THE BE VOLUTIONS OF THE CRUST OF THE EARTH. 345 



of the vapors and by the formation of fogs. To recapitulate, the equator 

 of heat remains thirty-four days longer to the south of the terrestrial 

 equator, and for as many days longer the humidity of the torrid zone is 

 transported to the north, where the excessive cold of the winter con- 

 denses it and precipitates it in the form of snow. The vapors condensed 

 south of the equator are only very seldom precipitated in the form of 

 snow, and then only in high latitudes, the cold of winter being insuffi- 

 cient to produce congelation. The accumulation of snow in the boreal 

 hemisphere and its almost complete absence in the temperate zone of 

 the austral hemisphere contribute in various ways to lower the total 

 heat of the boreal hemisphere. To the heat absorbed by the fusion of 

 the ice, by the interposition of fogs and clouds, we should add the re- 

 flection of the calorific rays from the snow. In fact, the winters becom- 

 ing more rigorous, the first result of this increase of cold will be the 

 depression of the equatorial line of snow-falls toward the equator; the 

 quantity of heat will not then be everywhere proportional to the sine of 

 the angle of incidence, for a part of the heat is reflected by the bed of 

 snow, an effect quite as important as the absorption by fusion of the ice 

 or by the interception of fogs. 



Let us remark also that the mean temperature of the summer can 

 never rise in the presence of large quantities of ice and snow. Scoresby, 

 in his voyage to Greenland, observed this curious fact, that the pitch with 

 which his vessel was coated melted under the influence of the direct 

 rays of the sun, while the ambient temperature was below zero ; a ther- 

 mometer exposed to direct radiation indicated 36° centigrade, and the 

 ambient air was at — 6°. This fact, in appearance so strange, accords 

 perfectly with our ideas of the diathermanous properties of the air. Tyn- 

 dal states that perfectly dry air is almost entirely incapable of absorb- 

 ing direct heat. The air is then cooled by the contact with the ice, but 

 it is not heated by the sun, hence the low temperature. * If by evapora- 

 tion the air is charged with humidity, still its temperature will not in- 

 crease, for in proportion as it absorbs heat it immediately gives it up 

 again to melt the ice, and the ambient temperature will remain invari- 

 able. If a stratum of ice were transported to the equator, the mean 

 temperature would be lowered to a point not equal to that of Central 

 Europe. 



Since, in accordance with what we have just said, the glacial periods 

 may depend upon a cause as general as the eccentricity of the earth's 

 orbit, is there no way of determining the epoch of these extraordinary 

 cold periods ? Messrs. Leverrier, Stone, and Croll have formed some 

 tables which indicate the value of the eccentricity at intervals more or 

 less remote. We recognize easily by them when the action of ice must 

 have been considerable at the surface of the globe, but we cannot tell 

 at what time the last great glacial period occurred. Was it between 

 80,000 or 240,000 years ago, or between 700,000 and 1,000,000? If we 

 adopt the first of these figures as nearest our period, then the second 



