JSCHINOIDEA. 237 



layers of the Cretaceous. The palaeozoic Palaeechinoidea are 

 very imperfectly known, but they comprise exocyclic as well as 

 endocyclic forms, though the majority are Endocyclic. They 

 make their appearance in the Upper Cambrian. 



From the late appearance of Spatangidae and Clypeastridae 

 it has been commonly assumed that the Endocycilca preceded 

 the Ectocyclica in evolution. This view is borne out by the fact 

 that Holectypus and the Cassidulidae, which are intermediate 

 between the regular and irregular type, preceded the Spatangidae 

 in the geological succession. But bearing in mind the fact that 

 Echinocystites is an exocyclic form from the Upper Silurian, we 

 should be prudent in suspending our judgment on this point, 

 until the Palaeozoic Echinoid fauna has been more fully inves- 

 tigated an attitude which is still further justified when we 

 remember that Collyrites, which is more modified in some re- 

 spects in the Spatangid manner than any of the Spatangidae, 

 preceded the latter in its first appearance, and is contemporan- 

 eous with (? before) Holectypus. 



A. Agassiz (Challenger Echinoids, p. 19) in discussing the origin of living 

 Echinoids calls attention to the hopeless nature of the attempt to re- 

 present the geological succession of forms either diagrammatically or 

 descriptively, and points out that this hopelessness is due to the great 

 number of different combinations of the various characters which have 

 existed in extinct forms. The structural features of living Echinoids are the 

 same essentially as those of extinct forms, but they are combined differently. 

 Features which have apparently disappeared reappear quite suddenly 

 and apparently in no connexion with the types which have immediately 

 preceded them. " We cannot hope," he says, " to trace the development 

 of any type through a series of forms each slightly different from its pre- 

 decessor ; we must only expect to be able to follow the changes of a single 

 feature and study it in its combination with other features, combinations 

 which from their very nature can never form an unbroken series, as their 

 terms are not synchronous. 



" If we examine in the same manner [i.e. by tracing it through the 

 Echinoids of all time] any one of the structural features which have once 

 made their appearance, we find that, without exception, they are either 

 persistent to the present day or can be traced in a somewhat modified 

 form in some one of the types now living, though the peculiar combination 

 of any definite number of these may have disappeared." 



Finally Agassiz goes on to say (p. 23) that " adopting this method of 

 tracing the development of a single structural feature at a time such as 

 the growth of the poriferous zone from the simple paired zone to the com- 

 plicated ambulacral zone of a Spatangoid, we shall find that the most 

 primitive ambulacral zone known still exists side by side with the exist- 

 ence at the present day of the resultants, if we may so say, of all the com- 

 binations which have taken place." 



