FLCETZ TRAP FORMATIONS. 507 



foreign substances, and I believe in some rare instances, 

 passing; into a coarse or coinilonieratc^ pu(l(liny;stonc. It 

 sonietiniis breaks into large rhombic masses (thoii2;b in this 

 case it must be acknowledged that it contains little or no 

 clay.) and these, on account of a more compact texture, 

 retain tlicir Ibrm lonticr than similar juasses of the other. 

 The precipicis t'ormed l)y Ijotii are rrecpientlv lolly and per- 

 pendicular, but the projections and angles of the red are 

 more worn and rounded than those of the variety now under 

 consideration. The narrow deliles and ravines which tlie 

 streams of water have excavated, are less tortuous as well 

 as narrower, when made entirely in this rock, than in other 

 instances. Tlic springs of water which it affords are some- 

 times saline, but more comnuiidy free liom mineral impreg- 

 naiions than sucli as are found issuing front th(> otiier va- 

 riety, it sometimes consists of glittering cryslahine parti- 

 cles, but. does not in tiiis case appear to i)e a chemical de- 

 posit. In line, it appears under an almost endless variety 

 of characters, vvhicii it would be in vain to attempt to enu- 

 merate. Though not in%arial)ly distinguished by the pre- 

 sence of an argillaceous inuiredient. yet this is often tlie case, 

 and it is constantly found accompanying the beds of soft 

 clay slate or bituminous shale, whenever these occur. Whe- 

 ther coal accompanies these beds in the neii^hbourhood of 

 the Hocky Mountains, as it usually does in other |)laces. we 

 are unai)Ie to say. It is however certain that tliey contain 

 similar impressions of vegetables, and in other respects close- 

 ly reseml)le the bituminous shale of many coal districts. 

 Aliout the sources of the Canadian, this shale occurs in very 

 narrow horizontal lieds. and contains charred veiretable mat- 

 ter which could not readily be distinguished from connnon 

 charcoal. 



ll'this formation of sandstone, ronsi-^ling of the two va- 

 rieties just mentioned, ever extended across the valley of (he 

 Mississippi to thf* Alh-ghany Mountains, as some ini!>;ht be 

 disposed to imagine, I cannot pretend to determine what was 



