'08 



DR HUGH ROBERT MILL ON THE 



In this summary, group 1 may be omitted from consideration on account of the small 

 number of cases which compose it, and because in water of such slight salinity the actual 

 amount of barium sulphate weighed in determining the sulphuric acid was very small and 

 difficult to weigh accurately. No reason appears for supposing that any of the other 

 determinations are subject to error from which the rest are free. 



Taking the mean of 



Table XXXV. — Relation of Sulphuric and Carbonic Acids to Salinity. 



Group. 



No. of 



Range of Chlorine 



Corresponding 



Range of 



Percentage 



of Normal 



Sea Water. 



Mean 



Mean 

 Percentage 



S0 3 



Amount of 



C0 2 



Milligrms. 



Cases. 



Grammes per litre. 



Chlorine. 



of 

 Sea Water. 



Chlorine 

 = 1. 



per 55'45 

 grs. CI. 





Surf. Deep. 















1 



8+ 



1-169-14-982 



4-0-73-1 



8-963 



42-0 



0-11255 



01591 



2 



10+ 



16-009-16-782 



82-7-86-0 



16-487 



85-7 



0-11740 



0-1478 



3 



5+ 1 



16-880-17-224 



86-5-88-0 



17-059 



87-2 



0-11753 



0-1495 



4 



5+ 8 



17-433-17-693 



89-4-90-8 



17-555 



90-1 



0-11671 



0-1458 



5 



14 + 15 



17-800-18-280 



91-4-93-9 



18-065 



92-8 



0-11727 



0-1466 



6 



7 + 23 



18-292-18-492 



94-0-94-9 



18-393 



94-5 



0-11713 



0-1473 



7 



1 + 14 



18-548-18-776 



95-2-96-6 



18-639 



95-8 



0-11727 



0-1478 



8 



2+ 4 



18-844 — 18-946 



97-0-97-5 



18-888 



97-2 



0-11678 



0-1449 



groups 2 to 5, which contain 57 cases, and of groups 6 to 8, which contain 51 cases, we 

 find that the proportions are as follows : — 





Chlorine. 



Percentage 



Sulphuric 



Carbonic 





Sea Water. 



Acid. 



Acid. 



2-5 



17-291 



88-9 



0-11723 



0-1474 



6-8 



18-640 



95-8 



0-11706 



0-1467 



Here there appears a distinct suggestion that in water of low salinity the proportion 

 both of sulphates and of carbonates is greater than in water of higher salinity. In other 

 words, the greater the proportion of river water the higher is the percentage of sulphates 

 and carbonates in comparison with chlorides. It is true that no regular change in the 

 proportion of the salts is shown by the individual data arranged in order of salinity, and 

 even the arrangement into groups of means, as in Table XXXV., fails to give a clue to 

 the numerical relation. The results, indeed, are so irregular that the relation pointed 

 out is rather suggested than demonstrated — the low values of both sulphuric acid and 

 carbonic acid in group 8 and their high value in group 3 are very difficult to understand 

 on the theory of the influence of river water. Groups 6 to 8 are almost entirely com- 

 posed of water from depths beneath the surface, while in groups 2 to 4 surface waters 

 predominate. Group 5 contains a nearly equal number of each kind, and, in the small 



