194 Prof. J. A. Pollock on the 



sulphuric acid have a surface tension which increases as the 

 concentration becomes greater; they may be said, then, to lack 

 the frothing property even more than water. The acetic 

 acid solution has a surface tension which decreases as the 

 concentration increases, so the photographs of Plate VIII. 

 may be taken as exhibiting the bubbles in three typical 

 liquids, and, by showing the similarity of the detail in the 

 three cases, proving a common method of production. 



Here the development, previously suggested as that of the 

 small bubbles of froth, is clearly apparent ; cylindrical pro- 

 tuberances are formed on the larger surfaces which lengthen 

 until the state of instability is reached, they then become 

 detached as separate bubbles. 



In seeking for an explanation to account for these pro- 

 trusions into the liquid of the surfaces of the gas bubbles, 

 any consideration of a local weakening of the surface tension 

 seems here out of the question. Firstly, a cause is not 

 apparent for an accidental diminution of the surface tension 

 in somewhat large patches, confined within narrow limits as 

 to size ; secondly, the effect is shown both with tap and 

 distilled water, used with every precaution to avoid con- 

 tamination, and no difference occurs when the gas is carefully 

 filtered : moreover, bubbles of the class under consideration, 

 produced in water and in non-frothing solutions, show no 

 special durability. That the phenomenon is not directly 

 dependent on any directed momentum of the gas is proved 

 by the fact that in the photographs the projections are shown 

 on the surface of closed bubbles, in which the gas can have 

 little, if any, general momentum relative to any part of the 

 surface. 



The outline in the photographs is fairly sharp as the ex- 

 posure was of the order of a millionth of a second, but in each 

 case the liquid was in a state of considerable turbulence. 

 With this fact in view, it is suggested that the bubbles, 

 whose development is shown, owe their origin to variations 

 in the external pressure on the parent surfaces due to the 

 motion of the liquid relative to these parts of the larger 

 bubbles. In this case the size and number of the bubbles 

 will depend on circumstances which determine the nature of 

 the liquid movement. 



This suggestion as to the origin of these small bubbles is 

 prompted by the result of the process of exclusion previously 

 outlined, and though impossible to prove, as the turbulence 

 of the liquid cannot be directly seen, is not improbable from 

 the appearance of the detail shown. 



