EFFECTS OF CHANGES IN THE DENSITY OF WATER UPON THE 



BLOOD OF FISHES. 



By G. G. SCOTT, 

 Department of Natural History, College of tlte City oj New York. 



The fact of variation in the number of human blood corpuscles in health 

 has been known for some time. Stierlin (1889) found variations in individual 

 men of i ,650,000 per cubic millimeter and somewhat larger variations in women. 

 Viault (1890) noted a marked increase in cases of residents at high altitudes. 

 Some physiologists suggest that the increased number is probably not due to 

 increased formation of red corpuscles but for one thing to evaporation of water 

 from the surface and loss of water from the blood. Yet others have found in 

 animals kept at low pressures that there occurs an increase in the number of 

 corpuscles. Hill (1906), reviewing this work, notes an increase in hemoglobin 

 and adds that it is a reaction on the part of the organism to compensate for low 

 oxygen pressure at high altitudes. Although the other side of this question, 

 the changes in the organism due to increased atmospheric pressure (caisson 

 disease) , has been investigated, the principal effect noted has been the presence 

 of air bubbles in the blood and other tissues, the fatal effect of which is well 

 known. The effect on the corpuscle count has not been noted. 



Dr. F. B. Sumner, director of the biological laboratories of the Bureau of 

 Fisheries at Woods Hole, Mass., who is interested in the cause of the death 

 of salt-water fishes in fresh water and whose investigations will be referred to, 

 suggested to the writer that he study the effect on the blood of fishes subjected 

 to changes of density in the water. Part of my time has been devoted to this 

 problem during the last three summers, with facilities and material afforded 

 by the Bureau of Fisheries. 



Sumner (1905) found that well-marked changes in weight resulted in certain 

 cases where the density of the water in which fishes were kept was increased or 

 decreased. He showed that although at first thought these changes might have 

 been expected on the assumption that we were dealing with an osmotic action 

 through semipermeable membranes, yet in other cases the changes did not follow 



B. B. F. 1908— Pt 2—30 I 145 



