BEARINGS OF PLANETESIMAL VIEW ON PETROLOGY 209 



and thence seek the prohal^le mode of genesis where the possibilities are 

 large and the embarrassments few. Each particular case presents purely 

 an a posteriori problem of its own. It is wholly unembarrassed by any 

 requirement that its factors shall sum up into a speculative primitive 

 magma. 



Bearings on the Problem of Isostasy 



A liguid earth should be an ideal example of perfect isostasy in the 

 highest sense — that is, isostasy in perfect horizontal as well as vertical 

 adjustment. The earth's crust, as it began to form, should have inherited 

 this quality in full perfection. Such distortions as later arose in the 

 crust from cooling and shrinking were the beginning of a long series of 

 actions that w^ere destructive of this perfect isostasy. Diastrophism is 

 thus made to function at the outset as an antagonist of isostasy. But its 

 normal function, as we find it in action today, is tributary to isostasy. 

 The surface forces of erosion, solution, transportation, and deposition, 

 that must have set in as soon as the deformation of the crust exposed it 

 to subaerial agencies, tended to counteract the effects of the first deforma- 

 tions and thus to reduce the deviation from the originally perfect isostasy. 

 The interplay of diastrophism and the surface agencies tends in general, 

 but not in all particular cases, toward the equalization of weight among 

 the earth columns— that is, toward isostasy. If diastrophism produces 

 inequalities at variance with isostasy, the surficial agencies and creep tend 

 to restore isostasy; if diastrophism produces inequalities in the interest 

 of isostasy, the surface agencies tend to smooth out the inequalities in 

 disregard of isostasy. The two sets of agencies tend to hold one another 

 in check. Starting with perfect isostasy and wdth little more than the 

 meager effects of cooling to actuate diastrophism, we should not expect 

 these mutually checking processes to permit inequalities to get far from 

 the original isostataic state. On the assumption of a once molten earth, 

 the first and greatest problem of isostasy is then to find an adequate and 

 consistent explanation of the profound differences of density that are 

 revealed by present evidence. Many years ago, when I still accepted the 

 molten theory, I worked industriously on this problem of the differentia- 

 tion of what must once have been very homogeneous material tlieoretic- 

 ally, but I never reached a solution that met the full requirements. The 

 first few steps seemed easy and promising, but very soon the counteract- 

 ing processes rose in efficiency and tended to destroy the gain that had 

 been made. The combination of processes seemed to confine the extent 

 of their differentiating effects to limits much short of the effects actually 

 in evidence. I am not aware that anv one has had better success. 



