190 On the Influence of Gravity on Magnetic Declination. 



a uniform law. I believe that I have now given the needed ex- 

 planation ; and since I have shown experimentally that the phe- 

 nomena are such as should be produced by gravitation, it is rea- 

 sonable to assume that they probably are so produced. The 

 probability is increased by the disappearance of the night oscil- 

 lation in summer (V.), the probability that when it is observed 

 it results from thermal disturbances, and the greater stability of 

 those critical hours which are nearest to the hours of maximum 

 sunward gravitation (VI.). The precise coincidence both in 

 time and direction of the lunar-diurnal declination and tidal 

 curves (VII.), the unavoidable inference that the moon has no 

 constant or specific magnetic action (VIII.), the " establishment" 

 of ten minutes at the Philadelphia station (IX.), the correspon- 

 dence of the lunar and solar curves in the diminished winter 

 amplitude (X.), and the uniformity at all stations of the semi- 

 annual variation- curves (XI.), are all necessary corollaries of my 

 hypotheses. 



The demonstration of a connexion between the daily and annual 

 variations of magnetism and gravitation would be incomplete if 

 our theory could not be so extended as to explain the decennial 

 and secular changes. For such an extension we are compelled 

 to wait until further study and observation have more pre- 

 cisely defined the character and value of those changes, and 

 suggested all the important gravitation- disturbances of long 

 period to which they may be plausibly referred. We may find, 

 however, in the attraction of Jupiter, one of the possible causes 

 of the 10 to II year period, while nutation, precession, geological 

 upheaval and depression, change of seasons, accumulation or 

 diminution of polar ice, and the shifting position of the centre 

 of gravity of our planetary system, must all necessarily con- 

 tribute to the production of gradual changes in the terrestrial- 

 gravitation currents. The belief does not, therefore, seem unrea- 

 sonable that the feeble vibrations of the tremulous needle may 

 not only furnish us with a delicate scale for weighing (as we 

 have already approximately done) the huge mass of the sun, 

 but may also aid us in the discovery and verification of other 

 important cosmical phenomena, and the assignment of their 

 appropriate laws. 



Although it is probable that terrestrial magnetism is mainly 

 owing to currents circulating above the surface, it is well known 

 that there are also earth-currents which exert an appreciable 

 modifying influence. I am confident that they will be found 

 equally obedient to the laws of gravitation, which affect every 

 particle of the earth's body, modifying the crystalline polarity 

 and cohesion of solids as well as the flow of liquids, and pro- 

 ducing internal tides, which may contribute largely to that meta- 



