I trust that I will not be criticized for saying that, in general, the argument between Hensen and Haeckel has been resolved. Hensen (1887, 1895) believed that true spatial and temporal heterogeneity of the plankton could be resolved by careful measurement and analyses of variance. Haeckel (1893) thought these analyses might give more information than the data possessed. Our present understanding of spatial heterogeneity, and its analysis, does not lead us to disagree with either. Investigations carried out by a number of marine ecologists have shown large-scale spatial variations in phytoplankton. Amongst many others, this feature can be seen in studies off the northeast coast of England (Cushing, 1955), in the English Channel and Celtic Sea (Cushing, 1957; Pingree et al. 1976), off Scotland (Adams, Baird & Dunn, 1975,1976) and off the coast of the Netherlands (Gieskes & Kraay, 1975). Nevertheless the statistics of sampling heterogeneous distributions is far from simple, estimates of density for instance can be greatly influenced by the sample size (e.g. Greig-Smith, 1964; Cassie, 1963). Bainbridge (1957) summarized his review by saying that patches occurred on all scales from a few feet to as much as 30 or 40 miles by 120 or 180 miles. Our present understanding of the nature of turbulence in the sea would reduce his few feet to centimetres.