Verplanck, W.S. (1958) Comparative psychology. Annual Review Psychology, 9, 99-118.

Comparative Psychology1

William S. Verplanck2

This chapter was prepared following a series of visits to laboratories and field stations where comparative psychology is under very active investigation. What has been observed, taken with this year's publications in the psychological, ethological, and biological journals, and the chapters of previous Annual Reviews, has given the chapter its form.

The advances in the past year in the topics usually allocated most space in this chapter, and that most psychologists think of when they think of comparative psychology at all have not been great. Some (1, 2, 3) might have been advances if they had appea red when the experimental work was completed. Another paper (4) provides a discussion of brain weights in more or less association with the report of some experiments on learning; the relation established is one of simple contiguity in the pages of a pap er. Little new appears on hoarding; investigations of bird navigation seem to have reached an impasse, where the only theory that seems adequate to the facts is untenable (5). [For a popularized summary, see Carthy (6).] The quasi-monopoly of the exper imental study of sex behavior in the male boojum, or rat, held by Beach and his colleagues, has been broken with the appearance of an excellent monograph by Larrson (7).

The general picture is a familiar one: advances have occurred, but at glacial pace; work on familiar variables inches along. A different picture arises, however, if the work of ethologists and zoologists in the field of behavior is examined in its own c ontext rather than according to psychologists' ideas of what should be important. Here the advances in the last year--in the last few years--have been rapid, remarkable, and of direct relevance to psychologists, whether comparative, experimental, social, or clinical. Ethologists, mostly European zoologists, study the behavior of a number of species from an objective (in the Watsonian or Pavlovian sense) point of view that does not exclude an active interest in physiological correlates of behavior. Prev ious chapters in this Annual Review will have familiarized the reader with some of their work, but they have placed, I think, undue stress on a few sets of investigations and misleadingly emphasized some now obsolete parts of ethological theorizing . What follows is based not only on the current publications of ethologists, but also on visits to their laboratories, and on many and long discussions with them.

Experimental facilities.--Through the past few years, several major research and teaching laboratories have been (or are now being) constructed, whose major features have been determined by the needs of research on animal behavior. The degree of e nvironmental control made possible in these laboratories extends considerably beyond the minimal air-conditioning all too rare in the United States; the very special environmental requirements for the maintenance of, and experimentation upon, a variety of species can be met.

Experimental animals.--Ethologists are more and more settling on a few laboratory species for their experimental investigations. This parallels the development of American animal psychology, and it has taken place for many of the reasons that hist orically determined the ascendancy of the rat. Analytic experimental research is becoming progressively limited to the stickle-back, the canary, the black-headed gull, the bee, the hen, and drosophila. Mammalian studies do not yet show such convergence- -perhaps because of the example we have given by our preoccupation with the rat, with its consequences.

Experimental design and treatment of the data.--Over the past few years, ethological research has become progressively analytic, with fuller and more careful control of the variables that must be taken into account in experimental design. The cont rols are both experimental ones (environmental manipulations) and statistical ones (appropriate designs). Together with these improvements in experimental technique has come a new emphasis on quantification, both in the measurement of behavior and in the association of probability levels with experimental findings. Methodologically, the ethologists have advanced significantly toward the experimental ideal of rigor, for lack of which they have frequently been criticized.

Theoretical orientation.--As one outcome of a number of papers (8, 9, 10) of a series of discussions at international meetings of various sorts3 and of any number of informal talks with experimental psychologists, as well as with each other, most e thologists have radically changed their viewpoint on the interrelationships between genetic structure and behavior: instinct is once again dead.4 Unlearned, or innate, means now to ethologists pretty much the same as it means to psychologists: what the animal brings with him to a set of observations that he doesn't seem to have had a chance to learn. That is to say, it merely sets a problem. Lehrman's (17) magnificent experimental exposition of how a complex species-specific behavior can be acquired has had its effect, and it is now commonplace to hear referred to as "probably learned" behaviors that a few years ago were categorized without qualification as instinctive or innate. Species-specific behavior is no longer solely attributed to genetic structure when the individuals of the species share prenatal environments and early experience, when they learn quickly and according to the same laws, when they show a common set of physiological mechanisms, and when they spend their lives in re markably uniform microenvironments. Unfortunately, psychologists do not seem to have learned as much from the ethologists as they from us, for we do not always show respect for species membership as a determinant of behavior (through whatever mechanism) that we might. [It is not, for example, appropriate to check statements about wild fowl, using domestic hens (18).] We do not match the growing sophistication of ethologists towards the many ways in which environmental events and physiological functions may determine both the structure and the behavior of the individual after fertilization.

The reader might conclude that, except for its superior technical facilities, ethology is becoming indistinguishable from comparative psychology as it is known in the United States. This is true only in part; there remain some characteristics of ethologi cal research that distinguish it sharply from most of what we do. The problems to which ethologists address themselves identify ethology as comparative psychology in the sense that comparative anatomy and comparative embryology are comparative sciences. It relates behavior not only to its immediate antecedent variables but to its broader biological context--to genetics, to taxonomy, to evolutionary theory, and to ecology. This comparative psychology, emphasizing the relation of behavior to general biol ogy rather than solely to physiology, returns to that familiar in the United States during the first decades of this century (but gone with the Journal of Animal Behavior). The present definition of comparative psychology in the United States seem s to be: all psychology dealing with animals that is not treated under learning, motivation, and physiological psychology; or, to state it another way, studies on animals that do not conveniently fit anywhere else. Again, ethologists are far readier to adopt working hypotheses and to drop them, once disconfirmed, as well. Finally, they are often concerned with certain specific behavioral phenomena (e.g., sign stimuli, displacement activities) that may no longer have any very special theoretical signifi cance but that still present live experimental and interpretative problems.

The research of ethologists, then, retains certain characteristics that differentiate it sharply from that of American psychologists. The biological emphasis has determined in large measure the selection of species and of problems for study. In this con text, largely unfamiliar to psychologists, some of the most interesting recent ethological research has been done. [For a general source on the biological context, see Carter (19).]

Several (necessarily overlapping) major areas of research may be distinguished. The first relates to general surveys of the behavior of one or of a number of species, usually descriptive. These are made under field or zoo conditions and range form super ficial surveys to intensive, protracted quantitative observation of large number of animals under controlled conditions. Such studies are the indispensable antecedents to analytic experimentation. More than once, in their absence, a psychologist has con trolled literally out of existence the necessary environmental conditions for, or necessary behavioral correlates of, behavior that he wished to experiment upon.5

The second area is the genetics of behavior, in a sense more restrictive than that in which strain differences in the lovableness of puppies or in the pugnacity of domesticated mice are referred to under the category "psychogenetics." Geneticis ts study drosophila; their genetic structure, and its relationship to morphology, are well known. What of their behavior?

The third area stems from the fact that related species behave alike. Taxonomy today is a science that deals not solely with the classification of species but with the most general problems of their origin and divergence in evolution. Behavior enters in to problems related to species in at least three ways; one of them is this: Behavioral differences between two populations that are morphologically identical may be the only way of telling them apart. The second relates to tracing out the evolutionary h istory of species or of a set of related ones. Finally, behavior is important in maintaining the reproductive isolation that serves to produce and maintain species. Much interbreeding that could occur does not do so only because the behavior (and associ ated morphology) of an individual of one group does not effectively control the sexual behavior of a member of another. Behavioral mechanisms contribute substantially to the development of new species and evolve with them by fostering genetic isolation.< BR>
The fourth major area relates the structure and behavior of members of a species to general evolutionary concepts: to adaptation to particular environments and to survival value--that is, to reproductive rates and to death rates. The generalization towa rds whose verification these studies are directed might be stated as follows: The behavior of an animal, like its structure, is the outcome of the interactions of the animal, and of its forebears, with its environment, animate and inanimate . . . with it s predators and prey, its mates and offspring. Pointedly, competition is omitted from this somewhat platitudinous statement of a not always obvious generalization.

A fifth research area cuts across the work of ethologists orthogonally to the above. One of the major contributions of ethologists is their detailed empirical account of the organization of particular individual responses into major response sets, identi fied by statistical covariance, and perhaps by common sets of antecedent variables. Besides such sets as attack, escape, and sexual behavior, the ethologists have identified a set called threat that occurs when tendencies to attack and to escape are appr oximately equal. (conflict). Threat includes a number of responses called displacement activities (at term that most ethologists qualify with an "in quotes"). The courtship behavior of both birds and fishes, analyzed in detail, includes respon ses that can readily be related to responses that appear in escape, attack, threat, and copulation proper. Aggressive behavior and fear, that is to say, are closely associated with sexual behavior; studies of sexual behavior, if pursued, lead to studies of aggressive behavior, and sometimes vice versa, unless the experimenter overcontrols the environment.

Displacement activities are those responses first encountered in a context in which their antecedents and consequents have become well understood but that also appear in other, quite different, situations where, in terms of the variables they had been rel ated to, they make no sense. A threatening gull may show behavior appropriate for nest-building; a bird in an approach-avoidance conflict (as food-shock) preens. The conditions required for the production of a given response or series of them as displac ement activities, remain a live experimental as well as a theoretical problem of special interest because many of them, ritualized, appear in the courtship behaviors that maintain genetic isolation.

A sixth line of endeavor looks toward the physiological bases of the relationships found--that is, it is concerned with providing a sound theoretical account of the immediate antecedents of behavior. Tinbergen's theory of instinctive hierarchies, expound ed in his well-known treatise (22), is obsolete and largely abandoned (especially by Tinbergen who emphasizes the theory's inadequacies perhaps more than do others). Alternatives are being sought.

Finally, something that might be termed applied ethology is developing. To be sure, a very large amount of research on behavior has been sponsored for good and sufficient economic reasons by schools of animal husbandry and the like, but, latterly, a new field of application has developed. This is the use of behavioral assays in pharmacology. A scattering of papers concerning behaviors that ethologists have worked on intensively has appeared, the work done by specialists in other fields. The inadequacy of these researches, as well as the striking behavioral phenomena observed, suggest that ethologists will find drug-studies an extremely fertile research tool--and vice-versa.

In the following sections we will present summaries of the past year's work in terms of these areas, recognizing that any classification forces one occasionally to be arbitrary in placing one or another study in this or that category.

General Studies

Hediger (23) reports a large number of observations on behavior in the field, in zoos, and in circuses. His tentative generalizations are a rich source for problems, especially on social behavior, that may be attacked directly in the laboratory. His book should be a valuable supplementary text. Further field observations are reported on whales and porpoises (24), three studies report briefly on the reproductive behavior of fish (25, 26) and scorpions (27), and two (28, 29) provide thorough and detailed descriptions of the behaviors of birds.

Two papers point up the usefulness of general surveys of the behavior of particular species. Dart (30) amusingly emphasizes that a reasonable set of data on the feeding habits of the hyena would settle a long-standing controversy on the origin of fossil deposits in caverns and hence provide a substantial contribution to our data on primitive man. [It is by no mans nonsense to refer to fossil behavior when its consequences are available for study.] The other (31), a very careful experimental study, show s that a minority of the domesticated albino rats and a majority of their wild cousins are both more systematic and effective in killing mice than are cats in killing rats. This difference conforms with many others observed and predicted between wild and domesticated strains. A stereotyped attack is usually followed by the eating of part or all of the victim, brain first. Nonkillers could not be induced to kill a mouse even though they ate fresh-killed ones presented when they were food-deprived. Nor could aversive stimulation that produced vigorous intraspecific fights induce these animals to attack mice. Nonkilling females, given live mice at parturition, adopted them, repeatedly retrieved them when they strayed from the nest, and added them to the litters, to the latter's disadvantage since the mice almost always moved into the warmest part of the nest and occasionally ate the new littermates. The survival rate of the litters into which the mothers adopted mice was low. The behavior of these rat s towards their mice was very clearly correlated with their endocrinological state. Amygdalectomy significantly reduced the incidence of killing by the killers. This behavior remains unrelated to others of the rat's behaviors that are known, and, in ter ms of them, is inexplicable; the fault perhaps lies in the absence of a sufficient variety of data on the animal's life history and behavior. The evidence suggests that the tendency of wild rats to kill mice is associated with the increased survival rate of the litters of mouse-killers in an environment where there are plenty of mice, with their tendency to eat neonate rats. This is a type of selection-pressure with respect to which albino rats have not been bred for many generations.

Genetics of Behavior

Strain differences have been investigated with respect to hoarding in rats (32) and audiogenic seizures in mice (33). Hinde (34) has studied the behavior of interspecies hybrid finches (Carduelines). If the parent species have identical behavior patterns, then these appear in unchanged form in the hybrids. Other responses, differing in the parent species, appear in intermediate form in the offspring, either quantitatively intermediate (where one species frequently shows the behavior and the other but rarely) or qualitatively (when the topographic patterns of the response vary). Behaviors that appear in only one of the parent species appear either unmodified in the hybrid or they do not appear at all. One significant result should be n oted-- "the relations between display components and behaviour tendencies are preserved (qualitatively) unchanged in the hybrids." That is, responses that are associated with aggressive behavior in the hybrid are also associated with aggressive behavior in the parents, and similarly for sexual and fear responses.

Of importance for the description of the action of the genes in the determination of behavior is the use of drosophila as experimental animals. Bastock & Manning (35) and Bastock (36) now report on the yellow mutant of Drosophila melanogaster, der ived from a single gene (sex-linked, recessive). Associated with the gene is a change in both the duration and frequency of the wing-display in the courtship of the male. This altered pattern of courtship is associated with reduced fertilization of norm al females by the male, although the courted female behaves toward the displaying mutant as she does toward a normal male. Two other papers also deal with the behavior of drosophila (37, 38); taken together these three papers suggest the fruitfulness of studying hereditary mechanisms in that species whose genetic structure is best known and most easily manipulable experimentally and whose behavior repertory is very small. It is still a bit difficult to see the scientific significance of genetic studies on some other species (39).

Behavior in Systematics

Taxonomic use of behavior.--Stokes (40) describes two morphologically identical species of gall-midges that parasitize different host-plants within the same genus; they have been classified separately because they produce quite different g alls. These species successfully breed only on a plant of the appropriate species. It is not yet known how this happens.

Simmons (41) treats with the head-scratching response of birds, of which there are two topographically differing forms, the same in all genera within a family but differing between families of the same order. Hinde (42), surveying the attack-flight-court ship-and-sex complex of behaviors as well as feeding, classifies the finches into two major groups, and Andrew (43) distinguishes among four separate classes of buntings. Elsewhere Andrew (44) treats in detail the intention-movements of flight that appea r in conflict situations; they act to maintain social groupings. Moynihan (45) presents preliminary data on the hostile aerial behaviors of a number of species of gull.

Mechanisms fostering genetic isolation (segregation of breeding).--For successful mating and for the segregation of breeding necessary for the evolution of a species, individual members of a given member of a strain of species must discrimin ate between co-members of their strain or species and members of other strains or species. They may or must also recognize individuals. Investigations of the variables and processes governing such recognition are of interest for reasons beyond the evolu tionary problems; they also provide data on learning under rather special constraints. Schloeth (46) describes the behavior shown when individuals first encounter other individuals of the same or of different species. This survey covers 814 different en counters. Hale's results (47) demonstrate that hens discriminate the breed of other hens and respond to them on the basis of their group-membership: behavior first displayed to one individual of another breed is generalized to other members of the same breed. In small mixed-breed flocks, one breed completely dominated the other. In larger flocks (about fifteen members), the effect was less complete. Alterations of appearance produced by dyeing or removing the combs of individuals altered the ability of other animals to recognize the individual but members of the other breed were still able to discriminate it from members of their own breed; they showed the appropriate dominant or submissive behavior to it. That is, disguised individuals moved withi n the social hierarchy with respect to members of their own breed but not with respect to members of the other. With trout, things are otherwise. Newman (48) shows that in general social behavior (dominance-submission relations and territorialism) brook trout tend to dominate the rainbows. One would welcome behavioral studies on the courtship and mating of these two species which must serve to maintain the reproductive isolation necessary for their maintenance.

A special problem of species recognition is that of imprinting. Some species of animals do not come equipped with a battery of species-specific behaviors serving to limit their social behaviors to members of the same species, but with a slight tendency t o respond (by following) to a broad class of environmental events (moving objects of a given size range) and a very strong tendency to learn to follow at high strength only those objects which they have had experience in following. In the laboratory, ind ividuals of these species learn to follow boxes, models, and people. One symposium (49) and a number of papers (50 to 54) now permit a set of generalizations whose significance is limited by the fact that not all investigators have worked on the same spe cies. First, imprinting, the greatly increased tendency to approach and to follow a given object, only rarely occurs in the absence of a history of extensive practice in following. Second, following is self-reinforcing; that is, no reinforcing stimulus has yet proven identifiable. It diminishes in strength with massed practice and with lack of practice; it is not irreversible. Third, practice in following must occur in the first hours and days of the individual's life. Fourth, the strengthened tenden cy to follow, produced by following a particular object, generalizes in typical gradient fashion to objects like it. Further experiments on imprinting will probably relate the behavior both to flocking (the running-together of members of a dispersed grou p of neonates) and to the vocalizations made by both the imprinted animal and by the moving object it learns to follow. Vocalizations may be significant whether they are distress cries or comfort (contentment) cries--the small notes given by aggregations of ducklings and goslings when they are together and when they are feeding or have just been fed. The latter vocalizations are analogous to the purring of cats and may be a form of behavior alternate to following in its correlation, once conditioned, wi th the whole body of social behavior.

Weidmann (54) points out that one probably significant difference between the behaviors of imprinted and nonimprinted ducklings is that when the latter are alone they stand or sit content, whereas the bird that has learned to follow some object gives dis tress calls, moves about, and in general shows appetitive behavior that ceases when the imprinted object comes into close proximity.

Marler (55) presents a general survey of the responses of birds that are sign stimuli for other birds of the same species. Such releasers tend to be ritualized, that is, highly stereotyped in form. Dealing with the same problem, Morris (56) discusses th e probable physiological basis of the feather postures and stereotyped somatic responses that are releasers for one or another behavior of other birds. He shows elsewhere (57) how a stereotyped magnitude of response may develop, both phylogenetically and ontogenetically--that is, through both race and individual learning--so that magnitude loses its usual quantitative relationship to other measures of response-strength such as frequency or rate. The topographic and quantitative stereotype of the releasi ng response is reinforced by its effectiveness in controlling the behavior of other individuals as a sign or discriminative stimulus. Buchholtz (58) illustrates the operation of releasers in maintaining reproductive isolation in dragonflies.

Evolutionary Studies

Survival value.--The three-spined stickleback, gasterosteus, stands motionless and erects its spines when it is strongly stimulated by mechanical means. The ten-spined stickleback, Pygosteus, with much smaller spines, respon ds similarly to the same stimuli. Fish of other species of comparable size, but lacking spines, do not stand motionless. Pike and perch eat small fish. Three investigators (59) provide quantitative data on the biological function of both the behavior a nd the spines of the stickleback and avoidance conditioning of pike and perch. With successive trials, the predators progressively avoid seizing the spiny fish when more comfortable prey is available. Just how effective this behavior is and how more eff ective three large spines are than then ten small ones is given by the day-to-day changes in the population of a single tank: initially--2 pike, 12 gasterosteus, 12 pygosteus, and 12 minnows; on day (1)--2, 11, 11, 9; on day (5)--2, 11, 11, 0; on day (7)--2, 10, 2, 0; on day (12)---2, 2, 0, 0; and on day (14)--2 pike, 0, 0, 0. Despined control pygosteus proved as acceptable as spineless fish of other species. Predators, on attenuated diets, learned to eat them--but stopped when alt ernative prey became available. A number of characteristics of the behavior of gasterosteus may be related to its relative immunity to attack: it is bold, the males nest on open bottoms, the females school and wander freely, and both sexes show c onspicuous nuptial coloration. By comparison, pygosteus is a timid, furtive, weed-frequenter.

The biological function of a quite different kind of behavior is shown by Kerruish (60)--in this case by measures of conception rates rather than of death rates, with domesticated bulls as subjects. A period of sexual stimulation (including the presentat ion of teaser-bulls) and of sexual display and foreplay produced shorter latencies of response to cows as well as enhanced conception rates when the semen collected was used in artificial insemination.

Adaptation of behavior to an environment.--Cullen (61) maintained the ledge-dwelling kittiwake under close observation through three complete reproductive seasons, making a point-by-point comparative study of its behavior with respect to that of th e black-headed gull. She was able to enumerate a series of differences between the two species in general behavior and in detailed responses, both releasers and not. Each difference can be related to the difference in habitat. Morphological differences and behavioral differences went together; e.g., the young kittiwake showed neither protective coloration nor cryptic behavior.

The behavior of the bee, and its evolution.--[For a recent summary, see (62).] There is increasing stress on studies that seek to establish the evolutionary antecedents of the performance of this insect in navigating by the sun, in telling time (p hysiological clocks), and in communicating the location of feeding grounds to its fellows, as well as in the less spectacular features of its social life. Before summarizing these, it is pertinent to bring the reader up to date on some other features of the behavior of bees. First, swarming is coming under investigation (63): prior to swarming, the bees display a greatly increased activity level and their behavior with respect to the old queen changes; the number of attendant workers falls off and they often fail to feed her. At the same time, virgin queens commence to pipe, progressively increasing in frequency up to the time swarming occurs; this piping produced a cessation of activity of nearby bees. [This observation may be related to the Frings & Little report (64) that pure tones between 300 and 1000 c.p.s., at 110 to 120 db. sound pressure at the source, produce a complete cessation of activity inside a nearby hive; at swarming time a lowering of threshold for such quiescence may occur.] Comm unication plays a very important role in determining the behavior of swarming bees (65). A number of scout bees go out from the swarm; after each encounters a possible nesting place she returns to the swarm and performs a dance whose vigor is related to the appropriateness of the nesting place: small, sheltered places produce excited and persistent dancing, with frequent starts of flight in the direction of the site; poorly adapted places (less sheltered and too large) produce less enthusiastic dancing . As more scouts return from various possible nesting sites, the competition between them has much the effect of political orations: the more, and the more excited, scouts that start off in a given a direction the more probable it will be that the whole swarm will follow after a number of preliminary starts and even of incipient fissions. The analogy with a New England town meeting is a tempting one.

Two further bits of information on communication must be included (62). First, the language of bees varies from strain to strain and there is even a suggestion of local dialects within a strain. Second, for some time before newly hatched workers go out on their first collecting trip, they follow the dances of returning foragers within the hive. Through this training period, they follow the dance of the communicating bee with increasing precision. Exchanges of food between bees form an important part o f the economy of a hive; Free (66) finds that the stimuli controlling both the begging and offering of food are received by the antennae of the beggar when it comes in contact with the lower part of the head of another bee (or head-model). Neither head-c olor nor shape is important in producing the begging behavior. Williams & Williams (67) present a mathematical model describing how selective breeding for a behavior such as food-offering may occur. This is an example of a mathematical evolutionary theo ry of a sort perhaps unfamiliar to psychologists.

Manning (68, 69, 70) has been investigating the behavior of bumblebees in gathering nectar. These experimental studies, employing models (baited and scented), provide evidence on the joint evolution of the bees and of the flowers from which they collect food and which they pollinate. These papers provide interesting sets of data showing that the foraging procedure of the bumblebee varies with the type of flower from which it is collecting, and that this behavior is mediated in part by learning processes .

Kalmus (71) presents an evolutionary problem: in the southern hemisphere the sun moves counterclockwise across the sky. Honey bees, established in the hemisphere since European colonization, maintain excellent orientation with respect to the sun throug hout the day. The offspring of queens recently imported from the northern hemisphere showed a large angular disorientation in the morning with respect to a feeding station where they had been fed the previous evening, and improved progressively through t he day until they were correct in the evening. Hybrids also showed such erroneous orientation.

Sheperd (72) addresses himself to the internal clock mechanisms that have been postulated to account for the foraging, navigational, and feeding behaviors of bees, and reports on a simple activity cycle in the distantly related fruit-fly that shows the sa me properties with respect to time as does the temporally-controlled feeding behavior of the honey bee. The bee's temporal behavior may have evolved from such a physiological rhythm.

Dethier (73), also working with flies, demonstrates behavior possibly antecedent to the communication dances. Flies walk circuitously after ingesting food, and their walking is a function of many of the variables that govern the communication dances of b ees. Dethier argues, from the "striking parallelisms between the gyrations of the fly and the communication dances of the bee," that common mechanisms are operating and that the bee's behavior in communicating can best be accounted for in terms of a stimulus-response analysis based on the simple behaviors exhibited by the fly and on the physiological mechanisms governing them. He notes that the basis of trophallaxis may lie in his observation that flies frequently regurgitate food and then cle an their appendages when their crop is full. Unfed flies in the vicinity become greatly excited, follow and move around the sated individual, and attempt to lick food from its mouth parts.

Organization of Behavior, Displacement Activities

Aggressive behavior.--Clarke (74) reports on the responses involved in threats and fights between voles. Godfrey (75), working on four races of the same species and of cross-breeds between them, finds that the offspring of matings between a female of an unaggressive less territorial race and a male of another race, although more hardy before weaning, showed higher mortalities afterwards as well as a failure to show the submissive postures strongly developed in the more aggressive paternal race. The evidence suggests that the higher death rate is attributable to the inability of the loser of a fight to stop the attack of the winner before it is killed, and (since the same effect does not occur with the genetically comparable hybrids of the opposite cross) that submissive postures are learned in the course of association with the mother. These voles show a marked preference (under olfactory control) for members of their own race as sexual partners.

In groups of chaffinches (76 to 79), a straight-line,6 peck-right hierarchy is set up as a consequence of repeated fights among members of the group. In it, males dominate females. The more dominant birds are less likely to attack animals markedly infer ior to themselves in the hierarchy.

In another experiment, in which the distance between adjacent food-boxes was systematically altered, Marler was able to show that the number of fights was inversely proportional to the distance between birds--a verification of Hediger's concept of individ ual distance. The distance-fight function has the form of a psychophysical threshold one. Hunger did not affect directly the tendency of the birds to fight, but it reduced the avoidance and fear behavior of socially inferior birds, thus bringing them in to closer proximity with a superior one and hence inducing fights. Fighting proved independent of food-getting behavior (although it may be reinforced by food) and of hunger. Marler was able to improve the position of a female with respect to other fema les in the hierarchy by dyeing her underparts red, simulating the male. Such disguised females won more fights with other females, but they showed fear and escape behavior to the red-breasted males and thus induced attacks. Only if such birds had been h and-reared and had no experience in being defeated by males, did they prove able to fool the usually more aggressive males and hence win a significant number of fights with them.

In a final experiment, Marler found that aggressive behavior--fighting-- may be reinforced by, for example, food-getting, so that a bird that has frequently fed quickly and well following a fight that drives off another bird eventually comes to look for a fight and, if confronted with two food-dishes at one of which a subordinate is feeding, will choose the same dish, driving the subordinate off. Marler finds no evidence for fighting as other than an instrumental response indulged in only with respect t o other activities, e.g., food-seeking and sex. He states: "It may be that aggression is never caused by frustration, except in a few higher animals, with more complex learning or perceptual capacities."

Sex and aggressive behavior.--Wood-Gush and his colleagues (82 to 85) and Fisher & Hale (86) have been concerned with the aggressive and sexual behavior of the male domestic fowl, gallus domesticus. The former find no relationship be tween mating-rank (frequency) and semen-quality, but a positive one between mating-rank and fertility, quantity evidently compensating for quality. A highly significant correlation between an index of comb size and mating frequency suggest an endocrinolo gical basis for the significant individual differences found in copulatory activity. In other studies no correlation was found between aggressiveness or place in the social hierarchy and mating frequency. Cocks reared in isolation and then introduced to females for the first time and who respond to them aggressively require far longer to attempt to mate or to mate successfully with the female. The less aggressive individuals showed sexual behavior more readily. Fisher & Hale, studying the releasers fo r aggressive behavior and sexual behavior, found that the posture of a second bird is the most important stimulus determining which will occur.

Among the mammals, Schwarts (87) reports that male rats can be conditioned to bar-press with opportunity to copulate as reinforcing stimulus. Hayward (88) reports that avoidance conditioning of young males with respect to oestral females reduces both pre copulatory behavior and the number of intromissions but does not eliminate sexual behavior altogether. Larsson (7) investigated the recovery of sexual responsiveness following copulation and ejaculation in the male rat; his procedure included a technique of pitting sexual behavior against a food-reinforced barpress. He finds two mechanisms necessary to account for his results; he distinguishes between the recovery functions of copulatory and ejaculatory reflexes. It is to be hoped that this work will r eceive the attention that it deserves. Beach & Jordan (89) also postulate two mechanisms, an arousal mechanism and a copulatory mechanism, to account for their data on much the same problem.

Closely associated with the aggressive-sex complex of behaviors as we have seen in Marler's work on chaffinches, is the problem of territorialism. This topic, foreign though it is to psychologists (although they are daily confronted with their children r efusing to share their toys and with the universal phenomenon of private property), is summarized in a series of papers in Ibis (90), as well as discussed by Hediger (91).

Displacement activities.--Two papers by Andrew (92, 93) reflect current ethological thought on these behaviors, which are now rarely interpreted as sparkings-over. Displacement activities occur in conflict and frustration situations; they are ofte n body-care behaviors. Andrew's point is a simple one: The animal, aroused to some activity by one set of stimuli and prevented from completing it by another set or by the absence of critical stimuli in a sequence of them, may be expected to respond to the one set of stimuli always present. When competing tendencies are balanced, these weak stimuli originating from the body surfaces can be effective--so the animal preens, face-washes, or the like. Andrew extends the argument to a variety of similar re sponses. His straightforward account suggests experimental tests of this view of displacement activities, and again shows how far removed ethologists now are from the theories expressed some years ago (22).

Physiological Basis of Behaviour

General.--Two papers are suggestive with respect to the development of a physiological theory of behavior. Weidmann (94) has made observations on gulls that have profound implications for the roles that hormones may be assumed to play in determining behavior. If one places three eggs in the nest of a pair of black-headed gulls several days before the gull lays, it "is possible to suppress egg-laying completely or partially." That is, if wooden eggs are placed in the nest at a time when they are an adequate stimulus for the production of brooding, then the brooding behavior in some as yet undetermined way inhibits the growth of the maturing follicles, and in fact causes them to degenerate, provided they have not yet reached a c ritical sage of maturity. A second and equally interesting implication of this observation comes out of the behavior of the male, whose brooding behavior parallels that of the female. The effects that are produced through hormonal action in the female a re produced in the male by the female's behavior.

The second paper is a lengthy theoretical survey by Prechtl (95), and again shows the shift in ethological thinking in the past few years. This thorough survey of the physiological literature deals with the problem of isolating known physiological mechan isms that have the properties that must be postulated on the basis of behavioral studies of stereotyped movements; it marks a clear departure from instinct theory and makes theoretical suggestions appropriate for learned behavior as well. Models embodyin g the principles of servo-mechanisms, together with well-established physiological processes, may now be developed.

New directions.--The effect of various drugs on behaviors such as the ethologists have experimentally analyzed with good effect have come under investigation in the hands of others than behaviorists. The results strongly suggest that ethologists m ay find drugs a valuable research tool and that they can make a very important contribution in this field. Two investigators (96) found that they could make guppies swim backwards. A more ambitious study of Siamese fighting fish (97) used the fighting r esponse as an index of the effectiveness of a number of drugs including reserpine and morphine. In general, the antihistaminics knocked out fighting behavior; morphine made the fish very aggressive. Enough is known of the behavior of the fighting fish, a nd of other fish, to suggest perhaps half a dozen mechanisms, alterations in whose function might produce such effects. A sophisticated observer could distinguish among them by simple observations. The following quotation about the effect of one set of drugs, read to ethologists, produces smiles: "another peculiarity was that they became pale upon exposure to a control fish exhibiting the fighting response." This is not peculiar--it is the standard response of a subordinate fish to a superior in a hierarchy, and it inhibits attack by the latter. Ethological studies of drugs may be expected to make a substantial contribution, not only to pharmacology, but also to the data indispensable for the development of an adequate physiological theory o f behavior (98).

Literature Cited

1. Beach, F. A., and Jaynes, J., "Studies of Maternal Retrieving in Rats. I. Recognition of Young," Journal of Mammalology, v. 37, pp. 170-80, 1956.

2. Beach, F. A, and Jaynes, J., "Studies of Maternal Retrieving in Rats. II. Effects of Practice and previous Parturitions," American Naturalist, v. 90, pp. 103-109, 1956.

3. Beach, F. A. and Jaynes, J., "Studies of Maternal Retrieving in the Rat. III. Sensory Cues," Behaviour, v 10, pp. 104-125, 1956.

4. Rensch, B., "Increase of Learning Capability with Increase of Brain-Size," American Naturalist, v. 90, pp. 81-95, 1956.

5. Kramer, G., "Experiments on Bird Orientation and their Interpretation," Ibis, v. 97, pp. 196-227, 1957.

6. Carthy, J. D., Animal Navigation. Allen & Unwin, Ltd., London, England, 152 pp., 1956.

7. Larsson, K., "Conditioning and Sexual Behavior in the Male Albino Rat," Acta Psychol. Gothobergenses, v. I, 269, pp. 1956.

8. Hebb, D. O., "Heredity and Environment," British Journal of Animal Behaviour, v. 1, pp. 43-47, 1953.

9. Lehrman, D. S., "A Critique of Konrad Lorenz's Theory of Instinctive Behavior," Quarterly Review of Biology, v. 28, pp. 337-363, 1953.

10. Verplanck, W. S. "Since Learned Behavior is Innate, and Vice Versa, What Now?," Psychological Review, v. 62, pp. 139-44, 1954.

11. Lorenz, K., "The Objective Theory of Instinct," in Foundation Singer-Polignac, L'Instinct dans le Comportement des Animaux et de L'Homme, pp. 51-56, Masson et Cie., Paris, France, pp., 796, 1956.

12. Lehrman, D. S., "On the Organization of Maternal Behavior and the Problem of Instinct," in Foundation Singer-Polignac, L'Instinct dans le Comportement des Animaux et de L'Homme, pp. 475-520, Masson et Cie., Paris, France, pp. 796, 19 56.

13. Schneirla, T. C., "Interrelationships of the 'Innate' and 'Acquired' in Instinctive Behavior," in Foundation Singer-Polignac, L'Instinct dans le Comportement des Animaux et de L'Homme, pp. 387-452, Masson et Cie., Paris, France, pp. 796, 1956.

14. Foundation Singer-Polignac, L'Instinct dans le Comportement des Animaux et de L'Homme, Masson et Cie., Paris, France, pp. 796, 1956.

15. Thorpe, W. H., Learning and Instinct in Animals, Methuen & Co., Ltd., London, England, pp. 493, 1956.

16. Beach, F. A., "Learning and Instinct," Nature, v. 179, pp. 387-89, 1957.

17. Lehrman, D. S., "The Physiological Basis of Parental Feeding Behavior in the Ring Dove," Behaviour, v. 7, pp. 241-286, 1955.

18. Hirsch, J., Lindley, R. H., and Tolman, E. C., "An Experimental Test of an Alleged Innate Sign Stimulus," Journal of Comparative Physiology and Psychology, v. 48, 278-280, 1955.

19. Carter, G. S., Animal Evolution a Study of Recent views of its Causes, Sidgwick & Jackson, London, pp. 368, 1951.

20. Riess, B. F., "The Isolation of Factors of Learning and Natural Behavior in Field and Laboratory Studies," Annals of the New York Academy of Science, v. 51, pp. 1093-1102, 1950.

21. Berry, P. C., An Exploratory Study of Food Transportation by Rats, Master's Thesis, Stanford University, Stanford, California, 1957.

22. Tinbergen, N., The Study of Instinct, Oxford University Press, Oxford, England, pp. 228, 1951.

23. Hediger, H., Psychology of Animals in Zoos and Circuses, Criterion Press, New York, N. Y., pp. 166, 1956.

24. Brown, D. H., and Norris, K. S., "Observations of Captive and Wild Cetaceans." Journal of Mammalology, v. 37, pp. 311-326, 1956.

25. Lowe (McConnell), R. H., "The Breeding Behavior of Tilapia Species (Pisces; Cichlidae) in Natural Waters: Observations on T. Karomo Poll. and T. variabilis Boulenger," Behaviour, v. 9, pp. 140-163, 1956.< BR>
26. Tavolga, W. N., "Pre-spawning Behavior in the Gobiid Fish, Bathygobius soporator," Behaviour, v. 9, pp. 53-74, 1956.

27. Thornton, I. W. B., "Notes on the Biology of Leiurus quinquestriatus," British Journal of Animal Behaviour, v. 4, pp. 92-93, 1956.

28. Frisch, O. von, "Zur Brutbiologie und Jugendentwicklung des Brachvogels (Numenius arquata L.)," Z. Tierpsychol., v. 13, pp. 50-81, 1956.

29. Weidmann, U., "Verhaltensstudien an der Stockente I. Das Aktionsysteme," Z. Tierpsychol., v. 13, pp. 208-71, 1956.

30. Dart, R. A., "The Myth of the Bone-Accumulating Hyena," American Anthropologist, v. 58, pp. 40-62, 1956.

31. Karli, P., "The Norway Rats Killing Response to the White Mouse: an Experimental Analysis," Behaviour, v. 10, pp. 81-103, 1956.

32. Stamm, J. S., "Genetics of Hoarding: II. Hoarding Behavior of Hybrids and Back-Crossed Strains of Rats," Journal of Comparative Physiology and Psychology, v 49, pp. 349-52 1956.

33. Frings, H., Frings, M., and Hamilton, M., "Experiments with Albino Mice from Stocks Selected for Predictable Susceptibilities to Audiogenic Seizures," Behaviour, v. 9, pp. 44-52, 1956.

34. Hinde, R. A., "The Behaviour of Certain Cardueline F1 Inter-Species Hybrids," Behaviour, v. 9, pp. 202-213, 1956.

35. Bastock, M., and Manning, A., "The Courtship of Drosophila melanogaster," Behaviour, v. 8, pp. 85-111, 1955.

36. Bastock, M., "A Gene Mutation which Changes a Behavior Pattern," Evolution, v. 10, pp. 421-439, 1956.

37. Smith, J. M., "The Effects of Insemination on the Behaviour of Female Drosophila subobscura, and Some Effects of Inbreeding on Male Behaviour," British Journal of Animal Behaviour, v. 4, pp. 37, 1956.

38. Tebb, G., and Thoday, J. M., "Reversal of Mating Preference by Crossing Strains of Drosophila melanogaster," Nature, v. 177, pp. 707, 1956.

39. Kallmann, F. J., "The Genetics of Human Behavior," American Journal of Psychiatry, v. 113, pp. 496-501, 1956.

40. Stokes, B., "Behaviour as a Means of Identifying Two Closely-Allied Species of Gall-Midges," British Journal of Animal Behaviour, v. 3, pp. 154-157, 1955.

41. Simmins, K. E. L., "The Taxonomic Significance of the Head-Scratching Methods of Birds," Ibis, v. 99, pp. 178-181, 1957.

42. Hinde, R. A., "A Comparative Study of the Courtship of Certain Finches (fringillidae)," Ibis., v. 98, pp. 1-23, 1956.

43. Andrew, R. J., "The Aggressive and Courtship Behaviour of Certain Emberizines," Behaviour, v. 10, pp. 255-308, 1956-1957.

44. Andrew, R. J., "Intention Movements of Flight in Certain Passerines, and their Use in Systematics," Behaviour, v. 10, pp. 179-204.

45. Moynihan, M., "Notes on the Behaviour of Some North American Gulls, I," Behaviour, v. 10, pp. 126-178, 1956.

46. Schloeth, R., "Zur Psychologie der Begegnung zwischen Tieren," Behaviour, v. 10, pp. 1-80, 1956.

47. Hale, E. B., "Breed Recognition in the Social Interactions of the Domestic Fowl," Behaviour, v. 10, pp. 240-254, 1956-1957.

48. Newman, M. A., "Social Behavior and Interspecific Competition in Two Trout Species," Physiological Zoology, v. 29, pp. 64-81, 1956.

50. Association for the Study of Animal Behaviour, "Proceedings of the Association for the Study of Animal Behaviour," British Journal of Animal Behaviour, v. 3, pp. 121-124, 1957.

51. Hinde, R. A., Thorpe, W. H., and Vince, M. A., "The Following Response of Young Coots and Moorhens," Behaviour, v. 9, pp. 214-242, 1956.

52. Jaynes, J., "Imprinting: The Interaction of Learned and Innate Behavior: I. Development and Generalization," Journal of Comparative Physiology and Psychology, v. 49, pp. 201-206, 1956.

53. Jaynes, J., "Imprinting: The Interaction of Learned and Innate Behavior: II. The Critical Period," Journal of Comparative Physiology and Psychology, v. 50, pp. 6-10, 1957.

54. Weidmann, U., "Some Experiments on the Following and the Flocking Reaction of Mallard Ducklings," British Journal of Animal Behaviour, v. 4, pp. 78-79, 1956.

55. Marler, P., "Specific Distinctiveness in the Communication Signals of Birds," Behaviour, v. 11, pp. 13-39, 1957.

56. Morris, D., "The Feather Postures of Birds and the Problem of the Origin of Social Signs," Behaviour, v. 9, pp. 75-113, 1956.

57. Morris, D., " 'Typical Intensity' and its Relation to the Problem of Ritualization," Behaviour, v. 11, pp. 1-12, 1957.

58. Buchholtz, C., "Eine Analyse des Paarungsverhaltens und der Dabei Wirkenden Ausloser bei den Libellen Platyonemes pennipes Pall. und Pl. dealbata Klug," Z. Tierpsychol., v. 13, pp. 13-26, 1956.

59. Hoogland, R., Morris, D., and Tinbergen, N., "The Spines of Sticklebacks (Gasterosteus and Pygosteus) as Means of Defense Against Predators (Perca and Esox)," Behaviour, v. 10, pp. 205-236, 1956-1957.

60. Kerruish, B. M., "The Effect of Sexual Stimulation Prior to Service on the Behaviour and Conception Rates of Bulls," British Journal of Animal Behaviour, v. 3, pp. 125-130, 1955.

61. Cullen, E., "Adaptations in the Kittiwake to Cliff-Nesting," Ibis., v. 99, pp. 275-299, 1957.

62. Frisch, K. von, "Lernvermogen und erbgebundene Tradition im Leben der Bienen," in Foundation Singer-Polignac, L'Instinct dans le Comportement des Animaux et de L'Homme, pp. 345-386, Masson et Cie., Paris, France, pp. 796, 1956.

63. Allen, M. D., "The Behaviour of Honeybees Preparing to Swarm," British Journal of Animal Behaviour, v. 4, pp. 14-22, 1956.

64. Frings, H., and Little, F., "Reactions of Hive Bees to Simple Sounds," Science, v. 125, pp. 122, 1957.

65. Lindauer, M., "Communication in Swarm-Bees Searching for a New Home," Nature, v. 179, pp. 63-66, 1957.

66. Free, J. B., "A Study of the Stimuli Which Release the Food Begging and Offering Responses of Worker Honeybees," British Journal of Animal Behaviour, v. 4, pp. 94-101, 1956.

67. Williams, G. C., and Williams, D. C., "Natural Selection of Individually Harmful Social Adaptations Among Sibs With Special Reference to Social Insects," Evolution, v. 11, pp. 32-39, 1957.

68. Manning, A., "The Effect of Honey-Guides," Behaviour, v. 9, pp. 114-139, 1956.

69. Manning, A., "Some Aspects of the Foraging Behaviour of Bumble-bees," Behaviour, v. 9, pp. 164-201, 1956.

70. Manning, A., "Some Evolutionary Aspects of the Flower Constancy of Bees," Pro., Roy. Phys. Soc., v. 25, pp. 67-71, 1957.

71. Kalmus, H., "Sun Navigation of Apis mellifica in the Southern Hemisphere," Journal of Experimental Biology, v. 33, pp. 554-566, 1956.

72. Shepherd, S. K. deF., " 'Clock'-Controlled Activity Rhythms in the Fruit Fly," Science, v. 124, pp. 172, 1956.

73. Dethier, V. G., "Communication by Insects: Physiology of Dancing," Science, v. 125, pp. 331-336, 1957.

74. Clarke, J. R., "The Aggressive Behaviour of the Vole," Behaviour, v. 9, pp. 1-23, 1956.

75. Godfrey, J., "Social Behaviour in Four Bank Vole Races" (Personal Communication)

76. Marler, P., "Studies of Fighting in Chaffinches. (1) Behaviour in Relation to the Social Hierarchy," British Journal of Animal Behaviour, v. 3, pp. 111-117, 1955.

77. Marler, P., "Studies of Fighting Chaffinches. (2) The Effect on Dominance Relations of Disguising Females as Males," British Journal of Animal Behaviour, v. 3, pp. 137-146, 1955.

78. Marler, P., "Studies of Fighting Chaffinches. (3) Proximity as a Cause of Aggression," British Journal of Animal Behaviour, v. 4, pp. 23-30, 1956.

79. Marler, P., "Studies of Fighting Chaffinches. (4) Appetitive and Consummatory Behaviour," British Journal of Animal Behaviour, v. 5, pp. 29-37, 1957.

80. Baron, A., Stewart, C. N., and Warren, J. M., "Patterns of Social Interaction in Cats (Felis domestica)," Behaviour, v. 11, pp. 56-66, 1957.

81. Chance, M. R. A., "Social Structure of a Colony of Macaca mulatta," British Journal of Animal Behaviour, v. 4, pp. 1-13, 1956.

82. Wood-Gush, D. G. M., and Osborne, R., "A Study of Differences in the Sex Drive of Cockerels," British Journal of Animal Behaviour, v. 4, pp. 102-110, 1956.

83. Wood-Gush, D. G. M., "The Agonistic and Courtship Behaviour of the Brown Leghorn Cock," British Journal of Animal Behaviour, v. 4, pp. 133-142, 1956.

84. Wood-Gush, D. G. M., "Aggression and Sexual Activity in the Brown Leghorn Cock," British Journal of Animal Behaviour, v. 5, pp. 1-5, 1957.

85. Wood-Gush, D. G. M., (Personal Communication)

86. Fisher, A. E., and Hale, E. B., "Stimulus Determinants of Sexual Aggressive Behaviour in Male Domestic Fowl," Behaviour, v. 10, pp. 309-323, 1956-1957.

87. Schwartz, M., "Instrumental and Consummatory Measures of Sexual Capacity in the Male Rat," Journal of Comparative Physiology and Psychology, v. 49, pp. 328-333, 1956.

88. Hayward, S. C., "Modification of Sexual Behavior of the Male Albino Rat," Journal of Comparative Physiology and Psychology, v. 50, pp. 70-73, 1956.

89. Beach, F. A., and Jordan, L., "Sexual Exhaustion and Recovery in the Male Rat," Quarterly Journal of Experimental Psychology, v. 8, pp. 121-133, 1956.

90. Anonymous, "Territorialism," Ibis, v. 98, pp. 340-530, 1956.

91. Hediger, H., "Instinkt und Territorium," in Foundation Singer-Polignac, L'Instinct dans le Comportement des Animaux et de L'Homme, pp. 521-556, Masson et Cie., Paris, France, pp. 796, 1956.

92. Andrew, R. J., "Normal and Irrelevant Toilet Behaviour in Emberiza Spp.," British Journal of Animal Behaviour, v. 4, pp. 85-91, 1956.

93. Andrew, R. J., "Some Remarks on Behaviour on Conflict Situations, With Special Reference to Emberiza Spp.," British Journal of Animal Behaviour, v. 4, pp. 41-45, 1956.

94. Weidmann, U., "Observations and Experiments on Egg-Laying in the Black Headed-Gull (Larus Ridibundus L.)," British Journal of Animal Behaviour, v. 4, pp. 150-161, 1956.

95. Prechtl, H. F. R., "Neurophysiologische Mechanismen des Formstarren Verhaltens," Behaviour, v. 9, pp. 243-319, 1956.

96. Keller, D. L., and Umbreit, W. W., "Chemically Altered 'Permanent' Behavior Patterns in Fish and Their Cure by Reserpine," Science, v. 124, pp. 407, 1957.

97. Walaszek, E. J., and Abood, L. G., "Effect of Tranquilizing Drugs on Fighting Response of Siamese Fighting Fish," Science, v. 124, pp. 440-441, 1956.

98. Abramson, H. A., Sklanofsky, B., Baron, M. O., and Gettner, H. H., "Blocking Effect of Brain Extract on Lysergic Acid Diethymalide Reaction of Siamese Fighting Fish," Science, v. 125, pp. 397-398, 1957.


1 The survey of the literature pertaining to this review was completed in May, 1957.

2 The writer wishes to express his appreciation to Drs. G. Baerends, J. van Iersel, W. H. Thorpe, R. A. Hinde, N. Tinbergen, and Konrad Lorenz, and to their colleagues, whose hospitality made it possible for him to observe what is going on in their labor atories. He also wishes to acknowledge his thanks to the institutions whose libraries and other facilities were made available to him, and to those who have reviewed the earlier drafts of the chapter.

3 One conference, now fully reported, marks a watershed in the "innate-learned" controversy [see (11, 12, 13) and their discussions]. The report (14) includes a number of authoritative summaries of major research areas.

4 The status of this concept among ethologists is not well represented in the Thorpe text (15). Neither the theoretical point of view to which most of the book is addressed nor the interpretations given to much of the experimental work summarized is wel l received by many ethologists, who dissociate themselves, often with asperity, from its propositions. The bibliography is a useful one. See also Beach's review (16).

5 Two examples come to mind. Riess (20) showed, in a carefully controlled experiment, that female rats that had never had the opportunity to pick objects up or to carry them about, prior to parturition, failed to build nests and to retrieve their young. He was careful, of course, to conduct the final critical tests in special living quarters provided with nesting materials; the animals were introduced into this just before parturition. Eibl-Eibesfeld repeated the experiment but, being more familiar wi th the behavior of rats, introduced the females into the experimental boxes several days before parturition, allowing them time to settle down to sleeping in one particular area of the cage. He introduced the potential housing materials at the same time as had Riess. These rats built nests and retrieved, although not as well as control animals. Again, Berry, working with the writer (21), found that rats eat small pieces of food at the place where it is found and transport large pieces to another place before eating it. This finding has implications for experiments on the effect of magnitude of reinforcement on habit strength as measured by running speed and resistance to extinction.

6 A very preliminary report (80) shows that such a simple social hierarchy is set up in cats under conditions that are quite restricted relative to the conclusions drawn. The more complex hierarchies, set up in freely moving groups, are illustrated by C hance (81).

Please address reprint requests, or any questions/comments, to: