How dependent are honey possums, Tarsipes rostratus, on their key winter food-plants?

The study aimed to examine the dependence of the honey possum, Tarsipes rostratus on the winter-flowering plants Banksia baueri, Dryandra plumosa and D. cuneata. Sampling of honey possum populations was carried out in six paired Grids in the Fitzgerald River National Park. The capture rate and various other population parameters was measured before and after the removal of flowers of these species in the Treatment Grids and then compared with populations in the Control Grids. Throughout the study, the flowering pattern of plants visited by honey possums was examined. Different plant species in the Proteaceae and Myrtaceae families had highly seasonal flowering patterns and there were several species in flower in each sampling month, suggesting that a continuous nectar supply was available to Tarsipes throughout the year. A peak in flowering occurred during the winter and spring months. Banksia baueri flowered most prolifically in July and the flowering intensity of D. cuneata ' increased through the winter to reach a peak in the early spring in most areas. The flowering of D. plumosa showed much less seasonal variation and flowered continuously for the duration of the study. During the winter, the population density of Tarsipes was limited by the supply of nectar and pollen and, during the manipulation, numbers were lower than expected and did not rise to their usual winter peak. The removal of B. baueri flowers had the greatest effect whereas removal of D. plumosa had the least effect. These changes in population size occurred as a result of the movement of individuals, particularly males, out of the cut Grids to areas with a greater abundance of nectar and pollen. Movement between Grids was negligible suggesting that honey possums that moved out of the Treatment Grids only moved short distances to the nearest flowering plants. Flower removal did not appear to influence dramatically the structure of honey possum populations. During the manipulation, the proportion of females in the Treatment populations increased as a result of the movement of males out of the area. However, the populations that remained in these Grids did not differ greatly in structure from those in the Control Grids. There was no change in the number of females with pouch young and, hence, the age structure of the populations remained the same. There was also no difference in the proportion of recaptures in each Grid. Flower removal also appeared not to have a deleterious effect upon the body condition of individuals. Tarsipes visited flowers from a wide range of species in the Proteaceae and Myrtaceae families and their diet was highly seasonal. When the food supply was reduced in the winter, honey possums did not feed on flowers from any new species of plant. Rather, they appeared to feed more intensively on species that were already a component of their diet. This applied particularly to Calothamnus gracilis and, to a lesser extent, D. nivea, B. gardneri var. hiemlis and B. coccinea. Finally, no significant relationship was found between the distribution of captures within each Grid and the distribution of B. baueri, D. plumosa and D. cuneata plants, suggesting that honey possums forage at random for nectar within their habitat