The spatial distribution of displaying Malaysian peacock pheasants was documented at a site in Peninsular Malaysia. Calling males were clustered and, therefore, were not evenly distributed throughout the available habitat. The location of these clusters differed between calling periods. An objective method of assessing micro-habitat structure at the forest-floor level was devised to investigate habitat use by displaying males. Habitat structure was measured at 197 points throughout the study site, both in areas used by the pheasant and in those areas not used by the pheasant. A DECORANA ordination revealed that two gradients explained much of the variation in micro-habitat structure. Male calling clusters were located in areas that were away from the river and more stable (e.g. fewer treefalls). Within these clusters, display scrapes were sited in areas that were more stable and were more influenced by the rivers than elsewhere within the area used by the calling males. It is suggested that the environmental gradients described by the ordination are more likely to determine the position of the calling cluster than individual habitat structure variables. Within calling clusters, however, it seems probable that the presence or absence of ground-level vegetation directly influences the siting of display scrapes.