CHEMICAL CHANGES IN THE COMPOSITION OF SEA-WATER. 495 



1*4585, the minimum 1*4535. Mr Dickson from an examination of 42 waters in the 

 English Channel obtained a mean D value of 1*4550, the maximum being 1*4580, the 

 minimum 1*4490.* 



Dr Konrad Natterer, in his papers on the chemistry of the Mediterranean water,t 

 gives the mean D value at a temperature of 17°*5 C. of forty-two Mediterranean waters 

 (after striking out the abnormal values) as equal to 1*3814, which is slightly higher than 

 that obtained by us for the waters examined ; the D at 0° C. is not given. 



The mean D A for the "Challenger" waters was 0*533, that for the above table 

 (D, page 493), excluding abnormal waters, 0*558, but the D A value is much more subject 

 to variation than the D value. The D A of the mud- water in Table IV. is as low as 0*05. 

 In the Mediterranean, Dr Natterer found values for D A ranging from 0*5622 to 0*6915, 

 with a mean of 0*59231. j 



In some of those waters in Table D, which have been preserved for several years, it 

 will be noticed that the alkalinity is decreased, in some increased, while in others, where 

 the bottles were quite full and the stopper tied round with parchment, there is no change. 

 In the waters with decreased alkalinity the bottles were half empty, or the waters had 

 partly evaporated, and there was a slight deposit of carbonates. In the waters with 

 increased alkalinity, small carbonate of lime organisms may have been bottled accidentally 

 along with the water, or a slight decomposition of sulphates may have taken place. 

 The water ought always to be strained through fine silk to remove the organic matter. 

 The waters which contained the remains of chitinous or calcareous organisms were high 

 in alkalinity, especially No. 25, in which we detected the remains of numerous organisms. 



The Reactions that take place in Blue Muds, and the admixture of Mud- Waters with 

 the normal overlying Sea- Water. — Should the water associated with the deposits at the 

 bottom of the sea pass, with its altered composition, into the water above, not only will 

 the overlying sea- water be greatly increased in alkalinity, but the carbonic acid will 

 likewise be increased in quantity above that found in normal sea-water. This may in 



* See Gibson, Proc. Boy. Soc. Edin., 1893, and Dickson, Journal Scottish Geographical Magazine, Jan. 1893, p. 17. 



t Chemische Untersuchungen im ostlichen Mittelmeer, Wien, 1 892. 



X Various constants are used by authors in expressing the relations which occur in sea-water analyses. 



1. The D value may be taken at any temperature, and may be expressed by D ( The D t (on the assumption that 

 there is no appreciable difference in the chemical composition of sea-water in different regions) is a constant for any 

 temperature at which the observations are made as well as at 0° C. ; but when the temperature is not mentioned, it is 



t S t - 1000 



understood to be D at 0° C. It may be represented by = D^ . 



/C 



2. The D A value, as stated in the text, is the density at 0° C. minus 1000, divided by the alkalinity (in milli- 

 grammes of carbonic acid per kilogramme of water), and expresses the relation of alkalinity to density, in the same way 

 as D expresses the relation of chlorine to density, and, like D, is unchanged by dilution or concentration of the water. 



3. The relation of chlorine per kilogramme to total salts per kilogramme is also expressed by a constant which 

 Dittmar gives as 1-8058. The chlorine (x), multiplied by this figure and divided by 10, gives the percentage salinity 

 of the sea-water. As has been pointed out by various writers, such as Pettersson, Ekman, Krummel, and Gibson, 

 and confirmed by ourselves, this constant may rise much higher in brackish water, such as the Baltic, varying from 1 - 801 

 to 2-15. (See Pettersson, Grunddragen af Slcagerachs och Kattegats Hydrografi, Stockholm, 1891, and Krummel, Geo- 

 physikalische Beobachtungen der Plankton Expedition, Keil, Leipzig, 1893.) 



4. A constant used by Krummel is that of the density to the total salts. Its value is given as 1312, and is found 

 by dividing the total salts per kilogramme by the density at 17° - 5 C. after deducting 1 (pure water at 17° - 5 C. = l). 



