What, then, were the facts which appeared to Romanes sufficient to justify a belief in the existence of a class of instincts dependent on inherited habit for their origin? He tells us that he only gives a few examples “amongst almost any number” that he could quote. It is certainly unfortunate that, out of more than one hundred and fifty pages devoted to instinct in his work on “Mental Evolution in Animals,” only three[45] are assigned to secondary instincts; or six, if we include one dealing with inherited peculiarities of hand-writing in man, and two showing the force of heredity in the domain of instinct, “whether of the primary or secondary class.” It is true that many pages are devoted to instincts of blended origin, but the co-operation of the Lamarckian factor is here rather assumed than proved. We must, however, be content to take the few examples that are actually given. They are four in number. First, that ponies in Norway are used without bridles, and are trained to obey the voice; and that, as a consequence, a race-peculiarity has been established, for Andrew Knight says that it is impossible to give them what is called “a mouth.” No details are given, and Romanes does not further discuss the evidence. Secondly, Mr. Lawson Tait had a cat which was taught to beg for food like a terrier. All her kittens adopted the same habit under circumstances which precluded the possibility of imitation. Supposing the facts to be correctly reported, and granting that the owners of the kittens, presumably aware of the maternal propensity, did not take some pains to teach the offspring of such a parent to beg (and this does not present much difficulty), one can hardly found a scientific conclusion on so slight an anecdotal basis. Thirdly, instinctive fear is said to be an inherited acquisition; which, fourthly, is lost by disuse. But, as we have already seen, modern investigation has placed this matter of so-called hereditary fear of natural enemies on a different footing. Pheasants, partridges, moor-hens, and wild duck show no fear of a quiet dog. If approached gently, in the absence of their parents, callow wild birds in their nest exhibit little alarm at the slow and gentle approach of man. Mr. Hudson’s opinion has already been quoted, but will bear repetition; it is, “that fear of particular enemies is in nearly all cases the result of experience.” And there is no evidence to show that, in those cases in which it is truly instinctive and not the result of experience, the instinctive behaviour is necessarily due to inherited habit and not to natural selection.