The biology and ecology of the endocoprid dung beetles of the Aethiopian zoogeographic region has been studied in the field and laboratory. Work has centred mainly on Oniticellus planatus and O. formoaus although infomation has also been supplied on O. egregius, O. pictus and Tragiscus dimidiatus. The taxonomy of Onticellus (s.str.) and Tragiscus has been discussed with the recommendation that O. planattus and O. pseudaplanatus be synonymised. It has been suggested that endocoprid distribution is chiefly controlled by temperature, rainfall and the density of suitable large dung masses. On the basis of distribution it has been suggested that the taxonomic status of the subspecies of O. pictus should be reviewed. Endocoprids (genus Oniticellus (s.str.) are found throughout the warmer regions of Africa and Asia. Tragiscus is found only in Africa. Methods are described for monitoring endocoprids in the field, for distinguishing freshly emerged from older specimens, and for breeding endocoprids under laboratory conditions. It has been shown that endocoprids breed within or just beneath large dung masses unlike most other members of the Scarabaeinae which actively bury dung and construct the their broods at depth in the soil. It has been shown that there is a restriction of breeding sites available to endocoprids due to the removal of dung by the Scarabaeinae beetles and the limited space available within dung pads. This has resulted in low endocoprid population numbers, a disadvantage, which has been countered by highly specialised but opportunistic breeding habits. The requirements of dung plasticity have limited the time available for breeding activity by most endocoprids (excluding O. egresius) so that both the reproductive system and the reproductive behaviour (cf. O. planatus) are geared to rapid egg and brood production. Large numbers of follicles are able to be produced over a short period. In the probable absence of feeding, this has possibly been allowed by the involvement of a well developed fat body. The broods are tended by the parent female during larval development and the follicles are gradually resorbed, probably to prevent starvation. Ovarian recovery is rapid once feeding recommences. Behavioural mechanisms have developed which boost the number of broods constructed and the rapidity with which they are produced. The duration of endocoprid colonisation and the timing of arrival at dung is shown to be related to the rate of pad desiccation which is chiefly controlled by temperature and degree of dung removal by other dung beetles. Precipitation may also play a part. Predation by vertebrates and other insects has been noted and a number of mechanisms are described which are probably protective. Two insect parasitoids of O. formosus larvae have been recorded. The potential use of endocoprids in the Australian dung and fly control project has been discussed and it is recommended that their introduction be considered of low priority. It is suggested that the endocoprids evolved from Euoniticellus type ancestors and that their behaviour complex nidification developed in response to harsh environmental factors, chiefly the danger of desiccation and competition for dung with other dung beetles. From a consideration of brood and brood chamber construction it is suggested that O. egresius is a relatively unspecialised endocoprid which may retain dry season aestivation and has not evolved much further than its probable Euoniticellus type ancestor. The broods are coated in clay and abandoned soon after construction. Greater specialisation is shown by O. planatus and O. formosus which exhibit brooding behaviour, increasing longevity, more specialised brood protection, increased potential fecundity and continuous activity throughout the year. O. formosus shows greater specialisation in brood production than O. planatus. Greatest specialisation is found in T. dimidiatus which exhibits a precise brooding period and low egg production. Brood construction is of a similar degree of specialisation to that of O. formosus. Activity is also continuous throughout the year. The comparative biology of the five southern African endocoprids is summarised in Table 35. Summary, p. 129-131.

1. Twenty four species of the genus Sisyphus known in southern Africa are listed with their synonyms and their taxonomy is discussed. 2. A key to the adults of 23 of these species is given. 3. The known distribution of 23 species and seasonal occurrence of 19 species are illustrated and briefly discussed. 4. The general biology of the genus Sisyphus is discussed and the major behavioural and biological differences between the two subgenera Sisyphus and Neosisyphus are given. 5. The laboratory and field investigation into variations in behaviour and biology in eight species occurring in the Mkuzi Game Reserve, with regards to interspecific competition, is presented. 6. The known possible natural enemies of Sisyphus are mentioned and the extent of competition from other genera, based on field and laboratory observations, is outlined. 7. Summary tables of all data with a final tabulated evaluation of the suitability of species for introduction into Australia is given. 8. Nine species have been highly recommended for Australia. Six more species could be considered but most of these still need further investigation. The remaining nine species were found to be unsuitable for transfer to Australia.