Species proposed on the basis of one or few individuals present a difficulty that transcends all modes of inference and all systematic data sources, including morphological and molecular data. This is particularly true for the liverwort family Lejeuneaceae, as these plants are inherently variable, differences between groups are frequently quantitative, and few specimens or limited material preclude destructive investigations, including molecular approaches. Here we use an extreme case of a single specimen to demonstrate how an integrated approach to character assessment and transparent decision processes can be used to provide sufficient evidence to support the hypothesis of species differences. To make progress on the single-specimen problem, an assessment of homology is essential to discriminate intra from interspecific variation in characters. Furthermore, an assessment of homology can substantiate differences between species of leafy liverworts in situations where the weight of character differences alone cannot because it is grounded in a phylogenetic framework. When homologous character systems and character states are equated with difference, the absence (or at least delayed development) of a stipe in Nephelolejeunea carcharias provides sufficient evidence for us to reject the hypothesis that this species is the same as N. talinayi because of what we know about the expression of perianth stipes in related species. As an inferential tool, homology exhibits considerable utility to liverwort taxonomists because it furnishes us with reasonable grounds on which to test hypotheses of relationship even in extremely observation-limited situations such as the one treated here.