We measured two aspects of dispersal in the alpine Australian scincid lizard, Niveoscincus micolepidotus: (1) natal dispersal, i.e. shift in home range over the lizard's first year of life, and (2) breeding dispersal, i.e. shifts of home ranges between breeding attempts as adults. On average, displacements were surprisingly small. Female neonates dispersed about twice as far as did males in the same cohort (means of 12 m vs. 6 m). A female's natal dispersal distance was not correlated with her body size or our estimate of physiological performance (sprint speed). However, larger, faster-running male neonates dispersed further than did smaller, slower males. As was the case for neonates, adult females moved significantly further between breeding seasons than did adult males (14.2 m vs. 9.6 m). Because of a female's long gestation period (more than 1 year), two groups of females occur simultaneously in the population, non-ovulated (i.e. with yolking folicles) and pregnant females (i.e. approaching parturition). Females that were not yet ovulated showed a markedly stronger dispersal in response to high reproductive effort (i.e. clutch size in relation to body condition) than did pregnant females. In adult males, body size was negatively correlated with dispersal distance, suggesting that although males have overlapping territories, they exhibit an increasing level of site tenacity with age and/or size. Thus, selection for the relatively more pronounced site tenacity in adult males may have resulted in the more marked philopatric behaviour compared to females also as neonates. © 2003 The Linnean Society of London, Biological Journal of the Linnean Society, 2003, 79, 277–283.