378 Mr. T. C. Fitzpatrick on the Action of 



recently, of two bottles full of alcohol, which had been dis- 

 tilled at the same time and then left to stand for a week, the 

 conductivity of the one at the end of this time was ten times 

 as great as that of the other. Tn all my measurements the 

 same glass cell has been employed ; and it is now found that 

 the resistance of the water or alcohol when left in the cell does 

 not fall nearly so quickly as it did in the case of the first 

 measurements, and it can be distinctly seen that the glass has 

 been acted on. 



The conductivities of these three solvents differ considerably. 



The average resistance-value for 250 cubic centim. of 



(1) Water introduced into the cell, 



R= 15,000 to 16,000 legal ohms. 



(2) Ethyl-alcohol, 



R= 50,000 to 60,000 1. ohms. 



(3) Methyl-alcohol, 



R = 4000 to 4500 1. ohms. 



It will be seen that the ethyl-alcohol is much the worst 

 conductor, whilst the water is intermediate between the two 

 alcohols. 



In the last* circular of the Electrolysis Committee Prof. 

 Ramsay makes mention of the work of Adolf Bartoii, that 

 benzene is a non-conductor, and methyl-alcohol conducts well, 

 and adds as a note, "is it pure?" 



I was surprised at finding the resistance-value for this 

 alcohol so low, and considered that it could not be pure ; but 

 after carefully distilling it from lime I obtained the same, or 

 almost the same, value ; and after again repeating the process 

 the value was not found materially different. 



It did not seem at first worth while working with a solvent 

 of such a low resistance, but I have lately employed it ; and 

 the results are. I think, interesting, the only objection being 

 the rapidity of its evaporation. 



Salts employed. 



It is a point worth noticing, that all the salts that are soluble 

 in ethyl- alcohol are deliquescent salts with the single exception 

 of mercuric chloride. This latter salt I experimented with, 

 as it is also the only salt more soluble in alcohol than in 

 water ; and I thought it might be possible, by comparing 

 the conductivities of solutions of this salt in equivalent quan- 

 tities in water and alcohol, to see if the difference of con- 

 ductivities of equivalent solutions depended at all on the 

 coefficient of solubility of the salt in the particular solvent. 



It was then found, as had been previously observed by 

 * Sixth circular, May 1887, p. 8. 



