changes in plasma glucose are shown in Fig. 6 and 

 verify that a physiologically significant stress was 

 imposed on the fish. Another difference between 

 the hatchery and wild fish was that the blood glu- 

 cose in the wild fish returned to normal after 3 hr 

 and stabilized there, while the blood glucose for the 

 hatchery trout remained unstabilized for the dura- 

 tion of the sampling period (5 hr). 



10 30 



180 



300 



TIME (MINUTESI 



FIG. 6 Effect of stress and repeated sampling on blood glucose 

 levels in a hatchery trout (top) and a wild trout (bottom); 



stressed group; control group. Each point represents 



the mean ±s.e. Asterisk (*) indicates significant difference from 

 initial level (P = 0.05). 



DISCUSSION 



Our results clearly demonstrate the presence of 

 increased blood clotting rates in salmonids after 

 periods of stress. The presence of increased num- 

 bers of thrombocytes at the same time was also 

 shown, but whether or not they cause the change 

 in clotting function is not clear. However, their 

 greater availability would enhance the ability of 

 the clotting system to produce the initial "plug" 

 of cells at a wound site to minimize any blood loss. 

 The increases in blood glucose during the experi- 

 ments showed that the degrees of stress imposed on 

 the fish were physiologically significant. The 

 changes in blood glucose may also indicate that 

 hatchery fish could be less resistant to stress than 

 wild fish. 



The significance of the blood clotting system in 

 relation to the gas bubble problems of fish is still 

 speculative and largely extrapolated from what is 

 known about clotting problems in people. It is 

 highly probable that the major immediate effect of 

 bubbles arising from external supersaturation is to 

 block blood vessels. Depending on the importance 



of the tissues supplied by the vessel and the dura- 

 tion of the bubble, the fish may die quickly, at the 

 one extreme, or not die at all, at the other extreme. 

 Once a blood vessel is blocked by a bubble, it is 

 also highly probable that a clot will form there. 

 The likelihood of clot formation and having the clot 

 spread further once it has begun would be greatly 

 increased, we believe, by the presence of the clotting 

 enhancement response induced by the shock of 

 bubble formation. The problem caused by clotting 

 is further aggravated by the slow rate of blood flow 

 found in the white muscle of fish. Once a clot is 

 formed, the severity of the resulting damage to 

 tissue due to intravascular coagulation increases. 

 Such clots might take a week or more to be removed. 

 Once the blood vessel is blocked long enough, the 

 tissue downstream from the blockage dies, with the 

 degree of importance depending again on which 

 tissue is involved. We believe it is significant that 

 many of the juvenile coho which were used in our 

 decompression studies and which survived the 

 decompression with few or no symptoms of gas 

 bubble disease died a week later with severe tail 

 rot. 



Finally our results suggest that the mechanisms 

 involved in the pathology and mortality associated 

 with formation of bubbles in fish chronically 

 exposed to low-level supersaturation are more com- 

 plex than can be simply explained by simple bubble 

 formation. The role of sublethal changes by the fish 

 in response to supersaturation may eventually be 

 shown to be as important to the survival of Columbia 

 River salmon as the lethal effects which have been 

 the subject of most recent studies. 



ACKNOWLEDGMENTS 



Contribution number 408, College of Fisheries, 

 Seattle, Washington. Supported by NIH Grant 

 #HL 16254-02. 



REFERENCES 



Meekin, T. K. and B. K. Turner. 1974. Tolerance of salmonid 

 eggs, juveniles, and squawfish to supersaturated nitrogen. 

 In: Nitrogen Supersaturation Investigations in the Mid- 

 Columbia River, Wash. Dept. Fish. Tech. Rpt. 12, pp. 78-126. 



Smith, L. S. and G. R. Bell. 1964. A technique for prolonged 

 blood sampling in free-swimming salmon. ). Fish. Res. Bd., 

 Can. 21(4):T\"\-r\7. 



Sovio, A. and K. Nyholm. 1973. Notes on hematocrit deter- 

 minations on Rainbow Trout, Salmo gairdneri. Aquaculture 2: 

 31-35. 



Wedemeyer, G. 1972. Some physiological consequences of 

 handling stress in the juvenile coho salmon (Oncorhynchus 

 kisutch) and steelhead trout (Salmo gairdneri). ). Fish. Res. 

 Bd. Can. 29.1780-1783. 



Blood Clotting Mechanisms 95 



