Recolonization by ants of rehabilitated mineral sand mines on North Stradbroke Island, Queensland, with particular reference to seed removal

The ant fauna of 12 sand-mined plots representing a range of rehabilitation ages and three undisturbed vegetation controls was surveyed during 1982. Physical and botanical parameters were also measured in each plot.

Sixty-four ant species were collected from the 15 study plots, of which 44 had colonized one or more of the mined plots. Ant recolonization proceeded rapidly in plots up to 6 years old and may have been influenced by the passage of time, plant cover, density and diversity variables, the amount of litter and by the paucity of logs. In terms of ant species composition, the mined plots were most different from the undisturbed areas. The older plots exhibited a lower ant species richness and this is believed to have resulted from interspecific competition with the tramp ant, Pheidole megacephala. Ant succession proceeded in a slower fashion in the plots dominated by P megacephala.

The influence of ants on seeds applied during broadcast seeding of rehabilitated areas was also investigated. Seed removal by ants was greatest for the arillate seeds of Acacia concurrens but moderate quantities of Allocasuarina spp., Eucalyptus spp. Xanthorrhoea sp. and Banksia spp. were taken also.

Seed removal by ants was low in the areas which had recently had topsoil applied except where ants foraged from adjacent rehabilitation areas; here they exerted their influence up to 50 m across the fresh topsoil. Seed removal rates in topsoil adjacent to forest were low. Removal rates in a revegetated area 2.5 years old approached those in forest. This indicates that the previous ant-seed relationship had been partially restored by this time, although the relative contribution of seed harvesters and elaiosome collectors still needs to be assessed.