ANEMONES 383 



If, however, they may be homologized with the latter, there can be 

 no representatives in E. adenensis for the four primary mesenteries 

 by which the transformation of the octamerous into the hexamerous 

 type is effected. On the other hand, if four of the eight mesenteries 

 in the lateral and ventro-lateral spaces do represent the fundaments of 

 these perfect mesenteries, then the rudimentary mesenteries in these 

 spaces do not arise in pairs, and we cannot account for the hexamerous 

 type any more successfully on this ground. 



E. sipunculoides is a still more aberrant form. The rudimentary 

 mesenteries do not occur in pairs, and are too variable to constitute a 

 stable foundation for a phylogenetic scheme. 



E. ddenensis 



£. beaufempsi 



Fig. 15. Scheme of the possible relationship of the species of Edxvardsia and the 

 Hexactiniae (through Halcampa). 



These ideas of the relationships of the species of Ednvardsia and 

 the Hexactiniae may be graphically expressed in the above diagram 

 (fig- 15)- 



Order ACTINIIN^ M. Edw. 



'^ Hexactiniae in which the tentacles are arranged in cycles, only a 

 single tentacle communicating with each endocoel."- (McMurrich, 

 1893, P- H^O 



