414 Mr. Gill on the Intensity of the Ultra- Violet Light 



The light emerging from the slit passed through a wire 

 gauze AA and fell upon the zinc plate BB connected to an 

 electrometer; the distance between the plate and gauze being 

 about 1 cm. AA was charged up to +40 volts by connexion 

 to a battery o£ small cells, and BB was initially at zero 

 potential and insulated. The plate and gauze were carried 

 upon ebonite supports. 



Method of conducting an experiment. 



In any set of observations the discharge was always passed 

 for a fixed time, usually 20 seconds. The first centimetre 

 of the discharge beginning from either electrode was placed 

 opposite the slit and light from it fell upon the zinc plate, 

 a certain number of negative ions n^ were set free, were 

 attracted to the gauze AA, and in consequence the electro- 

 meter showed a deflexion o\« After earthing and again 

 insulating the electrometer a deflexion S 2 was obtained, due 

 to n 2 ions being set free by the light from the second centi- 

 metre of the discharge, nothing except the tube having been 



moved. Then -= r 1 = — =the ratio of the intensities of the 

 o 2 n 2 



UV light from these two portions of the tube. 



As explained above, care was taken to keep the pressure 

 of the air inside the tube constant, and owing to the small 

 leak no other gas except air was used in the tube. [It is 

 possible that small changes of pressure that occurred were 

 due to chemical action.] 



The deflexions of the electrometer were always small, 

 which accounts for the choice of such a comparatively large 

 slit (1 cm. broad) in order to get a measurable ratio not too 

 much affected by errors of experiment. The relative inten- 

 sities of each cm. of tube were thus obtained, and in a few 

 cases of interest several cms. overlapping were taken as 

 well. 



It is obvious that the intensity depends upon three variables, 

 the pressure p, the current c, and the potential difference 

 between the electrodes V; these variables not being inde- 

 pendent but being connected by one relation. Looking 

 upon p and c as independent variables, a complete research 

 would determine the distribution of intensity for all values 

 of p and c, but the value of the results scarcely warrants so 

 laborious an undertaking, and ali the experiments were con- 

 ducted for a constant current, p being taken as the only 

 independent variable, A few preliminary experiments, how- 

 ever, showed that for the higher pressures (1 to 2 mm. 

 of mercury) the intensity was approximately proportional to 



