194 PROCEEDINGS OP THE GEOLOGICAL SOCIETY. [Jan. 10, 



the one represented in fig. 1, Plate XI Y. In this instance there are 

 two layers belonging to a single grannie. The innermost layer is 

 for the n?ost part perfectly compact and indefinitely fibrons ; and, 

 although not demonstrable, it cannot be considered other than a 

 modified form of the pertaining serpentine : the outer layer consists 

 of distinctly individualized fibres or aciculi, generally thick, occa- 

 sionally very fine, and separated from one another on the left half 

 (that is, of the figure), but closely adherent by their sides on the 

 right half, except at their points. The innermost layer is in one place 

 separately acicular, like the outermost one, showing clearly the 

 intimate relationship between the two extremes of structure assumed 

 by the ^' proper wall"*. 



Now such cases as this have been represented as " exactly cor- 

 responding " with the fine tubulation characteristic of the shells of 

 nummuline foraminifers ; but from what has already been adduced, 

 and other evidences to be brought forward, we, on the contrary, 

 contend for their being merely simulations of such tubulation. 



But this case is not to be compared with others we are acquainted 

 with. We have seen in Grenville Ophite some truly striking examples, 

 which show the aciculi of the "proper wall" far surpassing in their 

 mimicry of nummuline tubulation anything hitherto published — 

 cylindrical rods, uniform in length, thickness, and in their distance 

 from one another, and standing perpendicularly to the surface of the 

 ** chamber- casts." At first sight these instances, if studied by them- 

 selves, might be taken for " internal casts " of tubules ; but such a 

 view speedily vanishes when they are properly examined and studied 

 in all their relations. Besides being perpendicular, the rods are 

 occasionally seen leaning, arching, creeping, and twisting about in 

 all directions : they also unite wholly or partially, converging, 

 diverging, branching, and anastomosing, so as to produce conical, 

 cuneiform, dendritic, " brush-like," and an infinite variety of " bundles, 

 of which no two are precisely similar." Other aggregations occur, 

 forming small or wide-spreading patches, which have uniform, or 

 variously disrupted, surfaces. 



Whether viewed transversely or longitudinally, all these modifi- 

 cations are structurally alike. If examined with a magnifying 

 power of 350, the patches are seen to consist of " closely packed " 

 parallel fibres — in other words, to have a " uniform surface so close 

 in texture as to be with difficulty resolvable into the points of their 

 constituent aciculi "f. Place the bundles under the same object- 



* This is a case inexplicable on the idea that it has resulted from pseudopo- 

 dial tubulation ; but there is no difficulty in conceiving it to be one of double 

 allomorphism. The outer layer, which we assume to have been formed first, 

 has lost its finely fibrous structure, the fibres having become individualized or 

 transformed into aciculi, — a change characteristic of chrysotile. As the " proper 

 wall " of NummuUtes is formed of a number of laminse, it may be thought that 

 this double layer is a parallel case ; but the tubules, belonging to the former part, 

 pass uninterruptedly or continuously through the laminae ; whereas in the latter 

 each of the two layers is composed of an independent series of fibres, or aciculi. 



t It is quite evident Dr. Carpenter has observed examples similar to ours. 

 See ' Intellectual Observer,' vol. vii. p. 292, &c. 



