48 BRITISH STROMATOPOROIDS. 



On the other hand, in the genus Stromatopora, Goldf., where the coenosteum is 

 generally developed in " latilamina3," and where the radial pillars are so conjoined 

 with their horizontal arms as to give rise to a continuously reticulated skeleton, the 

 interlaminar spaces, as such, can hardly be said to exist. They are, in fact, repre- 

 sented only by the irregular branches of communication between adjoining zooidal 

 tubes (Plate V figs. 11, 15 and 17). Hence, in these forms the vitality of 

 the colony must at any given moment have been confined to the last formed 

 latilamina. 



In the Labechiidce, again, it is difficult to arrive at any certain conclusions as to 

 the true condition of the interlaminar spaces. If we regard the horizontal 

 processes or "arms" given out by the radial pillars of Labechia conferta, Lonsd., 

 as being actual plates, then there are no true interlaminar spaces. In place of inter- 

 laminar spaces, we should have a series of oblong or lenticular cells, occupying all 

 the intervals between the pillars, and resembling the intertubular tissue of 

 Plasmopora or the vesicular tissue of Cystiphyllum. If, on the other hand, we 

 consider the horizontal connecting-processes of the radial pillars of Labechia (Fig. 5, 



Fig. 5. 



r f 



Fig. 5.— Sections of Labechia conferta, Lonsd., enlarged twelve times. Wenlock Limestone, Benthall. 

 A. Tangential section. B. Vertical section, p p. Radial pillars, e c. Connecting-processes. 



c c) to be mere calcareous filaments, such as we see in Actino stroma clathratum, 

 JN T ich., then the interlaminar spaces of L. confert a, Lonsd., are represented by a loose 

 network of irregular, intercommunicating cellular cavities. Thin sections, unfortu- 

 nately, are not conclusive on this point. The balance of evidence at present, 

 derived both from thin sections and also from observations of the true surface or 

 of fractured surfaces of Labechia, would seem to me to be in favour of the view that 

 the connecting-processes in this genus are really in the form of tabular plates. If 

 this view be correct, then there are no interlaminar spaces, strictly so-called, in 

 Labechia, the condition of parts being very much what we find to exist in various 

 " tabulate " Corals, such as Fistulipora or Plasmopora, except that the large tubes 

 of the latter types are represented in Labechia by the " radial pillars." If this view 

 be correct, it would follow further that only the very last formed layer of the 



