16G GEOLOGY OF THE COMSTOCK LODE. 



Ill tlie experiments I have made the weight employed was about 5,000 

 times as great as that of a single slip. If a blunt edge, such as that of a ruler, 

 be now applied at right angles to the longer dimension of the slips, close to the 

 weight, with a light pressure, and be drawn away from the weight a fraction 

 of an inch, a slight relative movement will be perceptible. If this applica- 

 tion of energy to the system be repeated a score of times, the ends of the 

 pile of slips will be found to form a curved surface instead of a plane ^ 

 If the frictional resistance is proportional to the pressure, this curve must 

 sensibly coincide with that given by the equation 



y— 



\+xe 



for f — ., , Hnd will altoo:ether escaije detection. The thinness of the 



paper considerably obscures the character of the curve, but there is no 

 error in principle involved in plotting it on the assum^Jtion that the sheets 

 are of any thickness which may seem best adapted to bring out its geo- 

 metrical relations. For the given increment the curve will approximate 

 pretty nearly to the simple logarithmic curve. For the one hundredth-con- 

 tact the latter would give 



and the equation for inci'easing pressure 



y— 



1+0.02' 



yx=1.02?/. 



Unless the experiment is carried on until the lowest movable sheet has 

 traversed a sensible distance, the original position of the edges of the sheets 

 marked by the fixed slip gives the asymj^tote of the original curve. Fig. 4 

 on the next page shows a curve AB plotted from experiment witji its asymp- 

 tote, and a logarithmic curve CD of the form ?/zi:^-l«i~-^ plotted from its equa- 



'I noticed long since that pressmen in printing offices, by drawing the thnnib-nail across a pile 

 of sheets, force each of the upper sheets to project beyond the one next beneath it, so that one sheet at 

 a time can be removed conveniently and without delay. I observed that a regular curve resulted, but 

 presumed that it was a conic section. Having satisfied myself analytically that the curve produced 

 by faulting was logarithuiio, this observation recurred to me as a means of testing my results experi- 

 nientallv. 



