[page] 220

The Cause of an Ice Age

The letter of Prof. G. H. Darwin in your last issue states very clearly the argument on which Mr. Culverwell and himself rely as affording a demonstration of the inadequacy of the astronomical theory. It now seems opportune, therefore, to lay before your readers the general considerations which lead me to the conclusion that the whole argument they rest upon is unsound; and, further, that Sir Robert Ball's ratio of 63 to 37, representing the ratios of sun-heat received by each hemisphere in summer and winter respectively, is (contrary to Prof. Darwin's view) an important factor in any adequate discussion of the problem.

Accepting Prof. Darwin's estimate that the difference in the amount of sun-heat received in our latitudes during high and low eccentricity, would only give to Yorkshire the amount received by London or vice versa, I entirely demur to his statement that this would be also a measure of the amount of change in the climates of these places. To do so is to assume that the climate of a place, as regards the amount and distribution

[page] 221

of its temperature, is determined by one factor only—the amount of sun-heat it receives.

How very erroneous is this assumption, may be shown by the contrasted climates of places on the east and west sides of the Atlantic, due to the influence of both ocean-currents and prevalent winds; but even more strikingly by a comparison (which I made in my "Tropical Nature") between certain tropical and temperate climates. In Java, about 8° south of the equator, the altitude of the noonday sun in June is about 58 1/2°, while at London during the same month it is 62°, the length of the day at the same time being 5 1/2 hours greater with us. The sun-heat received in London must therefore be considerably greater than that received in Java, and, according to the rule that the amount of sun-heat determines temperature, London should then have the warmest climate. The fact, however, is that our mean temperature in June is more than 20° lower than that of Java and our mean highest temperature about 18° lower, a result due, as I have shown, to a variety of causes, of which the temperature of the atmosphere in all surrounding areas, the action of aqueous vapour in reducing the loss by radiation, and the accumulation of heat in the soil, are probably the most important. These facts prove, I think, that the amount of heat received by the whole hemisphere, through its influence on both oceanic and aerial currents, must be taken account of in estimating temperatures under different phases of eccentricity; and that any determination of the amounts of sun-heat received at particular latitudes, considered by themselves, are necessarily misleading and must usually indicate a difference of climate far below the truth.

But there is another consideration of even more importance which entirely invalidates the arguments of those who, like Mr. Culverwell and Prof. Darwin, treat the problem as one to be determined by a simple mathematical calculation of amounts of sun-heat received on the same area at different times. This is, the remarkable difference in the behaviour of air and liquid water on the one hand and snow and ice on the other, as regards climate; the former from their great mobility tending to the diffusion of heat, the latter by its comparative immobility to the accumulation and perpetuation of cold. Without this power of accumulation perpetual snow on tropical and temperate mountains, and glaciers in hot sub-alpine valleys and at only 705 feet above the sea-level in latitude 43° 35' south in New Zealand, would be impossible. In either of these cases, if an elevation of about a thousand feet should double the area of the snow fields, which might easily be the case, the outflowing glaciers would be greatly increased in magnitude and might either descend to much lower levels or spread out over large areas of the lowlands—and all this without any change whatever in the total amount of sun-heat received by the countries in which they occur.¹

For some years past there has been a persistent attack by astronomers and physicists on the explanation of the glacial epoch put forth by Croll and adopted with some modifications by many students of glacial phenomena. But as these writers have all treated the problem as a question of the direct effect of the amount of sun-heat received at different epochs in corresponding latitudes, completely ignoring the great distributing and accumulating agencies which are always and everywhere in action, their theoretical conclusions appear to us to be entirely beside the question. We have to deal with a highly complicated problem in physical meteorology, which cannot be solved by an appeal to the well-known facts of the amounts of sun-heat received, any more than can the June climates of London and Batavia or the general climates of Ireland and Manitoba or Terra-del-Fuego (in about the same latitude) be explained from similar data. The great merit of Croll was, that he fully realised the complexity of the problem; that he took account of the various relations and reactions of the oceanic and aerial currents, and the physical characteristics of air and water, snow and ice; and that he showed how these causes reacted on each other so that the winds and ocean currents of one hemisphere might have an influence on the accumulation of snow and ice in the other. Whatever errors he may have made in matters of detail, his method was undoubtedly a sound one, and it is because so many recent writers on the subject have wholly ignored his method without even attempting to prove that it is erroneous, that their views appear to us to be both retrograde and scientifically unsound.

¹ This remarkable property and its effects are explained in some detail in my "Island Life," p. 131 (second edition), under the heading "Properties of Air and Water, Snow and Ice, in Relation to Climate," and in the four following sections.