324 Mr. W. F. Barrett on some Physical Effects produced 



account of the extreme richness of this source in rays of high 

 refrangibility, I made use of the secondary spark from a powerful 

 RuhmkorfFs coil. But even here no blueness was evident along 

 the line of the spark. Moreover the passage of the discharge 

 over a piece of granite was unable to bestow or restore its power 

 of showing the blue radiance when the solid was afterwards 

 exposed to the hydrogen-flame *. 



An accidental experiment revealed the source of this curious 

 appearance. Amongst the various substances I examined, I 

 found that vulcanized india-rubber, when exposed to the flame, 

 exhibited the blue colour with great brilliancy. But here I had 

 reason to believe the blue colour was occasioned by the com- 

 bustion of the sulphur with which the india-rubber is asso- 

 ciated. To see if such were the case, a plate of platinum had its 

 surface rubbed with a stick of sulphur, and was then exposed 

 to the flame. Although no blueness was visible before, yet 

 after the rubbing the coloration shone out brightly for a short 

 time, having in every respect the same appearance as the 

 luminosity seen on granite, &c. Afterwards I found that the 

 slightest contact of sulphur, or a body containing it in the free 

 state, was sufficient to impart to any substance the power of 

 showing the blue coloration beneath the hydrogen-flame. When 

 a plate of platinum which showed the blue colour from being 

 wiped over with a piece of vulcanized rubber, was examined 

 with a high power under the microscope, some spots were seen 

 which were unaffected by a drop of water, but which readily 

 dissolved in bisulphide of carbon. These and other experiments 

 convinced me that, at any rate, the chief part of the phenome- 

 non must be attributed to the presence and combustion of 

 minute quantities of sulphur. It may be urged that if the 

 blueness be simply due to the combustion of sulphur, the dif- 

 ferent flames tried ought, equally well with the hydrogen, to 

 show the luminosity. But such is not the case, and as yet I 

 have no experimental grounds for explaining why this difference 

 in action should exist. It cannot be wholly due to the high 



* On granite and some other bodies, the spark, when caused to pass 

 through the substance, ramified for a short distance, more especially from 

 one pole, into beautifully sinuous lines of light, the current splitting itself up 

 as it struck the block, in order to pass through the paths of least resistance. 

 This occurred when the secondary terminals, separated beyond the striking- 

 distance, were caused almost to touch the surface of the granite. When 

 the terminals were approximated to each other until the spark crossed the 

 interval, grazing over the surface of the granite, it was found that the stri- 

 king-distance was considerably greater than in air alone. This may be 

 accounted for by the fact that the discharge deflagrated a portion of the 

 granite in its passage, rendering the spark not only longer, but far more 

 brilliant than when the granite was absent. 



