December 17, 1908] 



NATURE 



1S9 



LETTERS TO THE EDITOR. 

 [The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part o/ Nature. 

 No notice is taken of anonymous comtnttnications.l 



On the Salinity of the North Sea. 



The nccompanving chart of the mean salinity of the 

 surface of the North Sea has been constructed from the 

 international observations made during the years 1903-7. 

 A similar chart has been constructed by Mr. Martin 

 Knudsen (dealing with a somewhat shorter period), and 

 our two charts, independently prepared, agree with and 

 confirm one another in a very close way. 



The general features of the chart are extremely simple ; 

 the highest salinities are found, first, around the Shet- 

 land*, and, secondly, in the neighbour- 

 hood of the Straits of Dover, and the 

 values are somewhat higher in the 

 former region than they are in the 

 latter, where the connection with the 

 waters of the ocean is more remote. 

 The salinity falls off rapidly in the 

 Skager Rack, and is, on the whole, 

 low everywhere in the immediate 

 neighbourhood of the coast. 



We may the more easily compre- 

 hend and describe the form and dis- 

 tribution of the isohalines (or curves 

 of equal salinity) by comparing with 

 them the case of the distribution of 

 temperature or of potential in a bar 

 of metal subjected to a flow of heat 

 or of electricity. In such a bar of 

 metal, heated at one end and cooled 

 at the other (as in Forbes's classical 

 experiments), we obtain a series of 

 isorlierms running transversely to the 

 " thermal axis," and arranged in an 

 exponential series at increasing dis- 

 tances as we pass towards the cooler 

 end. If, in the next place, we apply 

 new sources of heat along the edges 

 • of the bar, it is obvious that the result 

 will be to bend the isotherms from 

 straight into curved lines, concave 

 towards the cooler end of the thermal 

 axis. Lastly, if we substitute for the 

 straight and elongated bar a square 

 plate, and apply our sources of heat 

 and cold at two of its adjacent sides, 

 then the thermal axis will be bent 

 into a curve, and the isotherms will 

 be crowded together upon its con- 

 cave, and comparatively remote from 

 one another on its convex, side. 



Now, neglecting the phenomena in 

 the Straits of Dover, which are of 

 comparatively small magnitude, we 

 have in the accompanying chart a 

 series of isohaline curves which corre- 

 spond very closely indeed with the 

 isothermal system just described. 



Our axis is traceable through the Cattegat and Skager 

 Rack, along a bent course in the middle of the North 

 Sea, to its termination in the Atlantic eastward of 

 Shetl.ind; the isohalines, which are essentially transverse 

 to this axis, are everywhere rendered convex towards 

 the ocean by reason of the influx of fresh water from 

 the shore, and these isohalines, while they are com- 

 paratively widely interspaced in the southern part of the 

 North Sea and off the east coast of Great Britain, arc 

 crowded together off the coast of Norway, that is to sav, 

 on the concave side of the axis. Furthermore, we notice 

 that the whole system of curved isohalines is thrust over 

 much nearer to Shetland than to Norway, firstly, I pre- 

 sume, because it is in the neighbourhood of Shetland that 

 the oblique north-easterly track of the so-called Gulf 

 Stream, with its highly saline water, lies nearest to the 

 North Sea, and, secondly, because the inflow of fresh 



NO. 2042, VOL. 79] 



water from the Norwegian coast predominates greatly over 

 that from the coast of Britain. We may follow the 

 parallel a little further, by noticing that, just as our axis 

 is bent within the North Sea, so it is also bent, but in 

 the opposite direction, as it passes from the Cattegat into 

 the Skager Rack. Accordingly, we find in this latter 

 region a disposition of the isohalines comparable, though 

 on a smaller scale, to that within the North Sea itself, 

 for they are crowded together on the concave side of the 

 bent axis, that is to say, towards the Danish coast, and 

 comparatively widely spaced on the Norwegian ; while at 

 the same time the whole system is thrust over towards 

 the Danish side by the greater infiow from the Swedish 

 and Norwegian coasts, to which disposition, no doubt, in 

 this case, the ■ course and direction of the outflowing 

 current from the Baltic contribute. 



A chart of the mean annual variation of salinitv, which 



Mean Surface Salinity of the North Sea, 1903-7. 



variation we can show to be, on the whole, regularly 

 periodic, is found to correspond very closely in its con- 

 tours with the chart of mean salinity, for the regions of 

 highest salinity are subject to the least variation, and those 

 of the lowest mean salinity to the greatest. At the mouth 

 of the Cattegat, where the mean salinity is about 25 "/„„ 

 (or 25 grams of chlorides in a thousand parts of water), 

 the mean annual variation is nearly 10 "/ ^„ ; at the mouth 

 of the Skager Rack, where tlie mean salinity is about 

 31 °/jj, the mean variation is about 5 "/j, ; in the middle 

 of the North Sea, with a mean salinity of 34-75 °/,„, the 

 mean variation is only about 0-2 °/„„ ; and in the region 

 of our highest salinities off Shetland, of about 35-25 °/„„ 

 the mean variation is less, and probably very considerably 

 less, than o-i °/„„. .\ further discussion of this subject, 

 including an account of the distribution of salinities at 

 various depths, and of the phases and other phenomena 



