Connexion with the Phenomena of Terrestrial Magnetism. 441 



ing projections of the calculated results. The agreement between 

 these two sets of curves is probably as great as could be expected 



iam 



3 



5 



7 



9 



H 



-Lftn 3 

 Derby 2* 



5 7 9 



IL 





















/^f 



















^^ 







,. 



,^' 



^ 



--~ 



N \ 



, Derby * 



Birmingham 

















\^ 



^^^"^ 





in the results of so short a series of observations ; and we seem 

 therefore entitled to conclude that the diurnal movements of the 

 two horizontal magnetometers are accounted for by electric cur- 

 rents traversing the upper strata of the earth. 



There is one point of difference to which it is important to 

 draw attention. It will be seen that the calculated curves are 

 for the most part above the observed. The reason of this will be 

 evident upon a little consideration. The zero from which the 

 calculated results are measured is the mean of the day ; whereas 

 that of the observed results is the true zero, corresponding to the 

 absence of all current. Now the chief deflections of the galva- 

 nometer needle (as appears from the latter curves) are those in 

 which the sun is above the horizon; and the zero line conse- 

 quently divides the area of the diurnal curve unequally, being 

 considerably nearer to the night observations than to those of 

 the day. If the calculated curves be displaced by a correspond- 

 ing amount, their agreement with the observed will be much 

 closer. 



The difference here noted is one of considerable theoretical 

 importance. Magnetometric observations furnish merely differ- 

 ential results, the magnitude and the sign of which have refer- 

 ence solely to an arbitrary zero. We are accordingly ignorant 

 even of the relative values of the effects, and are unable to com- 

 pare them with their physical causes, whether real or supposed. 

 In these respects the galvanometric observations have the ad- 

 vantage. In them, positive and negative are physically distin- 

 guished by the direction of the currents; and this, as well as 



Phil. Mag. S. 4. Vol. 22. No. 149. Dec. 1861. 2 G 



