are apparently produced suddenly. For instance, Livingston, who has bred' a great many new varie ties of the tomato, followed the practice of examin ing carefully his different plants for variations. Occasionally some striking new type differing from other varieties would be found. This was selected and used as the foundation stock for a new variety. Our good apples, pears, and peaches, have been found in many cases in fence-corners, and new varie ties of wheat, cotton and other crops have resulted very largely from the selection of strikingly good plants which, because of their superior quality, have attracted the attention of growers, and have been propagated. While many of these accidental discoveries are doubtless of hybrid origin, still it is probable that the majority are simply mutations or sports.
The third type of variation which is of impor tance to the plant-breeder is that produced by hybridization or crossing, and here we probably have the most prolific source of variations, and, therefore, the class of variation of the greatest importance and most consequence to the breeder. It has come to be an established policy to combine the good qualities of two races into a single race by hybridization and selection. Influence of environment.— It is a well-known fact that environment has a decided in fluence on the form and char acter of the plant. It is by no means cer tain, however, that these changes are of any value to the plant breeder. It seems certain that those changes which are the conse quence of en vironment purely are not hereditary. It is a well-known fact that if climbing or twining beans or viny cowpeas'are transferred from a south ern to a northern climate or from a lower to a higher altitude, they tend to produce a dwarfed type which will not show the twining or viny habit in such marked degree ; and in order to secure bush types by selection, breeders have sometimes advocated the transferring of types to more northern latitudes or to higher altitudes, where the experiments may be made under conditions that naturally lead to the production of a lower bush type. It is doubtful, however, whether such a transfer would be of material aid. While it is recognized that such variations are produced as an influence of the environment, it is also known that, on the whole, those variations which are produced as an immedi ate influence of the environment are not hereditary.
Individual variations and mutations are of greatest use to the plant-breeder. Without question, if the cowpea or bean were cultivated under southern conditions it would show individual variations in the degree in which it shows the climbing or twin ing habit. Even under southern conditions, certain individuals would doubtless show more of the bush type than others. It is believed by the writer that a bush type can be secured just as quickly under southern conditions by selecting from these lower and more bushy plants as it can by the same selection made in more northern localities or at higher altitudes.
Location of breeding plots.
It is important to consider the conditions under which the breeding patch or plat should be grown. Some growers are inclined to locate their breeding patches in the garden and give the plants the very best possible care, thinking that this is the best means of determining which plants are superior. Animal-breeders also isolate their breeding stocks and give them every possible care and advantage. On the contrary, some plant-breeders assert that it is best to have the breeding patch located on soils which are most like those on which the gen eral crop is to be grown. The writer has given this matter considerable thought, and he is strongly of the opinion that the most satisfactory method is to cultivate the breeding patch under the same conditions under which the ordinary crop is to be grown. Plants are fixed in one place, and are entirely dependent on the local soil conditions. If, therefore, the plant has been bred and adapted to one soil condition, it cannot be expected to give as good results under different soil conditions. If a variety is being bred for sterile soils, the selection should be conducted on similarly sterile soil in order to breed a race of individuals that are "gross feeders," as planters term it, and capable of deriv ing their nutriment from sterile soils and making a sturdy growth even under adverse conditions. If, for example, plants were being bred to adapt them to alkaline conditions, the breeding patch should not be placed in a sheltered, favored spot, where the soil does not contain alkali. The plants must be grown under alkaline conditions in order to discover, as a result of natural selection, those plants which do the best where the alkali is present, and thus guide us in the selection. The same would be true in breeding plants for arid regions. The plants should be cultivated in the arid region rather than in a moist region of heavy rainfall, or in a thoroughly irrigated patch.