lines on the scuta are more distinct than in the specimens already referred to. A few specimens differ from both of the varieties already mentioned in having the carina much broader and less acuminate, not barbed, and in having little or no trace of an internal tooth on either scutum. The radiating lines on both scuta and terga are fairly distinct. These specimens therefore differ from the typical form of L. anatifera in the absence of the internal tooth on the right-hand scutum and in the shape of the carina. In some respects they seem to come pretty close to L. testudinata Aurivillius, from the Cape of Good Hope. They differ, however, from that species, as described and figured by Gruvel, in the broader carina and in the shape of its fork, and apparently also in having 5 teeth on the mandibles. Neither of these points is of much importance, and but for the absence of the internal tooth on the scutum there is perhaps little to distinguish L. testudinata from L. anatifera. In my Kermadec Island specimens both the scuta have the umbonal angle somewhat incurved, but there is nothing that can strictly be called a tooth on either of them. For the present I prefer to look upon these specimens as a variety of the widespread and variable L. anatifera. It differs from L. hillii in having only two filaments. I have numerous specimens from the Chatham Islands that appear to be practically identical with this variety from the Kermadecs.

L. anatifera is almost cosmopolitan, but has not been recorded from New Zealand seas.