14 



Mr. C. Chree on the 



Time. 



1*1..— " 



log 10 -^. 







x n+l 



X n+Q 



*n+6~ x n+12 



h , u 









3 2 



•02551 



016611 



017507 



10 



•02838 



•16567 



•17068 



20 



•02958 



•16579 



•17784 



30 



•02563 



•16354 



•17327 



40 



•02969 



•16248 



•16745 



50 



•02732 



•16208 



•17072 



3 



•02507 



•16199 



•17532 



10 



•02850 



•16337 



•17933 



20 



•02733 



•16031 



17173 



30 



■02457 



•15916 



17422 



40 



•02929 



16024 



•17216 



50 



•02723 



•15822 



•16399 



4 



•02640 



•15609 



•15589 



10 

 20 



•02544 

 •02618 



15628 

 •15529 





Mean... 017136 



30 



•02565 



•15265 





40 



•02727 



•15502 





50 



•02510 



•15571 





5 



•02664 



•15618 





10 



20 



•02445 

 •02354 







Mean... 015980 





30 



•02802 







40 



•02796 







50 



•02557 







6 









Mean... '02668 







gradual decrease in log 



X n +6 



This would indicate that /a 2 , 



and so also k, which varies as /u, 2 , diminishes as the tem- 

 perature falls. Thus, strictly, k should have been treated 

 as depending on the temperature ; however, the variation is 

 not very rapid, and so Weber supposed it legitimate to take 

 the mean value supplied by the third column in calculating 

 the conductivity of water at the mean temperature of the 

 experiment. This gives // 2 = *3680, and thence £ = '0768, cor- 

 responding to a temperature of about 4°'l, according to 

 Weber's formulae. 



Lorberg considers the fourth column to give a correct 

 value for fi 2 , and explains the difference between it and the 

 third column by a theory presently to be considered. From 

 this one experiment he finds, by the method of least squares, 

 jju 2 =z -39556, with a mean error of '00067, and thence deduces 

 £ = •08317 — -0046 h. If, as he supposes, h be certainly less 

 than *006, the last term is of no practical importance. 



From the mean of a considerable number of experiments, 

 Weber finds £='0745 at 4°-l and ='0857 at 23°'67 ; thence 



