O'COWIJLL, h'i:\IS/()\ OF THE GEM ;S ZM'llh'lJXT I S 179 



synonymous with Zaphrentis. It was described first, however, merely 

 because Michelin took up the fossils of Sable before those of Tournay, 

 where the other species occur. These three others are C. coniu-hovis, 

 C. patula and (7. cornucopia^ given in the order in which they appear in 

 the "Iconographie Zoophytologique." Carruthers has recently stated 

 (22, 159, 166, 168) that C. comu-hovis represents a developmental stage 

 in the ontogeny of C. cornucopia', being the adult of the latter, and it 

 should, therefore, be included under C. cornucopias in all synonymies. 

 Carruthers not only studied the original descriptions and figures, but also 

 had the opportunity to examine the type material and many other speci- 

 mens from Tournay. N'evertheless, if Michelin's figure of the type of 

 C cornucopice (PI. 59, fig. 5) is of natural size, we cannot accept this 

 determination, though we can easily understand that the very young of 

 C. comu-hovis has the character of C. corjiucopia\ C. comu-hovis has, 

 however, the adult characters of the genus'.^Siphonophyllia as discussed 

 below. C. patula, on the other hand, agrees with f\ cornucopice. If C. 

 comu-hovis were the type of Caninia)<Sip]ionopliyUia would have no 

 generic standing. Lambe and other American authors have selected C. 

 patula as the type of Caninia, but there seems to be no good ground for 

 this selection. The question, however, is easily settled, for Michelin ex- 

 pressly stated (4) that cornucopia' was the type of Caninia^ and he gave 

 the following description : 



Stony polyp, free or fixed, sub-turbinate, simple, cylindrical, formed of super- 

 posed cellules (onps) each cell furnished marginally \yith lamellre sometimes 

 very short and twisted, sometimes reaching the center, but remarkable in that 

 they can be separated into little conoid funnels representing beyond doubt the 

 succession of the principal vital phases of the polyp, and set one into the other 

 around and forward from the central axis ; exterior striated.^ 



This description is composite and fits C. comu-hovis better than it 

 does the other two species ; but since the type of the genus was definitely 

 named, the generic characters must be determined by it. These are given 

 in the summary at the end of the paper. 



The selecting of patula as the type may possibly be explained in the 

 following "way : C. gigantea, as will 1)e shown below, does not belong to the 



^ "C cornu-copiw, Mich. Espece type de genre dedie au prince Cli. Bonaparte." 

 ^ "Ses caracteres sont : polypier pierreux, libre ou fixe, subturbine, simple, cylindrique, 

 forme de cellules superposees chaqne cellule garnie marginalement de lamelles, quelque- 

 fois tr&s courtes et sinueuses. quelquefois ^ atteignant le centre, mais remarquable en ce 

 qu'il est decompasable en petits conoids, representant sans douto la succession des prin- 

 cipales phases vitales du polype, et s'emboitant les uns dans les autres en dehors et en 

 avant de I'axe central ; I'exterieur est strio." (22, 166.) 

 8 Introduced by Carruthers for the sake of clearness. 



