The Scottish Naturalist. 249 



intense pressure to which the till was subjected during its grad- 

 ual accumulation under the ice. It is assuredly not the result 

 of aqueous action. Aqueous lamination is due to sifting and 

 winnowing — the coarser or heavier, and finer or lighter particles 

 being separated in obedience to their different specific gravity, 

 and arranged in layers of more or less regularity according to 

 circumstances. There is nothing of this kind of arrangement, 

 however, in the so-called stratified boulder-clay. If the clay of 

 an individual lamina be washed and carefully sifted, it will be 

 found to be composed of grains of all shapes, sizes, and weights, 

 down to the finest and most impalpable flour. It is impossible 

 to believe that such a heterogeneous assemblage of grains could 

 have been dropt into water without the particles being separated 

 and sifted in their progress to the bottom. Of course every one 

 knows that patches and beds of laminated clay and sand of veri- 

 table aqueous origin occur now and again in boulder-clay. I 

 suppose there is no boulder clay without them. I have seen 

 them in the till of Italy and Switzerland, where they show pre- 

 cisely the same features as the similar laminated clays in the till 

 of our own islands. But these included patches and beds point 

 merely to the action of subglacial waters, such as we know circu- 

 late under the glaciers of the Alps, of Norway, and of Greenland. 1 

 Again, I would remark that Mr Mackintosh has ignored all 

 the evidence which has been brought forward from time to time 

 to demonstrate the subglacial origin of boulder-clay, and to 

 prove the utter inefficacy of floating ice to account for the phe- 

 nomena. And he adduces no new facts in support of the now 

 discredited iceberg theory, unless it be his statement that flat 

 striated rock-surfaces (such as those near Birkenhead) have been 

 caused by floating ice — the dome-shaped roches moutonnees 

 being, on the other hand, the work of land - ice. As a 

 matter of personal observation, I can assure Mr Mackin- 

 tosh that flat striated surfaces are by no means uncom- 

 monly associated in one and the same region with roches 

 moutonnees. What are roches moutonnees but the rounded 



1 It is perhaps hardly necessary to say that I do not call in question the 

 fact of a great submergence having taken place in England during a certain 

 stage of the Glacial Period. The marine sands and gravels which occur be- 

 tween the lower and upper boulder-clays of Western England, and the high- 

 level shell-beds of Wales, are the proofs of that submergence. What I deny 

 is, that either the lower or upper boulder-clay, or the erratics which are asso- 

 ciated with these glacial deposits, were deposited in water- 



