Three new exotic species of ants are recorded for the Macaronesian island of Madeira: Cardiocondyla obscurior Wheeler, 1929, Lepisiota capensis (Mayr, 1862) and Pheidole navigans Forel, 1901. The well-known invasive Pheidole megacephala is now rare on Madeira, possibly because of competition with P. navigans. We did not come across the Argentine ant – Linepithema humile, which was once very common in the Funchal area. Despite extensive searching, the presumed endemic Temnothorax wollastoni (Donisthorpe, 1940) remains unknown other than from the type material. Presence, abundance and extinction of invasive and exotic ants are dynamic processes and we stress the importance of continuous sampling.

Cardiocondyla obscurior, exotic, invasive, Lepisiota capensis, Madeira, myrmecofauna, Pheidole navigans

Oceanic islands tend to have depauperate ant faunas and they are often notable for their assemblages of exotic and invasive species, which impact island faunas disproportionately (Williams 1994; Wetterer 1997; Rizali et al. 2010). The island of Madeira’s ant fauna has been well studied and it stands out for its history of observations on invasive species (Wetterer et al. 2006, 2007).

Madeira occupies some 737 km² and is situated in the North Atlantic Ocean (approx. 32°44'N, 17°00'W at its centre), west of Morocco and north of the Canary Islands. An Autonomous Region of Portugal, it forms part of the volcanic Macaronesian group of archipelagos, along with the Azores, Selvagens, Canary Islands and Cape Verde. The climate in Madeira is subtropical with a range of bioclimates, the northwestern part of the island being the wettest and the southeast the driest. The steep, northern half of the island hosts native laurel ‘laurisilva’ forests, which are humid, lush and green throughout the year.

Following the Portuguese settlement in the early 1400s, Madeira quickly became an important centre for commerce, particularly between Europe and the New World, with new plants and animals from across the globe imported to Madeira prior to distribution to mainland Portugal and the rest of Europe (Wetterer et al. 2007). The history of myrmecology in Madeira is summarised in Wetterer (2006) and Wetterer et al. (2006). Two highly destructive ant species, the big-headed ant Pheidole megacephala (Fabricius, 1793) and Argentine ant Linepithema humile (Mayr, 1868) – both identified among 100 of the world’s worst invasive alien species by the IUCN Invasive Species Group (Lowes et al. 2000) – were first recorded as exotics from Madeira, and they underwent apparent population explosions here in the 1850s and 1890s respectively (Heer 1852; Schmitz 1896; Wetterer et al. 2006, 2007). Madeira has since been synonymous with invasive ants, although Wetterer et al. (2006) subsequently found their distributions to be limited on the island, with little penetration of native habitats.

28 species of ant have so far been recorded from Madeira, of which ten are presumed native and one endemic (Wetterer et al. 2006, 2007). The last work on the Madeiran myrmecofauna was published by Wetterer et al. (2007), who recorded 23 species of ants including ten new to the island. Most of the material for that publication was collected in 2002. Therefore, it has been some 20 years since a survey was last conducted in Madeira and new invasive species may be expected. It is also a useful point at which to compare the current myrmecofauna to past observations. Here, we report three new exotic species for the island’s fauna. We also compare our observations of well-known invasive ants to those of previous authors. Finally, we comment on Temnothorax wollastoni (Donisthorpe, 1940), which remains unlocated since it was described.

Trips to Madeira were conducted by the authors in 2014 and 2021. During 2014, ants were collected sparingly from seven sites by KB. These included minors of a small Pheidole sp. which were collected from two locations in Funchal. Without majors, the species was impossible to identify but it appeared to belong to the flavens-group.

In 2021, ants were collected intensively at 22 sites by RG, from various locations around Madeira (Table 1 for all locations, 2014–2021). Methods focussed on direct manual searching, leaf litter reduction using a Winkler bag sieve, and baiting using tinned tuna (protein). Sampling was not systematic: sites were selected by combining visits to sites already sampled by previous researchers with opportunistic sampling of favourable sites holding a representative range of native habitats.

One of the primary goals of this trip was to collect majors of the unidentified Pheidole. Bait was used to attract majors of the unknown species of the flavens-group. We also searched for the presumed Madeiran endemic Temnothorax wollastoni (Donisthorpe, 1940), which hasn’t been recorded again since its discovery by Wollaston (BM 1858-21), but which is thought unlikely to have disappeared (Wetterer et al. 2006, 2007). The most similar species to T. wollastoni is Temnothorax gaetulus from northwest Africa (Santschi 1923) (Wetterer et al. 2007), which belongs to the exilis-group of species (Cagniant and Espadaler 1997). Temnothorax gaetulus is found in mountain steppe habitat with spiny, xerophytic vegetation (Cagniant and Espadaler 1997). We thus thought that T. wollastoni would likely be a ground-dwelling inhabitant of steppe or xeric habitat, but we used all methods that we typically employ to search for Temnothorax: searching on the ground, under stones, between cracks in rocks, under moss, on vegetation, in leaf litter, within hollow twigs on the ground or on shrubs and trees.

Ants were identified using all available resources (e.g., Antweb images http://www.antweb.org, Wilson (2003) and Sarnat et al. (2015) for the Pheidole, and comparison of ants in the personal collection of the first author for Cardiocondyla and Lepisiota). All material is leg. and det. R. Guillem, unless otherwise stated. All material collected is housed in the private collection of R. Guillem, which is held at the Gibraltar Botanic Gardens.

Sites and species recorded are provided in Tables 1, 2. An up-to-date checklist of all species recorded from Madeira so far is also provided in Table 3. New species to the island are discussed below.

Cardiocondyla obscurior Wheeler, 1929 (Fig. 1)

4 sites: 6. Madeira Botanic Gardens (23 Jun 2021), 2 workers on a wall; 10. Funchal Promenade (26 Jun 2021), 11 workers on ground; 11. Jardim do Almirante Reis (7 Sep 2014), 24 workers in lawn and planters; 18. Machico centre (22 Jun 2021), 5 workers in planters outside church.

The species was originally described from Thailand by Wheeler (1929) as Cardiocondyla wroughtonii var. obscurior and was raised to species level by Seifert (2003). Donisthorpe (1930) had also described Cardiocondyla bicolor from Israel but this was later synonymised with C. obscurior (Seifert 2003). It was first recorded for Macaronesia from Puerto de la Cruz, Tenerife in the Canary Islands in 1999 (Seifert 2003), where it was also recorded by Staab (2009), and very recently from Cape Verde (Wetterer and Espadaler 2021). However, Wetterer and Espadaler (2021) provide information on a species of Cardiocondyla reported by Collingwood and Van Harten (1993) from Cape Verde, which differ from Cardiocondyla emeryi, and have ‘tentatively assumed’ that these correspond to C. obscurior. Given this information, it is possible that C. obscurior has been present in Macaronesia for longer than initially thought.

In Europe, this species has been recorded from Germany (Seifert 2003), the Netherlands (Boer et al. 2018) and France (Blatrix et al. 2018), where it is an indoor exotic in greenhouses, as well as Spain where it is common in some urban environments, nesting outdoors (Valencia: Sánchez-García and Espadaler 2015; Alicante: Trigos-Peral and Reyes-López 2016; Barcelona: Espadaler and Ortiz de Zugasti 2019).

Material was compared to specimens from Gibraltar where it nests outdoors in irrigated, urban environments. This species has been found nesting in the calyx of pomegranates imported to the UK and Gibraltar from Spain (RG pers. obs). In all cases, whole nests have been found with queens, brood and males, demonstrating the ease with which a species can be transported to other countries.

Figure 1. 

Cardiocondyla obscurior from Funchal, Madeira, profile.

Lepisiota capensis (Mayr, 1862) (Fig. 2)

Site 6. Madeira Botanic Gardens (23 Jun 2021), 4 workers on paths.

Lepisiota capensis is an African species that is extremely common in South Africa (RG pers. obs.). It has been recorded from the Canary Islands (Espadaler and Fernández 2014; Schifani et al. 2018). Collingwood and Van Harten (1993) also report L. capensis from Cape Verde, but according to Wetterer and Espadaler (2021) this is likely to be a misidentification of Lepisiota canescens. A subsequent record from Cape Verde by Báez et al. (2005) may perhaps also be a misidentification.

Its arrival in the Canaries is considered likely to be recent (Espadaler and Fernández 2014) so it is probable that this is also a recent arrival to Madeira. Lepisiota frauenfeldi cfr. kantarensis Forel, 1911 was also reported for the first time from Fuerteventura and Tenerife recently (Schifani et al. 2018) and this too may arrive in Madeira in due course.

Specimens were compared to the authors’ material from South Africa. The petiolar spines in the Madeira specimens are longer than specimens from the Cape in South Africa but match the specimen illustrated in Espadaler and Fernández (2014) from La Gomera. The capensis-group of species is in desperate need of a revision and probably consists of several species (Hita Garcia et al. 2013).

Figure 2. 

Lepisiota capensis from Madeira Botanic Gardens, profile.

Pheidole navigans Forel, 1901 (Fig. 3A–D)

Funchal, 6 sites: 5. Pestana Carlton Hotel (5 Sep 2014), 12 minors in planter; 6 Madeira Botanic Garden (23 Jun 2021), 15 minors, 5 majors, 1 alate queen, under stone; 8. Parque Santa Catarina (7 Sep 2014), 1 minor in lawn; 8. Parque Santa Catarina (22 Jun 2021), 157 minors, 15 majors in a planter, baited; 9. Funchal Promenade (26 Jun 2021), 170 minors, 24 majors, in lawn, baited; 11. Jardim do Almirante Reis (23 Jun 2021), 7 minors in lawn; 12. Jardim Municipal do Funchal (23 Jun 2021), 19 minors, 2 majors, in lawn.

This species of Pheidole was first collected in Funchal in 2014. It was identified as a member of the flavens-complex, but couldn’t be identified to species level due to the lack of majors. Two Pheidole species from the flavens-complex were recently recorded from Tenerife in the Canary Islands: P. navigans and Pheidole bilimeki Mayr, 1870 (Hernández-Teixidor et al. 2020). We thus suspected that the Madeira specimens belonged to one of these species.

We used keys in Wilson (2003) and Sarnat et al. (2015) to identify the specimens as belonging to the Pheidole flavens-complex. Specimens were then identified to species-level using Sarnat et al. (2015), who restored P. navigans from synonymy with Pheidole flavens Roger, 1863 and elevated it to species rank. Our specimens match the descriptions of P. navigans provided. Major workers can be separated from those of P. flavens by the combination of predominantly longitudinal rugae on the posterolateral lobes, the more distinct and narrower antennal scrobe bordered mesially by strong, unbroken frontal carina, and the more continuously glossy scrobe depression (Sarnat et al. 2015) (Fig. 3B).

Figure 3. 

Pheidole navigans from Funchal, Madeira A major worker, profile B major worker, head C minor worker, profile D alate queen, profile.

This species is very common around Funchal, especially in lawns and the soil of planters. A nest was found under a small stone. Where it was present, it was the fastest ant to respond to the protein baits, followed by Lasius grandis. This species is thought to be South American in origin and has been introduced into the Palaearctic and Nearctic regions (Sarnat et al. 2015). It was originally described from quarantined specimens entering Germany from Mexico.

We recorded twenty species from Madeira, with three new records for the island – Cardiocondyla obscurior, Lepisiota capensis and Pheidole navigans – elevating the number of species recorded for Madeira to 31. All three species are exotics. Furthermore, the three have been recorded from the nearby Canary Islands (Seifert 2003; Staab 2009; Espadaler and Fernández 2014; Hernández-Teixidor et al. 2020) and therefore, their presence is not unexpected. Pheidole navigans is also reported to be spreading across the Southeastern United States (Sarnat et al. 2015) and the records from Macaronesia represent part of a more general expansion of the species.

We are uncertain whether these new exotic species will have an impact on other ant species, but the two best-known invasive ants certainly seem to be declining. We were unable to locate Pheidole megacephala in central Funchal. We visited some of the sites in the city reported for this species by Wetterer et al. (2007), including along the waterfront in the centre where they found it to be very common, but we did not find it. Furthermore, Wetterer et al. (2007) found Linepithema humile to be fairly uncommon in Madeira, except on the dry eastern tip of the island. Remarkably, we failed to record this species altogether.

We hypothesise that P. navigans could be displacing P. megacephala in the Funchal area, where the former now seems to be one of the most dominant ants and the latter has become difficult to find or disappeared altogether. Certainly, P. navigans responds extremely quickly to bait, recruiting many minor and major workers with remarkable speed. We only located P. navigans in the Funchal area and it would be of benefit to chart its expansion.

The most widespread and frequently encountered ant throughout Madeira continues to be Lasius grandis (Wetterer et al. 2007), a native species that dominates in the native laurisilva and other natural habitats but is also common in anthropogenic habitat. We share the view of Wetterer et al. (2006) that P. megacephala and L. humile are unlikely to have impacted on native ants significantly even at their peak and we are thus surprised that Temnothorax wallastoni remains unlocated, despite different surveys during the 21^st Century and targeted searches for Temnothorax. This species has only been found once and the description is based on two workers from ‘Madeira’. It seems unlikely that such a species would become extinct. Although it is recognized that Wollaston collected extensively on the island, the possibility that the specimen may have been collected elsewhere and labelled ‘Madeira’ in error cannot be discounted. We recommend that any future surveys for this ant on Madeira include extensive pitfall trapping in native habitats.

Regarding the old records of Camponotus sylvaticus and Messor structor (Table 3), we agree with Wetterer et al. (2007) that these species were sporadic introductions that have long since become extinct in Madeira. Wetterer et al (2006) argue that the record of C. sylvaticus was probably a misidentification of another exotic Camponotus species, imported to Madeira in infested timber, whereas the European native C. sylvaticus is a soil-dwelling species. It is interesting to note that Tetramorium bicarinatum was last recorded by Schmitz (1897) (as Tetramorium guineense). Having failed to record it, Wetterer et al. (2007) suggested that this tramp species may no longer be present in Madeira. It is thus possible that the species became extinct and has recently been re-introduced. However, Schmitz’s specimens were collected from Machico, where we also found it, so it is also possible that a population has remained there since its initial discovery.

New exotic species continue to arrive in Madeira and some of these could become invasive. We recommend regular surveying to intercept new arrivals and help elucidate the dynamics of interactions between invasive ants.

This work was carried out under the license N° 09/IFCN/2021 – FAU MAD and the declaration N° 07/IFCN/2021 – FAU MAD issued by the Instituto das Florestas e Conservação da Natureza, Madeira. We thank Albert Gonzalez (Gibraltar Botanic Gardens) for photographing our specimens. Dr Peter Stüben kindly advised on native habitats for sampling on the island.