252 PROCEEDINGS OF SECTION A, 



Suppose that the total heat received from the sun on one 

 hemisphere of the earth during the course of a twelve- 

 month be represented by 100, we proceed to consider how 

 these parts are shared between the seasons. I mean by 

 ''summer" in the northern hemisphere the interval from the 

 vernal equinox to the autumnal, and by "winter" the 

 interval from the autumnal equinox to the vernal. With 

 this understanding 63 parts of heat are received on each 

 hemisphere during its summer and. the remaining 37 parts 

 during the winter. It is the announcement of this law of 

 distribution which forms the novel feature in the present 

 communication. Herschel's erroneous statement was to the 

 effect that the heat was equally distributed, so that 50 parts 

 were received in summer and 50 parts in winter. Accurate 

 knowledge on this point is of essential importance in 

 attempting to understand the cause of an ice age. 



We have now to consider the other factor in the astro- 

 nomical theory : it arises from the perturljation of the earth's 

 orbit by the influence of the planets. At present the ellipse 

 which forms the earth's orbit differs Init little from a circle. 

 Its form is, however, not permanent. The disturbances 

 caused by the other planets, especially Jupiter and Venus, 

 makes the eccentricity to fluctuate when sufficiently long 

 periods of time are contemplated. In the lapse of hundreds 

 of thousands of years the shape of the elHpse oscillates 

 between the circle and the oval, so that not once or twice, 

 but many times, in the course of geological time the orbit 

 described"^ by our earth has assumed a relatively extreme 

 type. We are able to form an estimate of the eccentricity 

 possessed by the orbit when all the circumstances conspire to 

 make it as eccentric as possible. In the extreme case one 

 season of the year may amount to as much as 199 days, 

 leaving but 166 for the other. The contrast between this 

 condition of things and the present condition will be best 

 realised by observmg that the difference between the lengths 

 of the seasons at present is only seven days, while in the 

 extreme case it may amount to so much as 33 days. 



We have now obtained the numerical elements on which 

 the astronomical explanation of an ice age depends. We 

 have seen that under all circumstances 63 per cent, of the 

 total heat on the hemisphere must be received in summer 

 and 37 per cent, in winter. This statement is true whatever 

 be the shape of the ellipse described by the earth or what- 

 ever be the direction in which the line of equinoxes crosses 



