Nectar robbing by bees affects the reproductive fitness of the distylous plant Tirpitzia sinensis (Linaceae)

Abstract

Nectar robbing can affect plant reproductive success directly by influencing female and male fitness, and indirectly by affecting pollinator behavior. Flowers have morphological and chemical features that may protect them from nectar robbers. Previous studies on nectar robbing have focused mainly on homotypic plants. It remains unclear how nectar robbing affects the reproductive success of distylous plants, and whether defense strategies of two morphs are different. Nectar-robbing rates on the long- and short-styled morph (L-morph, S-morph) of the distylous Tirpitzia sinensis were investigated. We compared floral traits, the temporal pattern of change in nectar volume and sugar concentration, nectar secondary metabolites, and sugar composition between robbed and unrobbed flowers of two morphs. We tested direct effects of nectar robbing on female and male components of plant fitness and indirect effects of nectar robbing via pollinators. Nectar-robbing rates did not differ between the two morphs. Flowers with smaller sepals and petals were more easily robbed. The floral tube diameter and thickness were greater in L-morphs than in S-morphs, and the nectar rob holes were significantly smaller in L-morphs than in S-morphs. Nectar robbing significantly decreased nectar replenishment rate but did not affect nectar sugar concentration or sugar composition. After robbery, the quantities and diversity of secondary compounds in the nectar of S-morphs increased significantly and total relative contents of secondary compounds in L-morphs showed no obvious changes. Nectar robbing could decrease female fitness by decreasing pollen germination rate and thus decreasing seed set. Nectar robbing had no significant effects on male fitness. Robbed flowers were less likely to be visited by hawkmoth pollinators, especially in S-morphs. These results suggest that nectar robbing could directly and indirectly decrease the female fitness of T. sinensis, and different morphs have evolved different defense mechanisms in response to nectar-robbing pressure.

Keywords: defense strategies; distyly; nectar robbing; plant fitness; reproductive success.

FIGURE 1

Undamaged flowers (pink arrows) and flowers with obvious nectar rob holes (red arrows) of the long‐styled morph (a) and short‐styled morph (b) of distylous Tirpitzia sinensis. The long‐tongued hawkmoth Macroglossum pollinates the flowers of long‐styled morphs (c) and short‐styled morphs (d). Bumblebees (Bombus breviceps) rob nectar through the rob hole at the base of the corolla tube of the long‐styled morph (e) and the short‐styled morph (f). Carpenter bees (Xylocopa sinensis) rob the floral nectar of the long‐styled morph (g) and the short‐styled morph (h).

FIGURE 2

Comparison of the dynamics of nectar volume and nectar sugar concentration of unrobbed flowers (solid) and robbed flowers (open) in L‐morphs (triangles) and S‐morphs (circles) of T. sinensis from 08:00 to 20:00 h. Different letters indicate significant differences between different times. The capital (A/B) letters and lowercase (a/b) letters refer to comparisons of different times for unrobbed and robbed flowers respectively.

FIGURE 3

PCA score plots (a) of the nectar from L‐morphs of unrobbed (green), L‐morphs of robbed (purple), S‐morphs of unrobbed (orange), S‐morphs of robbed (blue) samples after UPLC‐Qtof analysis. (b) Heat map cluster analysis of the nectar from unrobbed L‐morphs, robbed L‐morphs, unrobbed S‐morphs, robbed S‐morphs samples after UPLC‐Qtof analysis. High and low abundance are indicated by red and blue colors, respectively. OPLS‐DA score plots of the robbed (triangles) and unrobbed (crosses) nectar samples from L‐morphs (c) and S‐morphs (d).

FIGURE 4

The direct effects of robbing on female components of L‐ and S‐morphs in T. sinensis. Comparison of pollen germination rate (a) and pollen tube lengths (b) of these two morphs with intermorph pollination treatment, intermorph pollination of pollen recipients with artificial rob holes, intermorph pollination of pollen recipients with nectar artificially removed, and comparison of seed sets (c) of these two morphs with the above three pollination treatments and open‐pollinated undamaged flowers as controls, and open‐pollinated flowers with obvious natural rob holes (different letters indicate significant differences among different treatments). The capital (A/B) letters and lowercase (a/b) letters refer to the comparison of different pollination treatments in S‐morphs and L‐morphs, respectively, as pollen recipients.

FIGURE 5

The direct effects of nectar robbing on two components of male plant fitness: pollen production per flower (a) and pollen siring ability (b). Comparison of the number of pollen grains between the anther of undamaged buds (control) and buds with artificial rob holes (a) and the seed sets with open‐pollinated controls and intermorph pollination with the pollen from undamaged intermorph individuals and from intermorph individuals with obvious natural rob holes (b) in L‐ and S‐morphs. Capital (A/B) letters and lowercase (a/b) letters refer to the comparison of different pollination treatments in S‐morphs and the L‐morphs, respectively, as pollen recipients.