1865.] Prof. Guthrie on Bubbles. 25 



volume of A, moreover, prevents the level of its water from undergoing 

 more than a very slight variation during a single experiment. 



Certain modifications were introduced in the apparatus for special 

 purposes, which will be described in their proper places. 



Judging by analogy from the results obtained with drops, we should 

 conclude the bubble-size to be influenced mainly by 



1. Rate of sequence, or value of fft. 



2. Chemical nature of bubble-gas, if homogeneous. Proportion between 

 its constituents, if heterogeneous. 



3. Nature of solid from which the gas is delivered. 



4. Size of orifice and geometric distribution of solid about its orifice. 



5. Temperature of gas and medium. 



6. Tension of gas, influenced by natural or artificial causes. 



7. Chemical nature of liquid, if homogeneous ; and proportion between 

 its constituents, if heterogeneous. 



As in the cases of SLG and SLL, the solid serves mainly as a support 

 to the dropping liquid, and influences the size of the drop by the various 

 ways in which it affects the liquid film which adheres to the solid ; the 

 actual disruption being between liquid and liquid : so in SGL the bubble 

 parts in truth from gas. 



The separation of a gas-bubble differs materially from that of a drop 

 in this respect. In the case SLG it is the persistent cohesion of the 

 liquid which gives the drop a spheroidal form, and thereby assists gravita- 

 tion to overcome the stubborn cohesion of the liquid. In the case of 

 SLL the separation is assisted by the persistent cohesion of the liquid 

 medium, which also tends to mould the drop into a spherical form, and is 

 hindered by the stubborn cohesion and weight of the medium, which, by 

 resisting its descent, increases its weight. In the case of a bubble, the 

 ascent of the bubble is due wholly to the descent of the liquid medium ; 

 and the spheroidal form of the bubble is due wholly to the persistent cohe- 

 sion of the liquid medium ; for this cohesion is most completely satisfied 

 when the cavity containing the gas is most spherical. 



We may now examine seriatim the influences of the seven conditions 

 noticed above. 



SGL. 



Influence of rate. To examine the effect of variation in gt we may take 

 common air as the gas, distilled water as the liquid medium, the tube H 

 of glass having an opening h of any convenient unmeasured size ; and as 

 we are not concerned with the absolute, but with the relative sizes of the 

 bubbles, the vessel M may be of an indefinite size. 



Table a shows the effect of variation in gt alone. Column 1 shows the 

 values of fft. Column 2 the sequence of the experiments. Column 3 the 

 number of bubbles. Column 4 the mean relative size of a single bubble 

 at the respective rates of column 1 . 



