A.— MATHEMATICAL AND PHYSICAL SCIENCES. 29 



Diurnal Variations. 



The data dealing with diurnal variations are very voluminous, for 

 records at Kew Observatory have been taken for many generations. As 

 the name indicates, a change in the magnetic elements takes place every 

 24 hours ; in the case of Kew, the horizontal force reaches a minimum 

 value about 10 or 11 a.m., and a maximum value about 7 p.m., the range 

 being about 35y in the summer and 12y in the winter. On the other 

 hand, the vertical intensity is a maximum about 10 or 11 a.m., and a 

 minimum at 7 p.m., this variation being also less in winter than in 

 summer ; thus, an annual variation is superimposed on the daily one. 

 It is observed that when the horizontal force increases the vertical force 

 diminishes, from which it follows that no general rise or fall in the intensity 

 of the earth's main field can explain the variations. The principal portion 

 of the variation occurs in the daytime, and the variation apjoears to be 

 beyond doubt a function of the sun's position above the horizon. 



Schuster's analysis of the data shows that the daily variation is 

 probably due to electric currents in the upper atmosphere, but in addition 

 to the magnetic effects of these currents there is an effect due to currents 

 induced in the earth by them. These induced currents are naturally in 

 the opposite direction to the inducing ones, and hence the magnetic 

 effects for the horizontal intensity are additive, while those for the vertical 

 force are opposed. 



Balfour Stewart first put forward this theory that the diurnal variations 

 are due to electric currents in a conducting medium above the surface of 

 the earth, and Schuster and Chapman have done much to develop it. In 

 general, the theory attributes the variations to convective motions of 

 conducting layers of air across the magnetic field of the earth and hence is 

 known as the ' dynamo ' theory. The general form and intensity of such 

 a current system can be inferred from an analysis of the variation data. 

 Chapman's analysis shows the system in the sunlit hemisphere to consist 

 of two closed circuits which (at the equinoxes) may be taken as symmetrical 

 with respect to the equator, their foci lying very nearly on the 11 a.m. 

 meridian. As the electric currents are supposed to be induced by the 

 movement of conducting layers of air in the magnetic field, such currents 

 must also be produced near the ground, but the conductivity of the air 

 near the ground is so low that their effect may be neglected. In the 

 upper regions the movements, while larger, cannot be regarded as 

 immeasurably greater than near the earth's surface, and the increase in 

 current intensity can only be attributed to an increase in the conductivity, 

 a view which Balfour Stewart was forced to adopt, although at the time 

 there was little evidence to support it. 



The magnitude of the dynamo effect is dependent on three factors — 

 (1) the horizontal movement of the air, (2) the conductivity of the air, 

 (3) the intensity of the vertical magnetic field. All these factors vary 

 with latitude, and hence it is to be anticipated that the magnitude of the 

 variations will also vary with latitude, which is the case. The intensity 

 of the field can be calculated with considerable accuracy, but the con- 

 ductivity and movements of the upper air are not known, although such 

 movements are attributed to thermal effects and hence will be a maximum 



