A series of paired-pond trials were completed during the winters of 2005 to 2008 to examine the effect of introducing artificial fish refuges on Great Cormorant Phalacrocorax carbo foraging behaviour, prey capture rates and fish losses in trial ponds, and to help refine their potential use as a tool in addressing cormorant-fisheries conflicts. Investigations comparing refuge volumes within a year provided clear evidence that larger refuges offered fish better protection than smaller ones. When compared across years, there was no clear relationship between refuge volume and prey consumption rate by Cormorants. There was, however, a very large difference in the consumption rate when ponds with no refuge were compared with any of the refuge volumes tested, suggesting that, in the absence of any other cover, the addition of even very small fish refuges can markedly reduce the quantity of fish eaten by Cormorants. Further trials indicated that introducing refuges as a single larger structure was more effective at reducing fish losses than an equivalent volume of refuge arranged as a number of smaller, widely-spaced, discrete units (>50% reduction in prey consumed per Cormorant visit). Trials further confirmed that the beneficial effects of fish refuges were not compromised by the presence of marginal habitat; indeed, prey consumption rates were >30% lower in the pond with additional simulated reed beds. The impact of refuge presence on fish loss rates varied between species, with Roach Rutilus rutilus benefitting in particular. The results are discussed in the wider context of interactions between Cormorants and their prey and with regard to the potential use of fish refuges as a management tool.