Animal Behaviour

Mr. E. J. Shellard observed[63] an act of similar import in a Scotch staghound, which “appeared at first to be the result of thought,” but which, on closer observation, was clearly seen to be the result of intelligence in the restricted sense of the term. The dog released the lever-latch of a yard door. “At first he raised his paws to the door and scratched violently, manifesting various signs of impatience. His scratches, which extended from the top of the door downwards, and over the whole area, would thus inevitably at some time or other reach the handle of the latch, which was thus struck forcibly downwards, the latch itself rising upwards. The door would then open from the weight of the dog pushing against it. The dog always opened the door in this manner from the time when the incident was first noticed until he left, a period of about three years. The door was opened with no greater ease at the expiration of that period than at the commencement. His paws would strike various parts of the door, and he never appeared to exercise any degree of judgment in the localization of his strokes, the fact of his paws striking the handle of the latch being a necessary result, providing the dog had sufficient patience and strength to continue.”

One or two more experiments with my fox terrier may be briefly described. I watched his behaviour when a solid indiarubber ball was thrown towards a wall standing at right angles to its course. At first he followed it right up to the wall and then back as it rebounded. So long as it travelled with such velocity as to be only just ahead of him he pursued the same course. But when it was thrown more violently, so as to meet him on the rebound as he ran towards the wall, he learnt that he was thus able to seize it as it came towards him. And, profiting by the incidental experience thus gained, he acquired the habit—though for long with some uncertainty of reaction—of slowing off when the object of his pursuit reached the wall so as to await its rebound. Again, when the ball was thrown so as to glance at a wide angle from a surface, at first—when the velocity was such as to keep it just ahead of him—he followed its course. But when the velocity was increased he learnt to take a short cut along the third side of a triangle, so as to catch the object at some distance from the wall. A third series of experiments were made where a right angle was formed by the meeting of two surfaces. One side of the angle, the left, was dealt with for a day or two. At first the ball was directly followed. Then a short cut was taken to meet its deflected course. On the fourth day this method was well established. On the fifth, the ball was thrown so as to strike the other or right side of the angle, and thus be deflected in the opposite direction. The dog followed the old course (the short cut to the left) and was completely non-plussed, searching that side, then more widely, and not finding the ball for eleven minutes. On repeating the experiment thrice, similar results were that day obtained. On the following day the ball was thrown just ahead of him, so as to strike to the right of the angle, and was followed and caught. This course was pursued for three days, and he then learnt to take a short cut to the right. On the next day the ball was sent, as at first, to the left, and the dog was again non-plussed. I did not succeed in getting him to associate a given difference of initial direction with a resultant difference of deflection.

I may here mention that, whenever searching for a ball of which he had lost sight in the road, he would run along the gutter first on one side and then on the other. A friend who was walking with me one day regarded this as a clear case of rational inference. “The dog knows,” he said, “the effects of the convex curvature of the road as well as we do.” I am convinced, however (having watched his ways from a puppy), that this method of search was gradually established on a basis of practical experience. No logical inference on his part is necessary for the interpretation of the facts; and we should not assume its presence unless the evidence compels us to do so.

Dr. E. L. Thorndike, in a monograph on “Animal Intelligence” published as a supplement to the Psychological Review (June, 1898), has fully described and carefully discussed a number of interesting experiments. The subjects (one might, alas! almost say victims) of some of these were thirteen kittens or cats from three to eighteen months old. His method of investigation shall be stated in his own words.

To Dr. Thorndike’s monograph we must refer those who desire detailed information as to apparatus and procedure. It must here suffice to state that the box-cages employed were rudely constructed of wooden laths, and formed cramped prisons about twenty inches long by fifteen broad and twelve high. Nine contained such simple mechanisms as Dr. Thorndike describes in the passage above quoted. When a loop or cord was pulled, a button turned, or a lever depressed, the door fell open. In another, pressure on the door as well as depression of a thumb-latch was required. In one cage two simple acts on the part of the kitten were necessary, pulling a cord and pushing aside a piece of board; and in yet others three acts were requisite. In those boxes from which escape was more difficult a few of the cats failed to get out. The times occupied in thoroughly learning the trick of the box by those who were successful are plotted in a series of curves, the essential feature of which is the graphic expression of a gradual diminution in the time interval between imprisonment and escape in successive trials. This is shown in Fig. 23, which is constructed from some of Dr. Thorndike’s data. In some cases the cats were set free from a box when they (1) licked themselves or (2) scratched themselves.

Dr. Thorndike comments on the results of his experiments as follows:—

Such experiments carried out on a different method give results in line with my own. The conditions are, however, somewhat unnatural, which I regard as in some respects a disadvantage. But we need experiments on different methods—the more the better,—and if the results they furnish are in accord, their correctness will be rendered the more probable. It is to be hoped that Dr. Thorndike will devise further experiments in which (1) the conditions shall be somewhat less strained and straitened, while the subjects are in a more normal state of equanimity (cannot “utter hunger” be avoided?), and (2) there shall be more opportunity for the exercise of rational judgment, supposing the faculty to exist. To establish the absence of foresight in the procedure of the cats, it is surely necessary so to arrange matters that the connections are clearly open—nay, even obvious—to the eye of reason. It appears that this consideration has not weighed sufficiently with Dr. Thorndike.

A series of experiments were made to ascertain whether instruction (in the form of putting the animal through the procedure requisite for a given act) was in any degree helpful. The conclusion is that such instruction has no influence. Those who have had experience in teaching animals to perform tricks will probably agree here—though some trainers give expression to a different opinion. It is, however, essential to distinguish carefully between showing an animal how a trick is done, and either stimulating its attention or furnishing accessory guidance (such as the occasional taps of the trainer’s whip when he wants a performing horse to kneel), or affording suitable conditions the results of which temporarily enter into the association complex. If the latter be eliminated the practice of trainers, I believe, bears out the general result of the experiments. Dr. Thorndike never succeeded in getting an animal to change its way of doing a thing for his. Nor was I, after repeated trials, able to modify the way in which my dog lifted the latch of the gate. He did it with the back of his head. I could not get him to do it (more gracefully) with his muzzle.

It may be said that the remarkable feats of performing animals imply the existence of faculties of a higher order than Dr. Thorndike and I are prepared to admit on the basis of our experiments. Mr. P. G. Hamerton many years ago described[64] how, in his own house, a cleverly trained dog would fetch in their right order the letters which spelt the English or German equivalents of common French words, and do other wonderful things. But the owner of the dog (M. du Rouil) admitted that there was a means of rapport between them which he was not prepared to divulge. It is just because the trainer has to lead up to and utilize chance experiences that such prolonged patience and care are required. The animal is but the instrument on which his clever trainer plays; an instrument of wonderful intelligence, but lacking in the higher rational faculty. The organized scheme is the master’s, not that of his willing slave. A rational being might not do more wonderful things; but he would learn them more rapidly and by a less wearisome method. As it is, the clever performing dog originates little or nothing, and repeats again and again the same stereotyped behaviour, which—if one witnesses the performance often—touches one with a profound sense of its lack of rational spontaneity.

As at present advised, therefore, I see no reason for withdrawing from the position provisionally taken up. The utilization of chance experience, without the framing and application of an organized scheme of knowledge, appears to be the predominant method of animal intelligence.

On this view, then, we may see in instinctive behaviour, and the multifarious automatic acts of animals, a means of providing experience of the right kind and on profitable lines. We may see in the play-instincts of the young a training ground for the more serious business of animal life—a theme developed by Professor Groos. We may see in the imitative tendency—the innate proclivity to follow a lead blindly and at first unintelligently—a further means of providing those useful items of experience which intelligence finds so serviceable. And we may see in the intelligence which can profit by chance occurrences that arise in these several ways all that suffices for the simple needs of animal existence.

With some differences of opinion Dr. Thorndike and I have much in common in the conclusions to which we have been independently led as to the method and limits of animal intelligence. We seem to be in essential agreement in the belief that the method of animal intelligence is to profit by chance experience without rational foresight, and that unless such experience be individually acquired, the data essential for intelligent progress are absent. While in our attempts to realize the general nature of animal consciousness there is a close similarity of treatment. In my “Introduction to Comparative Psychology” a good deal of space is devoted to an analysis of the psychology of skill “in order that we may infer what takes place in the minds of animals;” and I said:—“When I am playing a hard game of tennis, or when I am sailing a yacht close to the wind in a choppy sea, self does not at all tend to become focal. Hence, though I am a self-conscious being I am not always self-conscious. And presumably when I am least self-conscious, I am nearest the condition of the animal at the stage of mere sense experience. I am exhilarated with the sense of pleasurable existence, my whole being tingles with sentient life. I sense, or am aware of, my own life and consciousness, in an unusually subtle manner. Experience is vivid and continuous. Such I take it to be the condition of the conscious but not yet self-conscious animal.”

I can therefore cordially endorse Dr. Thorndike’s conclusions as expressed in the following passages:—

“One who has watched the life of a cat or dog for a month or more under test conditions, gets, or fancies he gets, a fairly definite idea of what the intellectual [intelligent] life of a cat or dog feels like. It is most like what we feel when consciousness contains little thought about anything, when we feel the sense-impressions in their first intention, so to speak, when we feel our own body, and the impulses we give to it. Sometimes one gets this animal consciousness while in swimming, for example. One feels the water, the sky, the birds above, but with no thoughts about them or memories of how they looked at other times, or æsthetic judgments about their beauty; one feels no ideas about what movements he will make, but feels himself make them, feels his body throughout. Self-consciousness dies away. Social consciousness dies away. The meanings, and values, and connections of things die away. One feels sense-impressions, has impulses, feels the movements he makes; that is all.”

And after an illustration from such a game as tennis, Dr. Thorndike adds: “Finally the elements of the associations are not isolated. No tennis-player’s stream of thought is filled with free-floating representations of any of the tens of thousands of sense-impressions or movements he has seen and made on the tennis-court. Yet there is consciousness enough at the time, keen consciousness of the sense-impressions, impulses, feelings of one’s bodily acts. So with the animals. There is consciousness enough, but of this kind.”

It may be said that between the method of intelligence and that of fully developed rational procedure there is a wide gap which must have been bridged in the course of mental evolution. Unquestionably. And in contending that the methods of the animal are predominantly intelligent, I am far from wishing to assert dogmatically that in no animals are there even the beginnings of a rational scheme. Indications thereof do not indeed at present appear to have been clearly disclosed by experiment. But the experimental development of the subject is still in its infancy. We shall probably have to await the further results which must be the outcome of patient and well-directed child-study. The human child does pass in the course of his individual development from intelligent to rational procedure. Here there is a bridge which is crossed by every child. When we know more about the stadia of this development we shall be in a position to apply the results obtained in child-study in the analogous field of animal-study. Till then we must possess our souls in patience, and base our provisional conclusions on the results of systematic investigation, rather than on those of casual observation and anecdote.

IV.—The Evolution of Intelligent Behaviour

No attempt can be made in this section to trace the successive stages of the evolutionary progress of intelligence from its lower to its higher developments. It is indeed questionable whether comparative psychology has, as yet, accumulated a sufficient body of data to render such a task profitable or even possible. And the lower the level of intelligence with which we have to deal, the less reliable are the scanty psychological data which we can obtain. To interpret the mental processes which accompany the acts of even the higher animals is a hard task, requiring careful psychological analysis. Still harder is the task to infer the psychological basis of the actions of the lower animals.

It is difficult to say where, in the hierarchy of animal progress, the beginnings of intelligence can first be traced. In the articulated animals, such as the insects, spiders, and crustacea, there is abundant evidence of intelligence of a relatively high grade. But even in their case, how hard it is to realize the nature of their experience—to get any adequate notion of their mental processes! We are inevitably forced to describe their psychology in the most general terms. So, too, with forms still lower in the scale of intelligence. Many molluscs unquestionably profit by experience. But can we clearly picture to ourselves the nature and manner of acquisition of this experience? The way in which limpets return to the scars on the rock which form their homes seems to show that they have acquired a practically adequate experience of their near surroundings. Romanes cites[65] some of the earlier observations which have been extended by Professor Ainsworth Davis.[66] I looked into the matter myself some years ago, at Mewps Bay near Lulworth in Dorsetshire. The method adopted[67] was to remove the limpets from the rock, and affix them at various distances from their scars. This can be done without difficulty or injury to the mollusc if one catches them as they are moving. But one must make sure that they are just leaving or returning to their proper homes, and are not taken in the midst of a more extended peregrination, as in that case their special scars cannot be noted. Failure to be careful in this matter vitiated my earlier observations, which are therefore excluded in the following table:—

From the nature of the rock surfaces the removal of a limpet to a distance of two feet almost invariably involved placing them on the further side of an angle. And though some returned over such an angle, the majority did not.

In most cases the individuals which failed to return to their respective scars took up new positions; and in several instances, when they were subsequently removed to a distance of a few inches from this new position, they returned to it. Their return to the scar was watched in many cases, and the course was fairly, but not quite direct. One limpet covered a distance of ten inches, over a somewhat curved course, in a little under twenty minutes. In another case the limpet on its return journey had to pass between two others, which necessitated the lifting of the shell to some height so as to clear one of them. On reaching the scar they twist and turn about so as to fit down in the normal position which is constant. When they come up the wrong way round they rotate pretty rapidly through the 180 degrees to get into position. One was observed to make a short excursion from and return to its scar under stillish water. But as a rule they seem to remain fixed when they are submerged, moving for the most part when the tide has just receded.

The greatest distance I have watched a limpet reach from its home was twenty-two inches. But I have found them at a distance of three feet from their scars—that is to say, from those to which they fitted perfectly. This was on a large flat surface.

When they move, the tentacles are projected out beyond the shell, and keep on touching and slightly adhering to the rock. On reaching the scar they carefully feel round it with the tentacles. By excision of these feelers Professor Davis was led to conclude that it is not through their instrumentality that the limpet finds its way back to its particular scar. But I am inclined to question these results. At any rate, further observations and experiments are needed to settle the point.

Snails will also return to special dark hollows or crannies in the wall after their foraging excursions. Such behaviour in molluscs affords evidence of something more than instinct. In popular speech, we should say that there is memory of the locality. And in any case it is difficult to interpret the facts without the assumption that the animals are conscious, and that re-presentative states are evoked through the mediation of presentative sense-impressions. But how difficult, if not impossible, it is to form anything like a satisfactory conception of the rudimentary mental processes of a limpet!

The most highly developed molluscs are the cephalopods. They have long sensitive mobile arms with which they feel for and capture their prey. “Now Schneider observed,” writes Dr. Stout,[68] “a very young octopus seize a hermit-crab. The hermit-crab covers the shell in which it takes up its abode with stinging zoophytes. Stung by these, the octopus immediately recoiled and let its prey escape. Subsequently it was observed to avoid hermit-crabs. Older animals of the same species managed cleverly to pull the crab out of its house without being stung.” Such cases afford evidence of profiting by experience through the exercise of intelligence.

Darwin’s careful observations on the manner in which earthworms drag leaves into their burrows seem to show that these annelids act intelligently, and deal with leaves of different shapes in different ways. The leaves of Pine trees, consisting of two needles arising from a common base, were almost invariably drawn down by seizing this basal point of junction; while the leaves of the Lime were, in 79 per cent. of the cases examined, drawn down by the apex; in only 4 per cent. by the base; and in the remaining 17 per cent. by seizing some intermediate portion. On the other hand, the leaves of the Rhododendron, in which the basal part of the blade is often narrower than the apical part, were in 66 per cent. of the observations drawn down by the narrower base. Triangles of paper were in the majority of cases seized by the apex. Commenting upon his observations, carried out with great care under experimental conditions, Darwin says,[69] “As worms are not guided by special instincts in each particular case, though possessing a general instinct to plug up their burrows, and as chance is excluded, the next most probable conclusion seems to be that they try in different ways to draw in objects, and at last succeed in some one way;” that is to say, they profit by experience based on the method of trial and failure. But Darwin adds that the evidence he obtained shows “that worms do not habitually try to draw objects into their burrows in many different ways.” And he seems to attribute to them an almost rational power of dealing with the circumstances in the light of general conceptions. “If worms,” he says, “are able to judge, either before drawing or after having drawn an object close to the mouths of their burrows, how best to drag it in, they must acquire some notion of its general shape. This they probably acquire by touching it in many places with the anterior extremity of their bodies, which acts as a tactual organ. It may be well to remember how perfect the sense of touch becomes in a man when born blind and deaf, as are worms. If worms have the power of acquiring some notion, however rude, of the shape of an object and of their burrows, as seems to be the case, they deserve to be called intelligent; for they then act in nearly the same manner as would a man under similar circumstances.”

Such power of perceiving the relation of the shape of a leaf or other object to the form of the burrow is presumably beyond the reach of an earthworm. It may be regarded as more probable that the earthworm inherits an instinctive tendency to draw down objects in special ways, and that this is subject to some modification under the play of experience, without the formation of anything so psychologically complex as a general notion, however rude. In any case the behaviour of earthworms in closing their burrows seems to afford indications of something more than instinct—of that profiting by the results of experience which characterizes intelligent procedure. More than this we cannot say.

Professor Whitman[70] has made some interesting observations on the leech Clepsine. “Place the animal,” he says, “in a shallow, flat-bottomed dish, and leave it for a few hours or a day, in order to give it time to get accustomed to the place, and come to rest on the bottom. Then, taking the utmost care not to jar the dish or breathe upon the surface of the water, look at the Clepsine through a low magnifying lens, and see what happens when the surface of the water is touched with the point of a needle held vertically above the animal’s back. If the experiment is properly carried out, it will be seen that the respiratory undulations (if such movements happen to be going on) suddenly cease, and that the animal slightly expands its body and hugs the glass. Wait a few moments until the animal, recovering its normal composure, again resumes its respiratory movements. Then let the needle descend through the water until the point rests on the bottom of the dish at a little distance from the edge of the body. Again the movements will cease, and the animal will hug the glass with its body somewhat expanded. Now push the needle slowly along towards the leech, and notice as the needle comes almost in contact with the thin margin of the body, that the part nearest the needle begins to retreat slowly before it. This behaviour shows a surprising keenness of tactile sensibility, the least touch of the water with a needle-point being felt at once.... If its back were rubbed with a brush or the handle of a dissecting needle, in order to test its sensitiveness to touch, the appearance would probably be that of insensibility and indifference to the treatment. Closer examination, however, would show that the flesh of the animal was more rigid than usual, and that the surface was covered with numerous stiff, conical elevations, the dermal papillæ or warts, which are so low and blunt in the normal state of rest as to be scarcely visible. It would be seen that the animal, although motionless, was in a state of active resistance to attack.... Clepsine has another and entirely different method of keeping quiet. The animal rolls itself up (head first and ventral side innermost) into a hard ball, outwardly passive, free to roll or fall whithersoever gravity or currents of water may direct it.... If by chance the animal has eggs, it will not desert them to escape in this way.... This species, then, has two quite distinct and peculiar ways of keeping quiet, and thus avoiding its enemies. If the animal has no eggs, or if it has young, it may adopt either mode of escape, while if it has eggs it has no choice but to remain quiet over them.... The act of rolling up into a passive ball may be performed (a) under compulsion, as when it is her last resort in self-defence; (b) under a milder provocation, as one of three courses of behaviour, as when the resting-place is turned up to light, and the choice is offered between remaining quiet in place, creeping away at leisure, or rolling into a ball and dropping to the bottom; (c) or finally, under no special external stimulus, but rather from internal motive, the normal demand for rest and shady seclusion, presumably very strong in Clepsine after gorging itself with the blood of its turtle host.”

Professor Whitman rightly regards the act of rolling into a ball as instinctive, and due to natural selection. But he does not undertake to discuss the question as to how much intelligence, if any, Clepsine may have. Nor, indeed, is it an easy matter to determine. The differential reaction according as the animal has eggs or not suggests intelligence; but it may be instinct varying according to the conditions of stimulation external and internal. The different behaviour which may be seen in different cases when a stone is turned to the light again suggests intelligence, but again may be determined directly by the conditions of stimulation. Prompted by Dr. Whitman’s observations, I endeavoured to determine whether a leech would grow accustomed to frequent gentle stimulation with a camel’s-hair brush, and cease to react under circumstances which were followed by no ill effects. But though I incline to think that this is the case, the observations were not such as to be satisfying and convincing. If intelligence be present we seem to find it in an early and rudimentary state.

Observation, we must confess, seems to afford little indication of the conditions under which intelligence first makes its appearance in the animal kingdom. And if we turn to general considerations, which at the best afford uncertain guidance, little light is thrown on the subject. If we accept the view already indicated,[71] that the nerve-centres which are concerned in the conscious control begotten of experience are independent of those primarily concerned in normal reflex action, we may perhaps believe that the simplest nervous system, worthy of the name, contains both these elements, and that in the course of the evolution of nervous systems in higher and higher grades, there go on pari passu the further differentiation of these elements, and the progressive integration of reflex and control centres into a closely connected and effective whole. Not that any expression of the facts, if such they be, in terms of an evolution formula, adds anything to our knowledge of the organic modus operandi. We know but little of the intimate nerve physiology of even the highest invertebrates. We see ample evidence of the control of behaviour in the light of individual experience. Of any detailed knowledge concerning the manner in which this control is effected we do not seem to possess more than the rude initial phases.

When we compare, however, the several grades of intelligence which observation suggests, and when we watch the conscious development of the more intelligent animals, we seem to find evidence of the growth of a system of experience, at first in very close touch with inherited modes of procedure, but gradually acquiring more of independence and freedom. Increase of the range and complexity of behaviour brings with it, not only increase in the range and complexity of experience, but also—what is, perhaps, even more essential to effective progress—greater unity and closer connection into a well-knit whole. And with this greater unity and closer connection there goes what one may term a condensation of experience by an elimination of detail and the survival of essential features repeatedly emphasized. This is analogous in the development of intelligence to the generalization and abstraction which play so important a part in the development of reason. It affords, in fact, the data which reflection utilizes in the purposive and intentional condensation and concentration of knowledge at a higher stage of mental development.

The omission of detail and the survival of the salient features is well known to us in the familiar facts of memory. We have seen thousands of sheep and oxen, no two of which are probably alike in all their external details as presented to vision. But we remember what a sheep or an ox looks like, and many of us can form a visualized image of either of these animals. This, however, is not the re-presentative image of any particular sheep or ox. It is what psychologists term a generic image. It is like a composite photograph made by superimposing on the same plate a number of individual images so that the salient features which all possess in common stand out clearly by their coincidence on the plate, while the distinctive details are but dimly presented. Thus does memory preserve the essentials common to many impressions while the distinguishing details are lost and fade, eliminated by forgetfulness. And thus in the experience which intelligence practically utilizes are the net results of a thousand particular impressions condensed in one effective image.

Condensation of experience is also effected by the elimination, under the guidance of consciousness, of those modes of behaviour which are not efficacious—a process to which Professor Mark Baldwin applies the phrase Functional Selection. There is a tendency at first to the overproduction of relatively useless actions. The multifarious random movements of the human infant, though their inexactness renders the child terribly helpless, afford a wide store of plastic material which intelligence can guide to its appropriate use. And the prolonged period of pupilage in the child is correlated with an unsurpassed range of combination and recombination of the abundant plastic material. The hereditary legacy, though it contains fewer drafts for definite and specific purposes than are placed to the credit of an animal rich in instinctive endowment, affords a far larger general fund on which intelligence may draw for the varied purposes of the freer financial existence of a rational being.

The relatively helpless young of many of the higher mammalia exhibit also much overproduction of seemingly aimless movements. But from these intelligence selects those which are of value for the purposes of life—those which experience proves to be effective. These—the relatively few—afford the motor impressions which by repetition stand out in experience, while the rest lapse from memory and are eliminated from experience as they are eliminated from practical performance. This is a great gain. Motor experience is rendered generic; the composite image that is retained is the net result of effective behaviour; and all that is valuable in the acquisitions of early life is condensed within manageable limits.

This process of rendering generic the particular items of a widening experience has a marked effect in the development of the conscious situations in the light of which behaviour is intelligently guided. It is not the master holding this whip or that ball which suggests to the dog a hiding or a scamper; it is a generic situation with interchangeable details. It is not this, that, or the other previously unseen cat that at once determines the situation for the fox terrier; the particular animal has never entered into his past experience: it is the fulfilment of the essential conditions of the generic image that is operative in behaviour. The experience of animals must inevitably become in large degree generic by the elimination of the unessential and survival in re-presentative consciousness of the salient elements in many slightly diverse situations.

Stated in terms of this conception, the familiar phenomena of mimicry are due to the fact that the mimicking form accords sufficiently well with the generic image to carry the same suggested meaning. As is well known, the model has been proved in many cases to be unpalatable or hurtful, while the mimic is in itself neither the one nor the other. The drone-fly, Eristalis, mimics the drone. And it has been urged that this cannot be a true case of mimicry, since the drone is harmless, though the female and “neuter” bees are possessed of stings. But I have satisfied myself by experiments with young birds, that (1) after experience with bees drones are avoided, and (2) that after similar experience drone-flies are also left untouched. Hence it seems that all three fall within the same generic image, the points of resemblance outweighing the differences in detail—as they do, indeed, with many men and women.

Such examples of mimicry belong to what is known as the “Batesian type”—so called after H. W. Bates, who, in 1861, discussed its occurrence among Amazonian insects in the light of the theory of natural selection. There are, however, certain groups of insects which, although themselves “protected,” possess common warning colours, causing them to resemble each other. These are sometimes classed under the head of “Müllerian mimicry”—so called after Fritz Müller, who, in 1879, first offered an explanation of the facts based on the theory of natural selection. He suggested that such mutual resemblance is advantageous to both protected forms, since it lessens the number of those which are killed by young birds and other animals while they are learning by experience what to eat and what to leave. For, as the result of careful observation, Mr. Frank Finn concludes “that each bird has to separately acquire its experience, and well remembers what it has learnt,”—a conclusion with which, as already stated, my own observations are entirely in accord. There is therefore a certain amount of destruction of even well-protected forms by young and inexperienced birds. If, then, two such forms resemble each other, the acquisition of experience is thereby facilitated and the amount of destruction reduced, on the assumption that the two fall within the same generic image. Upholders of natural selection are not, indeed, at one in accepting this explanation, and further observation is unquestionably needed. It is not improbable, however, that common protective coloration, such as the banding of yellow and black, seen in such different forms as the caterpillar of the cinnabar moth and the imago of the wasp, is of mutual utility. The following experiment was made with young chicks. Strips of orange and black paper were pasted beneath glass slips, and on them meal moistened with quinine was placed. On other plain slips meal moistened with water was provided. The young birds soon learnt to avoid the bitter meal, and then would not touch plain meal if it were offered on the banded slip. And these birds, save in two instances, refused to touch cinnabar caterpillars, which were new to their experience. They did not, like other birds, have to learn by particular trials that these caterpillars are unpleasant. Their experience had already been gained through the banded glass slips; or so it seemed. I have also found that young birds who had learnt to avoid cinnabar caterpillars left wasps untouched. Such observations must be repeated and extended. But they seem to show that one aspect of the Müllerian theory is not without some facts in support of it; and, so far as they go, they afford evidence that black and orange banding, irrespective of particular form, may constitute a guiding generic feature in the conscious situation.

It may be said that the generic condensation of experience here indicated implies the formation of general and abstract ideas, and that we cannot in face of the evidence accept Locke’s dictum that abstraction is “an excellency which the faculties of brutes do by no means attain to.” Romanes contended[72] that “all the higher animals have general ideas of ‘good-for-eating’ and ‘not-good-for-eating’ quite apart from any particular objects of which either of these qualities happens to be characteristic,” and he quoted with approval Leroy’s statement,[73] that a fox “will see snares when there are none; his imagination, distorted by fear, will produce deceptive shapes, to which he will attach an abstract notion of danger.” According to such views animals form concepts; and concepts belong to the sphere of rational thought. It is not my intention to enter at length into the refinements of psychological distinction. Many psychologists, however, seek to distinguish between, on the one hand, the predominance by natural emphasis, of certain qualities, such as that of being suitable for food, and, on the other hand, the intentional isolation of these qualities for the purposes of thought and rational explanation. Abstraction they regard as a deliberate process applied with rational intent to the material afforded by experience and reflection. Generalization, too, they regard as deliberate, and carried out with like intent. The result is not merely a composite or generic product, but something more subtle and less dependent on sense. “All trees hitherto seen by me,” said Noiré, “leave in my imagination a mixed image, a kind of ideal presentation of a tree. Quite different is my concept, which is never an image.” The concept “tree” is a deliberate synthesis of abstract qualities intentionally isolated, and recombined in accordance with the general relationships which subsist between them.

If we accept this distinction, if we regard abstraction and generalization as intentional mental processes carried out with the rational intent of discovering the relationships of phenomena with the object of explaining them and recombining their essential features in an ideal scheme of thought, we shall probably admit, with John Locke, that these are excellencies which the faculties of brutes do by no means attain to. But we shall none the less see that the predominance of certain salient features in experience by reiterated emphasis in association with natural needs, and the development of generic in place of merely particular re-presentations will afford the appropriate material for abstraction on the one hand, and generalization on the other. Intelligence supplies the embryonic mental structures from which, under the quickening influence of a rational purpose, abstract and general ideas may be evolved.

The essential features of the evolution of intelligence seem, then, to be, first, the development of controlling nerve-centres, by which the responsive action of reflex automatic or instinctive centres may be checked, augmented, or modified; secondly, the increased differentiation and integration of these control centres with extension of the range and complexity of experience in close touch with practical needs; thirdly, the condensation and concentration of experience by the formation of generic products through the reiterated emphasis begotten of recurrent situations having certain salient features in common, though differing in details; and fourthly, an increased plasticity of behaviour, especially in early life, enabling an animal to deal effectually with an environment far less simple than that to which the more stereotyped instinctive behaviour is fitted by inheritance to respond. And this evolution of intelligent behaviour is working its way up to, though as such it cannot reach, the succeeding phase of mental evolution in which the data, supplied by intelligence, are treated with a new purpose for higher ends in the rational thought which seeks to explain the phenomena, and frame an ideal scheme of their relations and interconnections.

Two further points may be noticed. First, that it is during the early and plastic days or months of life that intelligence is setting its seal on animal behaviour, and stamping it with its distinctive character. Adult life is very much what youth has made it; and old age is stereotyped through habit. In times of progress, the character of the race is determined by plastic possibilities of the young. Among them it is that the incidence of elimination makes itself felt, resulting in the survival of those whose intelligence can mould behaviour in accordance with the new circumstances of a wider life.

Secondly, this selection of the intelligent involves the survival of those in whose higher brain-centres there is room for a greater range and variety of interconnection by means of associating fibres. It involves a selective survival of the larger and more finely organized brains. It is probable, as Professor Ray Lankester has recently indicated, that the ridiculously small-brained mammals and reptiles of the past were creatures of instinct with little capacity for intelligent control. Their lives were simple, and their enemies and competitors no better provided with higher brain-centres than themselves. Stereotyped instinctive behaviour sufficed to enable them to hold their own, and meet the requirements of a life of dull and unprogressive monotony. Strength without cunning made these big-framed animals for a while masters of the situation. But among those existing animals whose skeletons indicate an analogous zoological position, there is none which exhibits a cerebral development so poor. And we may fairly conclude that the fact that these huge creatures have left no lineal descendants may be taken as evidence of the importance and value, in evolution, of that cerebral tissue which is the organic basis of intelligence. The higher brain contains the potentiality of that experience without which the evolution of intelligent behaviour in any race of vertebrate animals is impossible.

V.—The Influence of Intelligence on Instinct

We have seen that the relation of instinct to intelligence is essentially that of congenital to acquired behaviour. We have seen, too, that in the Lamarckian interpretation what is acquired in the course of life may be transmitted through inheritance, and thus the intelligent behaviour of one generation may become instinctive and congenital in the next. But serious biological difficulties stand in the way of the acceptance of this interpretation; there is, moreover, little or no evidence of the assumed transmission to offspring of any acquired modifications of structure or behaviour. We have, therefore, been led to infer that instinctive behaviour has been evolved through the selection of adaptive variations of germinal origin, the influence of intelligence being restricted to the fosterage of co-incident variations, that is to say, of those congenital variations which coincide in direction with the acquired modifications of behaviour due to intelligence. It is clear that on this interpretation the influence of intelligence on instinct is more indirect and less simple than that implied by the Lamarckian hypothesis. Intelligence and instinct are in large degree independent, though there is continual interaction between them. We have now to consider the nature of this interaction, and to this end we must indicate the relation of acquired modifications to the hereditary groundwork of the animal constitution.

The basal fact is, that the bodily tissues are subject to a certain amount of structural change during the course of individual life in accordance with the amount of functional strain put upon them. The labourer’s thickened skin, the enlarged and strengthened muscles of the athlete, the juggler’s acquired suppleness are familiar cases. Less familiar instances are afforded under abnormal conditions. Should one kidney from any cause be slowly destroyed, the other will slowly enlarge to carry on the increased work of elimination of waste products; when the larger shin bone of a dog has been removed after injury, the smaller bone becomes thickened to bear the added strain; new joint surfaces are sometimes formed where bones have been broken and the natural joints injured.

One may say that the normal development of any structure depends upon a due amount of use. But, since in the course of strenuous life any organ is from time to time subject to an abnormal amount of strain, it must be fitted to respond to a super-normal call on its strength and functional activity. Were the heart and the lungs, for example, unable to meet the greatly added drain on their energies, due to unwonted and severe exertion, collapse, perhaps death, would ensue if such exertion were imperatively demanded under special circumstances. And it is clear that many wild animals must be not infrequently placed in such circumstances as will subject their muscular structures and the functional activity of their organs of circulation and respiration to a strain nearly up to their extreme limits of endurance. The carnivorous hunter would often fail to secure his prey if his organization were unequal to a hard and prolonged chase; the hunted prey would not survive to procreate his kind if he fell a victim to the first pursuer through inability to stand the exertion necessary to enable him to make good his escape. It is thus, we may believe, through natural selection that a sufficiently high standard of strength and functional endurance is maintained. The failures in these respects are steadily eliminated. It is difficult to realize the great strain put upon a bird’s organization by the migration flight. Some ten times as many birds leave our shores in the autumn as return to them in the following spring. What proportion of these is weeded out in the act of migration we do not know; but we may be sure that only those fitted to stand a severe test of physical endurance return to rear broods which shall inherit in large degree similar vigour of constitution.

Two factors, then, determine the limits of efficiency in the bodily organs—heredity and use. And these two co-operate in such a way that we may say, either that due use is the essential condition of the effective development of the hereditary powers, or that heredity serves to condition their effective development through use. But though closely related, so that each may be regarded as conditional on the other, they are, if we accept the view that acquired characters are not transmitted as such, so far independent in that use adds nothing to, disuse subtracts nothing from, the hereditary store. It is, indeed, difficult to conceive how, on any view, the absence of the conditioning factor of normal use can be the efficient cause of a positive diminution of the balance at the bank of heredity. And Lamarckian thinkers have not succeeded in placing their conception of the matter in the clear light of a working hypothesis.

The amount of what we may term “modifiability” by use differs a good deal in the several organs and tissues. The teeth of carnivora and the antlers of deer may be cited as structures in which the conditioning effects of use form a relatively unimportant factor. On the other hand, the nervous system, with which we are here primarily concerned, is of all animal structures that in which what is acquired may attain the greatest importance in the successful conduct of life; the nature and the range of behaviour affording an index of the amount of modifiability in this respect.

We have already seen that instinctive behaviour is primarily a matter of the first occasion on which any given action is performed, and that many instinctive acts are subject to subsequent modification in the light of the experience gained during the early performances. The range of such modification varies both in different animals and also with respect to different modes of behaviour in the same animal. The more fixed and deeply rooted an instinct the less readily does intelligence obtain a hold on it, so as to direct the behaviour into new channels of better accommodation to the circumstances. M. Fabre describes how a Sphex, one of the solitary wasps, instinctively draws its prey, a grasshopper, into the burrow by its antennæ. When these were cut off the wasp pulled the grasshopper in by the jaw appendages; but when these were removed she seemed incapable of further accommodation to the unusual circumstances. It would seem an easy and obvious application of intelligence to seize the prey by one of the forelegs. But this was not done; and the grasshopper was then left. Intelligence did not seem equal to meeting the altered conditions presented by the maimed grasshopper. Still, there was some modification of the normal instinctive behaviour; and, as Dr. Peckham has shown, there may be more than Fabre noted. Let us assume the existence of an animal whose every act is instinctive, whose whole behaviour is marked out in strictly hereditary lines, no new departures being acquired in the course of individual life. This extreme case would afford an example of what we may term completely stereotyped behaviour. On the other hand, let us assume the existence of an animal with no hereditary definiteness of reaction, whose every act is intelligent, whose whole behaviour is the result of individual acquisition. This antithetical extreme case would afford an example of what we may term completely plastic behaviour. It is questionable, however, whether either of these extreme types occur in nature. What we find in our study of animal behaviour is some intermediate condition in which both factors co-operate, with a predominance either of stereotyped instinctive response on the one hand, or of plastic intelligent acquisition on the other hand. And in the latter case, as such behaviour approaches its ideal limits, we have modifiability under the circumstances of individual life at its maximum.

The evolution of intelligence as such runs parallel with the evolution of plastic behaviour; and this plasticity is necessitated by the variety and the complexity of the conditions of life—a variety and a complexity requiring many subtle modifications of response to enable the behaviour to reach accommodation to the changeful exigencies of diverse circumstances. To meet constant and relatively fixed conditions stereotyped instinctive responses suffice; and the elimination under natural selection of those individuals which fail to respond in fixed ways by specially adaptive behaviour tends to render definite the hereditary channels of nervous intercommunication. An inherited system of no little complexity may thus be evolved; of which we have seen examples in our study of instinctive behaviour. But the essential condition of the successful working of such a system is constancy in the environment to which the instinctive behaviour is adapted. Completely stereotyped behaviour, in its theoretical perfection, is in exact adaptation to the circumstances. Where instincts are only relatively perfect, further adaptation is secured through congenital variation and the survival of the individuals in which the behaviour is better adapted to the comparatively invariable circumstances. This is one line of evolution. But the evolution of intelligence is along independent lines of progress. Both, however, result from the functional activity of the same nervous system, they jointly determine the behaviour, they interact not only in the course of individual life but in the process of evolution, and they are both subject to the incidence of natural selection, which can determine whether the one line or the other shall preponderate—whether instinct or intelligence shall dominate behaviour.

If an answer must be given to the question whether instinct or intelligence has priority in the course of the evolution of behaviour, it may be urged that, on theoretical grounds, the claims of instinct are the stronger. Taking animals as we actually find them, however, they afford numberless examples of behaviour at first instinctive but subsequently modified, in greater or less degree, in accordance with the teachings of experience. Let us, first, assume that the environment is slowly changing, or has changed, in some definite manner. Such change would, of course, be relative, and might be due, either to new conditions brought to bear on the animal, or to the animal being itself brought, in the expansion of its life, within their influence. The old instinct is no longer quite adapted to the changed circumstances. If the change were sufficient in amount, and occurred somewhat suddenly, variations of instinct might not occur soon enough to enable the animal to reach adaptation by the gradual process of natural selection. If dependent on instinct alone the animal would, under these circumstances, be eliminated. But if intelligence were able to modify the behaviour to meet the new conditions this elimination would be prevented. In successive generations intelligence would constantly modify behaviour in the same manner and in a definite direction. Meanwhile congenital variations in different directions would occur. Those which were in directions antagonistic to that dictated by intelligence would tend to thwart accommodation and render it less effectual; but those which were coincident in direction would conspire with accommodation and render it more effectual. The individuals in which variations of instinct tended to thwart intelligence would be eliminated; while those in which coincident variations assisted and aided intelligent modification would survive. Thus intelligence would lead the way along lines which congenital variations would follow. And in the course of a number of generations the new instinct would reach the fully adaptive level, and further modification by intelligence would become unnecessary unless the environment continued to change yet more. Individual accommodation of behaviour would in this way determine the direction of instinctive variation; and yet throughout the process there would be, strictly speaking, no transmission of the intelligently acquired characters of the behaviour.

But though under constant and uniform changes in the environment the net result would be only a guided variation of the original instinct, under more variable and indefinitely changing circumstances the result would be different. The higher animals exhibit an intelligent plasticity which enables them to meet the requirements of the more complex environment into which their wider life has risen; for evolution lifts the animal from narrower into progressively wider spheres of activity and behaviour, so that its environment becomes relatively more complex. Here stereotyped behaviour would be rather a hindrance than an advantage. The winning animal in life’s struggle would be the one in which behaviour was most rapidly and most surely modified to meet particular needs—the one in which the teachings of experience were most promptly utilized in effective action. The inevitable tendency of the evolution of intelligence must be disintegration of the stereotyped modes of behaviour and the dissolution of instinct. Natural selection, which under a uniform and constant environment leads to the survival of relatively fixed and definite modes of response, under an environment presenting a wide range of possibilities leads to the survival of plastic accommodation through intelligence. It is not that intelligence has any direct influence tending to undermine the hereditary foundations of instinct, for acquired plasticity is not inherited as such; it is rather that when the stereotyped and the plastic are pitted against each other in the struggle for existence in the wider, freer, and more varied life of the higher animals the plastic survives and the stereotyped succumbs.

Imperfect as is our present knowledge of the manner in which the nervous connections implied in psychological associations are established, there can be no question that they are acquired in the course of individual life; they are modifications of nervous structure due to a special mode of use under the conditions of experience. Here, then, in the case of the nervous system, as in that of the bodily organs before mentioned, two co-operating factors determine the limits of efficiency—heredity and use. Just as the heart and lungs must inherit the power of standing abnormal strain if the animal is to avoid elimination in times of unwonted exertion, so must the nervous system inherit some reserve power of dealing effectively with unwonted circumstances by intelligent accommodation, if the animal is not to fall a victim to such circumstances. In other words, at times of heightened competition those animals which can draw on a reserve fund of intelligent accommodation will survive, while the stupid blunderers will be eliminated. We may term this reserve fund of intelligent accommodation, this inherited ability to meet specially difficult circumstances as they arise, innate capacity. From the nature of the case it must be indefinite, for it must carry with it the ability to meet unforeseen combinations of the environing forces by new combinations of the results of experience. Its distinguishing mark is plasticity, in contradistinction to the stereotyped fixity of typical instinct. And accompanying its evolution there is probably, as we have seen, a dissolution of its antithesis, instinct. Thus may we account for the fact that man, with his great store of innate capacity, has so small a number of stereotyped instincts.

But the dissolution of instincts is not complete. Residua are left in the inherited mental constitution. And these we term congenital tendencies and propensities. They differ from the typical instincts in the fact that the definiteness of response has been lost. They dictate a general trend of action, but the particular application in behaviour is due to intelligent accommodation. They are commonly spoken of as instinctive; and their mode of origin justifies the use of the adjective in association with the term “propensities.” But it must be remembered that the behaviour to which they lead is not, as such, wholly instinctive; it is a joint product of instinct and intelligence, the general trend being due to the instinctive propensity, while the mode of application is guided by intelligence.

There is, however, another way in which analogous propensities may be ingrained in the mental constitution, not as residual vestiges of old instincts, but as congenital rudiments fostered by new habits. It is a well-known and familiar fact that the frequent repetition of intelligent accommodation in certain definite lines begets habits, which so far simulate instincts as to be commonly described in popular speech as instinctive. Professor Wundt indeed places them in the category of “acquired instincts”—a usage which we regard as unsatisfactory, seeing that it tends to mask the distinction between the congenital and acquired factors in behaviour, and seeing that we have the well-defined term “habits” for acts rendered to a large extent automatic through repetition. Lamarckian thinkers regard habit as the mother of instinct, assuming that the acquired automatism of one generation may be transmitted to become congenital in the succeeding generation. This conclusion we provisionally reject, regarding the basal assumption as at present unproven. But though we cannot accept the view that habit is the mother of instinct, we regard it as not improbable that habit may be the nurse of congenital propensities. Remembering that similar habits are acquired by animals of the same species throughout a series of succeeding generations, and assuming that congenital variations are constantly occurring in many directions, it seems probable that some of these variations will be coincident in direction with the acquired habits. Thus would arise a congenital propensity to perform the habitual acts; and should they be of sufficient importance in the conduct of life to be subject to the action of natural selection, those animals in which such propensities were congenital would survive, whereas those in which no such propensities existed would be eliminated. It is unnecessary, however, to elaborate this conception further, since it is in line with that already discussed in considering the influence of intelligence in fostering a diversion of instinct under changing circumstances. Then we were considering how habit may lead to a congenital change in an old instinct; here we are dealing with the development of a new propensity.

Sufficient has now been said to illustrate some of the ways in which instinct and intelligence interact in the evolution of behaviour. Such interaction is further exemplified in the social life of animals, which will be dealt with in the next chapter.