in the conditions referred to the winters would exceed the summers by thirty-six days.
It is not claimed by Mr. Croll that a cold or glacial epoch is directly caused by the increased distance of the earth from the sun, but from physical agents thus brought into operation. Some of these we will proceed to mention.
As the winters increase in length, and the cold in intensity, the volume of snow-fall will become greater, and its area extended. The limit at which the summer sun melts it will move slowly southward. Behind it will be a gradual accumulation of snow forming into ice. Mountain-slopes will be covered with it, until it flows down into valleys and onward, a vast sheet of glacial ice, equally on lowlands and mountains.
Out of this condition will arise several results which powerfully react, increasing and intensifying the cold of the growing winter. A volume of snow and ice covering the ground chills the air by direct radiation, and by contact lowers its general temperature, thus delaying or arresting the process of melting by the summer's sun. It is a familiar fact that in snow-covered regions the direct rays of the sun may be intensely hot, melting pitch from timber, or heating rocks, while the temperature of the air is that of the ice upon the ground. The regions of Hudson Bay are sterile, not because the heat from the summer sun is not intense, but because they are covered with ice all the year. But for this the climate might be as genial as that of England. Ice and snow maintain steadily a temperature of 32°, no matter how hot the sun's rays may be, and a rock or piece of earth will become greatly heated, while a block of ice consumes the heat that falls upon it. The solar heat is, therefore, expended in breaking down the molecular structure of the ice, and must continue to do so until it disappears. By as much heat as is used over a region in this way is the heating effect of the sun's rays diminished, and a low atmospheric temperature is the result.
But the sun's rays falling on snow are to a considerable extent reflected back into space from its innumerable surfaces, greatly decreasing their heating effect.
The snow-sheet exerts another important influence on temperature by condensing the vapor of the air into fogs as the summers come on. In this way the solar rays are arrested, and their heating power dissipated. This occurs continually during the summers of arctic and antarctic regions. Dr. Scoresby observes in regard to the arctic regions that "the sun, when near the northern tropic, gives scarcely any sensible quantity of light from noon till midnight; it is frequently invisible for several successive days, and snow is so common that it may be boldly stated that it falls nine days out of ten from April to July." These are the conditions of climate in which glaciers grow and throw abroad their chilling influence. We are now to consider
