18/1.] MR. W. S. KENT ON NEW MADREPORES. 2/9 



whether the characters given may prove sufficient for the recog- 

 nition of two distinct genera on other species becoming known. 

 A. oculata (Ehr.) was the single form referred to the genus Allopora 

 by Milne-Edwards in his work just quoted. Recently Count de 

 Pourtales has discovered a second species, off the coast of Florida, 

 which he describes in the Bulletin of the Museum of Comparative 

 Zoology, Cambridge, U. S. for 1868, under the name of Allopora 

 miniata ; and my examination of the collection in the British 

 Museum has resulted in my detecting three other distinct and 

 undescribed species, undoubtedly referable to this same genus. 



The mass of evidence now accumulated demonstrates that Allopora 

 and Stylaster constitute two natural and easily defined genera. 

 The character of the presence or absence of ampullae, however, as 

 shown in the following description of A. explanata, is not trust- 

 worthy even for the purpose of making specific distinctions. The 

 whole and great difference must be based on their widely separated 

 mode of development or gemmation, which is easily recognized on 

 reference to figs. 1 a and 2a of PL XXV., and fig. 1 a of PI. XXIV. 

 accompanying this communication. 



In Allopora the corallum is aborescent, more or less massive, and 

 has the calices distributed irregularly throughout its surface, this 

 last character being likewise applicable to the mode of their first 

 appearance at the extremities of the branchlets. In Stylaster the 

 corallum is wanting in that massive and robust mode of growth 

 characteristic of Allopora, and the gemmation is invariably alternate 

 distal, as in Amphihelia, Lophohelia, and other Oculinidae, and 

 which, though sometimes disguised by the increase of the ccenen- 

 chyma in the basal portions of the corallum, is always apparent at 

 the growing termination. Even in the former parts the peculiar 

 primary mode of gemmation is betrayed by the more or less regular 

 disposition of the calices in a linear series on either side, rendered 

 sessile by the outgrowth of the ccenenchyma. This distinction 

 makes easy our appreciation of the, at first sight, somewhat obscure 

 characters of the form referred to the genus Allopora by Dr. Duncan, 

 to be presently referred to. In Allopora this latero-linear distri- 

 bution of the calices is altogether absent. 



Allopora nobilis, n. sp. 



Corallum arborescent, the main stem and branches very massive, 

 slightly flattened ; the branchlets subflabellate, thick, terminating 

 obtusely. Surface of the ccenenchyma smooth to the unassisted 

 eye, but presenting a delicate shagreened appearance when examined 

 with the pocket-lens. Calices densely distributed throughout the 

 surface of the corallum, slightly prominent ; very minute, scarcely 

 exceeding one-third of a line in diameter. Septa varying in number 

 from three to seven, more usually six ; their inner edges joining 

 laterally a little below the entrance to the calice, and forming minute 

 pit-like interseptal chambers, within each of which, as in Stylaster 

 erubescens (Pourtales), is enclosed a vertical fringe of small points 

 resembling hairs. Columella deeply immersed, cylindrical, its apex 



