ECHINODEEMA. 3 



frontal loop on to the dorsal surface, and also in the narrow oral field. But there are, 

 nevertheless, irreconcilable differences between the two forms. The ciliated band of 

 the Orotava larva in its lateral portions is produced into numerous secondary processes, 

 so that the whole animal acquires the appearance of a Nudibranch mollusc, whence 

 the name A. nudibranchiata. Further, in this larva the post-oral processes are 

 absent, but these are well marked (Plate I., fig. 1) in our larva, and though the 

 primary processes are deeply marked there are no tag-like secondary processes such 

 as exist in A. nudibranchiata. 



In spite of its considerable size (4 mm. long) our specimen is a young larva, 

 for the coelom is still in the form of a single unpaired vesicle {coe., fig. 1) com- 

 municating with the exterior by a pore-canal. 



Reviewing all the evidence, we arrive at the conclusion that our larva is a 

 new type of Auricularia, to which the specific name "antarctica" may be given. 

 It probably belongs to some large Holothurioid of the group Synaptidae. 



II. Post-larval Stages of Cucumaria sp. 



The specimens composing this collection were mostly in a state of such intense 

 contraction that their shape approximated to that of a sphere, but a few were 

 considerably less contracted, and of these sections and whole mounts were made. 



In PI. I., fig 3, one of the most successful of the whole mounts is shown. It will 

 be seen that at the oral end of the body there is an atrium overarched by five valves, 

 recalling the vestibule of a young Crinoid. In the intervals of the valves glimpses of 

 the buccal tentacles can be seen. Of these there are ten, as we learn from transverse 

 sections. The suckers of tube-feet can be seen to be developed along three radii. 



But the most striking feature in the preparation is the presence of numerous 

 calcareous plates embedded in the -skin, so numerous indeed as to constitute a veritable 

 cuirass, especially over the dorsal surface; the plates do not touch edge to edge but overlap. 

 Each oral valve is supported by a special plate. Between the rows of tube-feet there 

 are also calcareous bodies to be found, but these are not fully formed plates. Through 

 the semi-transparent tissues the outline of the alimentary canal can just be made out, 

 and we can observe the thick oesophagus, the long stomach, and the slightly bent 

 intestine. 



Now the identification of these specimens as the young of some species of 

 Cucumaria rests (i) on the number and shape of the buccal tentacles, (ii) on the nature 

 of the calcareous plates embedded in the skin. 



With regard to the first point, Cucumaria is characterised by the possession of ten 

 " dendrochirote '" buccal tentacles, i.e., tentacles which exhibit scattered lateral 

 branches. Further, these tentacles are devoid of specially developed ampullae, but on 

 the contrary the whole tentacle ring can be retracted as an introvert. With regard to the 



B 2 



