418 ON THE GEOLOGICAL RELATIONS 



esteemed a pure and proper mineral substance. It is 

 not carbon under its semimetallic, and therefore more 

 obviously mineral form of plumbago, at least; yet 

 even this might be the produce of vegetables, since it 

 is daily formed from coal in the Iron-furnace. If 

 Anthracite is not to rank with vegetable coal because 

 it contains no bitumen, this is equally true of un- 

 doubted vegetable coals, in Pennsylvania and else- 

 where. And therefore the early anthracite may be, 

 and probably is, an organic produce, and the record 

 of an antient vegetable creation ; while its very rarity 

 is an argument in favour of this supposition. 



Calcareous concave bodies, precisely resembling 

 fragments of shells, occur in hornblende schist 

 in Scotland, in Glen Tilt, not distinguishable 

 from those of the basaltic shales : had they been 

 found in a modern shale, no question would have 

 been entertained as to their nature. This is the 

 more remarkable ; as, under no other circumstance, 

 does lime, as a visible ingredient, occur in this cry- 

 stalline rock of fusion. I have proved that hornblende 

 schist is a fused slate ; and thus might it as well con- 

 tain shells, under that fusion, as the Lydian stones, 

 equally fused or semifused, so often do. This is a 

 second evidence of an early creation, the first, of ve- 

 getables, and the second of animals ; the one supports 

 the other, and they are confirmed still further by a 

 third fact. These are, the existence of orthoceratites, 

 abundant in a quartz rock in Sutherland, which follows 

 gneiss and is connected with it. And I consider 

 this evidence to be as perfect as we are entitled to 

 expect on such a subject, from the researches of one 

 individual, under the neglect of all other observers. In 

 any case, where no previous hypothesis to the con- 

 trary had existed, no one would hesitate in receiving it, 



