ELECTRICAL PROPERTIES OF HYDROGENISED PALLADIUM. 185 



These numbers prove two things — 



1. The greater the charge of hydrogen, the higher on the thermoelectric 



diagram is the hydrogenium line. 



2. The displacement of the line for a given increment of charge is 



greater at the higher charges. 



So striking is this second peculiarity that, whereas it requires the resistance 

 of the wire to be increased by 50 per cent., so as to bring the line half way to 

 the platinum line, it requires only 78 per cent, to make it move to the other 

 side of the platinum line to a distance greater by 50 per cent, than the distance 

 of the palladium line. This is indicated by the figures in the last column having 

 a negative sign. The great displacement in the hydrogenium line at high 

 charges, as proved by these experiments, is in agreement with the results of the 

 earlier experiments. 



It does not seem possible to draw any conclusion from the values given as 

 to there being any change in the inclination of the hydrogenium line. The 

 safest conclusion to draw would be that, on the whole, the ratio de'jde remains 

 constant through the corresponding range of temperature — that is, through 

 40° C. If this were so, it would mean that the hydrogenium line is not parallel 

 to the palladium line, since the latter is itself inclined to the platinum line. 

 The thermoelectric powers of palladium to platinum at 50° C. and 90° C. are in 

 the ratio of 70/75. If the hydrogenium line were parallel to the palladium line, 

 the corresponding ratio for it would be smaller for all the cases studied except 

 the last, for which it would be greater, being indeed greater than unity. There 

 is certainly no indication of such a property in the tabulated numbers. The 

 properties are indeed almost reversed. 



Again, the constancy of the ratio de'jde for all temperatures (if surely 

 established) would mean that the hydrogenium lines all pass through the point 

 of intersection of the palladium and platinum — a very remarkable fact, should 

 it be established by later experiments. We might then reason as to the 

 existence of a hydrogen line passing through the same point. Certainly the 

 constancy of this ratio is well marked in the last two cases, just the ones where, 

 upon the very natural hypothesis of parallelism, it should be most distinctly 

 variable. A closer study of the subject seems called for, and I hope before 

 long to attack the problem in a somewhat different manner. It may be said 

 that we can hardly expect to avoid irregularities due to a probable inequality 

 in the distribution of the hydrogen charge. Such irregularities will obviously 

 have more effect in thermoelectric measurements than in resistance measure- 

 ments. 



In connection with the subject of the thermoelectric properties of hydro- 

 genium, the following will make an instructive lecture experiment. A palladium 



