and when they attain the age of puberty, all the rabbits save one or two should be removed. Speedily those left with the hare will become with young, upon which they should be removed, and replaced by others. After this the hare should be kept in a hutch by himself, and a doe left with him at night only. As the hare is naturally a very shy animal, it will only breed when perfect quietness prevails. The half-bred produced in the first instance should now be put to the hare, and a cross, three parts hare, and one part rabbit, obtained. The permanent breed should then be obtained by crossing the quadroon doe leporide, if I may use the term, with the half-bred buck. I have directed attention to the production of the leporide because I believe that the problems in relation to it, which have been solved by M. Roux, have an important bearing upon the breeding of animals of greater importance than hares and rabbits. Here we find a race of animals produced by the fusion of two species, which naturally exist in a state of mutual enmity, and which differ in many important respects. The hare and the rabbit are respectively of but little value as food, at least they are of no importance to the feeder; yet a cross between them turns out to be an excellent meat-producing animal, which may be reared with considerable profit to the feeder. It is thus clearly shown that two kinds of animals, neither of which is of great utility, may give rise to an excellent cross, if their blood, so to speak, be blended in proper proportions. A half-bred animal may be less valuable than its parents, but a quadroon may greatly excel its progenitors. The goat and sheep are so closely related that they are classed by naturalists under one head--_Capridæ_. Some kinds of sheep have hair like goats, and certain varieties of goats have fleeces that closely resemble those on the sheep. There are sheep with horns, and goats without those striking appendages. The Cape of Good Hope goat might easily be mistaken for a sheep. It would seem, judging by the results of Roux's experiments, that there is no great difficulty in the way of obtaining a cross between the sheep and the goat. I do not mean an ordinary half-breed, but a prolific hybrid similar to the leporide. Of course, it is impossible, _a priori_, to say whether or not such a hybrid race, supposing it produceable, would be valuable; but as goats can find a subsistence on mountains where sheep would starve, it is possible that an animal, essentially a sheep, but with a streak of goat blood in it, could be profitably kept on very poor uplands. Whether a race of what we might term _caprides_ be formed or not we have derived most suggestive information from M. Roux's experiments, which I hope may be turned to account in what is by far the most important field of enquiry, the judicious crossing of varieties of the same species.