Such are the admitted characters of instinct. Though none of them is out of the reach of minute criticism; though none of them is absolutely true, still they are sufficiently exact to serve to distinguish instinct from all other psychological phenomena.

Instinct, so defined, is, beyond all question, transmissible, and subject to the law of heredity. The animal inherits the psychical dispositions, no less than the physiological constitutions of its parents. The naturalist takes account of the former characteristics, as well as of the latter. In his eyes it is as essential for the bee to extract the pollen of flowers, construct cells and in them deposit her honey, as for her to possess mandibles, six feet, and four wings. A worker-bee with the instincts of an ant would appear to him as strange a thing as a bee with wing-sheaths and eight feet. Every animal has two chief functions-one, nutrition, which preserves the individual; another, generation, which preserves the species. The latter transmits instincts together with physical forms-generation is moral as well as material. The beaver 'transmits to its young its anatomical and physiological characters as a rodent mammal, its constructive instincts, and architectural talent.

Thus we find at the outset a vast number of psychological facts -instinctive actions, strictly subject to the laws of hereditary transmission. It needs no long reflection to see how large is the domain of instinct: the Invertebrata seem to be completely restricted to this form of mental activity. In the sub-kingdom of the Vertebrata, the inferior classes, such as the Fishes, Batrachians, Reptiles, Birds, have oftentimes no other means save instinct, of supporting life, of attack, defence, and recognition of enemies. Finally, among Mammals, and even in Man, instinct gradually diminishes, but never entirely disappears. Its domain, therefore, is co-extensive with animal life; and this vast domain is governed by the laws of heredity.

Since it is an evident fact, universally admitted, that heredity is the invariable rule of the transmission of instincts, we need not cite instances to confirm our position. The tenacity of instincts is so great, and their hereditary transmission so certain, that sometimes they are found to outlive for centuries the conditions of life to which they are adapted. We have reason to believe,' says Darwin,

'that aboriginal habits are long retained under domestication. Thus with the common ass we see signs of its original desert life in its strong dislike to cross the smallest stream of water, and in its pleasure in rolling in the dust. The same strong dislike to cross a stream is common to the camel, which has been domesticated from a very ancient period. Young pigs, though so tame, sometimes squat when frightened, and thus try to conceal themselves even in an open and bare place. Young turkeys, and occasionally even young fowls, when the hen gives the danger-cry, run away and try to hide themselves, like young partridges or pheasants, in order that their mother may take flight, of which she has lost the power. The musk-duck, in its native country, often perches and roosts on trees, and our domesticated musk-ducks, though sluggish birds, are fond of perching on the tops of barns, walls, etc. We know that the dog, however well and regularly fed, often buries, like the fox, any superfluous food; we see him turning round and round on a carpet, as if to trample down grass to form a bed. . . . . In the delight with which lambs and kids crowd together and frisk on the smallest hillock, we see a vestige of their former alpine habits.'1

II.

Instead of dwelling unnecessarily on the heredity of natural and primitive instincts, it will be more instructive to inquire whether acquired instincts are transmissible. We have already said, when giving, according to F. Cuvier and Flourens, the characteristics generally attributed to instinctive acts, that none of them are absolutely true. Thus, instinct is not always invariable. The beaver changes, according to circumstances, the site and form of his house, and from being a builder becomes a miner. The bee can modify her plan of construction, and substitute for hexagonal cells pentagonal cavities. In the Island of Goree the swallows remain through the whole year, because the warmth of the climate enables them to find food at all seasons. In many species the mode of nestbuilding varies according to the nature of the soil, the locality, and the temperature of the country. Instinct is certainly not as pliant an instrument as intelligence; it cannot, like intelligence, adapt itself

1 Variation, etc., vol. i. P. 180.

to all media, conform to all circumstances, or vary and modify its actions in a thousand ways; yet it is capable of modification within certain limits, when subjected to strong and lasting influences.

Two causes chiefly produce these variations: external conditions and domestication. Climate, soil, food; the dangers which habitually surround the animal, and the impressions it receives, modify its organism and consequently its instincts. The action of man is still more powerful on the animal than that of Nature: by training, man fashions and bends it to his needs or his wishes. It is not for us to inquire here how these acquired or modified instincts are produced. We have only to ask whether they are hereditary. Experience answers in the affirmative; many facts show that acquired instincts, as well as those which are natural, are transmitted by heredity. Such are the following :

G. Leroy observes that in districts where a sharp war is waged against the fox, the cubs, on first coming out of their earths, and before they can have acquired any experience, are more cautious, crafty, and suspicious than are old foxes in places where no attempt is made to trap them. This he explains by the hypothesis of a language among animals. F. Cuvier has furnished the solution of the enigma by referring the fact to the heredity of modifications which are acquired by instinct. There is no doubt that the instinct of fear is acquired in many wild animals, and transmitted to their descendants. Knight, who for sixty years devoted himself to systematic observation of this class of facts, says that during that time the habits of the English woodcock underwent great changes, and that its fear of man was considerably increased by its transmission through several generations. The same author discovered similar changes of habit, even in bees. Darwin has established the fact that animals living in desert islands gradually acquire a fear of man, in proportion as they become acquainted with our methods of destroying them. He says that in England large birds are much more shy than small ones, and this, no doubt, because they are much more persecuted by man. The proof that this is the reason of the difference is found in the fact that in uninhabited islands large birds are not any more timid than small ones.1

1 Origin of Species, p. 260. Fifth Edition, 1869. Lucas, ii. 482.

C

When an animal is capable of education, that is, when its original instincts are capable of modification, it usually requires three or four generations to fix the results of training and to prevent a return to the instincts of the wild state. If we try to hatch the eggs of wild ducks under tame ducks, the ducklings will scarce have left the egg when they obey the instinct of their race, and take their flight. If they be prevented from flying away, and kept for reproduction, it will be several generations before we have tame ducks. The same may be said of free, or wild herds of horses. Their colts are broken with great difficulty, and even after taming they are far less docile than horses born in a state of domestication. Nay, even the mongrel progeny of wild and domesticated horses, or of wild and domesticated reindeer, take three or four generations before they entirely give up the shy habits of their natural state. On the other hand, colts bred of a well-broken sire and dam oftentimes come into the world with a marked aptitude for training; and some horse-trainers have even proposed to select brood stock exclusively from among horses that have been practised in the circus.

Originally man had considerable trouble in taming the animals which are now domesticated; and his work would have been in vain had not heredity come to his aid. It may be said that after man has modified a wild animal to his will, there goes on in its progeny a silent conflict between two heredities, the one tending to fix the acquired modifications, and the other to preserve the primitive instincts. The latter often get the mastery, and only after several generations is training sure of victory. may see that in either case heredity always asserts its rights.

But we

Among the higher animals, which are possessed not only of instinct but also of intelligence, nothing is more common than tỏ see mental dispositions, which have evidently been acquired, so fixed by heredity that they are confounded with instinct, so spontaneous and so automatic do they become.. Young pointers have been known to point the first time they were taken out, sometimes even better than dogs that had been for a long time in training. The habit of saving life is hereditary in breeds that have been brought up to it, as is also the shepherd-dog's habit of moving around the flock and guarding it.

Knight has shown, experimentally, the truth of the proverb “a good hound is bred so." He took every care that when the pups were first taken into the field they should receive no guidance from older dogs. Yet the very first day one of the pups stood trembling with anxiety, having his eyes fixed and all his muscles strained at the partridges which their parents had been trained to point. A spaniel, belonging to a breed that had been trained to woodcockshooting, knew perfectly well from the first how to act like an old dog, avoiding places where the soil was frozen, and where it was therefore useless to seek the game, as in such places there is no scent. Finally, a young polecat-terrier was thrown into a state of great excitement the first time he ever saw one of these animals, while a spaniel remained perfectly calm.

In South America, according to Roulin, dogs belonging to a breed that has long been trained to the dangerous chase of the peccary, when taken for the first time into the woods, know the tactics to adopt quite as well as the old dogs, and that without any instruction. Dogs of other races, and unacquainted with the tactics, are killed at once, no matter how strong they may be. The American greyhound, instead of leaping at the throat of the stag, attacks him by the belly and throws him over, as his ancestors had been trained to do in hunting down the Indians.

Thus, then, heredity transmits acquired modification no less than natural instincts. There is, however, an important difference to be noted the heredity of instincts admits of no exceptions, while in that of modifications there are many. It is only when variations have been firmly rooted; when, having become organic, they constitute a second nature, which supplants the first; when, like instinct, they have assumed a mechanical character, that they can be transmitted. If we note these differences in passing, we shall find them lead us hereafter to important conclusions.

III.

WE have just shown, from indisputable facts, that heredity governs the transmission of instincts, whether acquired or primitive. It might seem that in this portion of our inquiry, which has to deal only with the facts, we ought to be content with that exposition of the case. But certain theories, put forth by distinguished writers