The results of the High Flux Isotope Reactor (HFIR) Surveillance Programme have attracted considerable interest amongst regulators and utilities because of their implications for, in particular, light water reactor (LWR) support structures operating at temperatures below 100°C. The cause of concern was the apparently higher rate of embrittlement encountered in this programme at near ambient temperatures and at an apparently modest, by reactor pressure vessel (RPV) standards, neutron fluence. This was originally interpreted as a neutron flux effect. Because of these concerns the authors have undertaken a comprehensive review of the scientific literature on radiation embrittlement of RPV and structural steels in order to assess the potential impact on the integrity of the LWR/boiling water reactor support structures. Two main issues have been identified in this review. The first of these, the effect of neutron spectrum, relates specifically to the possibly unique character of the HFIR environment. The second relates to the nature of the embrittlement processes at low temperatures, in particular the occurrence or otherwise of the copper effect, which is known to be important at normal RPV temperatures of 150°C and above. Wilth respect to the effect of the neutron spectrum, it is now widely accepted that the highly thermalised neutron environment in HFIR played an important part in promoting the apparently higher embrittlement of the HFIR materials. It has been proposed that the lower energy recoils produced by thermal neutron ( n − γ) reactions are about 10 times more efficient in creating displacement damage than the high energy neutrons (E > 1 MeV), which normally account for most of the radiation damage in RPV and core components. This conclusion is firmly based on theoretical studies of displacement cascades and can be inferred from detailed studies of irradiation effects such as radiation induced segregation in different radiation environments. A major issue concerning the relevance of the HFIR results to RPV support structure embrittlement therefore centres on the nature of the neutron spectrum. To enable this issue to be resolved it is recommended that an effort be made to obtain more comprehensive and reliable dosimetry for RPV supports, particularly in relation to the thermal neutron component of the neutron spectrum. The literature survey has also demonstrated the confused situation that exists concerning the possible role of Cu in low temperature irradiation embrittlement. The only direct evidence concerning the role of Cu appears to be that of Russian workers who observed pronounced embrittlement in both commercial steels and model Fe/Cu/Ti alloys at 50°C. In the latter work a direct link to Cu content was identified at levels consistent with the predictions of models proposed for higher temperature embrittlement. However, the results also suggest a possibly strong inhibiting effect of interstitial elements such that embrittlement was only observed in alloys containing scavenging additions such as V and Ti. One analysis of the HFIR surveillance data has identified the possible effect of Cu and this can be shown to be consistent with the recent Shippingport neutron shield tank results. Similar analysis of other data from the literature appears to demonstrate a pattern of behaviour consistent with an undiminished copper embrittlement at low temperatures and reinforces the original concerns about RPV support structures raised by the HFIR surveillance results. The concerns are further reinforced by the fact that the support structures of many plant are manufactured from steels containing high (0·2-0·4% wt) Cu levels. It is concluded that there is a need to resolve the issues of neutron spectrum and low temperature Cu embrittlement effects, so that the structural integrity of RPV support structures in many operating nuclear plants may be better assessed.