There are strong links between habitat fragmentation, population size and genetic structure. However, to fully understand the long-term effects of fragmentation on population viability, it is necessary to first understand the relationship between life history traits and genetic characteristics in un-fragmented habitats. This is best done through comparing patterns of genetic diversity in sympatric species, since relative data may be maximally informative. We compared genetic structure and diversity among three sympatric small mammals - of which two were marsupial species - over a small spatial scale (<4 km) using microsatellite genotypic data from individuals sampled from four grids spaced in a linear fashion. Expected heterozygosity was high for all three species (average H, range: 0.781-0.886), but the smallest species had significantly higher genetic diversity (both heterozygosity and allelic diversity) than the two larger species, possibly due to it having; 1) large effective population size and little overlap in generations, and/or 2) high mutation rates in large alleles. Genotypic isolation by distance (measured as relatedness) was detected only at the smallest distance of 750 in, for any species. In the two marsupial species I F(st), relatedness and percentage of individuals assigned to site of capture were significantly lower in males than in females, corroborating previous reports of male-biased dispersal. However, sex-biased dispersal was not detected for the third species (the native bush rat Rattus fuscipes), and presence of significant heterozygous deficits suggested this resulted in inbreeding within kin-structured demes. We speculate that habitat fragmentation will have differing effects on population dynamics, social organisation and mating systems for the two marsupial species compared to the native murid rodent, due to their differing population structure and divergent inbreeding avoidance mechanisms.