1869.] in liberating Vapour from Boiling Liquids. 



247 



rather the top of the column is a little more heated than the bottom (since the 

 upper layers of hot water are at a higher temperature than the lower ones), 

 and the effect of this is that there are no convective currents ; cohesion is 

 diminished by expansion, not by convection. The whole column being 

 thus about equally heated at the same moment, vapour cannot form at one 

 part in preference to another, except at the surface ; but the whole column 

 of liquid goes on expanding under an increasing temperature until, becom- 

 ing more and more supersaturated with its own vapour, the increasing 

 elastic force suddenly overcoming pressure, cohesion, and adhesion, there 

 is a sudden burst of vapour. Or before this disruption takes place, if the 

 surface be touched with a chemically unclean solid, the vapour adhering 

 to it and thus set free, starts the vapour-giving action, just as touching a 

 cold supersaturated saline solution starts crystallization, and the action 

 once begun is propagated. 



If, however, the tube containing the ether &c. be not chemically clean, 

 if there be minute specks and points in the glass (as there often are) all but 

 invisible to the naked eye, and these be porous or not chemically clean, 

 vapour will stream from them long before the temperature of disruption is 

 attained, and there will be no disruption at all. These points and specks 

 account for many anomalous cases of crystallization which occur in operat- 

 ing with supersaturated saline solutions, and which puzzled Lowel and 

 other observers. We may have, for example, two tubes apparently pre- 

 cisely alike, cleaned in the same manner, containing a hot filtered solution 

 of the same salt, of the same strength, and exposed to the same cooling 

 influence. One of the solutions in cooling will suddenly become solid, 

 while the other will remain liquid, and continue so during weeks and 

 months. On examining the solidified solution, it will be found that crys- 

 tallization has been promoted by a minute speck or point at some part of 

 the tube, no matter where, and from this point, as from a centre, proceed 

 fine crystalline needles radiating in all directions. 



Soubresauts, — Liquids which render the surface of the vessel in which 

 they are boiled or distilled chemically clean, thereby favour the production 

 of soubresauts, or jumping ebullition. This is a mechanical action which 

 does not seem to have been sufficiently explained. Thus Gay-Lusaac 

 says, " When the liquid is above the boiling-point, it is in a forced state : 

 instantly a burst of vapour is formed, the liquor is thrown out, and the 

 vessel itself raised." 



But why should the vessel be raised ? The burst of vapour follows an 

 upward motion along the line of least resistance, which, so far from raising 

 the vessel, has a precisely contrary effect. It produces an equal reaction 

 in a downward direction, tending to force the vessel further into the ring 

 of the retort-stand, or other support, and it is the rebound from this that 

 causes the vessel to rise. If proof be required of the truth of this explana- 

 tion, it can easily be supplied by suspending, by means of an india-rubber 

 line or a bit of elastic, a tube containing crystals of sodic sulphate and a 



