Dixon and Poole — Photosynthesis and the Electronic Theory. 69 



necessary, liquids. A second rim, placed on the nnder side, prevents any 

 liquid spilled from reaching tlie insulation. As the plate is exposed to the 

 light beam, it is important that its support should expand as little as possible, 

 so it is carried on a short quartz tube, which is fixed in a sealing-wax plug 

 passing through the base plate. A thin copper wire down the axis of the 

 tube leads to the electrometer. 



The grid, seen in section in the figure, is made of brass strips 6 mm. deep 

 by 0'6 mm. wide, soldered together round tlie rim. This construction 

 ensures that any expansion of the strips only causes them to curve in a 

 horizontal plane without affecting their distance from tlie plate. This 

 distance is about 5 mm., so when the plate is covered with leaf-powder or 

 another metal plate the air gap is about 4 mm. Sealing-wax is used to 

 support and insulate the grid and connecting wire, and also to attach the 

 window covering the top of the vessel. In the earlier experiments with this 

 vessel this window was made of glass. 



A 500-watt " half- watt " focussing lamp was used as a source, the light 

 being concentrated into a beam of suitable diameter by means of a 4J-inch 

 condenser. A mirror at 45° above the testing vessel reflected tiie light down 

 on to it. A large beaker full of water standing on the window of the testing 

 vessel acted as a screen to reduce the infra-red radiation ; but its presence 

 apparently had little, if any, effect, showing that with this vessel the error 

 due to heating was very small. The beam was so concentrated that almost 

 all of it passed through the window, which is about 6 cms. in diameter. 



A battery of sixty pocket-lamp (three-cell) refill dry batteries was used 

 as a source of pressure, the grid being thus kept at about 230 volts. An 

 electrostatic voltmeter (not shown in fig. 1) was connected to the grid to 

 measure its potential and ensure that it remained sensibly constant during 

 a test, as airy variation in it would cause a large motion of the electrometer. 



The key K (fig. 1) consisted of a mercury cup in a block of paraffin wax 

 fixed on the slate bench on which the apparatus was set up. Into this 

 dipped the wires from the testing vessel and the electrometer, and also a 

 third wire connected to earth as shown. The latter was fixed in such a way 

 that it could be raised out of the mercury by means of a thread passing over 

 pulleys, thus isolating the quadrant system. The act of breaking contact 

 generally caused a deflection of the electrometer, which varied both in magni- 

 tude and direction ; but readings were never started uirtil the disturbance 

 so caused had died down, and the electrometer spot of light was steady near 

 its zero position. The key also served for connecting another wire for 

 charging the quadrant to some known pressure (generally 0*5 volt) for 

 calibration. 



