THECID^E AND HELIOPORID^E. 241 



useless to criticise his diagnosis of the genus in detail. I can 

 only say, therefore, that the structure of the type-species, T. 

 Swindernana, Goldf. sp., as above described, is such as to render 

 it apparently impossible that the forms ascribed to Thecia by 

 Dr Rominger can really be congeneric with it. In fact, Thecia 

 Swindernana does not agree in any important respect with 

 Thecia as defined by Dr Rominger, and especially differs in the 

 characters which I have above enumerated. If Dr Rominger, 

 therefore, has not been misled by the examination of specimens 

 greatly altered by fossilisation, it is clear that the species de- 

 scribed by him under the name of Thecia must belong to some 

 other generic type. Lastly, Dr Rominger asserts that Prot- 

 arcza, E. and H., possesses corallites which have mural pores, 

 and are provided with convex tabulae, and that it is a near ally 

 of Thecia. On this point, I can only say that I have failed to 

 discover any traces of tabulae in Protarcea, either in actual 

 specimens or in thin vertical sections ; that though the walls 

 are porous, there are no " mural pores," properly so called, in 

 any examples I have seen ; and that I am unable, therefore, to 

 accept the view that any alliance exists between this genus and 

 Thecia. 



So far as at present known, the gen-us Thecia is confined in 

 its geological range to the Upper Silurian rocks, the type- 

 species, T. Swindernana, Goldf., being an abundant and charac- 

 teristic fossil of the Wenlock Limestone of Britain and Sweden. 



HELIOPORID^E (Moseley). 



This well-defined and distinctly circumscribed group of corals 

 is characterised by the possession of a corallum composed of 

 two distinct sets of corallites, tenanted in life by two distinct 

 sets of zooids. The larger tubes are uniformly distributed 

 among a very much larger number of much smaller ones, and 

 are provided with delicate lamellar septa (" pseudo-septa"), 

 which are formed by infoldings of the wall, are typically twelve 

 in number, and are occasionally rudimentary or even obsolete. 



Q 



