128 THE VOYAGE OF H.M.S. CHALLENGER. 



anterior region we get forms like Pourtalesia carinata. Witli these lower anterior coronal 

 plates, combined with a great elongation of the plates, especially of those of the posterior 

 lateral, interambulacral, and ambulacral areas, we get such forms as Pourtalesia phiale, 

 while with plates of a uniform size in the different parts of the test we get such conical 

 forms as Echinocrepis, and when more or less elongate posteriorly, such forms as 

 Pourtalesia ceratopyga by the elongation of the posterior plates. 



In Spatagocystis the arched elliptical test is the result of a large number of hori- 

 zontally elongated coronal plates, somewhat larger in the posterior region of the test and 

 of uniform size anteriorly, while the more regular outline of Urechinus is the result 

 of a greater uniformity in the size of the plates of the anterior and posterior extremities, 

 which reaches its maximum of regularity in the nearly circular outline of Cystechinus with 

 its ambulacral and interambulacral areas composed of plates nearly aU of uniform size in 

 corresponding parts of the test. 



The genus EcJiinocrepis seems to form a passage between the Pourtalesise without 

 an anal snout and such Ananchytid genera as Cystechinus. We readily trace also 

 how such genera as Nucleolites do not differ so radically as they seem to do at first 

 sight from genera in which the anal groove is not developed. In fact when we 

 compare the fossU genera Echinohrissus, Hyhoclypus, Clypeopygus, and the like, with 

 the Pourtalesise, we' find in the latter family a sunken actinal groove, a feature once 

 so common among the genera of the Jurrassic and Cretaceous periods. This structure 

 has become less and less common until we find but few genera in the Tertiary formation 

 with a sunken anal groove, and at the present time only a few species of Nucleolites 

 in addition to the Pourtalesise stUl retain the sunken anal groove. This seems in 

 the Secondary and Tertiary periods to have been a character peculiar to the Echino- 

 lampadae and their allies. The existence of a deeply-sunken actinal groove or its repre- 

 sentative is, however, much more prevalent among the recent Spatangoids, and its modifi- 

 cation from the simple sunken ambulacrum of Asterostoma to the deeply-sunken actinal 

 groove of Pourtalesia can be easUy traced. It has left its trace in the greater number 

 of recent Spatangoids in the more or less sunken anterior ambulacral groove. This, 

 however, in most recent Spatangoid genera is limited to the abactinal .side of the test, 

 generally disappearing at the ambitus, taking its greatest development in the deeply- 

 sunken ambulacra of some of the Schizasteridae such as Moira and Schizaster, and forming 

 a gradual transition, as it were, between the existence of a single deeply sunken anterior 

 ambulacral groove situated on the actinal side and the more or less sunken petaloid 

 ambulacra. That is to say, the actinal groove is a modification at the actinal region 

 of the ambulacra similiar to the sinking of the plates of the apical part of the ambulacra 

 to form more or less deeply sunken areas ; only the pores remain single, and there is no 

 modification of the pores forming the petals, indicating a different function, as in the 

 normal Spatangoids. 



