538 Prof. J. J. Thomson on the Relation between the 



gas are negatively electrified, they may combine with positive 

 atoms of the metal to form a neutral compound, leaving an 

 excess of the atoms of the metal negatively electrified. Or, 

 if the atoms of the gas are positively electrified, then, if any 

 substance is present which can form conducting chains be- 

 tween the atoms of the gas and those of the metal, the atoms 

 will interchange their charges, the positive charge going to 

 the electropositive substance and the negative to the electro- 

 negative. Those atoms of the gas which were positively 

 charged might unite with those negatively charged to form 

 molecules, so that the gas itself would emerge electrically 

 neutral after contact with the metal. 



It will be seen that each of these methods requires some- 

 thing more than mere contact of the gas and metal, this seems 

 borne out by the phenomena exhibited by electrified gases. 

 It was shown by Giese that electrified gases from a flame 

 could pass through wire gauze connected with the earth, and 

 yet retain their charges : this fact is exemplified in a very 

 striking way by Lord Kelvin's electric strainers, as he finds 

 that electrified gases still retain an appreciable fraction of their 

 charges after passing through these strainers which consist of 

 many layers of wire gauze placed one behind the other. In 

 an experiment I described in the B. A. Eeport for 1894 

 ("The Connexion between Chemical Combination and the Dis- 

 charge of Electricity through Gases "), an electrified gas re- 

 tained its charge though enclosed in a vessel with a large pool 

 of mercury connected to the earth. I have also found that such 

 gases retain a large proportion of their charge after bubbling 

 through a mixture of sulphuric acid and water. In a paper 

 " On the Passage of Electricity through Hot Gases " (Phil. 

 Mag. [5] xxix. p. 444), I showed that when a current is flowing 

 through a hot gas a piece of cold metal interposed between 

 the electrodes completely stops the current, though when the 

 metal gets red hot the current passes through it with ease. 

 In all these cases the charged atoms would be striking against 

 conductors, so that they furnish very strong evidence that this 

 process alone is not sufficient to rob the atoms of their charges. 



The first of the processes alluded to above w T ould tend to 

 make a gas lose a negative charge or the metal a positive one, 

 while the second process would make the gas lose a positive 

 charge, the metal a negative one. This is on the assumption 

 that the metal is electropositive to the gas; if it were electro- 

 negative, the words positive and negative must be inter- 

 changed in the preceding statement. 



The first process involves chemical combination, thus any- 

 thing tending to promote combination, such as a high tern- 



