OSMOTIC PRESSURE OF SEA WATER. 267 



But of those kept in sea water of one half its normal concentra- 

 tion only three per cent, were dead at a time when all those in 

 the other two media had died, and seventy per cent, were kept 

 alive for four weeks, when the wounds were all healed and the 

 experiments discontinued. In these experiments therefore, no 

 deleterious effects obtain when the internal and external media 

 are approximately isotonic in spite of the injuries and free inter- 

 change between blood and aquarium water. In the hypotonic 

 and hypertonic solutions, however, distinct changes resulting in 

 death, take place. In the strong solutions (normal sea water) 

 microscopic examination showed that the blood corpuscles were 

 crenated. In fresh water the fish became greatly swollen indi- 

 cating the absorption of water. Whether laking or swelling of 

 the corpuscles takes place was not determined in this series of 

 experiments.* 



From these experiments we may conclude that in all proba- 

 bility the blood of Fnndulns does not suffer much if any change 

 in concentration when the fish is transferred from salt water into 

 fresh water or vice versa, provided the membranes are uninjured. 

 If these experiments admit of general application to migratory 

 toleosts they would indicate that these animals also are in some 

 way protected from changes in the osmotic pressure of the blood 

 and tissues and that the principal protective factor probably lies 

 in a lack of permeability of their membranes. We may further 

 conclude that in case of serious abrasion to the integument the 

 membranes become permeable and a change of osmotic pressure 

 of the blood results, a change which may induce the death of the 

 animal. The great mortality of the salmon after spawning in the 

 head waters of California's streams, is a well-known fact (Rut- 

 ter 23 ). Whether the generally battered condition of these fish at 

 the spawning season bears any relation to changes in the osmotic 

 pressure of the blood has not been investigated. It is not impossi- 

 ble that the actual cause of death lies in a decrease in the osmotic 

 pressure of the blood and that the injuries are responsible for 

 death only in so far as they permit the entrance of water and de- 



*The haematocrit would doubtless prove a valuable aid in making experiments of 

 this sort when it is impossible to obtain sufficient blood for freezing point <ietermina- 

 tions 



