Anode Rays to the Investigation of Isotopes. 859 



difficult to get them to give a beam at all unless mixed with 

 an alkali salt. From such a mixture the parabolas due to 

 the alkalis are generally considerably stronger than those 

 due to the alkaline earths. These differences appear to be 

 properties of the metals, and not to depend much on the 

 particular halogen compounds used, though some halogen 

 compound seems to be necessary if the metal is to appear 

 at all. The differences vary appreciably between different 

 anodes of similar composition and even for different photo- 

 graphs with the same anode, so that it is of no value to 

 attempt as yet to give a quantitative measure of these 

 tendencies. 



When a new anode is first set up and a discharge passed 

 through it, a considerable quantity of gas is evolved, and it 

 is necessary to run the discharge for some time before the 

 Gaede pump can keep the pressure down to a point at which 

 a photograph can be taken. The time varies from about 

 half an hour to four hours. The pressures at which photo- 

 graphs are taken correspond to dark spaces of from about 

 2 to about 5 cm. At very low pressures the current 

 suddenly ceases to pass, and a little air has to be let in to 

 start it again. It is generally found that the parabolas 

 improve in brightness for the first two or three photographs, 

 and a good anode will give ten hours running or more 

 before showing signs of exhaustion. As the anode gets 

 older, less and less gas is evolved, and it ceases to be 

 necessar}- to pump for more than a short portion of the 

 time of the exposure. 



The surface of the anode glows with a colour charac- 

 teristic of the metal used, and shows the corresponding 

 bright line spectrum, as does also the beam of rays itself. 

 The beam is emitted roughly normally from the surface of 

 the anode, and can be focussed by slightly coning out this 

 surface. It is often necessary to scrape the surface in order 

 to get the beam to fall on the end of the fine tube. At low 

 pressures only the centre of the end of the anode glows, 

 the rest remaining black. The colour of the rest of the 

 discharge is generally a yellowish grey. 



Electrical connexions. — The discharge was produced from 

 an induction coil. Two cathodes were generally used : one 

 at A (fig. 2) and the other an aluminum cap pierced by the 

 end of the fine tube B. They were usually both earthed. 

 The object of having the cathode A was to make the 

 discharge go easier when the dark space oi the other 

 cathode reached to the anode. Some photographs were 

 taken with only the cathode at A in use, the othor being 



3 L 2 



