40 
Proceedings of Royal Society of Edinburgh. [sess. 
introduction of any perceptible error, that for all the five samples, 
the value for the minimum deviation of the D line diminishes by 
31" for an increase in temperature of 1° C. The following Table 
E gives the observed and reduced values for the minimum devia- 
tion for the several waters. 
Table E. 
Series. 
Water. 
Observed deviations 
in secs, of arc. 
Temp, at 
which obser- 
vation was 
made. 
Deviations re- 
duced to 24° C. 
I. 
l v 
86,796" 
22°-9 C. 
86,761" 
2 V 
786 
22°-7 „ 
744 
3 V 
774 
22°'7 „ 
734 
4 V 
785 
22° '5 ,, 
737 
5 V 
800 
22°-3 ,, 
746 
II. 
l v 
86,804" 
22°«8 „ 
86,766" 
2 V 
768 
23°-l „ 
739 
3 V 
754 
23° *3 „ 
732 
4 V 
760 
23°*5 „ 
744 
5 V 
772 
23° -6 ,, 
759 
From this table it will be seen that the means of the reduced 
minimum deviations obtained from series I. and II. are — 
For l v 86,764 seconds of arc at 24° C. 
„ 2 V 742 „ 
„ 3 V 733 „ 
„ 4v 741 „ 
„ 5 V 753 „ 
Discussion. 
When investigating similar kinds of water by the optical method, 
we may express the differences observed in various ways ; but for 
our present purpose it will be sufficient if we consider two only. 
(a) We may select a prism having a strictly constant refracting 
angle of say 60°, and determine the values for the minimum devia- 
tions, 8, d lt d 2 , etc., of the D line for recently re-distilled water, 
and the waters under examination, a standard temperature being 
