Subspecies richness (average number of subspecies per species = SSP/SP) has traditionally been explained by niche breadth, characteristics of the preferred habitat ("habitat hypothesis"), dispersal ability, range size ("range size hypothesis"), etc.However, in several invertebrate groups, there are striking differences between SSP/ SP across families, which cannot be explained using classic arguments, and we might assume some hidden factor in the background. For example, some families of terrestrial molluscs, that possess conspicuously more subspecies than others, have higher degree of morphological complexity (shell size, number of diagnosable shell characters), for which we propose the "artefact hypothesis." According to this, morphological complexity plays an important role in taxonomists' subspecies recognition. We used shell size, number of characters mentioned in species descriptions and area extent in six European land snail families, and habitat characteristics (rockdwelling vs. non-rock-dwelling) in one largest European snail families (Clausiliidae) to test our hypothesis. We found that all three factors (range size, morphological complexity and habitat type) correlated positively with the SSP/SP ratio. Although the generality of our findings has to be tested in other invertebrate phyla, we provide examples of striking differences between SSP/SP ratios in butterflies (Lepidoptera) and beetles (Coleoptera) as well. Our findings imply that researchers (a) should use the number of SSP more carefully in comprehensive studies, and (b) should maintain and use the rank subspecies in order to provide a stronger status for the species rank, and avoid taxonomic inflation. Furthermore, the subspecies rank might be more useful for "cryptic species" (clades that are indistinguishable morphologically) than species.
K E Y W O R D Smorphological complexity, Subspecies-to-species ratio, taxonomic bias