But in the domestic state it is all otherwise. Free competition between individuals or between cells is not allowed. Thus, for example, among plants, crossings may be forced and hybrids made in gardens which would never occur in Nature. The florist prevents fertilization in the same kind and compels fertilization of a different kind. If male cells of the same kind were allowed to compete, the result would be different. Doubtless the same method would succeed in many lower animals. So also in higher animals free competition and sexual selection for union are often not allowed, and therefore animals of different species, such as the horse and the ass, unite, which would not do so if they were free to select as in the wild state. These two are widely distinct species, sometimes even called genera, and therefore the offspring is infertile; but two closely allied species, such as two species of wolf, or of the fox, in a domestic state would probably not only unite but produce indefinitely fertile offspring. In fact, it is almost certain that the dog was made by a mixture of several species of wolf, most, perhaps all, of them now extinct.[30] On the other hand, it is not at all certain that the extreme varieties of dogs have not passed the limit of greatest attraction, and therefore of greatest cross-fertility, and that, if allowed free choice, as in Nature, they would not breed true, or tend to breed true, with their own kind, and intermediate kinds die out in the struggle for life. =Law of Cross-breeding.=--Before going any further in this discussion, it is necessary to bring out another point of extreme importance in the formation of varieties, both natural and artificial--a point which I believe throws light upon the very significance of sex itself--I refer to the effect of cross-breeding. It is a curious and most significant fact that different varieties, both natural and artificial, are, up _to a certain limit_, not only cross-fertile and cross-attractive, but even more so than individuals of the same variety. Long experience has shown that very close breeding of the same variety for a long time fixes the _kind_ but _weakens_ the _stock_, especially in fertility, while judicious crossing of varieties strengthens the stock, increasing its fertility, and especially producing _plasticity_ or _variability_. Therefore breeders, if they wish to preserve a valuable variety, breed close; but, if they wish to make new varieties, cross-breed. But we have already seen that species are usually cross-sterile. Therefore there must be some regular law of increase to a maximum, and again decrease to zero. It is this law that I now wish to investigate.