RESISTANCE OF TUNGSTEN 
239 
lengths are much less than those of the radii and were therefore 
neglected in calculating the probable error of the density. The 
results are shown in Table 2. 
TABLE 2. 
SET 1. 
No. of 
Wire 
Length 
cm. 
Radium 
cm. x 10 3 
Masses in Grams 
Density 
Grams/ cm 3 
Left Plan 
Right Pan 
Mass Used 
A 
19.24 
2.396±0.003 
0.007012 
0.007012 
0.007012 
20.26 ±0.06 
B 
34.65 
7.836± .007 
.12922 
.12925 
.12923 
19.345± .034 
C 
29.90 
12.61 ± .015 
.29057 
.29063 
.2906 
19.477± .044 
D 
34.89 
17.61 ± .018 
.64655 
.64655 
.6465 
19.025± .039 
E 
38.65 
22.70 ± .025 
1.19529 
1.1953 
1.1952 
19.126± .042 
SET 2. 
A' 
35.26 
5.086±0.006 
0.05575 
0.05575 
0.05575 
19.47 ±0.045 
B' 
32.27 
10.33 ± 
.015 
.20848 
.20844 
.2084 
19.278± .056 
C' 
34.59 
15.03 ± 
.009' 
.46855 
.46853 
.4685 
19.091± .022 
D' 
40.40 
21.90 ± 
.075 
1.15222 
1.15220 
1.1522 ‘ 
18.96 ± .13 
E' 
38.62 
25.84 ± 
.023 
1.54572 
1.54571 
1.5457 
19.106± .034 
If the radii are plotted as abscissas and the densities as ordi- 
nates the curves in figure 40 are obtained. (Here too the radii 
of the circles represent the probable errors of the respective; 
densities. ) Both curves indicate that the density reaches a 
minimum when the radius is about 0.02 cm. The points for the 
curve of set 1 are scattered while those of set 2 fall closely along; 
the curve. In working with the same set of wires, Doctor Sieg 
finds that the rigidity increases as the radius decreases ; that the 
wires of set 1 are more irregular in their behavior than those 
of the other set, and that the relative positions of the radius- 
rigidity curves are the same as those of the radius-density curves 
as well as being of the same general shape. This is what we. 
would expect. 
