Deploying small-mesh drift nets in rivers is a well-established method for sampling drifting fish larvae and eggs. Quantitative comparisons are sometimes made on the basis of numbers of larvae captured per unit volume or time. In this study a GoPro ™ camera was mounted inside the drift net to record the change in flow over time (1 to 5 minute intervals for 3 hours). Although a small number of net nights (7 nights at 3 locations) were sampled, variance in the change in flow within and between sites was observed – even during soak times as little as 2 hours. In one case there was almost no change in flow over 180 minutes but at the most extreme, the flow dropped from 8.9 m3/min to 1.5 m3/min in just 160 minutes. Variance is probably due to the level of suspended particulates at different sites or times. If volumetric or temporal estimates are made on the basis of total flow only they could in some cases be misleading and at worst make comparisons almost meaningless. While there are dedicated data logging flow meters available they are prohibitively expensive for routine sampling. Researchers could consider the method used in this study to cost effectively assess the decay in net performance during sampling.