278 WITHDRAWN FROM THE ACTIOX OF GRAVITY. 



a millimetre, might seem sufficient of itself to establish iuflepenclence among these 

 films, but this thickness is here of no importance. I have caused a hexagonal 

 prismatic frame to be constructed in which the lateral edges were formed of mere 

 hairs, and yet everything has taken place in exactly the same manner. In this 

 frame, whose arrangement it would be difficult to represent by an engraving, the 

 upper base is raised by small springs, in such a way that the hairs, which are 

 elongated in the liquid, shall continue to be stretched. 



§ 25. But it may be readily conceived that in the case of a frame of which 

 the bases have less than six sides the incidents will be different, for then the 

 lateral faces making between them angles of less than 120°, the films which 

 proceed from two adjacent edges of the superior base, and which, along the cor- 

 responding vertical solid edge, would be in communication by the medium of the 

 liquid which moistens that edge, must tend to detach themselves from this same 

 edge, and be both directed towards the interior of the frame, to the end of re- 

 establishing between them the angle of 120° ; then, also, they will develop a 

 third film proceeding from the vertical solid edge in question, in such manner 

 that these films shall be united by an oblique liquid edge proceeding from the 

 point of junction of the three solid edges. This also experiment verifies: when 

 we withdraw from the liquid a prismatic frame with square base or the cubJc 

 frame, v/e see the films assume a re-entering direction as soon as the superior 

 base has emerged, and the effect is still more decided with the frame of the 

 equilateral triangular prism ; it will be observed, at the same time, that these 

 films conform to what 1 have advanced in the preceding paragraph — that is to 

 say, they become inflected in order to terminate at the surface of the liquid in a 

 vertical direction, and that they are concave in the direction of their breadth. 



As regards the regular pentagonal prism, for which the angle of two adjacent 

 lateral faces is lOS-", and consequently little inferior to 120'', it may be con- 

 ceived that the tendency of the films to re-enter must be feeble, and that hence 

 the thickness of the wires of which the solid edges of my usual frames are 

 formed, suffices in this case to establish independence between the films : thus, 

 with the pentagonal frame of Fig. 30, we obtain, while withdrawing it, only 

 plane films in the lateral faces ; but I have caused to be made a frame in Avhich 

 the lateral edges were of very fine iron wire, and then the films re-entered ; but, 

 as might be expected, they re-entered much less than in the two preceding 

 frames. This frame with fine lateral edges is represented by pj^_ 35 



Fig. 35. The superior base is sustained by two handles a 

 and I, by which the frame is to be held in immersing it. 

 § 2.6. Let us now see how these different systems are com- 

 pleted, when we continue to raise the frames. We will take, 

 first, the case of the hexagonal prism, and suppose that the 

 frame has the relative dimensions indicated at the end of § 

 20. When the inferior base rises from the liquid, a film will 

 be formed, as with the ring of § 23, extending from that base 

 to the surface of the liquid, a film which will continue to become thinner from 

 above downwards. If we still raise the frame, a point will soon be reached 

 at which the equilibrium of this film will be no longer possible, for it then 

 spontaneously and rapidly contracts, closes in separating from the liquid of 

 the vessel, and becomes a plane film in the inferior base of the prism. But 

 this plane film making right angles with those which occupy the lateral 

 faces cannot, according to what has been said in the preceding paragraph, 

 persist in this manner ; things must so shape themselves that it shall make with 

 the lateral films angles of 120°. Now, this is effected in the most simple man- 

 ner : the plane film in question ascends into the interior of the frame, diminishing 

 at the same time in extent, as if drawing to itself the lateral films, while other 

 films proceeding from each of the vertical wires attach themselves to these 

 lateral ones along the liquid edges Avhich unite them two by two, and equihbrium 



