802 B. N. PEACH AND J. HORNE ON THE 
drifts. The same remarks are applicable to the erratics of pink 
granite and quartz-felsite scattered over the surface between Fetha- 
land and Colifirth Voe. 
These accumulations were in all probability extruded at the 
snouts of the local glaciers when the great mer de glace had melted 
away from the immediate vicinity of Shetland. This is rendered 
all the more likely from the number of striated stones in the deposit, 
and its tolerably coherent nature, differing somewhat from the loose 
débris of the ordinary surface-moraines. But, in addition to these 
later deposits, there is abundant evidence to show that when the hill- 
tops had emerged from the icy covering which so long held sway 
during the primary glaciation, the severe frosts which prevailed 
caused an accumulation of blocks and rubbish on the surface of the 
attenuated glaciers. In course of time, as the glaciers melted back, 
loose heaps of rubbish were laid down, sometimes as isolated 
mounds, but frequently in concentric lines indicating pauses in the 
retreat. 
As might be expected from the size of the valleys and the limited 
elevation of the hills, the moraines are not large; but they are 
nevertheless very abundant; indeed there are few of the im- 
portant valleys draining a mass of high ground which do not contain 
well-marked groups. They consist of loose débris with angular and 
Subangular stones; and in some cases the deposit 1s merely an as- 
semblage of small stones without any matrix. Numerous examples 
occur, however, where the heaps show distinct traces of stratifica- 
tion and the stones are somewhat waterworn. 
In a small yalley about two miles in length, draining the east 
side of the Bonxie hills and flowing into the bay below Connings- 
burgh, a fine series of moraine heaps occurs, displaying the usual 
concentric arrangement. ‘They vary in length from 5 to 10 feet; 
and numerous blocs perchés of the Bonxie-hill schists rest on the 
mounds. . 
In the district of Delting, moraines are to be found in the main 
valleys and round the heads of the larger sea-lochs, as for instance 
the Dales, Colifirth and Swining Voes on the east coast, and near 
Voe, North Brae, and Voxter on the west coast. And so also on 
the banks of Vidlon and Dourye Voes in Lunnasting, similar de- 
posits are seen resting on the Boulder-clay. We frequently observed 
that the moraines became more numerous where two or more 
valleys converge at the head of a sea-loch, which is just what might 
be expected when the tributary glaciers must have coalesced at 
this point. At the head of Swining Voe two streams join the sea; 
and at the point where the valleys converge a well-marked series of 
concentric mounds is met with. The dividing ridge is strewn all 
over with innumerable moraines, doubtless the lateral moraines of 
both glaciers; and in the various burn-sections the morainic drift 
is found to rest on the Boulder-clay. The evidence derived from the 
disposition of these mounds shows clearly that at the time they were 
deposited the glaciers did not fill the valleys to any great extent, 
neither did the trunk glacier extend very far down the yoe. 
