45 
oscillations of temperature. It will be seen that every rise 
E or fall in one curve has a correspond- 
ing rise or fall in the other. It is, in 
fact, rare to meet with so close an 
agreement between curves represent- 
ing phenomena so widely different as 
the earth’s horizontal magnetic in- 
tensity and the mean daily tempera- 
ture of the atmosphere. 
The conclusions arrived at from this 
discussion may be briefly recapitu- 
lated as follows : — 
1st. — That the calorific intensity of the sun’s light is 
subject to periodical changes, the maxima and minima of 
which correspond respectively with those of solar spot 
frequency. 
2nd. — That the intensity of a ray of direct sunlight on 
its arrival at the earth’s surface, in the latitude of Oxford, 
is greater in April and September than in June, when the 
sun’s meridian altitude is greatest. 
3rd. — That the curve representing the mean monthly 
values of solar radiation on cloudless days has its times of 
maxima and minima corresponding with those of the curve 
representing the mean monthly diurnal ranges of the mag- 
netometer. 
4th. — It seems probable that the heating rays of the sun 
consist of two kinds differing considerably in intensity, and 
being subject to periodical changes, the times of maximum of 
one kind, and those of minimum of the other, corresponding 
respectively to the times of maximum frequency of solar 
spots. 
5th. — That the oscillations of mean daily temperature are 
intimately connected with the changes which take place in 
the earth’s horizontal magnetic intensity. 
I have said that the results derived from the Greenwich 
