This project is a contribution towards the understanding of the mechanisms involved in the chemocommunication systems of large mammals. Data are included on the social behaviour and use of scent marking for the African lion, P. leo as well as two other Panthera species namely the Siberian tiger, P. tigris altaica, and the leopard, P. pare/us. The study was conducted in three Zoos or Safariparks in Denmark: K0benhavn Zoo, Givskud Safaripark and Knuthenborg Safaripark. Data were collected in three ways. The first part of the project comprised behavioural observations of the undisturbed social and marking patterns of the study groups. Particular attention was paid towards Spraymarking, Scrape/urination and normal Urination but Clawing, Chinrubbing and Defaecation were also recorded. Methods were developed which made it possible to collect samples of scent marks from the study animals. The scent samples collected were used in the second part of the field work, which involved an experimental investigation in which the animals were presented with scent marks from foreign individuals of varying sex, age or reproductive status. These experiments were conducted both in the "natural" setting of the outdoor enclosures as well as in the artificial surroundings of the night cages of the animals. In the third and final part of the study the scent mark samples collected were subjected to chemical analysis using the "Headspace" procedure on a combined Gas Chromatograph and Mass Spectrometre. The activity and social interactions of the animals, although under influence of the captive environment, were generally close to what one would expect to see for wild animals. Each animal showed a distinct activity pattern and social repertoire, but more general differences between the sex/age groups were also found. Males tended to stand more than females or cubs and show higher levels of Investigation, whereas females generally moved more than males. Cubs played more than the two adult groups. These trends were seen in all three species. Each animal had a distinct marking repertoire, but generally the male patterns were dominated by Sprayrmarking followed by Scrape/urination, whereas the female patterns were much more variable. Male lions had higher rates of Spraymarking and significantly higher rates of Scrape/urination than females. No significant difference was found between Sprayrnarking rates of male and female tigers, but both had significantly higher rates than castrated tiger males. The leopard male had higher rates of the two marking types than the female. Male lions showed a bias towards Spraymarking over Scrape/urination in territorial contexts, but no such bias in social contexts. The females showed no bias in markings in either context. Investigation of experimental marks were dominated by Scenting and Flehmen and very little Licking and Overmarking was seen. In the chemical analysis 58 compounds were identified in lion urine samples. Lion male samples overlapped significantly more in compound composition with other male samples than they did with female samples. The majority of the lions overlapped more within their own samples than they did with samples from other lions, and for seven out of 16 lions this difference was significant. Male lions had a significantly higher absolute concentration of 2-butanone than females, and females had a significantly higher relative concentration of acetone than did males. 57 compounds were identified in tiger urine samples. All tigers overlapped more within their own samples than between theirs and other samples from same sex individuals, but only one significantly so. Castrated tiger males had higher concentration of 5-hydroxy-4-methyl-6-hepten-3-one than the intact male. In leopard urine samples 18 compounds were identified. Possible candidates for species and sex identifying compounds were put forward for lions and tigers. On average two lion samples overlapped significantly more in compound composition thall did a lion sample and a tiger sample. This was also the case for the overlap between two tiger samples compared to a tiger and a lion sample. Leopard samples overlapped equally with lion and tiger samples. The results are discussed in relation to what is known about the same species living in the wild. Furthermore the use of chemical signals are discussed in relation to their chemical and physical properties, and their function in a territorial context is commented upon.