CHEMICAL COMPOSITION OF RIVERS AND LAKES 



G5 



100 -f 



10 



EXPLANATION 



Discharge 

 Conductance 



-v' V 



OCT 



NOV 



DEC 



FEB 



Figtjke 2.— Relation of specific conductance to mean daily runoff of the Saline River 

 near Russell, Kansas, during part of 1946 and 1947. After Durum (1953). Re- 

 printed by permission of the American Geophysical Union. 



fluctuations of discharge on a daily basis, but daily 

 measurements of specific conductivity are available 

 for many rivers in the United States. An example is 

 provided by the Saline River, Kans., in figure 2. 



In addition to these temporal variations in the 

 chemistry of rivers, there are also spatial ones. It has 

 been known for a long time that the content of dis- 

 solved matter of river water tends to increase from 

 source to mouth. This tendency is particularly marked 

 in regions of interior drainage, but it is also present in 

 rivers emptying into the sea. A further complication 

 is introduced by heterogeneities in river water at any 

 particular level in the drainage profile. When two 

 rivers meet, or when large amounts of chemically 

 different water are introduced into a river in some other 

 way, for example, by a large spring or a sewage outflow, 

 there may not be complete mixing for a long distance 

 downstream, as Heide (1952) has shown. 



In any large river system the composition of the 

 dissolved salts is different in the various head-water 

 tributaries, but these local irregularities, which reflect 

 variations in the nature of the rocks in the various 

 parts of the drainage system, tend to cancel each 

 other as one proceeds downstream, and there is a 

 tendency for the composition of the water in the down- 

 stream parts of rivers to resemble one another. This 

 has led to the concept of a general or mean composition 

 of river water (Rodhe, 1949) and to some speculation 

 that ion-exchange reactions with the suspended load or 



Table 5. — Mayo River near Price, N.C. 



[Drainage area above sampling station 260 square miles. Note the relatively small 

 variations In water chemistry of this humid-climate stream with rather constant 

 discharge. Data from U.S. Oeol. Survey (1954b)] 



Date 



1949 



Oct. 1-10 



Oct. 11-20 



Oct. 21-31 



Nov. 1-10 



Nov. 11-20 



Nov. 21-30 



Dec. 1-10 



Dec. 11-20 



Dec. 21-30 



1950 



Jan. 1-10 



Jan. 11-20 



Jan. 21-31 



Feb. 1-10 



Feb. 11-19 



Feb. 20-28 



Mar. 1-10 



Mar. 11-20 



Mar. 21-31 



Apr. 1-10 



Apr. 11-20 



Apr. 21-30 



May 1-10 



May 11-20 



May 21-31 



June 1-10 



June 11-20 



June 21-30 



July 1-10 



July 11-20 



July 21-31 



Aug. 1-10 



Aug. 11-20 



Aug. 21-31 



Sept. 1-10 



Sept. 10-20 



Sept. 21-30 



with the soil might be buffering river water and reducing 

 it to a common composition in all parts of the world. 

 This implies a very large exchange capacity for the 

 participating solids in the system, which poses some 

 difficulty when we remember that the uniformity of 

 river water is most pronounced in the downstream 

 parts, where the simple geometry of streams renders 

 the contact between water and solids minimal. It also 

 seems to be unnecessary, for the similarity of the 

 various large river waters is apparently quite adequately 

 explained as a result of the integration of the chemical 

 composition of their tributaries. The larger the drain- 

 age basin of a river, the closer, on the average, will the 

 chemical composition of its rocks approach the mean 

 chemical composition of the surface rocks of the earth, 

 and the closer will the composition of the water which 

 it empties into the sea approach the mean composition 

 of all waters. The same line of argument holds for 

 that part of the salt which is of meteoric origin. There 

 does not seem to be any need to invoke sorption reac- 



