INTRODU("TION. 33 



'riie 11^ of these authors is the embryonic a^ of Ludwig-, Avhich they believe they 

 can detect in Tjapworfliura in a constriction of the above bow-shaped ossicles. 



My interpretation of the structure is as follows (PI. I, fig. 9) : All the 

 arnlnilacralia slope downwards towards the mouth [rule sujmi, p. 26). The first 

 pair of ambulacralia {cio of Sollas) are much larger than the immediately succeeding 

 ambulacralia and wholly or partially hide them. The second pair of ambulacralia 

 {a^ of Sollas) are particularly small, and only their most dorsal tip can usually be 

 distinguished. The third pair of ambulacralia (a^, of Sollas) are also overlapped, but 

 not quite so extensively as are the preceding pair. 



It may be argued that a post-mortem redeposition of calcite has contributed to 

 the apparent fusion of the apical crests of these ossicles as noticed by Sollas and 

 Sollas. The amount of overlap appears to vary, suggesting that during life there 

 was some play possible between the ossicles. Nevertheless,, with the very clear 

 example of Steitoiiferhefove us, it appears to me that the above authors are correct 

 in their conclusion that a fusion of several pairs of ambulacralia in the mouth 

 region may occur as a distinctive feature in many species of Palaeozoic 0{)hiuroidea. 

 The amount of fusion probably varies in different individuals of the same species 

 and most commonly affects the first two pairs of ambulacralia. 



Orally there is also a marked differentiation in the ambulacralia (PL I, fig. 10). 

 The base of the first ambulacra! is excavated into a large bowl-shaped depression, 

 doubtless occupied by a large buccal tentacle (the second buccal tentacle). The 

 first buccal tentacle occupies the same position as it does in Eoactis and Stenaster. 



The adambulacralia are almost as much modified as in recent Ophiuroidea. The 

 first adambulacralia (lateral buccal shields of Recent forms) are often fused with the 

 first ambulacralia and are not to be distinguished as separate plates. The second 

 adambulacralia are very small and appear as very thin plates in apical view. 

 It is not until we reach the fourth or fifth ossicles of the series that the adam- 

 bulacralia form a stout wall to the arm (PL I, fig. 9). 



A thin plate Avliich occupies the jjosition of a torus is apparently present, but 

 it is difficult to recognise it w^tli certainty. Stout spines (" teeth ") are situated 

 at the apex of the mouth-angle plates. 



As in Stenaster, the groove for the radial water-vascular vessel is very wide 

 near the mouth, but quickly narrows and becomes completely closed over. 



2'/ie Month-Frame of Aspidosoma grayse. — The mouth-frame of Aspidosoma will 

 be described in detail later. Meanwhile it is sufficient to note that its structure, 

 just as that of the arm (p. 24), is intermediate between that of Sfenaster and that 

 of Lapworthvni. PL I, fig. 8 shows the differentiation of the ossicles as seen in 

 oral view. 



