246 Prof. J. J. Thomson. [June 20 r 



fererices in the quality of the gas also conspire to produce these 

 patches is shown, I think, by the following phenomenon. A capillary 

 tube of fine bore containing mercury vapour and a little water vapour 

 developed a well marked red patch ; the tube was then heated for 

 some inches in the neighbourhood of the patch. In general heating 

 the tube makes the discharge yellow from the sodium vapour given 

 off from the glass ; in this case, however, the whole of the heated 

 portion, with the exception of the patch, turned yellow ; the patch 

 itself withstood the heating and continued to show the bright red 

 colour characteristic of hydrogen. 



Electrolytic Transport of one Gas through another. A tube of the 

 shape shown in fig. 1 was made of the finest bore thermometer- 

 tubing ; the extremities, C and D, of the tube in which the electrodes 



Fia. l. 



were fused were bent down so as to be parallel to each other, and so 

 near together that a slight motion of the tube suffices to bring either 

 of the extremities in front of the slit of the spectroscope. The tube 

 was mounted on a board moved by a lever ; by moving this the 

 observer at the spectroscope could readily bring the spectrum of 

 either the positive or negative electrode into the field of view. A 

 side tube, AB, was fused to the middle of the main tube and was pro- 

 vided with two taps ; in the space between these taps a small quantity 

 of any gas which it was desired to introduce into the main tube could 

 be imprisoned, and could, by opening the tap A, be introduced into 

 the discharge tube. The experiment consists in filling the main tube 

 with a gas at a low pressure, observing the spectra at the two elec- 

 trodes, then introducing by the side tube a very small quantity of 

 gas into the main tube, and again observing the spectra at the two 

 electrodes. 



A tube was filled with hydrogen and showed no trace of the 

 chlorine spectra ; a very small quantity of chlorine was then let in 

 through the side tube (in performing this experiment it is necessary 

 to be careful that only a very small quantity of chlorine is intro- 

 duced). After the discharge had been running through the tube 

 for a short time, the chlorine spectrum was found to be bright at the 

 positive electrode, though no trace of it could be detected at the 

 negative. When the discharge was kept on for some time, the chlo- 

 rine spectrum, though still visible at the positive electrode, got 

 fainter ; it-did not appear at all at the negative. If a considerable 



