1890. J 
129 
[Crosby. 
versity of composition. Bat as the trituration advances, the com- 
position becomes rapidly more simple and definite, the whole being 
reduced ultimately to quartz-flour and clay. 
The siphon residue or ninth product, which looks like clay, but 
is proved by the chemical test to be almost three-fourths rock-flour, 
appears under the microscope simply as a finer grade of pulverized 
quartz, with the grains somewhat more obscured by the dusty clay ; 
although much of the seeming clajf proves on closer examination to 
be simply quartz-flour which is too fine to be resolved by a low 
magnifying power. Thus the clay in the till constantly eludes us, and 
even in the tenth product or siphonate, which has remained in quiet 
suspension in water for at least fifteen hours, the proportion of com- 
bined water is not only too low for a pure clay, but finely divided 
quartz, a superfine quartz-flour, can be actually seen in spite of the 
masking clay. This accords with Daubree’s observation that the 
milky turbidity of the Rhine, even at a distance of hundreds of 
miles from the Alpine glaciers, is due chiefly to impalpable quartz. 
Comparison of the Upper and Lower Till . — The view now gener- 
ally accepted by glacialists that the upper or buff till simply meas- 
ures the extent of the superficial weathering or oxidation of what 
was originally a homogeneous deposit deprives this comparison of 
much of the interest that would otherwise attach to it. We may, of 
course, reasonably assume that the upper till embraces the largest 
proportion of englacial detritus, but it is impossible that the lower 
limit of this material should be sharply defined or coincide, as a 
rule, with the ever deepening line of oxidation. And the fact that 
the buff till is looser in structure, a less typical hardpan, should 
perhaps be attributed to the subsequent oxidation and the leaven- 
ing action of the frost quite as much or more than to its supposed 
englacial origin. But since drumlins, if not the till in general, are 
certainly due to a process of slow accretion, it is at least probable 
that the lower till is usually of more local origin and includes a 
larger proportion of large and angular rock fragments than the 
upper till. The scanty evidence afforded by the few deep sections 
of the till in this vicinity appears to lend some support to this as- 
sumption. 
That the lower or gray till has suffered no appreciable chemical 
change or decomposition since the disappearance of the ice-sheet 
is abundantly evident from the hard, unaltered ledges or rock-sur- 
faces on which the till now rests, and from the hardness and fresli- 
9 FEBRUARY, 1891. 
PROCEEDINGS B. S. N. H. 
VOL. XXV 
