June, 1892. 
THE ICE AGES, PAST AND FUTURE. 
127 
pulled inwards at others, but the combined effect of the two 
planets in this action becomes an extremely intricate 
calculation, on account of the great number of possible 
combinations of their relative positions ; and an enormously 
long period of years is required for all the possible changes of 
combination to be rung out. The average distance of the 
earth from the sun throughout each revolution in its orbit 
is unchangeable, and all the disturbance caused by the 
interference of the planets is simply a change in the shape of 
the orbit, making it more or less eccentric and elliptical. 
The extreme amount of the possible change in the orbit 
is also strictly limited, because of the overpowering control 
exercised by the attraction of the sun, which is more than 700 
times the weight of all the planets added together ; and 
he compels them to follow their due courses, although allowing 
some minor interferences to go on amongst them. 
For the production of an Ice Age in our northern hemi¬ 
sphere, a combination has to take place of the two causes 
that have been mentioned, namely the change in eccentricity 
of the earth’s orbit, and the change in direction of the earth’s 
axis. The eccentricity of orbit has to be so much that the 
distance of the earth from the sun at the extreme point is 
materially increased, and the proportion of heat received 
from the sun at that point is consequently materially 
diminished ; also the direction of the axis of the earth at the 
time has to be such as to give winter to the northern hemi¬ 
sphere when the earth is at the greatest distance from the sun, 
the north pole of the earth being directed outwards at the 
time instead of inwards as it is at this present time, when 
the earth is farthest from the sun. 
If the north pole happened to be directed inwards at that 
point of the earth’s orbit, it would then be summer in our 
northern hemisphere, and the result would be simply an 
exceptionally cool summer, and a corresponding very mild 
winter ; and this will be the case in the southern hemisphere, 
whenever an Ice Age occurs in the northern hemisphere. 
If the occurrence of an Ice Age depended alone upon the 
change in direction of the earth’s axis, it would occur 
every 21,000 years; and this will be still the shortest period 
possible for the recurrence of an Ice Age, but the requisite 
conjunction of the second cause, the change in eccentricity of 
the earth’s orbit, prolongs the possible period to nine and a 
half of these periods, or as long a time as 200,000 years. 
The remarkable calculations made by Croll of the changes 
in the eccentricitv of the earth’s orbit that must have 
*/ 
occurred during the last 200,000 years show that the eccen¬ 
tricity, which is l-60th part at the present time, was 
