144 Scientific Intelligence. 



6. On the Separation of Mixed Liquids. — Duclaux has 

 made a careful study of the conditions under which a homo- 

 geneous mixture of two liquids will separate into Iavo entirely 

 distinct layers, and has arrived at some very curious results. 

 He finds, for instance, that a mixture of 15 cubic centimeters 

 of amyl alcohol, 20 cubic centimeters of ordinary alcohol and 

 32-9 cubic centimeters of water, gives at the temperature of 20° C, 

 a molecularly unstable grouping, so that the least diminution of 

 temperature causes it to separate into two nearly equal layers : 

 He states that under these conditions the composition of the two 

 layers is invariably the same whatever the composition of the 

 initial liquid, the layers varying only in amount. The same fact is 

 also true of three as of two liquids ; though in this case the third 

 liquid takes no part in the separation, and remains the same in 

 each of the two layers as in the original liquid. Hence it is al- 

 ways possible to start with a given liquid such that by depres- 

 sion of the temperature, two layers of the same volume are pro- 

 duced. The range of variation of temperature necessary to effect 

 this separation is extremely minute, being much less than a tenth of 

 one degree Centigrade ! Moreover, the introduction of mere 

 traces of certain substances, as sodium and calcium chlorides and 

 other soluble salts, and the vapor of chloroform produce the same 

 effect as a lowering of temperature. So also a drop of water or 

 one of amyl alcohol will cause the separation. The author has 

 applied this phenomenon to the construction of an ingenious mini- 

 mum thermometer. By varying the amount of water present m 

 the above mixture for example, the temperature at which separa- 

 tion ensues may be varied. The solutions may be readily pre- 

 pared by taking the necessary quantities of aniyl and ethyl alco- 

 hol, maintaining them at the exact temperature required and 

 adding water drop by drop, until a slight turbidity appears, 

 which should dissolve upon the slightest heating. The mixture is 

 then placed in a tube and this is hermetically sealed. Ordinarily 

 the liquid is clear but it becomes turbid as soon as the temperature 

 falls below that at wbich it was prepared. A few drops of car- 

 mine in ammonia makes the separation more distinct, since the 

 lower layer only is colored. If ten parts of ether be mixed with 

 six of commercial methyl alcohol, and water be carefully added as 

 above, a liquid will be obtained acting as a maximum thermom- 

 eter, since it becomes turbid and separates when the temperature 

 rises above that at which it was made. This is colored with a 

 little blue ink. Several tubes of each kind would evidently be 

 exceedingly useful in maintaining a given temperature constant 

 for any purpose, since thev could be graduated to any interval— 

 C. JR., Ixxxi, 815, Nov., 1*875. g. f. b. 



1. Stationary Liguid Waves. Professor Guthrie has recently 

 communicated to the London Physical Society the results of his 

 observations on wave motion. If water in a cylindrical vessel, not 

 less than nine inches in diameter, be agitated by depressing and 

 elevating a flat circular-disk on its surface at the center, a form of 



