1899-1900.] Mr J. J. Manley on Examination of Sea - Water. 41 
maintained throughout ; or if the temperature is unavoidably 
variable, proper corrections determined from time to time, must 
be applied in order to reduce all the observed deviations to a 
common temperature. The ratios etc., may then be obtained 
o o 
and compared. This method is analogous to a determination of 
the relative density of a liquid at some standard temperature, and 
may be termed the relative deviation. 
(b) The refractive angle of the prism may be determined in 
addition to the minimum deviations of the D line by the waters, 
and from these data, the refractive indices, /z, p lf /z 2 , etc., for 
recently re-distilled water and the waters to be compared may be 
calculated. 
By the first or relative deviation method, a number of samples 
of water can be examined far more quickly than by the relative 
density method which is so generally adopted ; this is due to the 
ease and rapidity with which a minimum deviation observation 
may be made, and to the fact that the calculation is of the same 
simple order as that used for obtaining the relative densities. 
Let S D and represent the minimum deviations of the D 
line by sea-water and re-distilled water respectively, then the 
S 
ratio ^ — gives the relative deviation. Applying this method to 
D 
the waters under examination, and expressing the several minimum 
deviations in seconds of arc, we obtain the values shown in 
Table F. 
Table F. 
Min. deviation W^, for re-distilled water at 24° C. =85,018". 
Water. 
l v . 
2 V . 
3 V . 
4 V . 
5 V . 
Min. deviation S D at 24° C. 
86,764" 
86,742" 
86,733" 
86,741" 
86,753" 
Ratio — 5- 
W D 
1-02054 
1-02028 
1-02017 
1-02027 
1-02041 
