S6: Rotation of Liquids. 



er, it is obvious that its momentum must also be greater, and conse- 

 quently that when an oscillation is completed, as all the bodies, viz. 

 the glass vessel, the oil, and the vi^ater, are necessarily compelled to 

 commence their descending motion at the same instant that, among 

 them which has acquired the greatest momentum, must tend to con- 

 tinue its previous course, while the others are tending in the reverse 

 direction. Hence the water rises up on the sides of the vessel, 

 whose motion it encounters, just as it would do if encountering any 

 other obstacle. The lighter liquid, by taking the place of the heavier 

 in that part of the vessel (at x,) which is opposite to the elevated 

 point (y), allows the line xy to be much more inclined to the horizon 

 than would be consistent with the condition of equilibrium if a single 

 fluid were used. When a complete revolution is made by the jar, 

 the positions of the two liquids in the higher part of the course must 

 obviously depend on the relation of the /orce of gravity to the centri- 

 fugal force occasioned by the revolution of the apparatus ; if the 

 former predominate, the liquids will change places in the vessel, 

 otherwise not. Here is no departure from known laws. 



The second modification of the experiment mentioned, is that in 

 which a jar or other vessel containing two (or more) liquids is caused 

 to revolve horizontally on its axis. In this case the well known laws 

 of equilibrium and motion require that the upper surface should be a 

 concave paraboloid dependant on the form of the vessel and the quan- 

 tity of liquid. The experimenter, and probably many others have 

 supposed that the surface of separation must also necessarily take 

 the same concave form, and have formed their theories accordingly. 

 He was therefore not a little surprised to find the latter varying with 

 the nature of the liquids and the rapidity of rotation. Thus he found 

 that employing a cylindrical jar, and placing oil at the bottom and 

 alcohol above it, the rotation gave a concave surface of separation ac- 

 cording to theory ; while oil upon water revolving under similar cir- 

 cumstances gave a surface of separation convex upward, which rose 

 even to the upper surface forming the center of the latter, of water, 

 and the periphery of oil. Placing equal quantities of spirits of tur- 

 pentine in the one case upon oil, and in the other upon alcohol, 

 (both these liquids having the same density,) the surface of separa- 

 tion was, with the same velocity of revolution, concave in the former 

 case, and convex in the latter. 



It is stated in the notice above referred to, that " the chemical af- 

 finity of the liquids is to be regarded as the cause of these apparent 



