210 ON THE ABSOLUTE MEASUREMENT OF HARDNESS. 
Seebeck’s only advance on Frankenheim is a transfer of judgment from 
the hand to the eye, the latter being contessediy more skillful in mak- 
ing estimates. At best, however, the method thus established en- 
counters the following serious disadvantages. In the first place, the 
results obtained depend on a variety of minor conditions, foremost 
among which is the nature of the material out of which the stylus is 
made. Steel is most generally used, but steel can not be exactly 
defined, and therefore the observer has no right to assume that his 
stylus is a body of fixed properties. Moreover, the necessity of using 
both hard and soft steel in the apparatus introduces a further compli- 
cation, but,as a matter of fact, when a hard steel stylus is applied to a 
soft body the pressure under which the stylus moves must be reduced 
below the limit of measurement, whereas hard bodies are only scratched 
by hard steel. Franz used both a steel and a diamond point, and 
endeavored to co-ordinate the results of the two by measuring the 
hardness of a given suitable body in terms of each stylus. It is true 
that the numbers obtained in the two series of experiments show a 
constant ratio (cet. par.), but it does not follow that this would always 
be the case, and it is quite improbable for large intervals of hardness. 
The second difficulty encountered is the dependence of the results of 
the sklerometer on the degree of sharpness of the marking stylus. None 
of the above papers touch upon this matter, nor would it be possible for 
them to estimate this effect. Yet it is quite obvious that the pencils of 
different apparatus can not have been identically sharpened, and that 
the pencil of the same apparatus will soon become blunted by continued 
use. Measurements into which this serious discrepancy necessarily 
enters cannot therefore be comparable among themselves. Finally, 
the modus operandi, the velocity of the moving stylus and the diree- 
tion of the pressures are to be considered, and in some of the above 
papers hints relative to these points (motion, position, and inclination of 
pencil) are explicitly given. Barnes and Perlsin, however, first showed 
that the effect producible by varying the rate of motion of the stylus 
is so great as to be actually capable of inverting the data for hardness. 
Indeed, it has since become well known that the edge of a rapidly rotat- 
ing, relatively soft dise is scarcely touched by a file or a lathe tool, and 
that if the motion be rapid enough, it is the tool which suffers most. 
Nor is this phenomenon to be referred to an effect of temperature, for it 
finds its full explanation in consideration with the rates of motion to 
which it is due. The hardest cast iron can be turned off with a steel 
tool at a velocity as high as 2 meters per second of the moving parts. 
I am thus naturally led to the important question, whether the defi- 
nition of hardness given by the sklerometer is correct in principle. I 
believe this is by no means the case. Quite aside from the serious prac- 
tical difficulties which I have just summarized, it seems to me that 
hardness when determined by scratching is much too complex a concep- 
tion to be used as a basis for the definition of the property. Compli- 
