I confess that when I look round upon the varied habits of birds and mammals, when I see the frigate bird robbing the fish-hawk of the prey that it has captured from the sea, the bald-headed chimpanzee adopting a diet of small birds, a Semnopithecus in the Mergui Archipelago eating crustacea and mollusca, and the koypu, a rodent, living on shell-fish; when I consider the divergence of habits in almost every group of organisms, the ground-pigeons, rock-pigeons, and wood-pigeons, seed-eating pigeons and fruit-eating pigeons; the carrion-eating, insect-eating, and fruit-eating crows; the aquatic and terrestrial kingfishers, some living on fish, some on insects, some on reptiles;[JN] the divergent habits of the ring-ousel and the water-ousel; and the peculiar habits of blood-sucking bats;—when I see these and a thousand other modifications and divergences of habit, I question whether the theory that they have all arisen through the elimination of those forms which failed to possess them may not be pushed too far; I am inclined to believe that the inheritance of acquired modifications has been a co-operating factor. It is not enough to say that these habits are all useful to their several possessors. It has to be shown that they are of elimination value—that their possession or non-possession has made all the difference between survival and elimination.

On the whole, then, as the result of a careful consideration of the subject of instinctive and habitual activities, and in accordance with my general view of organic evolution as set forth in previous chapters, I am disposed to accept the inheritance of individually acquired modifications of habit as a working hypothesis. I do not think that absolutely convincing evidence thereof can at present be produced. But to the best of my judgment, the probabilities are in favour of the inheritance of modifications of existing activities, due to intelligence, instruction, and imitation; always provided that the exercise of these modified activities is sufficiently frequent and definite to give rise to habits in the individual.

I recognize three factors in the origin of instinctive activities—

1. Elimination through natural selection.

2. Selection through preferential mating.

3. The inheritance of individually acquired modifications.

Of these I consider the first quite incontrovertible; the second as highly probable; and the third as probable in a less degree. In all three, intelligence may or may not have been a factor. Some of the habits which have survived elimination under the first factor may have been originally intelligent, some of them from the first unintelligent. Some of the love-antics (so called), which, through their tendency to excite sexual appetence in the female, have been selected under the second factor, may have had a basis in intelligence; many of them probably have not. And though the great majority of individually acquired modifications of habits have owed their origin to intelligent direction, still it is conceivable that some of them have not. An animal may have been forced by circumstances to modify its habits, without any exercise of intelligence; and this modification, forced, through changed conditions, upon all the members of a species, may, through inheritance, have passed into the stereotyped condition of an instinct. Under each factor, then, we have two several categories.

1. Elimination..{a. of unintelligent activities.
b. of intelligent activities.
2. Selection..{a. of unintelligent activities.
b. of intelligent activities.
3. Inheritance..{a. of unintelligent activities.
b. of intelligent activities.

In all cases, however, where intelligence has been a co-operating factor, this intelligence has lapsed so soon as the activity became truly instinctive.

From the co-operation of the factors it is almost impossible to give examples which shall illustrate the exclusive action of any one. The following table must therefore be regarded as indicating the probable predominance of the factor indicated:—