18 
as 7 to 3, and for the whole half-year as 5 to 3. There 
is an excess or defect of about one-fourth of the mean value. 
Hence of 10 + f of 8 = 18' 5, will be the total heat for 
the northern,, and f of 10 + f- of 8 = 17*5, for the southern 
hemisphere, when the northern summer solstice is in the 
perihelion, and the northern winter solstice in the aphelion. 
Thus the northern half of our globe will receive from the sun 
one thirty-sixth, or nearly 3 per cent, of heat in excess of 
the mean value. Thus the period selected as the Ice Age is 
one in which the northern hemisphere receives from the sun 
an amount of heat exceeding by almost 3 per cent, its mean 
value, and greater than at any other period in the long course 
of 700,000 years. 
31 . Thus the result cannot depend on a lessened total amount 
of solar heat incident on the earth at the eras in question, for 
the total is increased. Sir J. Herschel, Arago, and other 
leading men of science, have failed to see that increase of 
excentricity within the actual limits could produce an ice age 
in either hemisphere. Mr. Croll admits that it could not, 
directly, be the cause of such a change ; but he argues that, 
indirectly, it may be the cause, by bringing other causes into 
operation. 
His reasoning is as follows. From the values of the 
excentricity at past periods he deduces the ratio of the direct 
solar heat at midwinter to its present amount. One column 
of his table gives the excentricity, from Leverricr’s formulae, 
at intervals of 50,000 years for three millions of years back- 
wards, and one forward, and of 10,000 years for one million 
backward. Another column gives the ratio of the midwinter 
solar heat at each period to what it is now. The temperature of 
space is assumed to be — 239° F. The excess above this 
limit is assumed to depend on the midwinter solar radiation, 
and to be strictly proportional to it. The midwinter heat of 
our countiy is taken at 39° F., or the excess as 278°. The 
ratios for the two selected eras, 850,000 and 210,000 years 
ago, are *837 and '864; hence the deficit at the two eras 
Avould be 45 0, 3, and 37 0- 7, and the results — G°‘3 and 
+ 1°‘3 F. for the midwinter heat of our country at thoso 
two eras. With such a degree of cold, ico and snow 
would rapidly form. Tho heat of the summer, Mr. Croll 
argues, would be unable to melt the winter ice, and it would 
go on accumulating through many successive years, till tho 
orbit and aphelion place wore changed, and the main condi- 
tion was thus reversed, after 10,000 years. 
Here Mr. Croll reverses his argument against PoissoAs 
theory, that space is not a body, and can have no temperature, 
