Although forested habitats in eastern Australia have attracted significant conservation-related research, this work has focused strongly on endothermic vertebrates. Threatened reptile taxa have received less attention, but information, on their habitat requirements should be central to management planning. The arboreal elapid snake Hoplocephalus stephensii is largely restricted to remnant forests in eastern Australia, and is listed as a threatened species under wildlife legislation. We collated 84 records of the current New South Wales geographical distribution of H. stephensii, and compared attributes of these locations to those of randomly chosen points within the same forests, in adjacent forests, in timber plantations and on freehold land across the geographical range of the taxon. Data on climatic and topographic characteristics of these sites were obtained from Geographic Information Systems databases and entered into a principal components analysis. Unsurprisingly, locations where snakes were recorded differed from the random sites in several respects (e.g. rainfall, elevation, seasonality of precipitation). Within a given forest, H. stephensii was generally found in areas similar to randomly chosen points. Comparison of vegetation communities used with those available within forests provided no evidence for active habitat selection. Comparisons of Geographic Information Systems-derived data for snake-collection localities along roads versus those within the forest revealed significant biases, and we warn that such methodological errors could generate spurious conclusions about non-random habitat use by threatened species. In combination with previous data from radio-tracking, we conclude that although H. stephensii is highly specialized in its arboreality and dependence upon hollow trees, its broad tolerance with respect to other factors (climatic conditions, vegetation communities, food types, etc:) allows populations to persist so long as large areas of forest with high numbers of hollow-bearing trees are available. These requirements are similar to those of many other components of the Australian forest fauna. Thus, the findings of this study support the idea that the same kinds of management programmes can effectively conserve a wide range of taxa, if such programmes protect critical habitat components at suitable spatial scales.