78 COMPLEX SOLUTIONS.—MACGREGOR. 
potassium sulphate by 2, concentration in gramme-equivalents 
per litre by N, and ionization coetticient by a. 
By Kay’s MeErTHop. } 
eee 
Ni Ne Ist 2nd 3rd 4th “f 
Approx. Approx. Approx. Approx. 
ay Q | % ay ay a, a, | %} ay | Ge |i 
O01 .03 .509 | .799 | .469 | al Nebr) |) HOSS ea allaoo all 026! ooTh 
OL 04 A ialeeoom) 2407 1) EGGH A085) 2767 es al etter 404).777 | 
.002 | .2 342) | 2658 | .281 | (658° || 22827) .658 ||... |5.....| e20u),o09 
2 | 002 | 242] .658 | 341] 712 | 941 | .712 |....|...-] 34n},72 
oA al 276 | .083 | .209 | .686 | .260 | .637 260. .626 .256).639 
i} 
The above table shows that for the solutions to which it 
applies, the ionization coefficients given by Kay’s method agree 
closely with those given by mine, the differences being in no 
case greater than a little over 1 per cent., and in most cases a 
small fraction of 1 per cent. It is worth noting that in the cases 
in which a difference exists, the second approximation values of 
Kay’s method are in general less divergent from mine than those 
given by higher approximations. 
If Kay’s method involved considerably less labour than mine 
it would be worth while to carry out a more extensive com- 
parison in order to determine its general trustworthiness. But 
the saving of labour, after a little practice with my method, is so 
slight, that such a comparison is unealled for. In eases in which 
either the available data do not admit of the determination of 
precise values of the ionization coefficients or only approximate 
values are desired, sufficiently good values may be obtained, 
with somewhat less trouble, by the use of Kay's method. But 
in cases in which precise values are desired, and the data are 
sufficiently exact to give them, the more exact method is to be 
preferred, notwithstanding the slightly greater labour which it 
involves. 
