M. Mendelejeff on the Compounds of Alcohol with Water. 139 



Two other sources of error increase the specific gravity of 

 alcohol, — first, the extraordinary tendency to absorb moisture 

 from the air (an alcohol of 0-78973 had, after standing half an 

 hour in an open flask, a specific gravity of 0*78992) ; and 

 secondly, its property of absorbing air to a considerable extent. 

 Alcohol of 0-78973 had, when shaken for five minutes in a closed 

 flask, air being excluded, a specific gravity of 0-79007. It is 

 known that alcohol saturated with air gives out air-bubbles 

 when diluted with water. Pouillet and Baumhauer determined 

 the specific gravity by hydrostatical processes, and their obser- 

 vations are thus charged with great errors. 



The author attempted, but in vain, to keep absolute alcohol 

 for a long time without alteration of weight : the best way was 

 to keep it over lime under a bell-jar. For each observation a 

 fresh portion of alcohol was rectified as rapidly as possible, access 

 of air being prevented. This was effected by means of a specially 

 constructed apparatus ; and the alcohol was in all cases tested 

 by the addition of water ; if globules of air were formed, the 

 specimen was rejected. In the distillation, the temperature of 

 the water-bath should not exceed 83° to 85° ; if it do, the dis- 

 tillate will certainly contain water, in spite of all the lime. 



The author has tested all the modes of dehydrating alcohol, 

 and communicates his observations. 



Calcined potash is well adapted for the dehydration of weak 

 alcohol; but it does not remove the last traces, and at most 

 an alcohol of 99° per cent, is obtained. Chloride of calcium 

 and anhydrous sulphate of copper act in the same manner. 



Caustic baryta, as proposed by Berthelot and Pean de St. 

 Gilles, is admirably adapted for preparing absolute alcohol. 

 When the supernatant liquid appears yellow, the alcohol has lost 

 the last trace of water. For dehydrating a litre of alcohol, 

 about 200 grms. baryta are needed. The author found it advan- 

 tageous not to pour the dehydrated alcohol off, but, after stand- 

 ing for about a day, to distil it directly. When the quantity 

 of baryta and alcohol was left to itself for a week, the liquid was 

 orange-red in colour, and on distillation furnished an alcohol 

 of higher specific gravity. 



Dehydration by sodium the author totally rejected. If sodium- 

 amalgam was used in order to moderate the action, some mer- 

 cury passed over; and in all cases the distillate contained sodium : 

 hence the determinations were too high. 



The most practical method is that of caustic lime. The 

 pieces must stand out of the alcohol, which must have at least a 

 specific gravity of 0*792 at 20°. In two days all water is re- 

 moved. But if it is desired to distil after the lapse of two or 

 three hours, a previous heating to 50° or 60° is necessary. The 



