THE DENSITY OF WATER AND THE BLOOD OF FISHES. II47 



two lots were made on the same slide at the same time and stained simultaneously 

 with Wright's blood stain, were then dried, and mounted in balsam at the same 

 time. A careful examination with the twelfth oil immersion showed no measur- 

 able differences in the corpuscles themselves. This was true in other similar 

 tests. In other words, whatever changes these different solutions have on blood 

 it may be said that these changes do not reach the corpuscles themselves. 



A series of experiments was made to ascertain whether there was any change 

 in the corpuscle count due to changes in water density. In the case of moun- 

 tain sickness it has been shown that at high altitudes there is a physiological 

 reaction on the part of the organism to the effect that per cubic millimeter 

 there are more corpuscles manufactured to take up more oxygen, and thus supply 

 the organism with its normal amount. If fishes are placed in fresh water, is 

 the blood diluted? This would be shown by a decrease in the number of cor- 

 puscles. To test this, hemocytometer tests were made upon the blood of 

 normal Fundulus heteroclitus. The average blood count of eight was found to 

 be 2,749,000. Four fish placed in distilled water for iX to 2^ hours were exam- 

 ined in the same way and the average blood count was found to be 2,171,000. 

 In two experiments with distilled water there was at first a decrease in the num- 

 ber of corpuscles counted, then a gradual increase, which increase later passed up 

 to and above normal. The last-named experiment indicates that there is at first 

 a reduction, due to influx of water. After the reduction in number of corpuscles 

 there is an increase, which indicates a reaction on the part of the organism to get 

 back to the normal condition. Experiments in which Fundulus heteroclitus was 

 subjected to fresh water show, in the main, that the gain in weight which is at 

 first noticeable disappears after a few days and the weight of the organism falls 

 below normal and the decrease continues. This is undoubtedly due to starva- 

 tion. It may be objected that these fishes can not live very many days in fresh 

 water. In distilled water, no, but in ordinary tap water, yes. On July 10, 

 1908, I placed ten Fundulus heteroclitus in a large rectangular dish containing 

 about 20 liters of fresh tap water. In the bottom was placed some gravel from 

 the shore near by, the gravel having been thoroughly washed. Most of the 

 fishes died during the next three or four weeks, but one fish was alive and appar- 

 ently in good condition on September 8, 1908, after sixty days. It should be 

 said that the tank in which this specimen was kept was washed out thoroughly 

 once a week and the fish fed. A rough test of the water with silver nitrate and 

 nitric acid was made toward the end of the period, and a slightly greater amount 

 of chlorides was indicated than was present in the fresh tap water. This agrees 

 with Sumner's statement that a slight addition of salt water to fresh water was 

 sufficient to keep the fishes alive, and therefore the death of fishes in dilute 

 solutions was probably not due to lowering of osmotic pressures. To show the 



