ABSTRACT1. The coastal waters surrounding Britain and Ireland became warmer during the 20th century and, according to the UK Climate Impact Programme 2002 scenarios of change and other sources, average annual seawater temperatures may rise a further 28C or more by the 2050s. This warming is part of a global rise in sea-and air-surface temperatures that will cause changes in the distribution and abundance of species.2. Initially, there will not be a wholesale movement northwards of southern species or retreat northwards of northern species, because many additional factors will influence the responses of the different organisms. Such factors include the hydrodynamic characteristics of water masses, the presence of hydrographical and geographical barriers to spread and the life history characteristics (reproductive mode, dispersal capability and longevity) of species. Survey data over the past century show how organisms react to changes of the order of 0.58C, and in the last two decades, when sea temperatures have risen by as much as 18C, there have been significant local changes in the distribution of intertidal organisms. These past changes provide a clue to more extensive changes expected in the future if global warming develops as predicted.3. Where species affected by climate change are dominant or key structural or functional species in biotopes, there may be a change in the extent and distribution of those biotopes. Some, dominated by predominantly northern species such as the horse mussel Modiolus modiolus, may decline and reduce their value as rich habitats for marine life. Others, characterized by southern species, for example the sea fan Eunicella verrucosa and the alcyonacean Alcyonium glomeratum, may increase in extent.4. Using information on the life history characteristics of species, their present distribution and other factors, a key supported by a decision tree has been constructed to identify 'types' of organism according to their likely response to temperature rise. Conspicuous and easily identified rocky substratum species are good candidates to track change. Using the key, many species are shown as likely to increase their range northwards significantly. In contrast, fewer will decline in abundance and extent in the north. If, as anticipated, global warming continues, then species with distributions already accurately mapped, or being mapped at present, will provide baseline data to test forecasts.