Measuring Surface- Tension and Angles of Contact. 145 



where h is the height of the centre of the meniscus above the 

 level of the liquid in the vessel A. h was found by moving 

 the microscope horizontally and focussing on the surface of 

 the liquid. Thus T may be found without applying any 

 corrections. 



In order to get rid of one of the observations parallel 

 light was used instead of using the mark 0. A prism P 

 with angles 90°, 45°, 45° was fixed with Canada balsam to 

 the bottom of the vessel A, and parallel light from the 

 collimator C^ was directed up the centre of the tube. In 

 this case u = vd and B=(/a — l)i?, so that only the three 

 readings corresponding to v and h need be taken. Here 

 T=gpJiR/2 = igp(ji — l)hv. Thus since Jw= const, the best 

 conditions for working are those in which h = v, or a tube 

 should be chosen which gives a capillary rise in the neigh- 

 bourhood of \/2T/gp(p,-l). 



The following rough readings show the result for water 

 with tubes of different diameters. 



Radius of Tube. 

 r (in cm.). 



Radius of Curvature. 

 R (in cm.). 



Surface-Tension. 

 T. 



•386 



•507 



70-6 



•318 



•378 



71-3 



•212 



•230 



70-8 



•123 



•124 



70-7 



•09 



•092 



71-2 



The objection to the method as described above is that it is 

 always difficult and in some cases (e.g. turpentine) impossible 

 to be sure of focussing on the centre of the meniscus. To 

 do away with the necessity for focussing on the meniscus, 

 another image, this time by reflexion, was formed. Light 

 from a second collimator, 2 , fell upon a piece of glass, Gr, 

 with parallel sides (a microscope-slide cover was found to 

 work excellently) placed at an angle of 45° with the axis of 

 the tube, whence it was reflected down to the surface of the 

 meniscus, which formed an image on the axis of the tube at 

 a distance B/2 from the centre of the meniscus : all readings 

 through the microscope must be taken through the plate Or. 

 Thus, if d be the difference between the vernier readings 



