288 



J. C. BRANNER — DECOMPOSITION OF ROCKS IN BRAZIL. 



the temperature and the time approximate the conditions in Brazil when 

 the rocks are largel}'' granitoid gneiss and where the temperature for rocks 

 openly exposed is often between 140 and 150 degrees Fahrenheit for three 

 or four hours during the da3\ '"^ Their results show that a mean tempera- 

 ture of 152.2°, Fahrenheit, on the surface of the rock will raise the tem- 

 perature at a depth of 20 inches 4.7°, Fahrenheit, in 3 hours and 46 min- 

 utes. Nothing is known b}^ direct observation of the temperatures of 

 rocks in Brazil at any depth beneath the surface ; but assuming that the 

 changes of 10 and 15 degrees take place at the same de2:)th as they do in 

 other parts of the world where observations have been made, let us cal- 

 culate the expansion of the rock which must be caused at the surface and 

 at depths of 3, 6, 9, 12 and 15 feet. The laws of the expansion of certain 

 rocks are known for ordinary temperatures. Gneiss expands one part in 

 from 187,560 to 228,060 parts for every degree Fahrenheit.f If we take 

 one part in eve'ry 200,000 as its average expansion for every additional 

 degree, the linear expansion for a surface of gneiss 300 feet long will be 

 as follows : 



Linear Expansion for 300 feet of Gneiss. 





Depth in feet. 



Temperature. 



Expansion. 



Surfiice 



Fahrenheit. 



108° 



15° 



10° 



7°(?) 



5° 



4°(?) 



Inches. 

 1.854 



3 



0.27 



6 



0.18 



9 



0.126 



12 



O.Oi) 



15 



0.072 









It must be confessed that these figures are disap})ointing, for tliey lead 

 one to suspect, if they do not show, that exfoliation on account of changes 

 of temperature and where small areas are ex]30sed is not so active an 

 agency in Brazil as the exfoliated peaks and bouhlers constantly suggest. 



It is no uncommon thing, however, for surfaces of rock a thousand feet 

 in length to be exposed, and in such cases there must l)e great strain be- 

 tween the upper and lower layers. Crevices are also o})ened whicli furnish 

 access to the rock beneath for that train of destructive agencies which 



*Lord Kelvin and Mr Murray have shown that the thermal conductivity of rocks is diminished 

 by increase of temperature between 50 and 214 degrees centigrade. (On the Temperature Varia- 

 tion of the Thermal Conductivity of Rocks. Lord Kelvin and J. R. E. Murray. Proc. Roy Soc, 

 vol. Iviii, no 349, pp. 1G2-1G7.) For objections to these results see Robert Weber in Nature, Sept. 

 12, 1895, pp. 458, 459. But as Dr Julien's data are taken from direct observations, we need not con- 

 cern ourselves in the present ease with variations of conductivity. 



f See also -'The Origin of Mountain Ranges." T. Mellard Reade. London, 1886, pp. 109-111; A. 

 J. Adie in Trans. Roy. Soc. Edin., vol. xiii, p. 3i36. .\c(^ording to the determinations of Lieutenant 

 Bartlett (Amer. Jour. Sci., vol. xxii, 1S3J, pp. 130-140), an expansion or contraction of one inch would 

 take place in every 107 feet with a change of 103 degrees Fahrenheit. 



