1890 .] 
119 
[Crosby. 
at Nantasket, twenty feet below the surface. This section is due 
to marine erosion. 
14. Gray till from the sea-cliff on the west end of Long Island, 
Boston Harbor, about twenty-five feet below the surface. 
15. Gray till from the sea-cliff on the north side of Nut Island, 
about fifteen feet from the surface. 
16. Gray till, slightly oxidized, from the south side of the 
Corey Hill drumlin, about twelve feet from the surface. 
These twelve drumlins are scattered across almost the entire 
breadth of the Boston Basin, from Somerville to Nantasket. The 
prevailing rocks within the Basin, and on which, for the most part 
at least the drumlins immediately rest, are the slate and conglom- 
erate series, with associated sandstones, and the numerous dikes 
of diabase by which they are intersected ; while beyond the limits 
of the basin, in the direction from which the ice which formed these 
drumlins must have come, we have, for the first ten miles, chiefly 
diorite, with some granite and less quartzite, and also numerous 
dikes of diabase. The geological conditions are thus favorable, 
except in the case of the diabase, for determining approximately 
what proportion of the till is of local origin. In this connection, 
however, it is important to observe that the Boston Basin is a val- 
ley region, a true topographic basin, in large part occupied by Bos- 
ton Harbor, and hence that the drumlins in question stand on an 
area distinctly lower than the crystalline rocks to the northward, 
thus, probably, enabling the latter to contribute more liberally to 
the formation of the drumlins than would otherwise have been pos- 
sible. 
METHODS OF ANALYSIS. 
In collecting the samples care was used to obtain entirely nor- 
mal material, except that all stones more than two inches in diame- 
ter were excluded. Stones and bowlders above this size are much 
fewer and more scattering than those below it, and it was found 
more satisfactory to estimate the proportion of this coarser mate- 
rial in the till from a careful examination of fresh sections, noting 
the number of stones of different sizes in a given area of the sec- 
tion and comparing the sum of their areas with the whole. Photo- 
graphs of sections are very useful in making these comparisons. 
In the case of artificial excavations, the larger bowlders, or those 
above one and one-half or two feet in diameter, are usually left 
