38 PBOF. p. M. Duncan's eevision of the 



and it is of course less wlien the pressure is localized ; they may 

 be bent inwards and swell out elsewhere correspondingly. During 

 life, moTements of the test occur, and it is doubtless due to the 

 contraction and dilatation of the internal longitudinal muscles, 

 noticed and drawn by Wy. Thomson, and so well described and 

 drawn by the Drs. Sarasin, as well as to gravitation. During 

 this contraction and also after death the edges of the plates, 

 especially the transverse edges, are approximated and slip over 

 or under each other as the case may be. Nothing can be clearer 

 than the drawings of A. Agassiz in regard to Fhormosoma tenue 

 (' Challenger' Eeport, p. 93, pi. xviii. a. figs. 1-13). The whole of 

 the transverse edge of a plate overlaps that of its fellow in suc- 

 cession. The fig. 4 has no overlap of edge, fig. 5 has it slightly, 

 but not figs. 2, 7, 8. The amount of overriding is small, and is 

 assisted to a small degree by the thinning of the edges of all the 

 plates, but there is no such thiug as a bevelled thick edge. 



In the drawing given by Sir Wy. Thomson (op. cit. 1874, pi. Ixv. 

 fig. 2), which shows the inner part of a test of Asthenosoma below 

 the ambitus, the pairs of pores situated in the transverse sutural 

 spaces are not forced out of their direction ; hence the overriding 

 must be very slight there. There is, on the other hand, definite 

 but limited overlap of the interradial plates. The amount of 

 interplate membrane varies in the species of Phormosoma, and it 

 is less on the whole than in those of Asthenosoma, and yet the 

 amount of overlap is only partial in this genus. The species 

 JBJiormosoma rigidwm, A. Ag., has a significant name, and it is 

 evident that the interplate membranes are reduced to their 

 utmost ; there is no appreciable space between the plates in the 

 drawing (Eeport on the ' Challenger ' Echini, pi. xii. a. figs. 1-4, 

 p. 104) ; all the other generic characters are present, but this is 

 not a flexible and "panting" form, and has most significant 

 alliances with the Diadematidse. 



The pistol-shaped outlined plates of Asthenosoma do not imbri- 

 cate on most of their transverse edges, but the overlap is perfect 

 at the median line of both areas," and a flap of plate must be more 

 or less permanently overlapped or underlapped by another there. 

 The ambulacral plates within the peristome, clearly overlap, 

 and in order that the movement shall be uncontrolled there, the 

 tentacles are associated with pores which are seen upon the plates 

 themselves, and they are not upon the interplate structure. 



In both genera of recent Echinothuridse a special character 



