492 ME. p. H. CAEPENTEfi ON NEW 



slightly beyond the edges of the axillaries. The distal edge is 

 much incised to receive the strong backward projection of the 

 quadrate second brachial. The following joints have markedly 

 unequal sides, the third (or occasionally the fourth) and the fifth 

 or sixth being syzygies. 



Disk naked. Mouth slightly excentric, with the anal tube 

 just behind it. Brachial ambulacra close down upon and between 

 the muscular bundles. Skeleton brownish white. 



OiF Cojima, near Havana. Depth 450 fms. One specimen. 



Although this species was discovered some four years earlier 

 than At. halanoides, I have thought it best to take the latter as 

 the type species of the genus, as it is represented by more abundant 

 and better preserved material, the single example of At. cuhensis 

 having lost all its cirri and all the arms above the first six or 

 seven joints. The specific name was given to it by Pourtales, 

 who referred it provisionally to the type of Antedon cuhensis, 

 which was obtained in the same dredging. The chief difference 

 between it and Atelecrinus halanoides is in the uniform height of 

 the basal pentagon, and the stronger backward projection of the 

 third radials and second brachials into the preceding joints. 



3. Ateleceintjs Wttilli*, n. sp. 



Description of an Individual. — Centrodorsal acorn-shaped, 

 4 mm. high by 3 mm. wide. The double rows of cirrus-sockets 

 are well separated from one another by intervening spaces, and 

 do not reach the dorsal pole. Four, or rarely five, sockets in each 

 row, the ends of which stand out prominently and give a serrate 

 appearance to the lateral edge of the plate. The upper portion 

 is uniformly smooth, without any interradial ridges ; but the edge 

 is marked by five slight incisions situated interradially. 



The basals are nearly uniform in height throughout their whole 

 width, but are somewhat arched in form. The apex of each arch 

 is interradial, and the interval between it and the notched edge 

 of the centrodorsal below is only occupied by perisome. Hence 

 the basal ring is really only in contact with the centrodorsal at 

 its five lowest points, i. e. at the interbasal sutures, immediately 

 beneath the middle of each first radial. The latter have exceed- 

 ingly high muscle-plates projecting inwards; but their dorsal sur- 

 face is barely half as long as that of the second radials. These 

 are nearly square, but deeply incised to receive the strong back- 



* I dedicate this very interesting species to the memory of Sir Wyville 

 Thomson. 



