Intelligence and Miscellaneous Articles. 



379 



A LITTLE ILLUSION. BY J. PLATEAU. 



Permit me to describe an experiment which, at the first glance, 

 seems as if it must realize perpetual motion. Into a capsule of 

 sufficient dimensions let us pour distilled water till its depth is 

 about 1 centimetre. In this let us immerse one end of a glass tube 

 of small internal diameter, 1*5 millim. for instance, and from 8 to 

 9 centim. in length; let us keep this tube in a suitably inclined 

 position, so that the water which will be caused to rise in it by 

 capillary action shall occupy a sufficiently large portion of its 

 length. Let us procure a second tube, having, say, an internal 

 diameter of 3 millim., and a length of some centimetres. Let us 

 draw out one end of this second tube, and bend back the drawn-out 

 thin portion so that it makes a more or less acute angle with the 

 wide part. Let us introduce the thin portion into the upper orifice 

 of the inclined tube till it reaches the water contained in this, and 

 give to the wide portion a vertically descending position ; this last 

 condition will be obtained by modifying either the above-mentioned 

 angle or the inclination of the other tube. Let us suppose that the 

 orifice of the vertical portion does not reach the liquid of the cap- 

 sule ; lastly, let us imagine the thin portion and the vertical tube 

 both full of water. For greater clearness, I represent in section, 

 in the annexed figure, the upper part of the system. 



a b is the inclined tube, and c d the vertical. In consequence of 

 the narrowness of the space comprised between the slender portion 

 and the inner surface of the tube a b, capillarity will maintain the 

 water near the orifice a of the latter ; suppose that it stops at a. 

 The lower extremity of the tube c d not reaching the liquid of the 

 capsule, that tube with its slender portion will constitute a siphon 

 of which the short branch dips in a liquid in equilibrium, while the 

 long branch descends several centimetres lower than the surface of 

 that liquid. Does it not seem, then, that the water must inces- 

 santly run through this siphon, to rejoin that in the capsule '? Kow, 

 if that took place, it would evidently be perpetual motion. 



But, the impossibility of perpetual motion being demonstrated, 

 I asked myself what would in reality take place under the above 

 conditions. I therefore requested M. Van der Mensbrugghe to 



