Springs: The Influence of Stratigraphy. — Hopkins. 367 
is not so clear why No. 13 should have so great an influence, 
as it is a very friable white sandstone, composed of almost 
pure silica. It is so friable in places that it can easily be 
rubbed, in the hands, into a loose mass of white sand resem- 
bling granulated sugar. The state geologist has suggested 
capillarity as an explanation of its impermeability. I think, 
however, that two other factors play a very important part : 
first, the indurated surface, since in nearly all places a thin 
hard crust has formed over the surface, by the deposition of 
silica, iron and possibly lime; second, the shale may be more 
abundant than would appear from surface exposures. Thus 
where the outcrops on the hillsides would indicate only traces, 
or the absence, of the shale, it may underlie a larger part, pos- 
sibly the greater part, of the hill ; its absence on the surface 
being due to local thinning of the bed, a characteristic of this 
formation, or to the weathering away of the shale near the 
surface, permitting the overlying limestone and chert to settle 
down on the underlying sandstone. This last factor is appli- 
cable only to the western part of the area, where the lime- 
stones, Nos. 11 and 12, are absent. 
In the eastern half of the area the shale is more variable 
in thickness than in the western half. Where it has a thick- 
ness of two or three feet or more, it is a prominent horizon for 
springs. Where it is absent or ver} T thin, the springs are ab- 
sent and the top of the sandstone No. 13 is the prominent 
spring horizon. 
The explanation for the excess of springs emerging on Nos. 
10 and 13, over the number emerging on the almost equally 
impervious shales of Nos. 2 and 7, is found in the character 
and distribution of the chert and limestone bed No. 8. In this 
bed the chert predominates in most places: and in all places 
it predominates on the surface, owing to its greater durability. 
It is very much jointed, and the weathering opens these 
joints, strewing the surface with a mass of loose angular frag- 
ments. The rain and snow water sink so quickly into this 
mass of broken stone that very little runs ofl* as surface 
water. This chert covers a much larger surface area than any 
of the other beds, probably due in part to old baseleveling, 
and in part to its greater durability. It covere a large area 
