266 ACCOUNT OF THE MINARALOGY 



ing immediately overlaid by the brick-coloured trap-tuff, pass- 

 ing into the superincumbent bed. At Eide and Waaij, the 

 appearances differed from those of Nalsoe. They did not oc- 

 cupy the surface, but were dispersed through the bed, at dif- 

 ferent elevations, yet all parallel, as if it had been formed of a 

 multitude of partial flowings, each moulding on the consolida- 

 ted surface of its predecessor, and extending, as the supply of 

 fluid matter permitted ; and, so far as we observed, this was 

 always in the direction of the declivity of the hill or cliff on 

 which we perceived them. 



By the regularity of the folds and wrinkles which characte- 

 rize their surfaces, and their usual horizontal position, they 

 seem to have flowed undisturbed j and although we never met 

 with any very extensive surface exposed at one place, in con- 

 sequence of finding them always on the declivity of a hill, yet 

 we had the most perfect evidence of their passing under the 

 incumbent rocks, but to an extent of which we could of course 

 form no estimation. There was enough, however, seen, to 

 mark it as a fact of a general nature, and one which, more than, 

 any other I have ever met with, denotes the origin of trap- 

 rocks. 



There is a wide difference between ascertaining the agency 

 by which a rock has been formed, and the manner in which 

 that agent has been applied. It is an inquiry perhaps of no 

 great utility, but, in prosecuting the one, the other naturally 

 presents itself, and the inquisitive mind cannot be expected to 

 stop short, particularly at a point, where, for the first time, it 

 has met with data quite unequivocal. Here, however, the multi- 

 plicity of proof in favour of the former, throws a difficulty of great 

 magnitude in the way of the latter. Had each bed been the 

 operation of a volcanic eruption, the appearances of fusion 



would 



