Now the theory of this is that the qualities of the parents are transmitted to the young in virtually the same proportion as they exist in the parents, but not with exact equality. Some young, either of the same or in successive births, will get a little more of quality A, others a little more of quality B, than the rest. If in the next generation one with a superabundance of quality A mates with one having an excess of quality B, the balance will be restored; and this is what usually happens in nature, so that variations are almost sure to be speedily ex tinguished. If, however, sovereign man sees to it that A mates with another A, then their young will be doubly endowed with that quality; and if the process is continued that quality will finally predominate, and will be maintained so long as outside mating is pre vented.
In practice, however, the matter is not quite so simple. Some animals resist change, or may be modified up to a certain point, and no further in that direction as in the well-known case of the blue Andalusian chickens, which 'will not breed true. It must be remembered, also, that negative as well as positive characters are transmissible, the weaknesses as well as the strength. Sometimes a fault seems inseparable from the sought-for virtue, as in the prevalent deafness of white cats.
The evidence as to inheritance of disease is too conflicting to be determinative. Probably no disease or its resulting lesion is transmitted; but a weakness of resistance, which amounts to a tendency, or at least a susceptibility to that disease, often develops in the descendants of animals afflicted with certain ailments, as ring bone in horses. Domestication has an effect on the procreative ability, or, at any rate, on the fecundity of animals, sometimes reducing it, but more often increasing it, owing, no doubt, to safer conditions and better food than the same creatures had in the wilderness; and the production of special breeds has occasionally been followed by exaggerations of this tendency. In varieties produced by hybridizing it is almost a rule that they are sterile. The more perfect in show-points is a kennel of toy-dogs the more delicate in health are they, and the harder it is for them to beget and nurture their puppies.
The mere production of a fixed artificial variety of animal is one thing, and the making of a strong, useful and valuable breed is quite another. It is probable that few new ones of practical importance will be effected.
or the establishment of new forms of plants, depends on essentially the same principles as have been outlined. The principal difficulty in producing a new variety of flower or fruit is to prevent mixture by the introduc tion of foreign pollen— the fertilizing element; but this may be guarded against. One source of a new variety is by seizing upon some con spicuous accidental variation in a plant, called by horticulturists a °sport," isolating it, and per petuating it by some of the methods presently to be mentioned. Many well-known flowers and fruits have originated from such sports.
°On an ordinary orange tree, not so many years ago, a single bud produced a branch which bore only seedless fruit, . . . cor related with the presence of a tiny accessory orange embedded almost wholly in the larger; that branch was grafted, as were the resulting branches ; and this is the origin of . . . the navel orange" (Ganong).
Reproduction by grafting or by cuttings is one method, then, of plant-breeding; but more commonly the new variety is originated or brought to perfection, or both, by artificial pollination — transferring by human agency desirable pollen to the stigma of the plant to be affected, whence it descends and fertilizes the seed. This is done in most .cases by touching the anthers of a ripe flower with a soft, camel's-hair brush, and then touching the stigma to be affected, to which the con veyed pollen will stick In some special cases the pollen of the plant's own flower is used; this ensures non-variability, and is called °in breeding"; but as a rule the pollen from an other plant of the same kind is taken. This is °cross-pollination," and tends to make an equally true and more vigorous stock, with an in creased tendency to (controllable) variation. Thirdly, the breeder may a given stigma from a different variety, or even (but not always) from a different species of plant. This is °hybridization," and the result is a °hybrid"— the source of many of the remark able varieties of plants that Luther Burbank and other scientific gardeners and experimenters have given us. It is the rule, however, as among animals, that these hybrids do not pro duce seeds, and must be perpetuated by planting bulbs or cuttings, or in the case of trees, by grafting, generation after generation. Success in these methods can be obtained only by pre venting all access to the stigma, and thence to the seeds of the plant operated on, of any pol len except of the one kind needed. This is most usually accomplished by simply enclosing the blossom to be treated in a cap or bag of paper or light gauze, so that no flying pollen or pollen-loaded insects can reach it.
In this review of methods the writer has necessarily given an impression of extreme simplicity, but the road to success abounds in difficulties and often tries the patience of the experimenter by its length.
Darwin, 'Variations of Plants and Animals under Domestication) (London 1885) ; Castle, (Heredity in Relation to Evolution and Animal Breeding' (New York 1911) ; Davenport, (Principles of Breed ing) (Boston 1907) ; Ewart, The Principles of Breeding and the Origin of Domesticated Breeds of Animals' (Annual Report of the Bureau of Animal Industry, United States Department of Agriculture, 1910) ; Kronacher, (Grundziige der (Ber lin 1912) ; Marshall, 'Breeding Farm Animals' (Chicago 1911) ; Pearl, 'Inheritance in Poul try' (several Bulletins of the Maine Experi ment Station) ; Wilson, (Principles of Stock Breeding' (London 1912).