88 GEOLOGY. 



early as the closing Paleozoic. Evidences of glaciation in northwestern 

 Europe, and also in China in about 30° N. Lat., at or near the base of 

 the Cambrian, has recently been presented. Less striking, but perhaps 

 not less significant, is the occurrence in the early Paleozoic, of extensive 

 salt and gypsum beds in rather high latitudes. These deposits seem to 

 imply severe and protracted aridity, and such aridity, especially where 

 north of the 30° belt, is not readily reconcilable with an enormous 

 equalizing atmospheric envelope. 



There seem, therefore, to have been, in Paleozoic times, much the 

 same alternations of very uniform with very diversified climates that 

 marked the Mesozoic and Cenozoic eras; in other words, the alterna- 

 tions of climate seem to have been of much the same order throughout 

 the known eras. The hypothesis of an enormous original atmosphere 

 suffering gradual depletion finds, therefore, but scant and uncertain 

 support in a critical study of either the biological or the physical his- 

 tory of the earth. 



II. A MODIFICATION OF THE PRECEDING VIEW. 



Can the gaseo-molten hypothesis be modified to meet these diffi- 

 culties? The especially troublesome phases of the hypothesis arise 

 from the assumption that the atmospheric and hydrospheric gases were 

 kept out or forced out of the hot molten globe, and so constituted a 

 vast atmospheric envelope. If amendment be possible, it must appa- 

 rently consist in assuming that lavas, however hot, may hold large 

 quantities of gaseous constituents which may be discharged later. 

 Existing lavas bring to the surface great volumes of absorbed gases. 

 They are disposed to discharge these, however, for the greater part, 

 under usual surface conditions; but the limits of this disposition are 

 not yet well understood; indeed, the whole subject of the absorption 

 and retention of gases in molten rock is imperfectly mastered. It is 

 not inconsistent, however, with present knowledge to suppose that a 

 molten globe of rock, arising from vaporized rock-substance, might 

 retain large quantities of the atmospheric gases so long as it remained 

 in the liquid state, and part with them only when it solidified. It is 

 admissible to go even further, and assume that much of the atmos- 

 pheric material might remain occluded in the solid rock after con- 

 gelation, since igneous rocks now contain notable quantities of gas. By 



