CAPILLARY CONSTANTS 441 



per centimetre. On one page of a recently-published authorita- 

 tive treatise, the well-known quantity " molecular surface 

 energy " is expressed in dynes per centimetre ! Two pages later 

 it is expressed in ergs ; and in recently-published research 

 papers, the differential coefficient with respect to temperature 

 of this same quantity is also given in ergs ! 



Moreover, in the discussion of the drop-weight experiment, 

 it is not uncommon- to find a statement such as the following: 

 " The drop falls when its weight just exceeds the surface-ten- 

 sion." This statement is probably a somewhat slipshod attempt 

 to expand into longhand the erroneous equation (iv), and as such 

 should read, " The drop falls when its weight just exceeds the 

 product of the surface-tension and the maximum horizontal 

 perimeter of the drop," but if English has any meaning it means 

 that the equation of equilibrium of the drop just before detach- 

 ment is given by mg = T, and it cannot be too strongly empha- 

 sised that a surface-tension is a force-per-unit-length, and is 

 no more a force than a velocity is a length. Such a statement, 

 therefore, involves a breach of the elementary theory of dimen- 

 sions which would, in most examinations, bring down a candi- 

 date's marks dangerously near to the limiting value zero. 



We have now reviewed rapidly some twenty methods for 

 the determination of surface-tensions. Whilst many of the 

 methods are specially suited to special cases, three methods 

 seem to stand out, in point of convenience and accuracy, as most 

 suitable for general use. These are Jaeger's method, the method 

 which depends on the estimation of the maximum pull upon 

 an anchor ring, and that which depends on the photographic 

 measurement of the partial depth of large bubbles. If one 

 could be quite sure that the drop-weight method would give 

 results which would be in accordance with those obtained by 

 these three methods, and not in general in accordance with those 

 obtained from capillary-rise, as its exponents claim, this also 

 would take its place as one of the most accurate methods yet 

 devised for the determination of capillary constants ; but, as 

 pointed out above, this question requires to be definitely settled 

 by experimental evidence. 



It is not necessary to discuss in any great detail the methods 

 used for the determination of the other important capillary 

 constant — the angle of contact — as comparatively little work has 

 been done thereon. Direct methods, such as that of Traube, 



