Pollinator and floral odor specificity among four synchronopatric species of Ceropegia (Apocynaceae) suggests ethological isolation that prevents reproductive interference

Abstract

Possession of flowers that trap fly pollinators is a conservative trait within the genus Ceropegia, in which pollination systems can be generalized or highly specialized. However, little is known about the role of plant-pollinator interactions in the maintenance of species boundaries. This study examined the degree of plant-pollinator specialization and identified the parameters responsible for specificity among four co-occurring Ceropegia species with overlapping flowering times. All investigated plant species were functionally specialized on pollination by Chloropidae and/or Milichiidae flies and each Ceropegia species was, in turn, ecologically highly specialized on only two pollinating fly morphospecies, though one plant species appeared more generalist. Species-specific fly attraction was due to the differences between plant species in floral scents, floral morphology, colour patterns, and presence of other functional structures, e.g., vibratile trichomes, which were shown to contribute to pollinator attraction in one study species. The combination of these olfactory and visual cues differentially influenced pollinator preferences and thus hindered heterospecific visitation. Furthermore, a pollinator exchange experiment also highlighted that species integrity is maintained through efficient ethological isolation (pollinator attraction). The mechanical isolation mediated by the fit between floral morphology and size and/or shape of fly pollinators appears less pronounced here, but whether or not the morphological match between male (pollinium) and female (guide rails) reproductive organs can impede hybridization remains to be investigated.

Figure 1

Photographs taken under a stereo microscope (A–D) and scanning electron microscopy images (E–F) of the most abundant pollinaria-carrying fly morphospecies collected from flowers of Ceropegia species at Pha Taem National Park, Thailand. Neophyllomyza sp. 1 collected from C. citrina flowers (A); Milichiella sp. 2 collected from C. tenuicaulis flowers (B, E); Chloropidae sp. 8 collected from C. boonjarasii flowers (C); Chloropidae sp. 7 collected from C. acicularis flowers (D, F). Red arrows indicate where on the proboscis the corpusculum of a pollinarium gets attached. Photographs by Rumsaïs Blatrix (A–C) and Manit Kidyoo (D–F).

Figure 2

Morphology of gynostegia and coronas in side view (A, D, G, J) and in top view (B, E, H, K), and the microstructure of pollinaia (C, F, I, L) of Ceropegia acicularis (A–C), C. boonjarasii (D–F), C. citrina (G–I) and C. tenuicaulis (J–L).

Figure 3

Floral morphological differentiation among the four sympatric Ceropegia species. (A) C. acicularis; (B) C. boonjarasii; (C) C. citrina; (D) C. tenuicaulis; (E) PCA biplot of individual flowers based on measurements (TFL total flower length, CTL corolla tube length, SDCT smallest diameter of corolla tube, LBI basal inflation length, WDBI widest diameter of basal inflation). Colored concentration ellipses are given around individuals for each species (grouping variable).

Figure 4

Non-metric multidimensional scaling (NMDS) based on the relative amounts of all volatile organic compounds (VOCs) detected from flowers of the four different Ceropegia species. The data were standardized prior to the analysis and Bray–Curtis distance was used. Only compounds in trace amount (< 0.1%) were not included in the analysis.