78 
MR. H. N. DICKSON ON THE CIRCULATION OF THE 
Arranged 
were :— 
Lat. N. 
No. of 
samples. 
100 SO: 
Mean 
Under 50° 
139 
11-74 
50° to 60° 
CO 
11-79 
Over 60° 
110 
11-76 
according to the chlorine contents of the samples, the mean values 
No. of 
100 SO3 
Cl. 
samples. 
Cl 
0 
to 5 
2 
11-49 
5 
„ 10 
0 
11-64 
10 
„ 15 
0 
0 
11-69 
15 
,, 17-5 
15 
11-77 
17' 
■5 „ 20 
286 
11-75 
Over 20 
14 
11-74 
of which the four samples with chlorine less tlian 10 may be regarded as untrust¬ 
worthy, on account of the small amount of the total sulphate v'eighed. 
The net result is to show that so far as the surface waters of the North Atlantic 
are concerned, the proportion of sulphates to chlorides (X 100) is 11'754, and that no 
variation takes place in this Cjuantity which can be determined by tliese methods. It 
may be interesting to note here that a mixture of samples collected in the Faeroe- 
Shetland Channel and North Sea on board tl.M.S. “Jackal” n 1893, gave the 
following values in January, 1896 :— 
19-473 
•00216 mean of 8 determinations 
LflO 
showing a diminution in the proportion of sulphates far beyond i^ossible limits of error, 
probaljly caused by reduction of the sidphates by organic matter. It seems therefore 
necessary that sulphate determinations pretending to high degrees of accuracy must 
be made fi-om fi-esh samples. 
In calculating salinities (^ 9 ) from the values of Cl, the tables in the previous 
work (32) were extended Ijy the formula and constants there emplo 3 'ed, and used 
throughout. As the corresponding tables for obtaining salinity from given values of 
4 ,S ^5 were computed from formulre based on an extension of the same data through the 
chlorine values (33), the values of 2 ^ obtained from the chlorines (p Cl) and from the 
Sprengel determination (p S) are not independent. As I fully expected when the 
work was begun that the ratio .^S^j/Cl found for the standard samples (34) would hold 
good all througlr, tlie diherences in the values of p Cl and j) S seemed an eas}* 
Cl = 
SO3 = 
100 SO3 _ 
~cl 
