336 and 408 hr, respectively), direct comparisons 

 could not be made with those that died after expo- 

 sure to 120%. However, visible gas emboli were not 

 found in the heart, gills, or major vessels of any 

 fish exposed to 115% and 110% supersaturated 

 water. Emboli were absent in the controls. 



At 120% and 115% supersaturation, gas accumu- 

 lations in the interray membranous tissue (emphy- 

 sema) and within the venules adjacent to the carti- 

 lagenous rays appeared more often in the tail than 

 in similar locations in the anal, dorsal, and paired 

 fins. Gross emphysema did not develop in the fins 

 of fish exposed for over 408 hr (17 days) to 110% 

 supersaturation. Subcutaneous emphysema along 

 the opercula was found in a significant number of 

 fish from all three levels and apparently is associ- 

 ated with longer periods of exposure than that 

 necessary to produce emphysema of the fins. Fin 

 and tail emphysema is apparently a more acute 

 lesion and is associated with high levels of 

 supersaturation. 



Exophthalmia, a lesion frequently associated 

 with GBD, occurred in six fish after 93 hr exposure 

 to 115% supersaturation. Exophthalmia associated 

 with GBD is caused by the accumulation of gas 

 within the fatty tissues of the periorbital space, 

 resulting in abnormal protrusion of the eye. Other 

 ocular lesions including blood in the anterior cham- 

 ber were seen only at 110% supersaturation after 

 long periods of exposure. 



The entire group of fish were fed immediately 

 prior to the experiment, but not during the experi- 

 ment. This afforded an opportunity to observe the 

 effects of supersaturation stress on intestinal motil- 

 ity. According to Klontz,* food should pass through 

 the stomach and intestinal tract of trout in 8 to 

 10 hr depending on the water temperature. Food 

 consisting of Oregon Moist Pellets® was retained in 

 the stomach for up to 54 hr at 120% and 93 hr at 

 115%. Food observed in the stomach appeared to 

 have undergone little change due to the digestive 

 process. A similar delay in the emptying of the 

 intestine was also observed. Controls and fish held 

 at 110% did not have undigested food in the stomach 

 when necropsied. The intestines of many fish also 

 contained an excess of thick bile-stained mucus. 



DISCUSSION 



The cause of death of fish by acute GBD has 

 been reported by many investigators as hemostasis. 

 This is caused by blockage of blood flow through the 

 heart and gills from the accumulation of gas emboli 

 within the capillaries of the gills and results in 

 anoxia and death. This study supports this conclu- 

 sion. All fish that died at 120% supersaturation level 

 had large accumulations of gas emboli in the heart, 

 ventral aorta, and gills. However, with the exception 



of two fish in terminal convulsions and one fish that 

 appeared normal, fish exposed to 120% and sampled 

 prior to death did not have visible accumulations of 

 gas within major vessels, heart or gills. Visible 

 emboli were not seen in any fish at 115% or 110% 

 levels. Based on these observations, the formation 

 and/or migration of macroscopic emboli within the 

 blood vascular system appears to be an acute 

 terminal or near terminal phenomenon. Otherwise, 

 gradual accumulation of gas emboli would be visible 

 in the gill filaments of fish still alive under super- 

 saturated conditions. No such lesions were found in 

 this experiment. 



The data indicate that initiation of bubble 

 growth from pre-existing nuclei and/or bubble dis- 

 lodgement from the periphery of the body may 

 occur under certain physiological conditions that 

 can trigger a "cascading bubble effect" of emboli 

 into the gill capillaries with resultant hemostasis. 

 The physiological conditions involved in the initi- 

 ation of this apparently irreversible cascading effect 

 are unknown for fish. Many studies have been con- 

 ducted on mammalian systems in connection with 

 decompression sickness or the "bends." Similar 

 studies should be done to define the physiological 

 processes occurring in fish dying of GBD and find- 

 ings should be considered in the determination of 

 acceptable levels of supersaturation. 



For example, supersaturation levels as high as 

 120% may be tolerated by some salmonid fish under 

 certain circumstances, whereas levels as low as 

 115% may be acutely lethal under a different set of 

 conditions. Circumstances that force fish to swim 

 rapidly while in sublethal supersaturated water such 

 as excessive water flow, flight from predators, etc., 

 may initiate the "cascading bubble effect" through 

 increased muscular activity. Muscular activity is 

 known to contribute greatly to the development 

 and release of visible gas emboli in cats during 

 decompression experiments (Harvey, 1944b). The 

 effect of muscular activity was especially important 

 at the lower levels of pressure change. The explana- 

 tion given by Harvey is that muscular contraction 

 favors bubble formation by further reducing the 

 hydrostatic pressure causing the formation of large 

 vapor cavities into which gas can diffuse. If the 

 vapor cavity persists long enough, a visible gas 

 bubble can be formed in liquids with low gas ten- 

 sion. The bubble may gradually dissolve, but before 

 this happens, it might move into the general circu- 

 lation. Gas accumulations are known to form in 

 muscle even at relatively low levels of supersatura- 

 tion or decompression (D'Aoust, 1974; Stroud etal., 

 1975). 



"Personal communication, George W. Klontz, Professor of 

 Fisheries, University of Idaho, Moscow, Idaho. 



Pathogenesis of Gas Bubble Disease 69 



