Mouse pup urine as an infant signal

Mouse pup urine as an infant signal

Physiology &Behavior, Vol. 45, pp. 579-583. ©Pergamon Press plc, 1989. Printed in the U.S.A. 0031-9384/89 $3.00 + .00 Mouse Pup Urine as an Infant S...

535KB Sizes 0 Downloads 46 Views

Physiology &Behavior, Vol. 45, pp. 579-583. ©Pergamon Press plc, 1989. Printed in the U.S.A.

0031-9384/89 $3.00 + .00

Mouse Pup Urine as an Infant Signal TIZIANO

LONDEI,

PAOLA

SEGALA

AND VINCENZO

G. LEONE

Dipartimento di Biologia, Sezione di Zoologia e A n a t o m i a comparata Universit~ di Milano, via Celoria 26, 20133 Milano, Italy R e c e i v e d 19 F e b r u a r y 1988 LONDEI, T., P. SEGALA AND V. G. LEONE. Mouse pup urine as an infant signal. PHYSIOL BEHAV 45(3) 579-583, 1989.- Drops of urine from donors of different ages and sexes were applied to the anogenital regions of 19-day fetuses, killed by freezing, that ordinarily are weak in stimulating parental care. These fetuses were then presented to virgin, naive albino mice (Mus musculus). In Experiment 1, fetuses treated with urine of 6-day-old pups were more stimulating than fetuses treated with water. Differences in contact, picking-up, retrieval, licking and self grooming were significant. The sex of the donor revealed no effect. In Experiment 2, urine from 10-day-old pups was found to be more effective than urine from 4-day-old pups, with different effects on contact and retrieval. Urine from pups was also more effective than urine from adults, because of different effects on contact, picking-up, retrieval, licking and nest building. The effects of urine from adult virgin females differed less from those of pup urine than did the effects of urine from adult males. Compared with the latter the adult female urine had different effects on picking-up, licking and cannibalism. Speculations are made about an infant factor that might be still active in the urine of adult females. Urine

Infant signal

Parental behaviour

Mouse

Weak stimuli

LESS attention has been paid to the effects of urine in the interaction between parent and infant rodents than to its effects on sexual and social interactions. Urine acts as a chemical cue in all these interactions. An additional, different function has been revealed by studies of the rat: the nursing female ingests urine from the offspring during anogenital licking and thus recycles such vital resources as water and electrolytes (9). Urine of both the lactating females [(2) in (3)1 and the adult males (3) has been shown to pass information to the young that causes them to decrease motor activity, which probably acts to maintain family cohesion. Different pheromonal properties in urine o f males and females influence the age of puberty in young female mice (6). As for the information transmitted from the young to the adults, M o o r e (12, 13, 15, 16) found in a series of experiments with rats that adult females, whether or not they had young, are more attracted by urine from male pups and that the dams spend more time licking the anogenital regions of male pups because of this difference in urinary cue. This maternal stimulation contributes to development of male sexual behaviour (14). The general purpose of our study was to investigate the stimulation o f parental care by urine from mouse pups. As we have said, it is already known that in the rat there is a pattern o f parental behaviour, licking o f the anogenital region o f the pups, that is influenced by urine. Even we assume that the phenomenon should be the same in the two species, we do not feel that the existence of a single activity is sufficient for classifying urine as a cue involved in the recognition o f the infant stimulus-object by the parent. Therefore, we have looked at several behavioural patterns. Logically, study of the urine as an infant signal should be based on some knowledge o f the natural situations in which the pup emits urine. A search of the literature reveals only that it is emitted in response to licking. We have also seen urination during retrieving when the pup was grasped by the mother in some inappropriate way (the pups

Naive females

squeaked and wriggled). These were chance observations, but they fit into the hypothesis that infant urine might be able to influence more than one type of parental activity. On the other hand, in view of the great variety of situations that can cause urine to be emitted, we did not feel it would be useful to do a systematic study of only one o f these at this time. For practical reasons we decided to first see whether or not there is an infant signal. In the future, studies of the different activities of the adult that can be influenced by pup urine might provide suggestions for ways to study the most important natural situations to determine the adaptive value of this signal. GENERAL METHOD All the animals used were CD-1 albino mice from Charles River. Dead fetuses were treated with urine obtained from donors of different ages and sexes and were presented individually to virgin and naive females and the reactions o f these females were recorded. Both the adults and the fetuses were used only once. The study was based on a randomized block design. The virgin females, 2 to 5 months old, had had no earlier contact with pups except their litter mates. Under these conditions, the parental responsiveness of virgin female mice is similar to that o f primiparous dams (1,17). For the three days before the behaviour test each female was isolated in a standard Plexiglas cage ( 3 8 x 2 2 x 1 5 cm), with food and water ad lib, with sawdust as bedding and w o o d shavings with which the mouse constructed a sleeping nest. The fetuses were taken surgically on day 19 of pregnancy, which is about one day before parturition. They were removed still alive, then were cleaned with gauze swabs and killed by freezing at - 2 0 ° C , after which they were stored for various periods at the same temperature. A b o u t 10 minutes before the test, each fetus was thawed in distilled water and washed to

579

580

LONDE1, SEGALA AND LEONE

remove any remaining traces of amniotic fluid, then dried on filter paper and finally a drop of urine was placed on the anogenital region. When presented to the adult, the fetus had just reached room temperature (20-24°C). A previous experiment ( l l ) had shown that under these conditions but without the drops of urine, even 18-day fetuses could elicit parental responses in virgin females, but much less effectively than pups frozen on the third day after delivery. In this study we chose to use weak stimulus-objects so that any increase in effect due to treatment would be easier to detect. The urine was collected so as to avoid contamination with extraneous material and frozen and stored at - 2 0 ° C in aliquots. Freezing has been shown not to block the pheromonal action of the urine (6). Each aliquot was used only once after it was thawed. Just before the behaviour test, the subject was removed from her cage and the fetus was placed in the farthest corner from the sleeping nest. The female was then replaced in the cage on the nest and the metal mesh cover was replaced. Observations were recorded for 15 minutes from that time on a check sheet, starting with the first contact of the female with the stimulus-object. In addition to the usual patterns of parental behaviour, we looked for self grooming and for two activities observed during our previous work (l l), namely the female burying the fetus under the bedding or eating it (cannibalism). Each contact outside the nest was recorded, to evaluate subliminal interest in the stimulus-object. The sequence of retrieving was divided into picking-up the fetus, moving it near or far from the original position, and carrying it into the nest (retrieval). This differentiation was made to help evaluate the uncertainty of the subject. Contact, picking-up and retrieval could easily be recorded when they occurred since they are short procedures and easy to see. Patterns that usually go on for longer, such as licking the body of the fetus, nest building (from sleeping nest to nurturing nest), nursing position and self grooming, were only recorded if a single pattern lasted for at least 20 consecutive seconds (17), long enough for us to be sure about what the mouse was doing. On the check sheet, for each of the 15 minutes of the test, each type of behaviour was recorded as having occurred or not occurred, regardless of frequency or duration during that minute. As a result, if a pattern was recorded to have occurred, its latency and duration (or, better, its repetitiveness) could have varied from 1 to 15 units. Latency of contact was measured from presentation. For the other patterns, latency was measured from the first contact. The analysis of variance of latency and repetitiveness produced a few significant results, that are given individually in the text. Only the number of subjects showing the various patterns and comparisons of their frequencies are given in tables. EXPERIMENTI The first phase of the study was to ascertain whether a urinary factor is one of the various infant stimuli that induce parental behaviour. At the same time, we considered whether or not the effects of urine of the different sexes of the pups might differ. A short communication of some of the results has already been published in Italian (10). METHOD

One hundred and twenty fetuses were divided into a control group and an experimental group which was again divided into two subgroups. Sixty fetuses were treated with drops of urine

from 30 male or 30 female 6-day-old pups (day of birth taken as day l). Pups of this age were used as donors because they are definitely still infantile but the metabolism should be sufficiently developed to produce effective urine. When picked up by the skin of the back, the pups emitted drops of urine that were drawn up into a Pasteur pipette before they came into contact with any appreciable zone of skin or with feces. The urine of each pup was stored separately and used to treat only one fetus. The sixty control fetuses were treated with drops of water. RESULTS AND DISCUSSION

Since none of the adults assumed the type of nursing position that virgin females presented with a pup assume, this response is not included in Table I. There were no significant differences in the effects of urine from donors of different sex and the great similarities of the results indicate that under these circumstances male and female urine has the same effect. Table 1 also contains the data for control fetuses (water). There was a highly significant preference for the fetuses treated with urine in terms of retrieving and licking. Even the difference between controls and urine-treated animals in inducing self grooming may indicate a positive effect of urine on parental behaviour. Self grooming is likely to be a displacement activity caused by some difficulties in expressing parental behaviour toward dead animals (1 l). Therefore, the greater frequency of this activity in the presence of a fetus treated with urine may indicate that adults receive greater stimulus for parental care from urine-treated fetuses than from water-treated fetuses. From the data in Table 1 we can also calculate the ratio of the number of fetuses retrieved into the nest to the number of fetuses only picked up. There were more such fetuses in the group treated with urine than in the group treated with water, X2(1)=5.07, p<0.025, indicating that retrieving was less certain in the second case. Burying mainly occurred at the beginning of the behavioural response and was associated with avoidance reactions. Cannibalism was never a result of attack (the opposite of predatory behaviour) but it followed licking out of or into the nest. In the first case the fetus was never retrieved, in accordance with the tendency of this species to carry no food to the nest. Considering that, the second case, first retrieving and then eating the fetus, indicates cannibalism not to be a primary motive for licking either. As previously (11) hypothesized, eating may be a mere consequence of persistent licking that has damaged the skin of a stimulus-object that cannot defend itself. Alternatively, even before the skin damage, yet after considerable licking, the subject might get some cue to eliminate the abnormal stimulus-object. In general, burying tended to delay and cannibalism to put an end to the parental care (licking included). The problem may arise whether the subjects showing either of these two activities (which were largely independent from each other) made some different discrimination, between experimental conditions, in comparison with apparently more maternal subjects, so that the data should be presented separately for different categories of subjects. Table 2, however, shows no significant heterogeneity. Though chi-square test is largely approximate with such small frequencies, it indicates that buriers, cannibalizers and the other females discriminated in the same way between urine-treated and control fetuses in connection with typical parental patterns. Further suggestion arises that such activities as retrieving and licking were parentally motivated in cannibalizers too. Otherwise, cannibalism being the motive, one would expect some preference reversal in com-

MOUSE PUP URINE AS AN INFANT SIGNAL

581 TABLE 1

NUMBER OF NAIVE FEMALES RESPONDING TO THE PRESENCE OF DEAD FETUSES TREATED WITH URINE FROM 6-DAY-OLD MALE OR FEMALE PUPS OR WITH WATER

Activity*

Male Urine N=30

Female Urine N=30

Pt

Urine N=60

Water N=60

Pt

30 22 19 25 12 27 3 6

29 19 19 22 12 27 5 6

ns ns -ns ns -

59 41 38 47 24 54 8 12

57 25 17 30 16 43 9 10

ns <0.01 <0.001 <0.005 ns <0.05 ns ns

Contact Picking-up Retrieval Licking Nest building Self grooming Burying Cannibalism

*Contacts were evaluated in proportion to the presentations (N), the other activities were evaluated in proportion to the contacts. 1-Chi-square test with Yates' correction; level of significance at p=0.05.

TABLE 2 NUMBER OF BURIERS, CANNIBALIZERS AND THE OTHER NAIVE FEMALES OF TABLE 1 (*) THAT SHOWED THREE TYPICAL PARENTAL ACTIVITIES OR NOT

Activity Retrieval Licking Nest building All three None

Buriers Urine Water N=7 N=7 3 4 1 0 2

0 1 2 0 4

Cannibalizers Urine Water N = 11 N=8 10 11 5 5 0

4 6 0 0 1

Urine N=40 24 31 18 12 7

Others Water N=40 13 21 14 7 14

Pt 0.25-0.50 0.50-0.75 0.10-0.25 0.10-0.25 0.75-0.90

*One subject in urine condition and two in water condition, the only ones that both buried and ate the fetus, are here excluded to achieve independence between categories. tChi-square test for heterogeneity; the resulting probabilities fall within the listed intervals.

parison with n o n c a n n i b a l i z e r s . T a b l e 2 also shows t h a t m a n y subjects were n o n m a t e r n a l without burying or eating the fetus. Male a n d female urine h a d also similar effects o n latency a n d o n repetitiveness o f the activities. Fetuses treated with water were licked m o r e quickly, F(1,75)=6.94, p < 0 . 0 2 , a n d were c o n t a c t e d m o r e times before they were definitely a b a n d o n e d or carried into the nest, F(1,114)=6.88, p < 0 . 0 1 . Both these p h e n o m e n a are f o u n d to be associated with weak retrieving (11). Licking lasted longer with fetuses treated with urine, F(1,75)=5.10, p < 0 . 0 5 . After retrieval, the fetus was often partially h i d d e n by the b o d y o f the adult or by the nest material, especially after nest building. Therefore, we did n o t try to differentiate between licking o f the anogenital region a n d licking o f other parts o f the body. This is one o f the m a n y differences between o u r experimental conditions a n d those o f M o o r e , so we c a n n o t explain w h y o u r results a b o u t a sex difference differ f r o m hers. F r o m the totality o f our data, it is clear that the urine of the m o u s e p u p has a positive effect o n p a r e n t a l b e h a v i o u r . EXPERIMENT 2 F r o m the results o f E x p e r i m e n t 1, the p r o b l e m arose o f whether or not the effect o f urine o n parental b e h a v i o u r might

be due to some properties that are unique to p u p urine a n d not to adult urine. W e could n o t be sure t h a t any type o f m o u s e urine might not be able to interact with other characteristics o f the stimulus-object to reinforce its i n f a n t effectiveness. O n the other h a n d , the effectiveness o f urine might change even in the first days o f life. W e decided to c o m p a r e the urine o f pups o f two ages a n d adults o f b o t h sexes. It was difficult to choose a d u l t d o n o r s because differences in social status a n d in physiological conditions m i g h t influence the effects o f their urine, even if in this experiment it was used in a n artificial situation. O n the other h a n d , to study m a n y categories o f d o n o r s would have made the experiment excessively cumbersome. W e decided t h a t at this stage o f our research it would be sufficient to look for any type o f urine from adults of either sex that would have a different effect f r o m p u p urine. METHOD O n e h u n d r e d a n d thirty-six subjects were divided into four experimental groups o f 34 each. G r o u p A fetuses were treated with urine f r o m 4-day-old pups, g r o u p B fetuses with urine f r o m 10-day-old pups. T h e urine was collected f r o m the pups a n d stored as described for E x p e r i m e n t 1. Because o f the preceding results, we did n o t take into a c c o u n t the sex o f the

582

LONDEI, SEGALA AND LEONE TABLE 3 NUMBER OF NAIVE FEMALES RESPONDING TO THE PRESENCE OF DEAD FETUSES TREATED WITH URINE FROM EITHER 4-DAY-OLD PUPS, OR 10-DAY-OLDPUPS, ADULT MALES, ADULT VIRGIN FEMALES

Activity* Contact Picking-up Retrieval Licking Nest building Self grooming Burying Cannibalism

A 4-day N=34

B 10-day N=34

C Ad.m. N=34

D Ad.f. N=34

A vs. B Pt

C vs. D pt

A+B vs. C+D Pt

33 22 19 24 12 26 7 18

33 29 28 26 19 27 4 12

34 13 10 9 6 27 11 6

34 26 18 22 10 26 8 14

ns <0.05 ns ns ns ns ns

<0.001 ns <0.005 ns ns ns <0.05

ns <0.02 <0.001 <0.001 <0.005 ns ns ns

*Contacts were evaluated in proportion to the presentations (N); the other activities were evaluated in proportion to the contacts. iChi-square test (hierarchic classification); level of significance at p=0.05.

infant donor. Group C fetuses were treated with urine from adult males and group D fetuses with urine from adult females. The adult donors were 2 to 3 months old, had never been mated and were housed in cages in groups of 5 mice of the same age and sex. Each adult was picked up by the back and held over a Petri dish. This caused most of the mice, but not all, to urinate. Five cages of males and seven cages of females provided a total of 19 male donors and 23 female donors. Urine from mice in a single cage was collected in a single Petri dish and mixed before being frozen in aliquots. If differences in condition of the adults might produce types of urine with different effects on parental behaviour, using pooled urine rather than individual urine might increase the effects of the less frequent types and decrease the variance of the results, thus increasing the possibility of seeing a significant effect. All the fetuses o f the four groups were treated with equal volumes of urine (0.01 ml). RESULTS AND DISCUSSION Three types of comparison of effects were made with these four types of urine: urine from 4-day-old and 10-day-old pups (A vs. B), urine from adult males and adult females (C vs. D) and the cumulative data for urine from pups and adults ( A + B vs. C + D ) . Since in this experiment again, as in Experiment 1, buriers, cannibalizers and the other subjects appeared to discriminate between the stimulus-objects in the same way, no separation is presented here. Table 3 does not include the nursing position because there were too few cases to make any comparison (zero in A, three in B, one in C and one in D). Retrieving was stimulated more by urine from the older pups than from the younger ones and by pup urine than by adult urine, whereas for the fetus to be picked up only, adult female urine was significantly more effective than adult male urine. Since the trend for picking-up appears to be a little different from that for retrieval, we also used the data in Table 3 to compare the numbers of fetuses retrieved as a proportion of the total number of fetuses picked up. The only significant difference was between A + B vs. C + D , X2(1)=4.15, p < 0 . 0 5 , indicating that retrieving was less certain in the second case. Table 3 also shows that pup urine was more effective in inducing licking and nest build-

ing. In addition, fetuses were cannibalized by more subjects if they had been treated with urine from adult females than with urine from adult males. Since the first of these were also licked by more subjects, cannibalism might well be a consequence of licking dead animals, supporting the hypotheses outlined in Experiment 1. Orthogonal contrasts were used for the analysis of variance. There were no significant differences in latencies of the activities. Since neither the mean latency of contact nor the mean latency of retrieval differed in the groups, the differences in repetitiveness of contact [ A > B : F(1,130)=5.80, p < 0 . 0 2 5 ; A + B < C + D : F(1,130)=19.33, p<0.005] appear to be due by and large to those cases in which the adult left the fetus outside the nest for the entire duration of the test but returned to make contact several times. As a consequence, it appears that the virgin females had not lost interest in these fetuses, but that their interest did not end into adoption. This interpretation might also explain how the repetitiveness of contact and the frequency of subjects showing retrieval can have significant differences o f opposite sign in the same types of comparison. Urine of pups also caused more repetitiveness of licking than urine of adults [ A + B > C + D : F(1,77)=7.66, p < 0 . 0 1 ] . We draw the following conclusions from the results of this experiment. Firstly, the effectiveness of pup urine on parental behaviour varies with the age of the donor. Secondly, pup urine is more effective than adult urine. This appears not only from combined group but also from individual group comparisons. The difference in the frequency of retrieval between D and B is greater than between A and B (Table 3). Therefore, urine from 10-day-old pups remains the most effective. Thirdly, adult urine has different effects on parental behaviour according to the sex o f the donor. Since adult male urine is less effective than adult female urine, the latter seems to be less different from infant urine than is adult male urine. Whether adult urine has positive, negative or no effect on parental behaviour is unclear, since it cannot be rigorously compared with the controis of Experiment 1. Concerning the occurrence of parental activities in the different conditions, the percentages that can be computed from Tables 1 and 3 make increasing sequences as follows. For retrieval: ad.m.=29.4, water-29.8, ad.f.=52.9, 4-day=57.6, 6-day=64.4, 10-day=84.8. For licking: a d . m . =

M O U S E P U P U R I N E AS A N I N F A N T S I G N A L

583

26.5, water=52.6, ad.f.=64.7, 4-day=72.7, 10-day=78.8, 6-day=79.7. For nest building: ad.m. = 17.6, water=28.1, ad. f. = 29.4, 4-day=36.4, 6-day=40.7, 10-day=57.6. The 6-day results fit the hypothesis about a continuous increase in the effectiveness of urine during the first days after birth. Since a sudden rise o f such a property between days 5 and 6 seems unlikely, 4-day urine should already produce positive effects on parental behaviour, as these percentages indicate. Finally, considering that no significant difference can be found between responses to 4-day and adult female urine from the data in Table 3, there is a suggestion that, unlike adult males, adult females retain some infant effectiveness in their urine. GENERAL DISCUSSION Experiment 1 showed that the presence of urine on the skin is one of the characteristics of the mouse pup that can stimulate parental behaviour. Experiment 2 showed that pup urine can be more effective than adult urine. Pup urine can be called an infant signal if we define a signal as a stimulus that is more active in eliciting a given type of response within a given sensory modality than other comparable stimuli, though not necessarily more active than other stimuli that act through other sensory modalities. In both experiments, the activities that were most influenced by urine were retrieving and licking. W e expected the effects on licking because of the results of studies of the rat, but the relationship between pup urine and retrieving opens up new research perspectives, for which we have some suggestions to make.

Since the pup can emit urine when it is transported too roughly, this might block the action o f the adult and cause the pup to be dropped. In this way the pup might avoid being injured (pup urine might also block aggression by alien adults) and at the same time maintain the interest of the parent in itself and the parent then might return to complete the retrieval. Having an acute effect does not mean that it cannot also have a chronic effect. In spite o f licking by the mother, the urine might leave a persistent chemical signal on the skin of the pup and this signal could contribute to symbolize the infant identity. During growth, when it has begun to move about autonomously, a young mouse becomes ever more reluctant of being carried back to the nest. In addition, the overall infant effectiveness decreases with age. From this point of view, the greater effectiveness of urine of 10-day-old pups than that o f urine of 4-day-old pups might indicate a compensation mechanism. We do not know at present what the nature o f the infant factor in urine is. However, it is known that the composition of rat pup urine changes during development [(7) and (18) in (8)]. The increasing infant property of the urine, during the first days after birth, might be a mere consequence of metabolic development. The decreasing effect, at some later age, might be caused, in part, by opposing factors depending on sexual development. The aggression-eliciting potency of male urine is known to be androgen-dependent (19). Urine of weanling females inhibits aggression in adult males, whereas urine o f weanling males apparently has no ability to either inhibit or stimulate aggression in adult o f the same sex (5). An inhibitory factor, probably the same one, is also present in urine of adult females and differs from the factor that arouses male sexual behaviour (4). It might be the infant factor suggested by the present study.

REFERENCES 1. Beniest-Noirot, E. Analyse du comportement dit "maternel" chez la souris. Monogr. Fran¢. Psychol. vol. 1. Paris: CNRS; 1958. 2. Cowley, J. J.; Wise, D. R. Pheromones, growth and behaviour. In: Porter, R.; Birch, J., eds. Chemical influences on behaviour: CIBA foundation study group. London: Churchill; 1970. 3. Csermely, D. Locomotor activity in newborn mice: II. Presence and influence of a male pheromone. Boll. Zool. 48:179-182; 1981. 4. Dixon, A. K.; Mackintosh, J. H. The relationship between the physiological condition of female mice and the effects of their urine on the social behaviour of adult males. Anim. Behav. 23:513-520; 1975. 5. Dixon, A. K.; Mackintosh, J. H. Olfactory mechanisms affording protection from attack to juvenile mice (Mus musculus L.). Z. Tierpsychol. 41:225-234; 1976. 6. Drickamer, L. C. Urinary chemosignals from mice (Mus musculus): acceleration and delay of puberty in related and unrelated young females. J. Comp. Psychol. 89:414-420; 1984. 7. Falck, G. Maturation of renal function in infant rats. Am. J. Physiol. 181:157-170; 1955. 8. Gubernick, D. J.; Alberts, J. R. Maternal licking of young: Resource exchange and proximate controls. Physiol. Behav. 31:593601; 1983. 9. Gubernick, D. J.; Alberts, J. R. Maternal licking by virgin and lactating rats: water transfer from pups. Physiol. Behav. 34:501-506; 1985. 10. Leone, V. G.; Londei, T.; Segala, P. Comportamento parentale nel topo (Mus musculus): effetto dell'urina infantile. Istituto Lombardo (Rend. Sc.) B 120:11-16; 1986.

11. Londei, T.; Segala, P.; Leone, V. G. Foetuses are less effective than pups in eliciting murine parental behaviour. Boll. Zool. 54: 65-67; 1987. 12. Moore, C. L. An olfactory basis for maternal discrimination of sex of offspring in rats (Rattus norvegicus). Anim. Behav. 29:383-386; 1981. 13. Moore, C. L. Maternal behavior of rats is affected by hormonal condition of pups. J. Comp. Physiol. Psychol. 96:123-129; 1982. 14. Moore, C. L. Maternal contributions to the development of masculine sexual behavior in laboratory rats. Dev. Psychobiol. 17:347356; 1984. 15. Moore, C. L. Sex differences in urinary odors produced by young laboratory rats (Rattus norvegicus). J. Comp. Psychol. 99:336341; 1985. 16. Moore, C. L.; Morelli, G. A. Mother rats interact differently with male and female offspring. J. Comp. Physiol. Psychol. 93:677684; 1979. 17. Noirot, E. Changes in responsiveness to young in the adult mouse. 1. The problematical effects of hormones. Anim. Behav. 12:5258; 1964. 18. Oh, W.; Stewart, L.; Baens, G. S.; Metcoff, J. Body composition and renal adaptation in the newborn rat. Biol. Neonate 8:65-80; 1965. 19. Smith, J. C. Senses and communication. In: Berry, R. J., ed. Biology of the house mouse. London: Academic Press; 1981:367-393.