Memoirs presented to the Cambridge philosophical society on the occasion of the jubilee of Sir George Gabriel Stokes, bart., Hon. LL. D., Hon. SC. D., Lucasian professor.

FROM RECORDS OF THE GREENWICH OBSERVATORY, 1871-1895. 131
VI. NUMERICAL APPLICATIONS.
Some investigators have cone to the conclusion that several meteorological and
magnetic phenomena shew a periodicity having a tiine not far different from 26 days
and, not uncommonly, this period is supposed to be connected with solar rotation.
I proceed to apply the methods of this paper to test the reality of this period.
Hornstein%, on the strength of the declination records for Prague, assigns to it an amplitude of '7 minute of arc or an intensity of '5. Such an intensity would be equal
to 500 times the expectancy, if an interval of 25 years is submitted to examination;
and if real and approaching Hornstein's value in magnitude, it should stand out above
the accidental periods to such a degree that every doubt would be removed. Adolph
Schmidtt was led by a discussion of Hornstein's results to a duration of 25a87 days
as being the most probable periodic time, while von Bezold finds a slightly shorter
period for the frequency of thunder-storms.
More recently Professors Eckholm and Arrhenius have published a paper in which
a periodicity of 25'929 is put forward as probable, or even proved. As opposed to these
investigators Professor Frank H. Bigelow gave a considerably longer time (26 68 days) to
the periodicity and has endeavoured to shew that it exists in many meteorological
phenomena.
To shew whether the Greenwich records confirm. or disprove these results, it is
necessary to calculate the intensities for each periodic time, and its corresponding half
period. This I have done, the results being collected in the first section of Table XIV.
TABLE XIV.
Square of Square of
Period Amplitude K Semi-period Amplitude K
25-87 -001001 '95 12-935 -000316 -83
25 929 -001027 -93 12-965 -000200 -55
26-68 -000242 -23 13-340 -000132 -35
25 809 -006168 5-86 12-905 -001060 2'80
25-825 -004182 4-07 12-913 -001286 3-39
26-181 -001144 1'09
26-255 -001081 1 04
26-814 -005936 5-64
27-061 -002943 2'80
* Wiener Ber. LXIV. p. 62 (1871). + Kongl. Svenska. Akad. Vol. xxxi. No. 3 (1898).
t Ibid. xcvi. p. 989 (1887).
17-2