STUDIES IN PLANT RESPIRATION AND PHOTOSYNTHESIS. 87 



one of potassium permanganate. The gasometer was then filled 

 with air drawn from out of doors. Special experiments were carried 

 out to insure that there was no stratification of carbon dioxid in the 

 gasometer and that this gas was of the same concentration at the 

 beginning and end of the experiment. The gasometer was sealed 

 by means of heavy mineral oil. 



By means of glass tubing, the gasometer was connected to a spiral 

 of metal tubing which stood in the water of the thermostat, so that 

 the air attained the temperature of the bath before entering the 

 leaf-container. The connections made by means of heavy wall 

 rubber tubing were wired and covered with several coats of Bakelite 

 paint. 



For these experiments a single leaf was placed in a chamber or 

 cell of special construction. This consisted, essentially, of a metal 

 frame, 15 by 25 cm., two sides of which were glass plates, 5 mm. 

 apart. This leaf-cell is shown in figure 21, in elevation (E) in section 

 (S) and in transverse section ( T) . The upper part of the metal frame 

 carries a metal trough, similar to that of the respiration chamber, 

 already described. This trough was also electroplated with copper 

 and nickel and covered with lacquer, so that it could hold mercury. 

 Into this trough fits a metal cover (C in fig. 21). Distilled water 

 was poured over the surface of the mercury, and when the cover was 

 placed in the trough a completely air-tight seal was made. Through 

 the cover passes a metal tube by means of which the air-stream 

 enters into the leaf-container. A narrow-bore tube through the 

 lower part of one edge of the container provides the means for 

 carrying off the air-stream. At the bottom of the leaf-container is a 

 glass tube with a flared opening. Into this was placed the petiole 

 of the leaf, which was suppUed with nutrient solution through the 

 tube N. This tube was filled with nutrient solution and sealed at 

 the upper end with a piece of rubber tubing and a screw-clamp. 

 During the course of an experiment, the level of the solution in which 

 the petiole stood went down; this was adjusted from time to time 

 by opening the screw-clamp at the upper end of the tube. 



To the exit-tube of the leaf-container is attached a water-trap con- 

 taining phosphorus pentoxide. The experiments were carried out 

 in a saturated atmosphere. This was assured by filling the gasometer 

 with moist air; moreover, the water over the mercury seal and the 

 nutrition solution provided ample water-vapor. The tubes through 

 which the air-stream passes after leaving the leaf-container are all of 

 small bore, 2 mm. It was therefore essential to avoid the formation 

 of water-drops by condensation in these tubes. This was accom- 

 plished by means of the water-trap filled with phosphorus pentoxide. 



From the leaf-container the air-stream passed to the control- 

 valve. This is shown in figure 22, and consists of a very fine needle- 



