484 MR. HEUBEllT L. HAWKINS ON 



Tornquist (57), aiid Bather (59). A great part of their ambulacra 

 is still built of primaries, which show no signs of their subse- 

 quent modification except in the arrangement of the primaiy 

 tubercles, one to each group of three ambulacrals. A large 

 number of these Liassic forms are grouped under the generic 

 names of Biademopsis and Hemipedina. These genera and their 

 Diademoid allies have been so exhaustively studied by Bather 

 {t. c.) that no detailed discussion of their characters or affinities 

 is necessary here. One of the most obvious features which 

 separate these early Diademoids from their descendants is the 

 structui-e of the perignathic girdle. The processes, although well 

 developed, are rendered quite inconspicuous by the considerable 

 elevation of the ridges. The latter structures are, of course, a 

 relic of Cidarid characters. This shows that the change from an 

 interradial to a radial position for the perignathic prominences 

 was a gradual one. The view that the increasing complexity of 

 ambulacral structure is connected with the growth of the peri- 

 gnathic processes, whicli hinder the passage of the ambulacrals on 

 to the peristomial membrane, is supported by Bather (i. c.) on 

 this evidence. 



There are, then, two orders of Echinoidea from which, on 

 stratigraphical evidence alone, the Holectypoida may have been 

 evolved. Of these orders, the Cidaroida were well established, 

 with their special structures stereotyped, before there is any 

 evidence of the existence of Irregular Echinoids. This fact alone 

 would seem to render unlikely any hypothesis which regarded the 

 early Cidarid* as directly ancestral to the Holectypoida. 



When consideration is taken of the essential features of a 

 Pygaster, a notable correspondence between them and the 

 structures of the early Diademoida becomes apparent. The 

 ambulacra are chiefly composed of primaries (with a triple ar- 

 rangement of tubercles), and towards the peristome a partial 

 compression of the plates into triads is seen. Triad formation, 

 in the same part of the ambulacra, is characteristic of all the early 

 Diademoids, and is one of the diagnostic features of the whole 

 order. The perignathic girdle of Pygaster shows well-developed 

 processes, but hardly appreciable ridges. This character, the 

 absolute antithesis to that of the Cidarida?, is known to have been 

 gradually attained by the Diademoids through their Triassic and 

 Liassic representatives. Again, the apical system of Hemipedina 

 often shows a prolongation backwards into the posterior inter- 

 ambulacrum. 



Most significant of all is the indication of affinity between the 

 two orders by the slight ontogenetic evidence already available. 

 In discussing the affinities of Hemipedina bonei, Wright (20) 

 admitted that he was uncertain as to the true generic relations 

 of the species. He was at one time inclined to class it with 

 Pygaster. Bather (59) has referred to this species, and is of the 

 opinion that H. bonei, if it is not a Hemip>edina, should be asso- 

 ciated with Pygaster. The species is a small one, and the shape 



