Manipulation of trait expression and pollination regime reveal the adaptive significance of spur length

The excel files provide data on: (1) Pollination success (pollen removal and pollen receipt) collected in a field experiment conducted in a short-spurred grassland and in a long-spurred woodland population of the orchid Platanhera bifolia on the island Öland in southeastern Sweden in 2017. In this experiment, artificial nectar spurs were used to vary the spur length of experimental plants. (2) Data on sugar concentration in nectar produced by flowers in a short-spurred grassland and in a long-spurred woodland population. This information was used to decide the sugar concentration to be used in "nectar" added to artificial spurs. (3) Pollen removal and receipt in flowers with intact nectar spurs and in flowers with artificial nectar spurs in plants included in the artificial-spur experiment. These data were used to check that pollination success was similar in intact and manipulated flowers within the natural range of spur lengths in each of the two study populations. (4) Data on pollen removal, female fitness, and floral traits in two grassland and in two woodland populations of P. bifolia on the island Öland in southeastern Sweden in 2016. These data were used to quantify phenotypic selection among open-pollinated control plants and among plants receiving supplemental hand-pollination, and from these results estimate pollnator-mediated selection, non-pollinator mediated selection, and net selection on spur length, plant height, number of flowers, and flower size.