192 REPORTS ON THE STATE OF SCIENCE.—1913. 
increase the resistance greatly. It is pointed out elsewhere in the Report 
(Note on ‘ Heat-Treatment ’) by Dr. F. Rogers that for adequate study of 
this branch of the subject very precise information concerning manufac- 
ture and of treatment previous to the specific treatment given must be 
available. 
Tests with Repeated Cycles of Combined Stresses. 
The question has been raised, notably by Turner (No. 90), whether a 
common factor may not be found for all kinds of stress systems when these 
systems are applied in simple harmonically varying cycles. Since the 
Wohler limiting ranges have been shown to coincide with the elastic ranges 
(at any rate for direct stresses) this question becomes very pertinent. 
_ The main result of Turner’s experiments is very briefly indicated in the 
comments on No. 90 in the bibliography. More experimental data are 
required. The Table (Appendix II.) gives all the information available at 
present. 
Alternate Stress with Repeated Impact. 
The very important conclusion (see No. 83) arrived at by Stanton and 
Bairstow seems to be well substantiated by Roos (No. 64). The practical 
importance of the discovery may be gauged from the following quotation 
from No. 83: ‘The authors are of opinion that conclusive evidence has 
been shown that materials which are strong under alternating stresses are 
in general strong under those shocks which are likely to be put upon them 
in ordinary machine practice.’ 
Practical Utility of the Alternate Stress Test. 
The practical bearing of Wohler’s results has long been recognised, 
witness the Launhardt and Weyrauch formule. (‘ Proc. I.C,E.’ Lxiii. 
1880-1. See also No. 3.) In view of the result of No. 83, it would appear 
that the repeated stress test ought to have enhanced importance. A 
Wohler test is rarely specified by engineers, who rely on the general result 
of research tests and on the convenient ‘ factor of safety.’ Resistance to 
sudden large shock is of at least equal importance with resistance to alter- 
nate stress, and these in general seem to be somewhat opposing require- 
ments. (See Nos. 3,83, and 66.) The former necessitates ductility, while 
the latter requires a high natural elastic limit. These exacting and in many 
cases apparently inconsistent conditions would appear to render the 
Wohler test, as well as a sudden large impact test (for the former does not 
detect brittleness), all the more necessary. But it is unlikely, however, 
that any test for resistance to repeated stress will be extensively used until 
a rapid, simple, and inexpensive test has been discovered. 
Rapid Means of determining Endurance under Stress Repetition. 
(1) Prof. J. H. Smith’s Method. (No. 74.) 
The method seems to be open to certain objection, and confirmation is 
required of its validity (see Notes on No. 74 in the bibliography) ; but 
there is promise that 1t may meet the need for a quick method of finding 
commercially the safe limits for alternating stress. 
(2) Professor J. O. Arnold’s Test. (Nos. 1, 2, and 3.) 
This test does not profess to give the elastic ranges, but only to be a 
practical substitute for the difficult Wohler test. ‘I'he test, however, is 
