164 Dr. H. A. Nicholson and Mr. R. Etheridge on 



corallites are very generally filled up with the matrix, they 

 are commonly destroyed, along with the tabulas, in the exterior 

 parts of the corallum. They can, however, be always readily 

 detected in thin sections taken across the corallites, provided 

 these are not quite superficial and are directly transverse to the 

 long axes of the tubes. The number of the septa (fig. 5), 

 varies, but is most generally four ; sometimes only three 

 or two are present, occasionally only one, and rarely as many 

 as five. In form the septa are quite short, extending only a 

 very small distance towards the centre of the visceral chamber, 

 attenuated towards their inner edges, but much thickened 

 towards their bases. This structure, along with an apparent 

 plication of the wall to correspond, gives to the transverse 

 section of the corallites a characteristically petaloid form 

 (figs, a, h)^ which sometimes alters to a reniform, rhomboidal, 

 or trefoil figure. The walls are thick and devoid of mural 

 pores, exhibiting the curious peculiarity that, as seen in trans- 

 parent transverse sections, they are composed of crystalline 

 calcite instead of granular carbonate of lime as is ordinarily 

 the case. Whether this is due to some secondary alteration 

 to which the specimens have been subjected, or not, it is 

 difficult to say ; but there is no ground for supposing that 

 these, any more than the other fossils in the same beds, have 

 suffered from any changes of a metamorphic nature. The 

 corallites are very long and basaltiform, increasing by fission, 

 and often showing a concentric arrangement round secondary 

 centres. As a general rule, the corallites do not appear to 

 grow to the same length ; so that the surface of the corallum, 

 instead of being uniformly rounded, is often very uneven, in 

 consequence of the unequal development of neighbouring groups 

 of tubes. It is also not uncommon for the secondary centres 

 of development above spoken of to become tubular, whether 

 by the absorption of a central corallite or otherwise, giving 

 rise in transverse sections to approximately circular hollow 

 spaces (fig. a), which become filled up with the matrix or 

 with calcite. . The tabulae (fig. c) are exceedingly well 

 developed and quite complete, averaging three or four in the 

 space of 1 line. No general epitheca, so far as we have been 

 able to observe, is present. The corallites, lastly, though 

 closely contiguous, are not amalgamated by their walls ; so 

 that a rough fracture exposes the exterior of the walls, and 

 not the interior of the visceral chambers. 



Form, and Loc. Common in the Cincinnati group of 

 Waynesville, Ohio, and the Hudson-river formation of the 

 river Credit and of Manitouwaning, Canada. 



Collected by, and in the cabinet of. Prof. Nicholson. 



