MTNTJTE STRUCTURE OF STROMATOPORA AND ITS ALLIES. 219 



are often placed at special points, and may correspond with the 

 " exhalant pores " or " oscula " of Sponges. Both also exhibit 

 the radiating surface-grooves and canals which are present in the 

 so-called Coenostromce. Similar radiating canals are present in 

 S. {Ccenostromd) discoidea^ Lonsd., S. granulata^ Nich., and 

 tuherculata, Nich., though in these instances they appear to be 

 confined to the superficial layer, and to be mere subdermal 

 grooves. In other cases no such structure can be detected. In 

 S. ostiolata, Nich., there are well-developed and regularly dis- 

 posed apertures of large size (" oscula " ?). In some cases (e. g. 

 S. nodulata, Nich., and 8. nuUiporoides, Nich.) the surface is 

 smooth ; but it is generally covered with miliary granules, 

 tubercles, or vermicular ridges. 



The geological range of Stromatopora proper, so far as our ow^n 

 personal observations go, is Upper Silurian and Devonian. 



(h) Caunopora, Phillips. 

 (' Palaeozoic Fossils of Cornwall,' &c., 1841.) 



Skeleton (" sarcodeme ") massive or expanded, its general 

 structure very similar to thac of Stromatopora proper, but the 

 concentric lamina) minutely undulated and inflected, and the 

 radial pillars often more or less rudimentary (as in Clathro- 

 dictyon). The reticulated skeleton is perforated by large tubes 

 half to one line in diameter), which have strong calcareous walls, 

 are open throughout, or crossed by a few remote horizontal parti- 

 tions, are sometimes connected with one another by lateral tubes, 

 are in the main perpendicular to the concentric lamina), and 

 finally open on the surface in the form of large rounded apertures. 



Type. — Caunopora placenta, Phill. 



In this section we must also place C. {Stromatopora) perforata, 

 Nich., from the Corniferous Limestone of North America. The 

 chief point by which this type is distinguished from Stromatopora 

 proper is the presence of vertical tubes, with definite walls, 

 perforating the entire mass. As has been previously stated, the 

 microscopic examination of the only Devonshire example of 

 Phillips's C. placenta at our disposal does not allow us at present 

 to accept Ferd. Roomer's suggestion that the genus is founded 

 upon specimens of Stromatopora which have enveloped a coral *. 



* As we have before remarked, we are now in possession of a fine series 

 of specimens of Caimojpora from the Devonian limestone of Devonsliirc, wliieli 

 seem to render it certain that the tubes of this form arc ideally part and parcel 

 of the Stromatoporoid itself, and that they are not due to the fact that the latter 

 has simply enveloped a coral of the type of Syringoijora. 



LINN. JOURN. — ZOOLOaY, YOL. XIY. IG 



