AbstractA survey of 1,189 lakes was conducted in the fall of 1987 as part of the Finnish Research Project on Acidification (HAPRO). Most of the lakes for the survey(n=987) were randomly selected by a two‐stage cluster sampling from two separate subregions, together covering the whole country. Lakes of sizes 0.01‐10 km2were included in the statistical sampling of the southern and central regions (Subregion 1). In northern Finland (Subregion 2) only lakes of sizes 0.1‐10 km2were included. The statistical sampling “frame” in each subregion consisted of the number of lakes in the grid net formed from a series of topographic maps with a scale of 1:50,000 (20X30 km2). The total number of maps was 472 in Subregion 1 and 175 in Subregion 2. In the first sampling stage, 100 maps were selected from Subregion 1 and 25 maps from Subregion 2, using the “probability proportional to size (pps)” method, selection of the maps being weighted by lake density. In the second sampling stage, all of the lakes with a size within the determined size range were numbered and eight lakes chosen by systematic random sampling from each previously selected map. A further 202 mainly head water lakes examined in previous acidification studies were also included. Water samples were taken from the surface layer (1m) of the lakes during fall overturn in 1987. An extensive analytical program was carried out on each lake‐water sample. Major watershed characteristics have been determined from topographic and soil maps. The statistically based survey design allows estimates to be made of the proportion of lakes fulfilling any predetermined criteria and the corresponding variances. The number of acidic (acid neutralizing capacity (ANC) ⩽ 0 μeq L‐1) lakes in Finland was e