36 GAS EVOLVED. [CH. II 



containing soda- water freshly drawn from a "syphon." 

 Devaux 1 has shown that this depends on the internal 

 atmosphere rapidly assuming the gas-pressure of the 

 water, by the diffusion of C0 2 from outside into the 

 intercellular spaces. For the same reason, apparently, 

 any movement of the water, e.g. stirring it with a glass 

 rod causes an increase in the escape of gas. It is on 

 account of this fact that we avoid the use of freshly drawn 

 spring water, which has, in a less degree, the effect of 

 "syphon" water on the yield of gas, and vitiates any 

 inquiry into the causes which increase or decrease the 

 rate of bubbling. 



(42) Light of different intensities. 



Now move the beaker into the shade, or cover it with 

 a sheet of white paper, and take a fresh series of readings, 

 and finally replace it in sunshine and record the rate once 

 more. White paper placed round the beaker (which 

 remains open above) may be expected to reduce the rate 

 by about one-half. In the absence of sunshine, an incan- 

 descent electric light of 5 10 candle power may be used. 

 We find however that the Incandescent Gas Company's 

 burner is the most convenient source of artificial light. 

 It is necessary to interpose a glass trough filled with 

 water between the light and the plant, to prevent undue 

 heating from the gas. For the same reason the water in 

 the trough must be constantly renewed by means of a 

 pipe connected with the water supply, and an overflow. 

 W T ith either the incandescent gas or the electric light the 



1 Ann. Sc. Nat. 1889. 



