1910] on Light Reactions at Loiv Temperatures 923 



case of oxygen is interesting, Tliis is transparent to heat, but has 

 many absorption bands for light in the visible spectrum, and also 

 has the power of absorbing the ultra-violet. Nitrogen, on the other 

 hand, is relatively transparent. An ordinary spectrum is shown 

 on the screen. A flask containing liquid oxygen is introduced 

 into the beam, and several dark bands are produced, showing the 

 absorptive power of the liquid. A similar vessel of liquid nitrogen, 

 however, shows no such bands. These two spherical vacuum flasks 

 of liquid oxygen and liquid nitrogen respectively can be placed 

 in the parallel beam of the arc, and it will be seen that they both 

 act as an ordinary lens and converge the light to a focus. On 

 holding a piece of black paper in the focus, very soon a hole is 

 burnt through and the paper ignites, the heat rays being evidently 

 transmitted. We will now project a few photographs of absorption 

 spectra of both liquid nitrogen and liquid oxygen admixed with it in 

 various proportions, also similar slides of absorption spectra of yeast 

 juice, gelatin, glycerin, etc., etc. These photographs were taken with 

 a quartz spectrograph, using for the most part a cadmium-magnesium 

 spark. All the photographs show marked absorption in the region 

 of the ultra-violet. 



Photo-electric Cells. — Cells can be constructed whereby light 

 energy can be transformed into electrical energy. This cell has 

 plates of silver coated with chloride of silver, placed in dilute sul- 

 phuric acid in a quartz tube, and is connected to a reflecting 

 galvanometer showing a spot of light on the scale. When light is 

 allowed to fall on one of the plates a deflection takes place. The 

 chemical decomposition produced by light is transformed into elec- 

 trical energy, causing an electric current. Other forms of light 

 cells lilled w^ith liquid mixtures sensitive to light have been used in 

 recent experiments. Their general construction is shown in Plate I. 

 figs. I. and II. A is a fine platinum wire (secured in the paraffined 

 cork in the base of the tube), drawn tight against the inner 

 surface, and fixed over the top edge by soldering on to stouter 

 copper wire wound round the outside of the top of the tube. 

 A thick platinum wire B is placed at the back to act as the second 

 electrode. The cell is conveniently mounted in a paraifin block in 

 which are two depressions for mercury cups to connect the two 

 electrodes. A is thus made to enclose a thin film of liquid between 

 itself and the wall of the tube. Any change in the composition 

 of this external film will not diffuse rapidly into the cell liquid, 

 and thus differences of electrical potential so produced will be 

 detected by means of the galvanometer. One of the two cells on 

 the table contains a saturated aqueous solution of chlorine peroxide, 

 the other a 20 per cent, solution of uranium nitrate in methyl alcohol. 



When a beam of light from the lantern is directed on to the 

 chlorine peroxide cell a good deflection of the galvanometer is ob- 

 served. On shutting off the light the disturbance soon passes away, 

 as a uniform diffusion in the cell is effected. The uranium nitrate 



