previous chapter
Chemistry of Sea Water
next sub-section

Constancy of Composition

It has been found that, regardless of the absolute concentration of the total solids, the ratios between the more abundant substances are virtually constant. The importance of this result cannot be overemphasized, as upon it depends the validity of the chlorinity:salinity:density relationships and, hence, the accuracy of all conclusions based on the distribution of density where the latter is determined by chemical or indirect physical methods such as electrical conductivity or refractive index.


The relative uniformity in the composition of the sea water was established by the investigations of Forchhammer, Natterer, and Dittmar. Although Forchhammer analyzed a large number of samples, his investigations were not complete because he did not determine certain of the abundant elements. Natterer made more detailed analyses, but it was Dittmar who laid the solid foundation for the present knowledge of the composition of sea water.

Dittmar (1884) made careful determinations on 77 water samples, representative of all oceans, which had been collected on the voyage around the world of H.M.S. Challenger. He determined the halides, sulphate, magnesium, calcium, and potassium. On composite samples he found the ratio of bromine to chlorine and estimated the carbonate. From the sums of the chemical equivalents of the negative and positive ions, he calculated the sodium by difference. This procedure was followed because he was unable to achieve satisfactory direct determinations for sodium. The results of Dittmar's work showed that there were no significant regional differences in the relative composition of sea water; consequently his average values could be used to represent the ratios between the major dissolved constituents. In table 33 are given Dittmar's average values in the units in use at the present time and referred to a chlorinity of 19.00 ‰. The percentages of the various ions are also shown.

Ion Original values Recalculated, 1940 atomic weights 1940 values
Cl = 19 ‰ % Cl = 19 ‰ % Cl = 19 ‰ %
C1 18.971 55.29 18.971 55.26 18.980 55.04
Br 0.065 0.19 0.065 0.19 0.065 0.19
SO42.639 7.69 2.635 7.68 2.649 7.68
CO3 0.071 0.21 0.071 0.21 …… ……
HCO3 …… …… …… …… 0.140 0.41
F …… …… …… …… 0.001 0.00
H3BO3 …… …… …… …… 0.026 0.07
Mg++ 1.278 3.72 1.292 3.76 1.272 3.69
Ca++ 0.411 1.20 0.411 1.20 0.400 1.16
Sr++ 0.013 0.04
K+ 0.379 1.10 0.385 1.12 0.380 1.10
Na+ 10.497 30.59 10.498 30.58 10.556 30.61
Total 34.311 34.328 34.482


Since 1884 the modification of atomic weights has affected the numerical results reported by Dittmar. Corrections for these changes may be made (Lyman and Fleming, 1940) as shown in the “recalculated” values in table 33. In the latter tabulation the sodium has been recalculated by Difference.

It is interesting to compare Dittmar's results with those obtained by modern methods of analysis as shown in the last columns of the table. The sources of these data are indicated in table 35. It is immediately seen that there are small differences for most of the elements determined by Dittmar and that certain other ions have been added to the list of major constituents. The bound carbon dioxide is reported as bicarbonate ion instead of as carbonate, strontium is given by itself instead of in combination with calcium, and fluoride and boric acid have been added.

The close agreement between the results of Dittmar and those obtained recently is remarkable when we consider the complexity of the problem and the great advance in knowledge of analytical chemistry. However, although the differences are small, they are significant, and hence the importance of Dittmar's work is that it showed the constancy of the ratios between the major constituents, and not that it led to accurate numerical values of these ratios.

In table 33 the composition is shown by referring the substances to a standard concentration, C1 = 19.00 ‰, and by means of the ratios between the different ions and the total dissolved solids. In most instances it is preferable to use a third method; namely, to give the ratios between the various substances and the chlorinity or the chlorosity (p. 52), and these ratios are known as C1-ratios and chlorosity factors, respectively. The Cl-ratio is the amount of any ion or substance per unit (gram) of chlorinity, and is obtained by dividing the concentration in grams per kilogram by the chlorinity, or the concentration in grams per 20°-liter by the chlorosity. Multiplication of the C1-ratio by a given chlorinity or corresponding chlorosity will give the concentrations as grams per kilogram or per liter, respectively. Concentrations in milligram-atom units are always on a liter basis, and, if divided by the chlorosity, yield the ratios that are called chlorosity factors. It may be noted that a chlorosity factor multiplied by chlorinity yields the concentration in milligram-atoms per kilogram.

The uniformity of relative composition in the oceans is the result of circulation and mixing. These operations are continuous, and tend to eliminate regional differences in composition, whatever the cause. Disturbing agencies bring about changes that are small compared to the bulk of the substances present and consequently will not materially affect the relative concentration of the major constituents. Further-more, many of the disturbing processes that tend to modify the relative composition are reversible. For example, the secretion of calcium

carbonate by organisms, which reduces the quantity of calcium in solution, takes place at a certain season or in certain parts of the sea, but upon the death of the organisms the calcium carbonate may dissolve in other regions. Although small regional differences may result from such processes, the effects are largely neutralized by mixing. On the basis of parallel determinations of density by means of interferometer (p. 54) and chlorine titration, Lotte Möller (Bein, Hirsekorn, and Möller, 1935) has shown that very small systematic differences exist in the composition of water masses of the North Atlantic, but as yet these are significant only as refined means for tracing water masses of certain characteristics.

The constancy of composition is, as already emphasized, of the greatest importance. Not only is it the basis of the chlorinity:salinity:density relationships, but it also affords a means of estimating the concentrations of all of the major constituents when the concentration of any one of them is known. Furthermore, results of studies on the composition or the physical properties of sea water in any locality are generally applicable to the water in any other part of the oceans.

Except in special areas, such as in the Baltic Sea, the Black Sea, and off the mouths of large rivers, it is not necessary to consider that the water represents special local types with properties that differ from those of sea water in general. Nevertheless, it should be remembered that the composition is not absolutely constant even for the major constituents listed in table 33. Various factors which will be discussed in detail later are always operating and always tend to modify the relative abundances. Rivers introduce dissolved material in proportions that are markedly different from those in the sea, and they also introduce sedimentary material that reacts in various ways with the dissolved constituents. The formation and melting of sea ice may bring about a modified distribution of the dissolved substances.

Thus far, comment has been largely restricted to those constituents of sea water that are present in large, or at least relatively constant, proportions. If we consider those elements which are present in small quantities and which are utilized by marine organisms, the concept of constant composition is no longer generally valid, because the concentrations of these elements vary widely, particularly near the surface. A great part of the work in chemical oceanography is now devoted to determining the space and time variations in variable constituents, and much thought is directed toward the solution of the problems related to the processes that control the observed distribution.

previous chapter
Chemistry of Sea Water
next sub-section