Many species of Eucalyptus, one of the dominant genera in Australian forests and woodlands, contain high levels of tannins and other phenols and are also heavily damaged by grazing insects. These phenols do not appear to affect insect attack because a wide range of concentrations of condensed tannins and other phenols in leaves of 13 Eucalyptus sp. influenced neither feeding rates of Paropsis atomaria larvae, nor their nitrogen use efficiencies. We discuss reasons why tannins may not appreciably reduce the availability of nitrogen (N) to these insects. Performance was directly related to leaf N concentration, and growth rates, N gains, and N use efficiencies all increased as leaf N content increased, although absolute feeding rates remained constant. These relationships differ from those found in insects feeding on other plants, and we suggest that the low N contents common in Eucalyptus leaves may be responsble. We propose that the extensive damage observed in many eucalypts is in part related to the high feeding rates maintained by individual larvae.

The activity of five seed-harvesting and ten nonharvesting species of ants was examined over a 2-year period. The ant community appeared to be stable in the long term, but there was considerable variability in the numbers and species of ants active at any one time, both on a seasonal and diel basis. Only four species were active throughout the year, with activity reduced in winter. Two species were consistently nocturnal, nine were mainly diurnal, and the remaining four species switched from diurnal to more nocturnal activity with increasing temperatures.The changes in the diurnal pattern and intensity of foraging activity were shown to be controlled by several climatic and biotic variables. These were classed as either 'stimulatory-inhibitory' which determined whether activity would occur or not, or 'regulatory', which set the physical limits to activity and controlled foraging intensity. Factors such as forage availability and colony satiation were placed in the first category, and soil temperature, moisture stress and light intensity were placed in the second.There was no simple response to the combination of variables, and individual species responded differently to the various factors. The resulting temporal separation probably reduces competitive interactions between species.

The trophic structure of an ant community near Deniliquin, in south western New South Wales, was found to be complex. Individual species covered a range of trophic levels and, apart from a few specialized predators, most species showed a broad diet. A total of twenty-seven species were placed in six different categories, according to food usage patterns:-specialist predators (2 spp.), generalist predators (1 sp.), predatorscavengers (4 spp.), omnivore-nectar collectors (10 spp.), omnivore-seed collectors (5 spp.) and seed harvesters (5 spp.). Almost 50% of ant colonies were of the five seed-harvester species and their food usage was examined more closely. They showed marked seasonal changes in composition of diet, corresponding closely to periods of seed production by different food plants. There were also differences in seed selection between species, and these were related to the size, phenology and probably chemical properties of the seed. Foraging behaviour by the various species showed differing adaptations to efficient utilization of a variable food supply.