Increased flood frequency and magnitude are predicted for Scotland, and the country contains several of the world's largest recruiting populations of freshwater pearl mussel (Margaritifera margaritifera). This study provides a unique flume experiment to measure the near‐bed velocities required for freshwater pearl mussel entrainment and factors affecting their movement. It represents the first quantitative attempt at examining the precise water velocities at which freshwater pearl mussel become vulnerable to displacement during high flow events. Measurement of the near‐bed velocities at which the mussels moved was undertaken using an indoor recirculating flume. The effect of the different parameters (bed substrate, mussel burial depth, mussel curvature, mussel alignment, shell curvature and the presence of a simulated foot) on entrainment velocity was tested in the flume and their significance was verified using Kruskal–Wallis and Mann Whitney tests. Bed substrate was found to have the biggest influence on mussel entrainment velocities with averages of 0.86 ms−1, 0.95 ms−1, 1.01 ms−1 and 1.42 ms−1 for sand, gravel, mixed bed and boulder beds respectively. Stepwise logistic regression showed that bed substrate, foot presence, mussel length, mussel burial depth and shell curvature were sufficient to explain mussel entrainment velocity. These findings provide valuable information for the modelling of freshwater pearl mussel dynamics in streams systems and assessing the vulnerability of endangered mussel populations to higher flows associated with climate change in Scotland. Copyright © 2015 John Wiley & Sons, Ltd.