The extensive acidification of fresh waters across Europe and North America during the 20th century led to extensive and widespread loss of species, which fundamentally changed the structure of resident communities. While attempts to limit or halt acidifying emissions have been successful in reversing the chemical consequences of acidification, any corresponding biological recovery has been patchy at best. The main aim of this thesis was to investigate potential ecological constraints on this biological recovery, ranging from interactions between individuals to ecosystem-level processes, using a model stream system that has been extensively studied for over 40 years. I used a combination of long-term survey data and experiments, both in the field and the laboratory, to provide evidence that the lag in recovery of acidified aquatic communities is due to intrinsic resistance to re-colonisation by acid-sensitive species, as a result of both direct and indirect interactions between predators and prey. However, there is also evidence of recovery at an ecosystem level, with rates of microbial decomposition increasing – a key link in the transfer of energy to higher trophic levels, which could be facilitating population increases in large-bodied predators. These findings increase our understanding of the drivers which govern the structure and function of ecological networks in response to an important climatic stressor. This is especially relevant given the rapid industrialisation of countries such as India and China, which could well soon experience anthropogenic acidification on a significant scale.