24 A. p. VINOGRADOV 



(Marshall Islands and others) have not shown any granite, the bore passing 

 straight into a basic olivine rock. The floors of the Atlantic and Indian Oceans 

 (except in their deep rifts) seem to have a granitic layer. The basaltic layer in the 

 beds of the oceans seems to be at least 10-15 ^ thick. There is, thus, no gradual 

 transition between the oceanic and continental parts of the crust of the Earth. 



The view was put forward a long time ago that there was a certain uniformity 

 of composition of the shells of the Earth — the granitic, the basaltic and any 

 other — throughout its whole extent. 



Undoubtedly the structure of the crust of the Earth, its thickness and compo- 

 sition, give an indication of the way in which it could have been formed. The 

 formation of the crust of the Earth is related to the differentiation of the sub- 

 stances of the Earth when hot. 



THE HEAT OF THE EARTH 



The course of the thermal balance of the Earth over 5 x 10^ years is a very 

 complicated problem. It is generally considered that there were two possible 

 sources of the heat of the Earth, (i) That it was the result of the radio- 

 active decay of U, Th and ^^^K and (2) that it was caused by the energy liberated 

 as a result of the compression of the Earth. Even as concerns radiogenic heat 

 there has been no agreement among scientists during recent years, mainly 

 owing to the absence of sufficiently acctirate data concerning the amounts of 

 U, Th and K in meteorites and in the depths of the Earth, and to the inaccuracy 

 of determinations of the constants of decay. If we assume that, at the present, 

 the mean content of radioactive substances in the silicate phase of the Earth 

 is the same as that of the stony meteorites or chondrites, i.e. U ~ 3 x lO"^ %y 

 Th '^ I X 10-5% and K^S-f x 10-2%, then elementary calculations will 

 show that 5x10^ years ago the radiogenic heat was enough to melt the material 

 of which the Earth is composed. I must add, though, that the mean values 

 assumed for the content of U, Th and K in meteorites are minimum ones.* 



There is even more room for doubt in the second question, namely, the 

 extent to which the mean values of the content of radioactive elements in chon- 

 drites is a permissible means of calculating the thermal effect on the Earth. 

 Unfortunately, however, we have no other means of doing so. Even in the case 

 of dunites, those ultrabasic rocks of the mantle of the Earth which sometimes 

 reach its surface, we have no reliable data as to their content of U, Th and K 

 or other elements, let alone mean values. I must point out that the obtaining of 

 such data is quite a difficult task. Our comparative estimate of the contents of 

 U, Th, K and a number of other elements in chondrites and dunites indicates 

 that the latter contain considerably less of radioactive and some other elements 

 than do the stony meteorites or chondrites. As we shall see later, there is, from 

 our standpoint, an explanation for this. We can also agree that the singularly 

 good and definite determinations of K in dunites carried out by Holyk and 

 Ahrens are correct. They give extremely low values for the K content of dunites, 

 namely ^ i x 10-^ %. 



* Even when analysing the same stony meteorites different workers have made esti- 

 mates of U content ranging between 3 x lO"* and 3 x lO"^ % [8, 14, 17]. 



