Torsion in Patella caerulea (Mollusca, Patellogastropoda): ontogenetic process, timing, and mechanisms

Abstract. Torsion is a process in gastropod ontogenesis where the visceral body portion rotates 180° relative to the head/foot region. We investigated this process in the limpet Patella caerulea by using light microscopy of living larvae, as well as scanning electron microscopy (SEM) of larvae fixed during the torsion process. The completion of the 180° twist takes considerably less time in larvae of Patella caerulea than previously described for other basal gastropod species. At a rearing temperature of 20–22°C, individuals complete ontogenetic torsion in ?2 h. Furthermore, the whole process is monophasic, i.e., carried out at a constant speed, without any evidence of distinct ‘fast” or ‘slow” phases. Both larval shell muscles—the main and the accessory larval retractor—are already fully contractile before the onset of torsion. During the torsion process both retractors perform cramp‐like contractions at ～30 s intervals, which are followed by hydraulic movements of the foot. However, retraction into the embryonic shell occurs only after torsion is completed. The formation of the larval operculum is entirely in‐dependent from ontogenetic torsion and starts before the onset of rotation, as does the mineralization of the embryonic shell. The reported variability regarding the timing (mono‐ versus biphasic; duration) of torsion in basal gastropod species precludes any attempt to interpret these data phylogenetically. The present findings indicate that the torsion process in Patella caerulea, and probably generally in basal gastropods, is primarily caused by contraction of the larval shell muscles in combination with hydraulic activities. In contrast, the adult shell musculature, which is independently formed after torsion is completed, does not contribute to ontogenetic torsion in any way. Thus, fossil data relying on muscle scars of adult shell muscles alone appear inappropriate to prove torted or untorted conditions in early Paleozoic univalved molluses. Therefore, we argue that paleontological studies dealing with gastropod phylogeny require data other than those based on fossilized attachment sites of adult shell muscles.