186 PALAEONTOLOGY OF OHIO. 



another. On the contrary, the corallites are more or less cylindrical, are 

 disconnected, sometimes placed quite apart, and each provided with a 

 separate epitheca, which is furnished with longitudinal stria and longi- 

 tudinal ridges. This may be a separate species, or may be only a variety 

 of F. stellata, but I have not yet had the opportunity of examining it 

 sufficiently. 



I Milne Edwards and Haime, in their great work, not only unite the 

 genera Favistella and Columnaria, but also regard F. stellata, Hall, as being 

 identical with Columnaria alveolata, Gold. I have above expressed the 

 opinion that C. alveolata, Gold., was originally founded upon examples of 

 Favistella stellata, Hall ; and the latter name should in this case be un- 

 doubtedly abandoned, according to the strict law of priority. In the 

 meanwhile, however, the name of Columnaria alveolata has become by 

 general consent firmly fixed upon a very well known and abundant coral 

 from the Trenton limestone, which, in my opinionj is clearly distinct 

 specifically from the coral described by Hall, from the younger deposits 

 of the Hudson River group, under the name of Favistella stellata. Whilst, 

 therefore, it may reasonably be doubted if Favistella is capable of being 

 generically separated from Columnaria, it appears to me that the course 

 which will cause the least confusion is to leave the coral now universally 

 recognized by American palaeontologists as Columnaria alveolata under 

 this name, retaining the Hudson River form as distinct, and leaving it 

 . an open question whether the latter should be called Favistella steltata or 

 Columnaria stellata. At the same time, as above stated, my own studies 

 . of these forms would lead me to think that Favistella, though undoubtedly 

 very closely allied to Columnaria, may reasonably be regarded as distinct, 

 provided the specimens originally described by Goldfuss under the name 

 of C. alveolata be. regarded as being erroneously referred to Columnaria. 

 If this course should not be adopted, the only alternative will be to re- 

 define the genus Columnaria of Goldfuss, so as to make it include Favis- 

 . tella stellata, Hall, and to form a new genus for the reception of the so- 

 . called Columnaria alveolata of the Trenton limestone and its allies. 



Position and lomlify: Hudson River formation (Cincinnati group), Cincinnati, Ohio. 



Genus COLrMNOPOR^V, Nicholson. 



(Geological Magazine, 1874.) 



Corallum aggregate, massive, composed of hexagonal corallites, which 

 have distinct walls, but are firmly and inseparably united with one 



