tication, we constantly meet with individual peculiarities of disposition and of habit, which in themselves are utterly meaningless, and therefore quite useless. But it is easy to see that, if, among a number of such meaningless or fortuitous psychological variations, any one arises which happens to be of use, this variation would be seized upon, intensified, and fostered by natural selection, just as in the analogous case of structures. Moreover, there is evidence that such fortuitous variations in the psychology of animals (whether useless or accidentally useful) are frequently inherited, so as to become distinctive, not merely of individuals, but of races or strains. Thus, among Mr. Darwin's manuscripts, I find a letter from Mr. Thwaites, under the date of 1860, saying that all his domestic ducks in Ceylon had quite lost their natural instincts with regard to water, which they would never enter unless driven, and that, when the young birds were thus compelled to enter the water, they had to be quickly taken out again to prevent them from drowning. Mr. Thwaites adds that this peculiarity only occurs in one particular breed. Tumbler-pigeons instinctively tumbling, pouter-pigeons instinctively pouting, etc., are further illustrations of the same general fact.
Coming now to instincts developed by lapsing intelligence, I have already alluded to the acquisition of an hereditary fear of man as an instance of this class. Now, not only may the hereditary fear of man be thus acquired through the observation of ancestors—and this even to the extent of knowing by instinct what constitutes safe distance from fire-arms—but, conversely, when fully formed it may again be lost by disuse. Thus, there is no animal more wild, or difficult to tame, than the young of the wild rabbit; while there is no animal more tame than the young of the domestic rabbit. And the same remark applies, though in a somewhat lesser degree, to the young of the wild and of the domestic duck. For, according to Dr. Rae, "if the eggs of a wild duck are placed with those of a tame duck under a hen to be batched, the ducklings from the former, on the very day they leave the egg, will immediately endeavor to hide themselves, or take to the water, if there be any water, should any one approach, while the young from the tame duck's eggs will show little or no alarm." Now, as neither rabbits nor ducks are likely to have been selected by man to breed from on account of tameness, we may set down the loss of wildness in the domestic breeds to the uncompounded effects of hereditary memory of man as a harmless animal, just as we attributed the original acquisition of instinctive wildness to the hereditary memory of man as a dangerous animal; in neither case can we suppose that the principle of selection has operated in any considerable degree.
Thus far, for the sake of clearness, I have dealt separately with these two factors in the formation of instinct—natural selection and lapsing intelligence—and have sought to show that either of them
