UNIFICATION OF METHODS IN TESTING MATERIALS. XLVII. 
pieces are measured to four millionths of an inch. It would be 
noticed that the limit proposed by Mr. Selman was considerably 
higher than that of Prof. Johnson, both of course being above the 
so called true elastic limit. There was no apparent advantage to 
be gained by adopting the limit proposed by Mr. Selman. It 
was simply a question as to how nearly parallel you could draw 
two lines and still have a definite point of secancy. At the 
previous meeting Mr. Selman had proposed that the one line 
should be made to slope ten per cent. more than the other, at the 
present meeting he substituted five per cent., and there seemed 
nothing to prevent some one else using three per cent. or one per 
cent. It was merely a question of draughtsmanship. 
With regard to Prof. Johnson’s proposed “apparent elastic 
limit,” it was to be regretted that a more detailed demonstration 
was not given in his book of the reliability of the statement that 
when the apparent elastic limit is located in the manner directed 
“it will be found at practically the same point on all tests of like 
kind on similar materials. 1t is therefore characteristic of the 
material.” Judging from some of the figures given in the book, 
(for example fig. 249), this particular elastic limit was subject to 
about the same variation as were the others. Prof. Johnson 
probably used the term “characteristic ” in a somewhat loose and . 
general sense. 
Professor WarkEN in reply to the discussion, said, in answer to 
Mr. Deane that in regard to chemical tests, it had generally been 
accepted that, if the physical tests as fixed by the specifications 
were satisfied, then, also, would the chemical composition be satis- 
factory, but this was not universally true. At the same time it 
was desirable to interfere as little as possible with the province of 
the steel manufacturer, whose business it was to produce steel in 
accordance with the — of aay edu tests specified, 
which should be sufficiently ter to develope 
the real nature of the material as applied to the particular purpose 
for which it is intended to be used. There were however, some 
few exceptions to this rule, such as for example in the case of 
