There is some controversy about the strength of the stock of mackerel in the Celtic Sea and off the coasts of Cornwall and Ireland, and it is difficult to find out if the numbers of mackerel in this area have fluctuated in the past (Johnson, 1977; Lockwood & Johnson, 1976; Lockwood, 1978; Coombs, Pipe & Mitchell, 1977,1979, 1980). The earliest complete survey of the south-west spawning grounds was made in 1937–9 (Steven & Corbin, 1939; Corbin, 1947) and was principally designed to show the seasonal and geographical pattern, not provide a basis for stock evaluation. Nevertheless, this survey is the only information we have about the breeding of mackerel in the days before the war when fishing was performed with drift-nets and lines, a period when we can assume exploitation was very much less intense than today when trawls and purse-seines are used. Attempts to use the 1937–9 data for estimation of the pre-war stock have been regarded as giving unsatisfactory results (Walsh, 1976). The data from the 1937–9 surveys were defective because we did not know the influence of the depth of fishing of the nets in relation to the vertical distribution of the eggs, nor the efficiency and catching capacity of the nets for sampling eggs and post-larvae.