122 Proceedings of Indiana Academy of Science 



di die -W2 



From the above — = r — -- = e''^'''^ 



ds doe -a3x/2 



d2 



— — pax 



d3 



di da da ^^ -. -_^ 



— = — = — =. . . = e'^': = e^ = K 

 dj ds d4 



di 

 Where X = otx = log^ —. 



From equation (1) 



do = die«^/2 = d,e^V2 

 since e^ = 1 + x + x'-/'-^ + • • . 



gX 2 = 1 _^ 1/2X if X is small, 

 and do = d,(l + 1/2X). 

 This equation is the one usually used in making correction for decrement. 

 Since e^ = eV2 x e^/^ = K 



eX/2 = »' K 

 and (in = di' K 

 why not multiply by the square root of the ratio instead of by one plus one- 

 half the logarithm of the ratio? 



