1200 REPORT—1885. 
that its breaking stress and its elongation or contraction should le between 
assigned limits. 
In the ordinary testing of materials, we observe the loads successively imposed 
and the corresponding deformation of the bar. If such a series of observations is. 
plotted with the loads for ordinates and the elongations for abscissz, and the points 
so found are connected by a curve, we get a stress and strain diagram showing at a 
glance the whole of the observations and indicating by its form the character of the 
material. 
Now it is easy to see that such a diagram, like the indicator diagram ofa steam 
engine, may be drawn autographically by the machine itself, and that the auto- 
graphic diagram will have some advantages over the diagram plotted from 
discontinuous observations. It will be a continuous record at all stages of the test, 
it will be free from accidental errors of recording, and at all events it will save a 
great deal of trouble in making numerous successive observations. 
The first autographie apparatus for testing machines was probably that attached 
by Thurston to his Torsion apparatus. But a Torsion machine is not usually 
employed in engineering tests. Not long after Polmeyer constructed a testing 
machine for tension tests with an autographic arrangement attached. This the 
author saw at Dortmund in 1883, Since then, Fairbanks in America, and Wicksteed 
in this country, have constructed autographic apparatus for ordinary testing 
machines. 
The author designed an apparatus for recording tension tests in January 1883. 
That apparatus gave a diagram two feet square. Since then he has constructed 
a smaller and handier apparatus which is placed on the table. 
In the author’s testing machine the loading of the specimen is effected by a 
single weight of a ton travelling along a lever or steelyard about 20 feet long. The 
motion of the screw which drives the travelling weight is transferred exactly to a 
brass cylinder like the paper cylinder of a steam-engine indicator. The amount of 
the rotation of this is, therefore, exactly proportional to the load on the specimen. 
The pencil slides at the same time axially with a motion proportional to the exten- 
sion of the specimen. This motion is obtained from a very thin wire, attached to 
the specimen, and strained over pullies, so that the motion of the pencil is double 
the actual extension. The wire is kept strained to a given tension by a weight of 
two or three ounces. By suitable choice of position of the autographic apparatus 
on the testing machine, the small vertical movement of the specimen in testing 
has no sensible influence on the record. It appears from check measurements that 
the record of the load is exact, and that of the extension has an error of at most 
sath of an inch. 
The accuracy of the record of extension depends mainly on two arrangements, 
(1) The selection of a position for the recording apparatus such that no horizontal 
movement other than that to be recorded is communicated to the wire. (2) On 
the use of an extremely fine wire, which can be strained to straightness and kept 
free from bends by a very small weight. The form of the clips which define the 
length on which the extension is measured is also of importance. In the clips 
shown the specimen is held between a plane, a knife edge and a point, so that the 
clips are rigid as regards any tilting on the specimen, and at the same time precisely 
define the length on which the measurement is made. 
3. Notes on Mild Steel. By G. J. Gorpon. 
The first note was with reference to the corrosion of steel, which was shown 
to have now been proved by experience to be about the same as for wrought iron. 
2nd. It was pointed out that shearing thick plates had the remarkable effect of 
rendering them cold short, and that this was completely cured by either annealing 
the plates or hy planing off the rough edges left by the shears. 
5rd. It was pointed out that the tendency of users of steel was towards em- 
ploying that of higher tensile strains, and that there was no reason why this should 
not be done provided the same ductility and toughness are insisted on, as in the- 
case of the milder steel now in use. 
