E. P. Cuherwell — Theory of the Ice Age. 59 



more than once lays down the law in the correct form in which 

 Ball states it, and in p. 87 of " Climate and Time " CroU incidentally 

 shows how well aware he was of the true nature of the disti-ibution 

 of annual heat. Speaking of the latitude of Edinburgh, he says : 

 " The quantity of heat received during winter is scarcely one-third 

 of that received during summer," and further down on the same 

 page he speaks of the " deficiency of heat received in winter from 

 that received in summer.^' 



In the first of these quotations from Croll we see one great 

 difference between his method and Ball's, a difference in which the 

 advantage seems to me to lie wholly with Croll. For, while Croll 

 cai'efully bears in mind the different relations which exist between 

 the winter and summer sun-heats in different latitudes, Ball lumps 

 together the lointer sun-heat over the entire northern hemisphere, 

 and compares it with the summer sun-heat over the entire hemi- 

 sphere. He does not suggest any reason for introducing the 

 sun-heat over the tropics in a question which relates only to the 

 temperature of the northern latitudes. He does not seem to think 

 it necessary in any way to justify or explain the application of the 

 numbers 63 and 37 (which, of course, do give the relative summer 

 and winter sun-heats received over the entire of either hemisphere) ; 

 he merely states that it is impossible to discuss the astronomical 

 theory of the Ice Age unless these figui-es form the refrain of every 

 argument (pp. 90, 91). But in fact these figures, 63 and 37, have 

 little or no relation to the astronomical theory of the Ice Age, which 

 is really the theory of the temperature of the northern latitudes in 

 the winter of great eccentricity. For the temperature at any latitude 

 is determined by four factors — 



(1) The gain of heat transferred to it from warmer latitudes by 

 means of air and ocean currents. 



(2) The loss by heat transferred from it to colder latitudes by 

 air and ocean currents. 



(3) The sun-heat received by it. 



(4) The heat radiated by it into space. 



Now neither the summer nor winter value of any one of these 

 quantities is represented by, or proportional to, the numbers 63 

 and 37. They vary from latitude to latitude, and it would be 

 impossible for any single pair of numbers to represent the combined 

 effect of the very complicated quantities over the different latitudes 

 to which the theory of the Ice Age must be applied. 



But if we are to take two numbers to represent the summer and 

 winter sun-heats which determine the temperature of the glaciated 

 regions, we should at least take figures which bear some relation 

 to the sun-heats which actually fell on those regions. Now, since 

 considerably more than half the annual sun-heat falls between the 

 equator and latitude 30° (as is evident from the fact that sin 30° 

 =:^), and since this amount is not far from equally divided between 

 summer and winter, it follows that out of 100 units of heat falling 

 annually on the northern hemisphere, say 25 falls in summer and 

 25 in winter within latitude 30°, leaving only about 63—25 and 



