378 MR. F. RTJTLEV ON THE ORIGIN OF [Aug. 1 894, 



grains is confused and quite irregular, just as one finds it in ordinary 

 cases of cryptocrystalline structure. 1 



These general characters seem to agree perfectly well with the 

 descriptions given by Mr. Griswold, who usually speaks of the 

 rhomhohedral cavities as having been occupied by calcite. On 

 p. 188 of his work, however, he says that " the rhombohedral cavities 

 undoubtedly contained crystals of calcite or dolomite." 



Without entering at present upon the question whether the silica 

 of the Arkansas and Ouachita stones is to be regarded as quartz or 

 as chalcedony, but accepting for the moment Mr. Griswold's con- 

 clusion that it is the former, it will be best to quote some of the 

 passages in his book which refer more especially to the origin of 

 these rocks. Evidently he has felt it important to compare them 

 with chert, and the following extract from his work, 2 headed 

 ' Differences between Novaculite and Chert,' is therefore given in 

 full :— 



" Defining chert according to the weight of opinion as a crypto- 

 crystalline siliceous rock formed by chemical action, and containing 

 a large percentage of the silica in the cbalcedonic form, it is now 

 possible to state wherein the Arkansas novaculites differ from chert, 

 and to present the theory of their origin. Chemical analyses tend 

 to show that the novaculites of Arkansas have a purer siliceous 

 composition than cherts, though if the calcite had not been dissolved 

 from the Ouachita stone it could not have been distinguished from 

 chert. The tests of solubility of the silica show a decided difference 

 between novaculite and chert. Microscopic examination shows that 

 the soluble silica of chert is in the form of chalcedony, while novacu- 

 lite is entirely without silica in this form. 



" As a result, apparently, of this difference in the form of silica, 

 novaculite has a fine gritty feeling, while chert is more glassy. 

 Owing also to the purer and more homogeneous composition of 

 novaculite, this stone is more translucent than chert. Novaculite is 

 not a tough rock like chert, and breaks more easily, though its 

 couchoidal fracture is as perfect as that of any chert or flint." 



With reference to Arkansas stone, there is one fact which seems 

 to me to be of considerable importance, namely, that, under the 

 microscope, the structure of Arkansas stone very closely resembles 

 that of flint, in those portions of the latter which are free from 

 organic remains (PI. XIX. figs. 1 & 2). So close is this resem- 

 blance that when such a section of flint has been examined in 

 polarized light, between crossed nicols, and a section of Arkansas 

 stone is immediately substituted, no difference can be recognized 

 between the two sections. One may, indeed, say that Arkansas 

 stone and flint are practically identical in structure. The structure 

 of Ouachita stone is similar, but somewhat coarser in texture. 



1 For these and for other specimens of hone-stones I am indebted to my 

 friend Mr. Wm. Berrell, M.I.C.E., who procured them from Mr. T. Hazeon, of 

 10, Bishopsgate Avenue, E.C., the importer. 



2 Op. cit. p. 187. 



