Ill 
2. The most satisfactory observations of this character have, 
however, been made by Professor Edmund Andrews on the 
ancient beaches of Lakes Huron and Michigan, in the United 
States, which were formed after the close of the glacial epoch. 
This calculation was based on the recession of the bluffs on 
the lake-shore, and on the amount of the sand thus washed 
away by the waves on the north, and deposited at the southern 
extremity or head of the lake. Dr. Andrews made a calcula- 
tion based on each of these data, and the result was about the 
same in both cases, which was, that the total time required for 
the formation of all the beaches (including the present) has 
been from 5,290 to 7,490 years. 
3. It is, not, however, to any of these calculations that I 
propose to call the attention of this Society at present ; to my 
own mind there is a simpler and more convincing method of 
solving this question than anj r of these, with regard to all 
of which there may be, in a greater or less degree, a residuum 
of scepticism arising from a want of implicit confidence in the 
accuracy of the observations. 
4. I propose to fix approximately the date of the glacial 
epoch without going into any calculations of this kind, but 
resting the determination on one single, well-ascertained fact, 
and I believe I can do so to the entire satisfaction of every 
impartial and unbiassed mind which will lend its attention to 
the subject. 
5. Before proceeding to elucidate the point I have in view, 
I may mention that the peat formations of Europe present a 
strong presumptive argument for the recent date of the gravel 
deposits of the river valleys in which the paleolithic remains 
are found. This peat is superimposed directly on the gravels, 
and no doubt commenced to form immediately on — or very 
soon after — the subsidence of the waters which deposited the 
loess and gravels which are found high up on the slopes of the 
valleys. The age of this peat will probably give us the time 
which has elapsed since the palaeolithic age. At the bottom 
of the peat and silt formations of the Somme valley, M. 
Boucher de Perthes found the traces of a pile-dwelling, resting 
immediately on the gravels. The “ lake-dwellers ” had suc- 
ceeded the cave-folk of the palaeolithic epoch. There is no 
geological formation to indicate any interval between the twc 
periods, although it is by no means unreasonable to suppose 
that, a brief interval — possibly a few centuries — had passed. 
The relics found at the bottom of the peat are none of them 
more ancient than the neolithic age. Much of the peat 
of Europe we know to be no older than the Roman period. 
i 2 
