418 RELATIVE TOXICITY OF HALIDES 



nite percentages of sodium chloride were replaced by the correspond- 

 ing molecular amounts of the other anions. The actual molecular 

 concentration of halide was therefore always m/7.8. With the degree 

 of dilution employed, differences in degree of ionization can be dis- 

 regarded. Table III shows the effects of such mixtures, for each of 

 which the molecular percentage of the anion studied is indicated as 

 "NaX." 



The chemicals used were of the following degree of purit}^: sodium 

 chloride, sodium hydrogen carbonate, potassium chloride, and calcium 

 chloride, Kahlbaum, "zur Analyse" (pre-war manufacture); sodium 

 fluoride, glucose, Merck's pure; sodium bromide, sodium iodide, 

 sodium chlorate, sodium iodate, and sodium nitrate, Merck's U.S. P. 



In all cases the animals were weighed before the experiment. As- 

 suming that the weights of the tissues used were proportional to total 

 body weight, there was no proportionality between weight of tissue 

 and duration of life. The extreme weight limits in all experiments 

 were 30 and 67 gm. In each experiment the extreme difference was 

 much smaller. Almost all the animals were male. 



DISCUSSION. 



Our results show a fair degree of constancy, as exhibited both by 

 the figures for the same conditions in different experimental series 

 (the experiments in any particular series were commenced at approxi- 

 mately the same time and therefore those in one series are somewhat 

 more accurately comparable) and the extreme variations obtained 

 under any one set of conditions. The actual difi'erences observed in 

 the three different experiments with chloride-Locke solution at 5°C. 

 are probably in part referable to different temperature fluctuations 

 within the slight limits mentioned (5 ± 1°C). The maximum survival 

 time of frog muscle in chloride-Locke solution was 230.7 hours; of 

 frog nerve, 228.5 hours; and of frog heart muscle, 58.4 hours. These 

 figures appear to be somewhat greater than any previously recorded, 

 the difference being almost certainly due to temperature. 



The period of survival decreases with increase of temperature in 

 all solutions. The results for frog muscle are shown in Fig. 1. Those 

 for heart and nerve give similar curves. (The mean temperature for 



