Summary 1.A fundamental goal of ecological network research is to understand how the complexity observed in nature can persist and how this affects ecosystem functioning. This is essential for us to be able to predict, and eventually mitigate, the consequences of increasing environmental perturbations such as habitat loss, climate change, and invasions of exotic species. 2. Ecological networks can be subdivided into three broad types: 'traditional' food webs, mutualistic networks and host-parasitoid networks. There is a recent trend towards cross-comparisons among network types and also to take a more mechanistic, as opposed to phenomenological, perspective. For example, analysis of network configurations, such as compartments, allows us to explore the role of co-evolution in structuring mutualistic networks and host-parasitoid networks, and of body size in food webs. 3. Research into ecological networks has recently undergone a renaissance, leading to the production of a new catalogue of evermore complete, taxonomically resolved, and quantitative data. Novel topological patterns have been unearthed and it is increasingly evident that it is the distribution of interaction strengths and the configuration of complexity, rather than just its magnitude, that governs network stability and structure. 4. Another significant advance is the growing recognition of the importance of individual traits and behaviour: interactions, after all, occur between individuals. The new generation of high-quality networks is now enabling us to move away from describing networks based on species-averaged data and to start exploring patterns based on individuals. Such refinements will enable us to address more general ecological questions relating to foraging theory and the recent metabolic theory of ecology. 5. We conclude by suggesting a number of 'dead ends' and 'fruitful avenues' for future research into ecological networks.

Chalk streams are among the most species-rich and productive of all temperate ecosystems. Despite this, a few keystone species have the potential to exert disproportionately powerful effects on community structure and ecosystem processes. Two of these are the bullhead Cottus gobio, a small benthic fish that is an extremely abundant, voracious predator, and the freshwater shrimp Gammarus pulex, which dominates the prey assemblage and is the principal detritivore. Field experiments detected a bullheadÁGammarusÁdetritus trophic cascade, with detrital processing rates slowed dramatically in the presence of the predator. In addition, survey data also revealed strong negative density-dependence between bullhead and brown trout, adding a further link in the cascade. However, although bullhead also depressed the abundance of a dominant grazer, the snail Potamopyrgus antipodarum, there was no cascading effect upon algal production, suggesting that autochthonous inputs were not controlled by topÁdown effects. This skewed effect of the predator upon autochthonous versus allochthonous basal resources stresses the need to consider both pathways of energy flux into the food web, whereas many previous studies have potentially overemphasized the importance of predatorÁherbivoreÁprimary producer cascades. The wider community food web contained 142 species and 1383 feeding links. This complex network exhibited ''small world'' properties, such as high clustering (unlike many other food webs) and shortest path lengths between species were small (in common with many other food webs). In particular, each of the four members of the detrital cascade could be connected to any other species by three links or fewer. Our data revealed that powerful cascading effects can be imbedded within even very complex ecological networks.

Body mass–abundance (M‐N) allometries provide a key measure of community structure, and deviations from scaling predictions could reveal how cross‐ecosystem subsidies alter food webs. For 31 streams across the UK, we tested the hypothesis that linear log‐log M‐N scaling is shallower than that predicted by allometric scaling theory when top predators have access to allochthonous prey. These streams all contained a common and widespread top predator (brown trout) that regularly feeds on terrestrial prey and, as hypothesised, deviations from predicted scaling increased with its dominance of the fish assemblage. Our study identifies a key beneficiary of cross‐ecosystem subsidies at the top of stream food webs and elucidates how these inputs can reshape the size‐structure of these ‘open’ systems.

Recent declines in Atlantic salmon Salmo salar populations are generally attributed to factors in their marine life-phase. However, it is postulated that factors affecting their freshwater life-phase might impact their marine survival, such as the influence of body size. While larger smolts are widely hypothesized to have higher marine survival rates, empirical support remains scant, in part due to inadequate data and ambiguous statistical analyses. Here, we test the influence of smolt body size on marine return rates, a proxy for marine survival, using a 12-year dataset of 3688 smolts tagged with passive integrated transponders in the River Frome, Southern England. State-space models describe the probability of smolts surviving their marine phase to return as 1 sea-winter (1SW) or multi-sea-winter adults as a function of their length, while accounting for imperfect detection and missing data. Models predicted that larger smolts had higher return rates; the most parsimonious model included the effect of length on 1SW return rate. This prediction is concerning, as freshwater juvenile salmon are decreasing in size on the River Frome, and elsewhere. Thus, to maximize adult returns, restoration efforts should focus on freshwater life-stages, and maximize both the number and the size of emigrating smolts.