24 BRITISH FOSSIL CORALS 



" Corallum very dense, and not bearing columnar processes, as in the preceding genus. 

 Calices polygonal, columella styliform, not projecting much. No pali. Septa thick, 

 apparently eight or ten systems, two or four of the secondary septa being as much 

 developed as the six primary ones. Walls thick and united as in Sti/locoenia." 



]\I. de Fromentel separated the genera Astrocoenia and 8tyloccenia, and retained the 

 latter amongst the Eusmilince aggregatm. There was no reference made, therefore, in his 

 generic diagnosis of Aslroccenia to the genus Stylocoenia. M. de Eromentel's descrip- 

 tion of the generic peculiarities of Astrocosnia are as follows : " Corallum massive, com- 

 posed of corallites united by their walls, which are prismatic in shape ; the calices are 

 polygonal ; the columella is styliform, and more or less projecting ; the septa are tolerably 

 thick, are few in number, and are dentate, especially near the columella ; there are no pali." 



Whilst investigating the Madreporaria of the Maltese rocks in 1865, I found that 

 the septa of the common species Sfgloccenia lolato-rotundata, Mich, sp., were dentate.^ 

 The species occurs also in the Chert of Antigua, and presents there the usual plain septa 

 considered to mark the family of the genus. If fossilization can remove the dentations 

 of the septa of one Stylocvenian^ it can do so in others, and it may be safely asserted 

 that all the Stglocoenians had dentate septa. 



This dentate condition of the septa brings the genera Astrocoenia and Stglocosnia 

 together again, although it removes them from the Eusmilina into the Astraacecs. 



MM. Milne-Edwards and J. Haime's generic description can thus stand, and its 

 concluding sentence respecting the thick walls of the genus which was omitted by M. 

 de Fromentel is very important. 



In some species, as in A. jmlchella, Ed. & H., the calices are so wide apart in some 

 specimens, and in certain spots in all the specimens, that there is evidently here and 

 there a coenenchyma between the walls of the corallites. The surface of the ccenenchyma, 

 which appears to arise from an hypertrophied condition of the adjacent corallite walls, is 

 usually ornamented either with prolongations of the costge, or with small papillose granules. 

 This is observed in other species, and it is noticed that the amount of coenenchyma 

 varies according to the shape of the corallum, and the rapidity of the multiplication of 

 the corallites. The presence of scattered granules, or of small papillae on the ccenenchymal 

 surface, and between the external terminations of the costae, is observed in some speci- 

 mens of a species, and not in others ; but the costse, although they may extend far over 

 the inter-calicular spaces (or, in other words, over the surface of the coenenchyma), never 

 unite, and run into those of adjoining corallites. There are modifications in the length and 

 straightness of thecosta), and where there is no coenenchyma, and the walls of the corallites 

 are thin, they may be so reduced in size as to appear to be simple terminations of septa. 



In many species the ccenenchyma, when non-costulated, and not ornamented with 

 granules, becomes slightly ridged, and foreshadows the condition which peculiarises the 

 genus Stylocoenia. 



' ' Ann. Mag. Nat. Hist.,' April, 1865. 



