CAPILLARY CONSTANTS 433 



for T. Many of them, of course, can be used with advantage 

 as research methods when the specific object of the research is 

 the determination of contact-angles, but that is not at present 

 the point under consideration. Those who desire further 

 knowledge of the practical details of these methods will find 

 them under the references in the bibliography appended to this 

 article. 



We now proceed briefly to discuss the various methods 

 which do not require a knowledge of the contact-angle. 



Concerning the three methods grouped under the heading 

 " Vibrations," it may be said of all that they are experimentally 

 of such complexity that more rapid methods which possess the 

 same order of accuracy are to be preferred. Method (i) makes 

 use of the fact that the period of oscillation about its equilibrium 

 figure of a small falling drop is a function of its surface-tension, 

 and this period may be determined by taking a rapid series of 

 instantaneous photographs of a falling drop. Method (2) de- 

 pends on the measurement of the wave-length of the standing 

 waves formed on the surface of a jet of liquid issuing from a 

 small elliptical orifice, and is, both in theory and in practice, 

 of such complexity, as to be prohibitive as a practical method. 



Method (3) depends on the measurement of the wave-length 

 of the ripples generated on the surface of a liquid by a dipper 

 vibrating with known frequency. Whilst it is more simple in 

 practice than the other methods, an accurate determination of T 

 demands considerable care in the measurement of the wave- 

 length, for it unfortunately happens that the most important 

 term in the equation which gives T contains the cube of the 

 wave-length, and therefore a i-per-cent. error in the determina- 

 tion of the wave-length burdens the corresponding value of 

 T with a 3-per-cent. error. 



Moreover, in each of the three above methods temperature 

 measurements are somewhat difficult. And it is to be remem- 

 bered that, in discussing questions of molecular complexity, 

 and of the relation between surface-tension and chemical 

 composition and constitution, the determination of the exact 

 value of the temperature-coefficient of surface-tension is quite 

 as important as that of the surface-tension itself. It is therefore 

 absolutely essential that the apparatus used shall be capable of 

 being placed in some form of thermostat which shall maintain 

 the liquid at definite and accurately-measurable temperatures. 

 29 



