quartz rock wherever occurring is generically the same, and that it differs no more from itself than the numerous varieties of gneiss, of mica slate, or of graywacke are found to do. It ought therefore to be designated in a geological system by one general name, since, as the designations of rocks are intended for the purposes of geology, rather than of mineralogy, it is preferable that one common term should be applied to the rock in question, than that the several varieties should each receive a separate one, a circumstance tending as much to confuse geological reasonings, as to render them tedious and intricate. It has already received various names, out of which it would be desirable to agree on one. It has been called granular quartz, transition sandstone, quartzose graywacke, and quartz rock. The first of these sufficiently expresses the character of one of its varieties; but its particularity excluding those modifications which are not granular, as well as those which contain felspar, is not well adapted for the purposes of geological description. Could we prove that it was in every case a rock of transition, and that the theory which the term transition implies was well founded, the name of transition sandstone would perhaps be the most applicable. But as I have ascertained, that it alternates at times with rocks, which the system here alluded to calls primitive, and as it is at all times desirable to keep clear of those terms which involve an hypothesis, I think we are bound to reject this name. The same argument is valid against the appellation of quartzose graywacke, a term in other respects perfectly inapplicable, since the very definition of graywacke decidedly excludes it; and there can be no greater evil than to confound under one name, substances of different qualities, particularly when that name itself, from its connection with an hypothesis, tends to blind our judgment and pervert our reasonings. I should therefore suggest the superior propriety of the term quartzy or quartz rock, a term involving no hypothesis, and which, at the same time that it is sufficiently general to include
