2 GO 
PROCEEDINGS: 
BOSTON SOCIETY NATURAL HISTORY. 
as small as a quarter or even an eighth of an inch. That they are 
not parts of larger pebbles which have been disintegrated by frost, 
or otherwise mechanically broken, is shown beyond all doubt by their 
well-rounded forms. In fact, a comparison between the pebbles of 
minimum size in the modified and unmodified drift shows no sub¬ 
stantial difference, though from the nature of the deposit the minute 
grains were, of course, more common in the former. 
Pertinent to the point in question are the observations made by 
Professor Crosby on the composition of the till in the region of 
Boston. Leaving out of account all pebbles over two inches in 
diameter he found that 10 per cent of the till, on the average, was 
composed of gravel between one quarter and one twelfth of an inch 
in diameter (Crosby, ’91, p. 123). 
Professor Slialer was inclined to regard the absence of small 
pebbles as due to the crushing force of the moving ice, which reduced 
to powder all fragments below a certain size. That this is not the true 
explanation seems to be indicated by the fact that in the Braintree 
train, although the ice was probably thicker than in Rhode Island, 
and although the red sandstone probably does not possess near the 
crushing strength of the iron ore, fragments of half an inch or less 
are common. Their absence in the Rhode Island train is probably 
to be accounted for in a large degree by the very limited area of the 
tract from which they were derived and the consequent rarity of 
fragments. When only a small fraction of one per cent of the till 
is made up of a given rock, it would certainly be very difficult, if 
not impossible, to find the smaller fragments. 
Distribution of Boulders and Pebbles. 
General distribution. — The distribution of the red boulders 
throughout the train shows in general a marked regularity, the 
amounts to be found in the till at a given distance from the source 
agreeing closely in the different parts of the train. It may be said 
that for a given distance from the source the number of red boulders 
varies directly with the amount of the till and will, therefore, as far 
as the absolute amount is concerned, show variations. Of course 
this relation of the number of boulders to the abundance of the till 
could not be expected to hold good up to the very boundaries of the 
train. The boundary, however, is much more sharply defined than 
