507 



differences are small, we give only the vapour tensions. Sometimes 

 the resistance is given here in fewer decimals than elsewliere; tlie 

 slightest change in indication of the oil-regulator described in Coram. 

 N°. 119 is the cause that the galvanometer in the region of the 

 great decrease of resistance does not settle down. 



The tables show (cf. also tig. 1) that the rest-resistance of In — 

 1922^/ above its vanishing-point temperature is much greater 

 than thai of the other wires, that for In — 1922 — / the temperature 

 at which the resistance diminishes most, has been shifted about 

 0,02 degree with regard to the corresponding one for In — 1922 — 

 II and — HI, and that the fall extends over a larger tempera- 

 ture region. Calculations with the available data by the aid of 



o.fe 

 VV 

 W, 



0.12 



0.00 



300 



p.. 



Fig. 1. 



Silsbee's hypothesis') or by the aid of curi-ent-densities render impro- 

 bable that the said displacement is caused by oxidation of /?i — 1922 — 1 

 throughout its length to such a degree that only a small nucleus 

 of indium remained'); the ratio of the W's in In — 1922 — 1, 



1) K. B. SiLSBEE. Scient. Pap. Bur. of Stand. No. 307 (1917). 



') In contrast with the other wires In — 1922 — I presents a dull oxide-like 

 surface. After the construction in July 1922 the resistance was preserved in 

 benzine; though this was supposed to have been distilled, it seems to have con- 

 tained impurities, which have attacked the wire. 



