Fuly 23, 1885] 
LETTERS TO THE EDITOR 
[ The Editor does not hold himself responsible for opinions expressed 
by his correspondents. Neither can he undertake to return, 
or to correspond with the writers of, rejected manuscripts. 
No notice ts taken of anonymous communications, 
[The Editor urgently requests correspondents to keep their letters 
as short as possible, The pressure on his space is so great 
that it ts impossible otherwise to insure the appearance even 
of communications containing interesting and novel facts. | 
Nomenclature in Elasticity 
In reference to a communication of mine which you published 
not long ago (see NATURE, vol. xxxi. p. 504) on this subject, I 
have pleasure in enclosing for publication, should you think fit, 
photos from three automatically recorded stress-and-strain dia- 
grams made in my laboratory. The originals were traced on 
smoked glass, the glass plate then varnished to fix it, and used 
at once as a negative. Test-piece No. 9461, of which Fig. 1 
shows the behaviour, was a very ductile piece of Swedish bar- 
iron, turned to }-inch diameter. The extensions were measured 
3.0 
vo 
o 
a 
rs) 
= 
2.5 
SWEDISH BAR IRON. 
NO o46l. 
2.0 BREAKING LOAD 18.85 TONS 
PER SQ. INCH. 
EXTENSION IN 10 INCHES 20.4PER CENT 
REDUCTION OF AREA AT FRACTCRE 
75.6 PER CENT 
EXTENSION DOUBLE FULL SIZE 
1G 
14 
TONS PER SQ. INCH 
Fic, r. 
on a length of ro inches, and recorded doud/e full size ; they are 
to be measured from the curved base-line, distances along which 
measure the total load on the piece (on a scale, as recorded, of 
about 19 tons per inch), and therefore the load per unit area 
(or, as I prefer to call it, the zwtensity of stress) up to the limit 
of elasticity, to which point the cross-section remains practically 
unchanged. The point where zmsfadiiity comes in is very 
marked, and also the release or going back of the stress after 
the material has “‘ broken down.” Lastly, the condition of 
local flow, or whatever it is to be called, is excellently shown. 
The material draws down in one place, so that the increase of 
extension, being confined to that place, is very small, and the 
total load diminishes, although the intensity of stress, on the now 
greatly reduced area, is much increased, as is shown further on 
in Fig, 4. 
Fig. 2 is an autographic diagram similar to the last, taken 
NATURE 
269 
from a piece of soft steel (No. 8397) 0°60 inch diameter and 
Io inches long. It shows most of the same characteristics, 
except that the breaking down is not preceded by any mterme- 
diate stage ; the loss of elasticity comes very suddenly. The 
whole load was taken off the piece and then reapplied a number 
of times during the experiment, after the limit was passed, and 
the curves show most distinctly by their parallelism the (practi- 
cal) constancy of the modulus of elasticity even up to the very 
maximum load. The curves show also the curious phenomenon 
—which I have often noticed in this form, and which, in some 
of its aspects, has been most carefully examined by Bauschinzer 
and others—of increase of load borne, for a very limited time, 
after the material has been allowed a short rest—here only a few 
seconds. If the rest be for hours or days a similar thing occurs 
in a much more marked fashion. 
Fig. 3 is a similar diagram from a piece of ‘*S. C. Crown” 
iron, showing the same features, although in a less marked 
degree. In this case the load was removed and reapplied after 
the piece had begun to draw down visibly, and the curve to turn 
back, with the result of showing the piece to be still elastic up 
to the load it had just borne. 
MILD STEEL BAR 
| NO. 8397 
GREAKING LOAD 27.52 TONS PER SQ: INCH 
EXTENSION IN |O INS 20 8 PER CENT 
49.0 PER CENT. 
REDUCTION OF AREA 
EXTENSIONS 
as 24Tons 
4 20 per sq. in. 
8 12 IG 
Fic. 2. 
Fig. 4, which is a copy of a diagram obtained by consecutive 
measurements, not automatically, illustrates a question asked, I 
think, by Mr. Ibbetson. The diameter of the bar under test was 
measured each time the extension was noted, and the curve of 
actual stress on actual area (II.) is plotted out, as well as the 
usual load and strain curve (I.). The most interesting point 
about it is perhaps the way in which the curve ends nearly 
parallel to the axis, an excessively small additional extension 
corresponding to a very great additional intensity of stress. This 
arises, of course, from the fact that the extension is here confined 
to a very small length of the bar, the diameter and length of the 
main part of the bar remaining practically unaltered. If the 
extensions were plotted to stresses in this main part of the bar, 
the curve would take the shape (III.). : 
All these curves illustrate distinctly, I think, a point not very 
generally known, that the non-elastic extensions form really a 
