STUDIES IN ANIMAL BEHAVIOR m O; II I O m O STUDIES IN ANIMAL BEHAVIOR S. J. HOLMES, Ph.D. ASSOCIATE PROFESSOR OF ZOOLOGY, UNIVERSITY OF CALIFORNIA BOSTON: RICHARD G. BADGER TORONTO: THE COPP, CLARK CO., LIMITED COPYRIGHT, 1916, BY RICHARD G. BADGER All Rights Reserved Made in the United States of America The Gorham Press, Boston, U.S.A. PREFACE present volume is largely devoted to sub- jects with which the writer's own investigations in animal behavior have been more or less closely concerned. For the interest of the general reader the special contributions of the writer have been made subordinate to the broader aspects of these subjects. Although there is little relationship be- tween the various topics dealt with, the different chapters are not devoid of a certain unity of aim. The several types of behavior here described were studied in the endeavor to get a fuller insight into their mechanism, or to interpret them from the ge- netic point of view. These two methods of attack are not opposed or mutually exclusive, as is some- times implied, but complementary. We cannot ob- tain a complete explanation of behavior by an analy- sis of the activities of the individual alone ; it is nec- essary to know also the evolutionary history of the species, and the various steps by which its present behavior has been acquired. The first chapter is historical and will, it is hoped, prepare the reader for a better appreciation of the general aim and import of what follows. Some of the chapters in this volume have appeared elsewhere as special articles. I wish to thank Dr. J. McKeen 3 4 Preface Cattell for his generous permission to republish chapter VI from Science, and chapter XIII and parts of chapter XI from the Popular Science Monthly. The Wisconsin Society of Natural His- tory has very kindly allowed me to republish chapter III which originally appeared in the bulletin of that society for June, 1912. CONTENTS CHAPTER PAGE I. ANIMAL PSYCHOLOGY, THE OLD AND THE NEW 9 II. THE EVOLUTION OF PARENTAL CARE . . 35 III. THE TROPISMS AND THEIR RELATION TO MORE COMPLEX MODES OF BEHAVIOR . 50 IV. THE PROBLEM OF ORIENTATION ... 69 V. THE REVERSAL OF TROPISMS 93 VI. THE BEGINNINGS OF INTELLIGENCE . .120 VII. SOME CONSIDERATIONS ON THE PROBLEM OF LEARNING 139 VIII. THE IMPLICATIONS OF TRIAL AND ERROR . 155 IX. BEHAVIOR AND FORM 166 X. THE BEHAVIOR OF CELLS . . . . 177 XI. THE INSTINCT OF FEIGNING DEATH . . 197 XII. THE RECOGNITION OF SEX .... 219 XIII. THE ROLE OF SEX IN THE EVOLUTION OF MIND 239 XIV. THE MIND OF A MONKEY .... 245 INDEX , 263 39677 STUDIES IN ANIMAL BEHAVIOR STUDIES IN ANIMAL BEHAVIOR i ANIMAL PSYCHOLOGY, THE OLD AND THE NEW PSYCHOLOGY is a science that has had few his- •*- torians, and the special province of animal psy- chology has never been accorded the dignity of a full and thorough historical treatment. It is far from the intention of the present writer to take up this neglected task. But there are a few salient fea- tures of the development of this branch of the sci- ence which it may be desirable to consider briefly in order to prepare us for an appreciation of the aims and methods of present-day animal psychology. The animal mind has enlisted a certain amount of interest from the earliest times. Do animals have souls? If so, Do these souls continue to live after death? How do the souls of animals differ from the souls of men? Do animals reason? — these are some of the questions which exercised the earliest philosophers, and they have been asked more or less persistently down to the present time. Among 9 io Studies in Animal Behavior comparative psychologists these questions, except perhaps the last, are no longer in the foreground of interest. Professor John Dewey has remarked that in philosophy we do not solve our problems; we get over them. And these old questions about the animal mind that have so long perplexed enquiring spirits we have now gotten over, rather than solved, and left behind in order to turn our attention to more fruitful subjects of investigation. Modern psychology troubles itself very little about the soul as an object of enquiry, and animal psychology concerns itself with this subject still less. There are a number of questions about conscious- ness which still occupy us. Where did consciousness begin in the course of evolution, if it can be said to have had a beginning at all? How is consciousness related to the bodily structure of the organism? What are the criteria by which its presence in an organism may be recognized? How does it influ- ence behavior, if we grant that it can influence be- havior?— these and many other related problems continue to perplex the scientific worker and the metaphysician alike. But many of the problems are beginning to recede from the foreground of inter- est, and to a considerable extent we may get over them in the future and leave them to one side, al- though at some time we may recur to them with re- newed insight and find that they wear a quite dif- ferent aspect. Even among the ancient Greeks we find much Animal Psychology, the Old and the New n the same diversity of opinion in regard to the men- tal life of animals that occurs in modern times. While it was contended by some that the animals occupy a position immeasurably below that of man, the Greeks in general were inclined to a more gener- ous estimate of the animal mind. Apparently there were no sustained studies of animal psychology among the ancient Greeks, with the single possible exception of the observations of Aristotle, most of which, however, were of a desultory character. Aristotle's three treatises on zoology contain nu- merous records of the habits of animals, and his general estimate of the mental powers of animals might well pass for that of a conservative psycholo- gist of the present time. Most animals, according to Aristotle, "appear to exhibit gentleness or feroc- ity, mildness or cruelty, courage or cowardice, fear or boldness, violence or cunning; and many of them exhibit something like a rational consciousness, as we remarked in speaking of their parts. For they differ from man, and man from animals, in a greater or less degree; for some of these traits are exhibited strongly in man, and others in animals." The ani- mal mind corresponds to the undeveloped human mind, for in speaking of infants Aristotle remarks, "nor does their soul at this period differ in any re- spect from that of an animal." ("Hist, of Ani- mals," Bk. 8.) The sage observation that the life of animals "may be divided into two acts, procreation 12 Studies in Animal Behavior feeding, for in these two acts all their interests and life concentrate," recalls the aphorism of Schiller that hunger and love are the ruling forces of the world. Aristotle was very far from regarding be- havior as determined by external stimulation. He had no conception of reflex action, no knowledge of the function of nerves, and he regarded the brain as an organ to temper the heat of the heart, the lat- ter being considered as the seat of sensation. The doctrine that animals reason found an ardent defender in Plutarch, who devotes two rather curious chapters of his "Morals" to a consideration of the mental powers of animals and their relation to the faculties of men. Different views are put forward in dialogue form much after the fashion of Plato. The discussions are of interest mainly as present- ing the views of animal intelligence current in Plu- tarch's time. Porphyry's views of animals were much like those of Plutarch, although they were interwoven with his Neo-Platonist doctrines concern- ing the nature of the soul. The Romans, like the Greeks, were generally in- clined to give credit to the lower animals for a considerable amount of intelligence. Pliny with his usual uncritical judgment related many stories of wonderful animal sagacity, and Celsus contended for the essential similarity of mind in brute and man. Galen shows an approach to the conception of in- stinct as it came to be understood by later writers. He dwelt at length on the adaptations of the struc- Animal Psychology, the Old and the New 13 tures of animals to the uses to which they are put, and upon the inborn proclivities of animals for using their parts in the proper way. Just as the muse incites the poet, so do the innate impulses of ani- mals lead them, without instruction, to perform the acts needful for their life. Similar views were ex- pressed in the writings of Cicero, (De Natura De- orum). Seneca wrote in much the same vein: "What practice teaches is slowly acquired and is made after many patterns; what nature teaches, that is the same for all, and, as soon as there, it takes place without reflection." "Nature teaches nothing but self-preservation and the knowledge necessary for this end." During the Christian era there were few recorded observations or speculations on animal psychology before the Renaissance. Interest was centered mainly in man, and especially in things affecting the wel- fare of his soul, and most of the attention that was bestowed on the mental life of animals was owing chiefly to the bearing of the subject upon theologi- cal teachings. The church emphasized the inferior- ity of the brute creation and taught that it was brought into existence solely for the service of man, although this doctrine was sometimes qualified by the admission that certain noxious creatures might have been produced by the devil, or came into ex- istence as a consequence of the Fall. The conception of instinct which had assumed more or less definite outlines among Roman authors 14 Studies in Animal Behavior came to be more clearly defined by the theological writers of the latter part of the Middle Ages. Of these St. Thomas Aquinas stands preeminent as an authority on the animal mind as upon most things else. Aquinas makes a sharp distinction between the sensitive soul (anima sensitiva) and the intellectual soul (anima intellectualis), ascribing the former to brutes, and the latter only to man. Animal and man are therefore separated by a broad and impassable barrier. Animals have sensations, sensory memory, but they have no reason (Summa Theologica, LXVI and LXXXIIB), no real freedom (/. c. CXIII), no responsibility. "No activity of the sen- sitive part can have place without a body. But in the souls of dumb animals we find no activity higher than the sensitive part. That animals neither under- stand nor reason is apparent from this, that all ani- mals of the same species behave alike, as being moved by nature, and not acting on any principle of art; for every swallow makes its nest alike. There- fore there is no activity in the soul in dumb animals that can possibly go on without a body." (/. c. LXVI.) "Sense," he says, "is found in all animals, but ani- mals other than man have no intellect; which is proved by this, that they do not work, like intellec- tual agents, in diverse and opposite ways, but just as nature moves them to fixed and uniform specific activities." "Sense is cognizant only of singulars, but intellect is cognizant of universals. Sensory Animal Psychology, the Old and the New 15 knowledge extends only to bodily things, but intellect takes cognizance of things incorporeal, as wisdom, truth and the relations between objects." The won- derful adaptiveness and perfection of instinct are not to be ascribed, in any measure, to the animal's own initiative, but redound to the credit of the Cre- ator. The ideas of Aquinas on animal psychology, like his ideas on so many other subjects, were greatly influenced by the teachings of Aristotle, but he em- phasized much more than his master did the differ- ences between -brute and human intelligence. Aris- totle greatly perplexed his followers and interpre- ters by his vague and somewhat vacillating treat- ment of the relations of the rational soul to the sen- sitive soul and to the body. But the learned St. Thomas promptly settles these problems with a de- cisiveness that leaves no room for doubt concerning his own standpoint. The writings of Aquinas served to define the position of the church in regard to the status of the animal mind. The opinions of this prelate were widely followed and have continued to influence opinion even down to the present time. In the writings of Descartes we find animal be- havior interpreted, for the first time, in terms of the functions of the nervous system. Descartes was an original investigator of the structure and func- tions of the nervous system and he arrived at many of our fundamental conceptions of nerve physiology. His celebrated doctrine that animals are automata, i6 Studies in Animal Behavior that their activities are entirely determined by their bodily organization without knowledge or will of their own, was simply the result of carrying out to its full logical consequences the mechanistic physi- ology that he had adopted. "I have diligently en- quired," he says, "whether all the motions of ani- mals came from two principles, or only from one ; and as I find it clear that they arise from that prin- ciple alone which is corporeal and mechanical, I can by no means allow them to have a thinking soul. Nor am I at all hindered in this conclusion, by the cunning and sagacity of foxes and dogs, nor by those actions done from lust, hunger or fear; for I pro- fess to be able easily to explain all these things by the sole conformation of their limbs." Although Descartes explained by a physical mech- anism what had been previously ascribed to the sen- sitive soul, his conclusion was equally acceptable to most of the dignitaries of the church, inasmuch as it preserved an essential distinction between the brute creation and man. In fact this disposition of the animal world was hailed with satisfaction by many church authorities, and it was rapidly followed by various writers several of whom developed even more extreme views. Malebranche, for instance, tells us that "Among cats and dogs and other ani- mals there is no intelligence, no spiritual soul as we are commonly told. They eat without pleasure, cry without pain. They grow without knowing it; they desire nothing; they know nothing, and if they act Animal Psychology, the Old and the New 17 with address, and in a manner that indicates intelli- gence, it is because God, in making them for self- preservation, has constituted their bodies in such a way that they withdraw organically and without knowing it from all that can destroy them and that they seem to fear." A reaction from such views was inevitable. They incurred the ridicule of La Fontaine, the protests of the good Father Bonjeant, who believed that ani- mals were inhabited by the souls of demons, and the criticism of Thomasius, Gassendi, Leibnitz and many others, of whom the French Inspector of For- ests, G. LeRoy, occupies a position of especial prom- inence. LeRoy's Lettres sur les Animaux contain many observations on the training of animals and other indications of intelligence, but their chief dis- tinctive feature is their attempt to explain the actions commonly attributed to instinct as the result of in- telligent experience. A more strenuous attempt in the same direction is found in Erasmus Darwin's Zoonomia. Darwin argues that the marked re- semblance in the form and functions of the bodies of man and animals should lead us to expect a simi- lar endowment of sensations, emotions and mental powers generally. He endeavors to show that the habitual acts of animals are learned like most of our own, and that certain habits appear innate sim- ply because we do not attend with sufficient care to the early stages of their formation. The chick walks soon after hatching, but before this time it moves 1 8 Studies in Animal Behavior its feet about while in the shell. It pecks at food, but before this it opens and shuts its mouth and swal- lows some of the white. The ability of calves and colts to walk soon after birth is attributed to their struggles while still in the uterus. Even the human foetus sucks in amniotic fluid and "learns to swallow in the same manner as we learn all other animal ac- tions which are attended with consciousness, by the repeated efforts of our muscles under the control of our sensations and volitions." Breathing, a more difficult case, is accounted for as follows: "The inspiration of air into the lungs is so totally different from that of swallowing a fluid in which we are immersed, that it cannot be acquired before our nativity. But at this time when the circulation of the blood is no longer continued through the placenta, that suffocating sensation which we feel about the praecordia when we are in want of fresh air, disagreeably affects the infant and all the muscles of the body are excited into action to relieve this oppression; those of the breast, ribs and dia- phragm are found to answer this purpose; and thus respiration is discovered; and it is continued through- out our lives as long as oppression begins to occur." Darwin is not daunted even by the clear indications of instinct presented by social insects, for he says, "If we were better acquainted with the histories of those insects which are formed into societies, as the bees, wasps and ants, I make no doubt but we should find that their arts and improvements are not so Animal Psychology, the Old and the New 19 similar and uniform as they now appear to us, but that they arose in the same manner from experience and tradition, as the arts of our own species, though their reasoning is from fewer ideas, and is busied with fewer objects and is exerted with less energy." Similar views were developed by Condillac in his Traite des Animaux published in 1766. What had come largely through the writings of Aquinas and his followers to be typical catholic doc- trine concerning the mental life of animals found a very able defender in S. H. Reimarus, a prominent prelate of the church, but at the same time an indus- trious worker in the field of animal behavior. His chief work, Allgemeine Betrachtungen fiber die Triebe der Thiere, remained for a long time the most extensive and thorough treatise on the subject. It passed through four German editions and was translated into French and into Dutch, and soon came to be regarded as a standard authority. The views of LeRoy, Erasmus Darwin and Condillac are subjected to a rigorous criticism which in most re- spects must be regarded as very well founded. The labors of Reaumur, Rosel von Rosenhof, Huber, Bonnet, Buffon, Spallanzani and other naturalists of the seventeenth and early part of the eighteenth centuries had added greatly to our knowledge of the instinctive behavior of animals, and Reimarus, from his wide acquaintance with literature and from his own experience, had little difficulty in making out a strong case for instinct. The key-note of the work 2O Studies in Animal Behavior is expressed in the title of an earlier treatise In- stinctus brutorum existensis Del, eiusdemque sapien- tissimi, indicem, issued in 1725. The wonderful instincts of animals cannot be explained on the basis of acquired habits ; they are innate endowments, per- fectly adapted to secure the welfare of the individual animal or its progeny, and hence an irrefragable tes- timony to the wisdom and beneficence of the Crea- tor. Reimarus was strongly influenced by Aristotle and St. Thomas Aquinas. He is at great pains to show that there is a fundamental difference between instinct and reason, and that man alone has true rationality. He concedes to the animals perceptions, memory, volition and the ability to learn; but he de- nies that they have abstract or general ideas or any power of passing from one representation to an- other. By numerous other writers animal instinct came to be dwelt upon as affording some of the most conclusive evidence of design. Paley dilates upon it at length in his Natural Theology, and one of the volumes of the Bridgewater treatises is de- voted to this fruitful topic. Throughout human history there have been sev- eral motives behind the various divergent opinions that have been held in regard to animal psychology. The impulse of the sympathetic lover of animals to attribute to his dumb friends and dependents a gen- erous meed of intelligence has always been a more or less potent influence. And then there is the temp- tation to tell as remarkable a tale as possible of the Animal Psychology, the Old and the New 21 performances of the creatures described. Both these tendencies frequently lead to reading into an animal's actions an unwarrantable amount of sagacity. But more powerful than these motives, especially during the Christian era, have been the various theological prepossessions of different schools. In the iyth and 1 8th centuries we meet with two opposed tendencies in the interpretation of animal behavior; one toward attributing the actions of animals to the same facul- ties that are possessed by human beings and leading to a limitation of the sphere of instinct, or even to denying instinct altogether; the other toward explain- ing animal behavior, so far as possible, in terms of instinct or automatism, with the limitation of reason and all higher mental attributes to man alone. The philosophical skeptics, in general, were partial to- ward the first; the defenders of the faith toward the second. The one class of writers attempted to link man and animal more closely together, and to show their fundamental kinship; the other tried to make the gap between man and animal as wide as pos- sible, and to show that there is an essential differ- ence between them. "After the error of atheism," says Descartes, "there is none which leads weak minds further from the path of virtue than the idea that the minds of animals resemble our own, and therefore that we have no greater right to a future life than have gnats and ants, while on the contrary, our mind is quite independent of the body, and does not necessarily perish with it," 22 Studies in Animal Behavior The same considerations that inspired this utter- ance of Descartes still have weight in determining the attitude of modern students toward problems of animal behavior. In the conclusion of his interest- ing book on the Psychology of Ants and of Higher Animals, Father Wasmann, one of the foremost in- vestigators of the behavior of social insects, writes in regard to modern evolutionary psychology that "By denying the existence of the essential difference be- tween animal and human psychic faculties this psy- chology not only raises brutes to the dignity of man, but degrades man to the level of the brute. Would to God that this were done in theory only; but alas! the practical consequence of this false theory is the demoralization and brutalization of man." There is no denying the serious import of the problems about which comparative psychologists do battle. But even if the victory should fall to the most materialistic of the contending parties it is hardly to be expected that the consequences would be as dire as Wasmann surmises. We are bid to beware of the same old scarecrow that has so often appeared in the path of scientific progress. But its aspect is growing less terrible as time passes, and it is always the part of the man of science to go straight ahead as if it were not in existence. The first to grapple in a very serious way with the problem of the evolution of instinct was La- marck, although we meet with suggestions in regard to the inheritance of the effects of habits in the writ- Minimal Psychology, the Old and the New 23 ings of LeRoy. Lamarck regarded the evolution of animals as brought about, to a very large extent, by psychic factors. His three primary divisions of the animal kingdom, the apathetic, the sensitive and the intelligent animals, are based on psychological distinctions. The apathetic animals, such as the in- fusoria, polyps, etc., are, like plants, irritable, but devoid of consciousness and spontaneity. The tran- sition from the apathetic to the sensitive animals comes with the development of a nervous system which was erroneously regarded as absent in some of the higher members of the apathetic group. Con- sciousness first appears in a vague form, but as or- ganization advances and the nervous system and sense organs become more specialized and perfected, consciousness comes to assume more of a guiding role in the process of evolution. Everywhere con- sciousness is considered by Lamarck as dependent upon the organization of the nervous system, but as he is an "interactionist" he regards the activities of animals as initiated to a large extent by their feel- ings and desires. As is well known, Lamarck explained instinct as inherited habit; but habits depend upon antecedent desires of the animal; these lead to efforts by which the desires may be satisfied, and by frequent repe- tition the acts thus prompted become habitual and are passed on as inherited proclivities to following generations. With the appearance of wants determination of 24 Studies in Animal Behavior action becomes gradually transferred from the out- side of the organism to the inside. New wants en- gender new habits and give rise to new organs. In- stinct is regarded as internal impulsion instead of response to outer stimuli. It stands sharply marked off from the activities, however complex, of the lower organisms whose behavior is entirely deter- mined by their environment. With the advent of the vertebrates with their well- developed brain we have the appearance of intelli- gence and free will. Lamarck sedulously avoids the anthropomorphism of many previous writers in maintaining that intelligence and will can arise only in higher animals which have the requisite nervous organization. With the perfecting of this organiza- tion the higher mental faculties become further de- veloped, and reach their culmination in the mind of man which is regarded as the outcome of a con- tinuous process of evolution. Although much had been written on the instincts and habits of animals in the half-century following Lamarck, there was but little contributed to the doc- trine of mental evolution until we come to Herbert Spencer, who undoubtedly ranks as one of the great- est of all genetic psychologists. Spencer's Principles of Psychology was published in 1855, but several years later after the Darwinian theory had been pro- mulgated it was reissued in a revised and consider- ably enlarged form. His treatment of life and mind from the common standpoint of the adjustment of Animal Psychology, the Old and the New 25 internal relations to external relations, his derivation of instinct from reflex action, his filiation of the process of reasoning with perception, his attempt to show that no sharp distinction can be drawn between instinct and reason, his efforts to give an account of the origin of the intuitions of space and time and the fundamental forms of thinking in terms of experience gradually accumulated through inheritance, and his theories of the genesis of moral impulses and aesthetic sentiments, are among the many notable features of this very original and closely reasoned book. Spencer was a believer in the transmission of ac- quired characters, and much of his psychological speculation is based upon this doctrine. Many of his cherished deductions, however ably wrought out, will have to be discarded if the effects of experi- ence, as so many biologists now believe, are not transmitted to the following generation. But if the foundation of many of Spencer's doctrines were to be removed, much of permanent value would nevertheless remain unshaken. The genetic method in the study of psychology, as well as in many other fields of thought, owes much to Spencer's illuminat- ing thought and stimulating influence. Only four years elapsed between the publication of Spencer's Psychology and the appearance of Darwin's Origin of Species. The latter work marked an epoch in the history of psychology as well as of biology. Darwin's theory of natural selection 26 Studies in Animal Behavior afforded a means of explaining the evolution of in- stincts, as of corporeal structures, by the slow accumulation of favorable variations. In the chap- ter on "Instinct" in the Origin of Species, Darwin showed that even wonderful and complex instincts such as those of the hive bee did not present in- superable difficulties to his theory. Could the theory be extended to explain the derivation and evolution of intelligence and possibly also the development of the human mind itself from that of lower forms? In the Descent of Man Darwin attempts to- show that the mind of man is fundamentally the same in kind as that of the animals, and differs only in de- gree of development. He attempts to show that the moral sense which had been often held up as man's peculiar possession and glory is an outcome of the social instincts and emotions found in higher ani- mals. Such conclusions, with their far-reaching con- sequences, naturally aroused strong opposition on the part of many conservative people, while at the same time they afforded an inspiring outlook to Dar- win's followers. Attempts to trace the genesis of human faculties were the natural outcome of the stimulus which Darwin gave to the study of compara- tive psychology, just as attempts to trace the lines of descent of various groups of the animal kingdom followed closely upon the early battles over the mutability of species. Romanes' volumes on Ani- mal Intelligence, Mental Evolution in Animals and Animal Psychology, the Old and the New 27 Mental Evolution in Man are among the best-known products of this movement. There was a strong tendency toward anthropo- morphism in many post-Darwinian writers, as there was in several of the skeptics of the eighteenth cen- tury. The effort to show that the human mind evolved from the animal mind led many to read an undue amount of intelligence into the activities of animals. The works of Perty, Biichner, Vogt, Brehm, and to a certain extent Romanes afford illus- trations of this failing. What Wasmann calls "hu- manizing the brute" became a favorite theme. Sev- eral writers of the early post-Darwinian period restrained within reasonable limits whatever bias they may have had toward anthropomorphism, and contributed observations and experiments which have thrown much light on the problems of animal psy- chology. Among these may be mentioned Lubbock, Forel, Lloyd Morgan, Dr. and Mrs. Peckham, Mc- Cook and many others. The scholarly works of Groos on The Play of Animals and The Play of Man are excellent examples of the application of Darwin- ian principles to the interpretation of behavior. Among post-Darwinian authors on animal psy- chology there soon arose a division between the neo- Darwinians who attributed evolution mainly or solely to natural selection, and the neo-Lamarckians who attached much greater importance to the in- heritance of the effects of experience. This division 28 Studies in Animal Behavior was foreshadowed by the different degrees of im- portance attached by Darwin and Spencer to the factors of natural selection and the transmission of acquired characters. Darwin accepted the latter the- ory, but assigned to the Lamarckian factor a subor- dinate role as compared with natural selection. Spen- cer, who had already elaborated his system of psy- chology on the basis of the Lamarckian theory, naturally attached great weight to that doctrine, but when the theory of natural selection was announced he cordially received it, but he appealed to it mainly as a helpful subsidiary hypothesis. Since Weismann made his attack upon the La- marckian theory he has been followed by a consider- able number of psychologists, such as Lloyd Morgan, Forel, Groos, Whitman, Baldwin and Zeigler, who reject entirely the doctrine that acquired characters are transmitted. Among the neo-Lamarckians who, though somewhat in the minority, still represent a flourishing school, there are all sorts of views re- garding the potency of natural selection, some writ- ers going so far as to cast the theory aside as a visionary and groundless speculation. Romanes, Eimer, Haeckel, Hering, Wundt, Cope, Semon and Pauly are among the principal writers of this school who have concerned themselves with comparative psychology. The combination of vitalism and La- marckism which is presented in the writings of Pauly and France affords, in the opinion of its adherents, a method of accounting for the development of Animal Psychology, the Old and the New 29 adaptations quite independently of natural selection. This is done by endowing the organism with a teleo- logical principle which makes all the adjustments re- quired to meet its needs ; these adjustments are then transmitted to the descendants and thus gradually effect a progressive adaptive modification of the species. Without raising the question as to whether the vitalistic explanation of adaptation really explains anything, it might be remarked that, since a teleo- logically working principle is assumed as the basis for the acquired adjustments of the individual, the same principle might also be evoked to guide the entire course of evolution without appealing to an agency of such questionable potency as the Lamarck- ian factor. Entia non sunt multiplicanda praeter necessitate™,. If we invoke any vitalistic agencies or teleological principles we might as well give them plenty to do. The theory of organic evolution having been firmly established, and the conviction having be- come quite general among psychologists that the hu- man mind has resulted by a continuous process of development from the mind of animals, the contro- versial interest in various questions that stimulated the earlier post-Darwinian students of the animal mind has to a considerable extent subsided. The effort to trace the evolution of particular instincts and mental faculties continues to afford an absorbing and fruitful occupation, but such work is pursued, 30 Studies in Animal Behavior not so much with the aim of establishing the fact of evolution in a particular field, as of illuminating the subject in the light of its historical development. In human psychology the genetic method of study is pursued to a large extent in the works of Stanley Hall, Groos and Baldwin, and in the recent volumes of Kirkpatrick on Genetic Psychology and of Parmelee on Human Behavior. Modern "phi- losophy of education" has been influenced in no small degree by studies and speculations in comparative psychology, and especially in this country by those of Stanley Hall and his school. The doctrine of re- capitulation figures quite largely in the writings of this school, and, while the results of its applica- tion have not always been happy, the general influ- ence of Hall and his followers has given a strong stimulus to genetic psychology and the scientific study of problems of education. Much attention has been devoted to the experi- mental analysis of instinct. The complex instincts of crustaceans and insects have been shown by means of operative experiments on the nervous system to be, to a large extent at least, capable of analysis into reflex activities of the various segments. A brainless crayfish will walk, eat food, reject innu- tritious substances, defend itself when seized, and perform various other complex activities. When any of the segmental ganglia of the nervous cord are cut off from communication with the others by severing the connectives, the appendages of the cor- Animal Psychology, the Old and the New 31 responding segment will respond adaptively to many stimuli that are applied to them. Many of the more complex activities of the animal may be explained as chain reflexes in which one act affords the stimu- lus to the performance of other acts. The be- havior of the whole animal might be regarded as the result of a sort of social cooperation of the activi- ties of its various quasi-independent nervous ganglia. One cannot say that the seat of instinct in such an animal is in the brain or any other part of the nerv- ous system. The instincts of the animal are the outcome of its general organization; they are the expression of the workings of the organic mechan- ism. Experiments indicate that the same conclusion applies to the instinctive behavior of vertebrates. Whether or not we regard instinct with Herbert Spencer, as "compound reflex action," we may be justified in concluding that it is everywhere a function of organization. With the same aim at analysis a large amount of work in recent years has been devoted to the sub- ject of tropisms. Loeb's celebrated theory of orien- tation attempts to explain these directed movements of animals in terms of reflex action; many instincts of animals are apparently the outcome of these tropic reactions; even in the higher animals it is probable that certain tropic responses are still re- tained. As the tropisms are discussed more fully in a later chapter this brief indication of their im- portance in the development of animal psychology 32 Studies in Animal Behavior may perhaps suffice. From the standpoint of analysis as well as that of evolution, a considerable interest attaches to the behavior of the simplest forms of life. Here, if anywhere, one might expect behavior to be capable of analysis into physical and chemical processes. The behavior of the Protozoa for this reason has attracted a number of careful workers who have endeavored to test how far such behavior can be explained in terms of simpler factors. Jen- nings in particular has done a large amount of careful observational work in this field; he has shown in the case of Amoeba, which is often re- ferred to as an almost structureless mass of jelly, that the behavior is surprisingly complex, and at present incapable of being explained in terms of physical and chemical laws. This does not imply that the behavior of Amoeba is, in the last analysis, incapable of such explanation; it simply means that our present knowledge is inadequate for this task. Even the simplest organism is a very complex struc- ture from the standpoint of the chemist. And it is not to be wondered at that the simplest creatures often act in ways which we can neither predict nor explain. Minute study of the behavior of the low- est organisms, however, has revealed a remarkable amount of uniformity. The activities of these forms are stereotyped, yet plastic, and while no intelligence has been proven to occur in any of them, their be- havior is often capable of modifications in various Animal Psychology, the Old and the New 33 ways in adaptation to changed conditions of life. The behavior not only of the Protozoa but of all higher classes of animals has been studied in the last few years with a great increase of zeal -and thoroughness. The activities of hydroids, jelly-fish, worms, mollusks, echinoderms, crustaceans, as well as insects, and vertebrates, have engaged the atten- tion of a small army of investigators who are rapidly amassing a vast store of detailed knowledge. A very few years ago there was established a spe- cial periodical, The Journal of Animal Behavior, devoted exclusively to papers on animal psychology, while an increasing amount of literature on the sub- ject is going into other channels. The days of anecdotal psychology, when it was the fashion to bring together stories from various sources illustrative of animal sagacity, are passing. The psychological interpretation of animal behavior is a subject that abounds in pitfalls for the un- wary. To find out what probably goes on in an animal's mind requires close and continuous obser- vation, and usually experiments under carefully con- trolled conditions. The careful experimental work of Thorndike, Cole, Yerkes, Hobhouse, Small and many other investigators, has given us a more ex- act knowledge of the mental activities of the ani- mals studied than would have been possible through the collection of any quantity of scattered observa- tions. The results have sometimes proven disap- pointing to zealous champions of the high mental 34 Studies in Animal Behavior development of their animal friends. There is little ground for believing that animals have abstract or general ideas, or the power of de- liberate reasoning, but there is considerable experi- mental evidence that they have ideas of a simple sort and a certain power of inference. Careful in- vestigators at present show a wholesome caution about ascribing to the mind of the animal more than the facts really justify. Lloyd Morgan has laid down the principle, since known as the principle of Morgan, that no act should be ascribed to a higher mental faculty if it can be satisfactorily ac- counted for in terms of a lower one. The burden of proof is thus placed upon those who contend for the superior endowments of the animal mind. What we want are not stories of performances which might have involved unusual intelligence, but records of achievements which cannot be accomplished ex- cept by means of unusual intelligence. In the latter case only are we justified in ascribing to the animal the mental attribute in question. In following the principle of Morgan we may often fail to give to the animal full credit for the faculties it may really possess, but our conclusions will be sound so far as they go. II THE EVOLUTION OF PARENTAL CARE "INSTINCTS for the care of young extend far •^•down in the animal kingdom and their origin therefore dates back to an early period in the history of the earth. The important role which these in- stincts have played in the evolution of animal life, — a role which has increased in importance as animals have become more highly evolved, — renders the sub- ject of their origin and course of evolution one of especial interest to the comparative psychologist. With a full realization of the fact that phylogeny is a treacherous field I have nevertheless ventured in the following account to outline the probable way in which animals came to care for their offspring, and to point out briefly how parental care has been in- strumental in shaping the more advanced stages of the evolutionary process. In several respects, notably in relation to social and ethical evolution, the institution of parental care has formed the foundation for higher stages of de- velopment. Organized societies in animals generally have their beginning in the expansion of the family. In some insects, for instance, such as the ants, bees, and social wasps, various gradations may be traced 35 36 Studies in Animal Behavior between the simple family and the highly organized social state. Many insect communities consist merely of an enormous family resulting from a single fe- male parent; and when the community includes more than this the condition is generally a secondary out- growth of the domestic group. In the care of parents for their young we prob- ably find also the first traces of altruistic instinct. Family life is impossible on the basis of purely egois- tic behavior. Some altruism, however weak and limited in its scope, is the essential condition of the family group. In low forms it is limited to the care for offspring; later it may include other members of the species beyond the limits of the family; but it is a long time before it extends its blessings at all widely. Most creatures care not a fig for the wel- fare of any but their own immediate kin ; and where we find any consideration bestowed upon alien forms, as in the fostering of aphids by ants, it is done for the sake of something to be gained in return. Ani- mals in general live under conditions in which they cannot afford disinterested benevolence. While the long hard struggle for existence may have bred tender feelings and unselfish impulses it has pro- duced them for the same ultimate purpose that is subserved by sharp claws and good teeth, — the sur- vival and perpetuation of the species. Parental care is one of Nature's devices for race survival. But very few of the devices which Nature has hit upon have influenced so profoundly the course of evolu- The Evolution of Parental Care 37 tion in the higher animals. We cannot of course follow step by step all the stages in the evolution of the parental instincts and feelings, but by a comparative survey of the animal kingdom it is possible to construct, with some degree of probability, the main outlines of this development. Among the lower invertebrates the offspring have to shift for themselves at their very first appearance upon the stage of life. There are sometimes devices such as brood pouches for the protection of the eggs or young, but any active solicitude of parents for their offspring does not appear until we reach the higher invertebrate animals. This lack of parental interest is well illustrated by the behavior of an am- phipod crustacean, Amphithoe, which the writer studied in some detail a few years ago. In this form the eggs, and also the young for a few days after being hatched, are carried in a brood pouch on the under side of the body. When sufficiently agile the young make their way out of the brood pouch and swim away, the mother paying no more attention to them than to any other animate object. Indeed the carnivorous habits of this species make it more or less dangerous for the young to tarry long in the vicinity of their parent. Several times I have caught the young in a fine pair of forceps and offered them to the mother, who ate them without the least com- punction. It is only in a grossly material sense, therefore, that Amphithoe can be said to be fond of its offspring. 38 Studies in Animal Behavior So far as the writer is aware the same utter lack of maternal sentiment characterizes all the crusta- ceans as well as the numerous varieties of worms and mollusks. Among the arachnids Fabre has described how the female scorpion assists her young to hatch by carefully tearing away the egg membranes with her jaws. When the tiny scorpions are liberated they have the curious habit of mounting upon the back of the mother, who for a period of several days remains closely confined to her nest. In the spiders, although as a rule only an attitude of hos- tility is manifested toward other members of their own kind, the running spider Lycosa carries her co- coon about with her and when the young spiderlings are hatched they cling in a squirming mass to her body. The mother does not feed or actively care for her young in any way, and it is doubtful if ma- ternal care goes further than a sort of good-natured tolerance of her living burden. There is much evidence for the supposition that the first step in the evolution of parental care was taken in the formation of instincts to secure the proper environment for the development of the eggs. Instincts for depositing eggs in places which afford food for the young, instincts for making co- coons or other receptacles for the eggs, and instincts for concealing the eggs from the attacks of other animals are common in animals which are too primi- tive to exhibit any care for, or even recognition of their own offspring. Many mollusks plaster their The Evolution of Parental Care 39 eggs on stones or aquatic plants or bury them in gelatinous masses in the sand, but they certainly have not the least glimmer of an idea concerning the free swimming larvse to which the eggs will give rise. The cabbage butterfly instinctively deposits its eggs on cabbages or other cruciferous plants, the hot fly oviposits upon the hairs of horses and cattle, and the May fly drops its eggs into a pond or stream, but it is utterly out of the question to suppose that any of these creatures has the remotest notion of the larvse that will issue from its eggs, much less of the relation of the environment of the eggs to the needs of larval life. A further step in the direction of parental care is taken by the solitary wasps. In many species the female digs a hole and then goes in search of a par- ticular kind of insect or spider to serve as food for her young grub. The victim is stung in such a way that it is paralyzed but not killed, so that the young can live upon food that is not decayed. Then the prey is stored in the hole, an egg laid upon it, and the hole filled up with dirt and left, the mother never coming back to see how things fare with her young, nor concerning herself further with its welfare. In fact, most of these wasps never see their own progeny, and show no signs of recognition when the larvae are shown to them. All the wonderful in- stinctive acts by which the solitary wasps provide so well for the needs of their larvae are blindly per- formed, and give no indication of any feeling of pa- 4O Studies in Animal Behavior rental affection, even of the most rudimentary sort. A more advanced stage in the evolution of pa- rental care comes with the appearance of instincts for more or less continuous care of the eggs after they are laid. If the cocoon of the running spider is taken away from her, which is done only after a certain resistance, she will eagerly seize it again if she happens to encounter it. The egg mass is treated as an object of interest, although the young which issue from it are regarded with indifference. In- stincts for remaining with the eggs for a longer or shorter time after they are deposited are not infre- quent among fishes, and especially among those spe- cies which expend some labor in the construction of a nest. Generally this task is performed by the male, as in the sticklebacks and the common Amia or dog- fish of our ponds and streams. The males of these fishes remain in or near the nest after the eggs are laid, and they keep on the alert to drive away any intruder that ventures too near. The Amphibia, which as a group are remarkable for queer breeding habits, show in many cases primi- tive instincts for guarding or at least remaining close by the eggs during the early stages of their develop- ment, although most species, as in our ordinary frogs and toads, simply abandon their eggs as soon as they are deposited. The newt Desmognathus lays its eggs in a small hollow and remains near them for some time. The snake-like caecilians coil about their egg masses, and certain species of frogs carry eggs The Evolution of Parental Care 41 in pouches on the back or in sacs connected with the throat. One of the most peculiar cases is afforded by the obstetrical toad of Europe in which the male carries the string of eggs coiled about his hind legs until the young are ready to hatch. While the Amphibia in some cases exhibit a cer- tain care for their eggs, they all appear to be utterly indifferent to their young. One reason for this is the fact that the young usually live in the water while the parents are often terrestrial, but another reason is doubtless to be found in the low psychic development of these animals. Reptiles as a class concern themselves very little about their progeny. The latter are quite well able to take care of themselves upon their first introduc- tion into the world. Alligators are said to watch ^ over the places where their eggs are buried in the sand, and pythons coil around their eggs and help to ^ incubate them by the warmth afforded by their bodies. Solicitude for their eggs is, however, very rare among the reptiles, and active care for their young is practically absent. With the birds, which as a group are remarkable for their parental and fostering instincts, care for the eggs appears to be universal, and with rare ex- ceptions the birds sit upon the eggs until they are hatched. There has been much speculation concern- ing the origin of this curious instinct of incubation. As Whitman has remarked, "the incubation instinct was supposed to have arisen after the birds had ar- 42 Studies in Animal Behavior rived and laid their eggs, which would have been left to rot had not some birds just blundered into cud- dling over them and rescued the line from extinc- tion." A little reflection makes it evident that such an origin is clearly impossible, and that we must go back probably to the cold-blooded reptilian ances- tors of the birds for the beginnings of the instinct. A comparative survey of the behavior of the more primitive animals toward their eggs makes it prob- able that the instinct of incubation grew out of the instinct to remain in or near the place where the eggs are deposited for the purpose of protecting them. Lying near or brooding over the eggs may have afforded, even in the cold-blooded ancestors of the birds, sufficient heat to make the eggs develop with increased rapidity. The advantage thus accru- ing to the species may have caused the protecting instinct to develop further into a true instinct for incubation. With the development of warm blood- edness which went on during the evolution of the birds from the reptiles the supplying of artificial heat, at first only a means of hastening development, became an indispensable condition of development. Dependence upon artificial heat must have been evolved pan passu with the development of the in- stinct for incubation. Certainly the latter never could have been developed after the former had been established. . There is much evidence to show that the next step toward the development of parental affection The Evolution of Parental Care 43 was brought about by the extension of care for eggs to what comes out of the eggs. Little more than a feeble beginning of such an extension occurs among some of the arachnids to which we have already al- luded. In most insects, although provision for the eggs is common, the parents, with rare exceptions, are utterly indifferent to their young. Among the social insects, however, such as the ants and bees, the young grubs are tended with scrupulous care. In the solitary bees and in many of the more primitive social species the eggs are laid with sufficient pro- visions for the entire life of the larvae and then left without further attention. The transition from this condition to one in which food is brought to the cells after the larvae are hatched is an easy and natural development. And as we pass to the more highly developed social groups such as the hive bees the instincts for taking care of the larvae become more specialized and more complex. Among the various species of fishes that bestow some care upon their eggs there are some forms that pay no particular attention to their newly hatched young which scatter as soon as they emerge from the coverings of the eggs. In other cases such as the dog- fish Amia the protecting instinct is extended to the young brood. The male parent swims about with his school of small fry and keeps many enemies at a safe distance, for he is an alert and valiant defender of his own. In the course of a few weeks, however, the family ties are broken, the little fishes become 44 Studies in Animal Behavior dispersed far and wide, and parental solicitude, hav- ing subserved its purpose of affording protection when it was most needed, is manifested no more. When care becomes extended from the eggs to the young a course of development is begun in which relatively more and more care is bestowed upon the offspring as we pass to higher forms. Instead of a large number of progeny left to shift for themselves with a consequent great loss of life, we find in the higher animals a decrease in the number of offspring combined with an increase in the care and attention devoted to each. The young become at the same time less able to take care of themselves and are de- pendent upon their parents for longer periods. And along with these changes there is an increase of sym- pathy, affection and the various emotions that come into play in the family relation. This is well shown among the birds. The lower birds lay many eggs either in crude nests or none. The young birds which are quite active when hatched do not remain long under their parents' care, and many are lost. The higher song birds, on the other hand, lay few eggs in a well-prepared nest, and the young which remain in a weak and helpless state for a considerable time after hatching are fed by their parents, kept clean, and protected from various ene- mies. There are few more fascinating pictures of domestic life than those afforded by the little family group in many of the higher birds. Among the mammals we may trace a similar line The Evolution of Parental Care 45 of evolution. The suckling of young presupposes a certain tolerance, if not regard, on the part of the mother for her offspring. Had there not been among the ancestors of the mammals a fairly close associa- tion between the parents and their offspring during the infancy of the latter there obviously could not have been evolved the mammary glands and other adaptations for suckling the young which are among the most fundamental and distinctive features of mammalian structure. Instinct and organization are everywhere closely correlated and act and react upon one another during the course of evolution. The mammals afford an interesting instance of the way in which instincts of parental care have been instru- mental in developing certain fundamental peculiar- ities of bodily organization. Certain writers of the associationist school of psy- chology have endeavored to explain why animals regard their young with affection on the basis of the relief the mother experiences in having the milk re- moved from her mammary glands. Bain would have us believe that maternal love was compounded out of numerous agreeable sensations of touch experi- enced in handling the soft bodies of infants. Why the mothers do not develop an equal fondness for velvet-covered cushions the theory does not make clear. Such explanations appear eminently absurd in the light of a comparative study of the relations of parents to offspring in various groups of animals. Parental care must have antedated the giving of 46 Studies in Animal Behavior milk, and it is probable that without parental affec- tion there would be little more contact between parent and offspring than there is among fishes and amphibians. From the genetic standpoint both ex- planations put the cart before the horse. Among animals generally parental affection is rather strictly limited to the period of infancy, after which there is a disregard or indifference that con- trasts strangely with our own natural sentiments. The affections of animals, like most of their other characteristics, are quite closely subordinated to the needs of the species; when the young are able to make their way in the world alone the function of parental love is past, and the feeling rapidly becomes extinct. As we approach man we find a lengthening of the period of infancy and a prolongation of the time dur- ing which the parents bestow their care and affection upon their offspring. As John Fiske has shown, the lengthening of infancy affords opportunity for the young to acquire experience and perfect themselves in the varied activities which are demanded in the life of a highly evolved animal. Where, as in higher forms, success in life depends relatively more on intelligence than blind instinct it is important that there should be a period of education in which the young animal is more or less shielded from the hard- ships and dangers with which it will have to cope in later life. Our simian cousins remain with their young for a long period, and exhibit a degree of The Evolution of Parental Care 47 tenderness for them that is little short of that shown by human beings. In the Descent of Man, Dar- win relates that "Rengger observed an American monkey (a Cebus) carefully driving the flies which plagued her infant; and Duvancel saw a Hylobates washing the faces of her young ones in a stream. So intense is the grief of female monkeys for the loss of their young, that it invariably caused the death of certain kinds kept under confinement by Brehm in North Africa. Orphan monkeys were al- ways adopted and carefully guarded by other mon- keys, both male and female." It is a far cry from the egg-protecting instincts of primitive animals to the love of a human mother for her children. But the animal kingdom presents us with many of the intervening stages of evolution. The most primitive instinct and the most developed affection work toward the same end, — the perpetua- tion of the race. Both are parts of the life process which is everywhere occupied with the business of its own maintenance. The two phases of this process represented by self- preservation and race preservation have become elab- orated pan passu in the course of evolution. If our preceding account is correct, parental care and all the social and ethical instincts to which it forms the pre- paratory stage of development may be regarded as an outgrowth of the process of reproduction. To the simple acts of egg laying there came to be added other activities which make for the welfare of 48 Studies in Animal Behavior the progeny, leading on to active solicitude for the young, and thence to social instincts which finally blossom out into the rich endowment of altruistic emotions and sentiments of highly evolved social life. It is in reproduction, which is essentially an altruistic activity, that we must seek for the roots of altruism. Egoism and altruism in their primal manifestations are coeval rather than successive phenomena. The primitive organism which grows and divides by fis- sion shows us the germ of both of these traits. REFERENCES BAIN, A. The emotions and the will. 4th ed., 1899. DARWIN, C. The descent of man. N. Y., 1874. FABRE, J. H. Souvenirs entomologiques, T. 9. FISKE, J. The meaning of infancy. N. Y., 1909. HOLMES, S. J. Observations on the habits and natural history of Amphithoe longimana. Biol. Bull. 2, 165, 1900. MITCHELL, P. C. The childhood of animals. London, 1912. MORGAN, C. L. Animal behavior. London, 1900. PECKHAM, G. W. and E. G. On the instincts and habits of the solitary wasps. Bull. No. 2, Wis. Geol. Nat. Hist. Soc., 1898. PYCRAFT, W. P. Infancy of animals. N. Y., References 49 REIGHARD, J. The natural history of Amia calva Linnaeus. Mark Anniversary Vol. p. 57, 1903. ROMANES, G. J. Animal intelligence. N. Y., 1883. SUTHERLAND, A. Origin and development of the moral instinct. London, 1898. WHITMAN, C. O. Animal behavior. Biol. Lec- tures, Marine Biol. Lab. Woods Hole, 1898. Ill THE TROPISMS AND THEIR RELATION TO MORE COMPLEX MODES OF BEHAVIOR 'HPHE subject of animal behavior has been of in- -*- terest to human beings from the earliest times, but it has not been taken very seriously until a com- paratively recent date. The ways of animals were considered curious, interesting and in many ways use- ful things to know about, but the great theoretical import of animal psychology was unsuspected until it came to be recognized that our own minds are the outgrowth of the animal mind, and that to obtain a truly scientific human psychology it is necessa-ry to have a clear insight into the psychology of the lower animals from which we are descended. Near the middle of the nineteenth century Herbert Spencer enunciated the principle that, "If the doctrine of evo- lution be true the inevitable implication is that mind can be understood only by observing how mind is evolved," and he boldly plunged forward upon an undertaking to remodel the science of psychology from the genetic standpoint. The result was the publication in 1855, four years before the appear- ance of the Origin of Species, of the Principles of Psychology, a work which for sheer originality, inde- pendence of treatment and profound grasp of the so Tropisms — Relation to Modes of Behavior 51 subject stands almost without a rival in the history of the science. Notwithstanding the work of Spencer, genetic psychology was given perhaps its greatest impetus by Darwin, not only through his influence in estab- lishing the general doctrine of organic evolution, but also through his careful work on, and illuminating treatment of the mental life of animals. The admir- able and original chapter on Instinct in the Origin of Species, the chapters on the comparison of the mental powers of man and the lower animals in the Descent of Man, and the work on The Expression of the Emotions in Man and Animals, were all sub- stantial contributions to the science, which were very influential in stimulating further work. It is not my intention to treat of studies of animal behavior undertaken from the standpoint of evolu- tion, but to discuss another and in many respects com- plementary aspect of the subject, that of analysis, or the effort to discover the causal mechanism of animal activities by resolving them into their component fac- tors. The analytical study of behavior is simply a consequence of extending to animal psychology the methods of experimental investigation so largely em- ployed in the physical sciences and which are coming more and more to be employed in biology and in the laboratory investigations of the psychology of man. The results thus far won may be meagre, but judg- ing from the increasing number of trained investiga- tors who are devoting themselves to the work, we 52 Studies in Animal Behavior may look forward to a rapid increase in our knowl- edge and insight. From the standpoint of analysis the subject of tropisms is one of great import. Certain stimuli exercise a directive effect upon the movements of animals, causing them to go toward or away from the source of stimulation. The moth flies toward a candle; infusorians gather in regions of dilute acids and avoid regions of too great heat or cold; certain caterpillars tend to crawl opposite the di- rection of the force of gravity. These directed movements are commonly called tropisms but there is a variety of opinions regarding the kinds of be- havior to which the term tropism may be applied and usage has not settled authoritatively upon any rigid definition of the word. We shall therefore use the word in a somewhat broad and indefinite sense. Tropisms have long been recognized in plants. The familiar phenomenon of the turning of plants to the sun was termed heliotropism by De Can- dolle in 1835, and he, in common with several other botanists in the early and middle parts of the nine- teenth century, regarded this turning as a direct and more or less mechanical effect of sunlight upon the tissues of the plant. Sachs on the other hand emphasized the aspect of irritability in tropisms, and maintained that it is the direction in which the rays of light penetrate the tissues of the plant and not merely the different degrees of illumination on Tropisms — Relation to Modes of Behavior 53 the two sides, that determines the direction of turn- ing. The work of Sachs on plants directed the at- tention of Loeb to the phenomena of tropisms in animals, and also furnished him with some of the fundamental conceptions of his own celebrated the- ory. Loeb's first paper on tropisms of any con- siderable length was entitled The Heliotropism of Animals and its Agreement with the Heliotropism of Plants. The publication of this paper marked an epoch in the analytical study of animal behavior. Previous to this time the tropic responses of ani- mals were interpreted as the expression of the pre- dilections or conscious choice of animals for certain kinds of stimulation. Graber, Sir John Lubbock and Paul Bert had studied the effect of colored lights upon animals and discovered that certain species congregated most abundantly under light of a cer- tain color, while other species would gather in greater numbers under light of a different color. These aggregations were therefore considered as an index of the kind of color most pleasing to the aesthetic or other sensibilities of the animals. Move- ments toward or away from lights were accounted for in a similar way. Earwigs and cockroaches were supposed to crawl away in secluded places be- cause they like the dark and butterflies were sup- posed to congregate in sunny spots because they en- joy the sensation afforded by the sunshine. A some- what more anthropomorphic interpretation of a tro- pism was suggested by Romanes in discussing why 54 Studies in Animal Behavior the moth flies into the flame of a candle. The con- clusion arrived at was that the moth was drawn to the fatal flame out of curiosity, or the desire of in- vestigating what manner of strange object a candle flame might be. The theory developed in Loeb's Heliotropism stands in a sharp contrast to the anthropomorphic views of his predecessors. Orientation of animals to light is supposed to take place in a more or less mechanical fashion like the orientation of plants. "These tropisms," he says, "are identical for ani- mals and plants. The explanation of them depends first upon the specific irritability of certain elements of the body surface, and second, upon the relations of symmetry of the body. Symmetrical elements at the surface of the body have the same irrita- bility; unsymmetrical elements have a different ir- ritability. Those nearer the oral pole possess an irritability greater than that of those near the aboral pole. These circumstances force an animal to orient itself toward a source of stimulation in such a way that symmetrical points on the surface of the body are stimulated equally. In this way the animals are led without will of their own either toward the source of the stimulus or away from it." The moth flies into the flame, not out of curiosity or any other conscious motive, but simply because it cannot help it. In a very instructive series of experiments Loeb showed that heliotropism in animals obeys the same Tropisms — 'Relation to Modes of Behavior 55 laws as heliotropism in plants. In both plants and animals it is the direction of the rays that controls the direction of movement. In both plants and ani- mals it is the rays nearer the violet spectrum that are the more potent in evoking the heliotropic re- sponse. In both plants and animals temperature, previous exposure to light and other external fac- tors influence reactions to light in much the same way. Back of all the differences of form and func- tion of plants and animals, and notwithstanding the higher organization of the animal world with its specialized sense organs and complex nervous sys- tems, the living substance of organisms possesses cer- tain fundamental common properties of irritability upon which the common and fundamental features of behavior which we call tropisms depend. The theory of Loeb would sweep away all higher psychic factors in the realm of tropisms, and re- duce the phenomena to comparatively simple mani- festations of reflex irritability. Further he would explain much of the so-called instincts of animals as a result of these tropisms. Since the prospect of finding a mechanical or causal explanation of any feature of behavior is always an alluring one, it will be of interest to pass in review some of these cases of tropisms with the end of determining how far the reflex theory will carry us. And then we shall consider the relation of these tropisms to more com- plex forms of behavior. An excellent illustration of a tropism is afforded by 56 Studies in Animal Behavior the. light reactions of the larvae of the marine worm Arenicola. These larvae are oblong in shape, with two eye spots at the anterior end. Near either ex- tremity there is a band of cilia by means of which the larvae swim through the water. The larvae are positive in their reactions to light, and will follow a light around in various directions. Orientation to light is brought about by bending the body to the stimulated side. If the larva is between two sources of light from which the rays intercept one another as they fall on the animal at an angle of ninety de- grees, the larva will take a course midway between the two lights. If one light is turned off the larva bends immediately to the other one. By arranging a mirror so as to throw a small spot of light on dif- ferent parts of the body, Mast has shown that when light is thrown into one eye there is a strong bend of the body toward the stimulated side. The parts behind the eye spot show no definite reac- tion. It is evident that orientation in this form is due to different intensities of illumination on two sides of the body. So far as can be ascertained ori- entation takes place directly and automatically, with- out any conscious decision on the part of the animal. The movements of the larva appear little more voluntary than the precise movements of certain Pro- tozoa or the swarm spores of algse. Let us pass to animals somewhat higher in the scale of life. Some years ago when on the Atlantic coast at Woods Hole, Mass., I studied the behavior of vari- Tropisms — Relation to Modes of Behavior 57 ous amphipod Crustacea of that region and particu- larly the reactions of the terrestrial species com- monly called sand fleas. It is a somewhat curious circumstance that the aquatic amphipods are nega- tive to light and tend to keep in the darkest part of their environment while the terrestrial ones are usu- ally positive. Positive phototaxis is the most pro- nounced in the most terrestrial of the species, the large Talorchestia longicornis, which lives in holes in the sand high up on the beach. When dug out of the sand these crustaceans usually lie curled up in a death feint, but when they become active they manifest a very strong tendency to hop toward the light. When brought into a room they may keep hopping toward a window with intervals of rest during the entire day. If they are placed in a dish one-half of which is shaded while the other half is exposed to the direct sunlight they will keep hopping toward the light until they are overcome by the heat of the sun's rays. The smaller Orchestia agilis, which lives nearer the water's edge and frequently manifests a nega- tive reaction to light, shows the same fatal degree of positive phototaxis when exposed for some time to strong sunlight. Does light orient these forms automatically and involuntarily as is apparently the case with the larvae of Arenicola? There are sev- eral facts which favor such an interpretation. The persistent and apparently unreasonable nature of the response, its sudden reversal by certain external 58 Studies in Animal Behavior agents, and especially the fact that the witless crea- tures continue to go toward the light even when they are brought thereby into a region where the heat proves fatal to them, seem to bear out the conclusion that the phototaxis of these animals is in the nature of an involuntary or "forced" response. This view is strengthened by the results of certain experiments on individuals which were blinded on one side. These experiments were undertaken with the view of ascertaining something of the mechanism of orientation. The amphipods do not become ori- ented by bending the body toward the light, but by the unequal activity of the appendages on the two sides of the body. In forms with positive photo- taxis it was found that blackening over one eye caused the amphipod to perform circus movements toward the normal side. In negatively phototactic species it was found that the same treatment caused circus movements in the reverse direction. It is probable therefore that impulses received by the eyes cross in the central nervous system and become carried to the appendages on the opposite side of the body, causing them to act with greater vigor, thus bringing the animal into a position of orienta- tion. This supposition led to the experiment of cutting the brain lengthwise through the center in several species of arthropods and it was found that, although sensitiveness to light could be shown to remain, all power of orientation to light was en- tirely destroyed. Tropisms — Relation to Modes of Behavior 59 It will be of interest in this connection to consider the light reactions of a somewhat more highly or- ganized arthropod, the water scorpion Ranatra. This insect lives near the banks of ponds and streams with the tip of its long breathing tube at the sur- face of the water and its raptorial fore legs held in a position for rapidly seizing any small passing creature which may be utilized for food. When Ranatra is taken out of the water it generally feigns death, assuming a perfectly rigid attitude which it retains through all sorts of maltreatment, even suf- fering its legs to be cut off or its body to be cut in pieces without betraying any signs of animation. By moving a light over the motionless insect it may gradually be brought out of its feint. The first no- ticeable signs of awakening are very slight move- ments of the head in response to the movements of the light. When the light is passed to one side of the body the head is rolled over ever so little toward that side. Move the light to the other side and the head tilts over slightly in that direction. Place the light in front of the body and the head bows down in front. Now carry the light behind the in- sect and the front of the head points slightly up- ward. These movements occur with perfect regu- larity in response to the movements of the light, and gradually increase in vigor and extent. After following the movements of the light with these definite movements of the head the insect slowly and awkwardly raises itself up and begins to follow the 60 Studies in Animal Behavior light with equally definite swaying movements of the body. If the light is to one side the legs on that side are flexed and the opppsite legs extended. Pass- ing the light over the body causes the reverse at- titude. Hold the light in front of the body and the insect bows down in front in an attitude of ab- ject submission. Carry the light behind the insect and it elevates the anterior part of its body and holds its head high in the air. These bodily atti- tudes are assumed with almost machine-like regular- ity. For each position of the light there is a cor- responding position of the head and body. After a little Ranatra will follow the movements of the light by walking slowly and awkwardly toward it, gradually increasing the vigor and rapidity of its response until it will rush toward the light with frenzied haste. It becomes oblivious to all else but the light, which seems to dominate its behavior en- tirely. If the source of light gives off a good deal of heat the insect will continue to go toward it until overcome by the heat. I have seen Ranatras when nearly killed by the heat of the lamp toward which they were attracted, slowly drawing themselves with the last remnants of their strength a little nearer to the fatal source of light. Nothing could seem more mechanical or more ob- viously the result of domination by outer agencies than the phototaxis of this form. There are, how- ever, some curious features of the behavior of Ranatra which are disclosed by other experiments Tropisms — Relation to Modes of Behavior 61 and which indicate that this insect is something more than a mere "reflex machine." If one eye is black- ened over there is a strong tendency for the insect to perform circus movements toward the normal side. Frequently as the insect veers over toward the normal side in going toward the light and thus brings the unblackened eye more and more out of the region of direct stimulation, a point is reached where there is hesitation, moving this way and that, accompanied by increasing uneasiness and excite- ment as if the creature were exasperated over its predicament. Sometimes the insect may get out of this situation by going completely around in a circle toward the normal side, or it may make a direct turn toward the blackened side and go toward the light. In several cases among Ra- natras and Notonectas individuals which at first performed circus movements and succeeded in going to the light by a very irregular route finally came, after a number of trials, to go to the light in a nearly straight line. Other individuals went to the light in a nearly straight line from the first. In some cases covering all of one eye and all but the posterior face of the other did not prevent the insect from going in a nearly straight path toward the light. Other specimens would do the same with only a small part of the lateral face of one eye uncovered. In the latter case neither the lightest nor the darkest part of the visual field was kept be- fore the eye. The insect behaved as if it were 62 Studies in Animal Behavior not guided by a mere reflex response, but had an awareness of the general space relations of its re- gion, the relative position of the light and itself and of the movements necessary to bring itself toward the light. If such a general typographical sense seems too high a psychical endowment to be cred- ited to so simple a creature, it must be remembered that other insects, notably the bees and wasps, have a much more definite and detailed cognizance of the topographical relations of their environment than anything in the behavior of Ranatra would call for. Simple and mechanical as much in the light reactions of Ranatra seems to be, there are many features of its phototaxis which are very difficult to explain on the basis of simple reflex orientation. We might expect a priori to find that somewhere in the course of evolution the tropisms become more or less subordinated to higher forms of behavior. It is quite evident that much in the behavior of animals may be explained as a more or less simple manifestation of phototaxis, geotaxis, chemotaxis, and so on. The daily depth migrations of pelagic animals is traceable, to a considerable degree, to variations in the sense of the response to light and gravity. One circumstance that leads copepods to swim to the surface at night and go down in the daytime is because they are positive to weak light and negative to strong light. The negative reac- tion of centipedes, termites and many other organ- isms keeps them in dark and secluded situations. Tropisms — Relation to Modes of Behavior 63 The positive reactions of many worms and crus- taceans to contact stimuli keep them in protected situations in various nooks and niches where they escape many of their enemies. The positive chemo- taxis of many animals leads them into situations where they may find their food. But one of the chief considerations which makes the study of tropisms of such importance is the fact that the tropisms enter as components into more complex activities of higher organisms. Tropisms in their purity are met with only in the simpler ani- mals. As we pass up the scale of life these primary tendencies to action become broken up and com- bined with other forms of behavior, so that they are lost sight of in the more complex activities into which they enter as component factors. A most interesting field of investigation in this con- nection is presented in the relation of phototaxis and vision. It is a field scarcely touched upon as only one investigator, Radl, has entered upon it with any seriousness. There seems to be a close connection in many animals, and especially in insects, between phototaxis and what are called compensa- tory movements. Place a lady-beetle on a turntable which is slowly rotated. The beetle begins to move its head and then its body opposite the direction of movement. Robber flies show the reaction espe- cially well. The reaction depends upon the eyes because it no longer occurs when the eyes are black- ened over. It does not depend upon the rotation 64 Studies in Animal Behavior of the insect's body. If the insect is placed in a cylinder on a stationary center and the cylinder ro- tated, the insect tends to walk around in the direc- tion of rotation. A frog under the same circum- stances will do the same thing. In these cases the animal reacts so as to keep, so far as possible, in statu quo with the visual field. A beautiful illustration of this is afforded by the so-called rheotropism of fishes. Many fishes have the instinct to head up stream against the current. This trait has been shown by Lyon to be dependent upon a visual reflex. He placed fish in an aquarium with the lower side made of glass below which could be drawn a long piece of cloth with alternate black and white stripes on it, giving the appearance of a moving bottom. As the strip was pulled along the fish swam in the direction of movement. Reversing the motion caused the fish to turn about and swim to 'the other end of the aquarium. In another ex- periment fishes were placed in a long bottle. When this floated down stream the fishes all swam to the up stream end. When it was pulled up stream the fishes all swam to the opposite end. Fishes in a stream, passively carried along, have no means of becoming aware of their movements except by means of objects in their field of vision any more than a man in a balloon who is carried along in a current of air. This automatic tendency to keep in constant relations with the objects in their field of vision keeps them from being passively carried down Tropisms — Relation to Modes of Behavior 65 stream. Many insects show the same trait in their flying against a breeze. Perhaps the instinct of hovering shown by many kinds of flies is an expres- sion of the same fundamental tendency. The automatic tendency to keep the body in a certain orientation to its field of vision which we find among crustaceans, insects and lower verte- brates, is to a greater or less extent replaced in forms with freely movable eyes by ocular movements which enable the moving animal to retain the same field of vision. Stalk-eyed crustaceans show com- pensatory movements of the eye stalks. Similar eye movements occur in fishes, amphibians and birds. A man at night more or less involuntarily directs his steps toward a single light in his horizon much as birds are drawn toward a lighthouse. Such ori- entation may be conscious and voluntary, but it can- not be denied that there is a sort of instinctive ten- dency toward it much as there is in all of us a strong tendency toward a certain orientation to the force of gravity. The reactions of animals to light have been pro- foundly modified by the evolution of the image- forming eye. It has been shown by Cole that if an eyeless form such as an earthworm or a form with simple eyes is subjected to stimulation by two sources of light of equal intensity but of different area, the animal is as likely to turn to the smaller light as to the larger one. In forms with image- forming eyes, on the other hand, it is the light 66 Studies in Animal Behavior larger area which is the more potent in causing the turning of the body. With the development of im- age-forming eyes it becomes possible for animals to respond to objects and not to mere differences in the amount of light and shade. The image-form- ing type of eye is stimulated by a decrease as well as by an increase of illumination on particular parts of its surface. This stimulation is generally as- sociated with an involuntary turning toward the source of stimulation. Hence, the automatic turn- ing of the head toward a new object in the field of vision, and the tendency to follow the movements of bodies with movements of the eyes. The eyes of animals are notoriously quick to respond to movements. The moving thing is the stimulating thing. With the evolution of the image-forming type of eye and the development of acute sensitive- ness to change of illumination of particular parts of the retinal surface, the general tendency to go toward or away from light may pass into an in- voluntary tendency to become oriented toward par- ticular moving objects in the field of vision. When an animal reacts in a definite way to objects im- pressed on its retina we commonly say that it sees. These reactions to objects come to be very com- plex and specialized. They come to depend upon the size, form and color of the moving object. But it is not improbable that they have their primary roots in the positive and negative phototaxis of simpler organisms. Josiah Royce in his Outlines of Tropisms — Relation to Modes of Behavior 67 Psychology has gone much farther than I should venture to do, in that he sees in the tropisms a set of tendencies which form a sort of fundamental background even in our own psychology. Objects of our own attention exercise a compelling force over us making us turn toward them. We involun- tarily turn toward a person or thing about which we are curious; in fact it requires some voluntary effort not to do so. Is this continual orientation to objects akin to orientation to light, odors, etc., in the lower animals? According to Royce these re- actions are fundamentally the same. Perhaps if we should follow the history of behavior closely enough in passing from lower to higher forms we should be able to fill in the intermediate steps. At present the connection is merely a suggestive hy- pothesis. Most of the work on tropisms that has been done thus far has consisted in determining the precise way in which tropisms are brought about, and the conditions by which they are modified. To find, as it were, what becomes of the tropisms in the course of mental evolution, how they are converted into higher forms of behavior, is a more difficult task. Voltaire has made the remark that we are governed by instinct as well as cats and goats. It is possible that we may be justified in going somewhat farther than the celebrated skeptic, in saying that to a cer- tain extent we are governed by tropisms as well as insects and worms. 68 Studies in Animal Behavior REFERENCES BOHN, G. ( i ) La naissance de Fintelligence, Paris, 1909; (2) La nouvelle psychologic animale, Paris, 1911. COLE, L. J. An experimental study of the im- age-forming power of various types of eyes. Proc. Am. Acad. Arts and Sciences, 42, 335, 1907. GEE, W. The behavior of leeches with special reference to its modifiability. Univ. of Calif. Pubs. Zool., ii, 197, 1913. HOLMES, S. J. (i) Phototaxis in the amphi- poda. Am. Jour. PhysioL, 5, 211, 1901; (2) The reactions of Ranatra to light. Jour. Comp. Neur. Psych., 15, 305, 1905. LOEB, J. ( i ) Der Heliotropismus der Tiere und seine Uebereinstimmung mit dem Heliotropis- mus der Pflanzen. Wiirzburg, 1890; (2) The mechanistic conception of life. Chicago, 1912. LYON, E. P. On Rheotropism. I, Rheotropism in fishes. Am. Jour. Physiol., 12, 149, 1904. MAST, S. O. Light and the behavior of organ- isms. N. Y., 1911. RADL, E. Untersuchungen uber den Phototropis- mus der Tiere. Leipzig, 1903. ROYCE, J. Outlines of Psychology. N. Y., 1903. WASHBURN, M. F. The animal mind. N.Y., 1909. WODSEDALEK, J. E. Formation of associations in the May-fly nymphs Heptagenia interpunctata (Say). Jour. Animal Behavior, 2, i, 1912. IV THE PROBLEM OF ORIENTATION TN treating of this subject one is compelled to en- •*-ter a field occupied by contesting parties. The subject of orientation has long been a bone of con- tention, but it is probable that, as has occurred in so many other controversies, a deeper insight will result from the conflict of opinion. Allusion has been made in the previous chapter to the theory of orientation to light and other forces, which has been developed by Loeb and which has influenced so largely recent work in the behavior of lower or- ganisms. This theory made the orientation of or- ganisms to light, heat, chemicals, etc., and the migra- tion of animals toward or away from the source of stimulus an entirely involuntary response based upon more or less simple reflex action. To construe be- havior in terms of tropisms and to explain tro- pisms as the inevitable effect of reflex action which in turn was supposed to be determined by the me- chanical and chemical constitution of the organism formed the general scheme of attack upon problems of behavior which Loeb and his adherents have at- tempted to follow out. Many features of the be- havior of lower organisms were found to receive a 69 70 Studies in Animal Behavior very plausible interpretation in accordance with this scheme. The studies of Jennings on the behavior of the simpler animals revealed many cases of apparent tropisms to which the tropism scheme of Loeb did not apply. Infusorians were found to collect in regions of dilute acid, not because they were ori- ented to the lines of the diffusing substance, but be- cause when they accidentally swam into the region of the dilute acid they remained there. The col- lections thus formed were the result of a sort of selection of chance movements. The organisms were found not to be able to orient themselves at all, although the aggregations formed resembled those which in other organisms are the result of orientation. Although the formation of collections by the in- fusoria and other asymmetrical organisms may occur without orientation, Jennings believes that in sym- metrical forms in which the body of the organism is obviously oriented to the stimulus the process does not occur in accordance with the tropism scheme of Loeb. "In the symmetrical Metazoa," he says, ".we of course find many cases in which the animal turns directly toward or away from a source of stimulation, without anything in the nature of preliminary trial movements. This is a simple fact of observation, which leaves open the possibility of many different explanations. Is the simple expla- nation given by the local action theory of tropisms The Problem of Orientation 71 one that is of general applicability to the directed reactions of lower and higher Metazoa?" "In considering the evidence on this question, we find that even in the symmetrical Metazoa the di- rection of movement with reference to external agents is by no means always brought about by a simple, direct turning. On the contrary, in many of the Metazoa, trial movements are as noticeable and important as in the Protozoa. This we have illustrated in detail for many invertebrates in the section devoted to this subject (Chapter XII, Sec- tion 2). For such behavior the local action theory of tropisms fails to give determining facts." Just how Jennings would explain orientation such as occurs when the moth flies into the candle he has nowhere made clear, further than saying that in such cases the organism "reacts as a whole," in- stead of being oriented by the local reactions of par- ticular organs. That the orientation of a symmetrical organism may be largely brought about by the indirect method of a sort of trial and error process has been shown, in the case of several organisms, by the present writer. In a paper on the "Selection of Random Movements as a Factor in Phototaxis" I have shown that in earthworms, leeches (Glossiphonia) , and the larvae of blow flies, there is a fairly definite orien- tation of the body to light rays as the animals move away from the source of illumination, and that it is not so much through a direct turning away from 72 Studies in Animal Behavior the light that orientation is effected as by the cir- cumstance that all movements, except those bring- ing the animal away from the light, are inhibited or reversed. If an earthworm is placed at right angles to a beam of light "it will be seen that the head frequently moves from side to side before ex- tension takes place. These movements may be very slight and ordinarily would escape attention. There is often a similar movement during the process of extension. Frequently the head is bent over towards the light during the first part of the extension and bent the other way and extended farther, or again it may be waved back and forth several times. Slight trial movements in all directions are continually being made. The reason why the worm makes more turns of a decided sort away from the light than towards it is largely because the little trials that bring the worm nearer the light are not followed up. Many of the turnings that would naturally be counted as negative are preceded by a slight posi- tive turn followed by a stronger negative one. In order to ascertain whether the negative reaction was manifested at the very beginning of the response, the following experiment was tried: A worm was allowed to crawl on a wet board. When it was crawling in a straight line it was quickly lowered into the beam from a projection lantern so that its body would lie at right angles to the rays. The ex- posure to the light was made in each case when the worm was contracted, and the first detectible The Problem of Orientation 73 movement of the head to one side noted. In the two specimens employed the first detectible turn was away from the light 27 times and towards the light 23 times. After a few extensions the worm in nearly all cases soon turned and crawled away from the light. The first detectible movement of the earthworm seems, therefore, to be nearly as likely to be towards the light as away from it. The slight preponderance of negative turns may be due to the fact that some of the smaller trial movements were overlooked, to a slight direct orienting effect of the rays, or merely to chance." In the leech Glossophonia, which crawls by a sort of looping motion, the anterior part of the body is frequently raised, extended, and moved about as if the animal were feeling its way. "If the animal turns it in the direction of a strong light it is quickly withdrawn and extended again, usually in another direction. If the light is less strong it waves its head back and forth several times and sets it down away from the light; then the caudal end is brought forward, the anterior end extended and swayed about and set down still farther from the light than before. When the leech becomes negatively ori- ented it may crawl away from the light, like the earthworm, in a nearly straight line. The extension, withdrawal and swaying about of the anterior end of the body enable the animal to locate the direc- tion of least stimulation, and when* that is found it begins its regular movements of locomotion. Of 74 Studies in Animal Behavior a number of random movements in all directions only those are followed up which bring the animal out of the undesirable situation." With the blow fly larvae the method of orienta- tion was much the same. "When a strong light is thrown on a fly larva from in front, the anterior end of the creature is drawn back, turned to one side and extended again. Often the head is moved back and forth several times before it is set down. Then it may set the head down when it is turned away from the light and pull the body around. If the head in moving to and fro comes into strong light it is often retracted and then extended again in some other direction, or it may be swung back without being withdrawn. If a strong light is thrown on a larva from one side it may swing the head either towards or away from the light. If the head is swung towards the light, it may be with- drawn or flexed in the opposite direction, or, more rarely, moved towards the light still more. If it is turned away from the light the larva usually follows up the movement by locomotion. Frequently the larva deviates considerably from a straight path, but as it continually throws the anterior end of the body about and most frequently follows up the move- ment which brings it away from the stimulus, its general direction of locomotion is away from the light. In very strong illumination the extension of the anterior part of the body away from the light is followed by a retraction, since in whatever direc- The Problem of Orientation 75 tion it may extend it receives a strong stimulus and the larva writhes about helplessly for some time. Sooner or later, however, it follows up the right movement. Occasionally the larva may crawl for some distance directly towards the light, but after a time its movements carry it in the opposite direc- tion. When once oriented, the direction of locomo- tion of the larvae is comparatively straight." It was not denied that these different forms showed a certain tendency to turn directly away from the light, but it was contended that orientation was mainly produced "indirectly by following up those chance movements which bring respite from the stim- ulus." The light reactions of the forms studied "may be interpreted as a resultant of two motor re- sponses; first, the activities of locomotion which are set up by the stimulus of the light, and second, the act of jerking back and bending the body from side to side in response to a strong stimulus from in front. Here are two instincts or reflexes, how- ever we may be pleased to call them, which are in a measure antagonistic in that the first is frequently overcome by the second. The direction of the ex- ternal stimulus determines which of these two in- stinctive tendencies predominates." The more recent work of Harper on the earth- worm Perichata bermudensis has shown that in strong light the worm turns away from the stimu- lus, but that in weaker light "orientation is the re- sult of a trial and error method." And Mast on 76 Studies in Animal Behavior the basis of his studies on the light reactions of an- other earthworm, Allolobophora, concludes that "se- lection of random movements, as Holmes, Harper and Jennings pointed out, undoubtedly plays a very large part in the orientation of the earthworm un- der ordinary conditions." The results obtained by Mast in his careful studies of the reactions of fly larvae confirm my own position as to the selection of random movements in these forms and add a number of interesting facts, one of which is that under certain conditions orientation may be to a considerable extent direct. Larvae at first exposed to strong light turn about as often towards as away from it and orientation is effected mainly by the method above described, but after longer exposure the percentage of direct turns from the light in- creases. This is accounted for by Mast, as follows : "If the larvae are carefully observed when they are suddenly exposed to lateral illumination by diffuse light, it is found that they respond immediately only if the anterior end is turned toward the source of light when the exposure is made. If this end is in any other position, there is no re- action whatever until the organism, in its normal process of locomotion, extends it toward the source of light. Then it is at once turned from the light to such an extent that it frequently makes a right angle with the posterior end. Later it is swung back, but only part way. The tip is, however, exposed and so the animal may be stim- The Problem of Orientation 77 ulated again, after which it again turns sharply from the source of light. This process is repeated until the organism has turned to such an extent that the anterior end is practically as much exposed when it turns in one direction as when it turns in the other. The great preponderance of lateral movements from the source of light and direct orientation in diffuse light therefore do not indicate that fly larvae have the power of differential response to localized stimulation." Loeb, who sets little store by such proximate categories of explanation as trial and error, has con- tended against the doctrine that orientation is ef- fected by this method. In speaking of the doctrine that orientation of an organism in strong light may be direct, while in weak light it may take place by the indirect method described above, he says: "If the photosensitiveness of the animal is lessened the animal may deviate for a longer period from the direction of the light rays. Such animals do event- ually reach the lighted side of the vessel, but they no longer go straight toward it, moving instead in zig-zag lines or very irregularly. It is therefore not a case of qualitative, but of a quantitative, difference in the behavior of heliotropic animals under greater or lesser illumination, and it is therefore erroneous to assert that heliotropism determines the movements of animals toward the source of light only under strong illumination, but that under weaker illumina- tion an essentially different condition exists." 7 8 Studies in Animal Behavior Now there are undoubtedly forms to which Loeb's remarks are entirely applicable. The possibility of the interpretation which Loeb has given was care- fully considered by the present writer while work- ing on the forms above described and was finally rejected as inadequate for the particular cases un- der consideration. It was pointed out that ua trop- ism of the direct sort is not necessarily a perfectly fixed and rigid affair. It may be a tendency more or less obscured by a lot of random movements arising from internal causes. An organism may be drawn to a certain point through a direct orient- ing reflex, but if there is at the same time a large element of random activity in its behavior it may seem to reach that point by the method of trial and error." A careful consideration of the movements of earthworms, leeches and fly larvae made it evident that orientation does not take place by the method just referred to. We are not dealing merely with the masking of a simple tropism by a lot of inconse- quential activity. It is largely by virtue of and not in spite of random movements that orientation is secured. If an organism were so constituted that whenever it extended toward the light it would jerk back and turn to one side, even though the di- rection of its turning were totally indeterminate, it would finally become oriented away from the light. To a considerable extent the reactions of the forms The Problem of Orientation 79 described are like those of such a hypothetical or- ganism. Whatever be the part played by the indirect meth- ods of orientation which have just been discussed, it is evident that very many animals orient them- selves by turning directly toward or away from the source of the stimulus. In fact this might be said to be the typical method for symmetrical organ- isms, and it has been shown to apply to several that are asymmetrical in structure also. Many forms will accurately follow a source of light, changing their direction promptly and without any preliminary trial movements whenever the position of the light is changed. The existence of such behavior is not of course proof of any particular theory of the mechanism of orientation. It is as consistent with the view that light is followed through deliberate vo- lition, or through curiosity as Romanes once sug- gested, as it is with the reflex theory. The latter interpretation has the principle of Morgan in its favor which is to the effect that "In no case may we interpret an action as the outcome of the exer- cise of a higher psychical faculty if it can be inter- preted as the outcome of one which stands lower in the psychological scale." But the law of par- simony in any of its forms, — and the principle of rf Morgan is one of them, — never affords proof; it only creates a certain presumption, and one, too, whose strength varies greatly according to circum- stances, in favor of any given conclusion. 8o Studies in Animal Behavior Several years ago it seemed to me very desir- able to study tropisms with the view of testing the possible presence of any more highly developed functions than are postulated by the reflex theory. For some reason this particular quest, which involves one of the most important theoretical considerations connected with the tropism theory, had been almost completely neglected. I have already alluded to some of the initial experiments which consisted in blacking over one of the eyes in several phototac- tic species of arthropods and then observing their reactions to light under various conditions of ex- posure. It was found that in several positively pho- totactic species blinding one eye tended to make the animal perform circus movements toward the nor- mal side. In several negative species similarly blinded, circus movements were performed in the reverse direction. This is what the reflex theory of tropisms would lead us to expect. That the re- lations discovered are fairly general is evinced by the discovery by other investigators as well as by myself of a number of other forms which react in a similar manner. Great individual variation was found in the degree to which the tendency to per- form circus movements was manifested, some forms after being blinded on one side continuing to go toward or away from the light in a nearly straight line. The fact observed by myself in Ranatra and Notonecta and subsequently found by Miss Brun- din in Orchestia, that individuals which at first per- The Problem of Orientation 81 formed circus movements later come to go toward the light in a nearly straight course, indicates a ca- pacity for regulation in behavior if not a primitive form of learning, and suggests the possibility that the normal tropisms of these forms are not wholly a matter of direct reflex action. Were these animals pure reflex mechanisms we should expect their cir- cus movements to continue. But judging from their behavior, these animals follow light more or less after the manner in which higher animals pursue their prey or any other object of interest. Possibly the same may be true of the fiddler crabs which move sidewise toward the light instead of orienting them- selves in the usual manner with their longitudinal axis parallel to the rays. Loeb is inclined to the conclusion that "in the fiddler crabs in the first place there is an entirely different connection between the retina and the locomotor muscles from that in other crustaceans, and that, secondly, there is a special peculiarity in regard to the function of the two re- tinas whereby they do not act like symmetrical sur- face elements." In the light of the facts mentioned above it is, I believe, doubtful if we need to assume any far-reaching modifications of nervous structure to account for the peculiar orientation of these crabs. This conclusion is supported, I think, by the experiments of Miss Brundin on Orchestia traskiana in which the amphipods which have a narrow com- pressed body were compelled to travel on one side by being confined between two horizontal glass 82 Studies in Animal Behavior plates. Although orientation in the usual way was impossible, the amphipods moved toward the light by pushing against the glass with their legs and by flexing and extending the body. The long axis was kept pointed approximately toward the light, but this position was maintained by movements quite different from those involved in orientation under normal conditions. Since the phototaxis of this species is readily modified by experience it is not im- probable that something more than direct reflex ac- tion is involved. Perhaps the explanation of why animals do the things that they like to do is involved in any complete account of the orientation of higher forms. For more primitive types in which behavior is lit- tle if at all modified through the agency of asso- ciative memory there is no reason for doubting that orientation is effected by means of reflex action. But granting this to be true there is room for a variety of ways in which orientation may be brought about. Many animals respond very readily to a sudden change in the intensity of light, while they show little or no response when exposed to constant illumi- nation. In some cases a sudden increase of light intensity produces a reaction, but more frequently the response is evoked when the intensity of the light is diminished. Hungry leeches (Glossiphonia) raise up and extend the anterior end of the body when a shadow passes over them, and mosquito larvae wriggle downward under the same circum- The Problem of Orientation 83 stances. Now organisms of any kind in going to or from the light are subjected to frequent changes of light intensity as they deviate to one side or the other of their course. Do the fluctuations of light intensity so caused produce stimulations that affect the orientation of the animal? The possibility that stimulations produced by such fluctuations in the in- tensity of light might play a role in orientation was suggested in my paper on the phototaxis of Rana- tra. As this insect, when one eye is blackened over, nevertheless, in some cases goes nearly straight to the light, it was pointed out that "were the insect so constituted as to respond to an increase of light entering the left eye by a turn to the left and to a decrease of light by a turn to the right, we can understand how, when once pointed to the light, a straight course might be preserved. If the insect turned towards the right there would be an in- crease of light entering the left eye which we might suppose stimulates the insect to turn in the opposite direction. Deviations to the left would cause a diminution of light entering the left eye, which we might suppose acts as a stimulus to turn to the right side. The right eye may be supposed to act, mu- tatis mutandis, in a similar manner." The conclusion reached, however, was that re- sponses to fluctuations of light intensity alone did not give an adequate explanation of the orientation of this form, although they might afford a co- operative factor. Such responses probably do play 84 Studies in Animal Behavior an important role in producing those orienting move- ments known as compensatory motions where the latter are, as in many insects, dependent upon the organs of vision. Mast, who is an opponent of the view that phototaxis is the result of light acting as a constant stimulus, is inclined to attribute orienta- tion very largely to intensity changes. In speaking of orientation in many lower forms, he says: "In many of these forms orientation is undoubtedly, and in all it is probably, a response to change of light intensity on some part of the organism. At any rate it has in no instance been demonstrated that it is, as Loeb states, 'a function of constant intensity,' that orientation to light is like orientation to an elec- tric current." There is much, I believe, that indi- cates that light stimulates quite apart from the shocks due to variations in its intensity, but the question as to the relative potency of the two factors involved, which Mast has done well to bring into greater prominence, is one that can be answered only by experiment. In the ordinary movements of animals to and from the light both these factors are free to come into play. The natural method of attacking the problem, therefore, is to exclude one of the possible agencies, and then to observe the effect of the other alone. Recently a series of experiments was carried on by Miss McGraw and myself, in which animals were exposed to illumination which was rendered constant so far as this could be done in the case of an ac- The Problem of Orientation 85 tively moving animal. In one set of experiments the insect was placed in a jar lined below and at the sides with white paper. "This was covered by a cone of the same material in the apex of which was placed an electric light. A small peep hole permitted the observation of the insects in the jar. In several experiments the insect was placed in a small circular glass dish in the center of the en- closure. Whether the insect turned to the right or to the left in this apparatus the amount of light entering the eyes was approximately the same. In- sects with one eye blackened over were placed in the jar and stimulated to activity whenever they came to rest by tapping on the jar, or when this failed by poking them with a wire. The very slight variations in the light entering the eye in the dif- ferent positions of the insect would have different effects, according to the theory of differential sen- sibility, depending on the position of the insect, and would not tend to produce a constant deviation of the path in any particular direction. The same may be said of variations caused by movements out of the horizontal plane. Since the slight effects of dif- ferential sensibility would tend to neutralize one an- other, any uniformly directed movements may be at- tributed, with considerable probability, to the con- stant stimulating effect of the light." Several negatively phototactic beetles placed in the apparatus showed a general tendency to turn toward the blackened eye. A Jerusalem cricket, 86 Studies in Animal Behavior Stenopelmatus, which is negative to light, was placed in the enclosure after having its left eye blackened over. When crossing the enclosure it invariably turned to the left and continued to go around in that direction when it came in contact with the edge. Several flies showed very decided circus movements toward the normal side. Thus even under condi- tions of practically continuous stimulation these forms kept up their orienting activities. In another set of experiments insects were held in a fixed position while their efforts at locomotion were given opportunity for expression by rotating a horizontal disk mounted on a pivot like the turn- table of the microscopist. "The apparatus was made very light and easy running so that even a small insect could set it in motion. By holding an insect over the disk with the head pointing either toward or away from the center, and having a light so that the rays fell on one side of the body, the movements of the legs which would ordinarily turn the insect toward the light would simply cause the disk to ro- tate in the opposite direction. With the insect held steadily, the stimulus afforded by the light would naturally remain constant, and if light oriented by its constant stimulating effect we might expect that the insect would keep rotating the disk in its at- tempts at orientation. "Butterflies proved to be very convenient to work with, since by grasping them by the wings folded to- gether above the body they could be held quite The Problem of Orientation 87 steadily, especially with the aid of a hand rest, above the disk. A cabbage butterfly, Pieris rap show a similar reyersal under different intensities of light. In certain cases chemical reactions are af- fected in specific ways by the different rays of the spectrum. "To explain," says Mast, "reversal in the sense of reaction on the basis of chemical reactions induced by light let us assume : ( I ) That Volvox contains substances X and Y, the chemical reaction between which is regulated by the intensity of light; (2) that The Reversal of Tropisms in a sub-optimum intensity favors the formation of substances represented by X and a supra-optimum those represented by Y; and (3) that the colonies are neutral in reaction when there are Y substances in one member of the equation and X in the other; positive when one member contains (X +) sub- stances and the other (Y — ), and negative when one contains (X—) and the other (Y+)." Vol- vox is positive in weak light and negative in strong, but when placed in light of supra-optimal intensity it does not immediately change its response. This, as Mast suggests, may be due to the fact that some time is required to transform the substance X upon which the positive reaction is supposed to depend. After more of Y was produced and the amount of X diminished through the action of more intense light the negative reactions would be initiated. This interpretation, which is confessedly very spec- ulative, not improbably contains elements of truth. It is probable that reversible chemical reactions ef- fect the general restoration of organic equilibrium which had been disturbed by the influence of the stimulus. Whether positive and negative photo- taxis depend on the relative proportions of the sub- stances belonging to a single equation is of course very problematical. It might be that the formation of a substance A which is favored by intensity of light brings about a negative reaction, while the posi- tive reaction might depend upon the formation of another substance B which was not connected with 112 Studies in Animal 'Behavior A at all. We are warranted in assuming that the ability to respond to light depends upon the exist- ence of substances that undergo photochemical changes. The effect of light would be to exhaust certain substances and to cause the accumulation of their products. These products when they exist in a certain degree of concentration may stimulate di- rectly or indirectly (possibly through their influence on oxydative processes) certain locomotor mechan- isms, while at a lower degree of concentration other mechanisms may be more stimulated. It may not be necessary to make any particular assumptions concerning the role of reversible chemical reactions, although the fact that such reversibility is a general property of chemical changes is naturally involved in any explanation of recovery from the effects of stimulation, and indeed in the interpretation of many other vital phenomena. To give a concrete illustration of how reversal may occur according to the preceding interpretation let us consider the reactions of Planaria to weak and strong mechanical stimuli. It has been shown by Pearl that Planaria maculata reacts positively to very weak mechanical stimuli, and in fact to very weak stimuli of many kinds, while to strong stim- uli it gives the negative reaction. The positive re- sponse is brought about by the contraction of the longitudinal muscles near the stimulated point. In the negative response the planarian turns away, not like a higher animal by the contraction of the muscles The Reversal of Tropisms 113 of the opposite side of the body, but by an actual lengthening of the stimulated side. This lengthen- ing is probably effected by the contraction of the dorso-ventral muscles which run from the dorsal to the ventral body wall, and possibly also by the cir- cular muscles. The response is local and near the stimulated area as it generally is in the lower inver- tebrates. The positive and the negative reactions depend either upon the functioning of two quite dis- tinct reflex arcs, or, in case the afferent impulses travelled in the same path, upon the diversion of the stronger stimulus into a new pathway. It is well known that strong stimuli often break over into additional pathways of discharge, and it is of course possible that such a phenomenon might occa- sion a reversal of reaction. What part the inhibitory influence of stimuli may play in reversal is uncertain. Loeb has alluded to the possible role of this factor in one of his earlier papers ('93), but without developing the suggestion further. When an earthworm turns away from the light we may assume, as in fact has been done by Davenport ('97), that the action of the muscles on the more illuminated side is inhibited, and that the animal accordingly turns away from the stimulus. There is no evidence, however, that such an inhibi- tory process occurs. A worm turns violently away when one side is touched with a hot needle, but the movement is not due to the mere inhibitory effect of the stimulus. If the turning away from the hot H4 Studies in Animal Behavior needle is due to the contraction of the longitudinal muscles of the opposite side of the body one would naturally suppose that the avoidance of a light was brought about in the same way. A few years ago I endeavored to study the pos- sible inhibitory effect of light in the phototaxis of Volvox. If the orientation of Volvox when swim- ming toward the light is due to the fact that light tends to inhibit the action of the flagella on the more strongly illuminated side, we should expect that, as the organism passed from a region of dim light to where the light was more intense, its gen- eral rate of locomotion would decrease, since both sides would be more strongly affected by the inhibi- tory stimulus. Specimens of Volvox were placed in a glass trough the bottom of which was marked off into equal spaces. The time required for a speci- men to traverse each space in its course toward the light was noted in a number of cases, and it was found that, on the average, the speed of the organ- ism was quite uniform in the varying intensities of light. The Volvox swam with much the same rapid- ity until they came near the optimum when their pace began to slacken. The results of the experi- ments were therefore opposed to the view that orien- tation is effected by the inhibitory effect of light on the movements of the flagella on the more illumi- nated side of the organism. Inhibition may function in other ways in the orientation .of Volvox and in changing the sense of its reactions to light and other The Reversal of Tropisms 115 forms of stimulation. There is much to indicate that inhibition is intimately related to reversal of tro- pisms in many forms, but I shall not venture upon any speculations, which at best could only be very tentative, as to its method of operation. It is quite possible that the explanation of reversal of tropisms may be quite different in different cases. Many rhizopods show a positive thigmotaxis to a weak mechanical stimulus, while they give a negative reaction if the stimulus is strong. Is this to be ex- plained in the same way as the reversal in the beat of the flagellum in a flagellate, or of the cilia in an infusorian? And does any of these cases have any- thing in common with the reversal of tropisms in an insect or worm? The mechanism of orienta- tion is quite different in different organisms, and it seems not improbable that the inner mechanism of reversal may be very different also, although there may be broad underlying features common to nu- merous apparently different cases. Up to the present time most of the work on the reversal of tropisms has been done with a view of ascertaining the various conditions under which re- versals may occur. Many interesting facts have been accumulated, but, as we have seen, they afford no very sound basis for generalization. We shall make little further progress by mere induction. We need to feel our way along by making hypotheses and testing them by appropriate experiments. But when one attempts to attack the problem in this way n6 Studies in Animal Behavior he soon finds himself hampered by the paucity of knowledge in those fields to which he would naturally turn for helpful suggestions. It would be very de- sirable to have a more adequate knowledge of the inner mechanism of orientation. It would be very helpful to have a deeper insight into the physiology of irritability. We know little of the processes cov- ered by the word stimulation. We are still looking for a satisfactory explanation of the curious phe- nomenon of inhibition. The behaviorist may discover many facts of interest and value regarding the re- versal of tropisms, but before arriving at an ade- quate explanation of this perplexing phenomenon he may have to wait until some of the more general and fundamental problems of irritability have been solved. REFERENCES BANCROFT, F. W. ( i ) The control of galvano- tropism in Paramoecium by chemicals. Univ. of Calif. Pubs. Physiol. 3, 21, 1906; (2) Heliotro- pism, etc., in Euglena. Jour. Exp. Zool. 15, 383, BAUER, V. Ueber die reflektorische Regulie- rung der Schwimmbewegungen bei den Mysideen, etc. Zeit. allg. Physiol. 8, 343, 1908. BOHN, G. La naissance de 1'intelligence. Paris, 1909. DAVENPORT, C. B. Experimental Morphology. N. Y., 1897-99. References 117, DICE, L. H. The factors determining the verti- cal movements of Daphnia. Jour. Animal Behavior, 4, 229, 1914. ESTERLY, C. O. Reactions of copepods to light and gravity. Am. Jour. Physiol., 18, 47, 1907. EWALD, W. F. On artificial modification of light reactions and the influence of electrolytes on photo- taxis. Jour. Exp. Zool., 13, 591, 1912. GROOM, T. T. and LOEB, J. Der Heliotropis- mus der Nauplius von Balanus perforatus und die periodischen Tiefenwanderungen pelagischer Tiere. Biol. Cent., 10, 160, 1890. HARPER, E. H. Behavior of the phantom larva of Corethra plumicornis. Jour. Comp. Neur. Psych., 17. 435. I907- HOLMES, S. J. (i) Phototaxis in the Amphi- poda. Am. Jour. Physiol., 5, 211, 1901; (2) Pho- totaxis in Volvox. Biol. Bull., 4, 319, 1903; (3) The reactions of Ranatra to light. Jour. Comp. Neur. Psych., 15, 305, 1905; (4) The reactions of mosquitoes to light in different stages of their life history. Jour. Animal Behavior, i, 29, 1911. JACKSON, H. H. T. The control of phototactic reactions in Hyalella by chemicals. Jour. Comp. Neur. Psych., 20, 259, 1910. LOEB, J. Ueber kiinstliche Umwandlung positiv heliotropischer Tiere in negativ heliotropische und umgekehrt. Arch. ges. Physiol., 54, 81, 1893. MAST, S. O. ( i ) Light reactions in lower or- ganisms. II. Volvox globator. Jour. Comp. Neur. iiS Studies in Animal Behavior Psych., 17, 99, 1907; (2) Light and the behavior of organisms. N. Y., 1911. McGiNNis, M. O. Reactions of Branchipus ser- ratus to light, heat and gravity. Jour. Exp. Zool., 10, 227, 1911. MOORE, A. R. ( i ) Concerning negative photo- tropism in Daphnia pulex. Jour. Exp. Zool., 13, 572, 1012; (2) Negative phototropism in Diapto- mus by means of strychnine. Univ. of Calif. Pubs. Physiol., 4, 185, 1912. MOORE, B. Observations of certain marine or- ganisms of (a) variations in reaction to light, and (b) diurnal periodicity of phosphorescence. Bio- chemical Jour., 4, i, 1909. PARKER, G. H. Reactions of copepods to vari- ous stimuli and the bearing of this on their daily depth-migrations. Bull. U. S. Fish Comm., 21, 103,' 1901. PEARL, R. The movements and reactions of fresh-water planarians. Quart. Jour. Mic. Sci., 46, 509, 1903. STRASBURGER, E. Die Wirkung des Lichtes und der Warme auf Schwarmsporen. Jen. Zeit. n. F. 5, 572, 1878. TERRY, O. P. Galvanotropism of Volvox. Am. Jour. Physiol., 15, 235, 1906. TOWLE, E. A study in the heliotropism of Cy- pridopsis. Am. Jour. Physiol., 3, 345, 1900. VERWORN, M. General physiology. London, 1899. References 119 YERKES, R. M. Reactions of entomostraca to stimulation by light. II. Reactions of Daphnia and Cypris. Am. Jour. Physiol., 4, 405, 1901. VI THE BEGINNINGS OF INTELLIGENCE "^TOTHING shews more the force of habit in ••• ^ reconciling us to any phenomenon, than this, that men are not astonish'd at the operations of their own reason, at the same time, that they ad- mire the instinct of animals, and find a difficulty in explaining it, merely because it can not be reduc'd to the very same principles. To consider the matter aright, reason is nothing but a wonderful and un- intelligible instinct in our souls, which carries us along a train of ideas, and endows them with par- ticular qualities, according to their particular situa- tions and relations. — David Hume, Treatise on Human Nature. We all have a certain curiosity regarding the evo- lutionary history of our various powers and attri- butes, but from many points of view an unusual interest attaches to the first development of intelli- gence. The word intelligence is used in a variety of senses by writers on comparative psychology, and any discussion of the origin of intelligence would be fruitless unless the meaning in which the term is employed be understood. One of the foremost 120 The Beginnings of Intelligence 121 of comparative psychologists, the acute Father Was- mann, defines intelligence as "the power of con- ceiving the relation of concepts to^ one another and of drawing conclusions therefrom. It involves ab- straction, deliberation and self-conscious activity." Intelligence, according to Wasmann, is the God- given attribute of man alone ; its possession separates man from brute by an impassable barrier. Many comparative psychologists, among whom we may mention Lloyd Morgan, Forel and Loeb, adopt as a criterion of intelligence the power of forming associations, or associative memory^ anxl we^sKalT follow the usage of these writers: it is obvious that the possession of this faculty marks an important step/ in advance upon the creatures whose actions are fatally determined by their in- stinctive make-up. From its beginning in forms in which the simplest associations are established only after a large number of experiences, intelligence has assumed a role of ever-increasing importance in the evolution of animal life, until in man, who is notoriously a weakling compared with the large beasts with which he has had to contend, it became the main factor to which the human species owes its supremacy in the struggle for existence. In considering the origin of intelligence one is naturally led to the subject of; the relation of in- telligence to instinct. Formerly it was the custom to contrast these two faculties as if they represented diametrically opposed types of activity. Instinct was 122 Studies in Animal Behavior regarded as something unalterably fixed, machine- like and practically perfect in its adaptation to the needs of the animal; intelligence was recognized as the antithesis of all these qualities — variable, plas- tic and eminently fallible. With the establishment of the theory of evolution writers became more disposed to discover the kinship and filiation of in- stinct! and intelligence and they have given us a variety of views as to the relation of these facul- ties. Basing his theory on Lamarck's doctrine that in- stinct is inherited habit, G. H. Lewes attempted to explain instinct as "lapsed intelligence." Per- formances which are learned with difficulty come, after sufficient repetition, to be carried out auto- matically and without any intelligent1 guidance. If the acquired facility of performing these acts is inherited and the acts are repeated generation after generation, it is probable that they might finally be performed by an individual without any previ- ous instruction at all; that is, they would become instinctive. An animal's instincts, according to this view, represent the stereotyped and mechanized be- havior which its ancestors found to be profitable; their adaptiveness rests upon the wisdom acquired by ancestral experience. More recently this view has been upheld by Eimer, and in a less extreme form by Romanes, Wundt and many others. One difficulty with the theory of lapsed intelli- gence is that it involves the acceptance of the doc- The Beginnings of Intelligence 123 trine of the transmission of acquired characters, which has come to be a very questionable biological theory. But another and more, fundamental dif- ficulty is revealed by recent work on the behavior of lower organisms. If instinct were derived from intelligence by a sort of mechanizing process we should expect, as Whitman has urged in his criti- cism of LewesV theory, to find intelligence domi- nant in lower forms of life, and that acts which are instinctive in the higher animals would be intelli- gently performed by; the lower ones. The work that' has been done on the behavior of lower or- ganisms enables us to state with confidence that such is not the case. In several large phyla of the lower invertebrates there has not, as yet, been demonstrated the least glimmer of intelligence ; and, as we pass up the scale of life, intelligence grad- ually supersedes instinct, not the reverse. We can say with some degree of assurance that, however the transition may have been effected, intelligence has grown out of purely instinctive behavior. It is not possible, however, to fix, except with the rudest approximation, the stage of evolution at which intelligence makes its first appearance. The transition from instinct to intelligence has been made, in all probability, not once, merely, but sev- eral times along different lines of descent. Intelli- gence in the vertebrates doubtless arose independ- ently from that of the insects, and the intelligence exhibited here and there among the mollusks prob- 124 Studies in Animal Behavior ably arose independently along a third line of de- velopment. Intelligence makes its appearance at a certain stage of organization along whatever line such a stage may have been reached. Up to the point at which the power of associa- tive memory becomes manifest there has been prog- ress along many lines which has prepared the way for the evolution of this new faculty. Behavior has not only become more complex, but it has be- come more plastic and capable of easy modification to suit new conditions. The lower organisms do not always react in a particular way to a given stim- ulus. What reaction occurs may depend upon the number of previous stimulations, the supply of food, exposure to different environing conditions, and numerous other factors which influence the internal state of the organism. The behavior of many lower animals is plastic and adaptive to a remarkable degree, and to a superficial consideration often gives the appearance of a considerable degree of intelli- gence, without there being any detectable power of associative memory. This plastic and varied be- havior not only simulates intelligence, but it se- cures for the organisms many of the advantages which intelligence confers. It adapts the animal to a more varied environment, and gives it the power of meeting a given situation in more than one way, so if one kind of response does not suit, another may be more successful. Let us glance briefly at some of the ways in which behavior may be modified. The Beginnings of Intelligence 125 A very general change of behavior in organ- isms consists in the habituation to any stimulus which is repeated at sufficiently close intervals so that the organism no longer responds to it. This is shown even among the protozoa. A Stentor or a Loxo- phyllum subjected to a light mechanical stimulus at short intervals soon fails to respond as at first, but the duration of the modification so produced is very short; in Loxophyllum it probably does not extend over two -or three seconds. Similar effects of re- peated stimulation but of longer duration have been observed in Hydra, several species of sea-anemones, planarians, annelids and various other lower inver- tebrates. As a rule failure to respond may occur more quickly and the effects of the stimulus remain longer as we pass up the scale of animal life. Occasionally the reverse phenomenon occurs when the response to a given stimulus is increased instead of diminished with repeated applications — a result which suggests the effect of the summation of stimuli. At times, as Bohn found in Cerianthus, there is an initial increase of responsiveness fol- lowed by a dulling of sensitivity. Bohn has at- tempted to subsume the effects of repeated stimula- tion under a general "law" to the effect that stimu- lation always produces at first increase of sensitivity to be followed later by a decrease. Sometimes, as Bohn claims, the initial increase is so short as to escape detection ; which may be true, but the burden, of proof is on M. Bohn. 126 Studies in Animal Behavior Repetition of a stimulus may call forth not only quantitative differences of response, but it may evoke responses of very different character. Animals are frequently provided with several modes of reacting to a given stimulus which may be called into play one after the other. Jennings has shown that if a Stentor is subjected to a light mechanical stimulus by causing fine particles of India ink to fall upon its disk from a capillary pipette it usually reacts first by bending a little to one side. If the parti- cles continue to fall on the disk the beat of the cilia covering the body may suddenly be reversed, thus creating a current tending to carry the offend- ing particles away. If in spite of this the particles still impinge upon the disk the Stentor may con- tract one or more times. Finally, if all these re- actions are tried in vain the infusorian may give a number of violent contractions, break loose from its place of attachment, and swim away. It would be an error to interpret the varied be- havior of this unicellular organism as a manifesta- tion of intelligence, although it is not unlike what the behavior of an intelligent creature might be under the circumstances. No power of learning by^ experience has ever been discovered in Stentor, or indeed in any other protozoan. The organism is provided with a number of different modes of re- sponse, and which one is set in action depends upon internal factors which are influenced by the crea- ture's previous activity. The organism which has The Beginnings of Intelligence 127 responded to a stimulus has become transformed into a different mechanism which may respond more or less readily than before or radically change its method of behavior. A striking illustration of varied responses to a given stimulus has been described by Jennings in the sea anemone Stoichactis. If a foreign body is placed upon its disk the anemone tries to rid itself of the object in various ways. The tentacles near the ob- ject collapse and the area between them extends, thus producing a relatively smooth surface so that the waves can readily wash the object away. If this does not occur the region under the object begins to swell, thus rendering the removal of the object still easier. If this reaction is unsuccessful the edge of the disk begins to sink so that a smooth sloping sur- face is formed from which the object can readily slide. Here, as in the case of Stentor, we have an organism capable of reacting in several ways to a given stimulus. What particular reaction is evoked depends upon previous stimulations. Modifications of behavior caused by different con- ditions of nutrition are found in the lowest mem- bers of the animal kingdom. Even the white blood cells after they have ingested a number of bacteria refuse to take in more. Whether there is a limit to the appetite of Amoeba has not been determined, but many infusorians such as Stentor, after having swept in a certain amount of food, react to food particles in a quite different way than when in a hun- 128 Studies in Animal Behavior gry condition. Hydra when not fed for some time extends the body, sways about in various directions and keeps up a restless movement of its tentacles, thereby increasing its chances of contact witfy the small creatures which serve as its prey. Instances of the non-intelligent modifications of behavior might be multiplied indefinitely. As we pass to higher forms the capacity for responding in different ways to a given situation becomes greatly increased. "Nature," says James in his admirable chapter on instinct, "implants contrary impulses to act in many classes of things, and leaves it to slight alterations of the conditions of the individual case to decide which impulse shall carry the day," and he points out that many animals lose the instinctive demeanor and appear to lead a life of hesitation and choice, not because they have no instincts, but because they have so many of them that they block one another's path. Intelligence in the accepta- tion of the term which we have adopted begins with the formation of associations. It does not make its appearance, so far as is known, until a compara- tively high stage of organization has been attained. The evolution along the lines of complexity of in- stinct and ready modifiability of reactions to suit new conditions, affords a substantial basis for in- telligent behavior. Without such evolution the power of associative memory would avail little. But with a large number of readily modifiable in- stincts, associative memory becomes the means of The Beginnings of Intelligence 129 affording a much wider and closer adjustment to the environment. The studies which have been made of primitive types of intelligence such as found in crustaceans, fishes and amphibians have shown that associations are formed by a gradual process of reinforcement or inhibition of a particular reaction to a given stim- ulus. The method followed is one which Lloyd Morgan has designated as "trial and error." It may be illustrated by the experiment of Yerkes on the formation of associations in the crayfish. In these experiments a box was employed into one end of which the crayfish was admitted through a narrow aperture. The other end of the box was divided by a median partition which gave the cray- fish a choice of two routes to a tank of water at the other end into which the creature was naturally desirous of getting. One of the two ways to the water was closed by a glass plate at its farther end so that the crayfish was afforded a choice of a right and a wrong path to the water. Would the crayfish after a number of trials learn to choose the right path and avoid the closed passage? In the first ten experiments the crayfish went as often to the right as it did to the left, but in the next ten trials the percentage of correct choices was somewhat greater. Finally after a large number of trials the animal came to choose the right path to the water, mak- ing but rarely any mistakes. Similar experiments with crabs, fishes and the 130 Studies in Animal Behavior frog have yielded similar indications of slow learn- ing. In some respects such learning resembles the slow formation of a habit rather than the judgment of a consciousness which "sizes up" the situation and determines upon a certain course of action. It is quite probable that such a primitive form of learn- ing does not include any association of ideas. It can be satisfactorily accounted for by assuming noth- ing more than an association of certain sense per- ceptions with particular movements. The animal may have no ideas to associate — nothing but sense impressions and motor impulses. Of course its men- tal content may include much more than this, but in interpreting the behavior of animals it is generally advantageous to follow the principle laid down by Lloyd Morgan — which is a sort of special appli- cation of the law of parsimony — that we should not assume the existence of a higher psychic function if the phenomena can be explained as well in terms of a lower one. The step from sensori-motor association to the as- sociation of ideas is not, I believe, a wide one, and comes about as a natural consequence of the elabo- rateness and what Hobhouse has designated as the "articulateness" of the mental process of adjustment. It is foreign to our purpose, however, to trace the increase in the number, delicacy, quickness and com- plexity of the processes of association which we meet in the various stages of mental evolution. Our prob- lem at present lies in the initial step involved in the The Beginnings of Intelligence 131 formation of a simple association. And it is a prob- lem which, despite its apparent simplicity, involves the consideration of some vexed and subtle ques- tions. In learning we have to do with two opposite processes of reinforcement and inhibition. A chick after it pecks at a caterpillar which is wholesome and savory pecks at a similar caterpillar more read- ily on a second occasion. Something has apparently reinforced the connection between the visual impres- sion produced by the caterpillar and the pecking im- pulse. If, on the other hand, the chick pecks at a caterpillar having a nasty taste it is apt to avoid pecking at it a second time. Something has hap- pened to inhibit the response that would otherwise occur. We commonly explain such behavior by ascribing to the creature feelings of pleasure and pain. We say that the chick pecks at one kind of a caterpillar because of the pleasant taste it de- rives, and avoids another variety because its taste is bad. Pleasure and pain apparently function as agents for the reinforcement of certain reactions and the stamping out of others. It is a general rule, though not without certain exceptions, that what affords pleasure is conducive to organic wel- fare, while that which is productive of pain is in- jurious. The upshot is that the associations that are the outcome of the pleasure-pain response are of just the kind that minister to the animals' needs. Beneficent arrangement! Apparently we have to do 132 Studies in Animal Behavior with a selective agency which preserves and intensi- fies certain kinds of behavior and rejects others on the basis of their results — a kind of "sorting demon" in the realm of behavior. What could be more tele- ological ! The fact that what is pleasant is usually beneficial and what is painful is usually injurious may be ex- plained with some plausibility as the result of nat- ural selection, as was first contended by Herbert Spencer. Animals which took pleasure in doing things which were bad for them and which experi- enced pain in doing things which were good for them would be very apt to fare ill in the struggle for existence. Natural selection would ever tend to bring about a condition in which the pleasant means the organically good and the painful means the re- verse. We should not expect the correspondence, if brought about in this way, to be complete, and it is rather in favor of the theory that we do not find it so. But granting this contention of Spencer, there is the important question still left unanswered, namely, Why do animals follow what is pleasant and avoid what is painful? In other words, why does pleas- ure reinforce and why does pain inhibit? Here is another fundamental problem and we find that Spen- cer with his usual appreciation of fundamental prob- lems was on the ground early with a theory. Pleas- ure, according to Spencer, is the concomitant of a heightened nervous discharge; pain the concomitant V The Beginnings of Intelligence 133 of a lessened nervous discharge. An act which brings pleasure causes an influx of nervous energy to the centers concerned in the movement; the lines of discharge become "more permeable," and upon a repetition of the conditions the same act follows with greater readiness than before. If the act is followed by pain with its concomitant of lessened nervous discharge, the diminution of nervous energy serves to prevent the performance of the act in re- sponse to the same conditions. Closely similar ex- planations of the physiology of the pleasure-pain re- sponse have been given by Bain and by Baldwin, the latter declaring that "pleasure and pain can be agents of accommodation and development only if the one, pleasure, carry with it the phenomenon of motor ex- cess— and the other, pain, the reverse — probably some form of inhibition or of antagonistic contrac- tion." The physiological concomitants of pleasure and pain have afforded a subject for numerous labora- tory studies and almost no end of theories. It has been impossible thus far to discover that either of these states is invariably accompanied by any defi- nite physiological condition. The theory of Spen- cer and Bain is open to obvious criticism, for the man who steps on a tack undoubtedly has a "height- ened nervous discharge," as much as a man who shouts for joy. And I believe I am safe in saying that no theory of the physiology of pleasure .and pain is on a sufficiently firm basis to warrant its be- 134 Studies in Animal Behavior ing regarded as anything more than a very tenta- tive working hypothesis. With our present knowledge of the psycho-physi- ology of pleasure and pain, the attempt to explain how these states or their physiological concomitants, whatever they may be, can act as agents of rein- forcement and inhibition seems rather a fruitless one. The process which we meet at the beginning of intelligence in simple associative memory may be formulated as follows: stimulus — reaction — pleasure — reinforcement « physiological state x stimulus — reaction — pain — inhibition physiological state y Spencer, Bain and others have endeavored to show how the organic accompaniments of pleasure and pain modify the creatures' subsequent responses. But as the problem was interpreted by these writers our ignorance concerning the physiological states x and y brings us to a standstill. In his valuable work on Mind in Evolution Hobhouse has presented a new point of view in considering this problem, which has the advantage of not involving any general theory of the physi- ology of pleasure and pain. It is essentially a the- ory of how behavior comes to be adaptively modi- fied through the formation of associations. It makes The Beginnings of Intelligence 135 no attempt to explain why pleasure is associated with certain experiences and pain with others. Such association may turn out to be as inexplicable as the problem why stimulation of the optic nerve gives rise to a sensation of light instead of some other kind of feeling. What it is feasible to attempt to explain is why certain responses tend to be repeated and others tend to be inhibited. And this can be explained with some plausibility as due to the con- gruity or incongruity of the reactions which come to be associated. For the sake of illustration let us consider again the chick which pecks at a nasty caterpillar. The irritation set up by the caterpillar in the chick's mouth evokes movements of with- drawal and ejection. The two responses of peck- ing and ejection become associated, but as the two movements are contradictory the result is inhibition. The pecking reaction no longer occurs in the pres- ence of a second nasty caterpillar, not because of any stamping-out influence of the physiological con- comitant of pain, but because it becomes joined with an antagonistic reaction. In a previous paper by the writer the attempt was made to extend the theory of Hobhouse to account for the reinforcement commonly held to be caused by pleasure. The assumption was made that this process is due to an organic congruity of the reactions. If the caterpillar pecked at is a savory one there is set up the reflex of swallowing. Pecking and swallowing form the normal elements 136 Studies in Animal Behavior of a chain reflex; when one part of the structure concerned is excited it tends to increase the tonus of the associated parts, and thus reinforce the origi- nal response. I have found that in the crayfish stim- ulation of the antennules, which are important or- gans of smell, sets up chewing movements of the mouth parts and grasping movements of the small chehe. Similarly stimulating the small chelae evokes chewing movements of the mouth parts and twitch- ing of the antennules, while stimulating the mouth parts directly may cause movements in both the other sets of organs. We have here as a matter of fact a number of reflexes which mutually rein- force one another. Suppose that in the chick the sight-pecking response and the taste-swallowing re- sponse are related as the feeding reflexes demon- strably are in the crayfish ; the second response would thus tend to reinforce the first, and if this tendency persisted we would have a case of learning by ex- perience. Animals in the course of their instinctive responses encounter stimuli which bring about other responses. These become associated. According to the nature of the nervous pathways involved, there may be re- inforcement of, or interference with the original re- action. Experience brings about an extension of the range of adaptations by the assimilation of con- gruent reactions and the elimination of acts whose secondary consequences are in the nature of an- tagonistic and thereby inhibitory responses. Such References 13? we may say, by way of expressing a tentative view- point, is the nature of primitive intelligence. But it will be seen that the capacity to form new adaptations rests upon the primary adaptive- ness of the instinctive reactions. The power of for- mation of associations alone would never lead to improvement. The adaptiveness of intelligence is based upon the adaptiveness of instinct; it may be said that intelligence is a means of enabling an ani- mal to live its life more completely and success- fully, but instinct furnishes the fundamental springs of action. Even complex creatures like ourselves form no exception to this rule. REFERENCES BAIN, A. The senses and the intellect. 3d ed., 1894. The emotions and the will, 4th ed., 1899. BALDWIN, J. M. Mental development. Meth- ods and processes. 2nd ed. N. Y., 1903. De- velopment and evolution. N. Y., 1902. BOHN, G. La naissance de 1'intelligence. Paris, 1909. HOB HOUSE, L. T. Mind in evolution. London, 1901. HOLMES, S. J. ( i ) Pleasure, pain and the be- ginnings of intelligence. Jour. Comp. Neur. Psych. 20, 145, 1910. (2) The evolution of animal intelli- gence. N. Y., 191 1. 138 Studies in Animal Behavior JENNINGS, H. S. ( i ) Contributions to the study of the behavior of lower organisms. Carnegie Inst. Pubs., 1904. (2) Modifiability in behavior. I, Behavior of sea anemones. Jour. Exp. Zool. 2, 447, 1905. (3) Behavior of lower organisms. N. Y., 1906. SPFNCER, H. Principles of psychology. N. Y., 1892. WASMANN, E. Instinct and intelligence in the animal kingdom. St. Louis, 1903. WHITMAN, C. O. Animal behavior. Biol. Lec- tures, Marine Biol. Lab., Woods Hole, 1898. VII SOME CONSIDERATIONS ON THE PROBLEM OF LEARNING ' I AHE essential nature of the process of learning *- constitutes a problem of such fundamental im- portance for psychology, to say nothing of physi- ology also, that any discussion which promises to contribute anything, however little, toward its solu- tion is abundantly justified. In the previous chapter an interpretation of the process of learning was briefly outlined, and it may be profitable to con- sider here some other proposed solutions of the same problem, as well as certain criticisms of the view set forth in the preceding pages. The formu- lation and criticism of different hypotheses regard- ing the mechanism of learning is especially desir- able, since the pathway ahead is none too clear, and since we have to rely to a large extent upon the method of trial and error in order to make prog- ress. An ingenious explanation of the learning process has been put forward by Thorndike,1 who, like Spencer and Bain, finds the explanation of the re- inforcement and inhibition of reactions in the pe- 1 "Animal Intelligence." The Macmillan Co., N. Y., 1911. 139 140 Studies in Animal Behavior culiar physiological processes supposed to accom- pany the satisfying and annoying experiences of the animal. Thorndike recognizes that while the ten- dency of animals to repeat responses that bring sat- isfaction or pleasure and to discontinue responses that entail pain usually leads to advantageous re- sults, this rule is not without its exceptions. "Many animals are satisfied by deleterious conditions. Ex- citement, overeating, alcoholic intoxication are, for instance, very potent satisfiers of man," and condi- tions which are very salutary often fail to produce satisfaction, and may even bring positive displeas- ure. All this is simply a matter of imperfect ad- justment. "Upon examination," says Thorndike, "it appears that the pernicious states which an animal welcomes are not pernicious at the time, to the neu- rones. We learn many bad habits, such as mor- phinism, because there is incomplete adaptation of all the interests of the body-state to the temporary interest of its ruling class, the neurones. So also the unsatisfying goods are not goods to the neu- rones at the time. We neglect many benefits be- cause the neurones choose their immediate advan- tage. The neurones must be tricked into permitting the animal to take exercise when freezing or qui- nine when in a fever, or to free the stomach from certain poisons. "Satisfaction and discomfort, welcoming and avoiding, thus seem to be related to the mainte- nance and hindrance of the life processes of the neu- Considerations on the Problem of Learning 141 rones rather than of the animal as a whole, and to temporary rather than permanent maintenance and hindrance." Now the modification of behavior through changes in the neurones is concerned chiefly with what affects the permeability of certain lines of communication in the nervous system. The seat of these changes in permeability is thought by many physiologists to reside in the synapses or membranes between the ends of anastomosing processes of the nerve cells. While the experimental evidence for this conclusion is rather meager we may adopt it provisionally as perhaps the most plausible view at the present time. The condition which permits the ready transfer of impulses from one neurone to another Thorndike calls the "intimacy of the syn- apse," and he formulates the following provisional hypothesis to account for the process of learning: "A neurone modifies the intimacy of its synapses so as to keep intimate those by whose intimacy its other life processes are favored and to weaken the intimacy of those whereby its other life processes are hindered. The animal's action-system as a whole consequently does nothing to avoid that response whereby the life processes of the neurones other than connection changing are maintained, but does cease those responses whereby such life processes of the neurones are hindered. "This hypothesis has two important consequences. First: Learning by the law of effect is then more 142 Studies in Animal Behavior fully adaptive for the neurones in the changing in- timacy of whose synapses learning consists, than for the animal as a whole. It is adaptive for the animal as a whole only in so far as his organiza- tion makes the neurones concerned in the learning welcome states of affairs that are favorable to his life and that of his species and reject those that are harmful. "Second: A mechanism in the neurones gives results in the behavior of the animal as a whole that seem beyond mechanism. By their unmodifiable abandonment of certain specific conditions and re- tention of others, the animal as a whole can modify its behavior. Their one rule of conduct causes in him a countless complexity of habits. The learning of an animal is an instinct of its neurones." Of course the assumption that the neurones re- act so as to make themselves more permeable to stimuli that are beneficial, and to make themselves less permeable to stimuli that are injurious has no direct evidence in its support. Its value consists in its serviceableness as an explanation of learning. But notwithstanding the fact that reactions on the part of cells are very frequently teleological the hypothesis is, I believe, not in accord with what is known of the physiology of the nervous system. So far as we know, nervous impulses are the same in character everywhere. We should expect that up to a certain degree of intensity, like functional stim- ulation in general, nervous impulses would enhance Considerations on the Problem of Learning 143 the life processes of the neurones. Only when over- stimulated would we expect that there would be any deleterious or annoying effects which are sup- posed to result in the reduction of the permeability of the synapse. Thorndike's doctrine then natu- rally leads to the position that the vitally good and pleasant stimulations are those of optimal or sub-optimal intensity, while stimulations of greater intensity would produce effects which are unpleas- ant and deleterious, at least at the time. Are the facts such as the theory would lead us to expect? Stimulation of pain nerves produces sensations which are generally disagreeable in all degrees of \ intensity which can be appreciated. Shall we say ^, then that all stimulation of the neurones involved in responses to pain giving stimuli are injurious to ; the neurones involved? If so, practically all func- \ tional stimulation of these neurones would be dele- terious to them. On the other hand, certain sen- N sory systems may be stimulated strongly for a long^.^/ time without producing results that are unpleas- &-^ ant. It is far from proven that the sensationsxa^^ aroused by stimulating the nerves of touch, heat, and certain of the nerves of taste and smell are ever disagreeable, however strongly the end organs are stimulated. It is probable that most of the unpleas- ant results alleged to follow from overstimulating a particular sense organ arise from the fact that Pjim_jierves become involved when the stimulating £, s£> agent is sufficiently strong. This is very probable 144 Studies in Animal Behavior in the case of heat; and there is more or less evi- dence for the same conclusion in regard to very in- tense visual and auditory stimuli, although it can not be regarded as entirely established. Even were it proven that overstimulation of any sense organ produces unpleasant effects (and avoiding reactions as a secondary result), the unpleasantness may be due not so much to the injury to the overstimu- lated neurones as to the breaking over of the stronger neural current into new channels. Certain tastes and smells are disagreeable even when they can barely be perceived at all, while others are welcome at almost any degree of intensity. It is very improbable that the disagreeable quality of the former is due to the supra-optimal stimulation of certain associated neurones. It is not so much the intensity of the effect produced by the stimulus that gives rise to disagreeable results and inhibitory effects as the channels through which the stimuli en- ter the nervous system. Were Thorndike's theory correct, we should expect to find comparatively mild stimulations producing, quite generally, pleasurable feelings, and strong stimuli producing disagreeable feelings, and, directly or indirectly, avoiding reac- tions. On the contrary, we find the organism pro- vided with sets of reflex arcs, each of which tends to be set into operation by its own kind of stimu- lating agent, and producing feelings of much the same quality, however their intensity may vary. Another consequence of Thorndike's theory is Considerations on the Problem of Learning 145 that reactions to disagreeable stimuli should be- come reduced the more often they occur. If a re- ceptor is connected with neurones A, B, C, and D involved in a reflex response to a stimulus that pro- duces disagreeable effects and an avoiding reaction, then as the neurones are, ex hypothesi, injuriously affected, the intimacy of their synapses would be weakened and the course of the impulse through the system more or less blocked. The disagreeable ef- fects and the vigor of the avoiding response would therefore tend to wear away after successive repeti- tions. In general it cannot be said, I think, that this occurs. Castor oil ought, after being taken a few times, to become much less unpleasant, but I very much doubt if this is a common experience. People suffer pain from a particular region for years with little or no diminution of intensity, and there are some things which up to a certain point may become more disagreeable, instead of less so, the more often they are experienced. Sometimes, it is true, things which are at first disagreeable come to be tolerated in time with little discomfort, but this is probably because the incoming stimuli be- come connected with other neurones than the ones which gave the original motor expression. In other cases it may be due to quite different organic ad- justments. It is doubtful if where the unpleasant stimuli continue to be responded to through the same channels, there is in general a weakening of the response or mitigation of its disagreeable qual- 146 Studies in Animal Behavior ity. According to the theory, responses involving unpleasant effects should come to inhibit themselves. Such a result might shield the neurones from fur- ther injurious stimulation, but it would hardly be conducive to the welfare of the organism in general, since it should keep on reacting so as to avoid sources of injury. That there is any general ten- dency for unpleasant responses to become reduced in vigor, apart from the purely temporary effects of fatigue common to all responses, is very ques- tionable. On the other hand, there is probably no general tendency for pleasant responses to become per- formed with greater vigor or to be accompanied by heightened satisfaction. Learning to like cer- tain articles of food as well as coming to dislike others is probably a matter of secondary associa- tion. Addiction to alcoholic drinks and other so- called habit-forming drugs is a phenomenon rather exceptional in character, the explanation of which need not concern us at present. Doubtless second- ary associations are involved here also. It is prob- ably not the initial pleasure which these drugs arouse that impresses the habit of using them (the sub- cutaneous injection of morphine, for instance, is the reverse of pleasant) , but some general physiological effect which is pretty closely associated with their injurious influence on the neurones. Under certain kinds of intoxication a person may experience feel- ings of pleasure that go along with cerebral con- Considerations on the Problem of Learning 147 ditions which are decidedly unwholesome. Ordinarily, pleasant tastes and smells retain very nearly their original algedonic quality. We may become more apt to seek a particular taste that we have had before, but this in no way goes to show that the original response set up by the pleasant substance takes place more easily or is attended with greater pleasure. Pleasures often wear away, and things once agreeable may come to pall upon us. There is little evidence that pleasure-giving stimuli in general tend to reinforce themselves, and where reinforcement occurs it is probably due to second- ary associations with other reflex arcs. The effects of agreeable and disagreeable states upon subsequent reactions to the stimuli immedi- ately producing them are much the same. Consid- ered as single responses both may be fatigued by repetition and influenced similarly by general bodily conditions. The notable thing about unpleasant responses is that they jexercise their inhibitory in- fluence on other reflex systems. It is chiefly in their influence on other associated reflex arcs that the dif- ferent effects of agreeable and disagreeable stimuli become manifest. The nasty caterpillar in the mouth of a chick may not become less disagreeable when picked up on subsequent occasions, nor the edible worm may not give a more intense thrill of satisfaction, but the experience with the nasty cater- pillar tends to check the pecking reflex, while the seizure of the edible caterpillar tends to reinforce 148 Studies in Animal Behavior it. Satisfying and annoying states do not affect the reactions by which these feelings are immediately aroused, but other reactions performed in close tem- poral relation to them. The various theories put forward to explain the learning process as an effect of the physiological cor- relates of agreeable and disagreeable states are all open, I believe, to serious objection. Aside from the fact that no theory as to what these physiologi- cal correlates are is on a firm basis there is no one standpoint that gives a consistent interpretation of the process of learning. It is not evident that the correlates of all the varied states that are annoy- ing have any general physiological characteristic which distinguishes them from the correlates of states which are pleasurable. And even if they have, it does not follow that these correlates are par- ticularly concerned with the mechanism of associa- tion. The view of learning sketched in the preceding chapter has at least the merit of avoiding any as- sumptions- in regard to the physiological concomi- tants of agreeable and disagreeable mental states. The attempt was made to show that, given an en- dowment of instinct plus the ability to form asso- ciations, an animal may acquire modes of behavior which adapt it to new conditions of life. A re- sponse which results in setting into action a strong instinctive proclivity is reinforced or inhibited, as the case may be, according to its congruity or incon- Considerations on the Problem of Learning 149 gruity with the proclivity thus aroused. Ordinarily a response A that is closely followed by an instinc- tive reaction B involving the liberation of a consid- erable amount of nervous energy is reinforced, probably as a result of the influence of this energy on the nervous connections simultaneously excited by the response A. If, however, the associated re- action B is opposed to A, as when an outreaching action is followed by a withdrawing response, the influence of B tends to inhibit the first response A. In both cases, A and B tend to become associated, but the different secondary reactions have different effects on the primary response with which they have become joined. Thorndike l has raised several objections to the view here discussed, but none of them, I believe, are fatal or even serious. "A secondary response R2 may bind Rx to Sj [the primary response to the in- itial stimulus], even though it is incongruous with it, and disjoin R! from Si though it is congruous with Rx. Thus a cat in a box, the door of which is opened, permitting escape and eating whenever the cat scratches herself, will soon come to scratch her- self as soon as put in the box, though there is no congruity between escape through a door and scratching." This objection is based on a miscon- ception of the sense in which the word congruity was used in the statement of the theory criticized. Viewed as two external acts of the cat's body, there 1 "Educational Psychology," I, 189, 1913. 150 Studies in Animal Behavior is no apparent congruity between scratching and get- ting out and being fed. But when these acts have been performed a number of times in close sequence we cannot assume that they will not form a con- gruous association. The investigations of Pawlow and his co-workers have gone far to show that al- most any two acts performed at nearly the same time may become associated. The salivary reflex of a dog may be set up through having become as- sociated with sights, sounds, stimulation of the legs, or even the application of a painful electric stimu- lus. How the second act influences the first after the association has been formed is the important consideration, and we need not be surprised to find that certain responses are reinforced by acts of very diverse kinds. "Again," says Thorndike, "if a cat is put into a box, X, with two alleys opening to the North from it, A and B, and if whenever it advances two feet into alley A it is hit from behind with a club and so runs on out of the North end of A, whereas, if it advances two feet into alley B, it is given a piece of meat and hit gently from in front, the cat will, when put into X, be less likely to advance into A and more likely to advance into B. Yet the response of advancing into A produced the congruous sec- ondary response of advancing further in the same direction, whereas the response of advancing into B produced the incongruous retreat into X." I think the experiment, however, is capable of a Considerations on the Problem of Learning 151 different interpretation. When the cat advances into alley A, gets a blow from behind and gives the avoiding reaction, she naturally comes to associate this avoiding reaction with the sight of that particu- lar alley. This on a subsequent occasion would cause the cat to give the avoiding reaction upon the sight of A, and consequently prevent her from en- tering the scene of her former mishap. When the cat enters alley B, is given a piece of meat, and hit gently from in front and driven back, she is forced, it is true, to make, in one sense, an incongruous re- sponse, although she is not prevented thereby from entering B a second time. Thorndike does not con- sider the essential element of the situation accord- ing to the theory criticized as well as according to his own view, — namely, the reaction to the meat. Without this, the act of driving the cat out of the alley would probably have inhibited her further en- trance. The association formed between entering the alley and eating the meat makes the entrance to the alley, as it were, a part of the meat-eating reaction, and the first response is repeated through having been coupled with the discharge of this strong instinctive propensity. Those acts which elicit an animal's natural in- stinctive reactions are particularly prone to become associated with the latter, owing to the greater dis- charge of nervous energy which these instinctive re- actions involve. The acts which are stamped in are those which are consistent with the performance of 152 Studies in dnimal Behavior an instinctive activity which they have been the means of setting into operation. Where an instinct is very readily discharged, as in the case of the food- taking instinct of a hungry animal, acts which oc- casion this discharge tend to become quickly and firmly associated. We may be justified in saying that certain instincts, such as that of devouring food, have a quite gen- eral tendency to reinforce acts which bring stimuli that form the occasion of their discharge. On the other hand, injurious stimuli may tend to in- hibit quite generally other activities with which they have been associated. Animals are supplied with an elaborate equipment of avoiding reactions as a part of their congenital make-up, and as a result of the associations formed through experience they come to shun the things by which these avoiding reactions have been aroused. There is nothing a priori improbable in the assumption that animals may be congenitally endowed with connections in the central nervous system which would enable them to link up their various trial movements in ways which, on the whole, are serviceable. That such connections should exist is no more improbable than the occurrence of any other form of adaptive or- ganization. Profiting by experience in an animal of a primi- tive type of intelligence we may conceive, then, to take place as follows: The creature is endowed with the capacity for responding to beneficial stimuli Considerations on the Problem of Learning 153 by aggressive, outreaching movements, and to in- jurious stimuli by movements of withdrawal, retreat and avoidance. All these are matters of pure in- stinct. Given the power of forming associations between responses, the animal acquires new habits of action by repeating those responses which arouse instinctive acts of a congruous kind, and by dis- continuing those responses which arouse instinctive acts of an incongruous kind. Modifications of be- havior brought about in this way would, in general, effect a closer and more adequate adjustment of the organism to the stimuli that act upon it. The asso- ciations formed would be in the direction of re- fining the creature's instincts and adapting them to varied conditions of life. Intelligence so developed would be quite generally subservient to instinct, as in fact we find it to be, especially in its more rudi- mentary stages of development. The new things an animal learns to do are done because they have been assimilated to its instinctive activities. The interpretation of learning which we have been considering ascribes an important role to re- inforcement and inhibition. The essential nature of these processes, despite much investigation and speculation on the part of physiologists, still re- mains an enigma. If we thoroughly understood their nature and knew something more of the changes (whether in the synapses or elsewhere) by which associations are established, we should doubt- less be able to penetrate more deeply into the mech- 154 Studies in Animal Behavior anism of intelligent behavior. I would not be un- derstood as claiming that the present account of the learning process is entirely adequate. There are too many little known elements in the process to warrant anything more than a tentative adop- tion of any one standpoint. But the view defended does not postulate the existence of any physiologi- cal processes beyond those known to be commonly manifested in the physiology of the nervous system. Nothing is assumed in regard to the physiological correlates of pleasure and pain, the teleological be- havior of the neurones, or the nature of the changes on which associative memory depends. It is of importance, I believe, to determine how far we can explain intelligent behavior without multiplying en- tities by calling other factors to our aid. At pres- ent it is far from clear that it is necessary to make such an appeal. VIII THE IMPLICATIONS OF TRIAL AND ERROR ' I AHE concept of trial and error is one that has •*• played a prominent role in modern writings on comparatively psychology, and especially those which concern themselves with the problem of how be- havior comes to be adaptively modified. The activi- ties of animals must obviously be so shaped as to preserve the life of the individual or its race. To a certain extent successful acts may be hit upon by sheer accident, but for the most part the purposive behavior of an animal is due to its congenital make-up. According to the reflex theory of instinct, which commands a wide following at the present time, instinctive activities are fatally determined by structural mechanisms which are set going either by external or internal stimulations. In either case instinct, even in its most wonderful and complex manifestations, is, in essence, nothing but response to stimulus. And if we ask for an explanation of the teleological character of the responses we are referred to "inherited organization" for an answer. As the machine is constructed, so will it work. Whatever the merits or demerits of the reflex 155 156 Studies in Animal Behavior theory of instinct, it is evident that many kinds of adaptive behavior occur in which the stimulus is responded to, not by a direct, appropriate act, but by varied movements which apparently have little re- lation to the source or nature of the stimulating agent. This is well illustrated by the behavior of the protozoan Paramoecium which has been so ex- haustively studied by Jennings. Paramoecium, as it swims through the water, rotates about its long axis and describes more or less of a spiral path. When it encounters an object or receives a sudden stimu- lus of any sort it reverses the beat of its cilia, swims backward, turns to the aboral side and then con- tinues on its way. It matters little on which side Paramoecium is stimulated; its reaction, which Jen- nings has called the "motor reflex," is practically the same. Sometimes the motor reflex brings the animal into closer contact with the stimulating agent. If so, the reaction is repeated, and continued so long as the animal is stimulated by the unfavorable conditions. Paramoecia, like many other simple organisms, form aggregations in regions of dilute acids. They are not drawn to these chemicals from a distance, but if a Paramoecium happens to swim into the area of dilute acid it passes through to the boundary between the acid and the surrounding water where it gives the motor reflex and swims in another di- rection. If it again encounters the boundary it gives the motor reflex and swims away, so it is prac- The Implications of Trial and Error 157 tically caught in a sort of trap. Other Paramoecia happening to enter the area are similarly caught, so that an aggregation of individuals is gradually formed. There is no orientation of the Paramoecia to the diffusing chemical. The acid does not attract the animals in any sense. The infusorians simply give the motor reflex when they swim from acid to water, and as a consequence of this simple re- action a collection of individual results. Jennings has shown that Paramoecium responds by the motor reflex to nearly all kinds of stimuli. Practically its whole behavior is based on this trial and error method. The organism does not directly avoid injurious stimuli. The motor reflex may or may not relieve it from the stimulating agent. But it keeps on repeating the reaction until in time it chances to be brought away from the injurious stim- ulation. When favorable conditions occur, its be- havior continues unchanged. The Paramcecium's philosophy of life is very simple. It consists in merely following the Pauline injunction: "Prove all things; hold fast that which is good." The same general scheme of reaction Jennings found to be very widespread among the lower or- ganisms. Although the reactions of the organisms may be simple and stereotyped, this scheme of re- sponse gives to the behavior of these forms a plas- ticity and adaptiveness that keep them away from injurious stimulations and in conditions favorable for their existence. Where there is "error," the 158 Studies in Animal Behavior organism tries again, and keeps on doing so until it attains ultimate success. "Error" generally means any act prejudicial to organic welfare. The lower organisms are like ourselves in avoiding things which are injurious and in remaining under beneficial conditions, whether or not they are influenced thereto by similar psychic states. Jennings feels "compelled to postulate throughout the series certain physiological states to account for the negative reactions under error, and the positive reactions under success," but the search for such states, as I have elsewhere attempted to show, is probably a vain quest. In behavior of the trial and error type, success is attained, not by a direct adaptive reaction, but by checking or reversing all reactions except the right one. The final outcome of the varied move- ments is adaptation. The method is roundabout and expensive, but it is better than nothing. It is Nature's way of blundering into success. The capacity to gain anything through the method of trial and error presupposes that an animal's re- actions to beneficial stimuli are different from its reactions to injurious ones. There is, as I hope to make clear, no way of acquiring this capacity unless there is a congenital basis of adaptive re- sponse to start with. We are led to the conclusion that the adaptive character of the indirect adjust- ments effected through the method of trial and error is, like the adaptiveness of direct instinctive The Implications of Trial and Error 159 responses, the outcome of inherited organization. This conclusion applies to intelligent behavior as well as to the more primitive forms of indirect adjustment. That the burnt child dreads the fire depends upon the fact that there is an innate re- flex tendency to jerk back the hand when it comes in contact with a hot object. We have therefore a bit of primary adaptive responsiveness to start with. Experience links up the sight of the object with this primary reaction. Association -per se has nothing teleological about it, but it is a means of effecting further adjustments, or rather perhaps of applying to new conditions the adaptive responses already present. If an organism were constituted so as to respond to all stimulations in a perfectly hit or miss fashion, without any reference to its own welfare, it is not likely, even if the creature had the power of forming associations, that it would be able to profit by experience. Suppose, for in- stance, it should react to the sight of an object by an act of seizure. Suppose that contact with the object, which we will suppose in this case to be food, should evoke an avoiding reaction instead of the usual purposive movements. The sight of the object becoming associated with the avoiding reaction would cause an inhibition of the first re- sponse by calling into play an antagonistic re- action. If food were responded to by an avoiding reaction, the associations acquired by experience, while they might modify behavior, would not help 160 Studies in Animal Behavior the animal out of its unfortunate situation. Merely associating experiences is of no particular value. There must be some principle of selective associa- tion if experience is to be turned to any account, and this principle is supplied by the animal's stock of congenitally adaptive reactions. What makes in- telligence of any value to its possessor is its ingre- dient of primary purposive responsiveness. With- out this ingredient, which is the real controlling hand in an animal's life, behavior would be a mere chaos of misdirected activity. It is really instinct that makes intelligence useful. If the adaptiveness of intelligence rests upon the adaptiveness of instincts and reflexes, and if the latter is determined by inherited organization, we must look to the forces that have moulded organi- zation, i. e., the factors of evolution, for the pri- mary source of adaptiveness in behavior. Aside from the very doubtful role of the Lamarckian fac- tor, we have at present no way of explaining how purposive organization can evolve, except through the operation of natural selection. Given varia- tion (which may be quite fortuitous), struggle for existence, and the survival of the best endowed, adaptive organization will be the outcome. Natu- ral selection is itself a sort of trial and error process, a method of getting a successful product out of vari- ability which, so far as adaptation is concerned, there is no reason to believe is other than of a ran- dom, hit or miss character. Whether the princi- The Implications of Trial and Error 161 pie of selection will account, directly or indirectly, for whatever there is of purposiveness in organiza- tion and behavior is of course an open question. But there is no other factor which is certainly opera* tive in shaping descent along adaptive lines, and as entities are not to be multiplied beyond necessity," it is, I believe, justifiable to adhere to th& theory of selection until its inadequacy is clearly established. The position to which we are led is that the se- curing of any advantage through the method of trial and error presupposes congenital modes of response which are adapted to secure the welfare of the in- dividual. The method is not the primary source of adaptive reactions so far as the individual is concerned. It cannot be the primary source of adap- tive behavior in the evolution of the race. A method of blundering into success instead >of attaining v it directly, it would be of no service unless the or- ganism were capable of turning to profit its fortu- nate trial movement. In order to do this the or- ganism must be provided for the situation by its inherited endowment. Certain writers on -genetic psychology have at- tempted to explain the beginnings of adaptive re- actions on the basis of individual experience alone. Given an organism which just responded to stimuli in a perfectly random manner, they attempt to show how the environment would discipline it into acting in accordance with its own welfare. Aside from those functions which Jensen has included under the 1 62 Studies in Animal Behavior primary purposive attributes of life (primare Zweckmassigkeit) , and which must be presupposed if an organism is to be an organism at all, we nat- urally suppose that during the early periods of evolution there has been a gradual substitution of adaptive for non-adaptive modes of behavior. If the new adaptations arise in and through experi- ence they must be transmitted by inheritance if they accumulate to the advantage of the race. But aside from the difficulties which beset the theory that such acquisitions are inherited, there is involved the fur- ther difficulty of understanding how, in the begin- ning, adaptations could have been acquired at all. To this the Lamarckian might of course reply that it is a fact that adaptations do arise through in- dividual experience, however we may account for them, and that it is not especially incumbent upon him, qua Lamarckian, to explain their origin. There is no gainsaying the pertinency of the Lamarckian's answer as regards the origin of many purposive forms of behavior. If, however, it can be shown that the ability to acquire special adaptations rests upon innate modes of reacting to stimuli, and that in so far as acquired characters are adaptive, their origin presupposes the existence of congenital adap- tive reactions, the Lamarckian can no longer dodge the responsibility of explaining how these congeni- tal reactions came to be of service to the organism. Unless the characters acquired by experience were useful their accumulation through inheritance would The Implications of Trial and Error 163 obviously be of no value. Lamarckism practically rests upon the assumption that an organism's nat- ural modes of response are teleological. But it gives no account of how they came to be so. Can an organism that responds totally without regard to its own welfare, if such an organism can be imagined, ever be hammered into an adaptively reacting mechanism through the influence of the en- vironment? Spencer's and Bain's theories of the origin of adaptive movements practically assume that it can. Both Spencer's and Bain's theories assume that adaptation is brought about by a process of selection from among the various random activi- ties performed by the organism. It is th"e acciden- tally adaptive movement that comes to be repeated. Both theories assume that the adaptive movement is repeated because it brings pleasure to the organ- ism. Hence there must be a correlation not only between pleasure and welfare, but between pleasure and the tendency to repeat an action by which pleas- ure is secured. To the question as to how pleasure came to be associated with organic well being, Spen- cer falls back upon natural selection for an answer. Organisms in which the pleasurable coincided with the organically beneficial were preserved, while the others perished, so there gradually came to be es- tablished the general relationship between the pleas- ant and the beneficial which we commonly find. So far as I know, no other plausible account of this relationship has been offered. The Lamarckian ex- 164 Studies in Animal Behavior planation of the origin of adaptive movements which Spencer himself adopts has to work with re- lationships which, as Spencer himself admits, have been established by natural selection. It is the con- genital make-up of the organism that determines which of its various random movements comes to be selected. In the absence of the proper hereditary endowment the organism might make random move- ments all its life without securing any desirable re- sult. No one has yet succeeded in showing how a bit of undisciplined protoplasm is able to acquire any form of adaptive behavior. We are brought to the conclusion that the abil- ity to profit by experience, both in its higher mani- festations as intelligence and in the simpler forms of organic adaptation, involves an organism moulded into at least partial conformity with its environment. The activities that are described as "trials" afford increased opportunities for adjust- ment, but the ability to take advantage of success- ful trials rests upon the basis of congenital endow- ment. It is inheritance that affords the means by which inheritance is improved. REFERENCES BAIN, A. The senses and the intellect. 3d ed., 1894. BALDWIN, J. M. Mental development. Meth- ods and processes. 2nd ed. N. Y., 1903. References 165 JENNINGS, H. S. The behavior of lower organ- isms. N. Y., 1906. JENSEN, P. Organische Zweckmassigkeit. Jena, 1907. SPENCER, H. Principles of Psychology. N. Y., 1892. TORREY, H. B. The method of trial and the tropism hypothesis. Science, N. S., 26, 313, 1907. IX BEHAVIOR AND FORM * I AHE idea has been recently emphasized that in •*• many cases the structure of an organism is, to a considerable extent, the effect of its behavior. We have, of course, long been familiar with the fact of functionally produced modifications, but it is only of late that behavior has been brought forward as a factor of importance in development, regenera- tion and other modes of form regulation. If the way in which an organism acts is determined by its structure, it may also be said that its structure is to a certain degree determined by the way in which it acts. The experiments of Child on the regeneration of planarians, nemerteans, and other forms have led him to the view that one very important ele- ment determining the way in which a part differen- tiates is the kind of movements it is called upon to perform during the period of its formation. A posterior cut end of a planarian, for instance, de- velops a tail because owing to the movements of the animal, the regenerating tissue is subjected to the same impacts and tensions to which the tail end of the entire animal is normally subjected. These 1 66 Behavior and Form 167 activities impressed upon the regenerating part are regarded as the directive agents in the process of dif- ferentiation, and if the animal were to act in a very different manner quite different formative processes would result. Child found that in pieces of Leptoplana which were caused to creep in a circular course owing to unilateral injury of the anterior end the new tail that was formed, instead of growing out posteriorly, developed in an oblique position in accordance with the altered activity. "The regenerating part grows in the direction of the principal tension, even though this forms an angle of 90° with the principal direc- tion of growth." In speaking of the ordinary activi- ties of Leptoplana, Child says : "It can scarcely be doubted that these movements play a part in shap- ing the regions in which they occur. A comparison between frequency, amplitude, and force of the un- dulating movements and the degree of lateral de- velopment in the regions in which they occur is most striking. According to the usual point of view, this correlation between structure and function is merely one of the many remarkable cases of adaptation, but in my opinion it is, at least in part, the direct result of function in the individual. . . . "As regards the manner in which the movement may affect the tissues it is not difficult to see that the movement of these parts to and fro through the water must subject them to tension in the lat- eral direction. This must affect in greater or less 1 68 Studies in Animal Behavior degree the distribution and arrangement of the plas- tic tissues composing the parts. A very simple phys- ical experiment serves to illustrate this point. A cylindrical or square stick of sealing wax moved to and fro in water sufficiently warm to soften it will undergo flattening in a plane at right angles to the direction of movement. The change in form is more strikingly shown if a rigid axis is present; a mass of wax molded in cylindrical form about a stiff wire will become in a few minutes a thin flat plate de- creasing in thickness towards the edges and with a rounded outline. The mechanical conditions re- sulting from the movement of the wax through the water are not widely different from those which the undulating margins of Leptoplana produce. If the wire axis of the wax be considered as the longi- tudinal axis the effect of movement through the water is lateral extension. In Leptoplana the un- dulating movement is confined chiefly to the lateral regions in the anterior third of the body and it follows that the conditions described are limited chiefly to these parts." "There can be little doubt, in my opinion, that these mechanical conditions constitute a factor in the formation of the broad lateral regions in Lepto- plana and more especially in other forms in which the undulating movements of these parts occur. In other words, the form is in some degree the result, not the cause, of the characteristic method of ac- tivity." Behavior and Form 169 In experiments on Planaria Child has shown that the head and pharynx do not attain their normal shape and structure when movement is largely in- hibited by anaesthetics, but he does not conclude that movement is the only factor involved. Regenera- tion and all other modes of the regulation of or- ganic form he regards as the outcome of functional regulation. Movement plays a certain role in shap- ing the outline of some organisms, especially those with physically plastic tissues, but "it is merely one of a great variety of functional factors." While studying the regeneration of the infuso- rian Loxophyllum I found that I was dealing with an organism in which regeneration and behavior are apparently closely connected. Loxophyllum is a flattened infusorian that moves by gliding on the bottom on its right side. It confines its activities usually to a small area for a considerable time; first it glides forward a short distance, then reverses its cilia and swims backwards, turns toward the oral side and then swims forward again in a new direction. As the animal swims forward the body is elongated, but as it goes backward the body is invariably shortened and widened, thus showing a constant association between the direction of the beat of the cilia and the contraction of the myo- nemes. In the changes in the form of the body it is the contraction and extension of the oral side that play the most important part, and it is along the oral side that the myonemes are especially 170 Studies in Animal Behavior abundant and of large size. In going forward the body bends orally and frequently undulates about in a very variable manner. The backward and for- ward movements of Loxophyllum may be performed in much the same way for a long period apparently without any external stimulation. The organism is capable of many kinds of behavior under different conditions, but its usual activities are all that con- cern us in the present connection. As has been found in other infusorians, the be- havior of pieces into which the organism is cut is closely similar to that of the entire individual. The pieces of this species regenerate so rapidly as a rule that one can actually watch the process going on. An excellent opportunity is thus afforded for study- ing whatever connection there may be between the activities of the piece and the changes of form that occur in it. In the regenerative changes of the posterior half of the body, for instance, we find that apparently as a consequence of the elongations and contractions that accompany the forward and backward move- ments, the general form elongates and becomes more narrow. The cut end is partially closed in by the drawing together of the two margins, the oral side extending over rather more than the aboral (Fig. i). In the movements of the animal it may read- ily be seen that the oral side is pushed ahead as the animal swims forward as if it were making active efforts to stretch itself out. Both margins, how- Behavior and Form 171 ever, take part in this active extension, but the oral side is the more active and soon it may be seen to push around the anterior end of the body, and its myonemes and cilia are thus brought into the characteristic curvature which is found at the an- terior end of the entire organism. Fio. 1. — Regeneration of posterior piece of Loxophyllum. The dotted line indicates the direction of the cut. ' The figures 2, 3, 4, etc., representing the successive stages of the process, show the gradual stretching of the ciliated margin, m, mouth. If the cut is made obliquely (Fig. 2) so as to limit considerably the extent of the oral side, the same method of regeneration is followed, the oral side is gradually stretched out so as to form the oral side and anterior end of the new individual. In the process of regeneration here followed the 172 Studies in Animal Behavior final form of the organism is reached by the sim- plest and most direct means, — there is a minimal amount of formation of new structures. The cut end does not differentiate any new organs, but the old differentiated parts are stretched out to form the structures of the new individual. It may read- ily be seen that the activities of the organism in FIG. 2. — Regeneration of Loxophyllum after being cut, as shown by the two dotted lines. The small part of the ciliated margin that remains is stretched but a short distance, while new cilia are formed along the new oral margin. extending, contracting the body and in constantly pushing the oral side farther ahead than the aboral, tend to mould the piece into the normal form of the species. In other words it might be said that the organism pulls itself into shape. The piece gets pulled into the form of the entire animal be- cause it behaves essentially like the entire animal. So far the results seem to support the view that form is the result of activity, at least to a consid- erable degree. But instructive developments are Behavior and Form yielded by further experiments. By making a cut so as to remove all the oral margin with its dif- ferentiated structures we compel the organism to follow a quite different method of regeneration (Fig. 3). At first, owing to the absence of the more contractile and extensile elements of the oral FIG. 3. — Regeneration of Loxophyllum after removal of entire oral margin. side, the aboral margin in the movements of the animal are pushed farther ahead than the oral. The organism elongates and becomes pushed over toward the oral side. Later, however, the oral side becomes thinned out and more transparent; cilia make their appearance in scattered groups, new trichocysts are developed, and contractile threads appear in the thinned out region. After these vari- ous structures are developed the oral side begins to extend and contract more than the aboral. Then it becomes pushed ahead more and more, and finally 174 Studies in Animal Behavior is carried around the anterior end of the body as in the normal animal. In this experiment the animal is forced to follow a method of regeneration very different from that pursued in the experiments previously described. The old differentiated parts are not stretched out to form the new ones, but the structures of the oral side of the new individual are formed de novo. It is an interesting fact that the time required for regeneration in the first case is short, the process being completed in a little over an hour, while in the latter case it was not completed until about fifteen hours. In the latter experiment the charac- teristic behavior of the oral side had to wait upon the differentiation of the oral region of the body. Where regeneration is apparently produced largely by the organism pulling itself into shape through its characteristic movements behavior seemed to lead the way in moulding the normal form. In other ex- periments the characteristic movements had to wait until the finer differentiations of the organism were developed by other factors; then behavior stepped in to help mould the already differentiated organ- ism into its normal outline. I think that the foregoing experiments (and I have described only a small part of the number that were made) indicate that the more fundamental factors in the regeneration of this organism are not so much its gross activities as various internal factors which we need not here attempt to specify. At the same References 175 time the behavior of the infusorian is a link in the chain of causes by which the final form is brought about, but it is concerned more in determining the general shape of the body than the finer details of its internal structure. What is true in this case has, I believe, a rather general application. The gross general behavior of an animal plays, I believe, a subordinate though at times an important role in the determination of organic form. But when we turn our attention to the internal processes which are responsible for the finer details of organization may we not again en- counter problems of behavior? May not the dif- ferentiation of these parts be, to a considerable de- gree at least, the result of the behavior of their com- ponent elements? The study of differentiation in higher organisms where it is possible to follow the activities of the various cells which make up the body makes it evident that such is the case. The role of this factor will be considered in the chapter on The Behavior of Cells. REFERENCES CHILD, C. M. (i) Studies on regulation. IV. Some experimental modifications of form regula- tion in Leptoplana. Jour. Exp. Zool. i, 95, 1904. (2) The regulatory change of shape in Planaria dorotocephala. Biol. Bull. 16, 277, 1909. (3) 176 Studies in Animal Behavior Analysis of form regulation with the aid of an- aesthetics. 1. c. 16, 277, 1909. HOLMES, S. J. Behavior of Loxophyllum and its relation to regeneration. Jour. Exp. Zool. 4, 399> X THE BEHAVIOR OF CELLS OINCE the epoch-making promulgation of the ^cell theory by Schleiden and Schwann biologists have commonly looked upon the component cells of the body more or less in the light of little organ- isms, each with its own individuality, each carry- ing on the business of its life to a certain degree independently of its fellows. Haeckel speaks of a cell soul which he regards as a quasi-discrete bit of psychic life. Binet in his Psychic Life of Micro- organisms writes of the various psychic faculties manifested by the white blood corpuscles and other cells of our bodies. And Lloyd Morgan in his book on Animal Behavior devotes a section to the be- havior of cells and shows how various formative processes are to a considerable extent the outcome of coordinated cell activities. The study of cell behavior has been greatly stim- ulated by the development of a branch of biological investigation, variously known as developmental mechanics, physiology of development, etc., which has for its aim the analysis of embryonic develop- ment, regeneration, and other formative activities, — -in a word, all those processes which lead to the 177 178 Studies in Animal Behavior production and maintenance of the normal form of the organism. The science of developmental me- chanics is of very recent growth. Its already numer- ous devotees endeavor to gain a deeper insight into the nature of development than it is possible to obtain by the older methods of descriptive embryol- ogy. However full and exact our knowledge may be regarding the cleavage of the egg, the formation of the germ layers, and the differentiation of or- gans; however thoroughly we may come to know the sequence of events which lead from the egg cell to the adult animal, such knowledge does not neces- sarily reveal the threads of causal connection that govern the course of development. The develop- ment of any organism is a wonderfully complex process. We may describe what goes on with the greatest fidelity to every detail, but the causal rela- tions are so involved that we can seldom discover them without special methods of analysis. The sci- ence of developmental mechanics set itself the task not merely of describing what takes place but of explaining why it takes place. Description it re- gards as subsidiary to explanation. Having for its aim the search for causes, it naturally adopts ex- perimentation as its chief method. Developing eggs are shaken apart, subjected to heat and cold, to chemicals and a great variety of other external agents; particular parts of the embryo are removed, displaced, or replaced by other parts, and all sorts of modifications of development are induced in or- The Behavior of Cells 179 der to discover something of the relations of de- pendence subsisting between the various activi- ties which lead to the formation of the embryo. When one watches the early development of the egg of a mollusk or annelid worm, follows the reg- ular and almost mathematically precise way in which cell division occurs, and the method by which the cells become arranged in a perfectly regular pat- tern, then observes the infoldings and overgrowth that lead to the gastrula stage, and the differen- tiation of particular cells to form the organs of the embryo, he cannot escape a feeling of wonder that a bit of simple material, such as the undivided egg appears to be, contains such powers of elaborate and well-ordered construction. A marvellous builder this bit of egg substance ! It is comparatively easy to make a sort of rough catalogue of the meth- ods it employs. Development may be said to be an affair of cell division, cell growth, cell differentia- tion, changes in the form of cells, changes in their position, etc.; but this cataloguing of processes is a mere preliminary to the business of further analy- sis. In order to build up an embryo, cells must di- vide, they must in most cases grow, and they must get into the right relative position and differenti- ate, some in this way and some in that, so as to es- tablish a harmoniously working mechanism. In most organisms formative processes involve a con- siderable amount of cell movement. Sometimes this is passive, as when cells are pushed or pulled by 180 Studies in Animal Behavior means of osmotic changes or growth processes occur- ring in other parts. But it is coming to be recognized more and more that many of the processes of embry- onic development are the result of active cell move- ment. Many cells in early development have a con- siderable power of locomotion, and, like the soldiers of a well-disciplined army, they move into their ap- pointed places at the proper time. To gain an in- sight into this feature of their development, to un- derstand why it is that these cells act as they do, we are naturally led to a study of cell behavior. To a considerable extent the form of our bodies is an expression of shall we say cell psychology? Or, at any rate, it is in part the expression of that kind of behavior which is embraced under comparative psychology when found among lower organisms. A few years ago Wilhelm Roux, an investigator of especial prominence among the experimental em- bryologists, found that if the cells of the frog's egg are shaken apart during early stages of seg- mentation, and placed in water a short distance apart they would slowly approach one another until they came into contact. This peculiar behavior, which was later observed also by Rhumbler, was called by Roux Cytotropism. Whether it is a form of chemotaxis, or just how it is brought about is un- certain. Cells of the early cleavage stages also show a marked tendency to come into contact with one an- other as closely as possible. While they round up during division, during the resting stages the cells The Behavior of Cells 181 fit together so as to leave no spaces between them. This trait not only tends to keep the cells in a com- pact mass, but it has other important functions as we shall see later on. The cells of the embryonic tissue called mesen- chyme have long been noted for their powers of mi- gration. These cells are usually irregular and changeable in outline, and are able to creep about much like an Amoeba. Usually they form masses lying between other cell layers. Their origin is varied, and they frequently move to a considerable distance from their source. An excellent illustra- tion of their behavior is afforded in the early de- velopment of the sea urchin. In the region of the hollow blastula which is being pushed in to form the primitive gastrula cavity there is given off into the interior of the embryo a number of amoeboid cells. These wander away from their point of or- igin and take up positions on opposite sides of the archenteron, where they form two groups. Why do the mesenchyme cells take up their position in these particular places? Driesch found that by vig- orously shaking the young embryos the mesenchyme cells became loosened and scattered about irregu- larly in the cleavage cavity. After the embryos had been left for some time, however, it was found that these mesenchyme cells were back again in their nor- mal position as if they knew what was their ap- pointed function and took up their station accord- ingly. We may conjecture that it was some chemo- 1 82 Studies in Animal Behavior tactic influence that drew these cells to certain re- gions of the embryo much as white blood corpuscles are attracted to substances produced by certain bac- teria; or possibly it may have been some form of reaction to contact stimuli ; but, anyway, the result is of interest in showing the role played by cell migra- tion in establishing the structure of the embryo. The mesenchyme cells of the embryo are the parents of most of the connective tissue cells of the adult organism. The latter have long been known to be more or less migratory under certain condi- tions, and the recent work on the cultivation of tissues outside the body has shown that they pos- sess considerable power of amoeboid movement. When a piece of tissue, especially from an embryo or young animal, is mounted in a hanging drop of blood plasma there soon appears around it a ring of more or less spindle-shaped cells radiating into the surrounding medium. These spindle-shaped cells are derived mostly from connective tissue, and the ring that is formed results chiefly from the out- wandering of cells, although in some cases the growth and multiplication of cells undoubtedly con- tribute to its formation. The writer has often watched the movements of these cells under the microscope. The locomotion is essentially amoe- boid, and the cells show a pronounced thigmotaxis, or tendency to keep in contact with solid bodies. The pigment cells of many animals are related to connective tissue cells both in origin and mode of The Behavior of Cells 183 behavior. In many crustaceans, fishes, amphibians and reptiles the pigment cells appear at times to be richly provided with branched processes, and at other times nearly all the pigment appears to be concentrated in a rounded mass. These changes often produce marked changes in the color of the animal. They are to a certain extent under the control of the nervous system, but they may also take place independently of nervous influence. The writer has succeeded in isolating pigment cells from FIG. 4. — Successive changes in the form of an isolated pigment cell from a frog. various amphibian larvae and also from the adult frog and in keeping them alive in hanging drops of blood plasma where their various changes in form could be followed with the greatest readiness. The pigment cells were seen to undergo changes in form similar to their changes in the skin, putting out processes here, drawing them in there, and in some cases creeping along the cover glass for a consid- erable distance. These cells, like their relatives, the connective tissue cells, have a strong positive thig- motaxis. Loeb found that in the embryos of the fish Fundulus they tend to appear along the course of 184 Studies in Animal Behavior the blood vessels of the yolk sac, thereby producing a certain color pattern characteristic of the species, a result probably due to some tropism, either a chemotaxis toward oxygen, or a thigmotactic re- action to the walls of the blood vessels. Both connective tissue cells and pigment cells when isolated tend to withdraw their processes un- der unfavorable conditions and assume a spherical form. Too high a temperature tends to cause this response, and the same reaction commonly results when the culture medium becomes contaminated by the accumulation of metabolic products. An Amoeba under similar conditions behaves in essentially the same way. As is well known, the cells of epithelial tissues are almost always arranged in definite layers one or more cells thick. Such tissues are found cover- ing the entire surface of the body, and lining all the inner surface of the alimentary canal and its various appended organs; epithelium forms the in- ner lining of the body cavity, the blood vessels and lymphatics; in fact, in almost every situation in which a free surface occurs it is covered by a layer of this tissue. If for any reason a part becomes denuded of its coating of epithelium it is usually rapidly cov- ered again by extensions of this tissue from con- tiguous areas. There have been various opinions as to just how this extension is brought about. Sev- eral investigators have concluded that it is mainly effected by the migration of epithelial cells, and the The Behavior of Cells 185 recent work of Leo Loeb, Oppel and Harrison tends strongly to confirm this view. A particularly favorable method of studying the problem is presented by the cultivation of epithelial cells in some nutrient medium outside the body. The writer has employed this method in the study of the epithelium of larval and adult amphibians. It was found that a piece of epithelial tissue kept in a hang- FIG. 5. — Strands of epithelial cells extending along and between fibers of cotton, tt and t' t'. The dark part represents the outgrowth from a piece of a frog tadpole cultivated in lymph. ing drop culture soon showed a fringe of flattened cells extending into the culture medium. In some cases the area of the extending tissue was over twenty times that of the implanted piece. Exam- ining with a high power of the microscope the mar- gin of the epithelial extension, numerous very fine processes or pseudopodia were seen extending in the direction of movement. These processes were seen to be in constant change. They are the active agents in causing the sheet of cells to be pulled out, for they are both adhesive and contractile. The effect of i86 Studies in Animal Behavior their activity is to cause epithelium to spread more and more widely, and in several preparations that were made, practically all of the available solid sur- face was covered by an investment of epithelium. If individual cells are isolated they frequently spread in every direction until they become flattened to an extreme thinness, and I have observed scattered individual cells flatten out until the adjacent cells FIG. 6. — Successive stages in the advance of a strand of epithelial cells from a frog tadpole. This represents on a more enlarged scale the end of the strand lying between the two fibers of cotton t and t' shown in Fig. 5. The dotted line indicates a fixed position. met, forming a continuous membrane, scarcely dis- tinguishable from certain epithelial membranes found in the body. The marked thigmotaxis of these cells leading them to spread over surfaces and to keep in close contact with one another is a very im- portant factor in bringing about the layered arrange- ment characteristic of epithelial tissue, as well as in causing this tissue to form a coating over the ex- posed surfaces of nearly all organs of the body. Epithelial membranes usually form surfaces of se- cretion and absorption which separate regions of The Behavior of Cells 187 different osmotic pressure. It is essential to the per- formance of many important functions that these membranes be absolutely continuous with no aper- tures at any point. This continuity is insured through the traits of behavior we have mentioned. FIG. 7. — A secondary membrane formed by the approximation of originally isolated cells from the ectoderm of Hyla, such as shown in B. Very fine pseudopodia are shown in the free margins. One very general and important feature of bod- ily organization therefore is traceable largely to certain peculiarities of the behavior of cells. In many animals the sex cells arrive at their final posi- tion only after considerable migration. In the sponges they are derived from amoeboid cells and in hydroids they often migrate from one germ layer to 1 88 Studies in Animal Behavior another. In the eggs of many insects where they are differentiated very early in development they have been observed actually to pass outside the embryo for a time only to wander back again. Hegner has performed the experiment of removing these cells at the time they passed out of the egg and found that no sex cells developed in the resulting ani- mal. In the embryos of several of the -lower ver- tebrates Dr. B. M. Allen has found that the sex cells arise in the entoderm. Then they wander out along the rudiment of the mesentery and take up their final position at a considerable distance from their point of origin. As to what causes the movements of these cells and guides them in their course we can only offer but very tentative conjectures. One of the most remarkable features of develop- ment is the way in which nerve fibers grow out from the central nervous system and push their way between various masses of cells to become con- nected with the proper end organs. The organism is permeated with a network of these fibers, spun out as fine as gossamer threads and connected with one another and with various parts of the organism in a very intricate way. In order to build up this elaborate and delicate system the nerve fibers as they grow out from the developing brain and spinal cord must take the right paths and connect par- ticular groups of cells in the central nervous sys- tem with corresponding parts of the body. Should nerve fibers not find their proper termination all The Behavior of Cells 189 sorts of functional disturbances would result. What is it that guides the nerve fibers to their proper goal? One widely accepted theory of the development of the nerve fiber is that it represents an outgrowth from the nerve cell. Harrison by cultivating in hanging drops of lymph small pieces of the nervous tissue from amphibian embryos has been able to observe the actual outgrowth of nerve fibers from the ganglion cells, and to study various stages in the process of outgrowth in living material. Har- rison observed that the extreme tip of the outgrow- ing nerve fiber was often expanded and irregular in contour, and that it underwent amoeboid changes as the fiber extended through the lymph. It is not improbable that this amoeboid activity is the cause of the drawing out of the nerve fiber.1 The factors, chemotaxis, thigmotaxis, or whatever they may be, which direct this amoeboid activity may be respon- sible therefore for the distribution of the nerve fibers in the body and the establishment of the nu- merous connections that occur within the nervous sys- tem itself. The architecture of the nervous sys- tem, the great controlling element in behavior, is not improbably itself a product to a large extent of the peculiar behavior of its component cells. 1 That the nerve fiber is actually drawn out in the way sug- gested has recently been shown by Mr. J. C. Johnson in his studies of the nerve cells of amphibian larvae. Reference may be made to his paper on "The Cultivation of Tissue of Am- phibians," published in the Univ. of Calif. Publ. Zool. Vol. 16, P- 55, 190 Studies in Animal Behavior Among the most motile of the cells of the body are the leucocytes or white blood corpuscles. These cells are very much like Amoebae in their appear- ance and their activities. They have the property of creeping about in the various spaces of the body and of passing through the delicate walls of the capillaries. Their power of engulfing bacteria and fragments of broken-down cells is well known. They act not only as the scavengers of the body, but by virtue of their power of destroying bacteria they defend the body against various disease germs that constantly invade it. Apparently they are drawn to centers of bacterial infection by^a sort of chemo- taxis. This is shown by an ingenious experiment by Massart in the following way: A tube of cul- ture medium containing a culture of the bacterium Staphylococcus pyogenes albus was introduced into the abdominal cavity of a rabbit. After a time it was found that the leucocytes had wandered into the tube in large numbers. A similar tube filled with the same medium but containing no bacteria was also introduced, but it was not entered by the leu- cocytes. It is probable, therefore, that some sub- stances produced by the bacteria caused the leu- cocytes to enter the tube. The species of bacterium used in the experiment is one of the common forms that give rise to the production of pus. This substance which so fre- quently gathers in inflamed areas is produced mainly by degenerated leucocytes which have accumulated The Behavior of Cells 191 at the seat of injury. The aggregation of these wandering cells may actually be observed under the microscope in the transparent living mesentery of the frog. If a part of the mesentery is drawn out of the living animal and stretched over the stage of a microscope the blood may be observed stream- ing through the capillaries, and the individual cor- puscles distinctly followed in their course. If a re- gion near a capillary is pricked with a hot needle the white corpuscles may be seen to pause in their course as they arrive near the injured area and pass through the capillary wall. Here again it is probably some chemotactic proclivity that causes the leucocytes to congregate. Combined with the power of these cells to devour bacteria this chemo- tactic tendency enables the leucocytes to play the part of watchful protectors in checking infections and destroying products of decay. Besides their role in maintaining the normal ac- tivities of the body the leucocytes often play an important part in development. In the formation of an organism the tearing down of previously es- tablished structures is often a necessary preliminary to further advance. The resorption of the tail of the tadpole is a process of involution in which the leucocytes play an essential role in destroying the fragments of the degenerating structure. The or- gans of the caterpillar are very largely destroyed in the quiescent pupa state in which the reorganiza- tion of the body occurs that results in the forma- 192 Studies in Animal Behavior tion of such a very different creature as the but- terfly. Here cells much like the leucocytes act as devouring cells or phagocytes which engulf the ma- terials of the degenerating tissues. Apparently th^ leucocytes are omnivorous in their appetite. One of my former students, Dr. Fasten, who has made a thorough study of the behavior of these cells, finds that they engulf even such sub- stances as sulphur and chrome yellow, which would be rejected by an Amoeba or almost any other free organism. This indiscriminate appetite would be fatal to a creature living a free life. Probably it is not particularly good for the leucocytes; but it must be remembered that the role of these cells is primarily altruistic. They work for the welfare of the body physiological. They are reproduced not so much from other leucocytes (although this proc- ess occurs) as by the division of cells in bone mar- row and certain other organs of the body. They are meant for sacrifice after a life of service. And Nature has made them rather more than usually unmindful of their individual welfare. However these cells do show many of the pur- posive reactions so commonly found in free organ- isms. Dr. Fasten by a delicate apparatus has suc- ceeded in bringing the point of a very fine glass rod against one side of the cell. The leucocyte thus irritated put out pseudopodia opposite the point of the rod and crept away from the stimulus. The method of response was practically the same as that The Behavior of Cells 193 of an Amoeba under the same conditions. Like Amoeba also the leucocytes were found to be nega- tive in their response to light, and when subjected to too high a degree of heat they withdrew their pseudopods and assumed a rounded form. It would be of interest to ascertain how the vari- ous other moving cells react to different kinds of stimulation. I have tried the effect of localized mechanical stimulation, light, and heat on pigment cells, epithelial cells and cells from connective tis- sue, but found no reaction to light, no marked ten- dency to crawl away from injurious mechanical stim- uli, but a general tendency to round up under too high a degree of temperature. The freely wan- dering tissue cells do not appear to possess that repertoire of adaptive responses which the leuco- cytes have in common with the Amoeba. Probably fuller investigation will reveal more adaptive be- havior in the cells of many forms. We know as yet next to nothing in regard to this field of inquiry, but it is one which promises to afford interesting results. In that fascinating group of primitive organisms, the slime molds, formative processes stand in a more obvious relation to cell behavior than in the higher organisms. The coming together of individual cells, their union to form a creeping plasmodium like a gigantic Amoeba, the transformation of this plas- modium into a definite form characteristic of a par- ticular species, — all these processes are for the most 194 Studies in Animal Behavior part matters of the behavior of cells. If we can interpret development anywhere in terms of the tropisms and other responses of individual cells this peculiar group of organisms would seem to present unusually favorable opportunities for attacking the problem. In a very interesting series of experiments Prof. H. V. Wilson has cut various sponges and hydroids into pieces, pressed the tissues through fine bolt- ing cloth so as to reduce them practically to masses of isolated cells. It was found that these cells be- gan to come together and form aggregations which subsequently differentiated into the form of the spe- cies from which the fragments were taken. Out of a hodge-podge of all sorts of cells one sees the gradual emergence of an organic body. The pro- ponents of the conception of a cell state could scarcely hope for a better illustration of their point of view. It is as if a group of independent cells had said among themselves: "Behold, let us creep together and form an organism. If some of us do this, some that, and others something else we will all get along in peace and harmony and more- over much to our mutual advantage." How a cell differentiates as well as what it does in the way of behavior depends largely on the re- lations in which it stands to other cells. The di- rection of differentiation which a cell follows may be looked upon as a response to environmental stimuli, just as the movements of a cell may be so regarded. References 195 The energy evolved as a consequence of stimulation may be employed mainly in producing structural modifications of the cell substance, or it may be ex- pended mainly in producing motor reactions. The response is usually purposive and social in either case. Zur Strassen in a discussion of "Animal Behavior and Development" before the Seventh International Zoological Congress at Boston called attention to the social behavior of cells and the analogies be- tween such behavior and the activities of social in- sects. The comparison is suggestive, but as Zur Strassen points out, there is a closer relationship be- tween the cells of the metazoan body and the pro- tozoa, which are not improbably the direct ances- tors of these cells. Were the protozoa social in their behavior, did the different behavior of the several castes depend upon mutual interaction as it does to a certain degree among social insects, the analogy between behavior and development would indeed be close. REFERENCES ALLEN, B. M. The origin of the sex cells of Amia and Lepidosteus. Jour. Morph. 22, i, 1911. HARRISON, R. G. The outgrowth of the nerve fiber as a mode of protoplasmic movement. Jour. Exp. Zool. 9, 787, 1910. HOLMES, S. J. ( i ) Observations on isolated liv- 196 Studies in Animal Behavior ing pigment cells from the larvae of amphibians. Univ. of Calif. Pubs. Zool. n, 143, 1913. (2) Be- havior of ectodermic epithelium of tadpoles when cultivated in plasma. I.e. n, 155, 1913. (3) The movements and reactions of isolated melanophores of the frog. 1. c. 13, 167, 1914. (4) The behavior of the epidermis of amphibians when cultivated out- side the body. Jour. Exp. Zool. 17, 281, 1914. LOEB, J. On the heredity of the marking in fish embryos. Biol. Lectures, Marine Biol. Lab. Woods Hole, 1899. MASSART, J., and BORDET, C. Recherches sur 1'irritabilite des leucocytes, etc. Jour. Soc. Roy. Sci. Med. et.Nat. de Bruxelles, 1890. OPPEL, A. Causal-morphologische Studien. IV Mitteilung. Arch. f. Entw-mech. 34, 133, 1912; V Mitteilung, 1. c. 35, 371, 1912. PETERS, A. Ueber die Regeneration des Epi- thels der Cornea. Arch. mikr. Anat. 33, 153, 1889. Roux, WM. Ueber den Cytotropismus der Furchungszellen des Grasfrosches. Arch. f. Entw- mech. 1, 1894. WILSON, H. V. ( i ) Development of sponges from dissociated tissue cells. Bull. Bureau Fish- eries, 30, 1911. (2) The behavior of the disso- ciated cells in Hydroids, Alcyonaria, and Asterias. Jour. Exp. Zool. n, 280, 1911. ZUR STRASSEN, H. Animal behavior and de- velopment. Proc. 7th Internat. Zool. Congress, Boston, 1912. XI THE INSTINCT OF FEIGNING DEATH ' I AHE so-called instinct of feigning death is one -*- which is very widely distributed in the animal kingdom. It crops out sporadically, as it were, in forms which are but very distantly related, and hence it must have been independently evolved a great many times. The expression feigning death is a misleading one to the extent that it is apt to give rise to the idea that the animal consciously adopts this device with the intent to deceive. But while it is probable that among the higher animals which sometimes feign death there may be an attempt to mislead their enemies, it is quite certain that among the insects, spiders and other low forms, there is no such aim in the creature's mind, if we grant (what some naturalists are disposed to deny) that these animals have minds. The unpremeditated character of this peculiar behavior was first shown by the ex- periments of Fabre on beetles. In order to ascer- tain if the duration of the feint was in any way affected by his own movements, Fabre made several observations on a large carab beetle, Scarites, which shows the death feigning instinct in a pronounced and typical form. When handled, the beetle would 197 198 Studies in Animal Behavior throw itself into an immobile state with its head bent down and its legs drawn in close to the body. It would remain in this attitude perfectly quiet for several minutes — sometimes for over an hour. Its awakening would be first manifested by a slight trembling of the feet and a slow oscillation of the antennae and palps ; then its legs would move about more vigorously, and finally the insect would arise and scamper off. Seized again, it would repeat the performance several times in succession, the dura- tion of the feint often increasing with successive trials. Finally, as if wearied, or convinced that the ruse were vain, the beetle would refuse to feign longer. Were the feints attempts to deceive its captor by simulating death? Fabre placed the insect on its back, went to a distant part of the room, and re- mained perfectly quiet. The beetle still lay as usual. He then went out of the room, carefully looking in at intervals to watch the course of events. Still the same immobility. In other cases he covered the insect so that it could not see out and then quietly went away. This was also found to make no dif- ference. In fact, whether the insects were sur- rounded by sounds and sights of moving objects, or entirely excluded from these influences made no dif- ference in the average length of time they would remain in a motionless condition. Similar experi- ments have been made on other insects by different observers,' who have all arrived at the conclusion The Instinct of Feigning Death 199 that conscious deception plays no part in the process. The attitudes assumed by insects and other forms when feigning death are usually quite different from those of dead specimens. This general fact was pointed out by Darwin, who says that "I carefully noted the simulated positions of seventeen different kinds of insects (including an lulus, spider and Onis- cus) belonging to distinct genera, both poor and first-rate shammers; afterward I procured naturally dead specimens of some of these insects, others I killed with camphor by an easy slow death; the result was that in no instance was the attitude ex- actly the same, and in several instances the atti- tudes of the feigners and of the really dead were as unlike as they possibly could be." The attitudes of animals in the death feint are frequently very characteristic. Many beetles as well as other forms feign with the legs drawn up to the body and the antennae closely appressed, so that the whole insect assumes as compact a form as pos- sible. The woodlouse, Armadillo, rolls itself up into a ball with its legs drawn into the center, a habit which has doubtless caused the name pill-bug to be given to this crustacean. A beetle, Geotrupes, ac- cording to Kirby and Spence, "when touched or in fear sets out its legs as stiff as if they were made of iron wire — which is their posture when dead — and remaining motionless thus deceives the rooks which prey upon them. A different attitude is assumed by one of the tree-chafers probably with the same end 200 Studies in Animal Behavior in view. It sometimes elevates its posterior legs into the air, so as to form a straight vertical line, at right angles with the upper surface of its body." FIG. 8. — Larva of a geometrid moth attached to a twig. Spiders usually feign by folding up their legs, dropping down, and remaining motionless. The caterpillars of some of the geometrid moths have the curious habit of attaching themselves to a branch The Instinct of Feigning Death 20 1 by their posterior legs and holding the body straight and stiff at an angle to the stem, thus forming a remarkably close resemblance to a short twig. Fre- quently the deceptiveness is increased by a marked similarity in color to that of the branch to which they are attached. While in most cases a species has a particular at- Fio. 9. — A water scorpion, Ranatra, feigning death. titude which it maintains when simulating death, there are some forms which feign in whatever pos- ture they may be when disturbed. A good ex- ample of this is afforded by the water-scorpion Ranatra. This insect has the two hinder pairs of legs, which are employed in walking and swimming, very long and slender; the first pair are fitted for grasping the small aquatic animals on which it feeds and are carried straight out in front of the body. It is only necessary to pick one of these insects out of the water to throw it into a stiff, immobile con- dition which usually lasts several minutes and some- 2O2 Studies in Animal Behavior times for over an hour. The legs may be closely pressed to the body so that the creature resembles a stick, or they may stand out at right angles to it, or be bent in any position, some in one way and some in another, depending upon how they happen to lie when the feint began. And no matter how awkward the position, it is rigidly maintained until the feint wears off. I have found that young Rana- tras, the first day they emerged from the egg and while their appendages were still soft and easily bent, showed the same death feigning instinct as the adults, although they did not persist in it for so long a time. The water-bug Belostoma usually feigns with the legs closely pressed to the thorax or else held folded at right angles to the body. In Nepa, on the other hand, the attitudes assumed seem to depend mainly on the position of the legs just previous to the death feint, so that it is difficult to distinguish a feigning individual from one that is really dead. Schmidt has been able to make the walking-stick Carausius feign in all sorts of ungainly positions which would often be maintained for hours at a time. Death feigning does not seem to occur among the lower invertebrate animals such as the Protozoa, coelenterates, molluscs and worms, although some of them may exhibit reactions which are prophetic of this instinct. Among crustaceans the instinct in its fully developed form is quite uncommon. Some The Instinct of Feigning Death 203 years ago I described the death-feigning of certain species of terrestrial amphipod crustaceans which are frequently found on sandy beaches near the sea- shore. On account of their peculiar hopping move- ments these crustaceans are commonly known as sand-hoppers or sand-fleas, although they have of FIG. 10. — A sand flea, Talorchestia, in the death feint. course no relation to the ordinary fleas of human experience. One of the largest species of sand- hopper, Talorchestia, is common along our Atlantic coast, where it lives during the day in burrows made in the sand, coming out only at night to feed upon the seaweed and other material washed ashore by the waves. When the Talorchestias are dug out of their burrows, they usually lie curled up with their 2O4 Studies in dnimal Behavior long antennae bent under the body and their legs drawn up so as to assume a compact form. They will lie in this way for several minutes, when they may be seen slowly to relax; the legs then move about, and soon the creature hops away by a sudden extension of its abdomen. When caught in the hand they will feign death again, and repeat the perform- ance many times in succession. Other species of sand-hoppers exhibit the same instinct, though less perfectly, and there are traces of it in many of the reactions of their aquatic relatives. The various species of wood lice exhibit the in- stinct of feigning death in various degrees. Some species are able to roll up into an almost perfect ball and will remain in that state for a consider- able time. Other species curl up, but make only a very imperfect approximation to a sphere, and they may maintain this attitude for only a short period. Some myriapods when disturbed curl up in much the same way. Among spiders death-feigning is not un- common, especially among the orb weavers. It is among the insects that the death-feigning in- stinct reaches its fullest development, occurring to a greater or less extent in most of the orders. It is especially common in beetles and not unusual among the bugs, but it is quite rare in the highest orders such as the Diptera and the Hymenoptera, or the ants, bees and their allies. It occurs in a few cases among butterflies and moths, both in the imago as well as the larval state. The instinct is The Instinct of Feigning Death 205 exhibited in different species in all stages of develop- ment from a momentary feint to a condition of in- tense rigor lasting for over an hour. Among the vertebrate animals death-feigning has been observed only rarely in the fishes. In the Am- phibia it is not exhibited in the striking way it oc- curs in insects and spiders, although frogs and toads may be thrown by the proper manipulation into an immobile condition more or less resembling it. A phenomenon apparently related to the death feigning of insects has long been known in certain reptiles. Darwin in his Journal of Researches describes a South American lizard which when frightened "at- tempts to avoid discovery by feigning death with out- stretched legs, depressed body, and closed eyes; if further molested it buries itself with great quickness in the loose sand." The Egyptian snake charmers by a slight pressure in the neck region are able to make the asp suddenly motionless so that it remains entirely passive in the hands of the operator. And similar phenomena have been found in other spe- cies. In birds the instinct crops out only here and there. A few summers ago when on the island of Peni- kese I was somewhat surprised to find the instinct well developed in the young terns which were hatched out in abundance on the hillsides. For a short time after being hatched the little downy fellows betray no fear of man and will cuddle un- 206 Studies in Animal Behavior der one's hand in perfect confidence. When the birds become larger and acquire their second coat of feathers the instinct of fear takes possession of them and they run and hide in the grass when you approach. Here they lie perfectly quiet; you may pull them about, stretch out their legs, necks, or wings and place them in the most awkward posi- tions, and they will remain as limp and motionless as if really dead. They will even suffer their wing or tail feathers to be plucked out one by one with- out a wince. But all of a sudden the bird becomes a very different creature. It screams, pecks and struggles to escape, and is very apt to succeed on account of the surprising quickness of the change. I have made several attempts to make a bird feign death a second time, but never met with success. According to Couch the land rail and skylark feign death, and Wrangle states that the wild geese of Siberia have the same habit during their molting season, when they are unable to fly. Hudson states in his most interesting Naturalist on the La Plata that the common partridge of the pampas, when captured, "after a few violent struggles to escape drops its head, gasps two or three times, and to all appearances dies. If, when you have seen this, you release your hold, the eyes open instantly, and with startling suddenness and noise of wings, it is up and away and beyond your reach forever." In mammals the instinct is so well shown in one of the lower members of the group, the opossum, The Instinct of Feigning Death 207 that the expression "playing possum" is familiar to every one. Foxes when trapped or hard pressed often drop down limp and apparently lifeless and will even endure a good deal of maltreatment with- out making any response. Hudson records that he was "once riding with a gaucho when we saw, on the open level ground before us, a fox not yet fully grown standing still and watching our approach. All at once it dropped, and when we came up to the spot it was lying stretched out, with eyes closed, and apparently dead. Before passing on my com- panion, who said it was not the first time he had seen such a thing, lashed it vigorously with his whip for some moments, but without producing the slight- est effect." Mr. Morgan in his book on the beaver gives the following instance on what he assures us is excellent authority: "A fox one night entered the hen-house of a farmer, and after destroying a large number of fowls, gorged himself to such repletion that he could not pass out through the small aperture by which he had entered. The proprietor found him in the morning sprawled out upon the floor apparently dead from surfeit; and taking him up by the legs carried him out unsuspectingly, and for some distance to the side of his house, where he dropped him upon the grass. No sooner did Reynard find himself free than he sprang to his feet and made his escape." Dogs are frequently deceived by this ruse of the fox and doubtless foxes have many times owed their 208 Studies in Animal Behavior lives to its aid. It has been often noticed that if one withdraws from a fox when it is feigning it may be seen to slowly open its eyes, then raise its head and carefully look around to see if its foes are at a safe distance, and finally scamper off. While in insects the instinct of feigning death is probably a simple reflex reaction to outer stimuli, it is doubtless associated in birds and especially mam- mals with a tolerably acute consciousness of the situ- ation. It involves a more or less deliberate inten- tion to profit by the deception, yet at the same time it is probably not a result of conscious reflection. The instinct is there, or else such a course of action would not occur to the animal's mind. Were it otherwise it would be difficult to understand why the ruse is adopted only by certain species while many others, equally intelligent and for whom it would be an equally advantageous stratagem, never mani- fest it. There can be little doubt that a fox which slowly opens its eye and warily looks around is act- ing with an intelligent appreciation of his predica- ment, but it is not to be inferred that he could have reasoned out his course of action "did not an innate proclivity in that direction form a part of his in- stinctive make-up. The physiological condition in what is called death-feigning is quite different in different forms. While there is a general relaxation of the muscu- lature in the sham death of some of the birds and mammals, the feint in most of the lower animals The Instinct of Feigning Death 209 is characterized by a tetanic contraction of the mus- cles. The attitudes assumed by many forms, such as rolling into a ball, keeping the legs and other appendages drawn close to the body, or, in some cases, holding them straight and rigid, are such as can be maintained only at the cost of considerable muscular effort. If a Ranatra is picked up by one of its slender legs it may be held out horizontally for a considerable time without causing the leg to bend. The situation is similar to that of a man seized below the knee and held out straight, face upward, without causing the knee to bend. As the legs of Ranatra are relatively exceedingly slender, the muscular tension which the insect maintains must be intense. Similar muscular rigidity is shown in the walking-stick Carausius studied by Schmidt. Specimens supported by the tip of the abdomen and the ends of the outstretched fore legs would lie straight as a stick for hours. Death feigning is markedly influenced by external conditions such as light, contact, moisture and espe- cially temperature. In experimenting with amphi- pods I have found that when the temperature is lowered the death feint persists for a considerably longer time. The same is strikingly true of the death feint of Ranatra. At temperatures higher than the normal the Severins found that the duration of the death feint in the water bugs Nepa and Belos- toma was greatly decreased. A sudden transition from a warm to a cold temperature, however, was 2io Studies in Animal Behavior found to diminish the duration of the feint in both of these species, owing possibly to the shock effect of the sudden change. The influence of light on the duration of the feint has been studied by the Severins in the forms just mentioned and by myself in the water bug Ranatra. The results agree in showing that the duration of the feint is greater in dim than in strong light, and that the feint is further shortened if a light is kept moving over the insect. The latter result, like the former, is probably due to the greater stimulation to which the feigning insect is subjected. As all these water bugs are positively phototactic, light tends to elicit an active response which antagonizes the instinct of feigning death. In a Ranatra which is induced to come out of the death feint by moving a light above it, the first signs of life manifested are orienting movements of the head which take place in perfect unison with the movements of the light. These are followed by swaying movements of the body, until finally the insect attempts to follow the light by walking or, if highly excited, by flying. It is a curious fact that while the water bugs, Nepa, Belostoma and Ranatra, feign death very readily when out of the water, they will do so much less readily when submersed in their natural ele- ment. In Ranatra repeated manipulations under the water usually fail to elicit anything but a mo- mentary and undecided response. Nepa under the same circumstances feigns for a somewhat longer The Instinct of Feigning Death ill time, and Belostoma, in exceptional cases, may feign for as much as a few minutes. Throwing any of these insects into water while they are feigning usu- ally terminates the feint at once or in a short time. The transition from air to water, even when no temperature changes are encountered, produces a marked effect on the reactions of many semi-aquatic forms. In Ranatra and certain terrestrial amphi- pods, as we have seen in a previous chapter, it pro- duces a sudden reversal of phototaxis, a change not improbably due to the influence of contact stimuli. It is not improbable that it is the influence of con- tact stimulation that terminates the death feint of aquatic bugs when they are placed in the water. Most insects and crustaceans which feign can be caused to do so repeatedly by stroking or handling them as soon as they show signs of activity. I have performed a number of experiments on amphi- pods, Ranatra, beetles and orb-weaving spiders to ascertain how long the death feint may be continued, and to determine also the lengths of successive feints. In all the forms experimented with there was found to be much variability in the behavior of different individuals, so that it was necessary to perform numerous experiments in order to arrive at trust- worthy conclusions. In general, all the forms stud- ied showed a gradual diminution in the duration of successive feints, until, often after several hours, it was no longer possible to evoke the response. Similar experiments undertaken at my request by the 212 Studies in Animal Behavior Severins showed the same phenomenon in Nepa and Belostoma. Fabre in his studies on Scarites found an increase in the duration of the first five feints, but his observations were not sufficiently numerous to eliminate the rather large amount of variability due to unknown causes. As the feints grow shorter the attitudes of the organism become less characteristic. Amphipods and pill-bugs do not roll up so completely or keep the appendages drawn so closely to the body. Spi- ders and beetles do not assume so compact a form, and in general it may be said that the muscular sys- tem gives evidence of diminished contraction. As the death feint is usually accompanied by a tetanic contraction of the muscles one would expect to find a diminution in the duration and perfection of the response as a simple consequence of fatigue. In a similar manner the diminution of the duration of the feint that occurs under higher temperature may be due to the fact that the muscles become exhausted more quickly when the metabolism is increased by the higher temperature of the body. The experiments undertaken to ascertain what part of the nervous system is most concerned in the death feint have yielded somewhat varied results. Robertson found that in the active species of spiders Epeira and Amaurobius the sham death reaction per- sisted after the removal of the brain, and was mani- fested in a weakened form when the subesophageal ganglion and even the first thoracic were removed The Instinct of Feigning Death 213 also. In a sluggish spider Celaenia Robertson found that "The sham death posture cannot be induced without the head ganglia." Schmidt finds that in Carausius the death feint entirely disappears after the removal of the brain. My own experiments on Ranatra showed that removal of the brain caused a marked diminution in the duration of the feint, although the response could still be induced in de- cerebrate specimens. If a feigning individual be cut in two across the prothorax the posterior part often continues to retain its rigidity for some time and may be thrown back into the death feint again if it is picked up and stroked. Similar results were ob- tained in Belostoma and Nepa by the Severins, who have investigated the role of the nervous system in especial detail. The cataplectic state which Preyer and Verworn found could be induced in decerebrate fowl may be allied to the conditions described above. One marked characteristic of the death feint is an apparent insensibility to pain. De Geer in writing of a boring beetle Anoblum pertlnax says that "you may maim them, pull them limb from limb, roast them over a slow fire, but you will not gain your end; not a joint will they move, nor show by the least symptom that they suffer pain. A similar apathy is shown by some species of saw-flies (Ser- rifera), which when alarmed conceal their antennae under their body, place their legs close to it, and remain without motion even when transfixed by a pin. Spiders also simulate death by folding up their 214 Studies in Animal Behavior legs, falling from their station, and remaining mo- tionless; and when in this situation may be pierced and torn to pieces without their exhibiting the slight- est symptom of pain." The Severins tried the ap- plication of heat to Belostoma, but the insect was invariably brought out of its feint and made strug- gles to escape, although it might endure more or less mutilation without making any response. I have found that a feigning Ranatra will allow its legs to be snipped off without betraying the least move- ment beyond an occasional twitch. Similar insen- sibility has been observed in Nepa (Severin) , Carau- sius (Schmidt) and other forms. As has been pointed out by Romanes, Preyer, Verworn, Schmidt and others, the instinct of feign- ing death is doubtless closely connected with much of what has been called hypnotism in the lower ani- mals. Crayfishes, frogs, lizards, certain snakes and many birds and mammals, may by a very simple process be thrown into an inactive condition from which they are not readily aroused by external stim- uli. In ordinary death feigning the animal falls into its immobile state upon slight provocation; a touch, or even a jar is sometimes all that is required. In) the so-called cases of hypnosis more or less manipu- lation is necessary. The exciting cause in both cases is generally some form of contact stimulus. In the hypnotism of animals, as Verworn and others have shown, there is diminished reflex irritability, and usually tonic contraction of many at least of the The Instinct of Feigning Death 215 muscles. Similar phenomena are observed in the death feigning of many forms, some of the insects, as we have seen, showing a lack of responsiveness that is truly remarkable. The independent development of death feigning along many different lines of descent makes it prob- able that we must look for the origin of this curious instinct in some fundamental and widespread mode of behavior. In a paper on the death feigning of terrestrial amphipods the writer has suggested that the death-feigning instinct in these forms had its origin in an accentuation of the thigmotactic re- sponse which is such a prevalent trait of behavior among the amphipods in general. It was found pos- sible to establish a series of stages between the typ- ical death feint of the large terrestrial amphipod Talorchestia and the ordinary thigmotactic reactions of the aquatic relatives of this species. In Orchestia palustris, a species not so exclusively terrestrial as the preceding, the body during the death feint is less closely curled up, the appendages are not so closely drawn up to the body, and the feint is not so per- sistent. In the small Orchestia agilis, which lives usually near the water line, the death feint is shown in a somewhat less decided way, while among the aquatic members of the Orchestiidae the same trait is manifested by a tendency to curl up and lie quiet when in contact with rocks or seaweed. To one who has studied and compared the attitudes and behavior of this series of forms there can be little doubt that the typical death feint of the most terrestrial of the 216 Studies in Animal Behavior species has its basis in the thigmotaxis of the aquatic forms. As the species studied become more terres- trial in habit, the thigmotaxis becomes gradually specialized into a typical instinct of feigning death. Many facts indicate that death feigning in insects and other forms has had a similar origin in the thigmotactic response. It is a rather striking fact that, with very rare exceptions, it requires some form of contact stimulus, it may be but a touch, jar, or even a breath of air, to elicit this instinct. One of the very few exceptions to this rule which has been recorded is afforded by Carausius, whose death feint, according to Schmidt, arises from internal causes, and cannot be induced by any discoverable environmental agency. Should it be definitely estab- lished that this case is truly one of "autocatalepsy," as it has been called by Schmidt, it would not be fatal to the supposition that it began originally as a reaction to contact; for it is not without precedent that an instinct having originated with reference to one feature of the environment should finally come to be set into operation by a quite different cause. Those cases of death feigning in which there is a limp and relaxed condition of the musculature, such as occurs in some birds and mammals, may have an origin quite different from that of the more preva- lent cataleptic type. The suggestion that death- feigning had its origin in the partial paralysis pro- duced by fear may perhaps apply to cases such as these, although this explanation cannot, I feel sure, References 217 be extended to the derivation of the typical form of this instinct. While there is simulation of death in both types of behavior we have described, it is prob- able that we have to do with two distinct kinds of reaction differing both in their physiological charac- ter and in their phyletic origin. REFERENCES DARWIN, C. Chapter on instinct appended to Romanes' mental evolution in animals. 1884. FABRE, J. H. Souvenirs entomologiques, T. 7. HOLMES, S. J. ( i ) Death feigning in terrestrial amphipods. Biol. Bull. 4, 191, 1903. (2) Death feigning in Ranatra. Jour. Comp. Neur. Psych. 15, 200, 1906. (3) The instinct of feigning death. Pop. Sci. Mon. 72, 179, 1908. HUDSON, W. H. The naturalist on the La Plata, 1892. KIRBY, W. and SPENCE, W. An introduction to entomology. 6th ed., 1846. LOHNER, L. Untersuchungen iiber den sogen- annten Totstellreflex der Arthropoden. Zeit. allg. Physiol. 16, 373, 1914. MORGAN, L. H. The American beaver and his works, Phila. and London, 1868. POLIMANTI, O. Studi di Fisiologia etologica. II, Lo stato di immobilita temporanea ("morte appar- 218 Studies in Animal Behavior ente"-"Totenstellen") nei Crostacei Brachiuri. Zeit. allg. Physiol. 13, 201, 1912. ROBERTSON, T. B. On the "sham-death" reflex in spiders. Jour. Physiol. 31, 410, 1904. ROMANES, J. G. Mental evolution in animals. 1884. SEVERIN, H. H. P., and SEVERIN, H. C. An ex- perimental study on the death-feigning of Belostoma (-zaitha aucct.) flumineum Say and Nepa apiculata Uhler. Behavior Monographs, Vol. I, No. 3, 1911. SCHMIDT, P. Katalepsie der Phasmiden. Biol. Cent. 33, 193, 1913. VERWORN, M. ( i ) Tonische Reflexe. Arch, ges. Physiol. 65, 63, 1897. (2) Beitrage zur Physiologic des Zentralner- vensystems. I Teil. Die sogenannte Hypnose der Tiere. Jena, I XII THE RECOGNITION OF SEX "LTOW do the males of the lower animals dis- ••• -"-tinguish the females of their own species from all the rest of the animate creation? Obviously the perpetuation of the race depends on the circum- stance that the male is correctly guided in the choice of a mate. A male beetle or bug will pass by with indifference thousands of other varieties of insects, but in the presence of a female of his own kind his interest is keenly aroused. It almost goes without saying that this power of discrimination is a matter of instinct. It is easily demonstrable in most animals that the element of experience is entirely unnecessary for the proper solution of this important problem. Apparently the male is guided by a sort of elective affinity to the right object upon which to bestow his attentions. What are the signs by which this object is recognized? An important factor in the discrimination of sex that naturally occurs to one is the sense of smell. The lower animals, or at least many of them, are influenced by odors to an extent which we with our comparatively obtuse olfactories find it difficult to TVD ' appreciate. The odors of most species are specific, TV^ 219 22O Studies In Animal Behavior and even individuals, in higher forms at least, have an odor peculiarly their own which a bloodhound is able to distinguish from among dozens of others. Differences in odor frequently characterize the two sexes, and hence afford a feasible means of sex dis- crimination. The males of many forms have much more highly developed olfactory organs than the females. The antennae of insects which contain the olfactory sense organs are frequently larger and more complex in the males. In the drone bee, for instance, the olfac- tory pits of the antennae are many times more numer- ous than in the queen or worker. Many male moths have large feathered antennae, whereas these organs are much smaller in the females. It is mainly through the sense of smell that the male moths are able to find their mates, and in some species this sense is developed to a degree that almost surpasses credence. In one of his most delightful essays the French naturalist Fabre tells of the nuptial flight of the males of the large and beautifully colored moth called the great peacock (le grand paon) . A cocoon which was kept in Fabre's study had produced a female moth which was placed under a gauze cover. In the evening the observer had his attention at- tracted by the call of his young son: " 'Come quick, come and see these butterflies ! Big as birds ! The room is full of them!' Several of the large moths had come into the room through the open window. The Recognition of Sex 221 Others had entered the kitchen, and still others were found in various rooms wherever there was a chance for them to enter. . . . "It was a memorable night — the Night of the Great Peacock! Come from all points of the com- pass, warned I know not how, here were forty lovers eager to do homage to the maiden princess that morning born in the sacred precincts of my study." The night was one of black darkness, yet the moths threaded their way through the trees surrounding the house, and came through open windows into darkened rooms without abrading in the least the scaly covering of their wings. But keen as the sense of sight in these insects may be it is not through this sense that the males are drawn toward their intended mates. "When," says Fabre, "I placed the females in boxes which were imperfectly closed, or which had chinks in their sides, or even had them in a drawer or a cupboard, I found the males arrived in numbers as great as when the object of their search lay in the cage of open work freely exposed on a table. I have a vivid memory of one evening when the recluse was hidden in a hat-box at the bottom of a wall-cupboard. The arrivals went straight to the closed doors, and beat them with their wings, toc-toc, trying to enter. / Wandering pilgrims, come I know not whe/e, across / fields and meadows, they knew perfectly what was behind the doors of the cupboard." Fabre cut off the antennae of several of the males of this and other species of moths and found that 222 Studies in Animal Behavior they failed to approach the vicinity of the female. The presence of other odors, which would make an almost intolerable stench in our own nostrils, failed to deter the males in the least from the object of their search. The odor, according to Fabre, seemed to be carried against currents of air, for a rare female specimen of the Lesser Peacock drew numer- ous males which flew with the wind to the place of her confinement. It is scarcely to be wondered at that Fabre considered the sense of smell to depend in part on other means than the wafting of odorous particles, a sort of force acting after the manner of rays of light or X-rays and capable of radiating to a great distance despite adverse currents of air. The role of smell in sex recognition among the crustaceans is more uncertain. The antennal sense organs which are very probably olfactory in function are in some species much better developed in the male sex. But little is known, however, in regard to the means by which most species find their mates. In some amphipod crustaceans with which the writer experimented a few years ago olfactory stimuli were found to play little or no part in the discrimination of sex. The males of this group have the curious habit of carrying the females about under the body. This act of transportation has no direct reference to the impregnation of the eggs further than to insure the proximity of the sexes when the proper time for fertilization arrives. This occurs soon after the fe- male casts off her skin, when the sperms are depos- The Recognition of Sex 223 ited by the male on the ventral side of the thorax of his mate. After the eggs are fertilized the male continues to swim about with the female as before. "The instinct of the male amphipod 1 to seize and retain hold of the female is one of remarkable strength. The male retains his hold, despite all efforts to dislodge him, with remarkable persistence, and will still cling to the female after the posterior half of his body has been cut away. My own obser- vations on the sexual behavior of amphipods relate mainly to three species, Amphithce longimana Smith, Hyalella dentata Smith, and Gammarus fasciatus Say. The sexual behavior of these three species is remarkably similar although they belong to as many distinct families. The female while carried about keeps remarkably impassive. Her thoracic legs are drawn up, the abdomen held strongly flexed, the whole body assuming as compact a form as possible. She takes no part in swimming; the movement of the pleopods when the body is strongly bent upon itself serves only to keep a current of water passing by the gills. She is carried about like a helpless burden, allowing her vigorous spouse to assume the entire labor of transportation and the responsibility for keeping her as well as himself out of danger. "The efforts of the male to seize the female and get her into the proper position to be carried have the effect of inducing her to throw herself into the 1 Quoted from an article by the writer on "Sex Recognition in Amphipods," published in the Biological Bulletin. Vol. 5, p. 288, 1903. 224 Studies in Animal Behavior characteristic bodily attitude and remain quiet. The attitude assumed by the female is similar to that observed in the ordinary thigmotactic reaction of amphipods and may, perhaps, be but the same form of response somewhat modified and specialized in relation to the function of reproduction. When the males are torn away from the females they soon seize their partners again and roll them about into the proper position and then proceed on their way in apparent contentment. The female, as soon as seized by the male, curls up and allows herself to be rolled and tumbled about without a show of resist- ance or protest. The males, as a rule, are larger than the females and usually get their partners into the desired position quite readily; but when a small male attempts to carry a large female he experiences much difficulty. I observed a male Hyalella en- deavoring to carry a female somewhat larger than himself. After seizing the female he would turn her around until she finally came into the proper position for transportation, but owing to the larger size of his partner the male could not reach around her body so as to carry her away. No sooner was the female properly adjusted than the male would lose hold of her round body and the same efforts had to be repeated. During all this performance the female remained dutifully passive. After watching the further struggles of the male for over half an hour I became convinced, although he was not, that he had undertaken an impossible task, and discon- The Recognition of Sex 225 tinued my observations." •That the male amphipods do not distinguish the females by sight was shown by blacking over the eyes of several males and then placing them in a dish with females. It was not long before each male had secured a mate. The possible role of the sense of smell was tested by removing from a number of males the first antennae which contain the olfactory sense organs. After the specimens had recovered from the slight shock of the operation they seized the females and carried them about in the usual man- ner. They reacted in the same way when the second antennae were removed also. In another experiment several females were con- fined within an enclosure of wire gauze which was placed in a dish of water containing several eager males. The males paid not the slightest attention to the females within the enclosure, but after it was raised and the females allowed to scatter through the dish, most of the males were found carrying their mates. "If one attentively observes Hyalellas as they are swimming about, it will be seen that the males do not pursue the females, great as their eagerness may be to seize and carry one of the opposite sex. Only when the two sexes collide in their apparently ran- dom movements does the male become aware of the presence of the female. When a male and a female collide, the female curls up and lies quiet while the male makes efforts to seize her. Should two females 226 Studies in Minimal Behavior collide they may curl up for a moment, but as they are not seized they soon pass on. When two males meet there is often a lively struggle. Each appar- ently attempts to seize and carry the other, but as neither will consent to remain passive they soon sep- arate. The different reactions of the two sexes to contact with other individuals is the factor which effects the union of the males with the females. Each reacts to the reactions of the other. The male has a strong instinct to seize and carry other individuals of the same species. The female has the instinct to lie quiet when another individual comes into contact with her, especially if she is seized. The instinctive reactions of the two sexes are complementary, and cooperate to bring about and maintain the peculiar sexual association characteristic of the Gammaridea. "If the association of the sexes is brought about by their peculiar modes of reaction to certain contact stimuli, it would seem probable that the only reason why males do not carry other males as well as females is that they are prevented from so doing by the active resistance of their intended mates. I was accordingly led to try the experiment of mutilating some male specimens so that they could no longer make effective resistance to seizure. The large sec- ond gnathopods (the principal means of defense) of several males were cut off and the mutilated indi- viduals were placed in a dish with several males which were recently torn from females. The muti- lated males were soon seized and carried about as The Recognition of Sex 227 if they were members of the other sex. In one case a mutilated male was carried about for over five hours. The mutilated males were more active than females are under the same conditions, and did not assume the same bodily attitude, but nevertheless their captors carried them without any manifest awareness of the deception to which they were sub- jected." In another group of crustaceans, the Copepoda, the mating instincts are more or less analogous to those of the amphipods. In the copepods, however, the males in seizing the females employ the first an- tennae, which are often enlarged and especially mod- ified for this function. In Cyclops fimbriatus, whose behavior was studied by the writer, the male clasps the female just in front of an enlargement at the base of the abdomen. Males show much eagerness in grasping the females, and they may be poked about roughly with a needle and the posterior part of the body may be cut off without causing them to leave their hold. "As the pairs of Cyclops 1 swim through the water the males are usually the more active. Frequently the female remains entirely quiet with the appen- dages drawn close to the body, and the body flexed ventrally, allowing herself to be passively carried about by her mate. At other times the female may swim as actively as the male. In general the be- 1 Quoted from Holmes on "Sex Recognition in Cyclops," Bio- logical Bulletin, Vol. 16, p. 313, 1909. 228 Studies in Animal Behavior havior of the females and their attitude while being carried closely resemble what is found in the Amphi- poda. So also does their behavior when the males come in contact with them and attempt to seize them. The female during the efforts of the male to clasp her around the base of the abdomen usually lies quiet with the appendages drawn close to the body. . . . "So far as could be detected the males do not seek or follow the females at a distance as Parker con- cluded they did in Labidocera. The association of the sexes seems to be due merely to chance collisions. Males often attempt to seize other copepods regard- less of sex. The males resist such attempts at seiz- ure and dart quickly away, while the females often stop and submit readily to the clasping propensities of their companions. Several males were injured so that they could not resist seizure, and in many cases they were seized by other males who worked indus- triously until they got their burden clasped around the base of the abdomen in the usual way. These associations did not last long, however; the active males, apparently appreciating that something was wrong, soon swam away. Recently killed females were often seized and in some cases carried about for a while, but they were finally dropped." There was no evidence that odor determined the sexual behavior of the males. The males paid no attention to a number of females that were enclosed within a tube whose end was covered with wire gauze. In another experiment several females were The Recognition of Sex 229 placed in a tube in which a small plug of loose cotton was inserted a short distance from one end. The males showed no tendency to enter the open mouth of the tube as they might be expected to do if they were attracted by the odor of the females. The ex- periment of removing the organ of smell, which was performed in the case of the amphipods, would be a fruitless one in Cyclops, as the seat of smell is located, to a considerable degree at least, in the same organs that are used for clasping. "It is evident that mating in Cyclops is brought about much as it is in the Amphipoda. The males have a strong tendency to clasp other copepods; the females tend to remain quiet in a condition some- what resembling the death feint while being seized by the males. It is not improbable that olfactory stimuli may cause the males to remain with the fe- male longer than they otherwise would, and they may render the males more prone to seize females than other males, but so far as could be deter- mined by watching the behavior of the animals the specific reaction of the two sexes to certain kinds of contact stimuli is the main factor in bringing about their association." In the crayfish the studies of Andrews and of Pearse have shown that sex recognition is effected by much the same method as is followed in the am- phipods. Extracts from the bodies of females added to water containing the males did not elicit the least response. Nothing in the behavior of the males indi- 230 Studies in Animal Behavior cates that they are drawn toward the females by the sense of smell. Mating apparently is dependent upon the chance meetings of the sexes. "During the mating season," says Pearse, "the instinct of the male is to grasp and turn over every crayfish that comes in his way. ... If this individual is a female of the same species the attempt may meet with suc- cess, but if it is a male or a female of another species the effort at sexual union is usually of short dura- tion." If males attempt to mate with other males, as they often do, they encounter active resistance, but if a female is attacked she usually remains pas- sive. In fishes the males are frequently distinguished by their more conspicuous coloration especially during the breeding season. The pugnacity and threaten- ing attitudes of the males at this time undoubtedly contribute to their mutual recognition, but there is evidence that the mature males often recognize one another by their peculiar markings. In her account of the breeding habits of the rainbow darter, Miss Reeves records several cases of young dull-colored males being mistaken for females, whereas the larger and more brilliantly colored adults usually recognize one another without difficulty. "The more nearly the behavior of a dull male simulates that of a female, as in the case of a male burrowing for food, the more is he likely to be taken for a female. Upon the near approach of the brilliant male the young male erects the first dorsal and rapidly escapes, The Recognition of Sex 231 modes of behavior not observed in the female. It appears, then, that the brilliant fish distinguishes the two by their behavior; a mode of sex recognition pointed out by Holmes ( 1903) in the case of amphi- pods. In the case of very young males the sex recog- nition must be wholly of this character, while males which already show some little sexual coloration are probably distinguished upon near approach by means of it as well as by behavior." 1 Among the amphibians the recognition of sex in the frog has been the subject of several interesting experiments by Goltz. In frogs and toads the males clasp the females during the breeding season until the eggs are discharged when the male sheds his sperm over them. What it is that induces the male to discharge his sperm at the opportune moment when the eggs are passing from the female has never been satisfactorily cleared up. The clasping of frogs insures that the male is on hand when his services in fertilizing the eggs are required. The clasping instinct is a temporary one, coming on early in the spring, and then ceasing after the short breeding period is past, when the sexes scatter and pay no further heed to one another's existence. As is well known, male frogs and toads often clasp various objects during the breeding season. Frogs have been found to clasp fishes the eyes of which they some- times gouge out with their thumbs ; and I have taken a male toad industriously clasping an old dried apple. 1 Biological Bulletin, Vol. 14, p. 35, 1907. 232 Studies in Animal Behavior The instinct to clasp is one of great strength and male toads may suffer their bodies to be cut in two without relinquishing their hold on the female. Correlated with the appearance of the breeding instinct there occurs increased development of the inner digital muscles and certain other parts of the fore legs of the males, a development which Nuss- baum has shown to be checked if the males are cas- trated a considerable time before the onset of the breeding season. Probably as a result of internal secretions of the reproductive glands, parts of the neuro-muscular mechanism become at this time pecu- liarly irritable, so that a particular form of reflex activity is very easily evoked. But notwithstanding this, a frog or toad seldom clasps for long anything but the female of his own kind. Other males may be clasped, but they are usually soon relinquished, while a female is clasped the more firmly the longer she is held. How does the frog distinguish male from fe- male ? Goltz has found that blinded frogs discrimi- nate between the sexes as well as normal frogs. After the olfactory nerves were cut he found that males can still distinguish females, so that neither sight nor smell is a necessary element in the recog- nition of sex. Even when the males were robbed of both sight and the sense of smell many of them succeeded in clasping the females among which they were placed. When the females were rendered mute by an operation they were no longer seized. The Recognition of Sex 233 Males, Goltz concluded, are not drawn toward the females through one sense alone, but by means of several senses, no one of which is indispensable. It is a striking fact that a frog whose head is cut in two so as to remove the cerebral hemispheres and eyes will nevertheless continue to clasp a female that is presented to him, while he soon rejects one of his own sex. The bodies of females are commonly plumper than those of the males. May the latter perchance distinguish the females by their form? Goltz tried the experiment of stuffing out the body of a male frog and giving it to another male, but he found that it was soon abandoned. Does the male frog have so delicate a tactile sense that even though nearly brainless he cannot be deceived as to which sex is within his grasp? Goltz is inclined to con- sider that such is the case. It may be open to doubt, however, if the possibility was sufficiently consid- ered that sex recognition may be a result of the behavior of the two sexes, much as it was found to be in amphipods. Despite the interesting experi- ments of Goltz the matter requires further investi- gation before a decided conclusion can safely be drawn. In the birds the sexes may often be easily dis- tinguished by sight, and in many species each sex is doubtless able to recognize the other by this sense alone. The discrimination may be aided by the ob- servation of differences in behavior. There is little 234 Studies in Animal Behavior evidence from the behavior of birds that the sense of smell is relied upon in this matter to any degree. In those birds in which the two sexes are much alike, as in most pigeons, differences in behavior ap- parently afford the chief means by which each sex distinguishes the other. Craig in his interesting ac- count of the expressions of emotions in pigeons 1 says that "If a cage containing an unmated male ring- dove be suddenly brought alongside another cage containing another ring-dove, of unknown sex, the male becomes highly excited at once, and gives vent to his excitement in all possible ways. First he bows and coos with all his might, and he con- tinues to do so for a long time. Then he charges about the cage, assuming the attitude peculiar to the charge, and frequently repeating the loud kah- of-excitement. At intervals, he stops to glare at the strange bird and sometimes to peck at it through the bars, but soon he starts again to bow-and-coo and charge. . . . "If left beside the stranger's cage for some hours, the male must sometimes rest and be silent; but even the intervals of rest and silence are broken fre- quently by series of perch-coos. This behavior on the part of the male is useful in that it stimulates the strange bird to respond, and, in responding, to reveal its sex. "If the strange bird be a male, it shows similar 1 Journal of Comparative Neurology and Psychology, Vol. 19, p. 29, 1909. The Recognition of Sex 235 excitement and aggressiveness. And the two males are sure to fight if they can reach one another. "But if the strange bird is a female, she acts far otherwise. She is at first very indifferent, unless she is particularly anxious to mate. And after some days, when she begins to show an interest in the male, she does not give the bowing-coo, nor charge up and down the cage, nor show other signs of pug- nacity and aggressiveness. So far from tending to aggress the male, her conduct is rather an expres- sion of submission to him. She shows a certain excitement; for instance when she utters the kah it is a kah expressive of gentle excitement. But she spends the greater part of her time in alluring the male by means of the nest-calling performance — the nest-calling attitude, seductive cooing, and gentle flip of the wings. She often tries to get through the bars of her cage to the male; and, failing to do so, she sometimes lies down with one side pressed against the bars. . . . "When the male sees the strange bird behaving in this submissive and seductive manner, he loses the intensity of his pugnacity; though he always con- tinues to be masterful. He spends less time now in the bowing-coo and more time in nest-calling and in trying to get to the female." In the mammals the sense of smell plays a much larger part in the recognition of sex than it does in birds and the lower vertebrates. Not only is the sense of smell as a rule acute, but scent glands 236 Studies in Animal Behavior of various kinds are of frequent occurrence in one or both sexes, and frequently the secretion of these glands is exceptionally abundant during the breed- ing season. Commonly scent glands are better de- veloped in the male sex. The strong odor of the male goat is notorious. In the male elephant there are glands on the side of the face which, in the breeding period, enlarge and emit a milky secretion. In the males of many species of deer and antelope there are facial glands that are especially active in the rutting season. Other species have scent glands on the feet and limbs, or near the tail. It is not improbable that the secretion of these glands, while not particularly agreeable to ourselves, may have an alluring influence on the opposite sex of the species concerned. There is abundant evi- dence that different species of mammals are able to recognize their own kind through the sense of smell, and it is a well-known fact that many mam- mals are exceedingly quick to detect the scent of an approaching enemy. Any one who has watched the behavior of dogs in taking a sniff at their different acquaintances, or in getting a fuller olfactory im- pression of a stranger, will realize somewhat to how great an extent experience with odors makes up the dog's mental world. Many facts indicate that mammals distinguish the opposite sex of their own species through the sense of smell, but as the sexes frequently differ in external appearance they are undoubtedly able to recognize one another by References 237 sight. A certain familiarity with the habits of street curs will convince one, I think, that the element of behavior, as in some of the cases previously de- scribed, plays a certain role also. The recognition of sex has been little analyzed in the mammals. The problem is more complex than in lower forms owing to the higher development of the mammalian mind, and the fact that several dif- ferent senses are usually involved. REFERENCES ANDREWS, E. A. Conjugation in the crayfish, Cambarus affinis. Jour. Exp. Zool. 9, 235, 1911. BANTA, A. M. Sex recognition and the mating behavior in the wood frog, Rana sylvatica. Biol. Bull. 26, 171, 1914. CRAIG, W. The expression of emotion in the pigeons. I, The blond ring dove (Turtur risorius). Jour. Comp. Neur. Psych. 19, 29, 1909. FABRE, J. H. Souvenirs entomologiques, 7 serie. GOLTZ, F. L. Beitrage zur Lehre von den Funk- tionen der Nervenzentren des Frosches. Berlin, 1869. HOLMES, S. J. (i) Observations on the habits of Hyalella dentata. Science, N. S. 15, 529, 1902. (2) Sex recognition among amphipods. Biol. Bull. 5, 288, 1903. (3) Sex recognition in Cyclops. Biol. Bull. 16, 313, 1909. PARKER, G. H. Reactions of copepods to vari- 238 Studies in Animal Behavior ous stimuli, etc. Bull. U. S. Fish Com. 21, 103, 1901. REEVES, C. D. The breeding habits of the rain- bow darter (Etheostoma coeruleum Storer), a study in sexual selection. Biol. Bull. 14, 35, 1907. XIII THE ROLE OF SEX IN THE EVOLUTION OF MIND / I AHE reason for the existence of sex is one of •*• those biological problems whose solution seems as remote as it did a century ago. Many remark- able discoveries have been made in regard to the microscopic structure and development of the germ cells in both* plants and animals. We have learned much of the general biology of sex, and the prob- able evolution of sex in the organic world. And substantial progress has been made with the old problem of the determination of sex. But to the question, Why came there to be two sexes at all? or in other words, Why do not organisms continue to reproduce asexually as it is probable they once did? we can only offer answers that, to say the least, are very hypothetical. While the fact that sex is absent in the lowest forms of life indicates that evolution has proceeded at least a certain distance without its aid, and sug- gests the possibility of the evolution of sexless forms of a high degree of organization, yet the general prevalence of sex, with but rare exceptions, in all but the most primitive organisms points to the con- clusion that sex has played a fundamental role in 239 240 Studies in Animal Behavior the evolution of the organic world. It is doubt- less futile to conjecture what the organic world would have been like if the institution of sex had never been evolved. Even if the processes of varia- tion and selection had gone on to the same extent — which is scarcely probable — the absence of sex would certainly have given to evolution a very different direction from that which was actually followed. Many of the most complex of the structural ar- rangements of organisms have especial reference to securing the meeting of the germ cells. The color and scent of flowers, and their many and beautiful adaptations to effect cross fertilization would never have appeared if plants were propagated solely by the asexual method. In animals the structural pe- culiarities associated with sex are, as a rule, among the most complex features of the body. Correlated with these structures we find mating instincts which frequently manifest themselves in complex modes of behavior. More acute senses have been evolved in many cases very largely in relation to securing the meeting of the sexes. The large antennae of male moths and several other insects, the larger eyes of the common drone bee, and the auditory appara- tus of the male mosquito are a few of the numerous illustrations of this fact. The various kinds of apparatus in insects for mak- ing sounds which are found in crickets, locusts, and cicadas are devices for drawing the sexes together, and the complementary development of auditory or- The Role of Sex in the Evolution of Mind 241 gans in the same insects has doubtless been greatly furthered through the evolution of these structures. The primary function of the vocal apparatus of the vertebrates was probably to furnish a sex call, as is now its exclusive function in the Amphibia. Only later and secondarily did the voice come to be em- ployed in protecting and fostering the young, and as a means of social communication. And the evo- lution of the voice in vertebrates doubtless influ- enced to a marked degree the evolution of the sense of hearing. It is not improbable, therefore, that the evolution of the voice, with all its tremendous consequences in regard to the evolution of mind, is an outgrowth of the differentiation of sex. In cases of degeneration through parasitism or other causes the female often proceeds much farther on the downward path than the male. In the scale bugs, for instance, the females have lost their wings and many other structures, while the adult male re- mains an active and graceful winged insect. The necessity for finding the female has kept the male from undergoing the degeneration that has over- taken the other sex. Much of the elaborate organization of the imago state of insects has reference to activities directly or indirectly concerned with mating and depositing the eggs in the proper environment for the develop- ment of the young larva?. There is a relatively long nymphal or larval period chiefly devoted to the vegetative functions of assimilating nutriment, 242 Studies in Animal Behavior growth; in many species the imago takes no food or need take none, before the eggs are fertilized and laid; and in several forms the mouth parts have become so atrophied that food taking is impossible. Some insects mate soon after they emerge from the pupal covering. In the May-flies, which live but a short time in the winged state in order to mate and deposit their eggs in the water, it is probable that the imago stage would long ago have disappeared were it not retained as a means of effecting the union of the sexes. So also with many other insects. In the winged state numerous new enemies are encount- ered and many lives are lost; in the pupa stage, which prepares for it, there is commonly an exten- sive tearing down of old structures and the building up of new ones during which the insect is helpless against many enemies and parasites. There are compensatory advantages in the possession of the imago stage in scattering the species into new re- gions, and in many other ways in the different groups, but were it not for the necessity for preserving the mating activities which occur in this period of the insect's life-history, it is probable that the complex organization of the adult state would very fre- quently have degenerated, or even been lost, if it had not failed to develop at all. The mating activities are almost everywhere among the most complex performances of an ani- mal's life. The opposite sex must be distinguished from all other creatures and responded to accord- The Role of Sex in the Evolution of Mind 243 ingly. Often pursuit and capture or winning over are the necessary preliminaries of mating. All this puts a premium so to speak on the sharpening of the senses, the development of strength and acute- ness, and the evolution of higher psychical qualities. Consider the mating activities of crustaceans, the courtship of spiders, the breeding activities of fishes, and still more the elaborate wooing of male birds, and it will become manifest how greatly the insti- tution of sex must have stimulated the evolution of more complex modes of behavior. All the facts here cited are trite enough even to the non-biological reader. But while it is sufficiently evident that the differentiation of the sexes has pro- moted the development of behavior in relation to mating, it may be well to point to the enormous in- direct consequence of this development in respect to the evolution of mind in general. In the evolu- tion of behavior one kind of instinct grows out of another just as new organs are usually formed by the elaboration of some pre-existing structure. A general elaboration of instinctive reactions in re- gard to any one sphere of activity affords a basis, therefore, for the differentiation of more complex or specialized behavior in respect to other activi- ties. The primary function of the voice, as has al- ready been pointed out, was to serve as a sex call. Later it became the means of various instinctive forms of communication and finally afforded the medium of articulate language. Had it not been for 244 Studies in Animal Bbha-vior its value in the mating of the lower vertebrates the voice might never have been evolved and man never have become man. While the specialization of senses, which, in cer- tain cases at least, has been carried on mainly for sexual purposes, naturally afforded a basis for the elaboration of many instincts, it is practically im- possible to trace in detail how various instincts, sex- ual and other, may have acted and reacted on one another's development. But we can discern enough of the influence of sex differentiation on the evolu- tion of behavior to feel assured of its importance. The necessity of solving the one problem that con- fronts all dioecious animals which do not simply shed their sexual products at random into the water has kept behavior in one sphere up to a certain mini- mum standard. The male must find and impregnate the female, and this fact sets a certain limit to his degeneration, at least in some period of his life. But besides acting as a check to degeneration, the necessity for mating has, in general, been a con- stant force making for the evolution of activity, en- terprise, acuity of sense, prowess in battle, and the higher psychic powers. We cannot pretend accu- rately to gauge its role in the evolution of mind, but it has evidently been a factor of enormous potency. XIV THE MIND OF A MONKEY T IZZIE was first seen in a store on Market •'—'Street, San Francisco, where she was confined in a cage with a small puppy which was put in for company. She was a specimen of bonnet monkey, Pithecus sinicus, and she had been recently imported from India, so her owner averred, but during her short captivity she had come to be quite tame and tractable. A few days later she became the prop- erty of the University of California, and was kept in a cage especially constructed for her reception in a sort of storeroom belonging to the department of zoology. Owing perhaps to the strangeness of her new surroundings, or to the loss of her old as- sociates, Lizzie frequently gave vent to a plaintive cry, but she seemed to be appeased when any one came near. When let out of her box she began to scamper about, climbing up tables and other objects, and examining things critically all about the room. If approached she would often utter a sound resem- bling a bark and stand with her mouth open in a threatening attitude, at the same time being on the alert to make her escape. She proved to be re- markably agile, even for a monkey, and very quick 245 246 Studies in Minimal Behavior to discover the least movement anywhere within her range of vision. She would move about almost con- stantly, but her attention was not directed to any one object for more than a few seconds at a time. Lizzie showed a strong aversion to being taken in the hands, although she soon came to jump upon my shoulder and ride about there quite contentedly. Often when I stretched out my hands to seize her she would bound past them to my arm and quickly scamper to my shoulder. It was difficult to get hold of her in that situation, for she would clamber about over my body in a very nimble way in her efforts to avoid seizure. She was fond of diving into my pockets and extracting articles therefrom and then scampering away with them. She appeared to take a certain pleasure in being pursued for the recov- ery of the stolen property. Most things which she took went straight to her mouth. She was especially fond of chewing up lead pencils, and took an ap- parent delight in breaking things or pulling them to pieces. After a detailed investigation of an object for some minutes, during which she turned it over and over with her hands and feet — for she was al- most as facile in grasping things with her feet as with her hands — she usually wearied of her play- thing and gave it little further attention. This sort of intellectual curiosity afforded her many things with which to occupy herself; and when no other object engaged her attention she would frequently busy herself with inspecting her own person in the The Mind of a Monkey 247 pursuit of possible parasites. One marked trait of Lizzie's behavior was the ease with which she became alarmed at any unusual object or occurrence. After some months of ac- quaintance, when she would sit contentedly on my shoulder, any quick movement would inspire her with fear. A certain instinctive dread of being taken un- awares seemed to be an ineradicable part of her mental make-up. Bred to a life of continued watch- fulness and fear in the forests of her native home, she was gifted, to a very unusual degree, with the faculties that make for the ready detection and avoidance of danger. For keenness of perception, rapidity of action, facility in forming good prac- tical judgments about ways and means of escaping pursuit and of attaining various other ends, Lizzie had few rivals in the animal world. She frequently surprised me by getting out at a half-opened door which I thought I had effectually guarded, or in grab- bing a bit of food from me which I was confident she could not reach. Her perceptions and decisions were so much more rapid than my own that she would frequently transfer her attention, decide upon a line of action, and carry it into effect, before I was aware of what she was about. Until I came to guard against her nimble and unexpected maneuvers she succeeded in getting possession of many apples and peanuts which I had not intended to give her except upon the successful performance of some task. In disposition Lizzie was gentle and tractable, and 248 Studies in Animal Behavior did not make more than a rather playful pretense of biting one's hand. Of affection or attachment to any one, such as has been described in other kinds of monkeys, Lizzie showed scarcely a trace. She seemed to enjoy the presence of people much as she would be gratified in examining a new kind of object; the pleasure she derived was an intellectual one rather than an emotional satisfaction such as a dog takes in his master. When she was irritated, which was easily done, she would give a sort of bark and face one with an open mouth and general attitude of attack. Fits of temper wore off very quickly, for she was very changeable in her moods as well as in her mental pursuits. She was also quickly over her fears. With her, sufficient not only for the day but for the moment was the evil thereof. She wasted no time in retrospection; she lived en- tirely in the present, and in one of very narrow span. While Lizzie was a most admirably efficient little mechanism for getting on in life under the con- ditions of a tropical forest, she proved to be quite stupid when tested according to certain other stand- ards. In order to get a clearer insight into her mentality, recourse was had to a number of experi- mental tests. The front of Lizzie's cage was made of vertical bars which were far enough apart to allow her to reach out her arm between them. A board was placed just outside her cage and a piece of apple put on the board beyond her reach. Lizzie grabbed at the apple, pushing the board to one side. The Mind of a Monkey 249 Then she tried to seize the board by one end and pull it toward her, but as the board was a little too heavy for her to move well in this manner, a nail was driven in the middle of the end of the board nearest the cage to serve as a sort of handle. When n B FIG. 11. — Diagram of Lizzie's cage with the board B, on which food was placed, n n, position of nails. board and apple were placed as before Lizzie reached immediately for the nail, pulled the board in and got the apple. A repetition of the same ex- periment was followed at once by the same result. During the third trial Lizzie attempted to seize the board by the side and pushed it away out of reach. When the board was replaced she pulled it by the 250 Studies in Animal Behavior nail and got the apple, as she did also at the next trial. In these experiments, even at first, Lizzie did not reach for the apple directly. She seemed to appre- ciate from her inspection of the situation that the apple could not be secured in this way. Her first efforts were directed to the means of attaining the desired end, and when the nail was driven into the board she seemed to apprehend at once how it could serve her purpose. There was no employment of the method of trial and error; there was direct ap- propriate action following the perception of her re- lation to board, nail and apple. After the fourth trial, when the board was in its usual position and before the apple was on it, Lizzie reached out and pulled it in. In the fifth trial, when the apple was replaced, she seized the board by the side and pulled it in after considerable effort and got the food. Then the board was replaced with no apple on it, but it was pulled in again in appar- ent expectation of the usual reward. This futile performance was repeated several times. Finally Lizzie grew weary of her wasted efforts and would no longer respond. Then a piece of apple was placed about six inches to one side of the board. After making an ineffectual attempt to reach the apple directly, Lizzie seized the board and pulled it in. She did this six times, after which she sat looking at the apple and whining. Then I placed the apple on the board, which was immediately pulled in by The Mind of a Monkey 251 the nail. After her appetite was whetted another piece of apple was placed six inches to one side of the board as before. Lizzie expectantly pulled the board in and repeated the performance six more times, but her actions became slower with each dis- appointment, until after the sixth trial she gave the board up and tried to reach the apple directly. Then I held the apple near the cage to give her a smell of it and replaced it near the board. Being thus stimulated, Lizzie pulled the board in by the nail three times, when she gave up the task. After being tempted as before she pulled the empty board in three times. In these experiments Lizzie showed that she had associated the act of pulling in the board by the nail with obtaining and enjoying the apple. But her persistence in pulling in the board when she could clearly see that the apple was several inches away from it showed that she exercised little discrimina- tion, and indicated that the associations she had formed were of a rather vague and hazy kind. In another experiment I placed the apple further out on the same board so that she would be unable to reach the food when the near end of the board was against the base of her cage. Lizzie pulled the board in at once and reached for the apple. Finding it too far out, she pushed the end of the board sidewise, at the same time keeping it against the base of the cage. This brought the apple nearer and she got it. The experiment was repeated sev- 252 Studies in Animal Behavior eral times and Lizzie solved her problem each time with little or no hesitation, as in the first trial. The problem was then made a little more difficult by placing the apple still farther out on the board so that she could not reach it even when she had pushed the end of the board as far to one side as the limits of her cage would permit. When she had pulled the board in and to one side, finding that the apple was still out of reach, she tried to seize the board by the side and to pull it in sidewise. It was too difficult for her to get a good hold of the board in this way, and her attempts were not suc- cessful. I then drove a nail near the middle of the board. Getting the apple involved pulling the board to the cage by the first nail, pushing it then to one side so as to bring the second nail within reach, seizing the board by the second nail and pull- ing it sidewise toward the cage until the apple was sufficiently near. At her first trial Lizzie pulled the board in by the first nail, then pulled it sidewise, and tried to seize the edge of the board. Appar- ently by accident her hand struck the second nail, which she seized at once, and by its means pulled in the board and got the apple. In the second trial Lizzie pulled the board in and to the left, then reached immediately for the second nail and pulled in the board toward her cage. In several subse- quent trials she secured the apple in just the same way. The problem was solved perfectly after the first trial. The Mind of a Monkey 253 In another set of experiments Lizzie was given a vaseline bottle containing a peanut and closed with a cork. In accordance with her instinct to bite at new objects Lizzie attacked the cork with her teeth, pulled it out and tried to chew it, holding the bottle meanwhile in one hand; then she noticed the nut when she transferred the cork to her feet, and tried to reach the nut, but the neck of the bottle was too small for her hand to enter. On turning the bottle over the nut dropped out unobserved and I replaced it and put in the cork. Lizzie immediately drew the cork and held it in her hind feet while she tried to reach the nut with her fingers. Finally, when she was holding the bottle upside down, the nut came within reach of her fingers and she got it out. When given another nut in a corked bottle she pulled the cork and tried to reach the nut with the bottle up- right, but in the course of her efforts she turned the bottle over so that the nut fell down within reach, when she got it. Without describing in detail her subsequent trials, I may say that Lizzie gradually came in the course of fifteen trials to turn up the bottle very soon after she received it and to get the nut much more quickly than at first. She never came to turn the bottle over and let the nut drop out. She was too busy trying to reach it with her fingers to get it by the easiest method. Even after she had come to get the nut rather quickly she often spent considerable time in attempting to reach the nut when the bottle was held 254 Studies in Animal Behavior upright. She did not pick out the essential acts that led to success. She perceived that the nut could be secured by going through a certain series of mo- tions, and the useless movements were gradually eliminated with an average shortening of the time necessary to gain the desired end. So far as her progress is concerned after she had removed the cork from the bottle, she gave no evidence of clearly perceiving how anything that she did furthered her purpose. Apparently she did not clearly apprehend that if she turned the bottle upside down the nut would fall down within reach of her fingers, al- though she had seen the nut fall dozens of times. In the course of her intent efforts her mind seemed so absorbed with the object of desire that it was never focussed on the means of attaining that ob- ject. There was no deliberation, and no discrimi- nation between the important and the unimportant elements of her behavior. The gradually increasing facility of her performances depended on the appar- ently unconscious elimination of useless movements. The previous experiments were modified by giv- ing Lizzie a nut in a screw-cap Mason jar without a cover. She could easily reach into the wide mouth and get the nut, but she picked up the jar instead and turned it about. Having accidentally dropped the jar, she scuttled away in alarm, but she cautiously approached it again, turned it over and got the nut. Then she picked up the jar, carried it to her perch, rolling it over and over with her hands and feet in The Mind of a Monkey 255 various ways. I took the jar and put another nut in it, but Lizzie continued to play with the jar and let the nut drop out unobserved. When the nut was replaced Lizzie tried to get it by biting the glass; in the course of her turning the bottle around, the nut dropped out and she picked it up. When an- other nut was put in Lizzie reached in at once and got it. Another nut was obtained in the same way, but at the next trial she turned the jar around in various ways until the nut fell out; and in numerous other trials on different days she sometimes got the nut by reaching it directly and at other times by turning the bottle around until the nut dropped out. Fifty or more trials did not teach her to secure the nut at once by inverting the bottle. While she came to get the nut by reaching into the jar more often than she did at first, she did not settle down to any uniform method of procedure. When a nut was placed in the jar and the cover screwed on very loosely, Lizzie tried to pull the cover off by using her hands and teeth. After much effort she succeeded and held the jar with a hand and a foot and the cover with the other hand. The novelty of the cover engrossed her attention and she let the jar drop. Soon she went to the jar, reached in and got the nut, and then resumed her investiga- tion of the cover. Another nut was placed in the jar and the cover screwed on very loosely as before. Lizzie took the jar to her perch, worked the cover off with her hands 256 Studies in Animal Behavior and teeth, and then reached in and got the nut, hold- ing the jar upright with her feet. The next trial resulted in practically the same way. The cap of the jar was then screwed -on farther. Lizzie at- tacked the jar industriously and finally removed the cover, although working quite unsystematically. The cover was put on as before and Lizzie worked at it about fifteen minutes, getting more and more ex- cited and impatient over her lack of success; some- times she tried to bite through the glass at the lower edge. After turning the cover this way and that she finally unscrewed it and got the nut. After numerous trials Lizzie never learned to unscrew the cover by turning it around uniformly in one direc- tion. She simply worked it back and forth until it happened to become entirely unscrewed. Lizzie was then set to the task of getting food out of a small box. Two sides of the box were made of strong wire netting; the rest was wood. In one corner was a small door which could be fas- tened by a hook passing through a small screw eye. In the first experiments a piece of apple was placed in the box and the door, which stuck rather tightly, was left unhooked. Lizzie looked at me while I put the food through the door and she opened the door at once and got the food. She did the same at the second trial, after which, when the food was replaced, the box was turned so as to lie on an- other side. This seemed to disconcert Lizzie and she tried biting and clawing at the wire netting, and The Mind of a Monkey 257 turning the box over and over until she became dis- couraged. I recalled her to the task by tapping on the box, but evoked only feeble efforts. When I opened and closed the door Lizzie observed me and went at once to the door and got the apple. Then I replaced the apple and closed the door and put the box in another position. Lizzie attacked the box in various places and then desisted. Soon she looked at the box, went to it as if an idea struck her, and tried to pull the door open by using hands and teeth ; finally, after some tugging, she succeeded. After a few more successful trials the door was fastened with the hook. Lizzie attacked the door with hands and teeth and turned the box over and over and often tried to get the apple through the wire. A renewed attack on the hook enabled her to get the door open and get the apple. The next trial on the succeeding day was followed by much the same method of at- tack. After biting at the hinges and various other parts of the box Lizzie loosened the hook and opened the door. Not to weary the reader with the recital of Liz- zie's misdirected efforts and slow progress, it may be said that she gradually came to concentrate her efforts on the door, but even after thirty trials she would bite at the hinges and edges of the door, and not infrequently she would turn the box over and bite at the wire netting. In all of her efforts at the hook she never learned to pull it to one side. She simply tugged at it this way and that with her 258 Studies in Animal Behavior teeth until it came undone. The mechanism of the thing, how the hook stood in the way of opening the door, she could not understand, simple as it was. When a button was substituted for the hook her mode of attack was much the same; and her prog- ress, such as she made in the course of thirty trials, was after the same slow method. She never per- ceived that when the button was turned in one di- rection it left the door free to come open, and that it prevented the door from coming open when it was in another position. She bit and worried away at the button, and pulled at the door until she got it open and got her food. The idea of the thing never got into her head. When both the hook and the button were used Lizzie had a very hard time to get her food. Oc- casionally after much varied and fruitless effort she would succeed. If she got the button turned right she would usually turn it the wrong way before she undid the hook. The experiments would probably have discouraged her observer had they not usually wearied their subject before she met with success. There was little hope that she would be able to solve more complex problems. A peanut that was hung below her at the end of a cord she obtained by pulling up the cord, hand over hand, the very first time she saw it. I tried to teach her to use a stick to pull in food with, as monkeys have sometimes been described as doing, but met with no success. Placing a bit of food out- The Mind of a Monkey 259 side the cage, I poked it about with the stick so as to give her a suggestion of how the stick might be employed to move the food within reach, but al- though the act was repeated many times, Lizzie never showed the least inclination to use the stick to her advantage. In fact, she never exhibited the least tendency to use any object as a tool. Next I tried suspending a piece of food beyond her reach and giving her a light box upon which she might mount and get the food. She did this readily enough when the box was in the right posi- tion. Then the box was pulled to one side in or- der to see if she would pull it back so that she could get upon it and reach the food. Although I fre- quently moved the box about to give her the sug- gestion and often put it in the proper place to en- able her to get the food, the idea of using the box in the way described never seemed to occur to her. Experiments with Lizzie were brought to a close by her death, but the results obtained were sufficient to give some insight into the nature and limitations of her mental endowments. While gifted with re- markably quick perception and in certain respects power of rapid judgment, nothing in her behavior gave any indication of the use of abstract or gen- eral ideas, or of deliberate reasoning. Neither did she exhibit the least tendency toward imitation, al- though I am not prepared to say that further ex- perimentation might not have revealed some evi- dence of this faculty. Some things, and even sim- 260 Studies in dnimal Behavior pie things, she apparently learned by the primitive method of the gradual elimination of useless move- ments after attaining a chance success. This type of learning is the one mainly followed by the less- developed mammals, but in the apes the curve of learning simple things usually shows a sudden de- scent from the start. One reason for her compara- tively slow progress in the experiments with the boxes and the bottles is, I suspect, that in her eager- ness to attain the desired end her attention was never strongly directed to the means employed. When we attempt to solve a puzzle we direct our attention to the means we employ and pass judg- ments upon them, but Lizzie never discovered the value of paying attention to method. Her impul- siveness and activity stood in the way of her at- taining any results that required a small amount of deliberation. The perception of very simple relations usually escaped her. She never clearly perceived that a hook could be unfastened by simply pulling it to one side, that a button would not hold a door closed when turned in a certain position; she probably never became clearly aware that when a bottle was turned upside down its contents would fall out. As we know these things, they involve a certain pre- vision, or representation to ourselves of how cer- tain things might happen if certain conditions were fulfilled. But this power was but slightly developed in Lizzie's mind. There are more indications of it The Mind of a Monkey 261 in Lizzie's performances in pulling in the board by the two nails. The quickness with which she learned the elements of the trick indicated that she perceived the way in which it might be done by simply in- specting the situation. But we should be cautious in our interpretations, because it was not known how near such actions might have been to her previous experience. Had she, for instance, been used to pulling in branches with fruit attached to them, pull- ing in the board might have been a particular ap- plication of some of her previous activities for which she may have had a strong instinctive bent. While it may not be safe to deny to Lizzie a certain amount of prevision in her performances with the board, we should hardly be justified in say- ing that they necessarily involved the drawing of an explicit inference. Should one ask if Lizzie were able to reason, the answer would have to depend on how reason were defined. That some of her acts are the outcome of simple inference, though per- haps not explicitly formulated in her mind, is quite probable. Even perception, as Spencer, Binet, and others have shown, is allied to inference; and Liz- zie's behavior evinces a much closer approach to the rational type than does the process of simple perception. Her behavior does not indicate so high a degree of mental development as that of several other monkeys that have been the subject of ex- periment. Whether her relative ineptitude for cer- tain tasks is an individual peculiarity or a trait char- 262 Studies in Animal Behavior acteristic of her species cannot be stated with any assurance. As I have remarked in another work, "We are apt to overestimate the importance of the ability to reason as if it were the chief thing of value in intelligent behavior. There are other mental traits which may enable an animal to get what it wants better than an increment of reasoning power. Gen- eral activity, power of attention, interest, quickness of forming associations, delicacy of discrimination, duration of memory, and the ability to form com- plex associations are all of the utmost importance in many situations of an animal's life. . . . Give a fox greater power of inferential thinking, but de- crease his alertness, curiosity, suspiciousness, and quickness of perception, and he might fall a victim to the hunter while his mind was employed on some other subject." Possibly more intelligence of the human sort would have been a positive drawback under the conditions of Lizzie's natural environ- ment. INDEX Allen, B. M., 188 Amoeba, behavior of, 32, 127, 184 Amphipods, phototaxis in, 57, 58 Andrews, E. A., 229 Anemones, behavior of, 125-127 Aquinas, St. Thomas, 14, 15 Arenicola larvae, phototaxis in, 56, 57 Aristotle, 11, 12, 15 Bain, A., 45, 133, 134, 139, 163, 164 Balanus larvae, phototaxis in, 98, 99, 102, 105 Baldwin, J. M., 28, 30, 133 Bancroft, F. W., 89, 107 Bauer, V., 106 Bert, P., 53 Belostoma, death feigning in, 202, 209-214 Binet, A., 177 Bohn, G., 125 Bonjeant, Father, 16 Brundin, M., 80, 81 Bunsen-Roscoe Law, 89, 90 Butterflies, orientation of, 86- 88 Carpenter, F. W., 93 Carausius, death feint, 209, 214, 916 Cells, behavior of, 177 ff. Celsus, 12 Chloroplasts, reaction of to light, 96, 97 Cicero, 13 Child, C. M., 166-169 Cole, L. J., 65 Compensatory motions, 63-66, 84 Condillac, E. B., de, 19 Copepods, sex recognition in, 227-229 Craig, W., 234, 235 Crayfish, intelligence of, 129, 136 Cytotropism, 180 Daphnia, phototaxis in, 98, 102, 105-108 Darwin, Ch., 25-28, 47, 51, 199, 205 Darwin, E., 17-19 Davenport, C. B., 113 Death feigning, 197 ff. DeCandolle, A., 52 De Geer, Ch., 213 Descartes, R., 15, 16, 21, 22 Dewey, J., 10 Dice, L. R., 102, 103, 105, 106 Differential sensibility, r61e of in orientation, 82-90 Driesch, H., 181 263 264 Index Drosophila, effect of intense light on, 93, 94 Earthworm, phototaxis in, 65, 71-73, 75-77, 113 Eimer, O. E., 28, 122 Electrotaxis, reversal of, 107 Epithelial cells, behavior of, 184-187 Eudendrium, effect of light on, 90 Euglena, orientation of, 89 Ewald, W. F., 90, 102 Fabre, J. H., 38, 197, 198, 212, 220-222 Fasten, N., 192 Fiddler crabs, phototaxis in, 81 Fishes, parental care in, 140, 143 recognition of sex, 230-231 rheotaxis in, 64, 65 Fiske, J., 46 Fly larvae, phototaxis in, 74-77 Forel, A., 121 Franc6, R. H., 28 Frogs, sex recognition in, 231- 233 Galen, 12 Gassendi, 17 Geotaxis, 61 Goltz, F., 231-233 Goodspeed, T. H., 107 Graber, V., 53 Groom, T. T., 98 Groos, K., 27, 30 Haeckel, E., 28, 177 Hall, G. Stanley, 30 Harper, E. H., 75, 76, 106 Harrison, R. G., 185, 189 Hegner, R. W., 188 Heliotropism, 52, 54, 96 Hobhouse, L. T., 33, 130, 134, 135 Holmes, S. J., 71-76, 80, 94, 114, 169 ff., 183, 185-187, 193, 201-206, 209-211, 213, 215, 222-228, 231 Hudson, W. H., 206, 207 Hume, D., 120 Hyalella, sex recognition in, 223-227 Hydra, behavior of, 125, 128 Hypnotism in animals, 214 Incubation, origin of, 41, 42 Inhibition, 131, 133-136, 139 in tropisms, 113-116 Instinct, analysis of, 30, 31, 51 origin of, 25, 26, 122, 123 Intelligence, beginnings of, 120 ff. critical study of, 33, 34 Jackson, H. H. T., 105, 106 James, W., 128 Jennings, H. S., 70, 71, 76, 89, 126, 127, 156, 158 Jensen, P., 161 Kirby, W., 199 Kirkpatrick, E. A., 30 Lamarck, J. B., 22-24, 122 Lamarckian factor, 28, 29, 160- 164 Learning, problem of, 139 ff. Leech, phototaxis in, 71, 73, 82 Index 265 Leibnitz, G. W., 17 Leptoplana, behavior of, 167, 168 Leucocytes, behavior of, 189-193 LeRoy, G., 17, 19, 22 Lewes, G. H., 122, 123 Loeb, J., 31, 53-55, 69, 77, 78, 81, 84, 89, 90, 98, 101, 102, 105, 113, 121, 183 Loeb, L., 185 Loxophyllum, behavior of, 125, 169-174 Lubbock, Sir J., 53 Lyon, E. P., 64 Malebranche, 16 Mammals, sex recognition in, 235, 240 Massart, J., 101, 190 Mast, 50, 56, 75, 76, 84, 89, 99, 109-111 McGinnis, M. O., 106 McGraw, K., 84, 85 Michener, E. R., 105, 106 Monkey, intelligence of, 245 ff. Morgan, C. Lloyd, 27, 28, 34, 121, 129, 177 Morgan, principle of, 34, 79, ISO Morgan, L. H., 207 Moore, A. R., 105, 107 Moore, B., 108, 109 Moths, sex discrimination in, 220-222 Nepa, death feint in, 209, 210, 212, 214 Nerve cells, behavior of, 188, 189 Notonecta, phototaxis in, 61, 80 Oltmanns, F., 99 Oppel, A., 185 Orchestia, phototaxis in, 57, 80- 82, 100, 101, 102, 105 Orientation, 54, 56-61, 69 if. Pain, relation to learning, 131- 137, 140, 143 ff. Paley, W., 20 Paramoecium, behavior of, 156, 157 electrotaxis of, 107 Parental care, 34 ff. Parker, G. H., 89, 103, 104, 228 Parmelee, M. F., 30 Pawlow, J. P., 150 Pauly, A., 28 Pearl, R., 112 Pearse, S. A., 229, 230 Pigeons, recognition of sex, 234, 235 Pigment cells, behavior of, 182- 184 Planarians, behavior of, 112, 166-169 Plato, 12 Pleasure, relation to learning, 131-137, 140, 143 ff. Pliny, 12 Plutarch, 12 Porphyry, 13 Preyer, W., 214 Protozoa, behavior of, 32, 33, 169 ff. Radl, E., 63 Ranatra, death feigning of, 201, 209-211, 213, 214 phototaxis in, 59-62, 80, 94, 95, 102, 104, 108 266 Index Random movements, 164 role in orientation, 70-78 Reason in animals, 9, 259-262 Reeves, C. D., 230, 231 Regeneration and behavior, 166 if. Reimarus, S. H., 19, 20 Rheotaxis, 64, 65 Rhumbler, L., 180 Robertson, T. B., 212, 213 Romanes, G. J., 26, 27, 28, 53, 79, 122, 214 Roux, W., 180 Royce, J., 66, 67 Sachs, J., 52, 53 Schmidt, P., 209, 214, 216 Severin, H. C. and H. H. P., 209, 210, 212, 213, 214 Sex in mental evolution, 239 ff. recognition of, 219 ff. Slime molds, behavior of, 193 Spence, W., 199 Spencer, H., 24, 25, 28, 31, 50, 51, 132, 133, 139, 163, 164 Spiders, death feigning of, 200, 212, 213 Stahl, E., 97 Stentor, behavior of, 125-127 Strasburger, E., 96, 97, 98, 101 Temperature, influence on tro- pisms, 94, 101-103 Terns, death feigning of, 205, 206 Thigmotaxis, 103-105, 115, 182, 184, 186, 189, 215 Thomasius, 17 Thorndike, E., 33, 139-142, 144, 149-151 Torrey, H. B., 89 Towle, E. W., 104 Trial and Error, 155 ff., 254 ff. Tropisms, relation of to in- stinct, 31, 32, 52, 62, 63 reversal of, 93 Verworn, M., 99, 214 Voice, primary function of, 241-243 Voltaire, 67 Volvox, reaction to light, 99, 107, 109-111, 114 Wasmann, E., 22, 27, 121 Weismann, A., 28 Whitman, C. O., 28, 41, 42, 123 Wilson, H. V., 194 Wundt, W., 28, 122 Talorchestia, death feigning of, Yerkes, R. M., 33, 104, 129 203, 204 phototaxis in, 57, 94 Zur Strassen, H., 195