Surface traps are recommended to monitor slugs in arable crops, but pastureland differs from arable land in many respects and these may affect the performance of sampling methods. Here, we report on the performance of non‐baited mat refuge traps, permanently placed at the same position over more than 2 years in four established pastures, for the assessment of slug numbers and biomass, in comparison with soil sampling and flooding over 3 days. Despite the high availability of alternative shelters provided by the vegetation, a great many slugs were captured in the traps over the study period and over a wide range of temperatures recorded under the traps. The catches of slugs in traps and soil samples showed significant positive relationships in terms of numbers and biomass per sample unit and also for the mean weight of the slugs registered in each type of sample, but traps showed a bias towards the larger individuals and underestimated the numbers of smallest slugs of each species (Deroceras reticulatum, Deroceras panormitanum, Deroceras laeve and Arion intermedius). The relationships between the temperature under the traps at the time of collecting and the numbers of trapped slugs were found to be species‐dependent. Besides the slugs, abundant slug eggs were regularly found beneath the traps at the four study sites, suggesting that permanent mat traps could be used to provide useful information about oviposition activity of slugs in the field.