https://orcid.org/0000-0003-2931-345X
![Sample our Bioscience journals, sign in here to start your access, Latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5925/TOC-Subject-Sample-2021-Bioscience.jpg"})
ABSTRACT
Wind, insects, or both? The pollination mode of chestnuts, an important genus of nut-producing forest trees of the Fagaceae family, is still unclear. We revisit this old question using an integrated approach, focusing on cultivated Castanea sativa trees and hybrids in South-western France. We first conducted a large-scale insect isolation experiment. We then monitored 16 trees, focusing on flowering phenology, flower abundance and insect visits. Half of these trees are male-sterile, helping explore the role of pollen in insect attraction. Finally, we characterized the pollination syndrome of chestnuts and contrasted it with that of wind-pollinated oaks using original and published data. Chestnut female flowers have erect styles resembling stamens from male flowers, a probable case of intersexual mimicry. The tree’s unusual phenology includes two peaks of pollen production. Pollinator exclusion experiments demonstrated a predominant role of insects in chestnut pollination. Flowering trees attract large numbers of beetles, bees and flies. In contrast, the few insects seen on female flowers (66 in 32 h of observation, <2% of the total) were mostly beetles. Compared to male-fertile trees, male-sterile trees attract fewer insects overall but their female flowers are more frequently visited and they have higher fruit set. All chestnut flower traits examined, such as the tiny pollen grains and the huge rate of pollen production, resulting in the highest pollen/ovule ratio ever reported in plants, are compatible with a beetle pollination syndrome. The high uncertainty of this pollination mode and its convergence with wind pollination explain the pervading confusion regarding chestnut pollination.
Acknowledgments
This paper is part of the PhD of CL. We thank C. Robin and T. Barreneche for their encouragement to explore the pollination biology of this fascinating tree. These experiments would not have been possible without the maintenance of the orchards by the Invenio team in Douville (24), the INRAE experimental unit (UE1442 Vignes Bordeaux) in Villenave d’Ornon (33), and X. Capdevielle of UMR Biogeco. For the netting experiment, we thank S. Cavaignac, N. Lebarbier, M. Mouravy, C. Gauthier, and all the staff who helped us count the fruits. For insect monitoring, we thank our Biogeco colleagues: I. Van Halder, who shared her expertise on insect ecology and helped monitor insect visitors in the field, and P. Chaumeil and F. Ehrenmann, who created the photo library and the associated database. We also thank M.-L. Benot and V. Fiévet for helpful discussions and advice as well as all colleagues that have helped with observations and granted access to facilities and equipment (D. Bert, M. Martin-Clotté, H. Christmann, B. Dencausse, C. Lalanne, Y. Mellerin, G. Saint-Jean, M. Trillat). We are grateful to the colleagues of the GDR Pollineco for great discussions and for B. Vaissières for advice, for pointing useful references and for critically reading and commenting the paper. We are pleased to acknowledge the help and support of the colleagues managing the citizen science program Spipoll, C. Fontaine (MNHN) and M. de Flores (OPIE). We are also very grateful to all “spipolliens” that contributed new observations on chestnut insect visitors following our request. Even if not directly used here, the data were very helpful when designing the study. We are indebted to our colleague J. Fernandez Lopez from Centro de Investigación Forestal de Lourizán (Galicia, Spain) for sharing her deep knowledge of chestnut ecology and genetics and for guiding us in one of the last natural chestnut forest in Galicia. Comments from two reviewers helped improve the manuscript. Finally, we are pleased to acknowledge the great support of our undergraduate students T. Menthonnex and J. Bonnier who measured thousands of catkins and flowers last spring, under conditions complicated by the Covid crisis.
Disclosure statement
No potential conflict of interest was reported by the authors.
Author contributions
CL performed all insect isolation experiments with Invenio staff in 2019. GB, RJP and CL surveyed insect visitors in 2019, GB established the photo library and performed all insect identifications from the photographs, EA and RJP carried out the measures of flower abundance with the help of two students from Bordeaux University, Julien Bonnier and Tanguy Menthonnex. RJP compiled information on chestnut’s pollination syndrome. CL and EA performed the statistical analyses, CL designed the figures in R. RJP and CL wrote the paper, with inputs from the other authors. Photographs of are by CL and all others are by RJP.
Supplementary material
Supplemental data for this article can be accessed here.