563 



of this gas would rapidly lose its toxicity when exposed in open dishes. 

 This was found not to be the case, however, for a saturated solution 

 of CO did not lose its toxic properties even after two weeks' exposure. 

 A liter of the saturated solution from A (Fig. i) was placed in each 

 of four 5 in. X 8 in. battery jars and the jars were set in a stream 

 of running water to keep the temperature constant at i8° C. A liter 

 of tap water was placed in a fifth jar and set beside the other jars. 

 Two small fish {Lcpouiis huniilis) weighing between 3.5 and 8 grams 

 each were placed in each jar. In the CO solution the fishes died 

 very quickly, while those in the tap water continued to swim about 

 normally. The dead fishes were removed at once and other indi- 

 viduals of the same size and species (except as noted) were placed in 

 the jars at intervals during the next two weeks. All of the fishes 

 placed in the CO solutions died, while the control pair was normal 

 tliroughout the entire time and for two weeks afterward, when they 

 were removed. 



Table II shows the procedure in one of the experiments. 



Table II 

 Showing the rate at which a liter of a saturated solution of carbon monoxide 

 loses its toxic properties when exposed to the atmosphere in a 5X8 inch battery-jar. 

 Fislies used, Lepomis humilis. Two individuals placed in the jar each time. 



This fish was Lepomis cyonellus, which is a more hardy fish than L. humihs. 



