854 



SCIENCE. 



[N. S. Vol. XIV. No. 361. 



data, Gauss took \ gt^ = 235, ^ = 53°33^ t = 4 sec- 

 onds, n< = f|f spatial minutes, and V = ^^^-g 

 radians. Gauss found E. D. =3.91 lines, the 

 experimental value being 3.99. The formula 

 gives S. D. = .00046 lines, the experimental 

 value being 1.5 lines. 



The resistance of the air was found by both 

 Gauss and Laplace to make no appreciable 

 difference in the S. D. Several writers deduced 

 formulae which seemed to give a much larger 

 S. D. than those of Gauss and Laplace, but in 

 every case some error in the reasoning has 

 been detected.* The deductions of Gauss and 

 Laplace have, thus far, stood the test of criti- 

 cism. 



Other than gravitational agents were con- 

 sidered by Oersted and Sir John Herschel.f 

 They suggested that the ' electric currents * * * 

 known to be circulating around the earth in the 

 direction of the parallels of latitude ' induce 

 currents in a falling metallic body and cause de- 

 flection to the south. But they became doubtful 

 of this explanation by the remarks of Grove, 

 who said that " inasmuch as a falling body was 



* See Gauss's criticism on Olbers, Gauss' ' Werke ' 

 Vol. v., p. 495 ; reference to Guglielmini's specula- 

 tions in Gilberfe's Ann., Vol. XII., p. 372; M. Petit's 

 articlein Comptes Bendus, Vol. XXXIII., p. 193, 1851, 

 and M. Dupr^'s criticism of that article in Vol. 

 XXXIV., p. 102, 1852, as well as Rosenberger's criti- 

 cism of Dapr6, in Gesch. d. Physik,No\. III., p. 436; W, 

 C. Redfield in Am. Journal of Science^ Vol. III., p. 283, 

 1847, and the correction on p. 451 of same volume ; 

 the theoretical part, contributed by Professor Cowie, 

 in W. W. Eundell's article, Mechanic' a Magazine, Vol. 

 XLVIII., p. 488. 



■\ Am. Journal of Science, Vol. III., 1847, p. 139 ; 

 Report British Assoc, 1846, Misc. Communicat, p. 2. 



moving between electrical currents, placed both 

 north and south of its line of fall, in his opinion 

 the effect of the one would counterbalance 

 that of the other, so as together to produce no 

 effect." 



I myself have been considering the effect of a 

 metallic ball falling through the varying mag- 

 netic field of the earth. Electric currents will 

 be generated in the ball. Resolve the motion 

 of the ball in the northern hemisphere into two 

 components, one component, AC, parallel to 

 the lines offeree, the other, CB, perpendicular 

 to them. The motion along AO produces no 

 current in the ball. That along CB generates 

 a current in a plane normal to the earth's mag- 

 netic lines of force. By Lenz's law, there is a 

 resistance to and diminution of the motion pro- 

 ducing the current. Hence, in all regions 

 north of the magnetic equator there results a 

 northerly deviation. Similar reasoning shows 

 that south of the magnetic equator a falling 

 metallic body experiences a southerly deviation. 



Moreover, if the dip is greater in the lower 

 levels of the atmosphere, then it will be seen 

 from the two positions of the ball in our figure, 

 that there is a relative motion of rotation be- 

 tween the ball and the earth's lines of magnetic 

 force. The ball has in the northern hemi- 

 sphere an apparent rotation about an axis point- 

 ing east and west, in a direction counter-clock- 

 wise when seen by an observer looking 

 westward. Hence by Lenz's law the ball will 

 experience a real rotation in a clockwise direc- 

 tion about the same axis. The interaction be- 

 tween the rotating ball and the air will cause 

 the ball in the northern hemisphere to drift 

 southward. 



I have been unable to secure accurate data 

 for the determination of the magnitude of the 

 two effects, but, taking the largest rate of vari- 

 ation in the magnetic intensity and the dip, 

 along a vertical line, given by Humboldt,* 

 both effects are found much too small to 

 cause a deviation measurable in an experi- 

 ment. 



It is evident that the problem of the southerly 

 displacement of falling bodies needs rein- 

 vestigation, experimentally, and perhaps also 

 theoretically. The Washington Monument, in 



* Cosmos, Vol. v., pp. 97, 115, London, 1872. 



