on Ionization Currents. 131 



If the ions start with the constant velocity v in all directions 

 from the plane yz the maximum distance they travel from this 



plane is — (tj + v) and the minimum distance is — (tf - v )• 



If v is not too large, therefore, we would expect to find that, 

 on keeping X constant and increasing H, the ionization current 

 reaching the electrode would remain constant until the mag- 

 netic field approached the value given by the equation 



d =^T> ' (1) 



where 2d is the distance between the plates, and that then the 

 current would diminish more or less rapidly to zero. The 

 values of X, H, and d corresponding to the drop in the cur- 

 rent strength when substituted in equation (1) should give an 

 estimate of the magnitude of the ratio e/m. 



The shape of the curves representing the experiments do 

 not admit of very accurate estimates being made of e/m. The 

 following conclusions, however, may be drawn from the data : 



In the first place sensibly all the negative ions at low pressures 

 must have masses of a totally different order of magnitude, 

 and very much smaller than those of the positive ions, even in 

 hydrogen. All the negative ions, therefore, must be corpuscu- 

 lar. This agrees with and extends the results obtained by 

 Townsend,* who used quite a different method, and ionized 

 the gases by X-rays. It is important as showing that none of 

 the negative ions are formed by the a-rays splitting up the 

 molecules into positive and negative parts each of molecular 

 dimensions. 



Let us apply the theory to the experiment in which the elec- 

 tric force was 25 volts per centimeter, and no effect of the mag- 

 netic field of 2600 gauss was observed on the positive current in 

 air. Since all the ions reach the electrode, d in equation (1) must 

 be greater than half the distance between the plates (l cm ), and, 

 therefore, substituting X = 2*5 X 10 9 and H = 2600 in equation 

 (1) we find that 



e/m < 740. 



Now the ratio of the elementary charge to the mass of the 

 nitrogen atom is about 700, and it follows that within the 

 limits of error of the experiment, none of the positive ions in 

 air can have a mass less than that of the nitrogen atom.f 



From the secoud experiment in air it appears that a con- 

 siderable number of positive ions are stopped by field strengths 



*Koy. Soc. Proc. A, lxxxv, 25-29, March 14, 1911. 



f Compare Sir J. J. Thomson, Phil. Mag.. Oct, 1910, Feb. 1911, and Oct. 

 1912. 



