at the Royal Institution, 1908-1916 771 



soon killed. Instead of being- intermittently luminoas, they then 

 became permanently luminous and the luminous effect was still obvious 

 three to four hours after death. The impression I gained was that 

 the " luminous material " might well be the product of a slowly- 

 decaying enzymic change. 



It is conceivable that the luminous appearance is the consequence 

 of an oxidative change of an oscillatory character — even that it is 

 conditioned by the formation of ozone. In this connexion, I may 

 refer to the mention made in my previous essay (p. 385) of Sir James 

 Dewar's fascinating demonstration of the luminous effect produced 

 by the oxidation of phosphorus under, certain conditions of low 

 pressure. I there remarked : " It may be supposed either that some 

 particular form of the phosphorons molecule is the origin of the glow 

 or that it marks a stage at which some special oxide is formed." I 

 have since been led to think that the " special oxide " may not be an 

 oxide of phosphorus but ozone, which is a well-known " product " of 

 the oxidation of phosphorous. 



I am thus led to the demonstration given by Sir James Dewar on 

 June 11, 1909, of the phosphorescent glow developed on passing an 

 electric discharge through air or oxygen and allowing this to stream 

 into a tall cylinder, placed on the plate of an air-pump, from which 

 the contents are rapidly removed, so as to reduce the pressure to 

 from 1 to 5 mm. of mercury, the air current rushing down at a 

 velocity of about 1000 feet per second (Fig. 4). 



A steady brilliant phosphorescent stream is seen down the whole 

 length of the cylinder. Hydrogen and all volatile organic compounds 

 containing hydrogen either stop or diminish the phosphorence. 



Carbon dioxide gives a marked but much less intense glow. It 

 is known from Sir Benjamin Brodie's experiments that ozone is 

 produced on subjecting this gas to the electric discharge. Cyanogen 

 and car])on tetrachloride vapour do not become luminous but both 

 sulphur dioxide and carbon disulphide vapour give good phosphores- 

 cent streams at suitable tensions. The glow in highly-purified 

 oxygen is short but distinctly more. brilliant than that given by 

 air. 



There can be little doubt that the phosphorescent glow in oxygen 

 or its compounds is due to the " production " of ozone. This gas, it 

 may be added, is a coloured substance and reversibly convertible 

 into oxygen : — 



^0 = dr^ 20^ jl^ 



It is conceivable that phosphorescence may arise, as in the case of 

 the fluorescent dye-stuffs, as a consequence of the repeated occurrence 



3 E 2 



