482 MK. HERBERT L. HAWKINS ON 



by tlie existence of a free-swimming larval stage. Of post-larval 

 changes in the Class but little is known. Agassiz (30), in the 

 Revision, summarized the state of knowledge of the "young stages 

 of Echini,''' and but little has been axlded since that date, at least 

 in the case of the Irregularia. Ontogenetic characters are always 

 difficult to observe and to appreciate among fossil forms, and far 

 more zonal collecting of young stages of the Jurassic and Cretaceous 

 Echinoids will be necessary before this line of evidence can be used 

 for their correlation. 



Some slight details are available at present, such as the Ilemi- 

 pedina-\)h&se of Pygaster semisulcatus and the young stages of 

 Conulus with an adapical periproct. (Valette, 69, has described 

 a young specimen of C. suhconicus in which the periproct is 

 already in the adult position, although the individual has a dia- 

 meter of only 10 mm.). Unlike the Mollusca and Brachiopoda, 

 the Echinoidea do not retain the first-formed portions of the test 

 throughout life ; so that, although new parts are continually being 

 developed, the acceleration by which these new portions assume 

 adult characters almost nullifies any recapitulatory featui-es they 

 may possess. In the matter of the interambulacral tuberculation, 

 which at first seems a promising structure for ontogenetic study, 

 this feature of acceleration renders the characters of the new plates 

 practically worthless. 



In addition to their sequence in time, it is therefore necessary 

 to consider the adult characters of each genus separately. The 

 features of an adult are divisible into two kinds. The first group 

 is that of adaptation to circumstances ; and the characters due to 

 this tendency, though interesting from other standpoints, have 

 little phylogenetic meaning. The second group of characters are 

 those which are unaffected, or are not necessarily affected, by the 

 surroundings of the oiganism, and which must in consequence owe 

 any peculiarities they possess to the line of evolution of the group 

 to which the individual belongs. Such features, which include 

 atavistic and vestigial structures, are of first-rate importance for 

 showing the phylogeny of a group. In the Echinoidea, the cha- 

 racters that would fall into the fii'st category would be those 

 directly concerned with assimilation, respiration, reproduction, 

 and locomotion. The characters of the second type would consist 

 of apparently trifling variations in the ornament or structure of 

 the test — variations of such a kind as not to afiect the vital 

 processes to any serious degree, nor be affected by them. Such 

 characters are the details of the plating of the ambulacra and the 

 variations, within certain limits, in the structure of the apical 

 system. These two characters are regarded as essential indices 

 of relationship in the present paper. 



There is, however, in the investigation of an extinct, annectant 

 group like the Holectypoida, an additional principle of evolution 

 that gives safe guidance. The two extremes of structure — those 

 of a Cidarid and of a Spatangid — are known. Generally speaking, 

 the Holectypoida should show a gradual tendency, in the course 



