6 
DB„ S. CHAPMAN ON THE SOLAR AND LUNAR 
not have been undertaken without considerable assistance. The heaviest part of the 
arithmetical work consisted in the computation of the lunar diurnal variations (§12) ; 
no reduced data of the desired kind were available, so that the variations had to 
be newly computed from the published hourly values of the magnetic elements, 
seven years’ records from each of five observatories being used. Skilled assistance 
was obtained in this and all the other work of computation, wherever possible. In 
regard to this I wish to make grateful acknowledgment of the help placed at my 
disposal by the Government Grant Committee of the Royal Society, by Dr. Schuster, 
and by the Astronomer Royal. Also in the preparation of the data of Tables I. to III. 
relating to the solar diurnal variation, I am indebted to the Astronomer Royal for 
computing assistance, and to Mr. W. W. Bryant, Superintendent of the Magnetic and 
Meteorological Department at the Royal Observatory, Greenwich, who personally 
shared in and controlled the reduction of the published data to this form. 
I wish also to acknowledge the courtesy of the following directors of observatories, 
who furnished me with manuscript records of such of their observations as I had 
need of, and which were at the time unpublished : Dr. Angenheister (Samoa), 
Mr. P. Baracchi (Melbourne), Dr. Schulze (Pilar and Laurie Island), and Mr. Skey 
(Christchurch, N.Z.). 
The preparation of even the tables of initial data, such as Tables I. and II. and 
(much more) IV. to VI., is a task of considerable magnitude, which I believe must put 
a serious obstacle in the way of further advance in the subject. In order that future 
workers may not be repelled by such initial difficulties, it is very desirable that 
directors of magnetic observatories should reduce their own observations more 
completely, and publish them in a readily available form. With regard to the lunar 
diurnal variations I have already made suggestions as to the manner in which this 
might advantageously be done. # The present discussion of these variations seems to 
confirm the desirability of further investigation. The same applies, though to a 
smaller extent, to the solar diurnal variations. In the latter case I would recommend 
that a threefold sub-division of the year should be adopted, as for the lunar diurnal 
variations in § 12 ; and that the variations from quiet days only for the mean of a 
number of years, should be used. 
Part I.— The Present State of the Problem. 
§ 2. Schuster’s first investigation (1889). 
The data used by Schuster were the Fourier coefficients of the first four harmonics 
in the solar diurnal variations of North, West, and vertical magnetic force, taken 
separately over the summer and winter halves of the year 1870, from the observations 
at Bombay, Lisbon, Greenwich, and St. Petersburg. The year 1870 happened to be 
* ‘Phil. Trans.,’ A, vol. 214, p. 295, 1914. 
