120 THE VOYAGE OF H.M.S. CHALLENGER. 



The dorsal processes belong in general to the dorsal ambulacra, but exceptions have 

 been found in Deima, Oneirophanta, Orphnurgus, and Pannychia, which carry processes 

 not only all over the dorsal ambulacra but also in a row situated above the pedicels 

 along each side of the body ; these lateral processes are in communication with the 

 ventral lateral ambulacra, which thus carry both pedicels and processes. Colochirus, 

 Troschel, &c, among the Pedata, proves that this is a peculiarity not exclusively charac- 

 teristic of these four forms. In all the Deimatidaa, as well as in several forms belonging 

 to the Psychropotidse, the dorsal processes are arranged in one or several rows along 

 each ambulacrum. Those animals which, as, for instance, Oneirophanta and Deima 

 are provided with processes of unusual size and length, have them disposed only 

 in a single row. Very rarely, and, as far as I know, only in Benihodytes sanguinolenta, 

 and possibly in Pannychia moseleyi, the processes are found scattered over the lateral 

 interambulacra. In the family Elpidiidse, on the contrary, traces of such an arrangement 

 in rows are obvious, but the processes are fewer in number and the rows have a tendency 

 to become dissolved. A closer examination of the representatives of the family in 

 question shows that the processes do not as a rule change their position, but are 

 situated at fixed places, either anteriorly, or both anteriorly and posteriorly, thus 

 leaving a greater or smaller portion in the middle of the back devoid of all processes. 

 At the same time it will be seen that the number of the processes is definite, as in 

 Scotoplanes globosa which always carries only three pairs, — a very remarkable peculiarity, 

 which is, doubtless, to be found in most of the representatives of the family in question, 

 though through lack of necessary material I have not been able to distinguish it ; in some 

 forms, as, for instance, in Elpidia glacialis, they vary a little in size as well as in position, 

 but even here can be found traces of a tendency to become definite in number and 

 position. In the Elpidiidse the processes are most evidently disposed in pairs, and though 

 the latter themselves are of a very variable size when compared to one another, 

 nevertheless each pair is made up of processes of equal size. 



As before noted another kind of dorsal ambulacral appendage is present, which, being 

 generally odd and very large, traverses the bivium from the one ambulacrum to the other, 

 and appears to be made up of one pair of processes or more, viz. , it is penetrated by 

 one or several pairs of wide tubes which are in communication with the two dorsal 

 ambulacra. It is most evident that since the animals are supplied with such a large 

 and long appendage, which either has its free end rounded or provided with lobe-like 

 processes, their appearance is highly characteristic. In Peniagone, Scotoanassa, &c, the 

 appendage in question is situated anteriorly, and is usually broad, flat, and made up 

 of several pairs of canals ; in Euphronides it is placed a little behind the middle of 

 the dorsum, and presents a more conical form, while in Psychropotes it protrudes near the 

 posterior extremity of the body, and is remarkable for its gigantic size and tail-like 

 aspect ; the appendages of the two last-mentioned genera are penetrated by a single pair 



