388 
ME. J. A. BROUN ON THE VARIATIONS OE THE 
time required for a synodical rotation of the sun and revolution of the moon differs hy 
upwards of two days, and the annual law of force obtained by him showed a maxi- 
mum in June and a minimum in December. Kreil then relinquished this idea, and 
sought for the explanation in the difference of the meridian altitude of new moon, 
which is low in winter and high in summer, his results showing a greater horizontal 
magnetic intensity for the moon south than for the moon north of the equator. This 
latter result for the moon’s declination did not accord with that afterwards obtained by 
me from the Makerstoun observations*. 
Although an examination of the Trevandrum observations confirmed me in the belief 
that the sun’s rotation was the great cause of the variations of daily mean horizontal 
intensity, yet trials for the greatest mean oscillation, employing periods of a few tenths 
of a day more or less than 27 - 2 days (the time given then by the best authorities for the 
sun’s synodical rotation), were not satisfactory f, and the more complete investigation 
was deferred till the discussion of a longer series might enable me to draw more 
certain conclusions. Meanwhile, having found that the observations made at Trevan- 
drum in 1844 and 1845 showed almost exactly the same variations as at Makerstoun, 
I returned to the examination of the observations made during these two years, and 
deduced from them, and from the observations made simultaneously in different Colonial 
observatories, that the mean duration of the oscillation was 25*96 daysj. This period 
differed so much from that then accepted for the sun’s rotation that I suggested some 
movement, regular or irregular, of the sun’s magnetic poles or of the meridian of 
maximum force §. 
In 1871 Dr. Hornstein deduced from his observations at Prague in 1870, by calcu- 
lation and graphic interpolation, a period 26 - 33 days, but varying from 2G‘0 to 26 - 7 days 
according to the instrument or method employed in the investigation. The mean 
period agreed almost exactly with that which had then been obtained by Spoerer for the 
sun’s synodical rotation from spots in the equatorial zone. This result, brought to the 
notice of the Royal Society, Nov. 16, 1871, by the Foreign Secretary, and referred to in 
two presidential addresses, was the subject of a note in the ‘ Proceedings ’ of the Society 
(vol. xx. p. 417, June 1872). In a postscript to this note the following passage occurs : — 
“ In my paper on the Horizontal Force of the Earth’s Magnetism, I stated that ‘ a 
careful investigation of a much larger series of observations leads me to believe that 
the period is variable within certain limits ’ || ; whether this variation is due to a change 
* Trans. Roy. Soc. Edinb. vol. xix. pt. ii. p. xxxvi. 
f Writing from memory, I have previously said “26 to 27 days” were tried; this was a mistake. Proc. 
Roy. Soc. xx. p. 418 (1872). J Trans. Roy. Soo. Edinb. vol. xxii. p. 543. 
§ Phil. Mag., Ang. 1858. I think there is little doubt that Kreil would have discovered the variations of 
the earth’s magnetism due to the sun’s rotation if he had not believed that period to occupy nearly the same 
time as the moon’s tropical revolution. My own belief that the oscillations observed by me were due to this 
cause in spite of the difference of periods, was founded on the comparison of observations in both hemispheres, 
where the agreement seemed inexplicable by the lunar action alone. 
|| Trans. Roy. Soc. Edinb. vol. xxii. p. 544. 
