482 
Proceedings of the Royal Society 
when it was possible for this to be done. Table IY. shows the 
.result of the observations, which are unfortunately very fragmentary. 
The rise of temperature between the hours 10 and 18 was found 
to be about 2° as an average. For the month the variation may 
be described as a gradual fall of temperature with a sudden 
depression on the 13th — the day of heavy spate already alluded to 
— from which there was a partial recovery. 
Conclusion. 
The importance to engineers of a knowledge of the relationship 
between salt and fresh water at the mouths of rivers is generally 
allowed; and much information regarding the variations of salinity 
in estuaries with tide and depth doubtless exists in the note-books 
of marine surveyors, but so far as we know, no definite investiga- 
tion of the matter has been published. Since our work was done 
without knowing the volume and velocity of the fresh water stream, 
there is a want of that quantitative element which gives precision 
to an engineering survey. The two objects are, in fact, distinct; 
but so much can be done by each, that it would form a definite 
advance in our knowledge of the circulation of water in the world 
if a conjoint study of the physical properties of a series of typical 
rivers and firths could be made simultaneously by a chemist and 
engineer, somewhat on the lines of Mr E. W. Peregrine Birch’s 
work on the Thames,* but in a more complete and satisfactory 
manner. 
A quantitative analysis of the soluble mineral constituents in the 
Spey river water is at present being carried out by one of us, with 
the view of ascertaining the influence of the river water on the 
ratio to each other of the principal components of the salt in the 
sea water near the estuary. 
Our work on the Spey corroborates the observations of engineers 
on the water at the mouth of tidal rivers, and shows more precisely 
than has to our knowledge previously been done, in this country 
at least, how sea and river water meet and mingle. We are also 
able to divide river estuaries into at least three classes, the condi- 
tions of all these being perfectly distinct. 
* Min. Proc. Inst. C.E . , lxxviii. 212, and lxxxi. 295. 
