Search

The prosobranch snail, Hydrobia ulvae, lays eggs on the shells of neighbouring individuals where they are held together and protected by a covering of sand grains. The eggs develop within this egg mass, hatching as veliger larvae for which there are several descriptions (Henking, 1894; Thorson, 1946; Pilkington, 1971; Fish & Fish, 1974). It is clear that there is considerable variation in the size, pigmentation and swimming behaviour of veligers from different localities. The material used in the present study was from a population of snails in the Dovey estuary, Wales, which have free-swimming veliger larvae similar to those described by Thorson (1946). These have a single whorl to the shell; a shell length at hatching of 140–170 µm; a semi-lunate band of purplish black pigment on each velar lobe; a narrow band of pigment between the eyes and black pigment associated with the hindermost part of the intestine. In contrast, those from Ply-mouth hatch with one and a half shell whorls and a colourless velum (Lebour, 1937; Pilkington, 1971); a shell length of 340–390 µm and show complete suppression of the pelagic phase, the velar cilia being short and incapable of lifting the larva off the bottom (Pilkington, 1971). It has been suggested by Pilkington (1971) that the differences in size and swimming behaviour might be the result of the Plymouth larvae hatching at a later stage. The present account is the result of investigations into the effects of temperature and salinity on the survival of embryos to hatching and the duration of embryonic development.

Variation in shape of the shell of Littorina saxatilis and Littorina arcana has been investigated\ud using shells from four shores where the species co-occur and seven where the\ud former only is present. Samples were taken from shores along the south-west and south\ud coasts of Britain, between Cardigan Bay and Kent. Measurements of eight shell variables\ud and of the operculum area were analysed using canonical variate analysis. The results of\ud the analysis suggest that the shells can be separated into three groups, namely L. arcana, L.\ud saxatilis where it is sympatric with L, arcana, and L. saxatilis where it is allopatric from L.\ud arcana. The separation of the two groups of L. saxatilis is based primarily on their relative\ud globosity, whereas the separation of L. saxatilis from L. arcana is based primarily on the\ud relative size of the operculum and width of shell whorl one (the first above the body\ud whorl). Sexual dimorphism has been considered and shown not to contribute significantly\ud to the between-shore pattern of variation observed. The relationships between\ud important variables were further investigated using ANCOVA, and the relationship of\ud foot area to the shell variables was also analysed at four of the shores. It is concluded that\ud there is evidence of character displacement in L. saxatilis, shown by an increase in\ud globosity in the presence of L. arcana, and that this may partly explain the continuing\ud uncertainty over the taxonomic and ecological identity of members of the rough periwinkle\ud group.

Nucella lapillus and Littorina littorea afford examples of turbinately coiled shells with no space between the columella and the inner face of the whorls. Three constants are sufficient to determine the form of such shells: λ, the ratio between diameters of successive whorls; β, the semi-apical angle and ρ, the ratio of aperture length and breadth in the apertural plane passing through the axis. The variation of these three constants with size was examined and shown to be relatively small and insignificant in Nucella, but in Littorina it was significant.For Nucella and Littorina these shell characteristics were used to relate for the nth whorl, the length along the shell spiral (ln) to the shell height (Hn), the latter being the measure usually employed in growth studies. This relation was derived mathematically and confirmed empirically aswhere α is the angle of the logarithmic spiral given by α = tan−1 2π sin β/loge λ. The increased resolution of the micro-growth band increments, which are measured along the shell spiral was directly related to the ln:Hn ratios.Since the formula for the ln/Hn ratio (= shell conversion factor) includes λ, β and ρ, the three constants necessary to determine shell form in turbinately coiled shells of the type described, the shell conversion factor may be a better index of shell shape than indices based on a single constant. In species where the shell grows isometrically this ratio will remain constant, whereas changes in shell shape with size will influence this ratio. Such changes in shell shape during growth can be quantified in terms of the shell constants and when included in calculating the shell conversion factor, will show the relative magnitude of shell shape changes brought about by growth. The shell conversion factor may also be used to study the effects of ecological variations on the geometry of the shell.The conical shell of the limpet, Patella vulgata, is geometrically a much simpler system in which direct shell measurements were used to relate length along the direction of maximum growth to the antero-posterior diameter of aperture.

The development of the hydroid of Amphinema dinema (Peron & Lesueur) has been followed in the laboratory until the production of young medusae and the hydroid identified as Perigonimus serpens Allman.The hydroid of Amphinema rugosum (Mayer) is described for the first time. It is very similar to P. serpens but more robust.The hydroid has been reared from Rathkea octopunctata (Sars); it is very minute and the hydranths possess a single whorl of long filiform tentacles/A small Cuspidella hydroid was reared from Mitrocomella brownei (Kramp).Certain points in the synonymy of these species are discussed. An account of the methods of rearing used is also given.