Some anthropogenic noise is now considered pollution. Evidence is building that noise from human activities such as transportation, construction and exploration can impact behaviour and physiology in a broad range of taxa . However, relatively little research has considered effects on development or directly assessed fitness consequences, particularly with respect to repeated or chronic noise. All fishes and many invertebrates use underwater sound for processes such as orientation and communication, and are thus vulnerable to anthropogenic noise pollution. Fishes and invertebrates detect the particle motion component of sound; this component has. been neglected, but must be understood alongside acoustic pressure if the potential impacts of noise are to be fully understood. As the first part of my thesis (chapter two), I provide a computer program which allows users to determine under what circumstances they should measure particle motion; I explain how these measurements can be made and provide a program for analysing this type of data. The main part of my thesis comprises experiments investigating the impact of repeated exposure to traffic noise, the most common anthropogenic source of noise in the marine environment, on fishes and invertebrates during development. In all three chapters involving experiments on fish in tanks and in the field, I found that a variety of behaviours were impacted by traffic noise playback. I also found that predictable noise can lead to different impacts on development from unpredictable noise, but that some species of fish may be able to habituate to traffic noise, while others suffer lower survival. Further, I found that the development and survival of sea hares (Stylocheilus striatus) can be negatively impacted by traffic-noise playback. Fishes 'and invertebrates provide a vital food source to millions of people and form crucial links in many food webs; studying their behaviour, development and fitness can give us an insight into population and community level impacts of noise that are relevant to species survival and evolution. Developing some of the novel ideas and techniques discussed in this thesis will enable us to advance this vital area of research.