Dr R. S. Allan has carefully collected from the many different beds in the Castlecliff—Kai-Iwi section, and his specimens show clearly that the apparent variability of multistriata from this locality has been due to careless collecting. Actually the shells from each bed are remarkably constant in characters and only in two beds, 6 and 10, were two forms collected. Four more or less distinct shapes can be distinguished and the following table shows their distribution.

From this it can be seen that the oval shell occurs fairly consistently in the shell beds and coarse bands, while the others occur mostly in fine muddy beds. Most Recent specimens seen (Awanui Bay, Great Barrier Island and Foveaux Strait) belong to the trigonal form, the northern shells differing from the southern only in having a bright, well-defined colour pattern. Shells in the Auckland Museum from mud, Cook Strait, 2½ miles north-west from Waikanae, are of highly elongate shape and agree very closely indeed with those from beds 1 and 5. Those classed as elongatetrigonal are intermediate in shape between typical trigonal and elongate forms. It may be that these differences in shape are directly due to the environment, but this does not explain the absence of the oval form from Recent faunules. Lower Pliocene shells from near Pahiatua seem to combine the characters of the oval and the trigonal forms, so it may be that we have here the emergence of a new species. In bed 6 at Castlecliff, the egenum sands, the elongatetrigonal and the oval forms both occur without grading, thus tending to confirm the indications that the two forms are distinct. Shell beds are liable to contain forms from different faunules, for many of them must be concentrations of mullusca brought about by the removal of the sands in which they were buried by increase in the rate of bottom currents; but the egenum sands do not appear to be of this nature.

A specimen in the Geological Survey collection from Star borough Creek is strongly subtrigonal, having broad, high beaks, a