In vertebrates, the steroidogenesis enzyme 5α-reductase converts testosterone to the more potent androgen 5α-dihydrotestosterone. Homologues of 5α-reductase genes have been identified in molluscs. However, recent findings suggest that vertebrate-type steroid androgens are not utilised in molluscan reproductive development. Genomic searches have revealed that molluscs do not possess many of the steroidogenic enzymes required to make testosterone, nor a nuclear androgen receptor. Consequently, the role of 5α-reductase in molluscs presents a mystery. Here, developmental exposures of Biomphalaria glabrata to selective pharmaceutical 5α-reductase inhibitors elicited a strong, highly reproducible phenotypic response characterised by the development of elongated "banana-shaped" shell morphology. In comparison to untreated snails, the shells are open-coiled and the whorls are unattached. Dutasteride (5α-reductase inhibitor) is approximately 10-times more potent at provoking the banana-shaped shell phenotype than finasteride, paralleling the pharmaceuticals' efficacy in humans. Other enzyme inhibitors with different modes of action were tested to investigate the specificity of the phenotype. However, only the pharmaceutical 5α-reductase inhibitors provoked the response. Dutasteride elicited the same phenotype in a second gastropod, Physella acuta. In the absence of evidence for de novo androgen steroidogenesis in molluscs, these findings suggest that novel substrates for 5α-reductase exist in gastropods, lending support to the contention that molluscan endocrinology differs from the well-characterised vertebrate endocrine system.