THE CONSERVATION OF ARBOREAL MARSUPIALS IN THE MONTANE ASH FORESTS OFTHE CENTRAL HIGHLANDS OF VICTORIA, SOUTH-EASTERN AUSTRALIA .5. PATTERNS OF USE AND THE MICROHABITAT REQUIREMENTS OF THE MOUNTAIN BRUSHTAIL POSSUM TRICHOSURUS-CANINUS OGILBY IN RETAINED LINEAR HABITATS (WILDLIFE CORRIDORS)

Hairtubing, a technique for detecting mammals, was used to census arbo real marsupials at 70 sites in the timber production montane ash fores ts of the central highlands of Victoria, south-eastern Australia. Of t hese sites, 41 were located in linear strips or wildlife corridors and the remaining 29 were in stands of contiguous montane ash forest. Fou r species of arboreal marsupials were detected in the hairtubing surve ys but only one, the mountain brushtail possum Trichosurus caninus Ogi lby, was recorded in sufficient numbers to allow subsequent analysis. Data derived for T. caninus from hairtubing studies were compared with those from earlier surveys where stag-watching was used to determine the presence and abundance of the species in retained linear strips. T . caninus was found to be absent from many retained strips during stag watching surveys but was subsequently detected by hairtubing. Movement by animals into retained linear strips from surrounding areas of cont iguous forest, probably whilst foraging, was the most logical explanat ion for this result. Statistical analyses revealed that T. caninus was most likely to move into strips that supported few trees with hollows and thus areas of poor habitat suitability where there were few resid ent animals. Such forays may represent an adoption of a retained area as part of an animal's home range. These findings suggest possible fun ctional differences in the use of retained linear strips by T. caninus where some sites provide habitat for resident animals and others, of poorer habitat quality, are used by individuals that do not live in th e strip. Within the survey sites, T. caninus was most likely to be det ected by hairtubing in those plots where there were numerous tree fern s, Dicksonia antarctica and Cyathea australis, and where the trees Aca cia dealbata and A. melanoxylon were present. Thus, some microhabitat features may influence the patterns of use by T. caninus within areas of retained forest. The implications of our results for wildlife conse rvation in timber production montane ash forests are discussed.