The extent and intensity of bottom trawling effort has continued to grow throughout the world, despite a consensus that this practice causes significant damage to marine benthos and commercial fish stocks. Many of the studies designed to test the effect of trawling on marine communities have been done on recently trawled grounds, which limits the impact of experimental trawls, and hence the decreases the ability of studies to confirm that trawling affects marine systems. The benthos is notoriously variable, which also hinders the ability of studies to detect changes in response to trawling, and necessitates careful design of experiments. This study tested the effect of trawling on macroinfaunal assemblage structure (types of taxa and relative abundances), in an area where little or no trawling had occurred in the previous 15 years. A spatially replicated Before-After, Control-Impact (BACI) design was used, incorporating three locations that each included a control corridor and an adjacent impact (trawl) corridor. Samples were collected from the same two small sites within each of the corridors before and after trawling to minimise confounding by spatial variation. Despite this rigorous design, changes consistent with an effect of trawling were not detected. Only at one of the three locations did the patterns in the types and abundances of infauna before and after trawling support the hypothesis that trawling causes changes to the structure of infaunal assemblages. These inconsistent results could possibly be due to differing vulnerability of the fauna and sediment at the locations to trawling-induced changes, or differences in the ability to detect a change. Given the high levels of natural variability inherent in infaunal assemblages however, the changes observed could be due to natural asynchronous variability. Making inferences about changes at only one location is not valid, given that natural variation can produce the same patterns. The combination of large spatial and temporal variability, in association with light trawling gear, means that prawn trawling in South Australia does not have consistent measurable effects on assemblages of infauna in GSV.