420 REPORT—1883 \ 
freely suspended magnetic needle will attain its most easterly position about 7 
or 8 in the morning, and its most westerly about 1 or 2 in the afternoon, while in 
the southern hemisphere the same end of the needle will be affected in a manner 
exactly the reverse of this, the extreme west being attained about 8 in the 
morning, and the extreme east about 2 in the afternoon. It thus appears that the 
type of this variation is of an opposite nature in the two hemispheres. Further- 
more, when the sun is north of the equator during our summer the northern type 
predominates, and to some extent invades the other, so that the variation in the 
northern hemisphere is increased, while thatin the southern is diminished. During 
our winter precisely the reverse takes place; the variation in the southern 
hemisphere being increased, while that in the northern is dimmished. Thus in 
either hemisphere the diurnal variation is greatest in summer and least in winter— 
that is to say, it is greatest when the sun acts most powerfully at the place 
of observation. 
Again, the range of this variation is greatest at times of maximum sun-spots, 
but the effect which the state of the solar surface produces upon the range lags in 
point of time behind its solar cause, so that a maximun of magnetic range does not 
take place until some time after a maximum of sun-spots. The most obvious 
inference from this mode of action would seem to be that the magnetic effect is due 
in some way to the indirect influence of solar radiation, and that this radiation is 
strongest when there are most sun-spots. So much for the best known effect of the 
sun upon the magnetism of the earth. 
The second effect to which I will now allude was first noticed by the late John 
Allan Broun, who showed that changes of the earth’s horizontal magnetic force, 
whether tending to its increase or diminution, takes place nearly simultaneously at 
the various recording stations of the earth. Here the horizontal component may 
in all probability be regarded as giving us a convenient means of measuring changes 
of total force, so that what these observations seem to imply is that the total 
magnetic force of the earth changes simultaneously at these various stations. I 
have recently found, on comparing Broun’s results with the state of the sun’s 
surface, that an increase of the earth’s horizontal magnetic force corresponds to an 
increase of sun-spots, and a diminution of the earth’s horizontal force to a 
diminution of sun-spots, the effect here, as in the previous instance, lagging some- 
what behind its cause in point of time. The difference between the two solar 
effects now described would appear to be that in the former (the diurnal change) 
we have a superposed variation of a different type from the earth’s system, while 
in the latter we have a variation having the same type as that of the earth, or at 
least which may possibly be regarded as having the same type. Now the earth’s 
magnetic system is a polar one, and hence if the sun affects this system as a whole 
we may imagine that he does so by a variation of his influence (whatever this may 
be) over the north magnetic pole, or over the south magnetic pole, or over both 
poles together. Should the state of the sun’s surface vary—say, for instance, in 
the direction of an increase of power—we may imagine that this would influence 
both poles. ~ 
But, apart from intrinsic changes of the solar surface, the sun during our 
summer may be imagined to exert a particularly powerful influence over the north 
magneticpole, and during our winter over the corresponding pole in the southern 
hemisphere. Again, a strong influence at either pole may reasonably be supposed 
to affect the whole system, so that we might, perhaps, on theoretical grounds 
expect a strengthening of the earth’s magnetic system twice a year—namely, at the 
solstices, the one being due to the polar action of the sun in the northern, the other 
to his polar action in the southern hemisphere. 
Now a semiannual variation of this nature has in fact been observed by Broun, 
who has made his analysis so carefully that his results cannot be attributed to mere 
instrumental changes. We have from these an increase of the earth’s horizontal 
component at the solstices as compared with the equinoxes. In order to elucidate 
this point, I have gathered together the various trustworthy determinations of 
the annual and semiannual variations of declination, horizontal force, and dip at 
stations in both hemispheres. These are exhibited in the following table :— 
