362 PROCEEDINGS OP THE AMERICAN ACADEMY. 



and in mine, so that the variations in Tables XXIX.-XXXI. are due 



to experimental error. The agreement in results between the two 

 sets is shown in Table XL., where I give in the first column the value 

 of the exponential factor w + 1 from the formula 



i^-hy) {y-s,xyiz"+'=c, 



and in the second column the values for the simplified integration 



As will be seen, the values of w + 1 are identical, the values for los 

 -ff, though very close, are not quite the same. This may be due to 

 inaccuracies in the work, but I am more inclined to attribute it to differ- 

 ences in temperature. It is not known at what temperature PfeifFer 

 worked, and it would take only a slight difference to account for the 

 variation. In Table XLI. I have tabulated the w + 1 values from 

 Pfeiffer's results, together with log C and log K. 



* My own measurements. 



