426 Sir John Conroy on a new Photometer. 



The table gives the results of eight observations made with 

 both photometers, the differences of each observation from the 

 mean, and also the squares of these differences. 



Bunsen's Disk. 



Differences from Squares of the 



centiin. the mean. differences. 



85-7 + *6 -36 



84-6 - -5 *25 



84-9 - -2 -04 



85*4 + -3 -09 



86-2 +1-1 1-21 



84-8 - -3 -09 



84-3 - -8 -64 



85-2 + 4 -01 



Mean . . 85-1 Sum . . 2*69 



New Photometer. 



85-0 --4 -16 



85-7 +-3 -09 



85-0 --4 -16 



85-3 -*1 -01 



85-2 --2 -04 



85-4 -*0 -0 



85-8 +-4 -16 



85-7 +-3 -09 



Mean . . 85*4 Sum . . 0-71 



The probable error of the mean result and the probable error 

 of a single observation were found by the ordinary formulae, 



0*6745 4 / sum of the squares of the differences , ^- ,, 



probable error of the mean result, n being the number of 

 observations. 



Bunsen's New 



disk. photometer, 



centim. centim. 



Probable error of mean result +0*148 +0*076 



„ „ single observation +0*418 +0*215 



The new photometer therefore appears to be twice as accu- 

 rate as the Bunsen's disk : it is only fair to add that, had the 

 measurements been made by an observer accustomed to work 

 with the disk, the result might have been different. 



