PLANETE3IMAL THEORIES OF THE EARTH'S ORIG1X. 



203 



suffice ... to melt the moonlet if it were composed of ordinary 

 volcanic rock, and provided all of the energy were applied to the 

 heating of the moonlet. Practically only a portion of it was thus 

 applied ; another portion produced heat in the contiguous tract of 

 the moon's material ; yet another was consumed in the deformation 

 of moonlet and moon resulting in the crater, and another resulted 

 in modifications of the moon's motions, changing its orbit, its 

 orbital velocity, its axis, and its rotational velocity. The energy 

 converted into heat might be regarded as the remainder after 

 deducting all other effects, and the resulting temperatures would 

 be further conditioned by the distribution of heat in the colliding 

 masses. 



"Since the area of the moon's surface directly struck by the 

 moonlet is a function of the square of the diameter of the moonlet, 

 while the energy applied to that area, being measured by the mass 

 of the moonlet, is a function of the cube of its diameter, more 

 energy would be applied to a unit of space in the case of large 

 moonlets than in the case of small, and the temperatures caused 

 by large moonlets would therefore be greater. To this relation I 

 ascribe the restriction of inner plains, indicative of fusion, to the 

 larger craters. . 



" In the breaking up of the postulated pre-lunar ring there were 

 at first many centers of aggregation — were the moon the only 

 center, the scars of impact would all be small. So long as the 

 masses were small the process of aggregation developed little heat, 

 for the heat of impact depended almost wholly on velocities 

 created by mutual attraction? That particular moonlet which 

 became the nucleus of the moon may therefore be conceived as 

 cold, or at least as sufficiently cool to be solid. As the moon's mass 

 grew, the blows it received were progressively harder, and for a 

 time their frequency also increased. The rate of heating probably 

 reached and passed its maximum while the mass was materially less 

 than now. During the whole period of growth the surface lost heat 

 by radiation, but the process of growth cannot have been slow 

 enough to permit the concurrent dissipation of all the impact heat. 

 On the one hand, there should have been some storage of heat in 

 the interior, and on the other hand, the stored heat can never have 

 sufficed for the liquefaction of the nucleus. Toward the close of 

 the process, when blows were hard but rare, liquefaction was a local 

 and temporary surface phenomenon, but the general temperature of 

 the surface was low. Impact heat, being evolved simultaneously 

 in the surface and the subsurface, was dissipated more rapidly from 

 the surface, so that there was a subsurface zone of relatively 

 high temperature. The zone thus inferred deductively is also 

 inferred inductively from the disparity of cavities and rims in the 

 case of large craters ; but, on the other hand, there is little evidence 

 of the wrinkling which, theoretically, should result from the 



