IN THE SOUTH OF ARRAN. 539 



seems rather to have been actually floated ofiP by the sinking of the land ; and the 

 first deposits thrown down on the bared rock are marine and shell-bearing. 



3. The boulder-clay was deposited in the sea. With the moving ice the debris 

 entangled in it and under it must have been in constant progress seawards, and 

 could never permanently have come to rest till thrust out into the sea under the 

 ice-foot. From this consideration alone we might have inferred the deposition of 

 the boulder-clay under water; but we have more direct evidence, for in Arran, 

 as in some other places, the boulder-clay contains marine shells in considerable 

 numbers, and in circumstances indicating that the animals lived and died in the 

 bed where they now are. Obviously, therefore, the clay in which these are found 

 was deposited under water. Corroborative proof of this fact is further found in 

 those beds of stratified sands and clay which are, comparatively, so frequent in 

 Arran, and which are seen elsewhere wherever an extensive section of the 

 boulder-clay is visible. These plainly imply the action of water, but of them- 

 selves, and without the shell beds, would still have left it doubtful whether the 

 water-action might not have been temporary and lacustrine. The two combined 

 confirm each other's testimony to the deposition of the boulder-clay in the sea. 



4. Though deposited in the sea, the boulder-clay has been compressed by ice. 

 At the edge of the ice-foot there must have been a growing bank of debris, on 

 which fresh heaps were constantly emptied. On a free and open slope, in such 

 circumstances, the mass of course simply rises in height, till, the angle of slope 

 become too great, the face of the bank breaks, and the upper part slips forward 

 over the lower. But in the case of the boulder-clay, the massive shelf of ice 

 above would prevent the free growth of the bank, which, partly under its own 

 weight, and partly under the pressure of the ice, must constantly have been sub- 

 siding, to some extent undergoing compression, to some extent yielding inter- 

 nally and bulging out in front, so as to undergo a complete disarrangement and 

 confounding of all its component parts, which is one of the characteristic features 

 of the boulder clay, and of which we may regard the bed shown in Plate XXI. 

 fig. 4. as an unfinished and transition example. There the pressure has evi- 

 dently been at once from behind and above, so that the beds have bulged out for- 

 wards, but the pressure had not been long enough maintained to work the various 

 beds into one another. 



The influence of the massive ice-foot, in working utter confusion in the bank 

 which lay below it, must have been all the greater from the different conditions 

 it must have assumed under the ever varying change of the seasons, and those 

 slower oscillations of rain-fall and temperature, which extend throughout years. 

 At times it would crush the oozy mass below with an enormous weight, till the 

 sludge was compressed almost to the solidity of stone. Again, it would be floated 

 off, and leave space below it for the deposition of stratified sands and clays, in 

 which even molluscan life might thrive, till a rush of debris, brought down by a 



VOL. XXIII. PART III. 7 G 



