62 MR. 0. W. RICHAEDSON ON THE IONISATION 



around the true value for the number of free ions per cubic centimetre. This would 

 give for the positive ions 10 18 and for the negative 10 S1 ; thus the number of free 

 positive ions would be insignificant compared with the number of free negative ions, 

 and the theory would at once account for this small value for the amount of one 

 metal transported into another by the electric current.* 



To test this theory, which has both simplicity and elegance to recommend it, both 

 the positive and negative ionisations were measured simultaneously in oxygen at low 

 pressures. The very great increase in the positive ionisation, produced by small 

 quantities of oxygen as compared with other gases, would indicate, if we accept the 

 above view, that the effect was probably specific and caused by the formation of an 

 electrical double layer. In this case, increasing the positive should decrease the 

 negative leak. The results of this experiment, which are given on p. 22, show that 

 although the oxygen altered the positive leak in a ratio of 10 to 1, the negative leak 

 was not changed by 20 per cent. As it is very unlikely that some other effect of the 

 oxygen would compensate so well as this, the writer considers that this experiment 

 renders the above view very improbable. It is chiefly on account of this experiment 

 that the view, that the positive ionisation is due to the absorbed gas, and only 

 indirectly to the metal, has been adopted in describing the results obtained in this 

 investigation. 



X. 17. SUMMARY OF THE PRINCIPAL RESULTS. 



The positive ionisatiou, i.e., the number of positive ions produced by 1 sq. centim. 

 of platinum surface per second, possesses a minimum value, which depends on 

 temperature and pressure, in most gases. The positive ionisation in oxygen at a low 

 pressure (< 1 millim.) is much greater than in the other gases tried. In oxygen at 

 low pressures, and temperatures below 1000 C., the ionisation varies as the square 

 root of the pressure ; at higher temperatures and low pressures it varies nearly 

 directly as the pressure, whilst at higher pressures at all temperatures the variation 

 with pressure is slower, so that at pressures approaching atmospheric the ionisation 

 becomes practically independent of the pressure. 



The variation with pressure in air is similar to that in oxygen. In nitrogen and 

 hydrogen the ionisation appeared to increase more rapidly with the pressure at high 

 pressures than in oxygen. In very pure helium at low pressures there was a positive 

 ionisation which was a function of the pressure. 



The experiments on ionisation by collisions indicate that the positive ions liberated 

 by hot platinum in oxygen are of the same order of magnitude as those produced by 

 the collisions. They are not great masses approximating to dust particles. 



The positive leak in oxygen always oscillated round a certain value under specified 



This view can easily he made to give a reasonable quantitative explanation of the change in the 

 positive ionisation produced by oxygen at different pressures. 



