278 
MEASUREMENT OF EtlGH TEMPERATURES. 
[bull. 54. 
since been proposed, the question regarding the thermal variations ofil 
viscosity which depends on the said law remains theoretically un% 
solved. Hence, between forms of an equation for rf' as a function of 
temperature, it is permissible to select the one among many applica- 
ble forms which confers the greatest practical advantage. 
In the present instance (to give an example bearing on the remarks 
just made) there is another form of equation which, besides its inherent; 
simplicity, might be applied to reproduce the observations in hand. This 
is rf'z=rjQ (1+yO") ( -y/i-f.tf/9"), which means that L=L (1+yd"), and it! 
seemed to me by no means idle to attempt to get some comparison with] 
regard to the relevancy of these two forms. I think this can best be 
n" 
done by computing the zero value of -L- (which must be a unit of course) 
Vo 
from the high-temperature values of J- , on the basis of each of the laws 
cited. Table 91, the data of which are taken from Tables 86 to 89, con- 
tains the results. • 
n" 
Table 91. — Calculated zero values of J—. 
[a =0.00367.] 
6" 
r," 1 
En or. 
r," 1 
tjo Vl+a0" 
V xl0 G 
„ o <l+afl")f 
r 
o 
565 
2.068 
0.98 
+0.02 
1.18 
318 
Air (Table 86) 
■i 
592 
995 
2.100 
2.678 
0.97 
0.96 
+ 3 
+ 4 
1.18 
1.24 
304 
243 
I 
1,216 
3.123 
1.01 
— 1 
1.34 
277 
f 
442 
1.969 
1.01 
—0.01 
1.22 
489 
Air (Table 89) . 
■! 
569 
982 
2.150 
2.713 
1.01 
0.98 
- 1 
+ 2 
1.22 
1.26 
394 
270 
1 
1,210 
3.219 
1.04 
— 4 
1.38 
315 
H 2 (Table 87) . . 
■i 
961 
1,212 
2. 727 
3.535 
1.00 
1.14 
±0.00 
-0.14 
1.28 
1.52 
295 
425 
f 
■< 
! 
418 
1.957 
1.05 
-0.05 
1.23 
550 
Ha (Table 88) . . 
512 
520 
2.122 
2.138 
1.05 
1.05 
- 5 
— 5 
1.25 
1.25 
490 
489 
1 
952 
2.796 
1.03 
- 3 
1.32 
336 
Turning first to the air points of Table 91, it appears clearly that 
the errors 
1 
1- 
7/ (l+«0") f 
in the results referring to Tables 80 and 89 are promiscuous in distribu- 
tion, and not larger than is quite compatible with the extreme difficul- 
ties of experiment. In the hydrogen points the same errors, though 
larger, are of a nature which can easily be interpreted. The large dis- 
crepancy at 1,212° in the results from Table 87 has been referred to 
permeability of platinum to hydrogen at this temperature. The errors 
(932) 
