Experiments on Positive Rays. 221 



mercury atoms may linger longer in the charged state than 

 the other atoms and last after the conductivity of the gas has 

 become very small in consequence of the disappearance of the 

 other atoms ; then on the next break of the coil, these charged 

 mercury atoms will be at first acted on by a much stronger 

 electric field than that which exists when the resistance of the 

 gas has been broken down and a considerable current passes 

 through the tube. 



On the Lines produced by the Negatively Charged Particles. — ■ 



There are on the plates curves which, as the direction of the 



deflexions shows, are due to negatively charged particles. 



In one sense all these curves may be regarded as secondaries, 



for it is unlikely that the particles acquire their negative 



charges before passing through the cathode, as if they had 



done so they would have been retarded by the strong electric 



field which exists in front of the cathode. But though they 



are all, in this sense, secondaries, we find on examining the 



plates that the lines corresponding to the negatively charged 



particles may, like the positive ones, be divided into two 



types. Some of the negative lines, corresponding to the 



secondaries among the positives, come close up to the origin, 



while there are others which, like the primaries among the 



positives, are finite arcs of parabolas, terminating abruptly 



when they approach within a certain distance of the vertical 



axis. In fact these lines on the negative side are frequently 



exact reproductions in shape and size of the corresponding 



lines on the positive. An example of this is shown in fig. 14 



(PI. IX.), where the curves a, /3 are the lines corresponding 



respectively to the positively and negatively electrified atoms 



of oxygen. When the discharge passes through very pure 



oxygen, it will be seen that every detail in the positive curve 



is reproduced in the negative, including the prolongation of 



the parabola corresponding to the double charge. We see 



from this that even the fastest of the neutral oxygen atoms 



are able to attract and retain a negative corpuscle. 



This identity in the shape of the positive and negative 

 curves for the oxygen atoms might suggest the view that 

 they are produced by a neutral molecule splitting up after it 

 has passed through the cathode into a positive and a negative 

 atom. We might be tempted to suppose that the negatively 

 charged atoms in general arose from the dissociation of a 

 molecule of a chemical compound in which the atom under 

 consideration was negatively electrified, and was in com- 

 bination with another positively electrified one : we might 

 in fact suppose that we had something analogous to the 

 dissociation of a molecule of an electrolyte in water. 



