STREAM POLLUTION 
375 
The variation in the amount of oxygen consumed by fresh-water fishes of dif- 
ferent species is apparent from the work of Gardiner, King, and Powers (1922) who 
give the oxygen consumption of the brown trout as from 90 to 200 cubic centimeters 
per kilo of body weight per hour at temperatures between 4° and 20° C.; of the 
goldfish as 16 to 90 cubic centimeters; and of the eel 9 to 60 cubic centimeters. 
The lower limit for dissolved oxygen — that is, the point at which the dissolved 
oxygen- — is so reduced as to present a lethal condition for fishes is equally difficult to 
define. Kupzis (1901) reports that in general the cyprinid, Leuciscus erythrophthalmus 
(the European roach), could live for sometime in water containing 0.7 cubic centimeter 
of dissolved oxygen per liter (1 p. p. m.) but that this species of fish died from asphyxia 
when the dissolved oxygen was reduced to 0.4 to 0.5 cubic centimeter per liter 
(0.57-0.71 p. p. m.). Plelm (1924) states that trout live best in water containing 7 to 
8 cubic centimeters of dissolved oxygen per liter (10-11.43 p. p. m.), but if the water 
be warm these fish show some discomfort when the dissolved oxygen is reduced to 
5.5 cubic centimeters per liter (7.86 p. p. m.). This author also states that carp live 
well in water containing 5 cubic centimeters dissolved oxygen per liter (7.1 p. p. m.), 
but show respiratory difficulties when the dissolved oxygen is reduced to 3 cubic 
centimeters per liter (4.3 p. p. m.) or lower. If the water be cold, carp can live for a 
short time in water containing only 0.5 cubic centimeter per liter of dissolved oxygen 
(0.71 p. p. m.). In order of their oxygen requirements, Plehn lists first the salmonids 
and coregonids, then the barbe, the rutte, the pike, carp, tench, goldfish, and lowest of 
all the eel. Gardiner and King (1922) give asphyxia! point for trout as from 0.8 
cubic centimeter of dissolved oxygen per liter (1.14 p. p. m.) at 6.5° C., to 2.4 cubic 
centimeters per liter (3.4 p. p. m.) at 25° C.; and for goldfish as 0.39 cubic centimeter 
per liter (0.56 p. p. m.) at 11° C., and 0.42 cubic centimeter per liter (0.6 p. p. m.) at 
27° C. Paton (1904) found that harmful to fatal conditions for young trout developed 
if the dissolved oxj^gen were reduced to 2 cubic centimeters per liter (2.9 p. p. m.) or 
lower, although some individuals were able to live in such waters for long periods of 
time. 
Thompson (1925) states that carp and buffalo have been found living in water 
carrying as little as 2.2 p. p. m. of dissolved oxygen. As a rule, he found a variety of 
fishes only when 4 p. p. m., dissolved oxygen were present, and the greatest variety 
of fishes were taken from waters carrying 9 p. p. m. of dissolved oxygen. His observa- 
tions made at Peoria Narrows, 111., in the summer of 1923 showed that fishes died 
over night in waters having less than 2 p. p. m. of dissolved oxygen. Of the various 
species of Illinois fish discussed by Thompson the dogfish, Arnia calva, seemed to be 
the most sensitive to low oxygen tensions and the carp, the most resistant. 
From the various observations cited above, which may be taken as representative 
of the voluminous literature on the oxygen requirements of fishes, it may be seen that 
the upper limit of dissolved oxygen at which asphyxia may be expected in fresh-water 
fishes if there be no unusual complicating factors, is in general about 3 p. p. m. at 25° 
C. In evaluating the dissolved oxygen level as a lethal hazard to fish life, this upper 
asphyxial limit must be considered rather than the lower minimal limit of dissolved 
oxygen which can be tolerated by some fish for varying periods, particularly under the 
conditions of rapid or abrupt oxygen reduction presented in so many of the experimen- 
tal studies, since the actual hazard to fish life begins at the oxj^gen level where 
