Magnetic Influence of Solar Light. 241 



time of their announcement, to be in Rome. The subject 

 was altogether a curious one, and Morenchini's experiments 

 were repeated in various quarters, and with various success. 

 MM. Carpa and Ridolphi produced results similar to those 

 alluded to above, but M. Berard of Montpelier, M. Hom- 

 bre Firmas at Alais, and Professor Configliachi of Pavia, 

 failed utterly in obtaining any evidence whatever of magne- 

 tic action. The balance of credit appears to have inclined 

 in favour of the latter observers, since the discovery fell into 

 disrepute until in 1825 it was again prominently forced on 

 the attention of scientific men by some ingenious and ap- 

 parently very decisive experiments of Mrs. Somerville. She 

 was at the time unacquainted with the details of Morenchini's 

 method of observing, but as it appeared to be improbable 

 that the same cause acting upon a needle could give rise 

 both to a northern and southern polarity, she protected one 

 half of each of those employed by her from the influence of 

 the sun's rays by covering it with paper, allowing the light 

 to act on the other half. Proceeding thus, she obtained the 

 most marked results, of which a brief detail may here be 

 given. Having covered half of a sewing needle, about an 

 inch long, with paper, she exposed the other half uncovered 

 to the violet ray of a spectrum thrown by an equiangular 

 prism of flint glass on a pannel at five feet distance. As 

 the place of the spectrum shifted by the motion of the sun, 

 the needle was moved so as to keep the exposed part con- 

 stantly in the violet ray, and after being thus exposed for 

 two hours, the previously unmagnetic needle exhibited de- 

 cided polarity, the exposed end having the properties of a 

 north pole. Repeated experiments were made with needles 

 of different sizes, and placed in different positions with re- 

 ference to the magnetic meridian, and with results uniform- 

 ly of the same kind as the preceding. Needles were then 

 in turn exposed to the other rays of the prismatic spectrum, 

 and it was found, that while the indigo ray was nearly as 



