We observed the behaviour of a colony of Mongolian gerbils (Meriones unguiculatlts) for 18 weeks, in an enclosure provided with facilities for burrowing, nest-building, foraging, wheel-running and other behaviour. Records were taken of both social and individual activities, and in particular of activities occurring during encounters between pairs of individuals. I. The colony originally consisted of eight animals, but severe fighting rapidly reduced the numbers to two males and two females. The colony then remained stable until week 18, when an adult male attacked the offspring of one of the females. Fighting was never observed between animals of opposite sex. The fighting was probably due to overcrowding, and suggests that in natural conditions gerbils do not nest in groups of more than a few adults. 2. Routine encounters between individuals revealed a number of sex differences, but few conspicuous individual differences. Agonistic behaviour was rare, and the observations gave little evidence for the existence of a dominance order. 3. When a female was in oestrus one male consistently attacked and chased the other, and consequently gained almost exclusive access to the female. The sequence of mating responses is described, and it is suggested that genital sniffing by the male enhances the receptivity of the female. Ventral marking by the female and drumming by the male occurred at unusually high levels during mating. 4. Each female produced two litters, but only the first two survived. The litters were dropped and the young housed in a nest-box outside the normal colony burrow, the entrance to which was kept blocked. Both litters were carried to another nest-box in the second week after parturition, and carried back again prior to the birth of the second litters, but the reasons for this behaviour were unclear. The males assisted in nest-building activities, but never entered the maternal nest-box. We suggest that the failure of the females to care for their second litters was due to overcrowding. 5. Ventral marking occurred at a high rate, and in conjunction with various activities. Marks were deposited over the entire enclosure and on a variety of substrates, but were concentrated at four main sites on mounds of earth near the burrow. There were no sex differences in total number of marks, but the males tended to distribute their marks more widely. Individual differences in amount of marking were correlated with differences in gland size, and with the outcome of aggressive encounters during mating and the occasional fighting. There was a strong tendency for fresh marks to elicit counter-marking by the same or another animal, but novel clean objects were also marked within a short time of being placed in the colony. There was no evidence that marked objects or areas were avoided by other animals or defended by the marker, and we suggest that marking is not primarily territorial in function. 6. Mutual marking, mutual grooming and sandbathing all occurred at the four most heavily marked sites near the burrow. We suggest that these activities, and the tendency for animals to counter-mark at specific sites, might serve to distribute a colony odour consisting of the odors of the individuals, and that this might reduce intra-colony aggression. We found no evidence that mutual marking or grooming reflected a dominance order. Drumming of the substrate occurred as an alarm response, but we saw no evidence of its use as an alarm signal. 7. The most frequent individual activity was digging. The burrow, which was inhabited by all the adults, consisted of from two to eight tunnels leading in towards a central nest of shredded paper, straw and leaves. The tunnels were repeatedly cleared and enlarged, and were often destroyed by digging at other locations. Freshly dug earth was often marked by the digger, and this seemed to elicit marking and digging by other animals. 8. Wheel running occurred only at a low rate, and did not develop a stereotyped character, but was performed by all animals. Contrary to expectations based on laboratory studies the majority of running occurred during the light phase of the cycle, and this was when the animals were most often active in the general sense. We suggest that gerbils are probably active day and night in natural conditions, and that the enhanced nocturnality shown in laboratory studies may be due to the lack of a dark burrow. 9. Other stereotyped activities such as shredding paper, gnawing and scrabbling were rare. 10. The results are summarized and discussed in relation to theories of dominance, ventral marking and sociability, and attention is drawn to the need for more field studies of behaviour in this species. We observed the behaviour of a colony of Mongolian gerbils (Meriones unguiculatlts) for 18 weeks, in an enclosure provided with facilities for burrowing, nest-building, foraging, wheel-running and other behaviour. Records were taken of both social and individual activities, and in particular of activities occurring during encounters between pairs of individuals. I. The colony originally consisted of eight animals, but severe fighting rapidly reduced the numbers to two males and two females. The colony then remained stable until week 18, when an adult male attacked the offspring of one of the females. Fighting was never observed between animals of opposite sex. The fighting was probably due to overcrowding, and suggests that in natural conditions gerbils do not nest in groups of more than a few adults. 2. Routine encounters between individuals revealed a number of sex differences, but few conspicuous individual differences. Agonistic behaviour was rare, and the observations gave little evidence for the existence of a dominance order. 3. When a female was in oestrus one male consistently attacked and chased the other, and consequently gained almost exclusive access to the female. The sequence of mating responses is described, and it is suggested...