its other sliglitly-slopiug upper side, first by two large plates ia direct 

 succession, the upper ouo of which is larger than the other, wider than 

 long, generally hexagonal in form, and gives off on the interaxillary 

 side another arm-like series of pieces, which pieces diminish rapidly in 

 size upward, and connect with a corresponding series from the other 

 division of the ray over the interaxillary space, so as to isolate the inter- 

 axillary pieces. At least, a few of the lower pieces of these little arm- 

 like series over the interaxillary spaces are fastened together and to the 

 adjacent parts, to form portions of the walls of the body. The next piece 

 of the main division of each ray, above that last mentioned, also gives 

 off, on the interradial side, another arm-like series of pieces, some of the 

 lower of which are fastened to those of the larger series first described. 

 Above this, the main divisions of the rays soon become free arms, com- 

 posed of much shorter, transversely-oblong, or slightly wedge-shaped 

 pieces, apparently every second one of which gives off alternately, on oppo- 

 site sides, short, free pinnules. Soon the arms become rounded, on the 

 outer or dorsal side, with a deep ambulacral furrow within ; while be- 

 yond they very gradually taper and extend to great lengths, apparently 

 without bifurcating. In all cases, they and the intermediate arm-like 

 divisions draw together rapidly at the upper \)avt of the body before 

 becoming free, and thus seem to reduce the vault or ventral area to com- 

 j)aratively small dimensions. 



The specimen represented by Fig. B shows one of the interradial 

 areas (or possibly it may be the anal) to be rather large, and occupied 

 by nine unequal plates, the lower one of which is wider than high, 

 hexagonal in form, and supports two nearly as large heptagonal pieces 

 in the next range, while those above are of various forms and sizes, and 

 more or less irregularly arranged. Mr. GrinnelFs Fig. 1 also shows 

 about the same number of pieces in one of these areas, but they are 

 somewhat differently arranged. 



Of the interaxillary spaces, our specimens show only two in one 

 specimen, and one in another, and each of these is occupied by only 

 two interaxillary pieces, generally of heptagonal form, and placed 

 directly one above the other. In order that the interradial and inter- 

 axillary series of plates may be the more readily distinguished from 

 those belonging to the radial system, I have had a light shade drawn 

 over the former in figure B, and a dotted line run up each of the rays 

 and its subdivisions. 



None of our specimens retain the arms much above the summit of the 

 body; but Mr. Grinnell says that in some of those studied by him they 

 can be traced to a length of eight inches, and jet have the ends broken 

 off", so that he thinks they may have attained a length of not less than 

 one foot. One of his figured specimens shows a large number of long, 

 slender, jointed pieces that seem to be portions of slender, detached, 

 simple arms, bearing short pinnules, apparently in some cases springing 

 one from each joint. As much as the small interradial arm-like divis- 

 ions of the rays look like a series of smaller arms springing from the sum- 

 mit of the interradial pieces, it is evident that they are really parts of the 

 radial system, or, in other words, divisions given off fi'om the secondary 

 radials; and the same may be said of those seen in the interaxillary 

 spaces. In fact, they all correspond to the pinnules of the free rays or 

 arms in all respects, excepting that they are larger, and united with 

 each other and with the main divisions of the rays for some distance 

 up, so as to constitute a part of the walls of the body. 



Some of the characters mentioned in these remarks are generic and 

 some merely specific, as we have not yet the means of giving separate 

 generic and specific diagnoses from such specimens as have been found. 



