50 



ANNALS NEW YORK ACADEMY OF SCIENCES 



were heated to boiling for about two minutes, after which the liquid was 

 allowed to cool, the flask was filled to the mark with distilled water and 

 the contents were shaken. The whole was then filtered and 50 c. c. of the 

 filtrate was used for an analysis of its chlorine by the Volhard method. 

 The amount of chlorine thus determined was multiplied by two, giving 

 the amount present in the original 5 c. c. of serum, and from this the 

 amount present in 1000 parts of serum was easily calculated. Table XIX 

 shows the results of the analysis of the chloride in five samples of serum 

 taken from seven fishes immersed in sea-water and in six samples of 

 serum taken from six fishes that had been transferred from sea-water to 

 fresh water for somewhat over an hour. 



Table XIX. — Chlorine content of the blood serum of dog-fishes in sea-ivater 

 and after immersion in fresh water 



Number 



Chlorine in serum 



from fishes taken 



from sea-water, 



in grams p. m. 



Number 



Chlorine in serum 



from fishes after 



immersion in 



fresh water, 



in grams p. m. 



1 



8.778 

 8.400 

 8.246 

 8.715 

 9.079 



1 



5.824 

 6.755 

 6.181 

 6.608 

 6.734 

 6.433 



2 



2 



3 



3 



4 



4 



5 



5 





6 





Average 



8.643 



Average 



6.422 



The average in the case of the first group is 8.643 p. m., while that for 

 the second group is 6.422 p. m., a difference of 25.7 per cent, representing 

 the loss of chlorine resulting from the immersion. 



The greater percentage of CI in the serum than in the blood is due to 

 the fact that practically all the chlorides of the blood are dissolved in the 

 serum. The significant feature of the two groups of analyses is that the 

 percentage loss in chlorine is approximately the same in the two cases. 

 The results warrant the conclusion that after immersion in fresh water 

 for about an hour, %. e., until near death, the blood contains about 25 per 

 cent less chlorine in solution than is the case with normal blood. This 

 means that the salt content of the blood is less than the urea and other 

 nitrogenous substances. If there had been no loss of salts by diffusion, 

 then there should have been a decrease of but 15 per cent in the salts at 

 the end of the period of immersion in fresh water. We are driven log- 

 ically to the conclusion that the excessive diminution in the salts of the 



