450 M. A. Secchi on the Periodical Variations 



ing Mairan, one cannot help seeing the serious difficulty which 

 he finds in explaining why the auroras have their maximum 

 at the equinoxes, and not at the epochs at which the earth 

 passes through the nodes of the solar atmosphere ; but do we 

 truly know the place of the nodes of the zodiacal light ? He 

 assumes that they are the same as those of the solar equator, 

 but this is not proved ; and if the zodiacal light constitutes a 

 ring, it might well be otherwise. On the magnetic hypothesis, 

 the greater frequency of the aurora at the epochs of the equi- 

 noxes would have relation to the position of the poles of the sun 

 with reference to the earth, these poles being in fact more 

 directed towards the earth at the equinoxes, and being more or 

 less oblique to it at other times. 



Those who hold the theory of the production of electricity by 

 vapours, may say that these become rarefied in the morning 

 and condensed in the evening; and hence may arise opposite 

 electric states, the fluid passing in the morning from the earth 

 to the atmosphere, and in the evening from the atmosphere to 

 the earth. This may be true ; but why should this condensa- 

 tion always take place at nine in the evening ? The hygrometric 

 curves of the different months show at all events a variation in 

 the hour of maximum according to the seasons. 



A hypothesis, however, can be found which would conciliate the 

 various facts, viz. that atmospheric changes may generate elec- 

 tricity, but that the direction of the current, which of itself 

 would be indeterminate, may be determined by the magnetic 

 action of the sun. But to expand this further into a hypothesis 

 would be at present premature. We will only say that it is not 

 improbable that the earth is subject to the magnetic action of 

 the sun in a manner unknown to us ; but now that magnetic 

 phenomena are developing themselves under so many aspects, 

 we may hope that the explanation of these mysterious actions 

 will soon be found. Not only magnetism, but diamagnetism 

 also may cooperate, and still more the induced currents which 

 exist in bodies of every kind. Two things only I wish to notice. 

 First, the value assigned by Gauss to the magnetism of a cubic 

 metre of the earth*, is such as to make one believe that the 

 whole mass of the earth is really magnetic, and that this force 

 results not only from ferruginous substances, but from the whole 

 globe itself. He proves, in fact, that the eighth part of a cubic 

 metre of the earth has a magnetic moment equal to that which 

 is possessed by a bar of steel 1 lb, in weight and 30 centims. in 

 length, magnetized to saturation. He justly observes, that such 

 a result must surprise physicists, aud that it would require 8464 



* General Theory of Terrestrial Magnetism, Taylor's Scientific Memoirs, 

 vol. ii. part 6. art. 5. p. 225. No. 31. 



