The African mole-rats (Bathyergidae) are a model taxon on which to test evolutionary theories of the origin
and maintenance of mammalian social behaviour, since species within the family range in social complexity from solitary
to eusocial. The aridity food-distribution hypothesis suggests that sociality within the Bathyergidae has evolved as
a consequence of habitat aridity, and the subsequent pattern of geophyte abundance and distribution. In more arid environments,
because of the patchiness of this food resource, and the fact that burrowing is only possible during brief
periods where rainfall has softened the soil, the risk of unsuccessful foraging is high. This risk may decrease when
large numbers of animals are involved in cooperative searching. We examined 25 burrows of seven species within the
Bathyergidae and used a simple box-counting method to determine their fractal dimension, which we propose as a
measure of burrow architecture that allows explicit testing of differences in African mole-rat burrows within the
theoretical framework of the aridity food-distribution hypothesis. We investigate the effect of habitat type and colony
size on this parameter, and conclude that habitat type is reflected in burrow architecture in the common mole-rat,
Cryptomys hottentotus hottentotus. We also found evidence that the burrows of species with larger maximum group size
explore the surrounding area more thoroughly. The results support the contention that larger group size offers adaptive
benefits in terms of how thoroughly the surrounding area is explored, and is not simply a consequence of the increased
costs of dispersal and colonization in arid areas.