Differences in the distribution of close relatives represent the outcome of factors acting during or after their establishment. In apomictic complexes, sexual diploids and their derivative apomictic polyploids often occur in a distribution pattern known as geographical parthenogenesis. This pattern is characterized by sexuals having a more southern and smaller range than their related apomicts, with apomicts often occurring primarily or exclusively in previously glaciated areas. The varied forces that could produce this pattern potentially include more rapid or more successful post-glacial colonization by apomicts due to advantages associated with uniparental reproduction, niche divergence that excludes apomicts or sexuals from the other's range, and reproductive interference that prevents sexuals from invading sites occupied by apomicts. We have been working to disentangle these forces in Townsendia hookeri, a perennial plant system from the Rocky Mountains' east slope that displays a classic pattern of geographical parthenogenesis. Here we present the results of a 5-year multiple transplant garden experiment that included a total of eight experimental gardens spanning the sexual and apomictic ranges. We transplanted greenhouse-grown seedlings from 12 source populations and measured survival, growth and reproduction annually. We used Aster models to summarize fitness over the five years in the transplant gardens. We found that the fitness of apomicts transplanted into the sexual range was lower than in their own range, but sexuals did show consistent declines in fitness when planted in the apomictic range. We also conducted a separate seedling establishment trial at each garden site.  Analysis of seedling establishment trials using generalized linear mixed models revealed that seedling establishment differed among garden regions, but not between sexuals and apomicts. In the context of range limit theory, our results suggest that dispersal limitation may limit the northward expansion of sexuals, while niche limits may prevent apomicts from extending into the sexual range to the south. Integrating the transplant garden results with additional data on the ecology, genetic structure, and reproductive dynamics of the system gathered to date suggests that apomicts likely originated near the northern range limits of sexuals, with favoured clones spreading northward into previously glaciated areas. Sexuals that disperse into the apomictic range may be precluded from establishing due to asymmetric reproductive interference via the pollen of apomicts, while the southern expansion of apomicts may be limited by biotic interactions.