24*2 Magnetic Influence of Solar Light, 



effective as the violet, the blue and green rays only occa- 

 sionally succeeded, and the yellow, orange, and red, never 

 did so, although the same needles were exposed to their 

 influence for several successive days. Subsequently, ex- 

 periments were tried with solar light, transmitted through 

 various coloured media, and it was found, that needles half 

 covered with paper, and exposed under a glass coloured 

 blue by cobalt, became after three or four hours feebly mag- 

 netic, and after six hours, sensibly and permanently so. Si- 

 milar results were produced with green glass, and also with 

 blue and green riband. Such is an outline, although no 

 more, of the experiments of Mrs. Somerville, but I regret 

 exceedingly, that I have been unable to procure any detailed 

 account of the various steps in her method of observing, 

 since although the results appear sufficiently decisive, it 

 would have been very desirable to have had the power of 

 remarking whether all sources of error had been foreseen 

 and eliminated by her ; especially as the method employed 

 by her is one in which many such sources exist, as will 

 afterwards be more particularly alluded to. 



In repeating the experiments of Mrs. Somerville, M. Baum- 

 gartner of Vienna discovered that a steel wire, some parts 

 of which were polished, while the rest were without lustre, 

 became magnetic by exposure to the white or undecom- 

 posed light of the sun, a north pole appearing at each 

 polished part, a south pole at each unpolished part, and 

 the effect was hastened by concentrating the light upon the 

 wire by means of a lens. In this manner eight poles were 

 obtained on a piece of wire eight inches long. 



The magnetic influence of solar light was exhibited in 

 another form by Mr. Christie of Woolwich in 1825, he 

 having found that the vibrations of a magnetized needle 

 when exposed to the sun's light, ceased in a much shorter 

 time than when it was vibrated in the shade, and this inde- 

 pendently altogether of the effect of temperature. This 



