210 Scientific Proceedings, Royal Dublin Socictij. 



have studied " live ice " mention the streams that flow from be- 

 neath the glaciers and ice-sheets. 



As to the second : Clarence King and others who have studied 

 the dying-out of ice-sheets have found that the ice melts from 

 above and below, leaving to the last patches and cakes on different 

 horizons. The waters due to such cakes, when they finally melt 

 away, must wash portions of the glacial drift into sands and gra- 

 vels formed and arranged subsequently to the overlying drift.^ 



Similarly in the till and moraine drifts we find on different 

 horizons, cakes and patches of gravel and sand, often very irregu- 

 larly stratified, as if there was a curling current, or that the roof 

 and sides of the cavities fell in, as the cakes of ice gradually melted 

 away. In various Papers on the English and Scotch drift such 

 cakes and patches of sand and gravel on different horizons are 

 appealed to as proofs of the existence of " middle gravels " of 

 intermediate age ; but, if my theory is correct, they are younger 

 than the overlying deposits. Also, if large debacles of glacial 

 mud slide down to form the upper drift, the water from the snow- 

 slush, or even from the mud alone, should form sands, "book 

 clays," and the like, between this soft matter and the harder floor 

 beneath.^ 



As to the third : gravels and other stratified drifts, when asso- 

 ciated with glaoialoid drifts, must be younger than the original 

 glacial drift with which they are connected ; as the glacialoid drifts 

 were formed from the debris of the last, while they are interstrati- 

 fied with the first. If the original drift cliff was perpendicular, or' 

 nearly so, the gravels would accumulate against it, whilst subse- 

 quently detritus, composed by weathering from the upper portion 

 of the cliff, would cover them up ; but if there were periodical falls 

 or slips of the cliff, the gravels and glacialoid drifts must in some 

 places be more or less interstratified, whilst elsewhere they would 

 blend or graduate into one another, as can be seen in numerous 

 places along the already-mentioned Irish coast cliffs. 



1 Some of the cakes may be isolated and surrounded by the clay, thus forming 

 ' ' lough holes' ' like those cut when making the railway under the Phoenix Park, 

 Dublin. These lough holes seem to have been filled for the most part with water, 

 there being only a little sand. 



2 If the floor is a rock surface we find in uneven places smaU patches of these 

 stratified drifts, which apparently were formed subsequently to that over them, as they 

 fit the roof of the cavity similar to matter found filling a rock cave. 



