One hundred and thirty-two shells of Athleta (Volutospina) spinosa (L.), have been examined, the majority being well enough preserved to permit of detailed measurements being made. The variation in the breadth/height ratio was found to indicate an increase with the age of the individual, approximately according to the equation B = 0·4444h + 0·00256h + 0·000027h. Spire height expressed as a percentage of the height is greatest in young shells, those less than one centimetre in height having values of thirty-three and thirty-four per cent, whereas fully grown individuals have a mode of twenty-seven per cent.

The variations in ontogeny are not considerable, in only one case have the differences from the normal type been recognized by earlier writers as worthy of specific distinction. It is probable that Athleta (Volutospina) trisulcata (L.) is a variety of Athleta (Volutospina) spinosa.

The correlation between the various ontogenetic factors is small, for example, that for the incidence of the fifth tubercle and its disappearance is only — 0·144. On the whole the variations in each of the several characters are independent of the variations in the other characters.

This series of specimens is clearly a group in which the latitude of variation is small, and may be contrasted with such groups as that of Planorbis multiformis, in which the variation was described by Professor Hickling (9).

I wish to thank Dr. A. E. Trueman for much helpful criticism and advice during the progress of this work, and Professor E. J. Evans, of the Physics department in this college, for the use of apparatus.