Deep-sea hydrothermal vents display extreme and highly variable environmental conditions that are expected to be among the most important factors structuring associated benthic populations and communities. We tested this assumption,focusing Oil the distribution of gastropods, as well as on the demographic population structure and reproductive biology of one dominant gastropod species ill zones characterized by alvinellid polychaetes and vcstimentiferan tubeworms. A total Of 14 biological samples from both types of habitats were collected at three sites oil the East Pacific Rise 13 degrees N vent field in May 2002. At all vents except one, the physico-chemical environment was described in two steps: (1) pH, total sulphide and reduced iron concentrations have been measured in situ in Alvinella habitats and correlations to temperature were assessed at the scale of each sampled vent, and (2) assuming the consistency of these relationships within a single edifice, ranges of physicochemical factors were estimated for each biological sample from the corresponding fine scale temperature measurements. A total Of 11 gastropod species were identified from all samples and 2 main faunal assemblages were distinguished: one dominated by Lepetodrilus elevatus in the alvinellid zone as well as in the vestimentiferan zone, and one dominated by the peltospirids Nodopelta heminoda, N. subnoda and Peltospira operculata confined to the alvinellid zone. Peltospirid gastropods were dominant over lepetodrilid gastropods in the more acidic, sulphide-richer, and hotter environments. Although this pattern could be related to specific physiological tolerances to temperature and sulphide toxicity, the weak correlation between community structure and physico-chemical variables suggests that additional factors are also involved. Particularly, the lost, species richness and the overwhelming dominance of L. elevatus in one faunal assemblage suggest that this species may outcompete peltospirids and greatly affect community structure. This hypothesis is supported by large differences in the demographic structure and reproductive biology of L. elevatus between the 2 faunal assemblages.