REPORT ON THE COMPOSITION OF OCEAN-WATER. 
201 
water samples by a more direct (chemical) method, and compare the resulting values 
with those computed from Buchanan’s specific gravities by my formula. This com- 
parison led to the very satisfactory result, that the “ probable error ” in any one of Mr. 
Buchanan’s specific gravities is rather less than ±0T ; that of pure water being taken 
as =1000. 
A little reflection shows that no number of analyses will enable one to calculate 
with any degree of exactitude the mean salinity of the ocean as a whole ; but even my 
160 salinity determinations, since they correspond to a great variety of places, suffice to 
give an idea of the limits between which the quantity fluctuates. Expressing the 
salinity in “parts of total salts per 1000 parts of sea-water,” I find that (of the 160 
values) 
The lowest (from the southern part of the Indian Ocean, south of 66° lat.) is . . 33 - 01 
The greatest (from the middle of the North Atlantic, at about 23° lat.) is . . 37'37 
(Some few samples from narrow straits or close to certain coasts are omitted, as being 
probably diluted to an abnormal extent with fresh water.) 
So much as to the ratio of the water to the sum total of the salts dissolved in 
it. Let us now inquire into the percentage composition of the salt mixture itself. 
A _ priori , we should say that this composition cannot be subject to any great 
variation; because, if there were no chemical changes going on in the ocean, and no 
gain or loss of dissolved individual salts, this composition would now, after thousands of 
years’ constant intermixture, be absolutely the same everywhere ; and what is going on 
in the shape of reactions and importation or exportation of individual salts, really 
amounts only to an extremely minute fraction of the whole, even in the course of a 
century. This conclusion is confirmed by the analyses of several hundred samples 
of surface-waters, which were carried out by Forchhammer in connection with a 
great research which he published in 1864.* According to his results, if we 
confine ourselves to the open ocean, we find that everywhere the ratios to one 
another of the quantities of chlorine, sulphuric acid, lime, magnesia, and total salts, 
exhibit practically constant values. With the view chiefly of supplementing Forch- 
hammer’s work, I have made exact determinations of the chlorine, sulphuric acid, lime, 
magnesia, potash, and soda in 7 7 samples of water collected by the Challenger from very 
different parts of the ocean : — 
12 from the surface. 
10 from depths of 25 to 100 fathoms. 
21 from depths of over 100 to 1000 fathoms. 
34 from greater depths. 
* Phil. Trans., 1865, vol. civ. p. 203. 
(PHTS. CHEM. CHALL. EXP, — PART I.— 1884.) 
A 26 
