and Attached Water \ 



273 



With this I made decimal mixtures ranging from 95 alcohol and 

 5 of water to 95 water and 5 of alcohol. These were submitted 

 in turn to the action of a cryogen, in order first of all to see at 

 what temperature the solidification of each mixture begins. For 

 if, as has been supposed, and quite recently by Melsens again 

 proved, a very weak (10 per cent.) spirit gives up only ice, 

 and since alcohol is notoriously not to be solidified by our most 

 powerful cryogens, it must follow that during the continuous soli- 

 dification of a weak spirit the temperature must fall continually. 

 Temperatures down to —19° were observed on the mercurial 

 thermometer. For lower temperatures, demanding the employ- 

 ment of solid carbonic acid and ether, use was made of an alcohol 

 thermometer, which was collated with the mercurial one at —22°, 

 The Table gives the temperature at which the solution began to 

 yield solid matter. What this solid matter consists of I have 

 afterwards to consider. Columns A x and A 2 are the values of 

 the first and second differences respectively. 



Table XI. — Temperatures at which Solidification begins in 

 Spirits of various strengths. 



Water per 



Alcohol per 



Temperature C. 







cent., by 



cent., by 



at which solidi- 



Ai- 



A 2 . 



weight. 



weight. 



fication begins. 







95 



5 



o 



- 2 



2-3 



+0-6 



90 



10 



- 4 3 



29 



+06 



85 



15 



- 72 



3-5 



+0-5 



80 



20 



-107 



4 



+0-7 



75 



25 



-147 



4-7 



-0-8 



70 



30 



-19-4 



3-9 



-0-2 



65 



35 



-23 3 



3-7 



+0-3 



60 



40 



-27 



40 



+20 



55 



45 



-31 



60 



-10 



50 



50 



-37 



5 



-20 



45 



55 



-42 



30 



+5-0 



40 



60 



-45 



80 



? 



35 



65 



-53 



? 





30 



70 



- 65 (not). 







From this Table it is seen that the temperature of initial soli- 

 dification sinks so regularly that the column of second differ- 

 ences shows for a long time the value 0*6. Only at lower tem- 

 peratures, which of course cannot pretend to the same degree of 

 accuracy, are serious variations visible. There is a rapid fall at 

 the ratio 35 water to 65 alcohol ; and at the ratio 30 water to 70 

 alcohol, I failed to effect solidification at —65° C. 



§ 97. So free from discontinuity are these numbers, that one 

 might be readily misled into the belief that the solid matter 



