183 



and the true explanation of the top slope must lie between 

 the two extremes. 



It seems clear, then, that the ionisation in the molecule 

 is proportional to the energy spent in it (/.c, to the stopping 

 power, or the amount of the effect A), to the velocity of the 

 a particle inversely, and to a quantity k, constant for any one 

 gas, but varying from gas to gas. It is this quantity which 

 is given in the last column of the tables above. 



The velocity of the a particle might enter into the for- 

 mula because A is effective in producing B in proportion 

 to the derangement of the atom or molecule consequent on the 

 presence of the particle within it, and therefore to the time 

 during which the intrusion lasts. There is something odd 

 about this conclusion, which suggests a reconsideration of the 

 position. 



At this stage, therefore, it is natural to raise the question 

 whether the effect A really is the cause of the effect B, whe- 

 ther, that is to say, the energy spent by the a particle goes 

 to the production of ions, or the ionisation energy comes from, 

 some other source and the a particle merely pulls the trigger 

 in its passage through the molecule. The fact that the ionisa- 

 tion produced varies as the time of passage is certainly indi- 

 cative of the truth of the latter hypothesis; whilst the occur- 

 ence of the stopping power in the expression for the ionisation 

 is not necessarily evidence against it, because the factor k 

 might be taken in conjunction with s, and ks might be found 

 to rejDresent not some derivative of the energy spent by the 

 particle wdthin the molecule, but some inherent property 

 of the molecule which determined the ionisation produced 

 in consequence of the pulling of the trigger. 



The quantity ks represents in the first place the specific 

 ionisation of the molecule ; that is a relative measure of the 

 ionisation produced in a molecule when an a particle passes 

 through it at a given speed. Now, it is an extraordinary 

 thing that the values of ks which I have obtained for different 

 molecules prove to be nearly related to already well-known 

 molecular constants, such as the molecular volumes, molecular 

 refraction constants, and so on. 



In the following table the values of k, s, and ks of a 

 nur^ber of substances are given in the first three columns; 

 the fourth contains the volumes of the molecular volume ?•, 

 and the fifth the ratio v/ks. The values of the volumes were 

 for the most part taken from the tables in Ostwald's Lehrbuch 

 der Allgemeinen Chemie, 2nd edition, p. 356, etc., but those 

 of C2II2 and C2II4 were calculated from the general equation 

 for obtaining the molecular volumes of organic compounds, 

 and the values for CO^, O^, and Hj were adopted on the 



