of a Liquid Mass devoid of Weight. 289 



be raised or lowered by a gentle movement. Place the first ring 

 upon its stand on the table, sustain the second at a convenient 

 height over it, and well moisten both with the glyceric liquid ; 

 then inflate a bubble of about 10 centimetres diameter, place it 

 upon the lower ring and withdraw the pipe ; now lower the upper 

 ring until it comes in contact with the bubble, which immediately 

 attaches itself to it ; at last gradually raise this ring, and the 

 bubble, which thus drawn out loses more and more its spherical 

 curvature, is converted, by a certain separation of the rings, into 

 a perfectly regular cylinder, having convex bases like the full 

 cylinders of oil. 



A rather larger diameter can be given to the bubble ; but 

 when it is too large, the cylindrical form is no longer obtained, 

 either because the cylinder which it is desired to obtain exceeds 

 its limit of stability*, or because, if it be still within this limit, 

 it begins to approach it ; in this last case, in fact, the figure- 

 producing forces becoming very little intense, the small weight 

 of the film exerts an appreciable influence, and the figure appears 

 more or less swollen at the lower half, and compressed at the 

 upper half. The tallest regular cylinder which can be formed 

 with the rings, before pointed out, has a height of about 

 17 centimetres. Let us state in this place, that, for the complete 

 success of experiments of this kind, the rings should have under- 

 gone a little preparation : when they leave the hands of the 

 workman they should be slightly oxidized on their surface by 

 dipping them for two minutes into nitric acid diluted with four 

 times its volume of water ; afterwards wash them in pure water. 



In the memoir will be found the way to produce, in the 

 laminated condition as well, the other forms of equilibrium of 

 revolution, namely^ those to which I have givenf the names of 

 catenoids, onduloids, and nodoids. 



These experiments are very curious ; there is a peculiar charm 

 in the contemplation of these figures, so slender, almost reduced 

 to mathematical surfaces, which make their appearance tinted 

 with the most brilliant colours, and which, in spite of their ex- 

 treme frailness, endure for such a time. These same experiments 

 can be readily performed, and in the most convenient manner. 



I now pass on to another application of my new process. 

 Procure a collection of frames of iron wire, each one of which 

 exhibits all the edges of a polyhedron — for example, of a cube, a 

 regular octahedron, of prisms with triangular, pentagonal, and 

 other bases. Each of these frames is to be fixed like the upper 

 ring in the before-mentioned experiment, by a fork attached to 

 two of its edges ; they ought also to be oxidized by nitric acid. 



* See Second Series in Scientific Memoirs, Part XXI. 



t See the abstract of the Fourth Series in Phil. Mag. vol. xvi. p. 23. 



