A new medium, designated Ribose Ornithine Deoxycholate agar (ROD), was developed and used in field trials at two fish farms where fish were known to have ERM. The medium indicated that Yersinia ruckeri could occur in faeces four to six weeks before appearing in the kidney. Fxurther epidemiological studies dealing with fish condition, performance, disease signs and water isolation are presented. These factors are discussed in relation to husbandry and management practices. The minimum uihibitory concentration (MIC) values for 124 strains of F. ruckeri were detemuned. The effect of oxolinic acid, oxytetracycline and a potentiated sulphonamide, on growth patterns over a 72 h period, was also determined. Results showed the bactericidal or bacteriostatic nature of each antimicrobial agent. It was possible to increase the MIC for oxoUnic acid, oxytetracycline and a potentiated sulphonamide usmg an in vitro technique. Attempts to decrease resistance to oxolinic acid were unsuccessful. However, it was possible to reduce the MIC's for o^etracycline and a potentiated sulphonamide. The recovery of Y. ruckeri, after artificial challenge, was less from the faecal and kidney material of a vaccinated group of fish compared with a non-vaccinated group. An E L I S A technique indicated that although there was no detectable serum antibody there was a local mucosal response i n vaccinated fish. Uptake of F. ruckeri antigen was demonstrated in vaccinated and non-vaccinated fish using an immunoperoxidase technique. However, only vaccinated fish appeared to take up the antigen by an active process. A survey of salmonid farms showed that there was a tendency for those sites where ERM had been diagnosed, irrespective of vaccination, to be larger table farms with production ranging up to over 200 tonnes p:a. Certain environmental and husbandry factors were reported as coinciding with the appearance of F. ruckeri. A slight tendency to consider that vaccines had failed i n some way was also indicated. This and the use of antimicrobial agents are discussed.

A group of 42 strains of Yersinia ruckeri were characterized and found to show homogenous biochemical profiles. The optimal conditions for in vitro culture of a representative strain of Y. ruckeri were established. The virulence was increased by passage through fish. A standard method for infection of rainbow trout (Oncorhynchus mykiss) with Y. ruckeri was devised. As a result cultures of Y. ruckeri grown on full strength of BHIA for 5 h at 30°C in static conditions were found to be most suitable for infection by both intraperitoneal injection and immersion. Virulence of stored cultures was maintained for over 1 year at -20 and/or -70° C and virulence was not lost after up to six consecutive subcultures. The effect on mortality of susceptibility of various fish stocks, water temperature and stocking density is described. The size of fish (between 5 and 60 g) had no effect on virulence. The injection and immersion protocols were used as in vivo models for the study of the effects of dietary vitamin E, with both laboratory prepared diets and commercially available diets. Comparative results of haematological and biochemical parameters, histology, mortality patterns, serum antibody levels and bacterial recovery from the four dietary groups of fish are presented. A comparative in vitro and in vivo study of intra and inter strain vanations among representatives of the serogroup 1 of Y. ruckeri was performed. Under the conditions used, some strains were virulent and other avirulent. Results of Western blotting and SDS-PAGE showed the presence of a heat sensitive factor (HSF), present only in the virulent strains of Y. ruckeri. To date this is the first putative virulence factor described for Y. ruckeri. Further work was undertaken in an attempt to characterize the HSF. It appears to be located in the periplasmic space and to have complex lipid and proteinaceous components. HSF is easily degraded, and is immunogenic and confers some protection when injected into rainbow trout. Selective media and detection methods for the HSF, and their potential use in laboratory and field studies are described.