Vol. 50.] CERTAIN NOVACTTLITES AND QTJARTZITES. 385 



stone appears to have a better right to be regarded as chalcedony 

 than as quartz. 



On p. 92 of his work, Mr. Griswold says : — " As the silica of 

 novaculite is neither crystalline, amorphous, nor chalcedonic, it must 

 be classed with those minerals which are structurally too fine to 

 show distinctive characters and are called cryptocrystalline. Thus 

 novaculite becomes a member of the same class of rocks with 

 chalcedony, flint, basanite or touchstone, chert, and jasper, a group 

 of rocks to which novaculite would, from its physical appearance, 

 seem naturally to belong." This statement appears to be true. It 

 may, however, be observed that, although this author says the silica 

 of novaculite is not chalcedonic, chalcedony is usually regarded as 

 one of the cryptocrystalline varieties of silica. It should, never- 

 theless, be noted that this evident slip of the pen is covered by the 

 concluding portion of the passage just quoted. Again, on p. 90 of 

 his work, Mr. Griswold states that " the remarkable fact demon- 

 strated by the analyses is the very constant, high percentage of silica. 

 Experiments on the solubility of the silica in caustic potash show 

 that it is almost entirely in the anhydrous form. There would, 

 therefore, be but little silica in the rock having the chalcedonic 

 structure, since chalcedonic silica is partly hydrous and pretty 

 soluble in caustic potash. Five per cent, is about the average 

 of soluble silica in the Arkansas stones as obtained by these tests. 

 This may seem rather high, but when it is considered that by 

 the same tests quartz-crystals, which are supposed to be practically 

 insoluble, give an average of about 4 per cent, of soluble silica, the 

 conclusion is reached that the silica of the novaculite is in a very 

 stable form. Two specimens of chert tested in the same way gave 

 over 30 per cent, of soluble matter, most of which was silica." 



The last statement is no doubt true, but cherts differ, and, in 

 one analysis by Mr. E. T. Hardman, only a trace of soluble silica was 

 present, while in another made by him, also of an Irish chert from 

 the Carboniferous Limestone, the amount of silica soluble in caustic 

 potash was merely 0*95 per cent. 1 



Mr. Griswold, on p. 91 of his work, quotes the opinions of 

 other writers, and among them the following : — 



" Mr. G. P. Merrill questions the theory of interlocking crystals to 

 account for the grit of the Hot Springs novaculite. He says that, 

 as a result of his own investigations, it is composed of ' a very fine 

 and compact mass of chalcedonic silica in which are embedded 

 widely-scattering angular grains of quartz '." " I agree with Mr. 

 Merrill, but, so far as the grit of the Ouachita stone is concerned, I 

 am inclined to think that the presence of minute grains of garnet 

 may have a certain influence. 



Mr. Griswold further says that " Dr. J. C. Branner, State Geolo- 

 gist, declares his disbelief in the theory that all the novaculites are 

 metamorphosed sandstones, and suggests that the ' compact varieties 



1 ' The Chemical Composition of Chert, etc' Sei. Trans. Roy. Dublin Soc. 

 vol. i. (new series) p. 90. 



2 ' Mineral Resources of the United States,' 1886, p. 589. 



