SYSTEMATIC SECTIONS. 



65 



comparison between the lineage-history of fossil fishes and that of Promopalseaster 

 is almost complete. It is probable that, as our knowledge grows, the same series 

 of changes will be followed in other Palgeozoic Asterozoa (see also p. 100). 





Size. 



Geological hoiizon. 



Oral surface. 



Apical surface. 



Hudsonasteridae 



Small 



Mainly Ordo- 



Odontophor helps to 



Primarv radialia and 







vician 



form the margin 



interradialia pronounced. 

 No accessory plates. . 



Mesopalseasterinae . 



Small to 



Ordovician to 



Odontophor enclosed bj 



Accessory plates, ad- 





large 



Carl)oniferous ? 



one pair only of infero- 

 marginalia 



radialia and inter- 

 marginalia present. 



Promopalseasterinse 



Large 



Ordovician 



Odontophor enclosed by 

 several infero-marginalia 



Accessory plates well 



developed. Radialia 



may break up. 



Xenasteridse . 



Large 



Devonian 



Ditto, veutro-latei-alia 

 also present 



Accessory plates usually 

 developed. 



(For Schuchert's families Palseasteridae, Gregory (emend.), and Neopalseastei-idae, see pp. 102-104). 



Pre 



M. Ordovician. 



-5 , 



M. &. U. 



Ordovician. 



Silurian. 



Devonian. 



Hudsonasteridae. 



XenasterideE. 



Mesopalaeasterinffi. 



PromopalaeasterinaB. 

 , Uranasteridae. 



Diagram illu.strating the sheaves of the various lineages and their derivation from a common stock. 

 The thickened portions of each line are meant to suggest that the stage of highest elaboration of each 

 lineage in individual lineages of a sheaf may be reached at varying periods. This is purely an illustra- 

 tive diagram, and our knowledge is not suilicient to work out the lineage-history of each sheaf in dotail.J 



(C) Changes in Ornament. 



The ornament on these early Palaeozoic forms is also of great interest. In, at 

 any rate, the more primitive species it recalls very strongly the pustulose ornament 

 of Edrioaster, which has been described by Bather (89, p. 167) as follows : " The 

 surface in well-preserved tracts is coarsely pustulate (pi. xi, fig. 1). The pustules 

 are distinct, and spring from an irregularly reticular surface. . . . These 



