516 Mr, A. Ferguson on the Theoretical 



We now proceed to give some experimental illustrations 

 of the above formulae, choosing water as the fluid for exami- 

 nation. It should be premised that, in spite of all the 

 experimental work that has been done on the surface-tension 

 of water, it cannot be claimed that the value of this constant 

 is known even with an accuracy of ono per cent. Restricting 

 ourselves to the determinations made during the last twenty 

 years, we find values given for the surface-tension at 12° C. 

 ranging between the limits 71*7 and 77 dynes per cm., with 

 a grouping of the majority of the results round 73 to 

 75 dynes per cm. 



It is not necessary to indulge in any lengthy criticism of 

 the methods used. It may be pointed out in passing, that 

 methods which involve any assumptions as to the magnitude 

 of contact-angles are obviously open to error, and that 

 measurements of capillary ripples are on several counts 

 open to criticism *. On the whole, measurements of bubbles 

 and drops seem to be least open to criticism. 



The majority of experimenters along this line have made 

 actual measurements of the bubble or drop itself. To the 

 writer it seemed preferable to photograph the bubble or 

 drop, afterwards making measurements on the photograph. 

 Such a method seems preferable for several reasons. The 

 surface is exposed only for a minimum time, the magnifi- 

 cation is increased, giving a corresponding increase iu the 

 accuracy of the reading, and a permanent record of the drop 

 or bubble is made, which can be checked at any time. 



In the apparatus used by the writer, the drop was blown 

 under a funnel placed vertically in a rectangular glass 

 trough with parallel sides. The size of the bubble could be 

 very exactly regulated by means of a tube and pressure-bulb 

 attached to the funnel. The bubble was illuminated by 

 means of a beam of light from a filament lamp and con- 

 densing lens, which beam impinged horizontally on the 

 surface oE the bubble, this latter condition being insured by 

 means of several narrow rectangular apertures placed hori- 

 zontally through which the beam passed. 



The plate (PI. VIII. fig. 3) shows some typical photographs. 

 A shows a bubble 35 mm. in diameter, which is rather too 

 small for the application of equations (ii./>) or (iii.)- B and 

 G show a larger bubble, over 50 mm. in diameter, at different 

 stages of extrusion. D and E show pendent drops of much 

 smaller size (4 mm. in maximum diameter) formed under 

 different pressures. To quote a few measurements, we have 

 the following results from bubble B (the measurements 

 * See, in particular, Bohr, Phil. Trans. A. 209 (1909). 



