THE NATURE OF CAUNOPORA. Ill 



but quite definite proper wall, either alone or with but a very thin secondary 

 lining. More commonly, the proper wall is strengthened by a dense secondary 

 deposit of light-coloured sclerenchyma, which may nearly obliterate its internal 

 cavity (Plate X, fig. 2). The tubes are attached inferiorly to irregular horizontal 

 stolons, which sometimes clearly have a proper wall, but which at other times seem 

 to be bounded only by the general tissue of the Stromatoporoid. The tubes 

 further give out lateral horizontal tubes, which may simply open into adjoining 

 tubes, or which may ultimately bend upwards and give origin to new vertical tubes. 

 Superficially, the " Caunopora-tubes " terminate in rounded thickened apertures, 

 which are flush with the general surface, or project very slightly above it. In the 

 few specimens in which the tubes appear to be prolonged above the surface at all, 

 it is probable that the fossil has been partially decorticated ; but the horizontal 

 connecting-processes certainly seem to occasionally lie above the last-formed layer 

 of the Stromatoporoid (Plate X, fig. 3). As to whether or not there exists any 

 communication between the " Caunopora-tubes " and the interlaminar spaces and 

 zooidal tubes of the enveloping Stromatoporoid, it seems impossible to arrive at 

 present at any absolutely positive conclusion. In most cases there certainly seems 

 to be no such communication. On the other hand, thin sections occasionally show 

 phenomena which would lead to the belief that the horizontal connecting-tubes may 

 open into the adjoining zooidal tubes, or that the main " Caunopora-tubes " them- 

 selves sometimes open inferiorly into the interlaminar spaces of the Stromatoporoid ; 

 but it is probable that the phenomena in question are delusive. 



As regards their internal structure, the " Caunopora-tubes " are probably always 

 tabulate. It is true that in a number of specimens " tabula? " cannot be detected, 

 but this is probably the result of mineralisation, as I have rarely failed to detect 

 these structures in well-preserved examples of all the forms of " Caunopora." The 

 tabular may be simply horizontal, or curved, but they are more commonly partially 

 funnel-shaped, a number of vesicular tabulae being placed on one side or on both sides 

 of the tube. Hence cross-sections of the tubes present appearances almost exactly 

 similar to those seen in corresponding sections of the corallites of Syringopora, and 

 sometimes in similar sections of Aulopora. Very often, the same " Caunopora- 

 tube " is partially furnished with flat tabula?, and partially with funnel-shaped 

 tabula?. As a general rule I have failed to detect the existence of septal spines in 

 the " Caunopora-tubes ;" and I am not aware that these structures have been 

 clearly recognised as occurring in any instance by previous observers. I have 

 never found any satisfactory indications of septa in any of the " Cannoporai " of the 

 British Devonian Rocks, but it may well be that this is the result of the extent to 

 which most of the Devonshire specimens have been altered by fossilisation. I have 

 also not succeeded in detecting such structures in a large number of specimens 

 from the Devonian Rocks of Germany, where the minute structure is very well 



