380 MR. F. RTTTLET ON THE ORIGIN OF [Aug. 1 894, 



In other cases in the same section I have obtained quite irregular 

 results. Such differences are doubtless due to differences in the 

 orientation of grains of the same character. 



As already mentioned, little if any distinction can be made be- 

 tween the most chalcedony-like variety of Arkansas stone and those 

 portions of a flint which exhibit no traces of fossil organisms. 

 From this extreme it seems probable that gradations may be traced 

 to coarser structural conditions, in which Ouachita stone represents 

 an intermediate and quartzite an extreme phase of coarseness. 

 This is partially shown on PI. XIX. figs. 1, 7, & 8. Pig. 7 

 represents the thin edge of a section of a pebble from a conglomerate 

 from Purtiall, in the Deccan. Except on the margin, this section 

 exhibits what may be termed a cryptocrystalline structure on a 

 large scale, while on the mai'gin it shows a microcrystalline struc- 

 ture on a large scale, as in fig. 7. In this case the cryptocrystalline 

 aspect is evidently due to the overlapping of crystals, while on the 

 margin the section is only thick enough to include a film of juxta- 

 posed, but not superposed, crystals. Pig. 8 is drawn from a section 

 of a quartzite from Xondweni, Zululand, and represents what we 

 may, for the time being, regard as the extreme phase of coarseness 

 in this series. 



I speak of these rocks as constituting a series, because I believe 

 that they have all had a common origin, that they are all siliceous 

 replacements of limestones. It is a generally recognized and, I 

 think, an incontrovertible fact that a large proportion of quartzites 

 are more or less altered sandstones, as clearly demonstrated by 

 Irving and Van Hise. 1 



The former of these authors, when speaking of the genesis of the 

 Huronian quartzites, stated that " all the true quartzites of the 

 Huronian are merely sandstones which have received various degrees 

 of induration by the interstitial deposition of a siliceous cement, 

 which has generally taken the form of enlargements of the original 

 quartz-particles, less commonly of minute independently oriented 

 areas, and still less commonly of chalcedonic or amorphous silica : two, 

 or even all, of the three forms occurring at times in the same rock.'' 2 

 Pully admitting the truth of this statement, which deals only 

 with 'true quartzites,' I am, nevertheless, inclined to think that it 

 is needful, not on mineralogical, but on genetic grounds, to divide 

 the rocks which might be generally termed ' quartzites ' into two 

 groups, including in the one group the indurated sandstones or true 

 quartzites, which might, for the sake of distinction, be termed 

 ' detrital quartzites ' ; in the other siliceous replacements of lime- 

 stones which, at times, may simulate detrital quartzites both in 

 mineralogical and structural characters. These might be termed 

 ' infiltration or metasomatic quartzites.' 



The necessity for such a classification I hope to render more 

 apparent by dealing at some length with the rhombohedral cavities 



1 ' On Secondary Enlargements of Mineral Fragments in certain Rocks/ 

 Bull. U.S. Geol. Surv. No. 8, vol. ii. (1884). 2 Ibid. p. 48. 



