222 THE FIGURES OF EQUILIBRIUM OF A LIQUID MASS 



It is first necessary to be able to give to the disc a considerable velocity of 

 rotation. To do this, we adapt to the upper part of the vessel a system of two 

 pulleys — one small, and fixed on to the prolongation of the axis of the disc at 

 the place of the handle, which is taken aAvay ; the other larger, and to the axia 

 of which the same handle is attached. In my apparatus the diameters of the 

 two pulleys are, respectively, 12 and 75 millimetres. In the second place, the 

 dianu-ter of the sphere being always nearly six centimetres, that of the disc 

 should be only two centimetres. Lastly, the disc should not have, as in the 

 preceeding experiments, its centre coinciding with that of the sphere. It should 

 be placed lower, toward the inferior part of the latter. 



Matters being thus arranged, the handle is turned with a velocity which ex- 

 perience soon enables us to determine. In my apparatus this velocity ought to be 

 about two turns and a half per second, which nearly corresponds to fifteen turns 

 of the disc in the same time. We then see, in general, a ring rapidly formed, 

 which extends itself, leaving in its centre a mass of oil, to Avhich it remains 

 ixnited by a thin pellicle. At the instant when the ring has attained a sufficient 

 development, (and by habit alone can this be correctly learned,) the rotation is 

 suddenly stopped. The pellicle then breaks, the ring remains completely iso- 

 lated, and the central mass forms inio a sphere. We have thus, during some 

 instants, a curious representation of the system of Saturn, except the flattening 

 of the ring. The ring returns rapidly, afterwards, upon itself, and is again 

 xmitcd to the central sphere. This experiment does not offer any great diffi- 

 culties. It requires, however, some skill to succeed perfectly.* 



22. In describing(§ 10) the experiment in which the flattening of the sphere is 

 effected by the immediate action of the disc, I have remarked that the movement 

 of the latter should not be continued too long, because the mass of oil then 

 comes to lose its form. Now, if Ave continue, nevertheless, to turn the handle, 

 with a view to observe the results of this disfigurement, we see manifested new 

 and very capricious effects. 



The sphere being Avell centred with relation to the disc, if we give velocities 

 of one turn in six, five or four seconds to the latter, Ave begin, after seven or 

 eight turns, to see the mass of oil elongate itself horizontally in one direction, 

 taking a form A\diicli*resembles much an ellipsoid of three axes ; and, Avhat is 

 more singular, this kind of ellipsoid is placed in an eccentric manner Avith rela- 

 tion to the axis of rotation. Figure 4 represents, for a A^elocity of a turn in 



Fig.- 4. 



four seconds, the mass viewed from three different sides ; that is to say, from 

 above and in the two lateral directions of the smallest and of the largest hori- 

 zontal axis : the dotted parts indicate the positions of the disc and of the axis of 

 rotation. The aspect of the mass seen from above shoAvs that it is slightly bent in one 

 direction; but this effect is evidentJy OAving to the resistance of the ambient liquid. 

 When once the mass has taken this form, it preserves it indefinitely as long 

 as the movement of the disc continues ; it continues to rcA^olve eccentrically 



* On comnmincatnig this very experiment to the academy, in the sitting of April, 1842, 

 (see thu Biilictius,) I stated tlint it Avas uecessary to vary tlic velocity of rotation. I have 

 since touud that, having adopted a convenient velocity, it was best to keep it uniform. 



