STKUOTURE OF MAINE GRANITES. 31 
Upon closer inspection it appears that the granite sheets are elongated 
lenses overlying one another, of which the upper one, as a rule, has its bulging 
part lying in the depression formed by the two underlying lenses where they 
come together. 
Branner a describes the exfoliation of the granitoid gneisses in 
Brazil, which he attributes only in part to changes of tempera- 
ture. He calls attention to the fact that the linear expansion of a 
mass of gneiss 300 feet long at a depth of 15 feet from the surface 
under a surface temperature of 103° F. would amount to only 0.072 
inch ; and he quotes the results of Forbes, Quetelet, and others to 
show that the annual change of temperature can penetrate rock only 
to a depth of 40 feet in temperate regions and still less in the Tropics. 
Merrill b describes Stone Mountain, in Georgia, as a boss of gran- 
ite 2 miles long by H miles wide and 650 feet high, which owes its 
form wholly to exfoliation parallel to preexisting lines of weakness. 
The mass appears to be made up of imbricated sheets of granite 
which he regards as the result of torsional strains. The bosslike 
form is incidental and consequent. Intermittent expansion and con- 
traction from changes of temperature have so affected the sheets 
that bound the mass at the sides that they have found relief in 
expansion in an upward direction. These ruptured sheets are rarely 
more than 10 inches thick, but are 10 or 20 feet in diameter. 
Herrmann a stuns tip his conclusions on the subject substantially as 
follows: The so-called sheets are thin near the rock surface, gen- 
erally only a few centimeters thick, but become gradually thicker 
with increasing depth. This downward increase in the thickness of 
the sheets is generally more rapid where the texture of the stone is 
coarser. The course of the sheets is not, as Vogt claims, parallel to 
the original surface of the consolidating rock. It is not governed 
by internal strains. The attitude of the sheets corresponds to the 
form of the present rock surface. The sheet structure is to be looked 
upon as the effect of the beginning and progress of weathering from 
the surface inward. These weathering cracks are determined by 
the form of the rock surface instead of that being determined by 
them. 
a Branner, John C. Decomposition of rocks in Brazil : Bull. Geol. Soc. America, vol. 7, 
1896; Exfoliation, pp. 269-277; Temperature and exfoliation, pp. 285-292. 
''Merrill, George P., Rocks, Rock-weathering, and Soils, 1897, p. 245. 
c For description and representations of Stone Mountain see Purington, Chester W., 
Geological and topographical features of the region about Atlanta, Ga. : Am. Geologist, 
vol. 14, 1894, pp. 105-108 and PI. IV ; also Watson, Thomas L., Bull. Georgia Geol. Sur- 
vey No. 9-A, 1902, p. 113, and Pis. T-VIII. See also description of another granite 
dome — Stone Mountain, in North Carolina — by Watson and Laney, in Bull. North Caro- 
lina Geol. Survey No. 2, 1906; PI. XXV. 
fl Herrmann, O., Steinbruchindustrie und Steinbruchgeologie, 1899, pp. 109-111. 
