REPORT ON THE COMPOSITION OF OCEAN- WATER. 
53 
The salinity of a sea-water could not possibly be more succinctly defined than by the 
statement of the percentage or permilleage of chlorine ; hence our chlorine-determinations 
could well be allowed to stand as they are. But they are not sufficiently numerous 
to have any value, except as a small addition to, and as affording to some extent a check 
upon, Mr. Buchanan’s extensive liydrometer-work. Hence it is not so much the values 
of y as the values y 1 — y in Column IX. which give the practically useful result of my 
analytical determinations. The constants which I needed for calculating the values y 1 , 
corresponding to Mr. Buchanan’s specific gravities, were ready to hand in the memoir* 
published by the chemists of the Norwegian North Atlantic Expedition ; and judging b}^ 
the general character of their work, I had no doubt in my mind about the substantial 
correctness of their results. But I thought I had better at least check these numbers 
by a series of experiments of my own ; and when I had once entered upon this work, I 
somehow allowed it to expand into a more general research into the matter, which I will 
now proceed to give in extenso, hoping that it may prove of some utility to physical 
oceanographers. 
The Specific Gravity of Ocean-Water, a Function of Salinity, Temperature, and 
Pressure. 
The pressure, as an independent variable, of course comes into consideration only in 
reference to an ocean-water as it is in situ ; and, for oceanographic purposes, it might, 
perhaps, be assumed to be measured with sufficient exactness by the depth of the super- 
incumbent layer of water. But this datum, in many of our Challenger samples, assumes 
rather high values, and its influence on the specific gravity then is greater than one might 
be inclined at first sight to think. According to Grassi (Ann. d. Chim. et de Phys., ser. 3, 
vol. xxxi. p. 437), sea-water of 1026*4 sp. gr. at 17°*5 C. when subjected to pressure, 
contracts by 43*6 millionths of its original volume per atmosphere of over-pressure. Now 
taking x as designating the number of fathoms (1 fath. = 72 in.), exerting a pressure equal 
to that of 30 inches of mercury, wc have 30 x 13*596 = 72 x 1*026 x x, whence x— 5*521. 
Hence a sea- water of 1*026 sp. gr. under the pressure of 8 fathoms of water of the same 
density undergoes compression to the extent of ^^ -- — 7*9 x 8 millionths of its volume ; 
its density consequently rises from 1*026 to 1*026 + ; when 8=1000 fathoms, for 
instance, we have an increase AS = 0*007 9 ; i.e., the actual density in situ would be = 1 *034 
instead of 1*026 at the ordinary pressure. And a good number of the Challenger samples 
were procured from depths of 3000 fathoms and more. It is very desirable that Grassi s 
research should be extended to waters of different strengths, and to different temperatures. 
* Den Norske Nordhavs-Expedition, 1876-78; Chemi, p. 47, &c. 
