Action: Increase the use of clover leys on farmland
We have captured no evidence that increasing the use of clover leys can enhance wild bee populations. One replicated trial in Germany showed that fields planted with a white clover grass mixture do not attract solitary bees to nest preferentially on site. A trial in Switzerland showed that if white clover is mowed during flowering, injuries and mortality of bees can be reduced by avoiding the use of a processor attached to the mower.
Supporting evidence from individual studies
As part of a larger study with 10 field types, Gathmann et al. (1994) placed bundles of reed stems Phragmites australis for cavity-nesting bees (and wasps) in four set-aside fields newly sown with a clover-grass mix. The mix was mostly white clover Trifolium repens, perennial rye grass Lolium perenne and alfalfa Medicago sativa. Four species of bee made nests in the reed stems in these fields, including one endangered species Megachile alpicola. However, in the same study, three of those four species also nested in reed stems placed in barley Hordeum vulgare fields. By contrast, 12 bee species nested in reed stems placed in 2-year-old set-aside fields mown in late June, and 16 species nested in reed stems in old meadows.
Fluri & Frick (2002) rotary mowed white clover crops during flowering with and without a mechanical processor, and monitored the death and injury to actively foraging honey bees Apis mellifera, on two 0.33 ha trial plots in Switzerland. During mowing with a rotary mower and processor (which crushes mowings to accelerate drying), 53-62% of the number of foragers recorded before mowing were found injured, dead or otherwise stuck in the mown grass after mowing. When mowing was conducted without a processor, the average number of bees left dead or unable to fly was reduced from an average of 1.4 bees/m2 (with processor) to 0.2 bees/m2 and many bees were observed foraging or flying away after passing through the mower. The effects of mowing with a processor (but not without) were also tested on a similar-sized plot of Phacelia tanacetifolia, on which bumblebees were recorded as well as honey bees. On average, 0.2 foraging bumblebees/m2 were recorded before mowing, and 'practically' no bumblebees were found in the mown grass.