PEESENT KNOWLEDGE OF THE EARTH WIECHERT. 435 



mathematical theory of gravitation. It tells us, first of all, that the 

 flattening would have to be one two-hundred-and-thirtieth if the 

 masses in the earth were uniformly distributed, and one five-hundred- 

 and-seventy-eighth if the main mass were located at the center. As 

 the real flattening is one two-hundred-and-ninety-eighth, or very 

 nearly that, it follows that the distribution of the mass in the interior 

 of the earth lies between the two extremes ; that is to say, the density 

 increases toward the center, but at such a rate that a notable part of 

 the mass is present in the outer layers. This, in fact, is the result at 

 which we had already arrived. A further conclusion to be deduced 

 from the figure one two-hundred-and-ninety-eighth is a certain state; 

 ment concerning distribution, which can only be formulated in its full 

 extent by means of mathematics, and which I am therefore unable to 

 place before you. The situation becomes much more favorable, if I 

 now make use of the suggestion, already referred to as a very natural 

 one, that the earth consists of a metal core enveloped in a mantle of 

 rock. As we know the thickness of the rocky mantle to a certain extent 

 from direct observation, we are enabled, on the basis of the figure 

 one two-hundred-and-ninety-eighth, for the flattening, to calculate the 

 size of the metal core and its density. It is found that the rocky 

 mantle on which we live must be 1,300 to 1,600 kilometers thick (800 

 to 1,000 miles), and that therefore the metallic core, so far as its 

 diameter is concerned, occupies about four-fifths of the globe. It also 

 appears that the density of the metal core must be a little more than 

 eight times as great as that of water. The density of iron under 

 the pressure conditions known to us at the earth's surface is a little 

 less than eight. Thus we see that we obtain for the density of the 

 metal core of the earth a figure corresponding to the density of iron 

 when somewhat compressed or somewhat alloyed with heavier metals 

 (for example nickel). We are thus led to the conjecture that the 

 metal core consists in the main of iron. In support of this conjecture 

 we are able to bring forward quite an array of additional arguments. 

 In connection with volcanic eruptions, rocks rich in iron are often 

 ejected from the depths of the earth. The meteorites which drop on 

 the earth from planetary space consist partly of rock and partly of 

 metal, iron being by far their predominant constituent. Analj^sis of 

 the sun's light by means of the spectrum shows that iron vapors have 

 a vast share in the composition of the sun. It thus appears that iron 

 is very strongly represented in the structure of our solar S3^stem, and 

 in particular that our earth is simply an iron ball coated witjti rock. 

 It simply represents on a larger scale a meteorite which consists of a 

 mixture of rock and iron. 



Interesting as these conclusions are, we must not forget that they 

 rest as yet on a very weak foundation. They would vanish at once, 

 for example, if the increase of density toward the interior of the earth 



