128 Stevenson Macadam on a new Theory of the 



proportion to the mass of the earth, which a sheet of paper 

 does to that of a medium-sized geographical globe. 



This theory is evidently quite inconsistent wdth our present 

 know^ledge of the laws of liquefaction. For when a piece of solid 

 matter is added to a portion of the same substance in a state 

 of fusion, and at a temperature above its melting point, the 

 solid, if not too great in bulk, soon becomes liquefied. Those, 

 therefore, who contend that the central liquid is at a greatly 

 higher temperature than the crust which encloses and is in 

 physical contact with it, assert a condition of matters which 

 all experience contradicts, and which is as inconceivable as 

 that a crust of ice should form or remain on the surface of 

 hot water, or a solid film of iron on the white-hot liquid metal. 



The adherents of the theory which places a limit to the 

 increase in temperature, contend that the crust of the globe 

 merely lies on the liquid mass, as we find a portion of ice does 

 on ice-cold water. 



This theory presupposes that the solid materials composing 

 the crust are, as in the case of ice and water, specifically 

 lighter than the liquid portion , and, on the same principle 

 that a cake of solid water can rest or float on a lake of liquid 

 water, so in regard to the earth, it is possible that the crust 

 can float upon the liquid mass. So far, accordingly, as 

 central heat is concerned, this theory might be recognised. 

 If, however, we look at what it can tell us regarding the 

 density of the globe, we shall there find it at fault. 



The density of the earth, as experimentally determined, is 

 about 5^ times that of water. On comparing this with the 

 mean density of the superficial parts of the crust, we find that 

 their density is only about one-half that of the entire 

 earth. But if we take into consideration the augmenta- 

 tion in weight which substances at some depth must under- 

 go, owing to the increased pressure towards the centre, we 

 are justified in asserting that the globe cannot be, as is sup- 

 posed by the advocates of the theory in question, a homo- 

 geneous liquid sphere, in physical contact with the crust 

 which encloses it, for, if it were so, it would far exceed the 

 determined density. It is requisite, on this view, to assume 

 that some powerful agent of expansion, such as heat, is at 



