Island bees: do wood nesting bees have better island dispersal abilities? Nikolaj Rauff Poulsen, Claus Rasmussen To cite this version: Nikolaj Rauff Poulsen, Claus Rasmussen. Island bees: do wood nesting bees have better island disper- sal abilities?. Apidologie, 2020, 51 (6), pp.1006-1017. 10.1007/s13592-020-00778-x. hal-03271949 HAL Id: hal-03271949 https://hal.archives-ouvertes.fr/hal-03271949 Submitted on 28 Jun 2021 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Apidologie (2020) 51:1006–1017 Original article * INRAE, DIB and Springer-Verlag France SAS, part of Springer Nature, 2020 DOI: 10.1007/s13592-020-00778-x Island bees: do wood nesting bees have better island dispersal abilities? Nikolaj Rauff POULSEN, Claus RASMUSSEN Department of Bioscience, Aarhus University, Ole Worms Allé 1, 8000, Aarhus C, Denmark Received 20 November 2019 – Revised4May2020– Accepted 28 May 2020 Abstract – For bees to reach isolated islands, they need to be able to cross large water barriers. However, functional traits such as nesting behavior, flight range, and body size can limit their dispersal. In this study, the bee faunas of seven different islands or island groups (Anholt, Canary Islands, Fiji Islands, Hawaiian Islands, Madeira, Malta, and Sri Lanka) were analyzed by comparing them to the mainland bee fauna. Nesting strategy and body size for each species were recorded. The relative distribution of traits among the island species were then compared to traits of the mainland species. We found that among islands located furthest from mainland, the proportion of wood nesting bees are higher than for those located closer to mainland, suggesting that wood nesting bees are more capable of island dispersal, most likely due to the associated rafting ability or the ease of being passively transported over water. The body size among the immigrated island species were shown to be commonly of moderate size, while a majority of mainland species were of small size, indicating that moderate sized bees—with longer flight range—were the more likely island migrators. The methods used could be useful for future studies of ecological influences on body size, as well as how other biological traits are adapted for migrant conditions, potentially serving as predictors in regards to climate change, dispersal ability, and control of invasive species. island biogeography / insect dispersal / rafting / Hymenoptera / Apidae 1. INTRODUCTION Cane and Neff, 2011), with some bees having more specialized nesting strategies, e.g., nesting Bees (Hymenoptera: Anthophila) are a large on cliffs (Roubik et al., 1985), in snail shells and diverse clade of insects, consisting of approx- (Müller et al., 2018), or in a diversity of self- imately20,000different species (Ascher and made aerial nests (Dressler, 1982; Wcislo et al., Pickering, 2019). Within the clade, a wide diver- 2012; Rasmussen et al., 2015). In terms of social- sity can be found in terms of morphological and ity, the majority of bee species, > 85%, does not ecological strategies for reproduction and forag- live in larger groups like the honey bee (Apis ing (Michener et al., 1994; Michener, 2007). Bees mellifera ), but live solitarily (Batra, 1984; usually nest in ground burrows or cavities in Danforth et al., 2019). The smallest known bees sticks, stems, and branches (Michener, 1964; are about 2 mm long (Michener, 2007), while some of the largest known bees include Wallace’s giant bee (Megachile pluto ), with a body length of Electronic supplementary material The online version of approximately 40 mm (Messer, 1984;Main, this article (https://doi.org/10.1007/s13592-020-00778-x) 2019). This diversity among the bees comes from contains supplementary material, which is available to authorized users. a long evolutionary timeframe, which is thought to have begun already during the Early Cretaceous Corresponding author: C. Rasmussen, (140 to 110 Ma), with the bees arising within [email protected] apoid wasps in the arid interior of west Manuscript editor: James Nieh Island bees: do wood nesting bees have better island dispersal abilities? 1007 Gondwanaland (Africa, South America) islands with harsh winter conditions (Michener, (Michener, 1979), or most likely in modern Africa 2007; Williams et al., 2019). In order for bees to (Danforth et al., 2006; Danforth and Poinar, have naturally dispersed across most of the conti- 2011). Their origin here coincides with the diver- nents and major continental islands, dispersal sification of flowering plants (Angiosperms) must have taken place during a period when the (Cardinal and Danforth, 2013). This synchrony continents were still adjacent, and most likely of the timing in the origin of the bees and angio- only separated by narrow water straits during the sperm diversification, represent one of the most end of Cretaceous and beginning of Eocene successful coevolutionary partnerships (Danforth (Michener 1979). For most bees, dispersal has et al., 2013), with bees dependent on food from presumably been slow across the continents and angiosperm flowers, and many angiosperms de- to the nearby landmasses, or by transport on mov- pendent on bees for pollination (Michener, 1979). ing continents (Michener, 1979). The bee faunas Because of this partnership, bees are considered of the Antilles and of central Indonesia show that among the most important pollinators of scattered islands between continents can serve as flowering plants—of which 78% in temperate stepping stones between the continents communities and 94% in tropical communities (Michener, 1979), allowing island hopping/jump depend on animal pollinators (Ollerton et al., dispersal (Whittaker et al., 2007). Importantly, 2011). some island groups like the Canary and Hawaiian Such plant-pollinator dependency can also pro- Islands had not formed during the Cretaceous, and vide a limitation for angiosperms that disperse to both island groups (Gillespie and Clague, 2009), new islands. Particularly, self-incompatible angio- especially the Hawaiian Islands, are located far sperms species are at a disadvantage when colo- out in the ocean, with no possibility for jump nizing islands, especially after long dispersal, be- dispersal. For Hylaeus (Nesoprosopis) to have cause self-incompatible species require a mini- arrived to Hawaii during a single rare event, pos- mum of two individuals in addition to an animal sible from Japan, some other dispersal mechanism vector, to become established (Inoue et al., 1996). would be necessary (Daly and Magnacca 2003). For bees to reach isolated or oceanic islands and Wind dispersal to islands of the smaller bees have become potential pollinators of angiosperms pres- been suggested as a possible dispersal mecha- ent therein, they need first to cross large water nism, but most bees prefer to stay in the nest barriers. While natural dispersal capacity is little during storms, only adventuring out to fly in good known for bees, some data exists about foraging weather, thus severely limiting the number of range, although this is necessarily shorter than individuals available for accidental dispersal dur- maximum flight range. For solitary bees, the max- ing storms (Michener, 1979). Yet, storms and imum recorded foraging distance is 1400 m for violent weather might still play a role in the island the larger species, with the maximum foraging dispersal of bees. Bees nesting protected inside range decreasing for smaller species with some sticks, branches, or stems could be rafted or blown only foraging between 100 and 300 m away from in severe storms, potentially moving bee nests to the nest (Zurbuchen et al., 2010). An exception to an isolated island where angiosperm pollen and this are some orchid bees (Apidae: Euglossini: nectar are available for food, providing opportu- Eulaema ) that have been observed returning to nities for the establishment of a bee fauna. Bee their nest as far away as 23 km or more (Janzen, species are also during the last 200 years recorded 1971;Wikelskietal.,2010). Even with a potential as exotic outside their native range, in particular flight distance of 23 km for the strongest fliers around commodity entry points, and are evidently among bees, bees would only be able to actively able to survive accidental or facilitated long- make it to a very narrow selection of islands that distance transport on trans-oceanic cargoes are not well isolated geographically (Holzapfel (Cane, 2003; Sheffield et al., 2011;Jensenand and Harrell, 1968; Ashmole and Ashmole, 1988; Madsen, 2018; Rasmussen et al., 2020). Such Peck, 1994). Yet, bees today are found on all human-mediated events may very well be fre- continents, except Antarctica, including remote quent and important. 1008 N. R. Poulsen, C. Rasmussen In this study, we test the hypothesis by the late the island and elsewhere. This was necessary as Charles Michener, world bee authority, regarding only traceable and immigrated species were of bees on chiefly oceanic islands: Bees of all sizes interest. Although of importance, we were not that primarily nest in wood, stems, or twigs are able to differentiate between species that have most likely to make major dispersal steps across been brought across oceans aided by humans large water barriers, while moderate and large- and those that presumably have arrived without sized ground nesting bees are less likely to do so human assistance. Such information is scattered (Michener, 1979). We addressed this by compil- and often non-existent or indirect (see for example ing species lists for each of the selected islands, Rasmussen et al., 2012), with the exception of, including species-specific biological traits.