The Avifauna of the Gulf of Guinea Oceanic Islands

Abstract

Although birds have always been one of the best-known taxa on the Gulf of Guinea oceanic islands, our understanding of their ecology and evolution has increased substantially in the last two decades. Intensive field-based surveys have allowed the first detailed island-wide distribution maps for most species and a much better grasp of habitat associations, highlighting the importance of native forests for many of the endemic birds. Molecular data have provided important insights into evolutionary history, leading to an extensive revision of the taxonomy of the islands’ endemic avifauna. Most speciation events are much more recent than the age of the islands, indicating a high species turn-over that is likely explained by the islands’ history of intense volcanic activity and their moderate distances to the mainland. These islands have the highest accumulation of endemic bird species for small oceanic islands: at least 29 endemic species occur in three islands with a total area of just over 1000 km². This may be explained by their particular geographic location: offshore from a species-rich continent at distances that allowed the colonization and evolution in isolation of many distinct lineages. All these contributions are now being used to ensure bird conservation, through updated species conservation status and species action plans for the most threatened species, and also to promote the conservation of the native forests on which most of the endemic birds depend.

Birds are one of the animal groups with which people are most familiar. This is especially true in the oceanic islands of the Gulf of Guinea where the avifauna has a prominent position in popular culture. The islands’ birds have also been a preferential target of scientific investigation since the colonial period, in part due to the exceptionally high number of endemic species. This richness has played an important role in bringing global attention to the unique biota of these islands. The state of ornithological knowledge on these islands was summarized at the start of the present century (Jones and Tye 2006), but since then a wealth of new information has been gathered by an increasing number of studies. Here we provide an updated overview of the diversity, ecology, and evolution of this remarkable avifauna.

We follow the taxonomy and nomenclature of Clements et al. (2021), which is identical to the most recent checklist of the birds of the region (de Lima and Melo 2021). Taxonomic authorities are presented in the checklist (Appendix).

History of Ornithological Research

The avifauna of the Gulf of Guinea islands was little known before the nineteenth century. These relatively isolated oceanic islands were logistically difficult to access, with rugged, inhospitable interiors, so that progress was slow and sporadic, with decades-long gaps between collections. By the end of the nineteenth century, the avifauna was well documented, although important collections made at various times in the twentieth century further contributed to recognizing the importance of the islands’ unusually high endemism. Throughout this period, it became clear that several endemic taxa were extremely rare and their forest habitats vulnerable to development, prompting conservation concerns in recent decades. Easier access to the islands and better logistical support have made it possible to run longer-term research and conservation projects, and the advent of modern molecular phylogenetics has enabled significant insights into the evolutionary history of the endemic bird community.

Apart from a few specimens collected earlier, the avifauna remained little known until the 1840s, when the islands were used as a base for the British Expedition to the River Niger. Expedition members collected several new species on São Tomé and Annobón but visited Príncipe only briefly and did not collect any birds there (Allen and Thomson 1848). Considerable confusion surrounded the provenance and identity of these specimens, however, due in part to the repeated loading and unloading of cargoes between ships and between islands. Other misidentifications resulted simply because African bird taxonomy was then in its infancy.

The most important early collections on Príncipe and São Tomé were made between 1847 and 1850 for the museums of Hamburg and Bremen by Carl Weiss, most of whose specimens were described in Hamburg by Gustav Hartlaub (1850, 1857), but confusion surrounded many Príncipe specimens that were wrongly ascribed to the African continent. These confusions were eventually resolved with the help of Heinrich Dohrn and John Gerrard Keulemans, whose collections on Príncipe between April and September 1865 were complemented by the field notes of Keulemans that proved to be a valuable source of information (Dohrn 1866; Keulemans 1866).

In 1885, Adolpho Frederico Moller obtained some birds from São Tomé while collecting plants for the Coimbra Botanic Gardens (Vieira 1887). Between 1885 and 1895, Francisco Newton assembled a comprehensive collection from all three islands for the Lisbon Museum, which were described in a long series of papers by José du Bocage (Bocage 1867, 1879, 1887a, b, 1888a–c, 1889a–c, 1891, 1893a, b, 1903, 1904). Newton was also a keen observer whose descriptions of behavior and ecology were partially published by Bocage as ancillary information. However, the original letters that Newton wrote to Bocage to accompany his specimens and almost all the specimens themselves were destroyed in the disastrous 1978 fire in the Museu Bocage in Lisbon. By the end of the nineteenth century, most landbird species had been described but an important collection was made for the museum at Genoa by Leonardo Fea between 1899 and 1901 (Salvadori 1903a–c).

Two significant collections were made in the early decades of the twentieth century. Boyd Alexander visited the main islands in 1909 collecting for the British Museum (Bannerman 1914, 1915a, b) and José Correia and his wife Virginia collected on Príncipe and São Tomé for the American Museum of Natural History in 1928–29 (Correia 1928–29; Amadon 1953). Both Alexander and the Correias kept extensive field notes on the distribution and status of the endemic bird species they collected at a time when agricultural exploitation of the islands was at its peak and some species had evidently become very rare. Correia’s valuable and entertaining diary (1928–29) remains unpublished, but his collection allowed Amadon (1953) to produce an important synthesis on the avian zoogeography of the Gulf of Guinea islands.

The next ornithological survey of Príncipe and São Tomé took place in 1949, when an Oxford University team published the first detailed field notes for most of the endemic birds (Snow 1950). Annobón has always been difficult to access because of its remoteness, so there was an even longer gap between Alexander’s 1909 expedition and subsequent visits by Aurélio Basilio in 1955 (Basilio 1957) and Hilary Fry in 1959 (Fry 1961).

A Portuguese scientific mission to Príncipe and São Tomé in 1954 collected birds and assessed the status of those that might require protection under colonial law (Frade 1958, 1959; Frade and Santos 1977). René de Naurois visited the islands several times in 1963 and 1970–73, publishing extensively on the ecology and systematics of most indigenous species, including all the endemic birds (Naurois 1972a, b, 1973a, b, 1975a, b, 1979, 1980, 1981, 1982, 1983a, b, 1984a–d, 1985, 1987a–c, 1988a–c; Naurois and Castro Antunes 1973; Naurois and Wolters 1975). He also wrote the first book to deal exclusively with the birds of all three oceanic Gulf of Guinea islands (Naurois 1994).

In June 1983, ornithologists from the Dresden Museum visited São Tomé (Günther and Feiler 1985) and again in March-April 1991, the second time as part of a multi-disciplinary expedition (Nadler 1993). Annobón belatedly received further attention with visits by Mike Harrison (1990) and Jaime Pérez del Val (2001). In 1996–97 an expedition by the University of the Azores conducted the first detailed study dedicated to seabirds (Monteiro et al. 1997).

The Red Data Book for African birds considered seven endemic bird species of these islands to be threatened (Collar and Stuart 1985), highlighting the fact that four single-island endemics on São Tomé had not been seen for over 50 years and were possibly extinct. This publication prompted conservation initiatives by the International Union for Conservation of Nature (IUCN—Jones and Tye 1988; Burlison and Jones 1988) and by the European Union. The latter acted mostly through ECOFAC, a program sponsored by the European Union promoting the conservation and sustainable use of forest ecosystems in Central Africa (Anon. 1994) that published the first field guide to the birds of São Tomé and Príncipe (Christy and Clarke 1998). Increased attention led to the rediscovery of all four “missing” bird species: São Tomé Short-tail Motacilla bocagii in 1987 (Eccles 1988), Sao Tome Ibis Bostrychia bocagei, Newton’s Fiscal Lanius newtoni in 1990 (Atkinson et al. 1994), and São Tomé Grosbeak Crithagra concolor in 1991, more than a century after it had last been seen (Sergeant et al. 1992). On Príncipe, the rare endemic Príncipe Thrush Turdus xanthorhynchus was rediscovered in 1996, after more than 50 years (Christy and Gascoigne 1996).

The number of ornithological studies has greatly increased since the turn of the twenty-first century, leading to numerous taxonomic changes and a much better understanding of this unique avifauna (Jones and Tye 2006; Melo 2007; de Lima and Melo 2021). However, there is still much to be discovered as exemplified by the existence of the Príncipe Scops-Owl, which was only confirmed in 2016, 90 years after Correia collected information from local people on its putative occurrence (Verbelen et al. 2016).

Species Diversity and Distribution

General Patterns

According to the latest assessment (de Lima and Melo 2021), the avifauna of the oceanic islands of the Gulf of Guinea comprises 146 confirmed species (Appendix). These include 66 resident species, which contain a remarkably high number of endemics: 29 (44%) species and 16 (24%) subspecies (Table 21.1). Seventeen species (26%) are possibly non-native, six are breeding migrant species (all of which are seabirds), four are regular non-breeding migrants, and 62 are vagrants. Eight species are of uncertain status, including five that may breed in the archipelago and three that have been recorded on the islands in the past. Additionally, the occurrence of 51 species remains unconfirmed, being based solely on uncorroborated observations (de Lima and Melo 2021).

Table 21.1 Numbers of bird species known from the Gulf of Guinea oceanic islands (de Lima and Melo 2021: Appendix). Percentages of endemics and possibly non-native species were calculated for the number of residents. One endemic species, the Príncipe Seedeater Crithagra rufobrunnea, has three endemic subspecies, which were not added to the tally of endemic subspecies

The most outstanding feature of the resident bird assemblage is the high level of endemism (Fig. 21.1). Almost all endemic species are single-island endemics, except for the Sao Tome Pigeon Columba malherbii, which occurs on all three islands, and the Sao Tome Spinetail Zoonavena thomensis and Príncipe Seedeater Crithagra rufobrunnea, both of which occur on Príncipe and São Tomé (Table 21.2). Endemic subspecies are also mostly restricted to a single island with two exceptions. The Little Swift Apus affinis bannermani occurs on Príncipe and São Tomé and is also considered the taxon present on the neighboring land-bridge island Bioko. Likewise, the African Emerald Cuckoo Chrysococcyx cupreus insularum is considered the taxon present on all three oceanic islands in the archipelago (Table 21.3, Box 21.1). Another defining feature is the unusually high phylogenetic diversity of these oceanic islands, with resident birds representing 28 families (Appendix).

Fig. 21.1

Some of the endemic birds of the oceanic islands of the Gulf of Guinea, including four Critically Endangered species (1–4), an undescribed species that was only confirmed for the first time in 2016 (8), two giants (3, 7) and a dwarf (1), and two phenotypically “aberrant” species (5, 6): (1) São Tomé Ibis Bostrychia bocagei; (2) Newton’s Fiscal Lanius newtoni; (3) São Tomé Grosbeak Crithagra concolor; (4) Príncipe Thrush Turdus xanthorhynchus; (5) Dohrn’s Thrush-Babbler Sylvia dohrni; (6) São Tomé Short-tail Motacilla bocagii; (7) São Tomé Sunbird Dreptes thomensis; (8) Príncipe Scops-Owl Otus sp. nov. Photo credits: (1, 2, 4, 5) Lars Petersson, (3, 8) Martim Melo, (6, 7) Paul van Giersbergen

Table 21.2 Endemic bird species of the Gulf of Guinea oceanic islands: Príncipe (P), São Tomé (ST), and Annobón (A), with their respective IUCN Red List category (IUCN 2021). Taxonomy and nomenclature follow Clements et al. (2021)

Table 21.3 Endemic bird subspecies of the Gulf of Guinea oceanic islands: Príncipe (P), São Tomé (ST), and Annobón (A). Taxonomy and nomenclature follow Clements et al. (2021); taxonomic authorities given in Appendix. Notes link to taxonomic comments in Box 21.1

Non-endemic native resident species include a large proportion of aquatic species, including three Ardeidae, one Phalacrocoracidae, and one Rallidae. They also include single representatives of Accipitridae, Psittacidae, Sturnidae, Nectariniidae, and Estrildidae. Possibly non-native species include five Ploceidae, three Phasianidae, three Estrildidae, two Columbidae, and single representatives of Apodidae, Psittacidae, Viduidae, and Fringillidae.

Breeding migrants include six seabird species: four Laridae, one Phaethontidae, and one Sulidae. Apart from the White-tailed Tropicbird Phaethon lepturus, which can breed in tree cavities and cliffs on the main islands, these species breed on bare offshore islets, such as Boné de Jóquei, Tinhosas (both near Príncipe), Sete Pedras, Rolas (both near São Tomé) and Tortuga (near Annobón). There are surprisingly few regular non-breeding migrants: three coastal waders (Scolopacidae) and the Barn Swallow Hirundo rustica (Hirundinidae). This contrasts with the large number of occasional migrants, comprising 62 species belonging to 27 families (Appendix), mostly shorebirds and passerines, whose number will continue to grow as currently uncorroborated records are confirmed. These patterns reinforce the hypothesis that migrants, whether Afrotropical or Palearctic, do not regularly cross the open sea of the Gulf of Guinea (Jones and Tye 2006).

Island Accounts

Príncipe has 88 confirmed species (Table 21.1, Appendix), including 32 resident species, of which 8 are single-island endemics and 3 are shared endemics with the neighboring islands (Tables 21.1 and 21.2). Príncipe also holds 9 (28%) endemic subspecies from species occurring on the mainland, together with two endemic subspecies of the endemic Príncipe Seedeater (Table 21.2). Five species (16%) are possibly non-native: two Columbidae, two Estrildidae, and one Apodidae. Príncipe, and especially its surrounding islets, holds breeding colonies of all seabird species that reproduce in the oceanic islands of the Gulf of Guinea, except Bridled Tern Onychoprion anaethetus. Tinhosas islets are remarkable for their seabird colonies, which hold around 200,000 breeding pairs of Sooty Tern O. fuscatus, accounting for 1% of the global population of this species and meeting the criteria for Important Bird Area (Valle et al. 2016). Both Tinhosas and Boné de Jóquei also hold important but declining breeding colonies of Brown Booby Sula leucogaster.

São Tomé has 96 confirmed species (Table 21.1, Appendix), including 50 resident species, of which 17 (34%) are single-island endemics and 3 (6%) are endemics shared with the nearby islands (Tables 21.1 and 21.2). This island also holds seven (14%) endemic subspecies of species that occur in continental Africa, together with one endemic subspecies of the endemic Príncipe Seedeater (Tables 21.1 and 21.3). Seventeen species (34%) are possibly non-native: five Ploceidae, three Estrildidae, two Phasianidae, two Columbidae, two Psittacidae, and single representatives of Apodidae, Viduidae, and Fringillidae. São Tomé and its surrounding islets, notably Sete Pedras and Rolas, hold breeding colonies of three seabird species: Brown Noddy Anous stolidus, White-tailed Tropicbird, and Brown Booby.

Annobón has 30 confirmed species (Table 21.1, Appendix), including 11 resident species, of which one is a single-island endemic and one is an endemic shared with São Tomé and Príncipe (Tables 21.1 and 21.2). It has three endemic subspecies, two of which are treated as full species by some authors (Table 21.3, Box 21.1), three native non-endemic species (Cuculidae, Rallidae, and Ardeidae), and three possibly non-native species (two phasianids and one estrildid). Seabird colonies include Brown Noddy, Black Noddy Anous minutus, Bridled Tern, and White-tailed Tropicbird.

Box 21.1: Comments on Taxonomic Uncertainties

The numbers link to the species in Table 21.3. Taxonomic authorities given in Appendix

• 1. Coturnix delegorguei histrionica. Endemic subspecies from São Tomé whose validity should be re-appraised combining multiple lines of evidence, including molecular data.

• 2. The systematic position of the Lemon Dove has never been satisfactorily resolved. Currently treated as Columba larvata, it was, for a long time, placed in its own genus, Aploplelia. Its current placement within Columba remains uncertain (Pereira 2013): it groups together with the Bronze-naped Pigeon superspecies (C. malherbii, C. delegorguei, C. iriditorques) within a larger clade encompassing the Old-World Columba and most Streptopelia species, but the exact affinities with these two genera remain unresolved. The plumage of Columba larvata provides little phylogenetic information as it varies widely, both between and within populations, and it may also change with age (Amadon 1953; Serle 1959). As such, it is imperative to conduct a full taxonomic revision supported by extensive molecular sampling (Baptista et al. 2020), where many of the current arrangements are unlikely to prevail. The subspecies from São Tomé have been treated by some authorities as a distinct species, but evidence supporting it is anecdotal (Baptista et al. 2020). Molecular evidence for the São Tomé and Príncipe populations detected large differentiation in mitochondrial haplotypes, suggesting at least two colonization events from the mainland (Pereira and Melo, unpublished results). The same data also showed that these populations are closely related to those from southern Cameroon but very distinct from those from Malawi (up to 3.2 my divergence), suggesting that there may be multiple distinct species on the mainland. No molecular data are available for the Annobón population, which is currently placed under C. l. inornata, occurring from Sierra Leone to Gabon.

• 3. Treron calvus has 15 recognized subspecies, many of which might not be valid (Hoyo et al. 2020). Molecular data placed the endemic Príncipe subspecies, T. c. virescens, together with birds from Bioko, currently treated as an endemic subspecies, T. c. poensis, and revealed two cases (in 14) of mitochondrial introgression from T. sanctithomae, endemic from São Tomé, into the Príncipe population (Pereira 2013).

• 4. Chrysococcyx cupreus insularum. Endemic subspecies of the three Gulf of Guinea oceanic islands, although most authorities treat or recommend treating C. cupreus as a monotypic species (HBW and BirdLife International 2020; Clements et al. 2021; Gill et al. 2021).

• 5. Apus affinis bannermani is a subspecies considered to be restricted to the Gulf of Guinea islands of Príncipe, São Tomé, and Bioko (Clements et al. 2021; Gill et al. 2021)—so not strictly speaking an endemic of the oceanic islands. The validity of this subspecies should be re-appraised as it has been considered indistinguishable from the neighboring mainland population (Amadon 1953).

• 6. Bostrychia olivacea rothschildi. Molecular studies on museum specimens are required to determine if this extinct endemic subspecies from Príncipe is valid, and whether it was more closely related to the São Tomé B. bocagei or to the mainland species B. olivacea.

• 7. Tyto alba thomensis. Recent molecular evidence suggested that this phenotypically distinct taxon, restricted to São Tomé, may constitute a separate species (Uva et al. 2018; Alves 2019).

• 8. Otus senegalensis feae. A recent assessment, using multiple lines of evidence but no molecular data, considered the scops-owl from Annobón as a valid endemic species, Otus feae (Collar and Boesman 2020). Both molecular data and phenotypic data (Freitas 2019) place it well within the intra-specific variation of Otus senegalensis, which indicates a recent colonization. The Annobón population is now very likely isolated from O. senegalensis, which is absent from the neighboring mainland. A deep phylogeographic study of O. senegalensis sensu lato is needed to understand its evolutionary history and clarify the taxonomic status of the Annobón population. In any case, several authorities already recognize it as a valid endemic species (e.g., HBW and BirdLife International 2020; Gill et al. 2021).

• 9. Corythornis cristatus thomensis and Corythornis cristatus nais are endemic subspecies from São Tomé and from Príncipe, respectively. Genetically they fall well within the nominate subspecies: mitochondrial divergences from samples from Malawi were only 0.3% for C. c. thomensis and 0.9% for C. c. nais, and divergence between the two island lineages is 0.8% (Melo and Fuchs 2008). The two subspecies have phenotypic differences: C. c. nais being intermediate between C. cristatus and C. leucogaster, while C. c. thomensis has darker plumage than mainland birds, especially the juveniles (Christy and Clarke 1998). It is possible that these populations represent recent and distinct colonization events that are now evolving in isolation. They are treated as separate endemic species by BirdLife International (HBW and BirdLife International 2020).

• 10. Halcyon malimbica dryas. Endemic subspecies from Príncipe, whose likely validity should nevertheless be confirmed using multiple lines of evidence, including molecular data.

• 11. Psittacus erithacus princeps. Mitochondrial data inferred a relatively simple, albeit curious, history for the gray parrots from Príncipe (Melo and O’Ryan 2007), which created the only real taxonomic conundrum for the avifauna of the Gulf of Guinea oceanic islands. This population is the result of two colonization events: one that occurred up to 1.4 mya and a contemporary one. The first colonization came from the mainland stock that evolved into P. timneh, whereas the recent colonization came from P. erithacus. We cannot exclude that the latter was an accidental introduction linked to the Portuguese trade of this species from Angola to Europe, which used to be done by boats that made a stopover in Príncipe (Melo and O’Ryan 2007). Most of the Príncipe birds (75%) have the Príncipe mitochondrial lineage, even though morphologically they are closer to P. erithacus. We still do not know the overall impact on the genome of the mixing of the two lineages. The International Ornithological Council opted to use the genetic evidence to treat the Príncipe population as a subspecies of P. timneh (Gill et al. 2021), while other authorities have kept the original treatment (HBW and BirdLife International 2020; Clements et al. 2021). We consider this to be an open issue that can only be sorted out with an extensive genetic investigation.

• 12. Dicrurus modestus modestus. The drongo present on Príncipe has been treated for a long time as an endemic species, although its taxonomic status was always considered unclear (Jones and Tye 2006). A recent molecular study on the D. adsimilis complex returned a new taxonomic arrangement for the group, in which the Príncipe population is conspecific with the populations occurring in the forests of the Lower Congo Forest Block, despite clear differences in bill and tail size (Fuchs et al. 2018).

• 13. Terpsiphone rufiventer smithii. The population of T. rufiventer from Annobón was, until recently, often treated as a separate species, and still is by some authors (Gill et al. 2021). As it is unlikely that there is regular gene flow with the neighboring mainland populations, this population is likely on an independent evolutionary trajectory.

• 14. Onychognathus fulgidus fulgidus. The nominate subspecies of this large forest starling was described from São Tomé, to which it is endemic. The birds on São Tomé are larger, more robust, and more vocal than those occurring on the African mainland (Amadon 1953; Christy and Clarke 1998), warranting molecular research to determine if it may constitute a distinct species.

Habitat Associations

The aquatic avifauna of the islands is species-poor but occupies a wide variety of ecological niches (Lima et al. 2021). Regularly breeding seabirds include two noddies, two terns, one tropicbird, and one booby, all of which breed in different sets of offshore islets (Monteiro et al. 1997; Jones and Tye 2006—Appendix), where they have distinct nesting microhabitats (Leventis and Olmos 2009; Valle et al. 2016; Bollen et al. 2018). The White-tailed Tropicbird also breeds on cliffs and trees on the main islands, and a putatively distinct form of the Band-rumped Storm-Petrel Hydrobates castro is thought to breed within burrows on the ground in São Tomé’s native forests (Flood et al. 2019). The resident Black Kite Milvus migrans parasitus is also commonly found foraging at sea. Other resident aquatic bird species include predominantly coastal species that also occur along larger rivers, such as Western Reef-Heron Egretta gularis and Long-tailed Cormorant Microcarbo africanus. Predominantly freshwater species also occur along the coast and in brackish waters, such as Common Moorhen Gallinula chloropus and Striated Heron Butorides striata. The islands receive remarkably few non-breeding aquatic species as regular visitors, most of which occur on the coast or along the lower reaches of rivers and streams. The lagoons on the northern coast of São Tomé and the bay of Santo António on Príncipe are the main localities where vagrant aquatic birds have been recorded (Jones and Tye 2006; de Lima et al. 2021).

Until the 1990s, assessments of the distributions and habitat associations of the terrestrial birds of the oceanic islands of the Gulf of Guinea were mostly based on non-systematic observations (Jones and Tye 2006) and focused on understanding the links between endemic species and land-use types to evaluate conservation status (Jones and Tye 1988; Atkinson et al. 1991; Peet and Atkinson 1994). Knowledge greatly improved following systematic surveys on both Príncipe (Baillie 2001; Dallimer and King 2008; Dallimer et al. 2012; Fundação Príncipe 2019) and São Tomé (Dallimer et al. 2009; de Lima et al. 2013, 2014; Soares 2017; Soares et al. 2020). These studies have shown that native species, including endemics, dominate the avifauna across the islands, while non-natives tend to be restricted to degraded environments, such as plantations and notably, to non-forested areas. A few of the endemics are more sensitive to anthropogenic influence and are currently restricted to the best-preserved forests. The Giant Weaver Ploceus grandis on São Tomé and the Príncipe Golden-Weaver P. princeps seem to be the only endemic species that are clearly more abundant outside forests, even though the Sao Tome Pigeon, the Sao Tome White-eye Zosterops feae and a few of the endemic subspecies are also frequently encountered outside native forest. As in other tropical forests (e.g., Newbold et al. 2013), species sensitive to forest degradation tend to be larger, and insectivorous or frugivorous, while non-natives are mostly small and granivorous. Other environmental factors are often correlated with land use, making it hard to disentangle their effect on bird assemblages but, overall, the highest proportions of endemics are found in remote steep areas at higher altitudes and with higher rainfall. Annobón is seldom visited by ornithologists, and thus the ecology of its avifauna remains the most incompletely documented (Sloan 2017).

Several studies have focused on the habitat associations of the islands’ Critically Endangered species: Sao Tome Ibis (Margarido 2015; de Lima et al. 2017), Newton’s Fiscal (Maia and Alberto 2009; Lewis et al. 2018), Príncipe Thrush (Dallimer et al. 2010; Rebelo 2021), and Sao Tome Grosbeak (Solé et al. 2012), as well as the Príncipe Scops-Owl, which has also been proposed to qualify as Critically Endangered (Freitas 2019). Other studies have addressed the habitat associations of the Gray Parrot Psittacus erithacus on Príncipe (Valle et al. 2017), the Annobón population of the African Scops-Owl Otus senegalensis feae (Rodriguez-Prieto et al. 2014), the distinctive São Tomé subspecies of Barn Owl Tyto alba thomensis (Alves 2019), and the endemic pigeons of São Tomé (Carvalho et al. 2014). In addition to providing an understanding of distribution and ecology, these studies have also helped in estimating population sizes (i.e., Azevedo 2015) and informing conservation strategies (BirdLife International 2014a, b; Fundação Príncipe et al. 2021).

The Endemic Birds

How Many Endemics?

Endemism in the oceanic islands of the Gulf of Guinea is restricted to resident landbirds (Jones and Tye 2006; de Lima and Melo 2021), although the still undescribed local population of the Band-rumped Storm-Petrel may represent an exception (Flood et al. 2019). Out of the 66 resident landbird species of the three islands, 17 are possibly non-native (de Lima and Melo 2021), and of the remaining 49 confirmed extant natives, 90% are endemic at the species or subspecies level (Tables 21.2, 21.3, and 21.4, Appendix). The exact number of endemic species varies according to different authorities (Hoyo 2020), with the most recent assessments recognizing 32 (HBW and BirdLife International 2021), 31 (Gill et al. 2021), and 29 species (Clements et al. 2021; de Lima and Melo 2021). These discrepancies are restricted to recent speciation events, in which some authors treat sister taxa as species, while others treat them as subspecies (Box 21.1). Such divergent taxonomic treatments are to be expected considering that speciation is a continuous process, and authorities agree that there is a total of 45 distinct evolutionary lineages (endemic species and subspecies, including the Príncipe Scops-Owl). Here we follow a conservative approach, where recent divergence events are treated as subspecies until further evidence emerges, resulting in a total of 29 endemic species (Tables 21.1 and 21.2) and 16 endemic subspecies (Tables 21.1 and 21.3). Additionally, the endemic Príncipe Seedeater has diverged into three single-island endemic subspecies: the nominate restricted to Príncipe, another to Boné de Jóquei (islet c. 3 km offshore from Príncipe), and the third to São Tomé (Table 21.3), a taxonomic arrangement supported by molecular data (Melo 2007).

Table 21.4 Divergence time estimates, in million years (Ma), between endemic birds of the oceanic islands of the Gulf of Guinea and their closest mainland relative(s); square brackets indicate clades. These estimates are illustrative only as they were derived from different markers and used different rates and estimation methods. When available, 95% confidence intervals are shown. Note that the ages of the oldest known sub-aerial rocks are 31 Ma for Príncipe, 15 Ma for São Tomé and 6 Ma for Annobón (Ceríaco et al. 2022)

Bird Endemism of the Gulf of Guinea Oceanic Islands in Perspective

Although confined to a land area of just over 1000 km², the 29 endemic species of Príncipe, São Tomé, and Annobón represent 60% of the endemic bird species of the vast Guinean Forests of West Africa biodiversity hotspot (area: 621,705 km²; IUCN 2015). It is not surprising for oceanic islands to be centers of endemism, but it is still instructive to compare the bird endemism levels of the Gulf of Guinea islands with those found elsewhere. The Galapagos have 22 endemic landbirds in 13 islands totaling some 8000 km², whereas the six largest Hawaiian Islands have 30 extant endemic species and 19 documented extinct endemics in over 16,000 km² (Stattersfield et al. 1998). The high level of bird endemism within the relatively restricted area of the Gulf of Guinea oceanic islands has no parallel, with the islands following a very distinct trajectory to the global pattern (Mayr 1965; Fig. 21.2). In a survey of the 45 small islands (<10,000 km²) that have at least one endemic species, the mean number of endemic species is 2 and the mode is a single species (Coyne and Price 2000). By comparison, São Tomé, 857 km², has 17 single-island endemic species and Príncipe, 139 km², has eight, plus three additional shared endemic species. A recent worldwide analysis of bird communities on oceanic islands further identified Príncipe and São Tomé as the only group of islands where the number of species, colonizations, and within-archipelago speciation all exceeded the predictions of the global model (Valente et al. 2020).

Fig. 21.2

The number of endemic bird species in the oceanic islands of the Gulf of Guinea has no equivalent worldwide. Left—number of endemic bird species in relation to island area for all the small oceanic islands of the world that have at least one endemic species; data from Coyne and Price (2000). Right—adaptation of Figure 1 from Mayr’s (1965) global analysis of species turnover on islands (percentage of bird endemism in relation to island area), where three island categories were defined: A—solitary, well-isolated islands; B—single islands near continents or large archipelagos; C—islands in the Gulf of Guinea. In both cases, the Gulf of Guinea islands follow a distinct trajectory, characterized by a much higher number of endemics than expected from their area alone

Why So Many Endemic Birds?

The 29 endemic species of the oceanic islands of the Gulf of Guinea belong to 20 distinct evolutionary lineages from 16 families. This means that the current diversity of the endemics required at least 20 independent colonization events from the mainland. Hence, the impressive level of bird endemism is spread across independent phylogenetic lineages rather than being concentrated within species-rich genera from just a few colonizations, such as in the well-known adaptive radiations that have occurred in other archipelagos. In the Galapagos, six colonizations gave rise to the 22 extant endemic species, dominated by the radiation of Darwin’s finches (Grant and Grant 2008), and in the Hawaiian archipelago, six colonizations diversified into more than 40 endemic species, dominated by the Hawaiian honeycreepers (Pratt 2005). The factor underlying these distinct patterns is geography. These two archipelagos are much farther from the mainland than the Gulf of Guinea oceanic islands. The oceanic islands of the Gulf of Guinea constitute an ‘intermediate island system’ (Melo 2007; Ricklefs and Bermingham 2007), whose biogeographical characteristics lie between very isolated and virtually independent systems, and those so close to the mainland that their diversity patterns are determined mostly by ecological factors. In addition, the diverse and species-rich ecosystems of West Africa and the Congo Basin to the north and east, respectively, provide ample sources for potential colonization to the Gulf of Guinea oceanic islands. This “surrounding landmass proportion” is not only very large, but it consists mostly of habitats that are similar to those of the islands themselves—the two most important parameters associated with an increased likelihood of successful colonization (Weigelt and Kreft 2013).

The linear arrangement of the Gulf of Guinea islands and the relatively large distances between them, which are similar to their distances from the African continent, appear to have favored independent colonizations from the mainland relative to dispersal between islands. Colonization by so many mainland species is expected to increase inter-specific competition, reducing the chances for adaptive radiation (Schluter 2000; Ricklefs and Bermingham 2007). Nevertheless, white-eyes (Zosteropidae) represent a five-species radiation in these islands (Melo et al. 2011—Box 21.2). Other instances of inter-island dispersal events leading to speciation include the Giant and Príncipe Golden weavers (Valente et al. 2020), and the Príncipe Sunbird Anabathmis hartlaubi and the Sao Tome Sunbird Dreptes thomensis (Rauri Bowie, unpublished data: sister species relationship supported by mitochondrial and nuclear markers.; MM and Luís Valente, unpublished data: sister species relationship supported by mitochondrial sequence data). Inter-island dispersal has also resulted in the differentiation of the Príncipe Seedeater into three subspecies (Jones and Tye 2006; Melo 2007). The distinct species of green-pigeons (Pereira 2013), scops-owls (Freitas 2019), thrushes (Melo et al. 2010), and Crithagra canaries (Melo et al. 2017) present on different islands may have been derived either from independent colonizations from the African continent or from inter-island dispersal. The same applies to the island subspecies of the Lemon Dove Columba larvata (Pereira 2013), and the Malachite Kingfisher (Melo and Fuchs 2008).

Although the proximity to a species-rich continent does increase the chances of successful colonizations, it also increases the probability that gene flow between island and mainland populations is maintained at levels that will prevent population divergence and eventual speciation. Hence, the most likely reason that the oceanic islands of the Gulf of Guinea support the highest concentration of endemic birds worldwide is their unique geographic location: they are close enough to the African continent to be colonized by a diverse array of species, but far enough to allow successful immigrants to evolve in isolation.

In summary, bird speciation in the oceanic islands of the Gulf of Guinea has in most cases occurred by independent divergence in allopatry (allospeciation: Mayr and Diamond 2001), when an immigrant population from the mainland reached one island and evolved there in isolation. This is the overwhelmingly dominant route for bird speciation on islands (Ricklefs and Bermingham 2007; Valente et al. 2020) and for birds in general (Price 2008). In addition, molecular data have now revealed previously unrecognized radiations on the islands, most remarkably among the white-eyes (Box 21.2). Although comprising only five species, this radiation may be the third largest globally for birds inhabiting small oceanic islands, and further stands out by having the fastest speciation rates recorded in birds and one of the highest in vertebrates (Appendix 5 in Melo et al. 2011).

Box 21.2: The Radiation of the White-Eyes (Zosteropidae) of the Oceanic Islands of the Gulf of Guinea

The spectacular radiations of the Hawaiian honeycreepers (Pratt 2005) and of the Galapagos finches (Grant and Grant 2008) could mislead us into believing that radiations are a common diversification process for birds on small oceanic islands—when they are in fact a very rare exception (Valente et al. 2020). The next largest oceanic island bird radiation worldwide is found in the Gulf of Guinea, where the five white-eyes species descend from a single ancestor that reached the islands within the last 0.7–1.1 my (Melo et al. 2011). Although modest in size, the white eye radiation boasts one of the fastest rates of speciation ever documented in vertebrates (Melo et al. 2011).

The radiation itself is a textbook example of the “archipelago radiation model” developed originally for island birds (Lack 1947; Grant 2001; Petren et al. 2005), which is also in agreement with current views of the speciation process (Rundle and Nosil 2005; Nosil 2012), where inter-specific competition is the engine of phenotypic diversification (numbers refer to photos below):

• (I) Descendants of the original colonization (6: mainland relative), island-hop and diverge in isolation (1, 3, 5).

• (II) Presumably because they occupy similar habitats, phenotypic changes are not pronounced—as illustrated by the Príncipe and São Tomé white-eyes (1, 3), which are indistinguishable in the field (Hering et al. 2018).

• (III) Inter-island dispersal events bring diverging populations together, setting the stage for inter-specific competition.

• (IV) The pressure of resource competition is felt more strongly in the outnumbered new arrivals. These undergo the most phenotypic change and at very fast rates. The most aberrant species of the group (2, 4) represent the most recent speciation events: they may have diverged from one another less than 0.3 mya and from a typical white-eye less than 0.5 mya—a timeframe typical of intra-specific but not inter-specific divergence in birds (Melo et al. 2011). This process of asymmetric divergence driven by resource competition has been predicted by theory (Doebeli and Dieckmann 2000) and is now empirically supported by the radiation of Darwin’s finches (Petren et al. 2005) and of the Gulf of Guinea white-eyes.

(1) Príncipe White-eye Zosterops ficedulinus. (2) Príncipe Speirops Z. leucophaeus. (3) São Tomé White-eye Z. feae. (4) Black-capped Speirops Z. lugubris (São Tomé). (5) Annobón White-eye Zosterops griseovirescens. (6) Northern Yellow White-eye Z. senegalensis, the closest mainland relative with the typical white-eye phenotype. Photo credits: (1–4) Lars Petersson, (5) Martim Melo, (6) Jake Selby

Systematics of the Endemic Birds: New Insights from Molecular Data

The taxonomy and systematics of the endemic birds of the Gulf of Guinea were mostly addressed by Amadon (1953) and Naurois (1994) and later summarized by Jones and Tye (2006). Evolutionary inferences were still then dependent on phenotypic data, and particularly on morphological traits, which are often adaptive and can quickly lose their phylogenetic signal. This issue is particularly problematic in birds (Bock 1967; Hafner et al. 1984) and on oceanic islands, where both rapid phenotypic evolution (Millien 2006; Melo et al. 2011; Garcia-Porta et al. 2016; Sendell-Price et al. 2020) and phenotypic convergence (Fleischer et al. 2008; Covas 2016; Benítez-López et al. 2021) are common. Molecular phylogenies are expected to constitute better hypotheses of evolutionary history than phenotype-based phylogenies, as they use genetic markers that are mostly independent from phenotypic traits, and not under direct selection (Bromham et al. 2002; Davies and Savolainen 2006).

The most important insight brought by molecular phylogenies was that most bird speciation events in the Gulf of Guinea islands are recent, having occurred since the late Pliocene (2.5 Ma—Table 21.4). The exceptions so far are the Príncipe and São Tomé thrushes and the Sao Tome Short-tail, which may have speciated in the mid-Pliocene (c. 3.5–4.0 Ma), and Dohrn’s Thrush-Babbler Sylvia dohrni, which may date back to the Miocene (c. 8 Ma—Table 21.4). Even without estimates of speciation times for some of the endemics, it is safe to conclude that the present species are all much more recent than the emergence of the islands they inhabit, which range from 31 to 6 Ma (Ceríaco et al. 2022). This pattern indicates that the oceanic islands of the Gulf of Guinea constitute a speciation center rather than a stable refuge for species that went extinct on the mainland. The absence of old species is nevertheless surprising and could be due to the recent and intense volcanic history of the islands (Lee et al. 1994; Barfod and Fitton 2014; Ceríaco et al. 2022). The patterns of genetic variation of the lizard Trachylepis thomensis within São Tomé have been linked to the impact of volcanic activity, with the extent of this variation being much lower than expected from the age of the island (Jesus et al. 2005). Likewise, volcanic activity on São Tomé has been linked to the evolutionary history of two sister caecilian lineages—Schistometopum ephele and S. thomense (O’Connell et al. 2021). In the absence of (sub)fossil evidence, it is not known if cycles of sea-level rise during post-glacial periods may have caused extinctions (Jones and Tye 2006; Ceríaco et al. 2022). Príncipe is likely to have been the most affected, having lost almost 90% of its land area as recently as 12,000 years ago (Norder et al. 2018). Likewise, during glacial periods Annobón was eight times larger, whereas the size of São Tomé did not change greatly. In any case, the relatively small sizes of the islands and their proximity to the mainland might make them susceptible to more rapid species turnover, which may further help explain the young age of most endemic bird species.

The main taxonomic consequence of the fast rates of phenotypic divergence inferred from molecular phylogenies is that all the endemic genera have been found invalid. This includes the Dohrn’s Thrush-Babbler and the Sao Tome Short-tail, formerly placed in the monotypic genera Horizorhinus and Amaurocichla, respectively, whose peculiar traits obscured to which families they belonged (i.e., classified as incertae sedis). Dohrn’s Thrush-Babbler, from Príncipe, is sister to a clade including the African Hill Babbler Sylvia abyssinica, an Afromontane forest species present also in the neighboring land-bridge island Bioko and Mount Cameroon (Voelker et al. 2009). In the case of the Sao Tome Short-tail, several of its traits, such as nine primaries (albeit a vestigial tenth primary is present) and ten rectrices with a protruding shaft, led to the suggestion that it could share an ancestor with the South-American Furnariidae (Naurois 1982). In actuality, it represents a recent speciation event from continental Africa, derived from the same ancestor as the Mountain Wagtail Motacilla clara, with which it shares the ecological niche of forest streams, and the Cape Wagtail M. capensis (Alström et al. 2015). The São Tomé Grosbeak, a Crithagra canary (Fringillidae) sister to the co-occurring Príncipe Seedeater (Moreau 1962; Melo et al. 2017), was originally placed in the monotypic genus Neospiza, and was often classified as a weaver (Ploceidae) (e.g., Bocage 1888b—he later placed it in Fringillidae: Bocage 1904; Sclater 1924; Bannerman 1953; Moreau 1962). The Sao Tome Weaver Ploceus sanctithomae, formerly in the endemic monotypic genus Thomasophantes (Amadon 1953; Moreau 1960; Naurois 1994), is now considered sister to the clade including the Forest Weaver Ploceus bicolor and the Red-headed Weaver Anaplectes rubriceps (Anaplectes being invalid as well; De Silva et al. 2019). The Sao Tome Sunbird is currently still classified under the only remaining endemic genus, Dreptes, which is known to be invalid, since the species is sister to the Príncipe Sunbird (genus Anabathmis) (Rauri Bowie, pers comm.; MM and Luís Valente, unpublished data). Finally, the genus Speirops, endemic to the Gulf of Guinea, is also no longer considered valid (Melo et al. 2011). It used to group four species of “aberrant” white-eyes (Zosteropidae): Black-capped Speirops Zosterops lugubris (São Tomé), Príncipe Speirops Z. leucophaeus, Bioko Speirops Z. brunneus, and Cameroon Speirops Z. melanocephalus (Mount Cameroon). However, molecular data show that the four species are not monophyletic and that the “aberrant” characters are the result of fast phenotypic divergence. On the oceanic islands, aberrant species represent the most recent speciation events (Box 21.2; Melo et al. 2011), rather than being derived from the oldest colonizations as previously assumed (Amadon 1953; Moreau 1957).

Conservation

The importance of the islands for conservation was first noted when the south-western forests of São Tomé were identified as the second most important for bird conservation in Africa (Collar and Stuart 1988). As a result, an IUCN-funded mission surveyed plants and vertebrates of São Tomé and Príncipe confirming the high endemism and the global importance of the biological diversity of the oceanic islands (Jones and Tye 1988; Jones et al. 1991; Jones 1994). For birds, each of the oceanic islands is listed by BirdLife International as an independent Endemic Bird Area (Stattersfield et al. 1998; BirdLife International 2021a). More recently, the moist lowland forests of Príncipe, São Tomé, and Annobón were identified as the third most important in the world for the conservation of forest birds (Buchanan et al. 2011), and the endemic birds were the main factor for the protected areas of São Tomé and Príncipe combined to be considered globally as the 17th most important protected area for the conservation of threatened species (Le Saout et al. 2013).

Some of the endemic birds were already extremely rare early in the twentieth century and remain at high risk, but only the Príncipe subspecies of the Olive Ibis Bostrychia olivacea rothschildi became extinct (de Lima and Melo 2021). The most recent assessment listed 14 threatened bird taxa for the islands (IUCN 2021): five Critically Endangered, including the Annobón Scops-Owl Otus feae—which we treat as a subspecies of O. senegalensis following Clements et al. (2021); four Endangered, including the Gray Parrot, which is common on Príncipe and the only non-endemic threatened bird on the islands (Valle et al. 2021); and five Vulnerable (Tables 21.2 and 21.3).

The number of threatened taxa has increased since 2000, when only nine were listed: three Critically Endangered (the Annobón Scops-Owl and the Príncipe Thrush were not recognized as species by IUCN), none Endangered (the Maroon Pigeon Columba thomensis was Vulnerable, the Príncipe and São Tomé white-eyes were treated as conspecific and Vulnerable, and the Sao Tome Green-Pigeon Treron sanctithomae and the Gray Parrot were Least Concern), and six were Vulnerable (including the Principe and Sao Tome white-eyes treated as the same taxon). Most of these changes have been due to improved knowledge and not necessarily to a deterioration of the situation of the species, despite indications that conditions might be worsening for several taxa (IUCN 2021). The conservation status of each bird species is reviewed every year (BirdLife International 2021b) and thus further changes are expected. Other taxa await assessment, namely all endemic subspecies and putative new species, many of which are likely to be threatened. These include the Boné de Jóquei Islet subspecies of the Principe Seedeater Crithagra rufobrunnea fradei, which has a highly restricted range (Melo 2007), the elusive Gulf of Guinea Band-rumped Storm-Petrel (Flood et al. 2019), and the Príncipe Scops-Owl, which is still being described but will likely classify as Critically Endangered (Freitas 2019).

Habitat loss, overexploitation, and introduced species are key threats to native birds both globally (IUCN 2021) and in the Gulf of Guinea. Given the habitat associations described in the previous section, forest loss and degradation are the main threats to the birds on these islands (e.g., Dallimer et al. 2012; Soares et al. 2020). Most of this habitat loss can be attributed to agricultural expansion and intensification (Oyono et al. 2014), both to supply the local markets (notably horticulture) and to produce export cash crops (e.g., cocoa, palm oil, and coffee). Logging, fire, mining, infrastructure development, urban and tourism expansion, livestock, and silviculture (e.g., oil palm wine and medicinal plants) also contribute to habitat loss. To halt ongoing habitat loss, it is vital to ensure the effective implementation of existing protected areas, and their possible expansion, since some important forests are not yet formally protected (BirdLife International 2020; de Lima et al. 2022). Furthermore, environmentally friendly practices should be promoted in extractive and agricultural activities to ensure that complex vegetation structures are not lost and that introduced species are kept under control (de Lima et al. 2014; Carvalho 2015).

Most bird species are hunted, but the effects of direct exploitation are more noticeable on larger species (de Lima et al. 2013), such as the Sao Tome Ibis (Sampaio et al. 2016; de Lima et al. 2017), pigeons (Palmeirim et al. 2013; Carvalho 2015; Fundação Príncipe 2019) and the Brown Booby (Bollen et al. 2018). The diffuse nature of hunting coupled with the rugged terrain makes it extremely difficult to enforce existing laws that regulate this activity (Albuquerque and Carvalho 2015a, b; de Lima et al. 2022). Given that bird hunting is mostly a cultural, commercial, and recreational activity that contributes little to protein intake, diverting hunting efforts to instead control populations of introduced mammals could have a dual positive effect on bird conservation (Carvalho 2015).

Introduced bird species are thought to be strongly associated with land-use intensification, having little or no impact on the native avifauna (Soares et al. 2020). On the other hand, the effects of introduced mammals (Dutton 1994) and plants (Figueiredo et al. 2011) have long been identified as potential threats, even though their impacts remain poorly understood (BirdLife International 2014a, b; Fundação Príncipe et al. 2021). Feral pigs and cows feed on understory plants and turn over the soil, disturbing key forest habitats that evolved in the absence of large terrestrial mammals. Rats Rattus sp. and Mona Monkeys Cercopithecus mona are likely to have direct effects through nest predation (Guedes et al. 2021). Introduced mammal and plant species also have the potential to change forest structure in the long term, namely through the disruption of seed dispersal and other processes linked to forest regeneration (Heleno et al. 2021).

Other factors, such as pollution and climate change have also been identified as potential threats to the endemic-rich avifauna of these islands (IUCN 2021). For example, the intensive use of insecticides was claimed to be responsible for a severe population crash of the Sao Tome Paradise Flycatcher Terpsiphone atrochalybeia in the 1970s (Naurois 1984a).

Because most threats to biodiversity act synergistically and often occur as a result of habitat loss, protecting the remaining native forest is the single most important measure to secure the future of these species (de Lima 2012). Fortunately, all the islands have significant proportions of their territory already dedicated to biodiversity conservation (UNEP-WCMC and IUCN 2021), and the protected area network is soon expected to expand to cover additional important habitats (BirdLife International 2020; de Lima et al. 2022). However, weak enforcement of environmental legislation remains a major concern (de Lima et al. 2017). Improving our knowledge of species ecology and of pervasive threats, such as hunting, invasive species, and climate change, will also be key to designing effective species-specific conservation measures. In this regard, conservation priorities in São Tomé and Príncipe have been identified through extended discussions, both for protected areas (Albuquerque and Carvalho 2015a, b) and for all Critically Endangered bird species (BirdLife International 2014a, b; Fundação Príncipe et al. 2021). Despite not being entirely fulfilled, these have been extremely useful in guiding conservation action, and continued revision will be essential. The success of ongoing conservation efforts ultimately relies on engaging the inhabitants of the islands, a process that is still in its infancy (de Lima et al. 2022). In this regard, birds are also being used to raise awareness locally and globally for the value of the unique biodiversity of the islands (e.g., Rebelo 2021; Ayres et al. 2022).

Concluding Remarks

The oceanic islands of the Gulf of Guinea constitute an outstanding example of an intermediate island system for birds, whose geographical location and rich rainforests maximize the accumulation of bird endemism. As such, they offer a valuable suite of phylogenetically independent replicates for testing hypotheses about evolutionary processes in speciation and adaptation. Knowledge gaps persist regarding Annobón and the status of some species, including potential undescribed endemics, past extinctions, and the origins of putative non-native species. Birds are still the best-known taxon in the archipelago, however, making them ideal exemplars that can guide future work on other groups. Our knowledge of environmental constraints and the history of human occupation of these islands also make them excellent models for understanding ecological processes and testing conservation strategies that can be used in a wider context, for instance in other small forested islands.

Appendix

Checklist of bird species on the oceanic islands of the Gulf of Guinea, excluding vagrant and unconfirmed species (de Lima and Melo 2021). Islands: P, Príncipe; S, São Tomé; A, Annobón. Status: E, endemic species; I, probably non-native; R, native non-endemic resident; S, endemic subspecies; X, extinct; B, breeding migrant; M, non-breeding migrant; ?, uncertain. Subspecies are only identified when they are endemic. Taxonomy and nomenclature follow Clements et al. (2021)

Higher taxonomy	Species/subspecies	P	S	A
Order Galliformes
Family Numididae
Numida Linnaeus, 1764	N. meleagris (Linnaeus, 1758)		I	I
Family Phasianidae
Coturnix Garsault, 1764	C. delegorguei histrionica (Hartlaub, 1849)		S
Pternistis Wagler, 1832	P. afer (Müller, PLS, 1776)		I
Gallus Brisson, 1760	G. gallus (Linnaeus, 1758)			I
Order Columbiformes
Family Columbidae
Columba Linnaeus, 1758	C. livia Gmelin, 1789	I	I
	C. thomensis Bocage, 1888		E
	C. malherbii Verreaux & Verreaux, 1851	E	E	E
	C. larvata inornata (Reichenow, 1892)			S
	C. l. principalis (Hartlaub, 1866)	E
	C. l. simplex (Hartlaub, 1849)		E
Streptopelia Bonaparte, 1855	S. senegalensis (Linnaeus, 1766)	I	I
Treron Vieillot, 1816	T. sanctithomae (Gmelin, 1789)		E
	T. calvus virescens Amadon, 1953	S
Order Cuculiformes
Family Cuculidae
Chrysococcyx Boie, F., 1826	C. cupreus insularum Moreau & Chapin, 1951	S	S	S
Order Caprimulgiformes
Family Apodidae
Zoonavena Mathews, 1918	Zoonavena thomensis (Hartert, 1900)	E	E
Apus Scopoli, 1777	A. affinis bannermani Hartert, 1928	S	S
Cypsiurus Lesson, R., 1843	C. parvus (Lichtenstein, 1823)	I	I
Order Gruiformes
Family Rallidae
Paragallinula Sangster, Garcia-R & Trewick, 2015	P. angulata (Sundevall, 1851)	?	?
Gallinula Brisson, 1760	G. chloropus (Linnaeus, 1758)	R	R	R
Order Charadriiformes
Family Scolopacidae
Numenius Brisson, 1760	N. phaeopus (Linnaeus, 1758)	M	M	M
Actitis Illiger, 1811	A. hypoleucos (Linnaeus, 1758)	M	M
Tringa Linnaeus, 1758	T. nebularia (Gunnerus, 1767)	M	M
Family Laridae
Anous Stephens, 1826	A. stolidus (Linnaeus, 1758)	B	B	B
	A. minutus Boie, F., 1844	B	?	B
Onychoprion Wagler, 1832	O. fuscatus (Linnaeus, 1766)	B	?	?
	O. anaethetus (Scopoli, 1786)	?	?	B
Order Phaethontiformes
Family Phaethontidae
Phaethon Linnaeus, 1758	P. lepturus Daudin, 1802	B	B	B
	P. aethereus Linnaeus, 1758	?
Order Procellariiformes
Family Oceanitidae
Hydrobates Boie, F., 1822	H. cf. castro (Harcourt, 1851)		?
Order Suliformes
Family Sulidae
Sula Brisson, 1760	S. leucogaster (Boddaert, 1783)	B	B	?
Family Phalacrocoracidae
Microcarbo Bonaparte, 1856	Microcarbo africanus (Gmelin, J. F., 1789)		R
Order Pelecaniformes
Family Ardeidae
Egretta T. Forster, 1817	E. gularis (Bosc, 1792)	R	R	R
Bubulcus Bonaparte, 1855	B. ibis (Linnaeus, 1758)	R	R
Butorides Blyth, 1852	B. striata (Linnaeus, 1758)	R	R
Family Threskiornithidae
Bostrychia G. R. Gray, 1847	B. olivacea rothschildi (Bannerman, 1919)	X
	B. bocagei Chapin, 1923		E
Order Accipitriformes
Family Accipitridae
Milvus Lacépède, 1799	M. migrans (Boddaert, 1783)	R	R
Order Strigiformes
Family Tytonidae
Tyto Billberg, 1828	T. alba thomensis (Hartlaub, 1852)		S
Family Strigidae
Otus Pennant, 1769	O. hartlaubi (Giebel, 1849)		E
	O. senegalensis feae (Salvadori, 1903)			S
	Otus sp. nov.	E
Order Coraciiformes
Family Alcedinidae
Corythornis Kaup, 1848	C. cristatus thomensis (Salvadori, 1902)		S
	C. cristatus nais (Kaup, 1848)	S
Halcyon Swainson, 1821	H. malimbica dryas Hartlaub, 1854	S
Ceryle F. Boie, 1828	C. rudis (Linnaeus, 1758)	?
Order Psittaciformes
Family Psittaculidae
Agapornis Selby, 1836	A. pullarius (Linnaeus, 1758)		I
Family Psittacidae
Psittacus Linnaeus, 1758	P. erithacus princeps Alexander, 1909	S	I
Order Passeriformes
Family Oriolidae
Oriolus Linnaeus, 1766	O. crassirostris Hartlaub, 1857		E
Family Dicruridae
Dicrurus Vieillot, 1816	D. modestus modestus Hartlaub, 1849	S
Family Monarchidae
Terpsiphone Gloger, 1827	T. atrochalybeia (Thomson, 1842)		E
	T. rufiventer smithii (Fraser, 1843)			S
Family Laniidae
Lanius Linnaeus, 1758	L. newtoni Bocage, 1891		E
Family Cisticolidae
Prinia Horsfield, 1821	P. molleri Bocage, 1887		E
Family Hirundinidae
Hirundo Linnaeus, 1758	H. rustica (Linnaeus, 1758)		M
Family Sylvidae
Sylvia Scopoli, 1769	S. dohrni (Hartlaub, 1866)	E
Family Zosteropidae
Zosterops Vigors & Horsfield, 1827	Z. ficedulinus Hartlaub, 1866	E
	Z. griseovirescens Bocage, 1893			E
	Z. feae Salvadori, 1901		E
	Z. lugubris (Hartlaub, 1848)		E
	Z. leucophaeus (Hartlaub, 1857)	E
Family Sturnidae
Onychognathus Hartlaub, 1849	O. fulgidus fulgidus (Hartlaub, 1849)		S
Lamprotornis Temminck, 1820	L. splendidus (Vieillot, 1822)	R
	L. ornatus (Daudin, 1800)	E
Family Turdidae
Turdus Linnaeus, 1758	T. xanthorhynchus Salvadori, 1901	E
	T. olivaceofuscus Hartlaub, 1852		E
Family Nectariniidae
Anabathmis Reichenow, 1905	A. hartlaubii (Hartlaub, 1857)	E
	A. newtonii (Bocage, 1887)		E
Dreptes Illiger, 1811	D. thomensis (Bocage, 1889)		E
Cyanomitra Reichenbach, 1853	C. olivacea (Smith, 1840)	R
Family Ploceidae
Ploceus Cuvier, 1816	P. princeps (Bonaparte, 1851)	E
	P. velatus Vieillot, 1819		I
	P. cucullatus (Müller, 1766)		I
	P. grandis (G. R. Gray, 1844)		E
	P. sanctithomae (Hartlaub, 1848)		E
Quelea Reichenbach, 1850	Q. erythrops (Hartlaub, 1848)		?
Euplectes Swainson, 1829	E. hordeaceus (Linnaeus, 1758)		I
	E. aureus (Gmelin, 1789)		I
	E. albonotatus (Cassin, 1848)		I
Family Estrildidae
Nigrita Strickland, 1843	N. bicolor (Hartlaub, 1844)	R
Estrilda Swainson, 1827	E. astrild (Linnaeus, 1758)	I	I
Uraeginthus Cabanis, 1851	U. angolensis (Linnaeus, 1758)		I
Spermestes Swainson, 1837	S. cucullata (Swainson, 1837)	I	I	I
Family Viduidae
Vidua Cuvier, 1816	V. macroura (Pallas, 1764)		I
Family Motacillidae
Motacilla Linnaeus, 1758	Motacilla bocagii (Sharpe, 1892)		E
Family Fringillidae
Crithagra Swainson, 1827	C. mozambica (Müller, 1776)		I
	C. rufobrunneaa (Gray, 1862)	E
	C. concolor (Bocage, 1888)		E

1. ^aC. rufobrunnea is represented by three endemic subspecies (Table 21.3)