84 TABULATE CORALS. 



about one millimeter in diameter, and thus agreeing with the 

 examples which I have seen from the Eifel. The statements 

 on this head made by Milne-Edwards and Haime would much 

 more closely apply to the form which they have described and 

 fieured from the Devonian rocks of Devonshire under the 

 name of F. ccrvicornis (Brit. Foss. Cor., PL XLVIII., fig. 2), 

 and which is so common in the limestones of Newton-Bushell 

 and Torquay. This form has large calices which are often 

 nearly two millimeters in diameter, and the thickening of its 

 walls (though present) is not strikingly marked. I am, how- 

 ever, very much disposed to doubt if this be specifically iden- 

 tical with the form figured by Goldfuss from the Eifel, and 

 my opinion is based both upon actual specimens and upon 

 thin sections. The Devonshire form is rather sublobate than 

 strictly dendroid ; its walls are not nearly so thick as those of 

 the typical examples from the Eifel ; its calices are on an aver- 

 age larger ; and its tabulse are much more numerous and more 

 closely set. Upon these grounds, though I have doubtfully 

 placed it under this species, I am inclined to think that it is 

 entitled to at least varietal if not specific distinction. The form 

 figured by Billings, from the Corniferous Limestone of Canada, 

 as F. ccrvicornis (Canad. Journ., new sen, vol. iv. p. no, fig. 

 9), and subsequently described by me under the same name 

 (Rep. on the Pal. of Ontario, p. 52, 1874), is apparently identi- 

 cal with, or closely allied to, the Devonshire form, and differs 

 from the Eifel specimens in the characters above alluded to as 

 distinctive of the British examples. 



Favosites polymorpha, Goldfuss, has often been regarded as 

 comprising dendroid forms, probably In part identical with the 

 present type. It seems clear, however, that this name, if it be 

 retained at all, can only be with propriety kept for examples of 

 a submassive or completely massive form, with uniserial pores, 

 and with the thin walls characteristic of Favosites proper. 



Favosites dzibia, De Blainv., again, comprises dendroid species 

 which are obviously of the Pachypora type, and it is a question 

 whether the species can be maintained as distinct from Pachy- 



