In this paper we revise the application of the UK37' and UK37 indices as sea surface temperature (SST) proxies in the Nordic Seas. In the summer of 1999 and 2000 we obtained samples of filtered particulate organic matter (POM) from surface waters (~6 m depth) of the Nordic Seas. A number of samples were collected from polar waters with up to 80% of sea ice cover. Alkenones were detected in all of the major water masses of the Nordic seas, across a spectrum of SST values from -0.5 to 13¿C and SSS values from 29.6 to 35.6 (psu). Concentrations of alkenones were similar in magnitude to those reported previously for the North Atlantic and the Southern Ocean. Values of UK37' from the new Nordic Seas POM data show no correlation with SST below 8¿C. In contrast, below this temperature a linear correlation exists between UK37 and regional SST, supporting previous suggestions that, overall, UK37 may be a more appropriate SST index for the region. It must be noted however that UK37 is calculated using the tetraunsaturated alkenone, and the dominant control on this compound is not yet fully understood. The new data highlight major differences between distributions of UK37 and UK37' in the water column POM and surficial sediments of the Nordic Seas. We also examine the geographical dependence of the UK37' versus SST relationship in the region's surficial sediments. Some areas are associated with unreliable SST estimates, whereas in others the UK37' versus SST relationship falls within the range of a global core top calibration. It is suggested that the breakdown of the UK37'-SST relationship in some regions is due to ecological and sedimentological factors. The latter relates to the extension of the habitats of alkenone producers in polar waters. The former to the resuspension of sediments and laterally transported alkenone inputs. This suggests that accurate alkenone derived SST estimates in the Nordic seas are geographically constrained.