“…This match between pollen size and style length within heterostylous species has long been recognized, and it has been considered to be an adaptation enabling compatible pollen to grow through the stylar tissue (Ganders, 1979; Dulberger, 1992; Barrett et al, 2000; Ferrero et al, 2009). Such within‐species selection, with a long style favoring large pollen, has been documented in comparative studies in numerous genera and families, suggesting that selection of pollen size is similar among species (Baker and Baker, 1982; Plitmann and Levin, 1983; Williams and Rouse, 1990; Bigazzi and Selvi, 2000; Roulston et al, 2000; Torres, 2000; Sarkissian and Harder, 2001; Aguilar et al, 2002; Yang and Guo, 2004; López et al, 2005; Jürgens et al, 2012; but see Cruden, 2009; Bedinger et al, 2011; Wang et al, 2016). Darwin (1877), however, showed that in seven distylous genera, pollen grain size was similar in the two morphs, while the pistil (style) of the L‐morph was about two or three times as long as that of the S‐morph, for example in Coccocypselum , Limnanthemum , and Linum .…”