Cultivar placement affects pollination efficiency and fruit production in European pear (Pyrus communis) orchards

European pear (Pyrus communis) requires insect pollination among compatible cultivars for fruit production. However, most commercial orchards have a limited number of cultivars arranged in monotypic blocks or rows. This can result in insufficient inter-cultivar pollination. We hypothesise that limitations in pollen transfer among cultivars could be explained by both insect behaviour and orchard design. We compared insect activity and pollination efficiency in two European pear cultivars, in orchards with different designs: (i) cultivars alternated in the same row or (ii) cultivars in separate rows. To assess limitations in pollen transfer, we also compared hand pollination with compatible pollen versus open pollination by insects. Insect visitors mainly foraged on neighbouring trees within a row, with few movements across rows (1%). Honey bees (Apis mellifera) and bumble bees (Bombus terrestris) visited significantly more flowers per tree (8.5 vs. 3) and more trees (2.1 vs. 1.3) than solitary bees (Andrena spp.) and hoverflies. Insect visitors deposited large amounts of pollen (∼500 pollen grains) on flower stigmas regardless of the insect type. Cultivar placement affected inter-cultivar pollination; less incompatibility signs were observed when cultivars alternated in the same row (5%) than when cultivars were in separate rows (38%). We observed limitations in pollen transfer as open pollination resulted in significant reduced fruit set, compared with hand pollination, in 'Conférence' (21% vs. 30.7%) and 'Doyenné du Comice' (7.2% vs. 16.8%). The foraging behaviour of the insects limited thus inter-cultivar pollen transfer in the orchards with cultivars in separate rows. Cultivars used for pollination (pollinizers) should be planted in the same rows as the main cultivar to increase inter-cultivar pollination.