1879]. 167 [Crosby. 
combination with the sulphates of-lime and magnesia, forming rather 
anomalous compounds which are decomposed only when the water is 
boiled nearly to dryness. The sulphate of lime is particularly efficient 
in thus locking up the free carbonic acid, and since we know that this 
salt was far more abundant in the older seas, it seems reasonable to 
conclude that it may have had an appreciable effect in preventing 
the greater volumes of carbonic acid in early days from decomposing 
the feldspathic sediments of the deep sea. 
The free silica or quartz so generally present in the pe'rosilicious 
rocks would, according to the view here advocated, have two prin- 
cipal sources,— the decomposing silicate minerals, and the silicious 
organisms always occurring in the abyssal deposits. 
These rocks, which are widely distributed over the globe and com- 
pose formations of great extent, are undoubtedly of marine origin. 
We can scarcely regard them as shore deposits, and therefore it 
seems natural and legitimate to conclude that they were formed in the 
deep sea; and I would submit that they are very fairly represented 
by the modern abyssal accumulations, especially if we take into ac- 
count the enormous period of time which has elapsed since their 
formation and the probable changes in the physics and chemistry of 
the sea which it has wrought. 
The petrosilicious rocks are often distinctly and beautifully banded. 
This structure usually results from the alternation of very thin and 
regular quartzose and feldspathic layers, and although doubtless orig- 
inating in, and determined in direction by, the sedimentary process, I 
think it can be proved that it has been made much sharper and more 
definite by a subsequent partial segregation of the ingredients, espec- 
ially the silica. In some cases these quartzose layers reach the con- 
dition of jasper, or even of vitreous crystalline quartz, and then their 
concretionary nature is very plain. In the vicinity of Boston the in- 
dividual bands or layers of the petrosilex vary from an almost micro- 
scopic thinness to nearly an inch thick; and there is a gradual pas- 
sage from banding of agate-like regularity and evenness, where single 
layers are continuous for yards, to that in which the quartzose layers 
are reduced to irregular lenticular sheets only a few times longer than 
thick, the perfect lamination changing to a distinct schistosity, and 
this to a structure where the layers become extremely irregular, end- 
ing abruptly or anastomosing and dividing in a systemless manner 
until the rock becomes a sort of quartzose reticulation having the 
meshes filled with feldspathic material. True globular coneretions or 
nodules, mainly of silex, are also common in our petrosilicious rocks. 
ee 
$3 =~ oo a ae ee 
=) 
