202 PEOF. H. A. NICHOLSON AND DE, J. MTJRIE OTf THE 



palaeontologists can no longer refuse to recognize and take into 

 account the possibility of the replacement of silica by carbonate 

 of lime. In this particular case, however, we should be compelled 

 to believe that this certainly unusual conversion of a siliceous 

 skeleton into a calcareous one had taken place just in tlie very 

 beds in which the fossils, as a rule, show no alteration whatever — 

 beds like the Wenlock Limestone of Britain and Sweden, and the 

 Trenton and Niagara Limestones and the Hamilton formation of 

 North America. On the other hand, when we meet with Stroma- 

 toporoids with a siliceous skeleton, we shiould have to believe 

 that this was their original constitution, and we sliould have to 

 account for the fact that these siliceous specimens are almost, if 

 not quite, exclusively found in strata in which the unquestionably 

 calcareous corals and Brachiopods are for the most part also sili- 

 ceous. The only other hypothesis, namely that the Stromatopo- 

 roids (like the Foraminifera) are sometimes calcareous and some- 

 times siliceous, we were at first not indisposed to accept ; but the 

 total absence of any composition out of distinct sand-grains in 

 the silicified specimens, and the existence of specimens partly 

 silicified and partly calcareous, have induced us to abandon this 

 view. This view, however, might be easily entertained if due care 

 were not taken to separate specimens showing the siliceous infil- 

 ling of the sarcode-chambers from those in which the skeleton 

 itself has been silicified. 



MiKUTE Stucture of the Stromatoporoids. 



In studying the minute structure of any Stromatoporoid, it is 

 necessary to make sections in two directions — namely, vertically or 

 radially, at right angles to the concentric laminae of the mass, and 

 liorizontally or tangentially, or parallel with the concentric 

 laminae *. Sections of the first order are easily prepared ; but 

 sections of the second order present greater difficulties, as the 

 concentric laminae are invariably more or less curved or undu- 



^ Owing to the very yariable form of the colonies of the Stromatoporoids, the 

 terms vertical and horizontal can hardly be used with propriety to designate the 

 two kinds of sections above referred to, unless they are understood simply as 

 expressing the relation of the sections to the concentric laminae of the fossil. 

 We prefer, therefore, to use the terms " radial" and tangential" the former 

 indicating all sections which are parallel with the " radial pillars," and which 

 cut the concentric laminae at right angles, whereas by the latter we understand 

 all sections which are approximately parallel with the concentric laminge, what- 

 ever may be the relation between the planes of these sections and the plane of 

 the general mass from which they are taken. 



