NEGATIVE ELECTRICITY FROM HOT CALCIUM AND FROM LIME. 169 



with the values calculated from the equation as is the case with platinum. This is 

 probably clue to the greater experimental difficulties attending the use of calcium. 



4. On oxidising the calcium on the cathode to lime there is an enormous increase 

 in the negative leak, the leak from a lime cathode at 950 C. being about the same as 

 the leak from calcium at 140UC. The variation of the negative leak from lime with 

 the temperature roughly obeys the Wilson-Richardson law, but the leak at any fixed 

 temperature is not constant, but gradually decreases with continued heating. This 

 diminution is not due to the same cause as the diminution of the leak from a new 

 platinum wire. In the case of platinum the decrease is caused by the gradual 

 evolution of gas probably hydrogen occluded in the metal. With lime it seems ta 

 be due either to a spluttering away of the lime from the surface of the platinum or to 

 a change in the nature of the lime itself. In this connection it is interesting to note 

 that a piece of lime subjected to a strong heat glows very brightly at first, but 

 gradually loses this property of glowing when kept continuously at a high tem- 

 perature. It is not improbable that there is some connection between these two 

 phenomena. Experiments are at present being made with this idea in view. 



5. The following are the values of the constants Q and A for platinum, calcium, 

 and lime respectively, obtained from observations of the negative leaks under a 

 potential difference of 40 volts in helium at a pressure of 3"236 millims. : 



Cathode. Q (calories). A. 



Platinum 1- 22xlO l'6x!0 



Calcium 7-29X10 4 1-7 x 10 4 



Lime 9-58x10* 6-4 xlO 11 



The constant Q represents the work done by a gramme molecular weight of 

 corpuscles in escaping from the surface of the cathode. We see from the numbers 

 given above that this is least in the case of calcium, but owing to the great variations 

 in the different values of Q obtained for lime and for calcium we cannot lay much 

 stress on the difference between the mean values for these two cathodes given above. 

 Tt would, of course, be expected that the corpuscles would escajn; more easily from 

 the metal than from the oxide, for we should expect the presence of the electro- 

 negative atom of oxygen in the molecule to act as an attracting force tending to 

 i-etaiu the escaping corpuscle. 



6. The greatly increased leak obtained by oxidising the calcium cathode into lime 

 is due to an enormous increase in the value of the constant A. Reasons have been 

 given for thinking that this constant cannot be proportional to the number of free 

 corpuscles per cubic centimetre of the cathode as follows from RICHARDSON'S theory 

 of the negative leak. 



7. The negative leak from lime in hydrogen is much greater than that in air or 

 helium. 



8. When the current density through the discharge tube reaches a certain value 

 (only obtained with a calcium or line cathode) the gas becomes luminous. This 



VOL. CCVII. A. Z 



