ABSTRACT
Background
Most bromeliad (Bromeliaceae) species have specialised flowers for specific pollinators, establishing a mutualistic relationship. However, other animals may also pollinate bromeliads, though it is logical to assume that specialised flowers are more compatible with one specific group of pollinators.
Aims
We compared the performance of a hummingbird (Thalurania glaucopis) and a butterfly (Heliconius erato) as pollinators of flowers classified as ornithophilous of the bromeliad Tillandsia stricta (Bromeliaceae).
Methods
We applied the single visit method with seed counting, and measured corolla tube length of flowers. We also conducted observations on plants to determine the effectiveness of pollination by the hummingbird and the butterfly.
Results
Both animal species pollinated flowers, although the hummingbird was a more frequent and effective pollinator, resulting in greater seed production. The corolla tube length of T. stricta was a significant predictor of seeds produced per flower following interactions with the hummingbird, but not with the butterfly.
Conclusion
Our results indicate a more intimate flower – pollinator relationship between T. stricta and the hummingbird than the butterfly in this population of the bromeliad in Rio de Janeiro, Brazil.
Acknowledgments
Research in Pedra Branca State Park (PEPB: Parque Estadual da Pedra Branca) was authorised by the Rio de Janeiro State Environment Institute (INEA: Instituto Estadual do Ambiente) through permit number 008/2015. The authors thank PEPB for logistical support. They are also grateful to Stephen Ferrari for his review of the English text. The authors also thank the editors (including the subject editor, Richard Abbott) and the anonymous reviewers for the valuable contributions which greatly improved the manuscript. Funding for this study was provided by the Brazilian Coordination for Higher Education Personnel Training (CAPES: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) by granting C.C.C.M. a doctoral research scholarship and C.J.O.M with a master’s scholarship (this study was financed in part by CAPES—Finance Code 001). The Rio de Janeiro State Research Foundation (FAPERJ: Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro) provided C.C.C.M. with a grant (DSC-10 process E_01/201.955/2017) and supported the research of M.A.S.A. (CNE processes E-26/203191/2015, E-26/202.835/2018, E-26/201126/2022). The Brazilian National Research Council (CNPq: Conselho Nacional de Desenvolvimento Científico e Tecnológico) provided M.A.S.A. with a research fellowship (PQ processes 305798/2014-6, 306.579/2018-9, 308615/2022-0), who was also supported by a productivity fellowship from Rio de Janeiro State University (Prociência, UERJ/FAPERJ).
Disclosure statement
No potential conflict of interest was reported by the author(s).
Authors’ contributions
CCCM conceived the study and designed the methods; CCCM collected the data; CCCM and CJOM analysed the data; CCCM and CJOM led the writing of the manuscript. MASA participated in the writing, contributing with important intellectual content and revision, and financial support through her research grants. All the authors contributed fundamentally to the drafts and gave final approval for publication.
Data archiving statement
The data were published in the Zenodo Repository (DOI: 10.5281/zenodo.6299296), with a one-year embargo (until 31 December 2023).