THROW-TESTING MACHINE FOR REVERSALS OF MEAN STRESS. 
counter over times ranging from 10 to 100 minutes, and the total time between 
starting the machine and rupture. 
The following tests carried out as part of Set B, Table III., have not been included 
in that table :— 
Reversals. 
Speed. 
a. 
132,700 
1855 
b. 
127,000 
1855 
c . 
17,800 
1777 
d. 
51,250 
1737 
e. 
329,000 
1775 
These tests were carried out after Nos. 24, 25, 26, and 27 of Set B. Owing to the 
specimen used in (a) breaking in a shorter time than was expected, the test was 
repeated with the result (b). The lower bearing, which had for tests (a) and (b) been 
locked, was now allowed perfect freedom to rotate, and tests ( c ) and (d) were carried 
out. The lower bush was now removed and a new keyway and key were made and 
very accurately fitted. Then test (e) was carried out. The large plate weights on 
the oscillating spindle were changed for others of smaller diameter but of the same 
weight, and the remaining tests of Table III., Set B, were completed. It is evident 
from the above that in the tests ( c ) and (cl) the free period of torsional oscillation of 
the specimen corresponded with that of the machine (see p. 276); 
In Tables IV., V., and VI. are given the corresponding results obtained for tests 
for Lowmoor iron, cast-steel, and cast-iron, respectively. 
The statical tests for these materials gave the following results :— 
Yield stress. 
Maximum 
stress. 
Breaking 
stress. 
Percentage 
elongation at 
maximum 
stress. 
Percentage 
elongation at 
rupture. 
f Maximum . 
16-46 
23-58 
23-08 
24 
29-4 
Lowmoor iron < Minimum . 
16-40 
23-55 
21-10 
21 
27-5 
[_Mean. . . 
16-43 
23-56 
22-27 
22-8 
28-5 
f Maximum . 
40-20 
60-80 
60-80 
— 
5-9 
Cast-steel . . < Minimum . 
39 • 45 
55-30 
55 • 30 
— 
2-5 
Mean. . . 
39-85 
58-10 
58-10 
— 
3-8 
The breaking stress for annealed specimens of the type used in the endurance tests 
was 23T tons for Lowmoor iron, and 48 tons for cast-steel. In the case of the cast- 
iron used, the breaking stress was 9'4 tons. As the specimens used were short, the 
extensions were not measured. 
By far the greater number of specimens broke without any appreciable change in 
diameter or length. A fair number, however, had their diameters greatly increased, 
VOL. CXCIX.—A. 2 P 
