By Klaus Immelmann

IFCB 1983

INTRODUCTION
In many vertebrates, especially in birds and mammals, knowledge of species-specific characters, i.e. of those features by which members of the same species recognize each other, is acquired by a learning process which is distinguishable from other forms of learning and -since Lorenz published his famous “Kumpan”-paper in 1935 – is known by its own name, imprinting.

In a brief and general way, imprinting can be defined as an early and fairly rapid learning process with a particularly stable result. The best-known examples are filial imprinting, by which the hatchlings of precocial birds learn the characteristics of their mother or their parents, and sexual imprinting, by which the young of both precocial and altricial birds learn the species-specific characteristics which enable them, in adult life, to recognize a conspecific mate (Fabricius 1951; Bateson 1966; Immelmann 1972b; Hess 1973).

Two main criteria have been described in the literature as the principal distinguishing characters of imprinting: its restriction in time, i.e. the fact that this form of learning is possible only at a certain stage during ontogenetic development known as the sensitive phase, as well as the great permanence of the results of this learning process, frequently referred to as irreversible (for review, see Immelmann 1972b).

The present paper does not attempt to present another review of the vast literature on imprinting which has accumulated over the years. It will rather try to characterize this early learning process by reporting details of one particular example of imprinting, the acquisition of sexual preferences in the Australian Zebra Finch (Taeniopygia guttata castanotis). This problem has been studied in our laboratory for many years, with emphasis on the two main criteria mentioned above.

EVIDENCE FOR SEXUAL IMPRINTING
In a pilot study we attempted to determine if and how early social experience has an influence on the choice of a mate. This was done in cross-fostering experiments with another species of estrildid finches, the Bengalese Finch (Lonchura striata f. domestica): male Zebra Finches were foster-reared with Bengalese Finches until they had reached nutritional independence. They were then separated from their foster parents and kept in individual cages.

As soon as they were sexually mature, they were tested in a double-choice situation with a female of their own species and a female of foster parents’ species. All males treated in this was showed a clear preference for the foster parents’ species and thus proved to be imprinted on the latter (figs. 1 – 4) (for details of methods and results see Immelmann 1969, 1972a).

Another study was concerned with female imprintability. This seemed to be important with regard to the fact that in the literature on imprinting, there has been considerable discussion as to whether in females, as in males, early experience may influence subsequent sexual preferences. Schutz (1965) published a theory based on his experiments with ducks according to which both sexes are imprintable in monomorphic species, whereas in sexually dimorphic species in which the male has a bright nuptial plumage with distinct color markings some kind of innate recognition of the male plays the most important role forthe female. These assumptions have been modified by Klint (1975), who found that female Mallards do modify their sexual preferences according to early experience, although they do indeed show a genetically determined preference for the nuptial coloration of the male.

Using a slightly different test situation, a multiple choice apparatus (fig. 5), Sonnemann and Sjolander (1977) tested the preferences of 44 Zebra Finch females raised by Bengalese Finches and of 17 controls raised by their own species. They found a clear preference for the males of the foster species in the first group and a clear preference for conspecific males in the control females. They have thus given the first experimental proof of female imprintability in a sexually dimorphic species of birds.

Figure 1.
Courtship behavior of Zebra Finches. The male (right) advances towards the female along the branch in a privoting dance, turning his feet and body with every jump and singing almost continuously. His tail is twisted towards the female, the crown-feathers are depressed and the feathers of the back and sides of the head are raised, thus making the black-and-white face markings and the chestnut ear-covers more conspicuous. The white belly-feathers are fully fluffed out. The female also turns her tail towards the mate.

Figure 2
A Zebra Finch male foster-reared by Bengalese Finches courts a Bengalese Finch female.

Figure 3
A foster-reared Zebra Finch male courts a Bengalese dummy in the presence of a dummy of a Zebra Finch female.

Figure 4.
Sexual reactions of Zebra Finch males raised by Bengalese Finch foster parents. The block diagrams give the number of courtship sequences (for description see fig. 1) towards the female of the foster parents’ species (B) and the conspecific female (Z) respectively. The diagram represents the results of twelve double-choice experiments of 30 minutes duration using different females each time. If there is no block above the symbol Z the conspecific female did not receive any sexual reactions at all (from Immelmann 1969).

Figure 5.
Multiple choice testing apparatus. The test-bird, a female Zebra Finch foster-raised by Bengalese Finches is placed in the central, octagonal compartment. Four Zebra and four Bengalese Finch males are presented in eight small cages arranged around the central compartment. A perch in front of each of the satellite cages was provided, with a photo cell arrangement, so that the time spent there by the female could be recorded (from Sonnemann & Sjolander 1977).

PERMANENCE OF EARLY PREFERENCE
In order to test the stability of sexual preferences established early in life, long-term experiments have been carried out with Zebra Finch males which had been reared by Bengalese Finches and, during the pilot study, had already proved to be sexually imprinted on their foster parents’ species. These birds were “forced” to mate with conspecific female: they were kept in a breeding cage with one Zebra Finch female and were provided with nesting facilities. The cages were placed in a room where no Bengalese Finches could be seen or heard, thus depriving the birds of any further contact with members of their foster parents’ species. Most of these males eventually mated with the female and jointly raised one or several broods. After six months they were separated from their conspecific mate and were tested again in a double choice situation as before. The results of these tests revealed the same preference for the Bengalese Finch females as was observed before the period of intra-specific contact (fig. 6) (for details see Immelman 1972a).

This means that the brief contact with the foster parents early in life clearly exerted a longer-lasting influence than did social contact of long-term duration during adult life. Some of the males were left with their conspecific mates for several (up to six) years. When given the choice again, even these males preferred Bengalese over Zebra Finch females. It can be concluded that in adult Zebra Finches the influence of early experience lasts fora period of time that probably exceeds the life expectancy of this species under natural conditions. Sexual imprinting in the Zebra Finch, therefore, is characterized by a degree of stability that justifies the term “irreversible”.

SENSITIVE PHASE FOR SEXUAL IMPRINTING
The next study was concerned with the possible occurrence of a sensitive phase for sexual imprinting, especially with its termination.
Using 337 individuals, this study investigated whether the same degree of irreversibility observed in adult birds can also be demonstrated in adolescents. Male Zebra Finches were again foster-raised by Bengalese Finches, but instead of being kept isolated after separation from the foster parents, they were immediately given intraspecific experience by being placed into a cage with several Zebra Finch males and females. Four series of experiments were run with three, seven, 30 and 60 days of intraspecific contact, respectively. The age at which the birds were transferred from Bengalese foster parents to the Zebra Finches varied from day 27 to day 73. The results of this study are summarized in fig. 7. They demonstrate that in adolescent males, in contrast to adults, it is still possible to alter a previously established preference and to “re-imprint” the birds on their own species. The success of such attempts depends mainly on two variables, the age of the bird and the duration of social contact with its own species. If only three or seven days of intraspecific contact are permitted, such contact must begin no later than around the 40th day of life in order to have the chance to exert a permanent effect on subsequent sexual preferences. On the other hand, if 30 or 60 days of contact are provided, changes in preference are still possible if the bird is placed together with con-specifics at a later age (57 or 71 day respectively).

Figure 6.
Sexual reactions of Zebra Finch males raised by Bengalese Finch foster parents before and after six months of breeding experience with a conspecific female. For each male, the left B-Z pair represents the results of the double-choice experiments before, the right pair those of the same number of experiments after the intraspecific contact (all other details in fig. 4) (from Immelmann 1972a).

It follows that the older the adolescent bird is, the more social contact is necessary to change a previously established social preference. In other words, with increasing age the social effort, i.e. the duration of social contact necessary to establish new preferences, increases and this process, as a comparison of the four series indicates, is a gradual one (fig. 8).

The main result of the study, however, is the demonstration of the fact that the sensitive phase for sexual imprinting in Zebra Finches comes to its close very early in life: even in the earliest isolates, before day 30, many individuals proved to be imprinted on Bengalese Finches so strongly that no subsequent change through intraspecific contact could be achieved any more. Between about day 30 and 40, some but comparatively fewer changes still occurred, whereas beyond that age the number of changes decreased rapidly.

An interesting side-result was the discovery of a remarkably large amount of individual variation: Some individuals seemed to be “closed” very early, and even an extended period of intraspecific contact (30 or 60 days) did not result in a change of their Bengalese Finch preference any more. Others seemed to “remain open” for a longer period of time and even brief intraspecific contact (three or seven days) led to a change in preferences. The possible functions of such differences have been discussed elsewhere (Immelmann 1979).

Figure 7.
Sexual preferences of male Zebra Finches raised by Bengalese Finch foster parents and subsequently exposed to conspecific females for three, seven 30, or 60 days respectively. Empty squares represent individuals which, during the double-choice tests, showed a more than 70% preference for Bengalese females. These males, obviously, had retained their preference for the foster parents’ species despite some early and, in case of the third and fourth series, extended contact with conspecifics, which means that their sensitive phase must have come to its close before the day of transfer. The males, represented by squares with a central black dot, in contrast, did show a preference for Zebra Finch females. Assuming that at the day of transfer they also had a preference for Bengalese Finches (in assumption seemingly justifiable on the basis of the results of the pilot study) they must have replaced their primary preference for Bengalese by a secondary preference for Zebra Finches as a result of social contact with the latter. The abscissa gives the age (in days) of transfer from the foster parents to the conspecific females (from Immelmann & Suomi 1981).

Figure 8.
The data demonstrate the effect of both the age and the duration of social contact for determining the sensitive phase: (a) with increasing age (shown here for the 60-day-group) the percentage of changes in preference decreases; (b) an extended duration of social contact with Zebra Finch females (shown here for the birds isolated from their foster partner between 28 and 40 days of age) increases the percentage of change in preference (from Immelmann 1983.)

The study of the sensitive phase for sexual imprinting has subsequently been extended into two directions: The first additional experiment was concerned with the possible influence of continuous social contact with the foster parents’ species: In the previous experiments, there was a complete change from interspecific to intraspecific contact at the day of transfer, which, in some of the individuals transferred at an early age, led to the “re-imprinting” on their own species described above.

In order to elucidate the possible influence of continuing contact with the foster parents’ species one of the series of the previous study (intraspecific contact for 30 days) was repeated with one slight alteration in methods: Instead of providing the foster-reared Zebra Finch male with Zebra Finch females only, it was now placed in a cage that contained three Zebra Finch, but also three Bengalese Finch females.

The results are shown in fig. 9. Despite the fact that these birds had the same amount of contact with the same number of conspecific females at the same age as in the original experiment none of them changed their preference for Bengalese Finches into a preference for Zebra Finches.

No definite conclusions as to the reasons of this difference are possible yet. The most likely explanation is that due to the presence of members of the foster parents’ species the young males concentrate their social activities on them, and although Zebra Finch females are as easily accessible they don’t engage in close social encounters with the conspecific females. As a consequence, their amount of real social contact with the latter is much smaller than in those individuals which are “forced” to direct all their social activities to conspecifics. Obviously, this means one more factor which might influence the termination of sensitive phases, i.e. the duration of “openness” to the influence of new secondary social stimulation. It may be concluded that not only the social subjects themselves but also the “surrounding conditions” may play a role in the acquisition of final, permanent preferences.

The second additional experiment was concerned with a species comparison. It was designed to elucidate whether the phenomenon of a sensitive phase for the acquisition of social preferences is unique to the Zebra Finch or whether in the second species used for the cross-fostering experiments, the Bengalese Finch, a similar situation exists. The study was conducted with Bengalese Finch males which in turn had been fostered-reared by Zebra Finches. The series with seven days of intraspecific contact, for which most comparative data are available, was chosen. The results are shown in fig. 10. They indicate that in the Bengalese Finch, under strictly identical conditions, the period of time during which a replacement of a primary preference through a secondary preference is possible is slightly more extended than in the Zebra Finch. The occurrence of an early termination of the sensitive phase, however, as well as the large amount of individual variation is the same in both species.

Figure 9.
Comparison between sexual preferences of foster-raised Zebra Finch males that, at different ages, were given 30 days of intraspecific contact only (above) and those that were exposed to Zebra and Bengalese Finch females (below) (details as in fig. 7).

IMPRINTING AND SEX RECOGNITION
In the sexually dimorphic Zebra Finch, in addition to species recognition, there is also the problem of how the sexes can be distinguished. From previous observations, more or less casual ones, it could be assumed that Zebra Finches obviously do not recognize members of the opposite sex by some kind of an inbuilt mechanism but that an early learning process is involved again. Zebra Finch males foster-raised by Bengalese Finches, for example, do court Zebra Finches if no Bengalese Finches are present (see above). But in this case, they court females and males and do not seem to make any distinction whatsoever.

In order to study the possible influence of early experience on sex discrimination we increased the sexual differences in plumage color above the natural amount by working with “wild-colored”, i.e. naturally grey, birds and members of a leucocistic morph of the Zebra Finch which are white in plumage color. Mixed pairs of parents with a white male and a wild-colored female and vice versa were used. The preferences of males and females raised in this way were tested in the usual double-choice or multiple-choice situation respectively.

The results revealed a clear difference between sexes. They showed that males have a clear preference for the plumage color of their mother whereas females prefer the color of their father (Immelmann & Sjolander 1983). Obviously, sons use their mother and daughters use their father as a model for establishing future mate preferences. Thus, in this dimorphic species there is a constraint on the outcome of sexual imprinting which produces different results in males and females, despite the fact that they are reared together by the same parents.

Figure 10.
Comparison between foster-raised Zebra Finch males (above) and foster-raised Bengalese males (below) that were exposed to conspecific females for 7 days (details as in fig. 7).

Table 1.
(a) Preference for wild-colored females in males raised by a white father and a wild-colored mother. Out of 42 males, 29 had a more than 90% preference for wild-colored females, another four males had a more than 80% preference and only one had a preference for white females. (b) The opposite situation: Out of 20 males raised by a white mother twelve had a more than 90% preference for white females (from Immelman & Sjolander 1983).

The mechanisms which might be involved in the sex-specific selection of a model for imprinting are not known yet. One possibility would be that the parents distinguish between male and female offspring and treat them differently. Recent observations by Balda (1982) indicate that there is a sexual dimorphism in the begging calls of young Zebra Finches which might provide a basis for sex-dependent parent-offspring-relations.

Another possibility is that the young in turn react sex-dependently to such behavior from the parents and perhaps also from other birds. More specifically there is a tendency in Zebra Finch fathers to show increasing aggression towards the young as independence approaches, and this aggression might well be sex-dependent, either because the sons are recognized as males or because they respond to aggression in a different way than daughters do. The aggression shown by the father may become associated with him and with males in general, and this could then act as a future deterrent for young males to approach other males. In females, on the other hand, similar tendencies could be overcome bythe courtship of the male, as has already been suggested by Morris (1954).

DISCUSSION
This paper has attempted, for one particular case of sexual imprinting, to elucidate the variety and multitude of factors involved. They comprise, amongst others, the age of the bird, the duration of its early social contacts, the continuity of social contact with members of the (foster) parents’ species, the possible influence of siblings and the obvious differences between the influence of father and mother on male or female offsprings respectively. Altogether the early acquisition of social preferences which has been thought to be a rather simple early learning process has proved to be a very complex phenomenon, much more extended in time than had been supposed during the early stages of imprinting research and dependent on many more variables than had been assumed originally.

As far as the general implications of the study are concerned, it is of course difficult and dangerous to draw conclusions from data collected on one species to possible parallel phenomena in other species of animals, and it is unlikely, therefore, that all details of the influence of early experience on subsequent social preferences elucidated for the Zebra Finch will be present also in other finches or other species of birds. Pronounced species differences have indeed been described, and they do occur even in closely related species. This refers, for example, to the duration of the sensitive phase for sexual imprinting which has been found to be of longer duration in Gray Lang Geese (Anser anser) than it is in the Mallard (Anal platyrhynchos) (Schutz 1965, 1970) and which, as mentioned above, is slightly longer in the Bengalese than in the Zebra Finch. Similar differences between species have also been found with regard to the relative importance of siblings on the establishment of preferences (see below).

Despite such differences in detail, however, the great amount of data on imprinting and imprinting-like processes which has accumulated over the years has shown that some general trends of early behavior development seems to be similar in many species of birds, mammals and even fish. One of these similarities is the general occurrence of phase-specificity, i.e. the wide-spread existence of sensitive phases in development during which the social environment exerts a stronger and longer-lasting influence on subsequent preferences than at other times. A second similarity is the fact that, as mentioned above for the Zebra Finch, such sensitive phases tend to occur very early in life. A third general phenomenon finally, is a tendency of sensitive phases to be geared very closely to the duration of parental care in the particular species.
In the Zebra Finch, for example, the young stay close to the nest only for the first days after fledging (which occurs around the twenty-first day of life); later on, they meet with young birds from other nests and form small groups of their own (Immelmann 1962). Because many Australian finches breed in mixed colonies or neighborhoods, these groups may consist of young from different species, and, as a consequence, the young Zebra Finches come into close social contact with members of other species. A brief sensitive phase that brings the learning of species-specific characteristics more or less to a close before the young leave their parents may, therefore, prevent the birds from learning inappropriate social signals form non-conspecifics. It is certainly not by chance, therefore, that the sensitive phase in this species, as our data have shown, is very brief indeed (which in turn makes it an ideal subject for experimental research on the nature of sensitive phases). An example almost opposite to the Zebra Finch is that of the above-mentioned Gray lag Goose, in which the parent-offspring bond is known to last for about ten months and to break apart only shortly before the onset of the following breeding season. For this species, the sensitive phase for sexual imprinting has been found to last at least until the 150th day of life (Schultz 1969).

Such temporal correlation between the duration of parental care and the occurrence of sensitive phases for imprinting does of course have an important biological function: While the young animal is still a member of the family group its opportunities to learn species-specific characteristics (as well as certain features of the environment) are greater than when it subsequently has to live on its own or in much looser groups. Natural selection, therefore, will certainly favor rapid learning during an early sensitive phase which comes to its close before or at the time when the young leave their parents.
Moreover, the very stable result which has been found to be characteristic of early learning may have additional effect: It may serve to terminate some of the essential learning processes before dispersal from the natal area. Such early termination may be important to prevent a major effect of stimuli from other sources at a time when – owing, for example, to deteriorating environmental conditions after the end of the productive season – the animal lives in mixed species flocks and comes in close contact with members of other species. In this case, irreversibility may help to “protect” the preference for the correct species, acquired early in life, from subsequent influence (Immelmann 1975). This has been found to be particularly important in song learning in birds, especially in those migratory species which share their winter quarters with members of other subspecies singing different song dialects. In this case, if singing starts before departure to the breeding grounds in spring, “mis-imprinting” on the “wrong” dialect may occur and has indeed been observed (Baptista 1974).

PRACTICAL APPLICATIONS
The rather general statements made above are supported by the results of numerous experimental studies on filial and sexual imprinting in many species of birds (for reviews, see Immelmann 1972b; Bateson 1978, 1979; Immelmann & Suomi 1981) and several species of mammals (Gilbert 1975; Huck & Banks 1980). They are also supported by countless casual observations on hand-raised animals or an individuals which, e.g. in zoological gardens or in private collections, had to be foster-raised with another species (Hediger 1950; Immelmann 1972b; Klinghammer 1973; Sambraus 1978). Under these rearing conditions, many individuals develop sexual preferences for humans or members of the foster species and often refuse to mate with their own species. As in other cases of imprinting, such preferences tend to remain very stable.

From these facts it becomes apparent that the phenomenon of imprinting may also have some practical implications. First of all, it necessarily leads to the conclusion that because of the “wrong” preferences they may create – in addition to several other disadvantages they may have – hand rearing and foster-rearing should be avoided whenever possible.

If, however, in some special cases these types of rearing cannot be avoided at all the following precautions should be observed:

1. The hand- or foster-raised individual should be brought in contact with members of its own species as soon as possible, e.g. as soon as it is able to look for itself and does not need any caretaking by fosterers any more. The earlier such intraspecific contact begins the greater is the chance that the animal will finally develop normal social preferences and other behaviors (e.g. the abilities to form social contacts or to integrate into social hierarchies). The best results can be expected to be achieved if the social contact starts before the age at which weaning occurs under natural conditions.
2. Intraspecific contact should last as long a possible, at least until sexual maturity has been reached.

(The most natural way of providing the young animal with conspecific contact already during the earliest stages of development would of course be to raise it together with siblings. In the vast majority of cases, this does certainly apply and should be observed. With regard to the influence of siblings, however, differences between species again seem to occur (see above). Whereas no immediate data are available for mammals some details have been described for birds:

In four species of pigeons, for example, Brosset (1971) found that siblings “play no role in the determination of sexual choice”. The same was found in color imprinting of domestic pigeons (Warriner et al. 1963) and Budgerigars (Melopsittacus undulates) (Stamm & Blum 1971). In the Lesser Snow Goose (Ansercaerulescens) both parental and sibling color influences mate choice (Cooke et al. 1976) and in the Mallard, Schutz (1965) and Klint (I likewise found a positive influence of siblings on subsequent sexual preferences.

In the Zebra Finch, finally, very much to our surprise, we have been able, in a large series of experiments involving several hundred male birds, to demonstrate a negative influence of siblings, especially of brothers. The results of this study clearly showed that the more male Zebra Finch siblings a foster-reared male Zebra Finch has the more it will subsequently refer his foster parents’ species, i.e. Bengalese Finches. Female siblings also exert a negative influence although not quite as strong as males (Immelmann 1983).

This strange result does of course require an explanation. It may perhaps be sought along the following line: Young Zebra Finches look different from adults. The biggest difference is in beak color which is black instead of bright red. A positive effect of siblings, therefore, could create the danger that young males also develop preferences for the specific characteristics of immature birds. The function of negative mutual influence at this age, therefore, may be to avoid the formation of such preferences. In those species, on the other hand, in which the young look similar to an adult female, as in the Mallard, mother and sibling imprinting “work together in the same direction” (Klint), and a similar danger does not exist. According to Klint (1978) this might be a reason for the great influence of siblings he found in his experiments. To come back to the practical conclusions, these results show that in those species of birds in which a strong age dimorphism exists and the young look very different from the adults the joint raising of siblings may involve some danger and an early contact with other adult con-specifics is even more important.)

3. In order to prevent the continuation of any contacts with the species of the foster parents, great care should be taken that no members of that particular species are kept in close vicinity (e.g. in the same or an adjacent room). Birds may be able to be in acoustic contact even over a rather large distance, and the same may be true with regard to olfactory contact in mammals.

It is obvious that not all of these actions will be equally necessary or equally successful in all species. It is to be expected, however, that the more of such precautions can be taken the more likely it will be possible to change a social preference for the foster species into a preference for the individual’s own species. Along these lines, laboratory research on imprinting may also have considerable practical utility for keeping and breeding animals in captivity. Its results may be of special importance with regard to the propagation of endangered species and the husbandry of animals for return to the wild (cf. Cade 1980).

SUMMARY
Australian Zebra Finches foster-reared by Bengalese Finches show a definite preference for members of the foster parents’ species. This statement applies to males as well as to females. The preference is very durable: In adult males no change could be achieved even through several months of reproductive contact with a conspecific female. In adolescent males, in contrast, a permanent change in preference towards Zebra Finch females is still possible. Its likelihood decreases with age, but increases with the duration of intraspecific contact.

In addition to “species recognition”, early experience also plays a role in the development of sex recognition. In an intra-specific cross-fostering experiment, Zebra Finches were raised by mixed pairs of a “wild-colored” (naturally grey) male and a leucocistic (white) female and vice versa. Their sexual preferences proved to be sex-dependent: males showed a preference for the color of their mother, females preferred the color of their father.

The biological significance of early sensitive phases, the restriction of optimal learning capacities to periods of optimal learning conditions, is discussed. Possible practical implications for foster-rearing practices are described: To raise naturally behaving birds, hand- or foster-rearing should be avoided but, when such rearing really becomes necessary, the bird should afterwards be brought into exclusive and long-lasting contact with its own species as soon as possible.

REFERENCES
Balda, R. 1982. Sex-differences in Zebra Finch begging calls. Paper presented at the 99th meeting of the AOU, Chicago, October 1982.
Baptista, L.F. 1974. The effects of songs of wintering White-crowned Sparrows on song development in sedentary populations of the species. Z. Tierpsychol. 34: 147-171.
Bateson, P.P.G. 1966. The characteristics and context of imprinting. Biol. Rev. 41: 177-220.
Bateson, P.P.G. 1978. Early experience and sexual preferences. In: Hutchinson, J.B. (Ed.) Biological Determinants of Sexual Behaviour. Wiley, London, 29-53.
Bateson, P.P.G. 1979. How do sensitive periods arise and what are they for? Anim. Beh. 27, 470-486.
Brosset, A. 1971. L’ “imprinting”, chez les Columbides – Etude de modifications comportementales au cours du vieillissement. Z. Tierpsychol. 29, 279-300.
Cade, T.J. 1980. The husbandry of falcons for return to the wild. Int. Zoo Yearbook 20: 2335.
Cooke, F., Finney, G.H. & R.F. Rockwell. 1976. Assortive mating in Lesser Snow Geese (Anser caerulescens). Beh. Genetics 6, 127-140.
Fabricius, E. 1962. Some aspects of imprinting in birds. Symp. Zool. Soc. London, 8: 139148.
Gilbert, B.K. 1975. The influence of foster rearing on adult social behavior in the fallow deer (Damn dams). In: V. Geist (Ed.): Determinants of Behavior. University of Alberta Press, Calgary.
Hediger, H. 1950. Wild Animals in Captivity. Butterworths, London. Hess, E.H. 1973. Imprinting. New York. Van Nostrand.
Huck, U.W. and E.M. Banks. 1 980.The effects of cross-fostering on the behavior of two species of North American Lemmings, Dicrostonyx groenlandicus and Lemmus trimucronatus: I. Olfactory preferences. Anim. Behan. 28: 1046-1052.
Immelmann, K. 1962. Beitrage zu einer vergleichenden Biologie australischer Prachtfinken (Spermestidae). Zool. Jb. Syst. 90: 1-196.
Immelmann, K. 1969. Ober den Einfluf3friDhkindlicher Erfahrungen auf die geschlechtliche Objektfixierung bei Estrildiden. Z Tierpsychol. 26: 677-691.

Immelmann, K. 1972a. The influence of early experience upon the development of social behaviour in estrildine finches. Proc. XV Int. Ornith. Congr., The Hague 1970: 316- 338.
Immelmann, K. 1972b. Sexual and other long-term aspects of imprinting in birds and other species. Adv. Study Behav. 4: 147-174.
Immelmann, K. 1975. The evolutionary significance of early experience. In: G. Baerends, C. Beer & A. Manning (Eds.) Function and Evolution in Behaviour. Oxford, Clarendon Press, 243-275.
Immelmann, K. 1979. Genetical constraints on early learning: A prespective from sexual imprinting in birds. In: J.R. Royce & L.P. Mos (Eds.): Theoretical Advances in Behavior Genetics. Asphen aan de Rijn. Sijthoff & Noordhoff, 121-136.
Immelmann, K. 1983. Sexual imprinting in Zebra Finches- mechanisms and biological significance. Proc. XVIII Int. Orn. Congr., Moscow 1982 (in press).
Immelmann, K. & S. Sjolander 1983. Imprinting on father versus mother in the Zebra Finch. Anim. Behav. (in press).
Immelmann, K. & S.J. Suomi. 1981. Sensitive phases in development. In: Immelmann, K, Barlow, G.W., Petrinovich, L. & M. Main (Eds.) Behavioral Development. The Bielefeld Interdisciplinary Project. Cambridge University Press, New York, 395-431.
Klinghammer, E. 1973. The effects of handraising animals on their behavior towards con-specifics and other species. Proc. Am. Ass. Zool. Parks & Aquariums Ann. Conf. Houston 1973.
Klint, T. 1975. Sexual imprinting in the context of species recognition in female Mallards. Z. Tierpsychol. 38, 358-392.
Klint, T. 1978. Significance of mother and sibling experience for mating preferences in the Mallard (Anas platyrhychos). Z. Tierpsychol. 47: 50-60.
Lorenz, K. 1935. Der Kumpan in der Umwelt des Vogels. J. Ornithol. 83: 137-213 and 289- 413.
Morris, D. 1954. The reproductive behaviour of the Zebra Finch (Poephila gutata) with special reference to pseudofemale behaviour and displacement activities. Behaviour 6: 217- 322.
Sambraus, H.H. 1978. Nutztierethologie. Parey, Berlin, Hamburg.
Schutz, F. 1965. Sexuelle Pragung bei Anatiden. Z. Tierpsychol. 22: 50-103.
Schutz, F. 1969. Triebstrukturen und Fehlleitungen der Sexulitat bei Tieren. In: A. Schelkopf (Ed.), Sexulitat, Formen und Fehlentwichlungen, Vandenhoeck and Ruprecht, Gottingen: 33-54.
Schutz, F. 1 970.Zur sexullen Pragbarkeit undsensiblen Phase von G’;Ansen und der Farbedes Pragungsobjekts. Verh. Zool. Ges. Wurzburg 1969: 301-301.
Sonnemann, P. & S. Sjolander. 1977. Effects of cross-fostering on the sexual imprinting of the female Zebra Finch Taeniopygia guttata. Z. Tierpsychol. 45: 337-348.
Stamm, R. & U. Blum. 1971. Partnerwahl beim Wellensittich: Der Faktor Korperfarbe (Melopsittacus undulatus (Shaw); Aves, Pisttacidae). Rev. Susse Zool. 78: 671-679.
Warriner, C.V. Lemmon, W.B. &T.S. Ray. 1963. Early experience as a variable in mate selection. Anim. Behav. 11: 221-224.
–

—