No. 3-] THE MORPHOLOGY OF THE ACTINOZOA. 119 



Boveri traces these forms back phylogenetically to a stage 

 antecedent to the appearance of the mesenteries in cycles and 

 in pairs, i.e., to a stage where the mesenteries developed, as in 

 the Edwardsiae, simply and bilaterally. By the development 

 of two mesenteries on each side, the Edwardsia stage was 

 brought into an hexactinian condition, and the formation of 

 another mesentery on each side of one of the Edwardsian 

 directives produced the Scytophorus condition. A further 

 continuation of this process, the development of another 

 mesentery on each side of the other pair of directives, resulted 

 in the Holactiniae, which, it must be assumed, later acquired a 

 tendency to develop additional mesenteries cyclically and in 

 pairs as do the Hexactiniae. 



Boveri consequently assumes the actual persistence of the 

 Edwardsian directives in both Scytophorus and Gyractis, the 

 absence of one pair in the former genus and of both pairs in 

 the latter being only apparent. With his views as to Scyto- 

 phorus I fully coincide, but believe that in Gyractis we have 

 to deal with an hexactinian in which both pairs of directives 

 have disappeared, - -just as one pair has disappeared in numer- 

 ous specimens of Metridium marginatum, the mesenteries 

 which really represent them having developed their longitudi- 

 nal muscles on adjacent faces. 



On reading Boveri's paper I recalled the fact that I had 

 been unable to make out directive mesenteries in RicSrdea 

 florida ('89), and I again subjected my preparations of that 

 form to a close scrutiny. The stomatodaeum of this species is 

 oval, its walls having numerous longitudinal ridges, but there 

 are no siphonoglyphes. Owing to the disc-like form of the 

 column and the moulding of the base over the irregularities of 

 the surface to which the animals attach themselves, it proved 

 difficult to obtain perfectly satisfactory transverse sections of 

 the entire column. In two cases, however, I succeeded in 

 getting sections from which I could be reasonably certain as 

 to the presence or absence of directives ; in one case I found 

 no directives and in the other a single pair, which, however, 

 was not opposite the end of the long axis of the stomatodaeum, 

 but to one side of it. 



