56 BRITISH STROMATOPOROIDS. 



diverse affinities, that they cannot, in my opinion, be employed with any advantage 

 as constituting by their presence a generic character. Hence I have not thought 

 it expedient to retain Winchell's genus Coenostroma, in the definition of which the 

 presence of astrorhizse is taken as the essential character. On the other hand, I 

 cannot agree with Prof. Ferd. Roemer (' Leth. Pal.,' p. 532) in thinking that they are 

 of quite variable occurrence, and that they have not even a specific value. My 

 experience is that the astrorhizse are, in general, quite constant in their absence or 

 presence, and also in their characters when present, in types which can be other- 

 wise shown to belong to the same species ; and that they can, therefore, be used as 

 marks of specific distinction. It must be admitted, however, that there are 

 Stromatoporoids which are otherwise very similar to one another in general 

 structure, but which in some cases possess astrorhizae, whereas at other times they 

 appear to be without these structures. In such cases, all that can at present be 

 said, is that a careful and extended series of microscopic observations will be needed, 

 before we can assert positively that such types are not distinguishable by any other 

 characters than the presence or absence of astrorhizal canals. 



(V7) Astrorhizal Tabulce. — In Stromatopora? dartingtonensis, Cart., Mr. Carter has 

 described transverse calcareous partitions as developed in the astrorhizal canals, 

 which in this particular type are usually of large size (' Annals and Mag. Nat. 



Fig. 7. 



Fia. 7. — Vertical section through the centre of one of the astrorhizse of Stroma- 

 toporella eifeliensis, Nich., enlarged twelve times, showing the central canal of the 

 astrorhiza and the numerous " astrorhizal tabula3 " in the larger radiating canals. 

 Middle Devonian ; Gerolstein, Eif'el. 



Hist.,' ser. 5, vol. vi, p. 339). The same observer has also compared these 

 transverse partitions with the " tabulas " in the zooidal tubes of Millepora. As the 

 astrorhizal canals appear to me to be in no way homologous with the zooidal tubes 

 of Millepora, and as many Stromatoporoids have truly tabulate zooidal tubes, I 



