Decembeb 11, 1896.] 



SCIENCE. 



863 



For ordinary arc 17.2 sec. 



Arc between Zn and C 2.3 '' 



Induction coil spark 10.7 " 



An uncharged Zn or Cn plate 30 to 40 

 cm. from tlie spark discharge became 

 strongly charged positively within a few 

 seconds. 



48. For detecting the dust developed by 

 the ultra-violet rays a glass vessel was pre- 

 pared containing a quartz window. The 

 vessel was filled under pressure with care- 

 fully filtered air. A dish of water within 

 the vessel kept the air nearly saturated 

 with water vapor. On allowing the air to 

 expand suddenly dust could be detected, if 

 present, by the fog resulting from moisture 

 condensed on the dustparticles. By proper 

 precautions air could be obtained dust free. 

 But when ultra-violet rays had passed 

 through the quartz window for about ten 

 minutes the presence of dust could always 

 be detected, even when no metal was 

 within the vessel. The authors came to 

 the conclusion that quartz itself gives ofi" 

 particles under the action of ultra-violet 

 rays, 



49. The use of the glass vessel in which 

 the body to be tested was to have been 

 placed was therefore abandoned. A jet of 

 steam was then used to detect dust. The 

 arrangement of apparatus is shown in the 

 figure. A quartz lens L was placed in an 

 opening in a grounded metal screen, and 

 condensed the rays from the source upon 

 the metal plate, M. Dust particles from the 

 lens were prevented from reaching -the 

 steam jet, J, by a glass screen, S. The plate 

 M was first carefully freed from adhering 

 dust. Preliminary tests showed that the 

 steam jet itself exerted no discharging ac- 

 tion. When the zinc plate M was nega- 

 tively charged, a strong dust reaction was 

 obtained on illuminating it. The dust first 

 showed itself in that part of the jet nearest 

 the plate, and gradually extended. The 

 effect was stopped by glass or mica placed 



in the path of the rays. No dust reaction 

 was detected when the Zn was grounded. 



|4r-----^i:::;i----r 



Fig. 1. 



When M was charged positively there was 

 also no effect. A potential as low as 300 

 volts (negative) was sufficient to enable the 

 dust to be detected. 



50. Various other metals were tested in 

 in the same way and showed the dust re- 

 action in different degrees. They are ar- 

 ranged below in descending scale, Ag being 

 least sensitive : 



Zn, Hg, Pt, Brass, Cu, Sn, Pb, Fe, Au, Ag. 



51. With a positive charge no effect was 

 observed in any case. 



52. With a very sensitive steam jet the 

 dust could also be observed from unelectri- 

 fied metals, being best seen with copper. 



53. Several liquids were tested, filter 

 paper being soaked in the liquid to be in- 

 vestigated, and then charged, Fuchsin 

 and Methylviolet showed the effect ; water 

 not at all. These liquids were tested also 

 as to the influence of ultra-violet rays in 

 producing negative discharge.* Water was 

 quite inactive, while the other two liquids 



* Liquid surfaces had previously been investigated 

 to some extent by Stoletow [C, R. 106, p. 1593] and 

 by Wiedemann and Ebert. [Wied, Ann, 35, p. 211.] 

 See below. 



