Electrified pulse trawls have replaced traditional tickler chain beam trawls in the North Sea fisheries for sole. This study investigates the mesh selection in pulse trawling of conventional cod-ends (80 mm cod-end mesh) used in the current pulse trawl fishery, and the effects of increasing the cod-end mesh size to 90 mm on catches of sole (Solea solea) and undersized plaice (Pleuronectes platessa). Cod-end selectivity was estimated for 79-80 mm and 87-88 mm cod-ends during two experiments on a commercial pulse trawler using a cover cod-end. The results show that with a mesh size of 79-80 mm the length where 50% of the individuals are retained (L50) for sole is 19 cm with a selection range (SR) of 4.9 cm. Given the observed length distribution of sole on the fishing ground this results in a 10% loss of marketable sole catches in the 24-27 cm length range. Increasing the mesh size in experiment one to 87 mm resulted in a L50 for sole of 22 cm with SR = 4.9 cm and in experiment 2 to a L50 of 26 cm and SR = 4.9 cm was found for 88 mm cod-end, resulting in a loss of marketable sole of 24% and 38% in experiment 1 and 2, respectively. These losses were detected in the 24-33 cm length range. Compared to sole, plaice showed steeper selection curve with a L50 of 14.4 cm (SR 2.5) and 14.1 cm (SR 2.1) for the 79-80 mm cod-ends in experiment 1 and 2, respectively. In the 87 mm cod-ends, this L50 shifted to 15.6 cm (SR 2.5) for experiment 1 and 18.7 cm (SR 2.1) for the second experiment. The ratio of plaice discards per kg marketable sole caught was 0.4 in experiment one for 80 mm cod-ends, and increased to 0.5 in a 87 mm cod-end. In the second experiment this was 2.3 for 79 mm and 2.5 for 87 mm. Increasing the minimum cod-end mesh to 90 mm thus increases the discard quantities of undersized plaice when the sole total allowable catch (TAC) is fully exploited.