296 



B. WILLIS DISCOIDAL STRUCTURE OF THE LITHOSPHERE 



rocks melt in laboratory experiments. Such masses may never have oc- 

 cluded water vapor, or they may have lost their more volatile constituents 

 by repeated heating and kneading. 



3. The observed conductivities, specific heats, densities, and melting 

 points of basic rocks, represented by basalt, and of acid rocks, repre- 

 sented by gneiss, may be taken as a basis of inference regarding the be- 

 havior of the two kinds of rock when subjected to rising temperature in 

 the asthenosphere — that is, at depths of 50 to 800 miles below the surface. 



To give form to the hypothesis, let it be assumed that there is a mass 

 of basic rock in contact with acid rock, both in a solid state, in the 

 asthenosphere. Let the temperature be rising by the conduction of heat 

 from below and, to establish uniform conditions to begin with, suppose 

 that an isogeotherm passes through both kinds of rocks. 



Now the temperature to which each kind of rock m411 be raised by the 

 same amount of heat depends on the conductivity, the specific heat, and 

 the density of that rock. The following values may be taken as a basis 

 of estimate :^^ 



Characteristic 



Basalt. 



Gneiss. 



Note?. 



Conductivity (k) 



0.00317 



0.000578 



k at 0° C. ; temperature co- 

 efficient for 1° 0. is, for 

 gneiss, minus 0.002803 ; for 

 basalt, plus 0.00001. 



Specific heat (c) 



0.199 



0.214 



At temperatures ranging 



* 







from 20° to 200° C. 



Density (d) 



3.0 



2.6 



At the surface, under one 



Melting point 



1,200 



1,300 to 

 1,500 



atmosphere pressure. 



The formula for the temperature to which a supply of heat will raise 

 a unit of volume in a unit of time is 



h/cd 



Inserting the above values, we obtain : for basalt, th = 0.0052 ; and for 

 gneiss, tg^^ 0.0014. That is to say, the basalt will heat up nearly four 

 times (3.7 times) as fast as the gneiss. 



This being so and the melting temperature of basalt in a dry melt being 

 lower than that of gneiss under atmospheric pressure, it follows that the 

 times required to melt them with a given supply of heat are at least 

 as 1 : 4. 



Assuming special cases, we may conclude: (1) If a body of basalt 



51 Landholt, Bornstein, and Meyerhoflfer : Tabellen, 1913. 



