Forces betiveen Atoms and Chemical Affinity. 761 



Positive Rays that in some cases the transference of elec- 

 tricity from one atom in the molecule to the other can take 

 place. For in the Positive Rays we find along with the 

 positively electrified atoms and molecules negatively electri- 

 fied atoms of such gases as hydrogen, oxygen, carbon, and 

 chlorine, when these gases are in the discharge-tube. These 

 negatively electrified atoms must have been positively 

 electrified at one time or they would not have been driven 

 through the cathode with a high velocity, they must at a 

 later stage have attracted a corpuscle and become electrically 

 neutral, and when in this state, though moving with a high 

 velocity, attracted and retained a corpuscle from the molecules 

 through which they were moving. If they can do this under 

 these unfavourable conditions, it is very improbable that 

 they should not be able when they are in combination with 

 suitable atoms to get a negative charge out of these atoms. 

 Thus we should expect that in the compounds of violently 

 contrasted elements the atoms in the molecule would be 

 charged some with positive, others with negative electricity. 

 We shall see later on direct evidence for the existence of 

 molecules of this type, which I shall call ionic molecules, to 

 distinguish them from molecules in which the atoms are not 

 charged. The process by which the atoms get charged I 

 shall call intra-molecular ionization. 



One very direct piece of evidence is given by the value of 

 the specific inductive capacity of compounds, evidence which 

 is particularly strong and easily interpreted when these 

 compounds are gaseous. The setting of the electric doublets 

 under the electric field will contribute to the specific in- 

 ductive capacity of the gas ; hence molecules in which the 

 atoms are charged, since they tend to have abnormally large 

 moments, will also tend to have large specific inductive 

 capacities. Again, since the setting of these doublets involves 

 the rotation of the molecules as a whole, they will move so 

 sluggishly that they are not likely to be affected by vibrations 

 as rapid as those of light-waves in the visible spectrum: hence 

 these doublets will not affect the refractive index in the 

 visible spectrum, while they will affect the specific inductive 

 capacity: we should expect therefore such compounds to 

 depart widely from Maxwell's law that the square of the 

 refractive index is equal to the specific inductive capacity. 

 Again, since the setting of the axes of the doublets under 

 the electric field will be hampered by the collision with other 

 molecules, and as these collisions are more numerous and 

 energetic at high temperatures than at low, the specific 

 inductive capacity of compounds of class II. ought to be 



