SUMMARY: Bacterial counts made, over a period of 27 months, on skin, gut and gill samples of freshly caught skate and lemon sole, using sea water-based and tap waterbased media in parallel, revealed a seasonal variation in the size of the bacterial populations on the fish throughout the year. Evidence is presented for the view that the occurrence of maximum bacterial populations on fish is correlated with plankton outbursts. Sea water-based medium was generally superior to a tap water-based medium for isolating bacteria from fish; the importance of this as evidence of a specific marine bacterial flora is discussed.Bacteria occur in appreciable numbers on the skin and gills and in the intestinal contents of fish (Harrison, Perry & Smith 1926). However, it is generally accepted that the internal organs and the muscle tissues of (healthy) live or freshly caught fish are sterile (Anderson, 1907;Proctor & Nickerson, 1935). The spoilage of fish as a foodstuff is due to the activities of bacteria normally present on the living fish which, after the animal's death, penetrate into the muscle tissues and decompose them. Consequently, the bacterial flora of different types of fish has been the subject of several investigations, the results of which were summarized by Shewan (1949). Most of these studies were confined to commercially important species of 'round' fish' such as cod, haddock and herring and have, in the main, consisted of qualitative analyses of the bacterial types isolated from them; flatfish such as skate, lemon sole, turbot, etc., have been little studied. However, the results of an investigation by Reay & Shewan (1949) into the numbers of bacteria on the skin of haddock a t various times of the year indicated that there is probably a seasonal variation in the quantitative load of bacteria on this fish. This conclusion is not entirely unexpected since seasonal variations in the numbers of bacteria in sea water have been observed by workers in different parts of the world (Lloyd, 1930;ZoBell & McEwan, 1935;Wood, 1953) and sea water is undoubtedly the source of many of the bacteria found on fish.Workers in the field of marine bacteriology have used a variety of media in investigations of the bacterial flora of sea water and of fish and this has made comparison of their results peculiarly difficult. Since it has been claimed that many bacteria of marine origin will not grow in the absence of sea water (ZoBell, 1946) the comparison of viable counts made in tap-water-containing and sea water-containing media is particularly difficult.The investigation described below was carried out to determine whether seasonal variations occur in the bacterial flora of skate (Raja spp.) and lemon sole (Pleuronectes mimocephalus) and to determine, so far as possible, what proportion of the bacteria on the fish require sea water in the media for their G. Microb. xv