Aphodius tasmaniae Hope is an intermittent pest of pastures in Canterbury. The defoliating biology of the larvae was investigated using remote sensing and controlled temperature methods in order to understand the mechanisms of damage. The larval development threshold was found to be c.5??C, and this showed that winter is unlikely to be an important growth period in Canterbury. The larvae were polyphagous and developed on all the common Canterbury pasture plants. Simulated pasture studies showed that annual turf composed of a mixture of hairgrass (Vulpia bromoides (L.) S.F. Gray), suckling clover (Trifolium dubium Sibth.) and storksbill (Erodium cicutarium (L.) l'Herit) was more susceptible to damage than perennial ryegrass (Lolium perenne L.). Tufted ryegrass plants were found to be less susceptible to damage than more open, less densely tillered, spreading plants. In perennial ryegrass and subterranean clover (Trifdlium subterraneum L.) experiments, thin stems ( < 1.0 mm) were found to be more susceptible to cutting by larvae than thicker stems. A method for studying the effects of larvae on plant yield under field conditions was developed, and using this technique, yield reductions of 20 - 29% in subterranean clover were recorded by densities of 350 - 500 larvae m????? for 87 days from mid-June to early September. Nocturnal defoliating activity was monitored in the field, using an automatic camera which took flash illuminated exposures at 30 min intervals. This device was used to record both emergences and defoliation events. Regarding emergences, the camera was used as a sampling tool. Defoliating activity declined to insignificant levels during frosts and rainfall, and was also significantly reduced at low soil moistures. Defoliating activity occurred under a much wider range of conditions than dispersive crawling, which was highly correlated with rainfall. This showed that a simple index of activity based on pitfall trap data would be unreliable as a predictor of defoliating intensity. Complementary laboratory studies confirmed the results of the field investigation on activity. They showed that defoliation would be insignificant below 2??C, and that foraging intensity was related to temperature. The field data showed that defoliating activity exhibited intermittent, pulsed characteristics, and hence that demand on pastures was not constant. This demonstrated the need for reappraisal of feeding models which assume that defoliation intensity is related mainly to stage of development.