1. No evidence was found of the presence of furunculosis in trout farms. Over 100 trout from four farms in S. England were examined with entirely negative results.

2. Of eighteen live trout taken from the Kennet, three were carrying the B. salmonicida in their blood. All three were in poor condition but alive and active, and two of them were so-called “black fish”.

3. Bacteriological analyses of Kennet water at eleven selected points were carried out in the month of August. They revealed no serious pollution nor any definite relationship between high bacterial content and the disease.

4. Laboratory experiments to test pathogenicity of B. salmonicida appear to show:

(a) That the bacillus is lethal to goldfish when injected into the muscle but not necessarily so when rubbed into a skin wound.

(b) That in the case of gobies infection from water polluted with pure broth cultures of B. salmonicida is possible and may cause death in about fifteen days after the introduction of the culture.

(c) That fish (gobies) which had survived the last experiment for a period of six weeks were capable of infecting healthy “contacts”, when kept with them in fresh running water.

5. A few laboratory experiments on the behaviour of B. salmonicida in sewage appeared to indicate that the bacillus persists longer in tap and distilled water than in crude sewage and sewage effluent.

6. Specimens of fish (salmon and trout) infected with furunculosis were also received during the summer from the Rivers Lledr (Wales), Lyon (Scotland) and a private hatchery on the Test at Whitchurch. The conditions in the two latter cases suggested that the outbreak was mainly due to carriers.

7. Measures suggested in the case of infected streams are: avoidance of over-stocking, bacteriological scrutiny of source of fresh stocks, removal of sick and ill-conditioned fish, and sewage control.