258 Terrestrial Magnetism in the Horizontal Plane. [Nov. 26, 



But in the cold months, the space included in the curve is much 

 smaller in many cases, probably not more than one-sixth of what it 

 is in the summer months ; and the curve often crosses itself in the 

 most bizarre fashion with irregular loops stretching out, three 

 crossings in one curve occurring very frequently. In the summer 

 •nonths there is a certain degree of symmetry ; but here is, constantly, 

 a preponderance on the west side, which leads me to imagine that the 

 magnetic effect of the radiant heat upon the sea is considerably 

 greater than the effect on the land. 



To obtain some numerical basis for a report, which though 

 exceedingly imperfect may convey some ideas on this wonderful 

 subject, I have adopted the following course. I have confined myself 

 to the months of June and July as probably the two hottest, and the 

 months of December and January as probably the two coldest. For 

 each of the curves applying to these months, I have laid down a 

 system of rectangular co-ordinates, corresponding to the Greenwich 

 meridian and to the line at right angles to the meridian (or the 

 geographical E. and W.). The extreme north ordinate and the 

 extreme south ordinate are measured, and their sum is taken, and 

 interpreted by a scale of measure formed in accordance with the 

 theory of the intruments ; and this interpreted sum forms the " Range 

 of Meridian Force " in terms of the whole Meridian Force. In the 

 same manner, the " Range of Transversal Force " is measured. As the 

 time of each of the two-hourly or hourly records is marked on the 

 curve, there is no difficulty in fixing approximately on the solar 

 times corresponding to the extreme N. and S. values, and the 

 extreme E. and W. values, mentioned above. These are all the 

 elements of the magnetic record which are described in the subjoined 

 table. 



The changes in the monthly records are very remarkable. They 

 leave no doubt in my mind that the diurnal magnetic changes are 

 produced by the sun. But I cannot account for every change that 

 takes place in the course of a day ; nor can I undertake to say 

 whether we can found, on these, the theory that general terrestrial 

 magnetism is a part of solar radiation, perhaps sometimes acting 

 through or sometimes impeded by the masses of land and sea on 

 which that radiation acts. 



Still I think that a considerable step is made by the establishment 

 of a connexion between terrestrial magnetism (on one hand), and the 

 radiation, or, at least, the visibility of the sun (on the other hand). 



