The effects of towed fishing gear on benthic fauna are under intense scrutiny and evidence is growing that trawling may significantly affect benthic communities in the North Sea. Most studies explore the current fauna or compare today's situation with that of 2 or 3 decades ago, when North Sea-wide information on benthos and fishing became available. However, in the North Sea, extensive mechanised trawling began more than a century ago. This study compared historical and recent records in order to explore potential long-term links between changes in the epibenthos and fishing. Based on reconstructed species lists from museum specimens, we compared epibenthos data from 1902 to 1912 with those from 1982 to 1985 and 2000. We analysed changes in average taxonomic distinctness (AvTD), a biodiversity indicator, and changes in biogeographical species distributions. Landings data were collated for round- and flatfish caught in the northern, central and southern North Sea from 1906 to 2000 as proxies for total otter and beam trawl effort, respectively. These indicate that the southern and much of the central North Sea were fished intensively throughout the 20th century, whilst the northern North Sea was less exploited, especially in earlier decades; exploitation intensified markedly from the 1960s onwards. For epibenthos, the mean AvTD decreased significantly from the 1980s to 2000, when it was below expected values in 4 ICES rectangles, 3 of these located in heavily trawled areas. Biogeographical changes from the beginning to the end of the century occurred in 27 of 48 taxa. In 14 taxa, spatial presence was reduced by 50% or more, most notably in the southern and central North Sea; often these were long-lived, slow-growing species with vulnerable shells or tests. By contrast, 12 taxa doubled their spatial presence throughout the North Sea. Most biogeographical changes had happened by the 1980s. Given that other important environmental changes, including eutrophication and climate change, have gained importance mainly from the 1980s onwards, we have concluded that the changes in epibenthos observed since the beginning of the 20th century have resulted primarily from intensified fisheries.