Mil. 0. W. RICHARDSON OX THE IONISATION 



Exj>eriments were also made to see if the negative ionisation in oxygen varied 

 with the pressure of the gas. The temperature was 1100C. and the saturation 

 currents with the wire charged positively and negatively respectively at the different 

 pressures were those shown in the accompanying table. 



The potentials used were +40 and 7 '5 volts respectively. These were tested and 

 found to produce saturation. With the potentials employed there was no possibility 

 of the measurements being vitiated by ionisation by collisions. The results show 

 that, whilst the oxygen increases the positive leak ten-fold, the negative remains 

 unchanged within the limits of experimental error. The independence of negative 

 ionisation and gas pressure, which had been previously observed by MCCLELLAND and 

 H. A. WILSON, will be found to be of considerable importance later in interpreting 

 the results on the positive ionisation. 



6. Current and Temperature. 



The last section forms a fairly complete investigation of the way in which the 

 positive ionisation from hot platinum in oxygen varies with the pressure, when the 

 temperature is kept constant, for a considerable range of temperatures. The 

 phenomenon was next investigated by measuring the ionisation at constant pressure 

 when the temperature of the wire was varied. 



The measurements were made with the glass tube apparatus previously described. 

 Rough experiments were made at several pressures, but only those at pressures of 

 1 to 3 millims. have been retained. At very low pressures irregular results were got, 

 doubtless owing to changes in the composition of the small quantity of gas present in 

 the apparatus, whilst at pressures comparable with atmospheric it was feared that :m 

 rrror might creep in owing to the walls of the tube Incoming heated. At pressures 

 of about 2 millims., however, several wires were tried and found to give consistent 

 results over a range from about 700 C. to 1250C. It was found convenient to 

 measure the negative current at each temperature along with the positive. The 

 results therefore enable us to compare the positive and negative ionisations from 

 wires under identical conditions. The ionisation would, no doubt, be proportional to 

 the area of the hot metal surface if the surface were uniformly heated. As, however, 

 the wires are colder ;it the ends nwing t<> the heat being conducted away through the 

 leads, the effects from wires .if different lengths and thicknesses will not be strictly 



