On some curious Effects of the Molecular Forces of Liquids, 435 



ground; for otherwise either the liquid sheet will not have 

 time to resolve itself into hollow spheres, or, should these 

 spheres be produced, it will be found impossible to observe 

 them and to notice their rupture. 



This experiment fully confirms the conclusion arrived at by 

 M. Felix Plateau, according to whom the simultaneous agglo- 

 meration of liquid molecules into perfectly closed envelopes 

 cannot be regarded as a necessary condition of the formation of 

 vesicles ; the admission of the generation of unclosed films of any 

 curvature whatever will suffice — a generation which assuredly 

 involves no impossibility. Without doubt the question of the 

 duration of the bubbles remains to be resolved. The larger 

 ones, it is true, burst in less than a second; but do the smaller 

 spherules likewise do so ? I have not been able to establish this 

 point, in consequence of the difficulty of judging whether these 

 small spheres are full or empty — a difficulty which is increased 

 by the circumstance that the envelope is much thicker than that 

 of the bubbles of soap-water. 



Apart from the meteorological question, the application of 

 the procedure above described to different kinds of liquid ap- 

 peared to me to possess some interest. I first tried soap-water, 

 and thus found that although bubbles form themselves very 

 well, their diameters are not great; and moreover they burst 

 as rapidly as when pure water is employed. If these results 

 differ from those described by M. Felix Plateau, it is merely 

 because, instead of turning rapidly round on projecting the 

 liquid, I simply threw it with a relatively moderate velocity, so 

 that the sheet was much less broad, and much thicker. I tried 

 also a very great velocity, so as to render the sheet very thin, 

 and I then obtained a very great number of small spheres 

 accompanied by some very light bubbles more or less large and 

 pretty durable ; I even succeeded, by rapidly projecting a con- 

 centrated solution of soap {qui s y etait prise en gelee), in pro- 

 ducing three bubbles, the largest of which had a diameter of 

 at least 25 centimetres, whilst that of the two others amounted 

 to about 8 or 10; all three had a duration of about half a 

 minute. 



My method succeeds very well with alcohol, though the bub- 

 bles burst very quickly. Amongst the volatile oils, I operated 

 with success upon oil of turpentine, and especially so upon petro- 

 leum oil. With heavy oils (of which I only tried olive oil) the 

 experiment is a little less successful : the bubbles which are 

 formed are very small, in consequence of the liquid being pre- 

 vented by its own viscosity from expanding into a broad sheet. 

 Lastly, I obtained good results with several saline solutions. 



I have not tried mercury. It appeared to me, in fact, scarcely 



