the Neighbourhood of Gamrie, Banffshire. 143 



a coarse, reddish-brown ferruginous and slightly micaceous sand. The sandstone underlying the 

 clay is coarse, micaceous, and of a deep red colour, as is likewise the matrix of the underlying 

 conglomerates. The rolled pebbles imbedded in them are commonly not larger than a marble, 

 and consist principally of quartz and clay slate. The bed No. 7. of the first section, is, however, 

 less compact, and more resembles No. 1. Below No. 7. (see section with wood-cut) the conglo- 

 merates gradually become coarser and harder, and inclose, with the pebbles of the schistose rocks, 

 a few rolled masses of gneiss. Only a small portion, however, of the lower beds can be seen in 

 either of the above-mentioned places, as the old red sandstone comes to the surface at a short 

 distance from the outcrop of the Ichthyolite beds (PI. X. Map, and Section No. 3.). The sand- 

 stone, at first sight, seems to rise from beneath the conglomerates, but is in reality brought into 

 juxtaposition by a fault. Owing to the soft and friable nature of its upper beds, its presence is at 

 once indicated, by the grassy slope of the sides of the ravines, instead of the bold and rugged 

 escarpment presented by the conglomerates. The latter may be seen in fuller development in the 

 chffs between Gamrie and Crovie, which rise to the height of about 400 feet. Near Crovie they 

 rest unconformably upon the lower beds of the old red sandstone (PI. X. No. 2.), and dip at an 

 angle of about 40° to the S.S.E., or from the face of the cliff as exhibited in the section ; but 

 upon approaching the fault, the strata rise rapidly westward or towards it. The beds are here of 

 great thickness ; and differ considerably from the upper beds in mineralogical composition, as 

 well as in the size of the inclosed, rolled pebbles. The matrix is generally siliceous and extremely 

 hard, occasionally of dark grey, and at other times of a white or red colour. In some of the 

 lower strata it occasionally consists of a white, crystallized carbonate of lime. White quartz 

 pebbles predominate in some beds, whilst the rounded fragments of granite, gneiss, micaceous and 

 argillaceous schists and hornblende rock, are mostly of the size of a cannon-ball, but sometimes 

 much larger ; a few which I measured being from three to four feet in circumference. Rolled 

 masses of a fine-grained, red and variegated sandstone, are also occasionally found in the lower 

 part of the series. This phsenomenon, together with the character of the cement of the lower 

 beds, would indicate that the matrix of this part of the red conglomerates, had been principally 

 derived from the destruction of the red sandstone upon which it reposes. 



Such are the general lithological characters of the deposit. There can be 

 no doubt, I should imagine, that the clay and shales inclosing the Ichthyo- 

 lites* belong to the upper part of this series, passing, as has been shown, ra- 

 pidly into the subjacent sandstones and conglomerates, in which occur remains 

 of fishes, similar to those in the beds above. The Ichthyolites are disposed in 

 the clay in nodular layers six or eight inches apart ; and most of the nodules 

 appear to have been formed upon a nucleus of a portion, or of an entire fish, 

 whose remains are, however, generally in a very imperfect state. Scales of the 



* The nodules are of a grey colour, flattish, and when they contain any portion of a fish, split 

 by a smart blow of the hammer in the plane of the major axis, and are broken with difficulty 

 in any other direction. The interior of most of the nodules is quite compact, and surrounded 

 with a crust of fibrous carbonate of lime, occasionally speckled by green carbonate of copper. 

 The species already figured in M. Agassiz's Poissons Fossiles, are as follows : — 



Cheirolepis uragiis, vol. ii. tab. 1^. fig. 1, 2, and 3. Cheiracanthus Murchisonii, vol. ii. tab. F. 

 figs. 3 and 4. and OsteolejJ's arenatus, vol. ii. tab. 2^. 



