INTRODUCTION. 49 



(4) The way in wliicli cups are formed for the podia (see pp. 22 — 25). 



(2) [/>] The passage of the side canals between every two aniljulacralia (an 

 Asteroid character, sec p. 13). 



OpUurold Characters. — (2) [^/.] The passage of the side canals through the 

 substance of the ambnlacralia (see p. 18). 



(7) The sharp separation of the calcifications of the disc from the arm (see 

 p. 15). 



For other Ophiuroid characters (not- described hy Schondorf) in Lapwoythiira 

 and Aspidosoma graijae., see pp. 24 — 26, 3 1 — 33. 



Special Characters. — (1) The enclosure of the radial water-vascnlar system can 

 be traced to an overgrowth of the longitudinal ridges, which, in the primitive 

 Asteroidea, ran alongside the ambulacral channel (see pp. 21 — 24). 



The character itself is curiously constant in Palgeozoic " wrigglers," and served 

 to protect the delicate water-vessel and its accompanying nerves and blood-canals. 

 It must be remembered that in Recent Ophiuroidea these structures are covered 

 over by ventral shields, and in Asteroidea they lie at the bottom of a deep groove 

 bordered by stout spines. Nevertheless, a similar enclosure of the water-vessel 

 may occur (possibly as a survival of an ancestral "tendency") in Recent Ophiu- 

 roidea and Asteroidea, e. g., in the third vertebra of Op)hiarachna incrassata , 

 Midler, and in Brisinga coronata, Sars (Schondorf, 67, p. 45). 



The ambnlacralia in many species of the Aspidosomatid^ undoubtedly have the 

 appearance of hollow half-cylinders. This is due to the fact that the radial canal 

 has become very large (see p. 24), and in consequence the dorsal median portions 

 of the and)ulacralia become thin and negligible. In forms such as Lapivortliura , 

 where the canal is relatively smaller, the ambnlacralia do not take on this appear- 

 ance. 



(6) The peculiar marginalia of the Aspidosomatidee ; have already been referred 

 to (p. 39). 



There is a convex disc of the usual Ophiurid type in Lapwortkuva ; Schondorf 

 is not correct (68, p. 213) in his interpretation of the disc of these forms as being 

 slightly concave. 



(5) I hesitate to accept this character. It is true that some species of the 

 Aspidosomatidge have sculptured ambnlacralia. This sculpturing extends on 

 the ambulacralia nearest to the mouth. I can scarcely believe that the sculpture 

 was external ornament. The disc must have been over the ambulacralia in this 

 region, unless we surmise either that the form was without a stomach or that the 

 disc Avas reduced to inter-radial pouches. 



The adambulacralia of Stenaster obtu.siis (PI. I, fig. 7) have a similar 

 sculpturing, and in this case there is no doubt that there was a skin with weak 

 calcifications, external and independent of the adambulacral scidpture. 



An analysis of the kind given above shows that the " Auluroidea " are really 



7 



