204 The Philippine Journal of Science leis 



a relationship to both Cornularia and Clavu- 

 laria. Its external characters are those of 

 Cornularia; that is, there are no spicules, the 

 Fig. 1. Part of a col- stolons and the proximal portions of the polyp 

 ony of Cornularia ^^q covcred with a homv envelope within 



mi nu t a sp. nov. i-iji tit 



Actual size. which the distal portion of the polyp is retrac- 



tile, and the polyps are connected by very 

 slender cylindrical stolons. Here the similarity ceases, and the' 

 other characters are those of Clavularia. The stolons contain a 

 number of endodermal canals instead of one as in Cornularia. 

 These are surrounded by a very thick mesogloea instead of the 

 thin lamella of Cornularia; the polyps have a general form more 

 like Clavularia than Cornularia, as they taper from the base to 

 the distal end, while in Cornularia the basal portion of the polyp 

 has the least diameter; the polyp in retraction has the form 

 characteristic of Clavularia — that is, the distal portion is re- 

 tracted within the proximal portion; and finally the perisarc, 

 while distinct and always present, is extremely thin and might be 

 considered as either disappearing or as being a recently acquired 

 character. The external characters, however, are exactly those 

 on which the generic definition of Cornularia is based, and we are 

 confronted with the necessity of founding a new genus inter- 

 mediate between Cornularia and Clavularia to receive the new 

 species, or of placing it in the genus Clavularia in spite of the 

 fact that in external characters it agrees with the definition of 

 Cornularia, or of placing it in Cornularia in spite of the fact that 

 it agrees with Clavularia in certain of its characters. Now that 

 the family Cornularidse, after much confusion and wasted effort, 

 has been reduced to a few well-defined genera, it seems unwise 

 to introduce new genera unless absolutely neces- 

 sary. The genus Cornularia as now constituted 

 contains only one or perhaps two species, and the 

 addition of this new species should not lead to 

 confusion even though it is atypical in some of 

 its characters. Of course, it is out of the ques- 

 tion to add to the already large genus Clavularia 

 a form which difi'ers so distinctly from all the y^a.-i. a par- 

 known species of that genus. For these reasons tiaiiy retracted 

 I have decided to consider the species under dis- ^°'7^. °^ ^°'" 

 cussion as an atypical species of the genus Cornu- uta. showinK 

 laria, forming a connecting link between it and ^^^ . ""■'"k'^' 



. perisarclike 



the genus Clavularia. covering of 



As I have stated above, the genus Cornularia ^^^ '"'^''' p*""" 



tion. Much 



has contained until the present time but one well- enlarged. 



