Play fighting in juvenile golden hamsters (Mesocricetus auratus): Effects of litter size and analysis of social interaction among males

Behavioural Processes 43 (1998) 265 – 273

Play fighting in juvenile golden hamsters (Mesocricetus auratus): Effects of litter size and analysis of social interaction among males Mauro Luı´s Vieira a,*, Emma Otta b a b

Departamento de Psicologia, Uni6ersidade Federal de Santa Catarina, 88040 -900 Floriano´polis, Santa Catarina, Brazil Instituto de Psicologia, Departamento de Psicologia, Experimental, Uni6ersidade de Sa˜o Paulo, 05508 -900 Sa˜o Paulo, Sa˜o Paulo, Brazil Received 25 March 1997; received in revised form 12 February 1998; accepted 24 March 1998

Abstract Previous studies have shown that different social experiences, such as the interaction involving partners of different ages or among individuals of different sexes, significantly alter the play rate. In the present study, two experiments are described, the hamster being used as subject, to investigate the effects of litter size and to analyse the play fighting behavior among males. It was found that there was no significant difference in the time spent in play fighting by an animal when it was placed in groups of 2, 4 or 6 individuals (Exp. 1). Moreover, the majority of animals did not show preferences during play fighting when they were placed in groups formed by four males (Exp. 2). Through these data it has been noted that the hamster distributes the total time spent in play fighting among the available partners. This fact can be explained by the hamster’s social organization, because it is a solitary animal, which does not develop a clear and durable interaction with its playmates. Also, through other results, a certain stability was found in most of the behavioral categories registered such that they were not significantly affected by the manipulated variables. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Play fighting; Golden hamsters; Litter size; Social interactions

1. Introduction

* Corresponding author. Fax: +55 48 3319751; e-mail: [email protected]

In several species of mammals, the young stay together throughout their development. In this period, the physical proximity among them, mainly after birth, facilitates their thermal regula-

0376-6357/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0376-6357(98)00020-5

266

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273

tion when the mother is absent from the nest or shelter. Through competition for maternal milk and solid food, the young develop physical abilities. Another means of reaching the same end is through play. According to West (1974), play supplies physical exercise for many parts of the body. Also, social play becomes adapted in such a way as to maintain the proximity of individuals, facilitating the mother’s task in protecting them. Different social experiences, resulting from the absence of siblings or from interactions among siblings of different sexes, have significant effects on the physical and social development of the young. For example, the total absence of partners, sibling or nonsibling, during the juvenile period when play is normally more frequent, has evident consequences on the individual and on the social behavior of animals. The privation of physical and social contacts with siblings can produce atypical behaviors and stereotyping in cats (Guyot et al., 1980). In hamsters (Vieira, 1991) and in cats (Mendl, 1988), it was found that the mother did not totally compensate for the absence of siblings. Besides the absence and the interaction with individuals of different ages, it has been found that the partner’s sex is an extremely strong determinant in the modulation of the play. In most of the studies carried out with several species, from rodents to primates, it has been found that the males play more than the females (Meaney et al., 1985). Due to this difference, the behavior of partners of the same sex becomes more attractive to members of the opposite sex (Jamieson and Armitage, 1978; Meaney and Stewart, 1981; Guerra et al., 1992). Caro (1981), analyzing the behavior of cats, found that the frequency of playful interactions among females increased when at least one male was present. On the other hand, the play fighting of the males, in a group with two males and one female, was not very affected by the presence of the female. However, the play of the males decreased when two females were present. Some factors that have received little attention, with reference to the modulation of play,

are litter size and the analysis of play among partners of same sex. The two experiments carried out had the objective of investigating these two variables in the playful interaction in juvenile golden hamsters.

2. General methods

2.1. Subjects In total, 120 and 32 juvenile male golden hamsters were used in Experiments 1 and 2, respectively. Their ages were between 25 and 32 days. The reason for the choice of this age group is related with the frequency of play fighting behavior. Although this behavior begins to appear around the 15th day of life of the hamster (Guerra and Vieira, 1990), it reaches a maximum between 25 and 35 days, and after the 40th day, aggressive interactions become more frequent (Pellis and Pellis, 1988). Thus, after the 40th day, play fighting can be easily confused with aggression, although these two types of behavior are different in adult life. All the animals used in our experiments were obtained from the stock of the Laborato´rio de Psicologia Experimental at the Universidade Federal de Santa Catarina. From the period of mating to the end of the observation sessions, the subjects were housed in rooms (with an ambient temperature of 2394°C) under a 12-h light, 12-h dark cycle, with light coming on at 07:00 h. The weight difference among the partners did not surpass 10%.

2.2. Materials The animals were placed in polypropylene home cages that measured 40× 30×17 cm, containing a standardized amount of wood shaving and pellets of food. Food and water were available ad libitum. Standard note-sheets and chronometers were utilized during the experimental sessions for recording behavior. In Experiment 2, a video camera was also used to film the animals.

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273

2.3. Beha6ioral categories The choice of the behavioral categories was made based on empirical works from the literature and on experimental works accomplished in our laboratory. Pilot experiments were realized in order to test the viability and the operating efficiency of the experiments, reliability tests being carried out by four observers. The index of reliability obtained was at least 85%. The categories selected for analysis in the experimental studies were: (1) Play fighting: the subjects engaged in vigorous physical interaction, normally facing one another and making fast and frequent movements of the front paws. Light body bites sometimes occurred, usually directed to the partner’s ventral or front part. Typical behaviors of attack, defense and counter-attack took place, without, however, any wounds being inflicted, resulting in the complete submission of one of the partners involved. The duration and the frequency of the play fighting bouts were recorded. Also, the forms of this type of social interaction were quantified. A focal animal was observed in each one of a number of positions in relation to his partner: on top (active pinning), underneath (passive pinning), or side-toside. It was also observed how many times the focal animal changed these positions during the play fighting; (2) situation subsequent to play fighting: after the interruption of the play fighting, it was recorded whether the animals remained near each other or whether they withdrew. It was considered that they were distant when the minimum distance between them was more than the length of the body; (3) physical contact: the time that the focal animal spent in corporal contact with his partners was recorded, independent of the part of the body involved. Animals could be in passive social interaction (sleeping, for example) or in active social interaction (eating, nest building, play fighting, etc.). For the purposes of analysis, in Experiment 1, the time spent in physical contact: (a) with a partner and (b) with two or more partners, was differentiated; (4) locomotion: through coordinated movements of the paws, the focal animal changes the position of the body from one place to another. A locomotion bout began when the subject moved starting from an

267

immobile position and it ended when the subject came back again to the rest position, with a duration of 3 s or more.

2.4. Statistical treatment The data obtained were analyzed using one-way ANOVA and Tukey’s HSD post hoc test. On the other hand, comparisons between only two behavioral measurements were made with the Student’s t-test. Finally, for analyzing the level of relation between two behaviors, Pearson’s Correlation was used.

3. Experiment 1: Effects of litter size on play fighting It has been very well established in the literature that the litter size influences the modulation of the parental behavior. It also influences the development of the young. For example, the Mongolian gerbil pup has a more rapid physical development when in litters with two siblings, in comparison with those in litters composed of one or five other siblings (Elwood and Broom, 1978). The objective of the present experiment was to investigate the influence of litter size (2, 4 or 6 pups) on the frequency and the duration of the play fighting in golden hamsters. We expected to find a higher play fighting rate among groups with more individuals, where there was a greater possibility of social interaction.

3.1. Procedure When they had completed 25 days of age, the hamsters were separated from their original litters and distributed equally in three groups, composed of 2, 4 and 6 animals, denominated: LS-2, LS-4 and LS-6 (LS = litter size). Each group was composed of ten litters. For the composition of the new litters, only animals that were not siblings were used. After this experimental manipulation, the animals were taken to a separate room, where they stayed until the end of the experiment. The observations began 24 h after the experimental manipu-

268

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273

lation. Observation was for a duration of 20 min a day and was carried out for seven successive days, until the animals had reached 32 days of age. In each one of the litters, independent of the number of animals, one of the subjects was chosen as the focal animal for registration of the behavioral categories. To facilitate discrimination of the focal animal, a subject that was different in color or coat was chosen. When this was not possible, a small nick was made in one of the subject’s ears.

3.2. Results The total time spent in physical contact was affected by the litter size [F(2,27) =4.74; P B 0.025]. Through the two-to-two comparison of the groups, it was found that this difference existed only between the groups LS-2 and LS-4 (DHS= 138.1; P B0.025). Physical contact was higher in the group that had a larger number of individuals (Fig. 1). In relation to the percentage of time the focal animal spent in physical contact with the partner being compared or with several simultaneously, in the groups TN4 and TN6, it was found that a significant difference only existed in the latter group. In TN4, the focal animal spent 51.3 9 2% in physical contact with a partner and another

Fig. 1. Mean percentage ( 9S.E.M.) of observation time that the focal subject spent in physical contact with partner(s) in each of the experimental groups.

47.79 2% with two or more partners [t(9)= 1.38; P\ 0.05]. However, in TN6 the focal animal spent significantly more time in contact with various partners simultaneously than with a single partner. With a single partner, the focal animal spent 42.49 3% of the time, the other 57.69 3% being distributed at the same time among several partners [t(9)= 2.69; PB 0.05]. Unlike the case of physical contact, in relation to play fighting there were no significant differences among the three groups [F(2,27)= 1.15; P\ 0.05]. On the other hand, in all the groups, after the end of the playful interactions, the animals were more often close to one another than they were distant. Through statistical analysis, it was found that differences were significant [TN2: t(18)= 5.96; PB 0.01; TN4: t(18)= 4.18; P B 0.01; TN6: t(18)= 7.77; PB 0.01]. The frequency of locomotion bouts was another behavioral measurement that did not present significant difference among the groups [F(2,27)= 0.25; P\ 0.05]. The number of partners in the home cage was not a sufficiently important factor to alter the rate of the individuals’ displacements in space.

4. Experiment 2: Play fighting among males In hamsters, it was found that both males and females spent approximately the same amount of time in play fighting (Guerra et al., 1992). However, this only happened when the pairs were of the same sex. When individuals of different sexes formed the pairs, the play rate was significantly lower. This suggests that the partner’s sex is a variable that modulates the occurrence of playing. In rats, there is variability in the frequency of the play behavior among siblings, and the most playful individuals engage more in play with others that present similar levels of this behavior, instead of playing with many different partners (Poole and Fish, 1976). In Experiment 1, it was found that there was no significant difference regarding the play fighting rate among hamsters in litters of different sizes. However, it was not discovered whether there were differences or preferences among individuals of the

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273

269

Table 1 Mean percentage ( 9S.E.M.) of time spent in play fighting by subjects of litter 6 with each of the partners and the associated probability values Subjects

1 2 3 4

Partners

Statistical analysis

A

B

C

(F-value)

14.2 9 5 52.1 9 9 52.7 9 11 18.0 9 5

67.29 7 26.29 7 25.9 9 8 37.19 13

18.698 21.6 9 5 21.4 9 4 44.99 12

F(2,10) =13.11* F(2,12) =4.00** F(2,12) = 2.67 F(2,12) =1.11

* PB0.001; ** PB0.05.

same litter. The objective of this experiment was to investigate the time spent in play fighting by the hamster. In individuals of the same sex, it would be distributed equally among the partners or, as is the case with the rats, there would be a preference to play with some particular individuals. However, in the event of an individual having this preference, would the other hamster necessarily respond in a similar fashion?

4.1. Procedure In this experiment, when the hamsters reached 25 days of age, they were also taken away from their original litters and placed with other partners of the same age. None of the animals in each new litter were siblings. In total, eight new litters were formed, each with four individuals. After the formation of each litter, the animals were taken to a separate room, where they stayed until the end of the experiment. Twenty-four hours after this experimental manipulation, each litter was videotaped for 20 min once per day, from the 26th to the 32nd day of age, and each tape was observed four times each for a different focal animal. To facilitate discrimination, subjects of different colors were used. As there were four individuals in each grouping, in total 560 min of observation by grouping were accomplished, with a total of 4480 min for the whole experiment.

4.2. Results Through the statistical analysis of the data for

the time spent in physical contact among the individuals of each one of the litters, it was found that there was no significant difference in any of the litters [F(3,24): Litter 1 (LS-1)= 0.40; LS-2= 0.71; LS-3= 0.46; LS-4= 0.47; LS-5= 1.50; LS-6= 0.36; LS-7= 2.44; LS-8= 2.45; in all the cases P\ 0.05]. The average percentage of the time spent in physical contact, all the individuals being analyzed simultaneously, was 66% of the total time of observation in the 7 days of the experiment. Through a comparison among the eight litters it was also found that there was no significant difference among them [F(6,42)= 1.08; P \ 0.05]. On the other hand, in the analysis of the time spent in play fighting, it was found that there was a significant difference in litter 6 only [F(3,24): 4.47; PB 0.05]. In contrast, in the analysis carried out on the eight litters, no significant difference was found [F(7,48)= 0.21; P\ 0.05]. The average percentage of the total time all individuals spent in play fighting was 18.8%. Analyzing litter by litter, in order to discover what preference there would be in the play fighting among several equal pairs, it was found that a preference existed in just one of the litters [LS-6: F(3,36)= 6.54; PB 0.001]. Through a detailed analysis of this litter, using post hoc testing to identify which of the equal pairs played more, it was found that one of the pairs played significantly more than the other pairs. There was no significant difference among the other equal pairs. In this litter, significant difference was also registered when the total time spent in play fighting by the individuals with each one of the partners was

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273

270

Table 2 Mean percentage ( 9S.E.M.) in relation to the alternate pinning position during play fighting in each of the litters that demonstrated a significant difference. The associated probability values are indicated in each of the comparisons Litter

1 4 6 7

Subjects

1 1 1 1 2 3 4

Alternate pinning position

Statistical analysis

Active pinning

Side-to-side

Passive pinning

F(2,10)

29.5 95 27.3 92 28.09 4 24.09 2 28.19 3 27.7 9 4 31.69 3

46.6 9 3 44.7 9 4 44.7 9 2 44.1 9 2 46.5 9 3 45.3 9 4 42.5 9 2

23.9 9 6 28.0 9 3 27.4 9 4 32.0 9 2 25.3 9 4 27.2 9 3 25.9 9 3

4.58*** 6.89** 5.96** 18.62* 7.39** 5.62** 6.04**

* PB0.001; ** PB0.025; *** PB0.05.

analyzed (Table 1). Through the identification of the partners, it was found that there was reciprocity in relationship to the preference. The subjects 1 and 2 played more significantly with each other than with other partners. In relation to the forms of play fighting behavior, of the 32 individuals analyzed separately, in 78% of the cases there was no significant difference in the positions assumed in playful interactions (on top, underneath or side-to-side) for the focal animal. In those cases in which there was significant difference (22%), the most frequent position was side-to-side, as may be visualized in Table 2. It is important to point out that all the individuals presented significant differences only in litter 7. To exemplify what happened in the previous experiment, the subsequent situation to the playful interaction being analyzed, it was found that in all the litters the individuals more frequently stayed close to one another rather than distant. Therefore, through this data, it was found that the individuals’ permanence close to one another after the play fighting is a general phenomenon, and that this happened with almost all the individuals, independent of the litter they belonged to. Finally, the general analysis of the frequency of the individuals’ locomotion in the eight litters did not indicate significant difference [F(7,48) = 1.40; P\0.05]. However, when the four individuals of each were analyzed, it was found that there were significant differences in two of them.

In both cases, it was noted that one of the individuals moved around significantly more than the other three. One might ask whether the individual that moves around more also plays more. Crossing these two behavioral measurements, a positive correlation was found in just six cases (18.8%). In this case, the largest locomotion rate was directly related with the largest play fighting rate. However, in the great majority of cases (81.2%) this fact was not found. Also, in five cases (15.6%) the correlation was negative, but not significant. When the locomotion rate was high, the play fighting rate was low and vice-versa. Moreover, comparing the 32 individuals simultaneously, absence of significant correlation was found between the two behaviors (Pearson’s index correlation= 0.09).

5. Discussion The litter size was a factor that interfered significantly in the time that the animals spent in physical contact. The rate of this behavior was higher in the litters that had more individuals. This is explained by the higher probabilities for interaction with partners. In the group LS-6, the focal animal spent more time in contact withseveral partners than with just one. The vast number of individuals sharing the same space, and also the tendency to nestle together while resting, can explain this fact.

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273

Opposite to the situation of physical contact, the play fighting rate was not significantly affected by litter size. The focal animals spent about 19% of the total time of observation in playful interaction in both experiments. This value can be due to the period of the day in which observation was made, that is, between 16:00 and 20:00 h. As the hamster is a twilight animal, this schedule corresponds to the period of the day of greater activity (Lerwill, 1974). In natural situations, it has been found that the time spent in play fighting in various species is less than 10% of the total time throughout the day (Fagen, 1981). Through measurements of oxygen consumption among cats, it was found that the daily energy spent in play fighting was between 4 and 9% (Martin, 1984). Among deer, activity in moving around consumes 3% of the energy used every day (Miller and Byers, 1991). In rats, social play, according to Siviy and Atrens (1992), accounts for between 2 and 3% of the total daily energy budget. In percentage, Thiels et al. (1990) has confirmed that among rats, 3% of the time each day is spent in play fighting. With reference to Experiment 2, in a general way, it was found that the individual differences were not very significant among the several behavioral measurements registered. In the case of the preference in the partners’ choice for playing, it was found that in just one of the litters was preference significant. However, it appears that this phenomenon is an isolated case, not representing a characteristic in the social interactions among hamsters. Although physical proximity is quite frequent among hamsters during the juvenile period, this does not necessarily mean that there are preferences. Perhaps what is important is the physical contact and the opportunity to play, the partner’s identity not being so important. This occurs among individuals of the same sex. It is known that, in the case of the hamster, the partner’s sex interferes in social interaction (Guerra et al., 1992). Dyads of the same sex exhibited a significantly high level of physical contact and engaged in less play fighting than did dyads of different sexes. It can be suggested, therefore, that some behaviors in the hamster may be modulated,

271

mainly, by environmental factors, such as the partner’s sex, rather than by endogenous factors (testosterone or androgen). Based on these data and on the results obtained in Experiment 2, it can be expected that striking individual differences and preferences exist in relation to play fighting when hamsters of different sexes are compared. Perhaps the results of Experiment 2 would have been different if the litters had been composed of both males and females. Through the analysis of the results of Experiments 1 and 2 jointly, it is suggested that, in the case of the hamster, a standard time to be spent in play fighting would exist and that this would be distributed proportionally among the number of available partners for the social interaction, when individuals of same sex are present. An explanation can be given, based on the way of life of the hamster, that a solitary species is being considered. In effect, it would not develop effective and durable interactions with its partners during the development period when the youths stay together. The stability of some behavioral measurements in the hamster seems to go beyond the manipulated variables in the present study. This is the case for the positions adopted by the hamsters during the playful interactions. There was no significant difference among the positions adopted by the hamsters during play fighting, registered in Experiment 2. In another study it was noted that, independent of the partner’s sex, in none of the dyads did the hamsters show dominance (Guerra et al., 1992). Nor did blindness produce any alteration in that result (Guerra et al., 1989). One of the aspects that characterizes the play fighting, and that differentiates it from agonistic interaction, is exactly the break of dominance–subordinance relationships (Burghardt, 1984). Besides the reversion of positions during the play fighting, other behaviors also seem to be less influenced by the size of the litter, by the hamster’s individual differences, or possibly by its sexual interest during the juvenile period. This is case for the subsequent situation. The results of the experiment have demonstrated clearly that after the playful interactions, the hamsters remained close to one another, indicating it to be an

272

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273

amicable interaction. This fact was also found with rats. After the end of the majority of the play fighting bouts, the partners stayed close (Hole, 1991). In relation to the general activity, the locomotion bout can be used as an indicator of the rhythm of the individuals’ activity. This behavioral measurement was used in these experiments with the purpose of detecting individual differences, which perhaps did not appear in other behaviors, directly related with the play fighting. However, in the great majority of cases there was no significant difference in the locomotion frequency among the individuals of the same litter or among litters, nor was there any correlation between locomotion and play fighting. Therefore, in the case of the hamster, it does not seem that there is a direct relationship between level of general activity (locomotion rate) and play fighting. This result was also found in another study (Guerra et al., 1992). In this case, hamster females in the company of other females were significantly more active (moved about more) than males in the company of other males. However, this did not mean that they played more significantly among themselves. In a general way, one of the implications of the results obtained has to do with the sociability of the hamster. Although the hamster is considered a solitary species during adult life, it is suggested that during the juvenile period, physical contact is very important, and is not discriminating during play fighting, when male individuals of the same age are involved. It remains to be discovered whether these results can also be extrapolated for females or mixed groups. In contrast, the comparative analysis between the hamster and other species suggests that among social species, such as the rat, preference may exist.

Acknowledgements The beginning of this research was supported by CAPES, through a post-graduate research scholarship at the doctorate level, and later by a grant from CNPq (522017/96-7) to Mauro L. Vieira.

References Burghardt, G.M., 1984. On the origins of play. In: Smith, P.K. (Ed.), Play in Animals and Man. Blackwell, Oxford, pp. 5 – 41. Caro, T.M., 1981. Predatory behaviour and social play in kittens. Behaviour 76, 1 – 24. Elwood, R.W., Broom, D.M., 1978. The influence of litter size and parental behaviour on the development of Mongolian gerbil pups. Anim. Behav. 26, 438 – 454. Fagen, R.M., 1981. Animal Play Behavior. Oxford University Press, Oxford, 684 pp. Guerra, R.F., Vieira, M.L., 1990. Some notes on mother-infant interactions and infant development in golden hamster (Mesocricetus auratus). Ci. Cult. 42 (12), 1115 – 1123. Guerra, R.F., Vieira, M.L., Gasparetto, S., Takase, E., 1989. Effects of blindness on play fighting in golden hamster infants. Physiol. Behav. 46, 775 – 777. Guerra, R.F., Vieira, M.L., Takase, E., Gasparetto, S., 1992. Sex differences in the play fighting activity in golden hamster infants. Physiol. Behav. 52, 1 – 5. Guyot, G.W., Benett, T.L., Cross, H.D., 1980. The effects of social isolation on the behavior of juvenile domestic cats. Dev. Psychobiol. 13, 317 – 329. Hole, G., 1991. Proximity measures of social play in the laboratory rat. Dev. Psychobiol. 24, 117 – 133. Jamieson, S.H., Armitage, K.B., 1978. Sex differences in the play behavior of yearling yellow-bellied marmots. Ethology 74, 237 – 253. Lerwill, C.J., 1974. Activity rhythms of golden hamsters (Mesocricetus auratus) and Mongolian gerbils (Meriones unguiculatus) by direct observation. J. Zool. Lond., 174, 520 – 523. Martin, P., 1984. The time and energy costs of play behaviour in the cat. Z. Tierpsychol. 64, 298 – 312. Meaney, M.J., Stewart, J., 1981. The descriptive study of social development in the rat (Rattus nor6egicus). Anim. Behav. 29, 34 – 45. Meaney, M.J., Stewart, J., Beatty, W.W., 1985. Sex differences in social play: the socialization of sex roles. Adv. Study Behav. 15, 1 – 56. Mendl, M., 1988. The effects of litter-size variation on the development of play behaviour in the domestic cats: litters of one and two. Anim. Behav. 36, 20 – 34. Miller, M.N., Byers, J.A., 1991. Energetic costs of locomotor play in pronghorn fawns. Anim. Behav. 41, 1007 – 1013. Pellis, S.M., Pellis, V.C., 1988. Play-fighting in the Syrian golden hamster Mesocricetus auratus waterhouse, and its relationship to serious fighting during post-weaning development. Dev. Psychobiol. 21, 323 – 337. Poole, T.B., Fish, J., 1976. An investigation of individual, age, and sexual differences in the play of Rattus nor6egicus (Mammalia: Rodentia). J. Zool. Lond. 179, 249 – 260. Siviy, S.M., Atrens, D.M., 1992. The energetic costs of roughand-tumble play in the juvenile rats. Dev. Psychobiol. 25, 137 – 148.

M.L. Vieira, E. Otta / Beha6ioural Processes 43 (1998) 265–273 Thiels, E., Alberts, J.R., Cramer, C.P., 1990. Weaning in rats: II. Pup behavior patterns. Dev. Psychobiol. 23, 495–510. Vieira, M.L., 1991. Comportamento de brincadeira em filhotes de hamsters dourados (Mesocricetus auratus). Unpublished

.

273

Master Dissertation, Universidade de Sa˜o Paulo, Sa˜o Paulo, 142 pp. West, M., 1974. Social play in the domestic cat. Am. Zool. 14, 427 – 436.