CHEMICAL COMPOSITION OF RIVERS AND LAKES 



Gil 



OVERTURN 

 PARTIAL COMPLETE 



0.4 0.2 0.2 0.4 



Figure 6.— Redox potential near the mud-water interface and concentrations of dis- 

 solved substances in the water just over the mud surface of Esthwaite Water, 

 England, during 1940. Redox potential (E?) in volts, Os=dissolved oxygen, 

 Fe=fcrrous iron, Mn=manganese, P = phosphate as PiOj, NHi=ammonia, 

 NOj=nitrate nitrogen, all is in parts per million. After Mortimer, 1941-42. 

 Reprinted by permission of Blackwell Scientific Publications, Ltd., Oxford, 

 England. 



GENEEAL ANALYSES 



NORTH AMERICA 



An overwhelming mass of chemical data exists for 

 the rivers of North America. Most of the analyses 

 have been made as part of a systematic sampling 

 program of the Water Resources Division of the U.S. 

 Geological Survey and have been published in a series 

 of Water-Supply Papers: Collins, Howard, and Love 

 (1943); Collins and Love (1944); Howard and Love 

 (1945); Howard (1948); U.S. Geological Survey (1947 

 1948]; 1949b; 1950 [1951]; 1952; 1953a, b; 1954a, b, c, 

 d; 1955a, b, c, d, e; 1956a, b, c; 1957a, b, c; 1958a, b, c, 



d, e; 1959a, b, c, d, e, f; 1960a, b, c, d, e). In addition, 

 many reports deal with individual states. The follow- 

 ing is a selection from a very large number of references; 

 generally only the most recent ones of a series are 

 cited. 



U.S. Geological Survey (1960f, g, h); Saunders and 

 Billingsley (1950); Geurin (1951); Geurin and Jeffery 

 (1957) ; Smith and others (1949) ; California Department 

 of Water Resources (1956, 1957); Lamar (1944); 

 Cherry (1961); Hershey (1955); Lamar and Laird 

 (1953); Lamar, Krieger, and Collier (1955); Hembree, 

 Colby, Swenson, and Davis (1952); Lamar (1943); 

 Pauszek (1952); Pauszek and Harris (1951); McAvoy 

 (1957); Woodward and Thomas (I960); White (1947); 

 Lamar and Schroeder (1951); Ohio River Valley Water 

 Sanitation Commission (1950 [1951]); Murphy (1955); 

 Dover (1956, 1959); Beamer (1953); Pennsylvania 

 State Planning Board (1947); Pauszek (1951); Hastings 

 and Rowley (1946); Irelan and others (1950); Irelan 

 (1957); Hughes and Jones (1961); Connor, Mitchell, 

 and others (1959); Lamar and Whetstone (1947); 

 Whetstone and McAvoy (1952); Kapustka (1957). 



The older data are collected in Clarke (1924a). 

 Inventories of published and unpublished data may be 

 found in Northcraft and Westgarth (1957), U.S. Federal 

 Inter-Agency River Basin Commission (1948, 1954, 

 1956), and Westgarth and Northcraft (1956). 



Lake sampling has not been nearly as comprehensive. 

 There are extensive sets of data both in government 

 publications and the limnological literature, but most 

 of the analyses are deficient because some important 

 major ions have been neglected, or because of insuf- 

 ficient sampling. Despite the wealth of data there are 

 serious gaps in the coverage even of the rivers. There 

 does not exist, for example, any really adequate series 

 of chemical data for the lower Mississippi. 



Outside of the United States information is far more 

 scarce. It has not been possible to locate a reasonably 

 complete analysis of a single river in Mexico, and the 

 coverage for Canada is very poor. There are only a 

 few scraps of information for the whole MacKenzie 

 River system, and the chemistry of most of the northern 

 lakes and rivers is completely unknown. This is 

 particularly unfortunate because of the opportunities 

 Canada affords for the study of the geochemical 

 regimen of rivers that are completely within the tundra 

 zone. Even the waters of the well-settled parts of 

 the country, however, are represented by only a few 

 spot samples, and it is not possible to draw up a reliable 

 long-term mean for any Canadian river. 



ST. LAWRENCE RIVER BASIN 



A representative series of analyses is presented in 

 tables 6 and 7. The St. Lawrence basin is a well- 



