130 
MINUTES OP PROCEEDINGS OF 
with so low a weight that the motion was almost imperceptible, while in the 
case of others, as oak and oak, cohesion could not be overcome until such 
weight was in the scale pan as produced high rates of motion. 
The resistance of friction as proved by M. Morin,* no doubt is indepen¬ 
dent of the rate of motion and depends on the space passed over, and in 
the higher rates of motion the weight in the scale pan was in excess of that 
absolutely needed by friction alone, and therefore expended itself in doing 
the work in a shorter time. Such weights, therefore, cannot be taken as 
giving true comparative values of the resistance of friction of the various 
substances, and the coefficients worked out from them cannot strictly com¬ 
pare unless the rates are identical. 
The whole of the experiments demonstrate that it is quite impossible 
to ensure constant results, because it is impossible to control all the 
conditions producing such results; the variation is, however, greater in some 
than in others. 
There is no doubt that the principal cause of the variation is the fact that 
after passing over each other under heavy pressure the surfaces change from 
their normal state, and when they slide past each other a second time they 
do so under new conditions; different parts are brought in contact, and 
certain raised portions of the surfaces are more condensed. So much is 
this the case, that it was found the progressive variations due to continued 
use were so great as to prevent the effects of different pressures from being 
distinguished. To partially obviate this and other sources of error (as 
varying rates), Series e was arranged and carried out. To use fresh 
specimens in place of repeatedly using the same would, as was found, 
import greater variations in the results. 
The velocities with which the surfaces of compressors slide past each 
other during the recoil of the gun are so much greater than the highest 
rate attained by the specimens under trial, that, as in the case of the 
specimens, adhesion was overcome at either high or low rates; therefore 
in the quicker movement of compressor plates there can be no retarding 
effect from this force, except during the first and last instants, unless, 
probably, much greater pressures per square inch are applied than was 
the case in the experiments; but this is not so in any compressors at 
present in use; the Elswick pattern producing, with 100 lbs. on the lever, 
about 80 lbs. per square inch, while a pressure of 175 lbs. was used in the 
experiments. 
The manner in which the various specimens passed through the others 
was very different. 
Smooth regular motions always resulted from the pieces capable of giving 
low rates, and the resulting coefficients were more uniform, while also the 
period at which adhesion ceased, and the blow no longer required, was 
tolerably well marked. 
On the other hand the specimens incapable of yielding the low rates of 
motion passed along with a jerky and irregular motion, causing much 
vibration in the apparatus, and great variation in the resulting coefficients. 
The point at which the blow ceased to be required was also very irregularly 
marked. 
* “Nouvelles Experiences sur le Frotfcement.” M. Morin. Paris 1833* 
