Abstract

Fragmentation of natural vegetation is currently one of the largest threats to plant populations and their interactions with pollinators. Plant reproductive susceptibility to habitat fragmentation has been investigated in many species; however, the response of wild mass‐flowering species is poorly known, with research limited to mainly boreal plant species.Here, we studied twelve remnant populations of the threatened mass‐flowering shrub Conospermum undulatum in the southwest Australian biodiversity hotspot, each presenting different population size, level of isolation, and floral display. We assessed the impact of fragmentation on (a) fruit and seed production; and (b) seed germination. To gain a deeper understanding of factors influencing the reproductive success of C. undulatum, we performed pollinator exclusion and self‐pollination treatments to experimentally assess the mating system of this threatened shrub.We found C. undulatum to be strictly self‐incompatible and totally reliant on pollinators visiting with an outcrossed pollen load to complete the reproductive cycle. Further, we found that fruit production dropped from 35% to <20% as a result of decreasing floral display. A reduction in population size from 880 to 5 plants and from ~700 to 0.21 in the floral display index led to a decrease in seed output, while a similar reduction in seed output, from 6% to 3%, was observed as a result of increasing isolation index from −21.41 to −0.04. Overall, seed germination was positively related to population size, and a negative relationship was found between germination and isolation. Synthesis and applications. Our results demonstrate the important relationship between pollinators and floral morphology in plants of southwest Australia that have coevolved with native pollinators and developed characteristic flower morphologies over long time frames. Indeed, due to its characteristic pollination mechanism, the self‐incompatible C. undulatum can only rely on specialized native pollinators for pollen flow and cannot rely on its mass‐flowering trait to attract generalist pollinators from coflowering species; neither can it compensate for the lack of visitors by promoting geitonogamy. Consequently, fragmentation has a significant effect on the reproductive output of C. undulatum, and size, isolation, and floral display of populations are important factors to be considered when planning conservation actions for the species.

Highlights

  • Overall, seed germination was positively related to population size, and a negative relationship was found between germination and isolation

  • | 11495 the self‐incompatible C. undulatum can only rely on specialized native pollinators for pollen flow and cannot rely on its mass‐flowering trait to attract generalist pollinators from coflowering species; neither can it compensate for the lack of visitors by promoting geitonogamy

  • To account for possible effects of temporal isolation, we recorded the reproductive phenology of this species once a week for the entire flowering season and we evaluated the flowering synchrony between populations using a modified version of the method proposed by Freitas and Bolmgren (2008), replacing individuals with populations

Read more

Summary

| MATERIALS AND METHODS

The study was conducted in southwest Western Australia within the Swan Coastal Plain bioregion (Figure 1). In a medium‐sized population of C. undulatum (216 plants), we randomly selected ten plants per treatment 2 weeks before anthesis, and we placed fine mesh bags around three inflorescences per plant In this way, we prevented insects from visiting the flowers (i.e., PE treatment). Full models for fruit and seed production contained all the retained population descriptors (i.e., population size, isolation, and floral display) as the explanatory variables. TA B L E 1 Reproductive output of Conospermum undulatum in term of fruit and seed production for pollinator exclusion (PE), exclusion and triggered flowers (PET), and exclusion and hand self‐pollination (PES) treatments. The probability that a flower will develop a seed showed a significant relationship with all the explanatory variables of population size, isolation, and floral display (F1,173 = 29.80, p < .001; F1,173 = 14.88, p < .001; F1,173 = 16.80, p < .001, respectively).

| DISCUSSION
Findings
CONFLICT OF INTEREST
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call