190 Professor R. W. Wood [May 19, 



a distance of nearly 2 cm. It was now necessary to prove that this 

 was not light scattered by the dust particles in the air. To do this 

 we have only to photograph the spectrum of the aureole. If it is 

 similar to the spectrum of the spark, we are safe in attributing it to 

 scattered light. If it differs we know that it must be fluorescence, or 

 the genesis of waves of different wave-length from any present in the 

 light of the spark. A photograph of the region surrounding the 

 spark was made with a quartz spectrograph, and it was at once found 

 that the spectrum was wholly different from that of the spark — in fact 

 it was almost identical with that of the oxy-hydrogen flame. For the 

 further study of the phenomenon, a piece of apparatus was devised 

 by which the light of the spark could be more effectually shut off. 

 A small hole was bored through a plate of aluminium fastened to the 

 end of a short vertical brass tube. This plate formed one electrode, 

 the spark passing between an aluminium rod lying along the axis of 

 the tube and the under side of the plate at the point perforated by 

 the hole. 



In a perfectly dark room, if the eye was held a little below the 

 plane of the plate, no luminosity could be seen in the air above the 

 hole, if it was reasonably free from dust, yet a photograph taken with 

 a quartz lens showed a bright beam, or squirt, of light issuing from 

 the hole. A photograph of the phenomenon is here shown, and you 

 will notice the strong resemblance which it bears to a comet. (Fig. 8,/7). 



Many weeks have been spent in an attempt to determine the 

 exact origin of this radiation, and the question has proved to be the 

 most baffling one which I have ever attempted to solve. The work 

 is still in progress, and many remarkable observations have been 

 made, each one leaving us more in the dark than before. As an illustra- 

 tion I may mention a circumstance discovered by Dr. Hemsalech and 

 myself last winter in Paris. We found that if a jet of air was blown 

 through the squirt of light the luminosity was destroyed in the 

 region traversed by the moving current of air, but was of un- 

 diminished intensity both above and below this region. This makes 

 it seem as if the emanation which comes from the spark, and which 

 causes the luminosity of the air, must act for a brief time upon the 

 air in order to cause the luminosity. It also shows that the emana- 

 tion, whatever may be its nature, is not swept aside by the air current. 

 We have also found that other gases become luminous when sub- 

 jected to the spark emanations, the spectrum in each case being 

 different and peculiar to the gas used, electrolytic hydrogen, for 

 example, giving a strong luminosity. 



It is thus apparent that by employing this " photographic eye " of 

 quartz many new phenomena may be brought to light which have 

 previously hidden themselves behind the limitations of the human 

 eye. A study of the absorption by the candle-flame of ultra-violet 

 has also been made. In this case the light emitted by the candle 

 falls out of the problem, for its flame emits little or no ultra-violet. 



