144 Prof. BischofF oti the Temperature of' 



consequently, a transmission of heat during one lialf of the 

 year from below towards the surface ; and during the other 

 half, from the surface towards the centre of the earth ; but 

 from the interior it is emitted constantly from the centre to- 

 wards the surface. That the latter is actually the case, is very 

 clearly proved by the above-mentioned springs, which partici- 

 pate in the variations of temperature in the outer crust, but 

 yet are thermal. 



To find the law of the increase of temperature for a given 

 place, entirely free from all external influence, the depth must 

 always be first determined by observations, to which the im- 

 mediate external influences penetrate. In Chap. VIII. we have 

 already seen that this is a difficult problem. 



If the external causes which have an influence on the tem- 

 perature of the earth in a certain district, penetrate to a cer- 

 tain depth t feet ; then, all other circumstances being the same, 

 this depth would be the same in the mountains of that dis- 

 trict as in the plains. Let us suppose the district only to be 

 of such an extent, that t have the same value at every point of 

 it ; the limit of the external influences would then be parallel 

 to the form of the surface, running horizontally beneath the 

 plains, and rising under the mountains. 



Supposing the decrease of temperature to depart from the 

 centre of the earth, or from any spheroid concentric with the 

 surface of the earth ; it is immaterial what the temperature of the 

 centre or of the nucleus bounded by that spheroid may be ; this 

 decrease of temperature, all other circumstances being the same, 

 cannot proceed according to the same law towards all points of 

 the earth's surface in concentric sjiheroids. For in that case 

 there must be the same temperature at all points of the sphe- 

 roid which forms the limits of the influence of the external 

 temperature, whether under the equator, or under the poles. 

 The decrease of temperature from the interior of the earth to- 

 wards its surface must, therefore, follow different laws in dif- 

 ferent latitudes ; that is to say, the temperature must increase 

 more slowly in proportion to the depth, under the equatoi", 

 than at the poles. 



If we imagine curves drawn from the equator towards the 

 poles, all points of which having the mean temperature at the 



