1 82 REPORTS ON INVESTIGATIONS AND PROJECTS. 



in its later stages, is found, by progressive investigation, to take 

 the form of oscillations between extraordinary polar mildness and 

 uniformity, on the one hand, and remarkable stages of glaciation and 

 aridity in low latitudes on the other. The solution rests in the dis- 

 covery of conditions competent to preserve subtropical life in the 

 polar latitudes through the six months, polar night on the one hand, 

 and the production of extensive sheets of ice on relatively low tracts 

 on the borders of the torrid zone on the other. 



In lieu of the more familiar hypothesis of a primitive atmosphere 

 of vast extent and hot, vaporous nature, carrying most of the con- 

 stituents that have since been consumed in the carbonation and oxida- 

 tion of the outer part of the lithosphere, the planetesimal theory 

 substitutes an atmosphere of essentially the present magnitude gath- 

 ered gradually during the growth of the earth and unaffected by the 

 extraordinary conditions attending a molten sphere. By hypothesis 

 the atmosphere was maintained by a system of feeding from within 

 the earth and from cosmical sources, and suffered loss from its contact 

 with the lithosphere, chiefly by carbonation and oxidation. Thus 

 its fluctuations in mass and constitution were dependent on the bal- 

 ance between supply and consumption. In this it furnishes the basal 

 conditions under which coordinate agencies are thought to have pro- 

 duced the oscillations between climatic mildness and uniformity, on 

 the one hand and extraordinary diversity and severity, on the other. 

 The atmospheric supplies are presumed to have varied essentially wit h 

 the volcanic activity of the earth, modified possibly by variations 

 in the cosmic supplies, but of these we have no knowledge and little 

 basis for intelligent conjecture. The consumption of the more active 

 constituents is assumed to have varied with the amount of effective 

 contact between the atmosphere and the lithosphere, which in turn 

 varied chiefly as the extent of the land and its degree of protrusion 

 above the sea-level changed. In a general way it seems probable that 

 volcanic activity increased with the deformations which augmented 

 the protrusion of the land, and hence that, in a general way, increase of 

 supply was correlated with increase of consumption, and so the atmos- 

 phere was protected against excessive variations and the continuity 

 of life was preserved. But the geological record forces the recogni- 

 tion of very marked oscillations of climate, as stated, and hence the 

 inference that the correlation of atmospheric supply and consumption 

 was not perfect and that oscillations of composition and mass marked 

 the earlier as well as the later stages of the atmospheric history. This 

 is based on the assumption that the atmosphere exercises a control- 

 ling influence over climate, and that, though not the only factor, it 

 supplies the basal conditions under which other factors operate. 



