On Static Friction. 33 



impurities. By following up this clue it was found that 

 none of the three substances mentioned above had when 

 chemically pure any power of lessening the pull needed to 

 cause the one glass face to slip over the other. Chemical 

 substances in fact fall into two classes according to whether 

 they are active or inactive in this respect under the con- 

 ditions of the experiments. Water is an inactive substance,, 

 and this is the basis of the tea-cup experiment. 



In our observations watch-glasses weighted with lead run 

 into them whilst hot were used. .A small arm projecting 

 from the lead was attached to a silk thread which passed 

 over a pulley to a pan which held the weights. We were at 

 some pains to procure a glass plate for the watch-glasses to 

 slide upon having an " optical" face. But it was found that 

 ordinary plate glass gave the same value as the plate with an 

 accurately plane surface. Measurements were carried out 

 on a levelling table. 



Plates and glasses were cleaned by washing with soap- 

 powder and then rubbing under a vigorous flow of tap-water 

 with the, finger-tips, previously washed, until the peculiar 

 clinging stage was reached which sets a finger-bowl vibrating 

 when the wet finger-tip is rubbed round the edge. Both 

 plates and glasses were drained and dried in air. Con- 

 tamination will creep over the clean surfaces from solids 

 touching them, therefore contact during the drying process 

 must be reduced to a minimum *. 



A solid surface dried in air will retain a layer of water in 

 equilibrium with the aqueous vapour, and some of the fluids 

 used, such as acetic and sulphuric acids, absorb water. For 

 these reasons it was necessary to carry out the measurements 

 in air free from aqueous vapour. This was done by using a 

 chamber through which a rapid stream of dry and dust-free 

 air could be passed. The results recorded below will not be 

 obtained unless this precaution be taken. In such a chamber 

 a cleaned surface retains its purity for hours, the substances 

 used to dry the air seeming completely to remove the 

 lubricating matter which condenses to form the " cn-ease " 

 film on surfaces exposed in the ordinary way. 



According to the accepted view a "clean" surface as 

 defined above would differ from the "raw" surface produced 

 by splitting a solid in a vacuum by the presence of a layer 



* In using a Pockell's trough to measure the surface tension of water 

 under various conditions, in order to obtain really steady readings, one of 

 us found it necessary to reduce contact between the trough and the table 

 by interposing a couple of glass rods. 



* Phil. Mag. S. 6. Vol. 38. No. 223. July 1919. I) 



