Insects in the family Drosophilidae, commonly known as vinegar flies, are used as model systems for a diverse range of research fields, especially medicine (Hewitt & Whitworth 2016). More recently, there has been growing concern surrounding invasive or pest Drosophilidae species in South Africa, as several invasive Drosophilidae species have been detected in other parts of the world and can target ripe soft-bodied fruit, resulting in economic losses for the agriculture sector (Farnsworth et al. 2017). Consequently, current species composition, and spatial and temporal patterns of Drosophilidae biodiversity are important baseline information. South African Drosophilidae are a case in point: there is surprisingly poor current knowledge of Drosophilidae species diversity, geographic range extent of common or rare species or how dynamic population abundances are in this group (McEvey, Potts & Rogers 1988; Tsacas 1990). Furthermore, what little information is available is likely outdated or not necessarily representative of the different regions of the country. The most recent, scattered records show 76 species in the country documented between 1900 and 2013 in an unpublished document updated in 2016 and provided by Dr Shane McEvey (Australian Museum Research Institute), while the most recent accessible published data indicate 70 known species from South Africa (Tsacas 1990). From the unpublished document, more recent records include D. immigrans (2010), D. simulans (2013), D. punctatonervosa (2013) and Scaptomyza oxyphallus (2013). Although an invasive drosophilid species, D. flavohirta (1983), has been recorded in South Africa previously, no recent information regarding this species’ abundance or its potential impacts on native biodiversity is available (Tsacas 1990).

Drosophilidae species were sampled sporadically in two urban-agricultural regions of South Africa using fruit-filled bucket traps (Table 1) for a different project, but also aimed at generating biodiversity knowledge. Species were identified using the Universal cytochrome oxidase I (COI) primers (LCO1490 and HCO2198; Folmer et al. 1994) by InqabaBiotech^TM and the South African Sugarcane Research Institute’s (SASRI) biotechnology department. PCR conditions were an initial denaturation of 5 min at 94°C, followed by 35 cycles of 30 s denaturation at 94°C, 30 s annealing at 50°C and 60 s of extension at 68°C, and a final elongation at 68°C for 10 min. Sequences were aligned and edited in BioEdit (Hall 1999) and closest sequence matches obtained using BLAST (Altschul et al. 1997) in NCBI. Morphology of some species was confirmed against relevant species identification keys (Yassin & David 2010). Positive matches were downloaded to include in a phylogenetic tree to confirm identification based on both neighbour-joining (NJ) and maximum likelihood (ML) methods using MEGA X (Kumar et al. 2018). The resulting trees were similar, and therefore we only show the tree based on ML (Figure 1).

Once species were identified, it was evident that some species had no prior published records regarding their occurrence in South Africa (Tsacas 1990). TaxoDros was also consulted and showed no records for some species in South Africa. This was the case for three species, namely Drosophila ananassae (Figure 2a), D. nasuta (no image available) and Zaprionus taronus (Figure 2h). Zaprionus taronus has never before been reported from South Africa (McEvey, pers. comm.); however, this species is surrounded by ongoing taxonomic uncertainty. Zaprionus taronus is probably not invasive and more likely overlooked or previously mis-identified as it has a confirmed Afrotropical distribution (Yassin & David 2010). The expertise of the person doing the identification, how the determination was made and what material was used for the determination are key information that needs to be established. Prigent, Suwalski and Veuille (2017) and Yassin et al. (2010) have highlighted the importance of sequence data to establish species delineation in the Drosophilidae.

Zaprionus taronus was recorded in a home garden in Stellenbosch, Western Cape province, South Africa, while the other two species were sampled near buildings at the South African Sugarcane Research Institute in Mount Edgecombe, KwaZulu-Natal province. We are unable to determine whether these species should be classified as alien or invasive as there is limited data covering Drosophilidae species’ historical native distributions in the region. The current distribution of D. ananassae is cosmopolitan, while D. nasuta largely occurs in the Afrotropical region and south Asia (Brake & Bächli 2008 cited in Vilela & Goñi 2015). The species collected were bycatch from another research study, and thus the bait variety was quite restrictive and sample sites selective; therefore, this may be the reason for the exclusion of other more commonly reported South African drosophilids (e.g. Z. capensis). A single sampling site has been provided per species as the primary project only required a single site per species.

Although there are 70 Drosophilidae species known from South Africa (Tsacas 1990), little is known about the current composition of species nor how this varies across space and through time. From our limited effort of a small number of sites with few traps, three of the 13 species we identified were found to be first records for South Africa. This highlights the importance of surveying Drosophilidae in various biomes, and across seasons or climatic conditions, and using diverse baiting methods to capture the full range of species diversity in the family. With improved knowledge of species composition, species interactions and natural diets (land/habitat use) and other ecological functions, the potential impacts of invasive species in this group can be better understood.