549 
Physics. — ‘Note on the melting point of palladium and Wirn’s 
constant c,.” By G. Horst and E. Oostreruuis. (Communicated 
by H. KaAMERLINGH ONNzs.) 
(Communicated in the meeting of March 25, 1916). 
1. Recently E. Warsure') has published new rules for the 
standardizing of thermometers by the Physikalisch-technische Reichs- 
anstalt at Berlin. In the notes appended, it is stated that the inten- 
sities of radiation of the black body at the melting point of palladium 
and that of gold for 40,6563 u are in the ratio of 81.5 to 1. 
From Wien’s radiation-formula it follows that: 
2 / y Epa mp 4 1 1 7 
a7 910 Seas re p= men —— Lis 
M Eau mp ‘ 1 Au mp 4 Pl mp 
where M= log,,e and L is a constant. 
It follows from the data given by Wareure that 1 = 2,8880. 
This constant may also be derived from measurements of other 
observers. W. W. Copientz’) has made a number of determinations 
of c, which are based on the melting points of palladium (1549), 
copper (1083), antimony (630.0) and zine (419.2) as a scale of tem- 
peratures. All observers agree that on the scale which is fixed in 
this manner the melting point of gold lies at 1063°. Calculating /4 
from his value of c,—=14465 and the melting points of palladium 
and gold, we find Z = 2 8880. 
In the Astrophysical Journal, Vol. 42, p. 300, 1915 E. P. Hypr, 
F. E. Capy, and W. E. Forsyrar publish some measurements, from 
which Z may again be derived. It follows from their results that 
FE, mp 
RDE 
== 76,9 and therefore L = 2,8869. The 
Au mp 
differences. between these values for / may be explained by a devia- 
tion of the melting of palladium of only + 0°.25. A better con- 
cordance, therefore, cannot be expected. *) 
Whatever therefore the thermodynamic temperature of the melting 
point of palladium may be, it will always be necessary to assign 
a value to c, such that 
1) For instance Ann. d. Phys. (48), 1034, 1915, 
2) Bull. Bur. of Stand. (10), 76, 1914. 
3) Other series of observations (see F. Henning, Temperaturmessung p. 240), 
also yield values for L which do not deviate much, with the exception of that of 
HorBorN and Vaentiner, which is 2.7 °/, larger. 
