seek for the cause of secular climatal variations in the earth's astronomical relations. In the same chapter, but more fully in the Appendix, Dr. Croll goes on to show that the variable length of the seasons consequent upon the ellipticity of the terrestrial orbit had begun to attract attention before the close of the last century; and that, as early as 1830, Sir Charles Lyell had expressed the idea that the long winters and short summers of the southern hemisphere might have some influence in lowering the temperature of that portion of the globe.[1] Sir John Herschel and others, however, soon after demonstrated that the light and heat received by any portion of the earth's surface during any year is practically invariable, whatever the eccentricity of the terrestrial orbit; the greater proportionate length of winter in the hemisphere whose winters occur in aphelion being exactly counterbalanced by the greater proximity of the sun in summer. This is, indeed, a legitimate deduction from Kepler's second law, and was long ago demonstrated by D'Alembert. The hypothesis, therefore, fell into disrepute. Over fifteen years ago, however, the author of the work under consideration began to point out, in a series of papers (chiefly in the "London and Edinburgh Philosophical Magazine"), the substance of which is reproduced in "Climate and Time," that while the variable length of the seasons resulting from this cause could never produce a glacial epoch directly, yet the same cause might, especially when intensified by a high degree of eccentricity, bring into operation a chain of physical agencies which could not fail to very materially affect the climate of the globe.
As a further introduction to that portion of the work devoted to the elucidation and application of the above-named astronomical and physical principles, and as an illustration of the efficiency of one of the secondary agencies on the operation of which the theory is based, the heat-conveying power of ocean-currents is discussed at length in the second and third chapters. The importance of these currents is shown to be immense. Thus, according to Professor Dove's "Temperature Tables," the temperature of the British Isles, and of western Europe generally, is 12° Fahr. above the normal—or, more properly, the mean—for that latitude, while the temperature of corresponding portions of eastern North America is nearly as much below the normal. Dr. Croll attributes this difference to the effect of the Gulf Stream in warming western Europe, and of the cold counter-current in chilling our American coasts. The same subject is recurred to frequently throughout the volume, notably in Chapters XI. and XII., in the latter of which Mr. Findlay's objections are answered by calculating from his own data that the heat liberated from the Gulf Stream in the North Atlantic is equal to more than one half of that received directly from the sun in the same latitude. An analogous condition of things exists on the shores of the North Pacific, which are similarly
- ↑ "Principles," first edition, 1830, vol. i., p. 110.
