Assessing impacts of two major new exotic pollinators in Fiji

A project undertaken by Flinders University, South Australian Museum, University of the South Pacific, University of Adelaide, and Secretariat of the Pacific Regional Environment Program. Project leaders are Mike Schwarz (Flinders University), Mark Stevens (South Australian Museum), Scott Groom (University of Adelaide), Marika Tuiwawa (University of the South Pacific) and Posa Skelton (SPREP)


Introduced exotic species that become ‘weedy’ often have very negative impacts on native ecosystems.  On the other hand exotic pollinators, such as the honeybee Apis mellifera and the bumble bee Bombus terrestris, are often valued for their ability to increase crop yields.  Balancing these two factors can be difficult, and land/resource managers need to be informed by solid research on likely impacts.

Our project aims to assess the likely benefits and threats posed by bee species that have been recently introduced into Fiji, how they are adapting to Fijian habitats, and how this might impact on endemic pollinators.

Our project builds collaboration between Flinders University, University of the South Pacific, University of Adelaide, the South Australian Museum, and the Secretariat of the Pacific Regional Environment Program (SPREP). It will also contribute to the training of Fijian and Australian research students in the fields of agriculture, ecology, biodiversity, and conservation genetics.

Progress to date

We have recently shown that the distributions of the exotic bees Braunsapis puangensis and Ceratina dentipes are much more extensive than had been previously realized (da Silva et al. 2016).  These bees have been able to disperse throughout most of Fiji in a short time and we have also reported their recent dispersal into French Polynesia (Groom et al. 2016).  Itseems likely that both exotic species will become widespread throughout many SWP archipelagos.

We have also modelled the likely distribution of Braunsapis puangensis in the Indo-Pacific region under various future climate change scenarios (Silva et al. 2016).  Our modelling shows that B. puangensis is likely to extend its distribution in the future.  Whilst this may have some negative ecosystem impacts, this long-tongued bee species has the potential to provide important crop pollination services if there are any threats to the honey bee, Apis mellifera, for example via spread of honeybee pathogens.

We have examined social organization of Braunsapis puangensis in Fiji and shown that it is a casteless species, living socially but with no social hierarchies (da Silva et al. 2015).  This has implications for understanding social evolution, especially with respect to gain and loss of hierarchical structure.

Our most recent studies (currently in preparation for journal submissions) have demonstrated some unexpected findings with potentially major implications:

  1. The endemic bee fauna of Fiji is substantially greater than earlier studies suggest, but with most of the undescribed new species being morphologically ‘cryptic’ and restricted to highlands.  We are now exploring whether this unexpected species diversity is due to Pleistocene cycles of climate change.
  2. At least some endemic Fijian bee species (all in the halictine genus Homalictus) are able to effect pollination of solanaceous pasture weeds, such as Solanum torvum, and that may help explain why such weeds have become so widespread in Fiji, even though they usually require buzz pollinators (which are not endemic to Fiji) (Staines et al. 2017).  Our most recent data (Hayes et al. in prep.; Francis et al. in prep.) suggest that endemic Fijian bee species are indeed ‘super-generalists’ and this may indicate why some tropical archipelagos, such as Fiji, are especially vulnerable to invasion by exotic weeds.
  3. Amegilla pulchra is another bee that has been introduced to Fiji and is a buzz pollinator, so has the potential to increase seed sets for solanceous weeds.  However, our most recent work shows that it feeds entirely on introduced garden ornamental plants or introduced non-solanceous weeds (Groutsch et al. in review).  It is therefore unlikely to increase the spread of Solanum pasture weeds, and their spread now seems to be entirely due to endemic Fijian bee species.
  4. Behavioural studies on the introduced bee Braunsapis puangensis reveal complex interactions among nestmates, including trophollaxis (mouth-to-mouth food exchange).  Combined with molecular phylogenetics and studies on Australian bees, this finding implies that such complex behaviour has arisen at least 30 million years ago, indicating an ancient origin for complex social behaviour.
  5. Our field work has also explored the nesting and social biology of endemic Fijian halictine bees for the first time.  All species that we have examined are communal nesters (cooperative nesting but without queen or worker castes) and this indicates that the so-called “Australian Enigma”, which involves the puzzling ubiquity of egalitarian bee societies in Australia, extends to the tropical Pacific island bee fauna.  Understanding why this form of sociality, which is rare in the rest of the world but is common in Australia, and now the southwest Pacific, forms an important challenge to current theories of social evolution.
  6. Our most recent genetic work has revealed the existence of at least 15 new endemic bee species in Fiji – most restricted to highlands.  This indicates a greatly unexpected diversity in highland bee fauna, and these are all likely to be very vulnerable to climate change.  Analyses indicate that the formation of these species is driven by climate specialization and probably augmented by cross infections by the bacterial parasite Wolbachia.
Recent publications

Our recent publications from APSF-funded work includes the following papers.  If you are unable to access these journal papers via the web, please email Mark Stevens, Scott Groom or Mike Schwarz for a pdf reprint.

Groom, SVC, Stevens, MI and Schwarz, MP (2014) Parallel responses of bees to Pleistocene climate change in three isolated archipelagos of the southwestern PacificProceedings of the Royal Society B: Biological Sciences 281 (1785)

Groom, SVC, Ngo, HT, Rehan, SM, Skelton, P, Stevens, MI and Schwarz, MP (2014) Multiple recent introductions of apid bees into Pacific archipelagos signify potentially large consequences for both agriculture and indigenous ecosystemsBiological Invasions 16 (11), 2293-2302

Groom, SVC, Tuiwawa, MV, Stevens, MI and Schwarz, MP (2014) Recent introduction of an allodapine bee into Fiji: A new model system for understanding biological invasions by pollinators.  Insect science  DOI: 10.1111/1744-7917.12136

Groom, SVC, Hayes, SE, Ngo, HT, Stevens, MI and Schwarz, MP  (2014) Recipe for disruption: multiple recent arrivals of megachilid bees in Pacific archipelagosJournal of insect conservation 18 (4), 613-622

da Silva, CRB, Stevens, MI and Schwarz, MP (2016) Casteless sociality in an allodapiine bee and evolutionary losses of social hierarchies.  Insectes Sociaux, 63:67-78

da Silva, CRB, Groom, SVC, Stevens, MI and Schwarz, MP (2016) Current status of the introduced allodapine bee Braunsapis puangensis (Hymenoptera: Apidae) in FijiAustral Entomology  55: 43-48

Silva, DP, Groom, SVC, da Silva, CRB, Stevens, MI and Schwarz, MP (2016) Potential pollination maintenance by an exotic allodapine bee under climate change scenarios in the Indo-Pacific regionJournal of Applied Entomology, doi: 10.1111/jen.12337

Groom, SVC, Stevens, MI, Ramage, T and Schwarz, MP (2016) Origins and implications of apid bees in French PolynesiaEntomological Science, in press

Staines, M, Vo, C, Puiu, N, Hayes, S, Tuiwawa, M, Stevens, M I and Schwarz, MP (2017). Pollen larceny of the tropical weed Solanum torvum by a Fijian endemic halictine bee with implications for the spread of plalnts with specialized pollinator requirements. Journal of Tropical Ecology, doi: 10.1017/S0266467417000098

Where to next?

Before our APSF-funded research the understanding of Fijian bees was limited to a handful of now very old papers.  Only four endemic Fijian bees had been reported and nothing was known of their biology.  We now have a much more detailed picture of bees in Fiji.  We now know that there are many introduced species and that the number of endemic species is dramatically larger than had been anticipated.  We also know that endemic species are supergeneralists and likely to greatly augment the spread of exotic weeds, even if those weeds have specialized pollinators that are not present in Fiji. At the same time we know that some introduced bees are unlikely to enhance the spread of weedy exotics, and another species may provide fall-back pollination services for crops if managed honeybees decline with the future introduction of honeybee pathogens.  There are also good prospects for using endemic bees to increase crop pollination by manipulating nesting site conditions.

We suggest the following topics as areas that are important for future research:

  1. The recent multiple speciation events involving highland endemic bees is likely to yield important insights into the role of past climate change as a river of island biodiversity.  Our latest results suggest that such biodiversity may result from an interaction between climate specialization and the effects of Wolbachia as a promoter of pre-mating isolating mechanisms
  2. Endemic Fijian bees hold enormous promise as species managed for crop pollination at both village and broadscale levels, and are likely to be much more effective pollinators than the introduced honeybee, and are not vulnerable to honeybee pathogens that have decimated honeybee populations in other regions of the globe.  Study is required to measure their impacts on crop yields.
  3. Endemic Fijian bees are super-generalists and this is likely to enhance the spread of exotic weedy plants..  This vulnerability to exotic plants indicates a need to more effectively guard against further accidental plant introductions.
Opportunities for collaboration

We welcome all opportunities for collaborating on research into the biodiversity, conservation, and pollination biology of south western Pacific terrestrial biota.  If you are interested please contact Mike Schwarz ( or any other members of our team.

Captions for Figures

Figure 1. The endemic Fijian bee Lasioglossum (Homalictus) hadrander.  This bee is one of several species that are restricted to only the highest plateaus in Fiji and will be vulnerable to any global warming.

Figure 2. Team member Ella Deans collecting bees from the introduced weed ‘Wedelia’ (Sphagneticola trilobata).  Wedelia is a major environmental weed and very frequently visited by the exotic bee species Braunsapis puangensis.  Exotic bees in Fiji have the potential to increase the abundance of weed species.

Figure 3. Current distribution of the exotic bee species Braunsapis puangensis and Ceratina dentipes in Fiji.  These two species have been able to disperse into nearly all ecosystems in Fiji and have become widespread in other Pacific islands.  They represent a challenge for understanding how exotic pollinators affect native ecosystems, but may also help boost pollination of agricultural crops.

Figure 4. Team members (from left to right) Mark Stevens, Carmen da Silva, Mike Schwarz, Celina Rebola and Ella Deans at the high-elevation Nadarivatu field site used to study nesting biology of Fijian endemic bees and invasiveness of exotic bee species.  Endemic Fijian bees commonly nest in areas where natural or man-made events lead to exposed soil.

Figure 5. Collecting bees in the Koroyanitu highlands.  Our expedition was the first to sample bees from this National Heritage Park, and revealed multiple new bee species, all vulnerable to climate change.

Figure 6. Solanum torvum is a major introduced weed species in Fiji that is toxic to stock.  It normally requires specialist ‘buzz pollinators’, which are absent from Fiji.  However, one Fijian endemic bee is able to pollinate Solanum via ‘pollen larceny’, and this may explain why Solanum has been able to become so widespread in Fiji.

Figure 7. Face-view of an undescribed Fijian Homalictus bee. Most Fijian Homalictus species have striking metallic colours, but they are very small.  (Figure courtesy of James Dorey who is researching Fijian bee speciation and is also a macrophotographer)

Figure 8. (Left) Surface area of land in Fiji at present (black regions) compared to the Last Glacial Maximum (grey regions) when sea levels were lower by ~150m (adapted from Ash 1992). (Right) Plot of population size of Homalictus fijiensis showing a dramatic increase in effective population size since the Last Glacial Maximum, with greyed region indicating 95% HPDs.  This change in population size is inconsistent with the huge reduction in land area of Fiji, but it is consistent with adaptation to a warming climate since the LGM.

Figure 9. Vehicle problems have been a constant trouble for our work in the Fijian highlands where roads can be very rough and steep, even for large 4WDs.


Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.