G. F. Becker — Impact Friction and Faulting. 203 



of the weight depending from it and the curve shown by a 

 longitudinal section of the rod will be continuous with that 

 of the logarithmic column, instead of showing a salient point 

 at the junction as would be the case in a corresponding posi- 

 tive impact. Similar principles apply under the slightly more 

 complex conditions existing when a glass rod is softened in the 

 middle before a lamp and drawn out from the cold ends. 



Experiments on friction. — More or less exact experiments can 

 readily be made to test the statements as to friction here made. 

 If a flat bar (such as a standard steel foot rule) be laid upon a 

 frictional surface (blotting paper for instance), and a yielding 

 mass (say a rubber pencil eraser) be drawn along its upper sur- 

 face, it will be found that a slight pressure suffices to set the bar 

 in motion and that there is simultaneous relative motion be- 

 tween the rubber and the bar and between the bar and the 

 blotter. If a board about as thick as the bar is broad is cov- 

 ered with thick paper of the same kind as that on which the 

 bar rests, and the edge (thus prepared for the sake of abetter 

 contact) is pressed upon the bar and moved in the direction of 

 its longer dimension, relative motion on each side of the bar 

 will again take place. In these experiments if the pressure 

 applied is very light, no relative movement will occur below 

 the bar because a certain amount of energy is needful to strain 

 the minute projections of the paper beyond their elastic limit. 

 These trials show that motion is really produced by friction if 

 under that term the whole phenomenon is understood, or con- 

 comitantly with friction if the word is limited merely to the 

 dissipation of energy accompanying the action. Still further 

 to test the character of this transmission of energy, I arranged 

 a pile of seven steel bars one foot long 1J" wide and -£%" thick, 

 inserting between each a- slip of blotting paper of the same 

 width and length as the bars. These slips of paper were con- 

 sidered necessary to effect fairly complete contact which could 

 have been attained otherwise only by grinding the bars to- 

 gether with polishing powder. To adjust the strips of blot- 

 ting paper to the steel surfaces and to remove grosser inequal- 

 ities in the paper, each bar was rubbed a few times backward 

 and forward on the underlying slip. A long strip of wood 

 was then laid upon the strip of paper covering the topmost bar 

 and this was uniformly weighted over the bar with some ten 

 pounds of lead. The wooden strip was then drawn away 

 by a horizontal traction for three or four inches. The energy 

 was distributed throughout the system, relative motion taking 

 place from the first instant on 13 out of the entire 15 contacts 

 and the ends of the bars were manifestly arranged on a curve 

 asymptotic to their original plane and almost certainly recog- 

 nizable as the logarithmic curve to an eye familiar with its form. 



