Fruit flies of the tribe Dacini (Tephritidae) include many agricultural pests, but also crucial pollinators of orchids and other plants. Surveys for Dacini fruit flies were conducted using methyl eugenol, cue lure, and zingerone as male attractants across Sabah, Borneo, Malaysia in 2018 and 2019, in habitats ranging from primary forest in highly protected Conservation Areas, selectively logged forest, secondary forest, and highly disturbed sites. Because 2019 was a mast year with mass fruiting, our surveys of that year collected more than 33,000 flies, compared to just more than 500 flies in 2018 – with similar trapping efforts. Our work adds 46 species to the 43 previously known from Borneo, bringing the total for the island to 89. Three new species are described: Bactrocera (Bactrocera) melanobivittata Doorenweerd, sp. nov., Dacus (Mellesis) danumensis Doorenweerd, sp. nov., and Zeugodacus (Zeugodacus) cataracta Doorenweerd, sp. nov. The new species are only found in conservation areas; B. melanobivittata is attracted to methyl eugenol and D. danumensis and Z. cataracta are attracted to zingerone. A discussion on how biogeographic affinities of the species in the checklist support a strong biogeographic boundary across Wallacea is provided, significant records are highlighted, and the relevance of these fly species for protecting agriculture as well as native ecosystems is discussed.

Bactrocera, biodiversity, biogeography, Dacus, Oriental fruit fly, pest, taxonomy, Zeugodacus

With a surface area of 743,329 km², Borneo is the third-largest island in the world, behind Greenland and just shy of New Guinea, and was formed through a complex and dynamic geological history heavily influenced by climatological change (Morley 2000). What is now the island of Borneo was part of Sundaland on the Sunda Shelf continental plate up to around 45 Ma, when a warming planet and sea level rise mostly isolated the land mass between 45–25 Ma, only narrowly connected to mainland Asia (Wilson and Moss 1999; De Bruyn et al. 2014). During this period, large parts of what is now the southern part of Borneo were submerged and formed shallow reefs (De Bruyn et al. 2014). In the past 25 Ma, tectonic plate movement in the region has caused an uplift of the entirety of Borneo and included the formation of one of the tallest peaks in Southeast Asia at 4,095 m; Mount Kinabalu near the northern tip of the island (Cottam et al. 2013). This dynamic formation history and intermittent connections with other landmasses gave rise to a diverse array of habitats and the evolution of complex ecosystems, and the present-day islands that were part of the historical region of Sundaland are jointly recognized as one of 25 global biodiversity hotspots with high levels of endemicity (Myers et al. 2000; De Bruyn et al. 2014; Merckx et al. 2015; Neo et al. 2020).

Borneo has not been immune to the consequences of global and regional anthropogenic development. During the last century, extensive habitat loss has been well-documented for plants, birds, and mammals (Scriven et al. 2015; Ocampo-Peñuela et al. 2020). Presently, large areas have been converted to agriculture, primarily palm oil plantations, but there is a patchwork of selectively logged and partially conserved secondary forest that remains (De Bruyn et al. 2014; Chung et al. 2020). Arguably, two of the best-conserved areas of remaining primary tropical rainforest are the Danum Valley (438 km²) and Maliau Basin (390 km²) in the Malaysian State of Sabah, which have been designated as Conservation Areas since 1984 (Marsh and Greer 1992). Such areas now form crucial refugia on which many species rely for their persistence and these parks offer fleeting opportunities for science to study species before they become extinct. All groups of life on Borneo are still understudied, and dozens of new species of, for example, fish (Hanafi et al. 2022), plants (Rodda and Rahayu 2022), snails (Marzuki et al. 2021), and insects (Tan et al. 2021) are still being discovered and described.

There are more than 1,000 species in the (sub)tropical fruit fly tribe Dacini (Tephritidae: Dacinae) in four genera: Bactrocera, Zeugodacus, Dacus, and Monacrostichus (Doorenweerd et al. 2018). The tribe is mostly known for the threat some representatives pose to global agriculture (White and Elson-Harris 1992; Ekesi et al. 2016). However, many of the species in this tribe are also crucial pollinators, particularly for the most diverse orchid genus on the planet, Bulbophyllum (Nakahira et al. 2018; Katte et al. 2020; Thawara et al. 2024). The genera Bactrocera and Zeugodacus are particularly speciose in Southeast Asia and Oceania, with a center of diversity in New Guinea (Drew and Romig 2022). Chua (1998) provided an annotated checklist of Dacini for Malaysia, including the Bornean part; Drew and Romig (2013) broadly studied Dacini in Southeast Asia including Borneo. Between these two major works and several smaller studies, 43 species had been recorded from Borneo; considerably less than the 296 species recorded from New Guinea and adjacent islands (Drew and Romig 2022).

We conducted Dacini fruit fly surveys across Sabah in Malaysian Borneo in 2018 and 2019 using three male attractants; methyl eugenol, cue lure, and zingerone, to better understand Dacini species distributions and their biogeographic connections across Asia as essential data for pest management as well as for conservation. We integrate our data with literature records from across Borneo to present an updated checklist for the island and we describe three species new to science.

With the new records from our survey, the 89-species Dacini diversity of Borneo is now similar to other areas in Asia where recent surveys have been conducted, such as Sulawesi with 83 species (Doorenweerd et al. 2020) and Vietnam with 93 species (Leblanc et al. 2018). However, it should be noted that our survey method only targets species that respond to the lures that we used, and the true diversity is likely significantly greater. The new Dacini we record for Borneo are likely not recent introductions but previously overlooked native species. Thirteen species (15%) on the checklist are endemic to Borneo. The majority of species, 68 (76%), are also found in mainland SE Asia and/or the Philippines, indicating the strong biogeographic link from the historical region of Sundaland. Eighteen species (20%) are shared with the neighboring island of Sulawesi, part of the biogeographical region of Wallacea (Doorenweerd et al. 2020). Only three species (3%) are also found east of Wallacea, crossing Lydekker’s line (Lydekker 1896) that runs just west of New Guinea: B. frauenfeldi - mango fruit fly, B. umbrosa - breadfruit fly, and B. terminaliae. The first two are common human commensals and may represent semi-recent introductions west across Lydekker’s line (Krosch et al. 2018; Doorenweerd et al. 2023b). Little is known about B. terminaliae, which was previously only recorded from New Guinea and no hosts are known (Drew 1989). COI sequence data (Doorenweerd et al. 2024b) as well as morphology suggest a very close affinity of B. terminaliae to B. quasiinfulata, which is widespread in Southeast Asia and may prove conspecific in further studies. Overall, the biogeographic affinities of the Dacini of Borneo reflect the biogeographic boundary across Wallacea, particularly Lydekker’s line. This is congruent with other studies that found a fauna largely isolated by Lydekker’s line (Michaux and White 1999; Krosch et al. 2012, 2018; Doorenweerd et al. 2020; Starkie et al. 2024). Dacini surveys of the Maluku Islands and other parts of Indonesia, which have been little studied so far, will be crucial to fill further knowledge gaps in the current and historical distribution of Dacini across this region.

Only two species were found in all five survey areas across Sabah: Bactrocera dorsalis and B. melastomatos. Bactrocera dorsalis, Oriental fruit fly, is the most economically damaging pest within Dacini due to its extreme polyphagy and invasion records spanning continents – but Borneo is considered part of its native range (Clarke et al. 2019). We found it to be the dominant species during our surveys; 40% of all specimens collected were identified as B. dorsalis based on morphology. Although this may include some misidentifications as members of the B. dorsalis complex cannot all be separated reliably based on morphology alone (Doorenweerd et al. 2023a), our results suggest that B. dorsalis, even in its natural environment, is a dominant species in Borneo ecosystems. Bactrocera borneoensis and B. incognita are two recently described species that were previously confused with B. dorsalis (Doorenweerd et al. 2023a) and have undoubtedly been confused in our counts of B. dorsalis. It is unclear whether they are pests of commercial crops. We only found B. borneoensis – confirmed with genomic data (San Jose et al. 2023) – commonly in the more natural areas of Danum Valley, Maliau Basin, and Tenompok Forest Reserve, suggesting it is native to Borneo and is not a pest species. Bactrocera incognita was only revealed with genomic data (Doorenweerd et al. 2023a) on Mount Silam; although it remains unclear if it is native to Borneo, its low abundance does not suggest it is currently a pest of commercial concern. Bactrocera melastomatos is genetically indistinguishable from B. rubigina and B. osbeckiae, both based on COI (Doorenweerd et al. 2024b) and also from whole genome data (Dupuis et al. 2018). The three are distinguished by the extent of black markings on the scutum, with B. rubigina completely or almost completely red, B. osbeckiae with ~ 50% red and ~ 50% black, and B. melastomatos mostly black (Drew and Romig 2013). This small group of species is common throughout Southeast Asia and requires further taxonomic study, but there are no records of any of them being economically damaging.

Although our surveys were designed to collect as many species as possible and not to compare diversity or abundance between sites, there are some notable trends in our results. The Maliau Basin was the most diverse area during our surveys with 45 species, followed by the Sepilok Forest Reserve with 28 species and Danum Valley with 27 species. The Tenompok Forest Reserve, at the base of Mount Kinabalu, only had five species but at 1,265–1,385 meters in elevation this cooler area was expected to be less diverse for a (sub)tropical group of flies. The large difference in diversity between the two most undisturbed areas that we surveyed, Maliau Basin and Danum Valley, is likely due to the climatological events that occurred between the two survey years and may not reflect habitat quality. In 2018, when we visited Danum Valley our surveys yielded a total of 502 flies, whereas the 2019 surveys, when Maliau Basin was visited, yielded 33,403 flies. The trapping efforts between both years were similar; this large difference in fly abundance is possibly a consequence of 2019 being a mast year. Mast years in Borneo occur every 4–10 years after a period of drought and result in > 80% of all the angiosperms flowering simultaneously, followed by a super abundance of fruit (Kettle et al. 2010), and evidently, extremely high fruit fly densities. Little is known about the importance of mast years for the functioning of tropical ecosystems (Appanah 1993), but our results suggest that they have massive impacts on insect population dynamics that warrant further study.

Two polyphagous pest species were not as widespread as might have been expected: B. frauenfeldi and Z. cucurbitae. Bactrocera frauenfeldi, the mango fruit fly, is a common pest that is widespread across New Guinea and surrounding islands, the Solomon Islands, and northern Australia. Borneo is the (present) northwestern limit of its distribution. Zeugodacus cucurbitae, melon fly, is a common pest of cucurbits and other fruits and is believed to be native to Southeast Asia but has invaded Africa and several Pacific Islands (De Meyer et al. 2015). In our surveys, B. frauenfeldi and Z. cucurbitae were only found in Sandakan Rainforest Park and the Sepilok area and possibly represent recent introductions that have not spread outside this region and could potentially be controlled to prevent further spread. Species not native to Borneo have the potential to become invasive causing agricultural damage and, in natural forests, disrupting both native orchid pollination as well as fruit production and dispersal. However, we cannot rule out that B. frauenfeldi and Z. cucurbitae may already be more prevalent in agricultural areas across Borneo, as our sampling mostly focused on less disturbed regions. Another polyphagous pest species, Z. tau, was only found in the Danum Valley and Maliau Basin Conservation Areas; suggesting that Borneo is (part of) its native range. It is widespread in Southeast Asia and causes significant economic losses of particularly cucurbits (Liu and Ji 2024), but we did not encounter it outside the natural areas or in numbers comparable to pests like B. dorsalis and B. carambolae.

Continued Dacini surveys across Borneo will undoubtedly bring more species records and better resolution to current species’ distribution and movement over time. Such information provides essential background for pest management in agriculture as well as management of natural resources. Although more labor intensive, rearing fruit flies from collected fruit in particular would be invaluable in providing new host records, and recording fly species that are not attracted to any of the attractants used during surveys. Further surveys of Sarawak, Kalimantan, and Brunei are likely to yield more undescribed species and new island records as these areas have not been surveyed extensively in recent years, and never with zingerone as a male attractant. Undescribed species can be expected particularly in undisturbed areas; the three newly described species from our study were only found in the nature preserves of Maliau Basin and Danum Valley.

We thank the Sabah Biodiversity Centre for providing collection permits [Ref. JKM/MBS.1000-2/3 JLD.3 (88); JKM/MBS.1000-2/2 JLD.7 (109)], and export permits that made this study possible. The CCF of Sabah Forestry Department, the Director of Sabah Parks and Yayasan Sabah are acknowledged for their support. We would like to thank Michael San Jose and Daniel Nitta for helping to prepare for the surveys. Unforgotten, Suzika Juiling (internship student) and Ian Brandon (FRC), who kindly volunteered during the fieldwork in Sepilok. We thank Ms. Eyen Khoo for donating some of the tubes for sampling. We thank Stadia Maps for making the Stamen-Terrain tiles available for non-commercial academic use. We thank an anonymous reviewer for careful comments that helped to improve the manuscript.