138 



THE GENUS MONTICULIPORA. 



of the walls of the corallites, not the slightest trace of any line of 

 demarcation between adjoining tubes being recognisable. The 

 corallites are as a whole very uniform in size, and are polygonal 

 or sub-polygonal in shape. Small corallites, in limited number, 

 are usually intercalated among the larger tubes ; but the amount 

 of these varies even in different parts of the same section, and 

 as they show no peculiarities of structure, they are apparently 

 merely young tubes. At almost all the angles of junction of 

 the large tubes, however, and commonly in the thickness of 

 their lateral walls, are developed large hollow spines (" spini- 



form corallites"), which exhibit 

 a central dark or light space 

 surrounded by a conspicuous 

 thickened wall (fig. 24, and PI. 

 I. fig. 1(5). The upper ends of 

 these peculiar structures pro- 

 ject as intracalicine spines upon 

 the surface, and I have not 

 been able to detect that they 

 are open above ; though they 

 Fig. 24.— Portion of a tangential section of are clearly hollow, and they 



Montlciilipora vioniliforntis, Nich., taken r u A ' \ 



just below the surface, showing the in- Can Olten be traCCCl Ul lOll- 



tracalicine spines, enlarged fifty times, aitudinal SCCtionS aS distinct 



From the Hamilton Group of Ontario. *-' 



tubular cavities in the axis of 

 the thickened wall of the corallites (see fig. 3, b). 



Longitudinal sections (PI. I. fig. \c) show that the corallites 

 in the axis of the branches are thin-walled, with comparatively 

 few tabulce. In the outer part of their course, however, where 

 they bend outwards to reach the surface, they become thickened, 

 and the walls assume the peculiar fibrous aspect characteristic 

 of all those MonticuliporcB in which a complete amalgamation 

 of the corallites occurs. In this part of their course, also, the 

 tabulae increase considerably in number ; but they are never 

 very closely arranged, and they are always complete. 



The only species known to me with w^iich the present 

 form would be likely to be confounded is the ]\I. Barrandi, 



