A MONOGRAPH OF THE EXISTING CRINOIDS 95 



The specimen from off Madras has 17 arms 120 mm. long, 7 IIBr 4(3+4) series 

 being present. One of the postradial series is in process of adolescent autotomy. The 

 cirri have 62-68 segments and are up to 63 mm. in length. I noted that this specimen 

 seems without any doubt to be a young example of C. rustica. 



Two specimens from Siboga station 226 were described as a new species under the 

 name of Thalassometra marginalis. Their characters were as follows: 



The centrodorsal is low, rounded conical. The cirrus sockets are arranged in 10 

 columns of 2 each, the two columns of each radial area being in close contact with those 

 of adjacent radial areas but separated interiorly by a bare triangle basally about half 

 as wide as a cirrus socket that comes to a point just beyond the last functional socket 

 and is continued thence as a narrow line to the dorsal pole. 



The cirri have 60 segments and are 50 mm. long. 



The ends of the basal rays are veiy prominent as high dorsoventrally elongate 

 tubercles which are usually rather longer than the dorsoventral length of a cirrus socket. 



In one specimen a small portion of the lateral end of the radials is visible as two 

 elongate tubercles meeting in a triangle over the ends of the basal rays. The proximal 

 borders of the IBri are produced into a narrow rim that projects downward over the 

 radials almost or quite concealing them, but cut away in the interradial angles to 

 accommodate the ends of the basal rays. 



One of the specimens has 17 arms about 95 mm. long; two of the IIBr series are 

 2 and five are 4(3 + 4). The other, of the same size, has 18 arms with all eight IIBr 

 series 4(3 + 4). 



I wrote that the two individuals of Thalassometra marginalis have about half the 

 arm and cirrus length of Crotalometra rustica, though apparently they represent a 

 species very closely related to it, and also to C. magnicirra. The new species was sup- 

 posed to differ from C. rustica in having the proximal edge of the IBri produced down- 

 ward into a narrow rim that projects over the radials entirely, or almost entirely, con- 

 cealing them. In the interradial angles the IBri are cut away to accommodate the basal 

 rays, whereas in C. rustica they are separated from the basal rays by the radials, which 

 form a continuous band all around the calyx. These differences are scarcely of specific 

 importance. 



A specimen from off Great Nicobar in 1,040-1,120 fathoms was described as a new 

 species under the name of Crotalometra eridanella as follows: 



The centrodorsal is hemispherical. The cirrus sockets are arranged in 10 columns, 

 usually three to a column. There are two columns in each radial area, one along each 

 side, each of these closely crowded against the adjacent column in the adjoining radial 

 area. The central portion of each radial area is occupied by a bare sunken strip which 

 is somewhat less in width that the breadth of the adjacent cirrus sockets. 



The cirri are about XXV, 45-57, large and stout, probably from 55 to 60 mm. in 

 length. The first segment is very short and those following increase in length to the 

 sixth, which is about as long as broad, or slightly longer than broad, and increase still 

 further to the tenth, a transition segment, which is about half again as long as broad. 

 From the eleventh onward the segments gradually decrease in length so that those in 

 the terminal third of the cirri are about twice as broad as long. On the tenth the dorsal 

 portion of the distal edge is slightly produced, and as the segments become shorter this 

 production of the distal edge extends proximally and at the same time rises so that the 



