407 



this is true, there is, however, much variation, and some of the largest 

 fishes died first. 



A number of compounds show still further irregularity and in the 

 cresols the larger fishes are very often first to die. In carbon bisul- 

 phide this appears to be the rule with the concentrations studied. In 

 the case of naphthalene the larger fishes usually die first. Xylene, ben- 

 zene, and toluene show considerable variation in this respect though 

 the results with a given concentration and given weight of fish are 



particularlv constant. 



Table IV 



Shoioing typical results of lillinf/ Lepomis humilis in various compounds. In sulphur 

 dioxide the smaller fishes die first. In xylene there in some variation and in carbon 

 bisulphide the larger appear to die first as a rule. 



In the case of carbon dioxide and low oxygen, Wells ('13), found 

 that the results of dividing the time to death by the weight of the 

 fishes, secured as the time to kill a gram of fish, not a constant but a 

 value which increased rather regularly in most cases, as the size of the 

 fish decreased. In the case of the coal-tar products there is much more 

 irregularity than in the case of the conditions studied by Wells. 



Since the smallest individuals of given species usually die quickest 

 the importance of contamination is greatest in connection with the 

 younger stages. The effect upon the development of eggs is likely to 

 be great. Gortner and Banta found that various phenolic compounds 

 killed the eggs of frogs and other amphibians in concentration as low 

 as 50 parts per million. The critical point for further study is in 

 connection with the most sensitive stage. This is the weakest link in 

 the life-history chain, and represents its strength. It is the stage on 

 which minimum fatal concentration must he zuorked out. 



