Vegetable Assimilation and Respiration. 313 



the initial oxygen diffusion effect solutions were made up as required by 

 diluting this stock with water richly charged with oxygen, as already described.' 

 In the preparation of bicarbonate solutions the exact amount of sodium 

 bicarbonate required was weighed out each time and dissolved in the 

 oxygenated water just the moment before it was used. 



Many shoots of water plants are curved near the tip. It is important to 

 set up the shoot with the plane of the curvature at right angles to the light 

 rays, so that the distance from the light remains constant after any re-setting 

 of the shoot or alteration of its curvature. 



Artificial light is needed for prolonged uniform illumination. To obtain 

 strong enough light some observers have placed cylindrical vessels of water 

 in front of the water-chamber containing the plant. These condense the 

 light, as lenses, into a vertical strip upon the shoot. Experiments with this 

 type of apparatus showed that the rates of bubbling were very sensitive to 

 the slightest displacement of the cylinder or of the plant. A variation of the 

 rate of bubbling by as much as 100 per cent, might be produced by shifting 

 the plant 1 to 2 cm. laterally while the rest of the apparatus was unmoved. 

 Such condensing cylinders were therefore abandoned. The light used 

 in these experiments was that of a cluster of three large incandescent 

 mantles fed with gas under a pressure of 8 inches of water (the Keith light). 

 This gives a radiating surface about 3 inches wide by 5 inches high, and is 

 very suitable and steady. The three burners are grouped on an equi-lateral 

 triangle plan and the distance of the plant from the centre of this triangle is 

 recorded. In some cases, where it has any significance, the intensity of 

 illumination is expressed by calculation on the "inverse square of the 

 distance" basis. The intensity at a distance of 10 cm. is arbitrarily taken as 

 equal to 10 units. This law does not, of course, apply with any accuracy to so 

 large a source of light. 



If the temperature of the water and plant rise much during experimenta- 

 tion, errors of various natures, biological and physical, will creep into the 

 readings. It is important to have an effective running water screen in front 

 of the plant chamber to cut out most of the rays absorbable by glass and 

 water, but it is inevitable that the shoot should be somewhat warmed by the 

 additional radiation that it absorbs. In these experiments the experimental 

 solutions did not rise in temperature more than J° to 1° C. The temperatui'e 

 of all the experiments lay between 15° and 18° C. 



The rate of bubbling was timed at intervals with a stop-watch. The 

 number of readings shown on the figures may appear rather scanty, but 

 bubbles came off so regularly and slowly that each point recorded is in reality 

 a series of as many readings as bubbles observed. 



