PROTARTHRASTER LONGIMANUS. 



L65 



apical surface show the position of the original ossicles very clearly (Text-fig. 1<> I). 

 Their position is almost diagrammatically simple. There is a distinct centrale 

 which, like all the other apical plates, has a raised central area. Around the 



lot; 



107 



I.M. A , d - 



Text-fig. 106. — Plan of ossicles on the oral surface of Prolarthraster longimanus. Ad., adamhulacralia ; 



I.M., infero-marginalia ; 31. P., mouth-angle plates ; O., odontophor. x 12. 

 Text-fig. 107. — Wash drawing of a portion of the oral surface of Protarthraster longimanus. Ad., adra- 



dialia ; 1. 31., infero-marginalia. x 18. 



centrale are grouped five primary radialia forming a closed circle. In the inter- 

 radial angles just distal to the primary radialia are five triangular plates which, as 

 suggested above (p. 163), probably represent the fused pairs of adradialia from 

 adjacent arms. The remaining adradialia are added in the same manner as in 



Mr? 



Text-fig. 108. — Wash drawing of the mouth-angle plates of Protart/i raster longimanus. 31. P., mouth- 

 angle plate ; O., odontophor ; T., torus, x 8. 



Urasterella. All the plates of the apical surface are closely touching without any 

 interspaces for the protrusion of papulae. 



The paxilla- shafts of the radialia are drawn in Text-fig. 104 as being trans- 

 verse ridges. This is as they appear on one arm. On other arms the paxilla-shafts 

 of the proximal radialia appear to be knob-like as in Salteraster ? coronella. I am 

 unable to say whether this appearance is natural or whether it is due to a destruc- 



22 



