553 
of Edinburgh, Session 1874-75. 
whose occurrence in our paper I am aware. We said that in his 
researches on the conductivity of solutions of different density, he 
was not careful to keep to exactly the same temperature throughout 
a whole series of observations, so that his results did not admit of 
accurate graphic representation. This is a mistake, proceeding 
probably from a misunderstanding of one of his tables, and I am 
happy to have this opportunity of making the correction. In 
speaking, however, of our having made observations at constant 
temperatures, Professor Beetz’ statements are somewhat too sweep- 
ing. He proves it himself; for the confidence which he places in 
Paalzow’s* numbers shew that this is not the cause of his want of 
confidence in ours. Our experience is contrary to his opinion. We 
found the method of constant temperatures somewhat slow, but met 
with no great difficulty in making our measurements always within 
a small fraction of 10° C. Had it not been so, it would have been 
easy to adopt the method of interpolation. 
The second remark was merely a statement of fact, and not 
intended, as Professor Peetz thinks, as a reproach. Which of the 
two kinds of formula is the better, is, of course, simply a matter of 
opinion. We wished merely to state that Professor Beetz, having 
adopted the usual one, had not availed himself of what we regarded 
as one great advantage of the other. He thinks, however, that we 
not only made the worse choice, but did not produce good speci- 
mens of the kind which we had chosen. “ It would be better,” 
he says, u if there was more agreement between their observed and 
calculated numbers.” As a comment upon this remark, I give 
the tables of conductivity (observed and calculated) of the solutions 
which both he and we have examined. His results we take from 
the table given in his paper, f the latter part of which has been 
omitted, because our formula (a manifest disadvantage) does not 
apply to solutions beyond that of maximum conductivity. Our 
numbers are those given in our agreement-table, $ reduced to the 
same form as his, the first four being omitted because they corres- 
pond to solutions weaker than any that Professor Beetz examined. 
All the given numbers must be multiplied by 10 ~ 9 . 
* Monatsberichte der Berliner Akademie, 30. Juli, 1868, p. 488. 
t Pogg. Ann. cxvii. 1862, p. 20. 
X Trans. Roy. Soc. Edin. xxvii. part i. 1872-73, p. 64. 
