GEOLOGY 17 
Rhynie and between the Correen-Bennachie and the Foudland ridges 
through the wind-gap at Oldmeldrum to the lower Ythan; (4) the 
valley of the Don ; (5) the valley of the Dee. 
In course of time the present drainage system has been evolved by 
(a) development of subsequents on the sedimentary fillings of pre-Old 
Red valleys in the schist floor ; (b) adjustment of streams to the structure 
of the schists laid bare by removal of the sedimentary cover, e.g. capture 
of streams flowing at higher levels by tributaries working back from 
others that had cut their beds lower, these depredations usually being 
favoured by belts of weak rocks or weak rock-structures (shatter-belts 
along fault-lines, etc.). 
Of such adjustments in times geologically recent but in all cases pre- 
glacial there are many instances. The annexation of the Tarf drainage 
area by the Tilt ought to be regarded as the classical example of river 
capture in the British Isles : it would indeed be difficult to find a clearer 
or more striking case anywhere. Map-study reveals self-evident cases 
-of beheading at earlier dates of the primeval arf. 
The capture of the upper Geldie by the Feshie has often been cited 
as an obvious case of river capture, which it is; but it is commonly 
stated to be of recent (i.e. post-glacial) date. That is not so. Capture 
was pre-glacial, and glacial erosion followed by glacial deposition has so 
altered the contours about the elbow of capture that absolutely none of 
the marks of recent capture are to be recognised. 
In the complicated history of the Spey one point only can be noted. 
Above Grantown the gradient of the river is low ; from Grantown to the 
sea there is an almost uniform fall of 16 ft. per mile. In this part 
of its course the river has been rejuvenated and here we find the charac- 
teristic mark of rejuvenation—an inner, narrow, young valley incised in 
the floor of an outer, wide, old valley. The inner valley is deepest in 
the vicinity of Knockando, and the axis of uplift must cross the river there, 
not at Grantown as usually stated. 
B. GLaciaTION.—Within most of our area the distribution of erratics 
from the outcrops of all rocks yielding readily identifiable boulders shows 
that the direction of ice-movement varied widely at different periods 
of the Ice Age. For example, boulders from the Huntly basic rocks 
have been transported towards all points of the compass between N. 50° W. 
round by E. to S. 20° E. The diorite of Netherley (on Burn of Rothes) 
has been carried in similar but still more widely divergent directions. 
From this wide dispersal one infers the action of more than one ice- 
sheet, an inference confirmed by study of striz and boulder clays. At 
least three ice-sheets have successively traversed the area. Hence it is 
to-be expected that the drift series will be complicated and difficult to 
interpret. ‘Three characteristic drifts, however, can be identified with 
certainty, though between the Findhorn and the North Esk only four 
open sections show all three in direct superposition and only two show 
three superposed boulder clays. 
(1) First Ice-sheet—The transport of erratics, the direction of striz, 
and the character and contents of the ground moraine clearly prove that 
the ice traversed our region roughly from N.W. to S.E. This was no 
B 
