THE PUBLIC AQUAEIUM 325 



An interestin<r phase. illiistratin<j: the linkinfr of residual chlorine 

 with the character and use of the \Yater, is shown at the Bay City 

 hatchery. The whitefish battery is so arran(red that the jars of eggs 

 are placed in superimposed tiers, with the fresh water entering the 

 topmost series and passing through the whole battery, to be dis- 

 charged ultimately in the overflow of the lowest tier. It was found 

 that the mortality\vas highest in the upper sections, diminishing pro- 

 gressively, until the final group was largely unaffected. Manifestly, 

 in exerting its toxic action the chlorine reduces its concentration until 

 it becomes innocuous to succeeding lots of eggs. The same aspect is 

 sho^^^l in experiments when a rainbow trout of 18 centimeters length 

 reduced an initial chlorine content of 0.3 part per million to 0.05 part 

 per million in one hour. The dose was fatal, however, without fur- 

 ther increment of chlorine. Other practical instances of the cumula- 

 tive action of chlorine appear when a lot of fingerlings without 

 previous subjection to chlorine was divided. One group was placed 

 in a trough supplied with the regular flow of constantly chlorinated 

 water as fed to the whole aquarium. The others were placed in 6 

 liters of water from the same source in glass jars and maintained by 

 aerating the water artificially and by infrequent changes. Within a 

 Aveek the specimens in running water were dead, while from the other 

 lot there were survivors at the end of three weeks, the majority having 

 been utilized in experiment. A nonlethal concentration thus becomes 

 lethal under constant administration. 



The process of autoneutralization exemplified in the successful 

 preservation of fingerlings in the jars indicates that chlorinated 

 water may be used with little hesitancy for the balanced home aqua- 

 rium, where changes are made only at rare intervals. If the tap 

 water is allowed to stand several days, residual chlorine should ordi- 

 narily be far below the critical concentration. The same action, of 

 course, is observed in the presence of nonliving organic matter, which 

 may act as a reducing agent, but the marked decrease of the chlorine 

 content with fish may probably be credited to physiological factors 

 as well as interaction with organic matter, such as the mucus of the 

 fish. A quantitative determination of the relative importance of the 

 two influences is lacking at present. 



CHEMICAL ASPECTS OF CHLORINE REMOVAL 



Very small concentrations of free chlorine, if maintained, are toxic 

 to fish. However, larger doses exhibit their toxic effects more rapidly, 

 and a fish whose " gassing " has progressed to a serious stage gener- 

 ally will not recover even if transferred to water to which no chlorme 

 has been added. Hence, in studying methods for counteracting or 

 removing the chlorine it was decided to use fish in small a(i[uaria and 

 to introduce chlorine only once but in sufficient quantities that there 

 would be no doubt as to the effect. Rainbow-trout fingerlings seem 

 to be most susceptible to chlorine and for this reason were used in all 

 the experiments. 



The chlorine used was generated by the action of hydrochloric acid 

 on potassium permanganate, washed to remove acid and dissolved in 

 water to form a concentrated chlorine solution. This was then stored 

 in glass-stoppered bottles in the dark until needed. Two methods of 



