DIURNAL VARIATIONS OF TERRESTRIAL MAGNETISM. 
43 
As regards the former, the sources of information are the diurnal variations of 
barometric pressure, of air temperature, and of wind. Angot* * * § and HannI have 
studied the daily barometric variations in great detail, and Dines | has discussed the 
daily changes of wind at St. Helena. Gold§ has examined the theoretical relations 
between these phenomena and the air-temperature variations. 
The 24-hour component of the barometric variation is very irregular in its distri¬ 
bution over the earth, varying greatly both in amplitude and phase with season, 
situation (continent, land or ocean, mountain or valley), and weather conditions. The 
12-hour component, on the contrary, is one of the most regular of all meteorological 
phenomena. Its phase is almost exactly the same over the whole region between 
latitudes ±60 degrees (at least), its amplitude shows a regular diminution with 
increasing latitude, while it is practically independent of longitude, weather con¬ 
ditions, and local situation. Mountain records indicate that the 24-hour component 
diminishes with increasing height, vanishing and re-appearing with reversed phase. 
I'he 12-hour component likewise diminishes in amplitude, but almost proportionately 
to the pressure, || while its phase is gradually retarded. Gold assigns 90 degrees as 
the probable total diminution in the corresponding phase. 
The annual changes in the 24-hour barometric variation are not very regular ; 
those in the 12-hour component, on the contrary, are simple and definite. The 
phase is constant throughout the year, while the amplitude has maxima at the 
equinoxes and unequal minima at the solstices, the total variation, however, being 
small. The solstitial minima are simultaneous in the two hemispheres, the principal 
minimum occurring at aphelion in June. 
Angot has shown that there is also a harmonic of period eight hours having 
a regular annual variation, but it is too small to require consideration in this paper. 
The dependence on latitude of the amplitude of the 24-hour component is rather 
uncertain ; Angot gives the law as sin 2 0, 6 being the co-latitude. Schuster, in 
his second memoir (§ 6), used the harmonic Q/ or sin 6, stating however, that the 
harmonic Q : / might also be present. Expressed in millimetres of mercury, the value 
actually used was, at the equator, 
(28) 0‘3 sin t. 
Angot found that the amplitude of the 12-liour component was mainly proportional 
to sin 4 6, but contained in addition a term proportional to sin 2 9, as Adolf Schmidt 
* Angot, ‘Annales du Bureau Central Meteorologique de France,’ 1887. pp. 237 -344. 
t Hann, “ Lehrbuch,” and also numerous papers in the ‘ Met. Zeitschrift ’ and the publications of the 
Vienna Academy. 
+ Dines, ‘ Meteorological Office Publication No. 203,’ 1910. 
§ Gold, ‘ Phil. Mag.,’ 19, p. 26, 1910. 
I! The diminution of amplitude seems to be slightly more rapid for the 12-hour amplitude than for the 
total pressure. 
G 2 
