the fry were divided into two groups, one -half 

 to act as controls and the other half used to de- 

 termine if the disease is transmitted by the diet. 

 All fingerlings received a diet of beef and hog 

 liver until mid-March after which time the con- 

 trol groups continued to receive an all -meat 

 diet without salmon products, while the exper- 

 imental groups were fed a mixture of 50 percent 

 salmon products and 50 percent hog and beef 

 products. It was thought fairly warm water tem- 

 peratures are required for the utilization of the 

 Cortland diet so it was withheld until the latter 

 part of July, the beef and hog liver diet being 

 fed until this time . 



Increased weight and size of fish neces- 

 sitated a reduction in the number of fish per 

 trough by early July. At this time a group of 

 surplus experimental fingerlings was planted 

 and 80 troughs of experimental fish were moved 

 outside into 20 small rearing ponds (4 troughs 

 per pond). TTie fish remaining in the hatchery 

 were distributed throughout the hatchery. As in 

 all other moves each pond and trough on the ex- 

 perimental diet was accompanied by the identical 

 movement of its control. Nineteen of the 20 

 ponds of both groups were screened with 2 -inch 

 poultry netting to prevent birds from having ac - 

 cess to the fish, as they previously had been 

 observed dropping moribund and dead fish in un- 

 infected ponds . The remaining pond in each 

 group was covered with a fine-mesh plastic cloth 

 to keep out insects . 



Winthrop hatchery 



After the eggs spawned at the Winthrop 

 station and those received from the Leavenworth 

 station had hatched, the young fish were divided 

 into 56 troughs: 28 on one side of the hatchery 

 and their 28 controls exactly opposite them on 

 the other side. The experimental diet of 50 per- 

 cent meat products and 50 percent mixed salmon 

 viscera was begun the last week in February, 

 1954. These viscera were from the same lot 

 that had been fed the previous year at the Win- 

 throp, Entiat and Leavenworth hatcheries during 

 which time all three hatcheries had been affected 

 by the epizootic . On April 20, the fingerlings 

 were moved outside into 14 small Foster-Lucas 

 ponds -- the control fish on one side of the hatch- 

 ery and the viscera-fed fish on the other side. 

 On June 16, 1954, one pond of viscera -fed fish 



was planted, on June 18 three ponds of control 

 fish were planted, and on July 7, two more ponds 

 of healthy, viscera -fed fish were planted. There 

 remained two ponds of fish from the Leavenworth 

 stock and two ponds from the Winthrop stock on 

 the viscera diet and the group of four ponds of 

 control fish. These fish were retained for over- 

 winter holding. 



General laboratory procedure 



In conjunction with the large-scale hatch- 

 ery experiments, additional data were obtained 

 in the laboratory. Of particular importance were 

 the factors influencing the spread of the disease 

 and the fact that the virus perpetuated itself 

 despite the extremely high mortality of infected 

 fingerlings . 



Whenever a higher than average daily mor- 

 tality occurred in any troughs or ponds at the 

 experimental hatcheries, the fish were examined 

 immediately in an effort to determine the cause 

 of death. If examination failed to disclose a path- 

 ogenic agent, moribund fish and recent mortalities 

 were brought to the Seattle laboratory for testing. 

 They were combined for one minute in a blender 

 and centrifuged for one minute . Dlilutions were 

 made of the supernate to give a mixture of 1 part 

 fish in 99 parts nutrient broth; 0.05 to 0.10 ml. 

 of this suspension was inoculated into each of ten 

 disease-free fingerlings. The controls consisted 

 of an equal number of fish inoculated with sterile 

 nutrient broth. If mortalities occurred in the sus- 

 pension-inoculated group, the same procedure 

 was repeated with the remaining fish and the 

 diluted homogenate was passed through a sterile, 

 7-pound Mandler filter. The filtrate was inoculated 

 into a minimum of ten fingerlings with appropriate 

 controls . Material from ensuing mortalities was 

 serially transferred to exclude, by total dilution, 

 any toxic effect of the original filtrate. The healthy 

 fish in these experiments were obtained from the 

 control troughs at the Leavenworth hatchery. 



To check the possibility of the virus gaining 

 entrance into the hatcheries via an air-borne or 

 aquatic vector, plankton and bacterial counts were 

 made periodically throughout the year at all ex- 

 perimental hatcheries and, in addition, insect and 

 small mammal collections were made at the 

 Leavenworth hatchery . Plankton samples were 

 obtained by placing a plankton net under one of the 



