Translator Disclaimer
10 December 2021 Taxonomic revision of the Red-fronted Tinkerbird Pogoniulus pusillus (Dumont, 1816) based on molecular and phenotypic analyses
Alexander N. G. Kirschel, Michaella Moysi, Sifiso M. Lukhele, Matteo Sebastianelli, Tsyon Asfaw, Louis Hadjioannou, Kim G. Mortega, Ara Monadjem, Robert G. Moyle
Author Affiliations +
Abstract

Red-fronted Tinkerbird Pogoniulus pusillus (Dumont, 1816) presently comprises three recognised subspecies, of which two are found in East Africa and one occurs disjunctly in southern Africa. Based on their respective distributions and phenotypic differences, a taxonomic reassessment of the species is warranted. We performed a phylogenetic reconstruction using the mitochondrial genes ATPase 6/8 based on 33 samples from across the distribution of Red-fronted Tinkerbird and four outgroup samples, and then determined correspondence between genetic distances and differences in song and morphology among clades using the Tobias et al. criteria. Our phylogenetic analyses revealed 4.4% sequence divergence in mtDNA between northern and southern populations, with plumage, morphometric and song differences of a similar magnitude to those between P. pusillus and Yellow-fronted Tinkerbird P. chrysoconus, and above species-level thresholds according to the Tobias et al. criteria. Furthermore, the molecular phylogeny supports recognition of a synonymised taxon (P. p. eupterus) as a distinct, but phenotypically cryptic, subspecies in East Africa, with c.1.5% sequence divergence from P. p. affinis and P. p. uropygialis, which in turn differ less (1%) from each other. We propose that northern and southern Red-fronted Tinkerbirds are treated as separate species, and that the subspecies eupterus is resurrected. Our findings suggest that P. chrysoconus as presently constituted may also merit taxonomic revision.

Red-fronted Tinkerbird Pogoniulus pusillus (Dumont, 1816) has a disjunct distribution (Fig. 1), with two subspecies, P. p. affinis (Reichenow, 1879) and P. p. uropygialis (von Heuglin, 1862) from East Africa currently recognised by the IOC world bird list, and which are separated from the third, the nominate southern African race, by at least 1,500 km (Table 1). P. p. pusillus (including niethammeri Clancey, 1952), ranging from Eastern Cape, South Africa, to Sul do Save, Mozambique, differs in body size and plumage from the taxa in East Africa (Short & Horne 2001), but these differences have not hitherto been quantified to determine if they meet species-level criteria (e.g. Tobias et al. 2010). Furthermore, the transition zone between the two currently recognised subspecies in East Africa, P. p. affinis ranging from southern Ethiopia and Somalia south to the Rufiji River in Tanzania and west to Oldeani (Tanzania) and north Uganda, and P. p. uropygialis in northern and central Ethiopia to Eritrea and northern Somalia (White 1965), has remained unclear (Ash & Atkins 2009). These subspecies differ primarily in the yellow dorsal streaking of P. p. uropygialis, vs. white in affinis. Two additional subspecies have been recognised but since synonymised: subtle morphological differences led Grote (1928) to describe eupterus (type locality Ukerewe Island; see Fig. 1, Table 1), from inland Tanzania as distinct from P. p. affinis, which he considered a coastal race (Grote 1939). Grant & Mackworth-Praed (1942), finding overlap in wing length between coastal and inland forms, disagreed and synonymised it within affinis but instead recognised lollesheid (van Someren, 1931, 1932) from the Juba River, Somalia, based on the latter's shorter wing. Subsequently, White (1965) synonymised lollescheid (sic, possibly an unjustified emendation) within affinis, citing the variation in size as clinal. Thus, with several forms described from East Africa, a taxonomic reappraisal is warranted, ideally incorporating molecular phylogenetics.

Figure 1.

Map of Africa illustrating the disjunct distribution of Red-fronted Tinkerbird Pogoniulus pusillus (shaded). Points represent localities for 37 samples used in the study, including all currently recognised subspecies of P. pusillus (P. p. pusillus, P. p. affinis and P. p. uropygialis) and the separate clade of eupterus identified in this study as differentiated from affinis (circles), Yellow-fronted Tinkerbird P. chrysoconus (P. c. chrysoconus, P. c. xanthostictus and P. c. extoni—yellow symbols), and Speckled Tinkerbird P. scolopaceus (diamond). Also shown with the symbol X are type localities for the taxa pusillus (A), affinis (B), eupterus (C), lollesheid (D) and uropygialis (E). The dashed line represents the approximate contact zone between affinis and uropygialis, and the hatched area covers arid savanna separating the distributions of eupterus and affinis, within which the presence of which form of Red-fronted Tinkerbird, if any, is unknown.

img-z2-1_428.jpg

TABLE 1

Currently recognised taxa (by the IOC world bird list) and synonyms, years of description, authors, and type localites.

img-z3-2_428.gif

Red-fronted Tinkerbird had been distinguished from its presumed sister taxon, Yellow-fronted Tinkerbird P. chrysoconus, primarily using plumage differences, most notably forecrown colour. It is most parsimonious for a derived trait, such as the red forecrown colour, to have evolved once, but with the red forecrown feathers in this species explained by a gene that converts yellow dietary carotenoids to red ketocarotenoids (Kirschel et al. 2020a), it is possible that homoplasy, wherein different mutations lead to a convergent phenotype, explains such parallel patterns. It is therefore possible that disjunct populations of tinkerbirds with red forecrowns have resulted from independent mutations among genes functioning in the conversion of carotenoids. A taxonomic reappraisal of Red-fronted Tinkerbird should thus incorporate all currently recognised subspecies of Yellow-fronted Tinkerbird (P. c. chrysoconus, P. c. extoni and P. c. xanthostictus) to determine whether they might actually be closer relatives of any of the subspecies of Red-fronted Tinkerbird.

We re-examine the relationships between populations of P. pusillus using molecular genetics and relate our findings to morphological variation according to Tobias et al. (2010) criteria. We then suggest a taxonomic reappraisal based on our findings. Our results suggest southern and eastern Africa forms have long been diverging, reflected by phenotypic differences, whilst there are genetic breaks in regions of East Africa that have seldom been identified as playing a role in the avian biogeography of the region.

Methods

Field work and sampling.—Field work was performed in Ethiopia in January and April 2019, and in Eswatini (formerly Swaziland) in July 2019. Blood samples were collected from the brachial vein of tinkerbirds. Of the samples collected by the University of Cyprus (UCY; see Table 2), seven P. p. uropygialis, five P. p. affinis and one P. c. xanthostictus from Ethiopia, along with one P. p. pusillus from Eswatini were then pooled with five samples of presumed P. p. affinis and one of P. c. extoni collected during field work in Kenya in 2011–14 and Tanzania in 2013 (described in Nwankwo et al. 2018), and three of P. p. pusillus collected in South Africa in 2015 (see Nwankwo et al. 2019). Two additional samples were provided by museum collections (a blood sample of P. p. pusillus from Morgan's Bay, South Africa, provided by the Museum of Vertebrate Zoology of the University of California Berkeley (MVZ), and a tissue sample of Speckled Tinkerbird P. scolopaceus from Equatorial Guinea, provided by the University of Kansas Natural History Museum; KU), for the outgroup.

TABLE 2

Pogoniulus samples included in the study, including GenBank accession numbers for the sequences.

img-z4-2_428.gif

Lab work.—DNA was extracted from blood and tissue samples using a Qiagen DNeasy blood and tissue kit, following the manufacturer's protocols (Qiagen, Valencia, CA, USA). Polymerase chain reaction (PCR) was then performed to amplify DNA of ATPase 6/8 genes in 10 µL reactions containing: 1× reaction buffer, 25 mM MgCl2, 10 mM dNTP, 1 U/µL Taq polymerase (Kapa Biosystems), and 1 µM of each primer. The primers ATPC02GQL (5′-GGACAATGCTCAGAAATCTGC-3′) and ATPC03HMH (5′-CATGGGCTGGGGTCRACTATG-3′) (Eberhard & Bermingham 2004) were used, and PCR amplification was performed using an Applied Biosystems 2720 Thermocycler. The thermal protocol involved an initial denaturation step at 95°C for five minutes, followed by 35 cycles (98°C / 40 seconds, 52°C / 45 seconds, 72°C / 40 seconds) and a final extension step at 72°C for five minutes. Resulting bands from gel electrophoresis of PCR products were visualised on a 1% agarose gel and extracted using a Macharey-Nagel gel extraction kit for Sanger sequencing. Resultant sequences were pooled, with 12 further sequences obtained from previous studies (see Nwankwo et al. 2019, Kirschel et al. 2020b), resulting in a total of 38 sequences (Table 2).

Phylogenetic reconstruction.—Sequences were aligned using MUSCLE in MEGA v.7 (Kumar et al. 2016) and imported as Fasta into Geneious, where they were converted to phylip format for maximum likelihood phylogenetic analysis using IQtree (Minh et al. 2020). Blocks were partitioned by codon position (Chernomor et al. 2016), then model selection was performed (Kalyaanamoorthy et al. 2017) based on best Bayesian Information Criterion (BIC) score, and maximum likelihood estimation was performed using 1,000 ultrafast bootstraps (Hoang et al. 2017).

Bayesian analysis was performed using MrBayes (Huelsenbeck & Ronquist 2001) in Geneious (Kearse et al. 2012), running four heated chains for one million generations, sampling every 1,000 generations with a burn-in of 250,000 generations. We also calculated uncorrected pairwise sequence divergence using distance estimates provided in Geneious.

Morphometrics.—We measured wing chord, tarsus, tail, bill (maxilla and mandible) length, and bill width of 128 P. pusillus and 148 P. chrysoconus specimens (Table 3) from the Natural History Museum, Tring (NHMUK); Natural History Museum of Los Angeles County, Los Angeles (LACM); Field Museum of Natural History, Chicago (FMNH); National Museum of Natural History, Smithsonian Institution, Washington, DC (USNM); American Museum of Natural History, New York (AMNH); Museum of Natural Science of Louisiana State University, Baton Rouge, LA (LSUMZ); Peabody Museum of Yale University, New Haven, CT (YPM); and the University of California Los Angeles Dickey Collection, Los Angeles (UCLA). One of us (ANGK) took all morphometric measurements. In accord with Tobias et al. (2010), we used Cohen's d statistic to quantify the effect size of morphological differences and assigned magnitude scores to plumage data and behavioural and ecological differences, to determine if different populations might qualify for species status according to the yardstick approach (Tobias et al. 2010). Specimens of both sexes were included in analyses after first running tests for differences between the sexes; in these monomorphic species, sex was determined by collectors based on gonad identification, and comprised 171 males and 105 females. Comparisons were made between P. chrysoconus and P. pusillus to establish the yardstick, then between P. p. pusillus (n = 23) and the East African populations (n = 125), and finally between three sets of East African populations that accorded with clades identified from the phylogenetic reconstruction.

Song analysis.—Songs of Pogoniulus tinkerbirds are very simple in structure, typically comprising ‘pulses’ (short tones, see Fig. 2) repeated at a constant frequency and rate (Kirschel et al. 2009, 2020b, Nwankwo et al. 2018, Sebastianelli et al. 2021), and are innate (Lukhele et al. unpubl.). The songs of P. chrysoconus and P. pusillus are the simplest in the genus, with the pulses repeated continuously without separation into discrete songs or bouts (Monadjem et al. 1994). They can thus be characterised simply by their pace and pitch. We analysed peak frequency and the temporal measurement of inter-onset interval (IOI), a measure of pace, calculated as the time between the onset of two consecutive ‘pulses’, from 386 recordings of P. chrysoconus and 187 P. pusillus, collated for a parallel study (Kirschel et al. unpubl.). To detect the large number of tinkerbird ‘pulses’ present in each recording (typically 100+), we used Raven Pro (K. Lisa Yang Center for Conservation Bioacoustics 2019) to detect tinkerbird ‘pulses’ automatically using the built-in band-limited energy detector (BLED) and extracted separately high-resolution spectral (DFT size = 65,536) and temporal measurements (DFT size = 512). The BLED detects sound signals when their amplitude in a specified band of frequency exceeds a specific signal-to-noise ratio against background noise (in dB). However, given the non-selectivity of the BLED, both natural and anthropogenic sounds with frequency similar to tinkerbird ‘pulses’ might be detected erroneously or might lead to modification of the time duration of the signal detected, if overlapping with tinkerbird calls. Therefore, signal detection results were visually inspected for false positives and modified where necessary to match the actual start and end times of ‘pulses’. We then used Cohen’s d statistic to quantify the magnitude of differences between first the two species, then between the 40 recordings of P. p. pusillus and 147 recordings of East African populations (35 P. p. affinis, 87 presumed P. p. eupterus and 25 P. p. uropygialis), and lastly between the three East African populations corresponding to the clades identified by phylogenetic reconstruction.

Figure 2.

Spectrograms showing song structure of Yellow-fronted Tinkerbird Pogoniulus chrysoconus and the four subspecies of Red-fronted Tinkerbird P. pusillus.

img-z6-1_428.jpg

TABLE 3

Means ± standard deviation of morphometric (in mm) segregated by sex (M = male, F = female) and acoustic measurements per taxon used in analyses by this study.

img-AdgI_428.gif

TABLE 4

Summary of the Tobias et al. (2010) taxonomic scoring system. Step 1 involves quantification of a trait's differences based on effect size (Cohen's d). In step 2, a score is given based on geographical relationships between pairs of taxa, whereas during the third step ecological and behavioural differences are classified as either minor or medium. According to this scoring system, a minimum total score of 7 is required for two taxa to be considered as separate species, and minor differences alone cannot be used to trigger this status.

img-AMz2_428.gif

Quantitative analysis.—Cohen's d statistics were calculated in R 4.0.1 (R Core Team 2020) using the ‘effsize’ (v.0.8.1) package (Torchiano 2016). The values were then converted into Tobias et al. (2010) levels of magnitude and plumage difference were scored according to the scales they suggest therein (Table 4). Ecology and behaviour, and geographical relationships, were also scored using Tobias et al. (2010) guidelines, based on differences evident from our field work but also described in field guides. We also compared morphology between groups using Principal Components Analysis (PCA), based on the correlation matrix of the six morphometric measurements using R 4.0.1 (R Core Team 2020), which allowed us to reduce the dimensionality of our dataset, as well as to discriminate the different clades in two-dimensional space. Ellipses (based on 95% confidence intervals) were drawn around the different clades identified in the phylogenetic reconstruction.

Plumage coloration differences.—These were scored qualitatively by ANGK based on experience from measuring the specimens, live birds in the field, and from quantitative analyses of all specimens as part of published (Nwankwo et al. 2019, Kirschel et al. 2020) and unpublished work.

Results

Maximum likelihood and Bayesian analyses.—We resolved a concordant phylogeny using Bayesian and maximum likelihood approaches (Fig. 3). The P. pusillus clade was nested within P. chrysoconus, and southern populations of P. p. pusillus were sister to northern populations. There were three distinct clades within East Africa, with ‘P. p. eupterus’ sister to a clade including P. p. affinis and P. p. uropygialis. Mean pairwise distances between these three populations were of the same magnitude (1.0–1.6%, see Table 5) as between established Yellow-fronted Tinkerbird subspecies, P. c. chrysoconus and P. c. xanthostictus.

Phenotypic differences.—All morphometric comparisons between the sexes resulted in ‘small' or ‘negligible’ Cohen’s d statistics (range 0.11 < d < 0.38) according to the ‘effsize’ package (Torchiano 2016), suggesting these species are sexually monomorphic, so all further tests were run on the entire specimen set with sexes pooled, although we provide means and standard deviations per sex (Table 3). We found greater differences in morphometrics and song between the larger and faster singing P. p. pusillus and East African populations with smaller body size and slower songs, than between P. pusillus and P. chrysoconus, but differences between East African Red-fronted Tinkerbird populations were more subtle (Table 6; see also Fig. 4 for differences in IOI). Between P. pusillus and P. chrysoconus we estimated a Tobias et al. (2010) score of 8, based on low effect sizes for morphometric (1) and song differences (1), 3 for plumage, based on the distinct forecrown colour difference (Nwankwo et al. 2019, Kirschel et al. 2020a), 1 for ecology, based on differences in their preferred habitats, savanna woodland (P. chrysoconus) vs. forest (P. pusillus) (Short & Horne 2001) and 2 for their narrow hybrid zone (Nwankwo et al. 2019). The score for (southern) P. p. pusillus vs. East African (northern) populations was also 8, based on greater morphometric and song differences than between the recognised species (both scored 2); a plumage score of 3, based on marked facial differences in throat and eyebrow colour (white in the north, yellow in the south) and the golden-yellow wing patch in the south vs. yellow wingbar in the north (Fig. 5); and a score of 1 for ecology, with southern Red-fronted Tinkerbird primarily a tall-forest bird, whereas northern populations mostly are found in acacia woodland (Short & Horne 2001). The disjunct distribution of northern and southern Red-fronted Tinkerbirds means that no score is given for geographical relationship. Within northern populations, scores were more difficult to assess because geographical relationships are not clear. We scored 5 between P. p. affinis vs. P. p. uropygialis, based mostly on wing length differences, and the latter’s yellower dorsal streaking (Fig. 5), but whose geographical relationship we expect involves either a broad (1 point) or narrow (2 points) hybrid zone in the Great Rift Valley; 4 for ‘P. p. eupterus’ vs. P. p. uropygialis, which again differ in wing length and dorsal streak colour, and we believe are allopatric; and 2–3 for P. p. affinis vs. ‘P. p. eupterus’, which we expect are either parapatric (3 points) or form a narrow hybrid zone (2 points), but with little or no differences in phenotypic traits (all Cohen’s d <1). The first two principal components extracted explained 79.6% of morphometric variation (Table 7), and PCA supported the results of the Cohen’s d analysis, with much overlap between the two species, but with P. p. pusillus almost completely distinct from all northern Red-fronted Tinkerbirds (Fig. 6).

TABLE 5

Uncorrected mean pairwise genetic distances among populations of Red-fronted Tinkerbird Pogoniulus pusillus, with pairwise distance from sample J31301 P. chrysoconus (Yellow-fronted Tinkerbird) included as a reference. Values in lowest diagonal indicate intra-taxon variation (not estimated in chrysoconus, represented here by a single individual).

img-AJHg_428.gif

Figure 3.

Bayesian consensus phylogenetic tree showing relationships within the Red-fronted Tinkerbird Pogoniulus pusillus clade and outgroup samples of Yellow-fronted Tinkerbird P. chrysoconus and Speckled Tinkerbird P. scolopaceus. Posterior probabilities of major bifurcations are shown above, and maximum likelihood bootstrap values below, branches. Catalogue / sample numbers are included for cross reference with Table 2.

img-z9-1_428.jpg

TABLE 6

Cohen's d statistics for morphometric (wing chord, tarsus, tail, and exposed length of maxilla, mandible length, and bill width) and song characters (peak frequency and inter-onset interval = IOI) between populations of Red-fronted Tinkerbird Pogoniulus pusillus.

img-z9-4_428.gif

Figure 4.

Inter-onset interval differences in songs between (A) Yellow-fronted Tinkerbird Pogoniulus chrysoconus and Red-fronted Tinkerbird P. pusillus, (B) P. p. pusillus and the three subspecies proposed for P. uropygialis in East Africa, and (C) among the three East African populations.

img-z10-1_428.jpg

Figure 5.

Representative photographs of individuals of Red-fronted Tinkerbird taxa included in the study. (A) P. p. uropygialis (sample AS32086, Table 2), (B) P. p. affinis (AS32041), (C) P. p. pusillus (AR93132), and (D) ‘P. p. eupterus’ (J31312) (Alexander N. G. Kirschel)

img-z10-3_428.jpg

Figure 6.

PCA plot of the first two principal components from morphometrics, explaining 79.6% of the variation. Ellipses are based on 95% CI. Yellow-fronted Tinkerbird Pogoniulus chrysoconus broadly overlaps with the various taxa of Red-fronted Tinkerbird (listed in the legend according to clades identified in the study), but there is almost no overlap between pusillus and other forms of Red-fronted Tinkerbird.

img-z11-1_428.jpg

TABLE 7

Factor loadings, eigenvalues and proportions of variance explained by the first two principal components extracted using PCA from morphometrics of study skins of Red-fronted Tinkerbird Pogoniulus pusillus.

img-AAe5_428.gif

Discussion

P. p. pusillus is highly divergent from East African populations of Red-fronted Tinkerbird. Although our analysis of mtDNA suggests the pusillus clade is monophyletic, consistent with the findings in Nwankwo et al. (2019), by contrast, preliminary evidence from ongoing work (ANGK unpubl.) based on whole genomes is discordant with the topology found here and suggests it is polyphyletic. Indeed, whole genome data support a sister relationship between P. p. pusillus and P. c. extoni, and a sister relationship between a northern clade of P. chrysoconus, comprising P. c. chrysoconus and P. c. xanthostictus, and the northern clade of P. pusillus, rendering both P. pusillus and P. chrysoconus polyphyletic. If P. p. pusillus is not even sister to East African populations of Red-fronted Tinkerbird, this further supports our findings here demonstrating their genetic distance in mtDNA, and their phenotypic divergence. Using Tobias et al. (2010) criteria, we estimate a score of 8, the same as the difference assessed here between the presently recognised species P. chrysoconus and P. pusillus, and exceeding the Tobias et al. (2010) species-level threshold score of 7. Morphological differences, such as in the greater tarsal length of P. p. pusillus, may correspond to expectations under Bergmann's Rule, but with no intermediate Red-fronted Tinkerbird populations occurring between their ranges, we suspect these size differences are biologically significant. Likewise, songs differ in pace between disjunct populations of Red-fronted Tinkerbird more than they do between Red-fronted and Yellow-fronted Tinkerbirds, with P. p. pusillus song much faster than those of affinis, eupterus and uropygialis in East Africa, and such differences might serve to differentiate these populations. We thus propose that Red-fronted Tinkerbird is split into two species, Southern Red-fronted Tinkerbird Pogoniulus pusillus (Dumont, 1816), and Northern Red-fronted Tinkerbird P. uropygialis (von Heuglin, 1862), including subspecies P. u. affinis (Reichenow, 1879) and P. u. eupterus Grote, 1928.

East African populations that span the range of eupterus, described by Grote (1928) from Ukerewe Island, Tanzania, as an inland race encompassing the Kilimanjaro area (Grote 1939), form a distinct group that is genetically more distant from affinis and uropygialis (c.1.5–1.6% sequence divergence) than the latter taxa are from each other (1%). Pairwise distance between eupterus and the other two is comparable to that between P. c. chrysoconus and P. c. xanthostictus (1.6%). Therefore, on the basis of equivalence and parsimony, eupterus should be resurrected as P. u. eupterus. We believe this status is merited based on the genetic divergence between eupterus and the clade including affinis and uropygialis. We appreciate, however, that based on our phenotypic yardstick analysis, morphometrics, plumage and songs differ less between affinis and eupterus, whereas uropygialis is more distinctive. Our analysis of morphometric data does not support Grote's (1928) suggestion that eupterus and affinis differ in wing length.

It should be noted that the distribution of eupterus as understood here differs from the initial description by Grote (1928). Although this taxon occurs inland as far west as northwestern Uganda in the north, and to Mbeya in the south, we infer from data presented here and in parallel work (ANGK unpubl.) that it also extends east across much of Kenya and Tanzania to the Indian Ocean coast, from the south side of Kilifi Creek, Kenya, south to the Rufiji River, Tanzania. P. u. affinis occurs from north of Kilifi Creek to the type locality at Kipini, near the Tana River, and to Serenli along the Juba River in southern Somalia, from where lollesheid (van Someren, 1932) was described (Fig. 1). The distribution of affinis extends inland from Somalia to southern Ethiopia, to the edge of Harena forest on the south slope of the Bale Mountains, where its preferred habitat of acacia woodland meets montane forest. We were unable to sample along the southern Rift in Ethiopia for logistical reasons, but samples of Red-fronted Tinkerbird from Wondo Genet, Bishangari and north across the Rift to the east side of Lake Tana and Lalibela were all uropygialis, which extends north into Eritrea and south-east along the Red Sea and Gulf of Aden coasts to northern Somalia.

The relative distributions of the three northern taxa of Red-fronted Tinkerbirds suggest that the Great Rift Valley may have played a role in their divergence. On the other hand, genetic divergence either side of Kilifi Creek is unexpected, and studies are ongoing to determine if there is gene flow between these forms (ANGK unpubl.). Tinkerbirds in this complex have been shown to hybridise, with P. p. pusillus and P. c. extoni interbreeding freely in their 50–100 km-wide contact zone in southern Africa (Nwankwo et al. 2019, Kirschel et al. 2020a), and introgressive hybridisation between the more closely related and phenotypically similar eupterus and affinis would not be unexpected if they do indeed come into contact.

Taxonomists are in much disagreement regarding how to define species limits, for example regarding the applicability of the Tobias et al. (2010) criteria (Rheindt & Ng 2021). Indeed, previous work on Pogoniulus tinkerbirds provides a relevant example in the case of ‘White-chested Tinkerbird P. makawai’. Considered a species by some based on its morphological distinctiveness (Collar & Fishpool 2006, del Hoyo & Collar 2014), but according to others it is an aberrantly plumaged Yellow-rumped Tinkerbird P. bilineatus (Dowsett & Dowsett-Lemaire 1980, 1993), a viewpoint supported by phylogenetic analysis (Kirschel et al. 2018), which demonstrated that it is nested within P. b. mfumbiri. In the present study, both a molecular phylogeny and phenotypic differences support species-level status for northern and southern forms of Red-fronted Tinkerbird according to the Phylogenetic and Evolutionary Species Concepts, and potentially even the Biological Species Concept, based on the phenotypic differences identified. Geographically parapatric cryptic taxa are more difficult to differentiate, yet such phylogenetic conservatism can result from vicariance and subsequent secondary contact, and we believe the genetic distance and even non-sister relationship of eupterus and affinis is sufficient to separate them subspecifically. The phylogeny also suggests Yellow-fronted Tinkerbird is paraphyletic with respect to the entire Red-fronted Tinkerbird clade, or polyphyletic according to unpublished work with whole genomes, meaning further research will be needed to elucidate species limits therein.

Acknowledgements

We thank the following museums (and their staff) for loans of study skins, samples, and access to their collections: Natural History Museum (Mark Adams), Natural History Museum of Los Angeles County (Kimball Garrett), Field Museum of Natural History (Dave Willard), National Museum of Natural History (James Dean), American Museum of Natural History (Paul Sweet), Museum of Natural Science of Louisiana State University (Steve Cardiff), Museum of Vertebrate Zoology of the University of California Berkeley (Jérôme Fuchs) and the Peabody Museum of Yale University (Kristof Zyskowski); we especially thank Kathy Molina at the UCLA Dickey Collection for managing the study skin loans. We thank Emmanuel Nwankwo, Athanasios Karageorgos, Andreas Ioannou, Athina Efstathiou and Anna Stylianou for assistance in the lab, Phumlile Simelane, Samuel Jones and Birhan Dessalegn for their help in the field, Guy & Yvonne Levene of Bale Mountain Lodge for their hospitality, Lincoln Fishpool for assistance with the literature, the Ethiopian Wildlife Conservation Authority (EWCA) and Eswatini Big Game Parks for permits, and Wondo Genet College of Forestry and Natural Resources, Hawassa University, for permits and support. We also thank Lincoln Fishpool, Bob Dowsett, Françoise Dowsett-Lemaire and two anonymous reviewers for comments that helped improve the submitted manuscript.

References:

1.

Ash, J. S. & Atkins, J. D. 2009. Birds of Ethiopia and Eritrea: an atlas of distribution. Christopher Helm, London. Google Scholar

2.

Chernomor, O., von Haeseler, A. & Minh, B. Q. 2016. Terrace aware data structure for phylogenomic inference from supermatrices. Syst. Biol. 65: 997–1008. Google Scholar

3.

Collar, N. J. & Fishpool, L. D. C. 2006. What is Pogoniulus makawai? Bull. Afr. Bird Cl. 13: 18–27. Google Scholar

4.

Dowsett, R. J. & Dowsett-Lemaire, F. 1980. The systematic status of some Zambian birds. Gerfault 70: 151–199. Google Scholar

5.

Dowsett, R. J. & Dowsett-Lemaire, F. 1993. Comments on the taxonomy of some Afrotropical bird species. Tauraco Res. Rep. 5: 323–389. Google Scholar

6.

Eberhard, J. R. & Bermingham, E. 2004. Phylogeny and biogeography of the Amazona ochrocephala (Aves: Psittacidae) complex. Auk 121: 318–332. Google Scholar

7.

Grant, C. H. B. & Mackworth-Praed, C. W. 1942. Notes on East African birds. (1) On the races of Pogoniulus pusillus pusillus (Dumont) occurring in eastern Africa. Bull. Brit. Orn. Cl. 63: 19–20. Google Scholar

8.

Grote, H. 1928. Neue Formen von Ostafrika, Aldabra und Madagaskar. Orn. Monatsb. 36: 77–79. Google Scholar

9.

Grote, H. 1939. Ueber Pogoniulus pusillus affinis und P. p. eupterus. Orn. Monatsb. 47: 83–84. Google Scholar

10.

Hoang, D. T., Chernomor, O., von Haeseler, A., Minh, B. Q. & Vinh, L. S. 2017. UFBoot2: improving the ultrafast bootstrap approximation. Mol. Biol. & Evol. 35: 518–522. Google Scholar

11.

del Hoyo, J. & Collar, N. J. 2014. HBW and BirdLife International illustrated checklist of the birds of the world, vol. 1. Lynx Edicions, Barcelona. Google Scholar

12.

Huelsenbeck, J. P. & Ronquist, F. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755. Google Scholar

13.

K. Lisa Yang Center for Conservation Bioacoustics. 2019. Raven Pro: interactive sound analysis software (Version 1.6.1) [Computer software]. Ithaca, NY: Cornell Lab of Ornithology, Ithaca, NY. Available from  http://ravensoundsoftware.com/Google Scholar

14.

Kalyaanamoorthy, S., Minh, B. Q., Wong, K. T., von Haeseler, A. & Jermiin, S. L. 2017. ModelFinder: fast model selection for accurate phylogenetic estimates. Nature Methods 14: 587–589. Google Scholar

15.

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P. & Drummond, A. 2012. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28: 1647–1649. Google Scholar

16.

Kirschel, A. N. G., Blumstein, D. T. & Smith, T. B. 2009. Character displacement of song and morphology in African tinkerbirds. Proc. Natl. Acad. Sci. USA 106: 8256–8261. Google Scholar

17.

Kirschel, A. N. G., Nwankwo, E. C. & Gonzalez, J. C. T. 2018. Investigation of the status of the enigmatic White-chested Tinkerbird Pogoniulus makawai using molecular analysis of the type specimen. Ibis 160: 673–680. Google Scholar

18.

Kirschel, A. N. G., Nwankwo, E. C., Pierce, D., Lukhele, S. M., Moysi, M., Ogolowa, B. O., Hayes, S. C., Monadjem, A. & Brelsford, A. 2020a. CYP2J19 mediates carotenoid colour introgression across a natural avian hybrid zone. Mol. Ecol. 29: 4970–4984. Google Scholar

19.

Kirschel, A. N. G., Nwankwo, E. C., Seal, N. & Grether, G. F. 2020b. Time spent together and time spent apart affect song, feather colour and range overlap in tinkerbirds. Biol. J. Linn. Soc. 129: 439–458. Google Scholar

20.

Kumar, S., Stecher, G. & Tamura, K. 2016. MEGA7: Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets. Mol. Biol. & Evol. 33: 1870–1874. Google Scholar

21.

Minh, B. Q., Schmidt, H. A., Chernomor, O., Schrempf, D., Woodhams, M. D., von Haeseler, A., Lanfear, R. & Teeling, E. 2020. IQ-TREE 2: new models and efficient methods for phylogenetic inference in the genomic era. Mol. Biol. & Evol. 37: 1530–1534. Google Scholar

22.

Monadjem, A., Passmore, N. I. & Kemp, A. C. 1994. Territorial calls of allopatric and sympatric populations of 2 species of Pogoniulus tinkerbarbet in southern Africa. Ostrich 65: 339–341. Google Scholar

23.

Nwankwo, E. C., Mortega, K. G., Karageorgos, A., Ogolowa, B. O., Papagregoriou, G., Grether, G. F., Monadjem, A. & Kirschel, A. N. G. 2019. Rampant introgressive hybridization in Pogoniulus tinkerbirds (Piciformes: Lybiidae) despite millions of years of divergence. Biol. J. Linn. Soc. 127: 125–142. Google Scholar

24.

Nwankwo, E. C., Pallari, C. T., Hadjioannou, L., Ioannou, A., Mulwa, R. K. & Kirschel, A. N. G. 2018. Rapid song divergence leads to discordance between genetic distance and phenotypic characters important in reproductive isolation. Ecol. & Evol. 8: 716–731. Google Scholar

25.

R Core Team. 2020. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. Google Scholar

26.

Rheindt, F. & Ng, E. Y. X. 2021. Avian taxonomy in turmoil: the 7-point rule is poorly reproducible and may overlook substantial cryptic diversity. Ornithology 138: ukab010. Google Scholar

27.

Sebastianelli, M., Blumstein, D. T. & Kirschel, A. N. G. 2021. Higher-pitched bird song towards the coast supports a role for selection in ocean noise avoidance. Bioacoustics https://doi.org/10.1080/09524622.2021.1879680 Google Scholar

28.

Short, L. L. & Horne, J. F. M. 2001. Toucans, barbets and honeyguides. Oxford Univ. Press. Google Scholar

29.

van Someren, V. G. L. 1931. New races of birds from eastern Africa. J. E. Afr. Uganda Nat. Hist. Soc. 9: 193–197. Google Scholar

30.

van Someren, V. G. L. 1932. Birds of Kenya and Uganda, being addenda and corrigenda to my previous paper in Novitates Zoologicae XXIX, 1922. Novit. Zool. 37: 252–380. Google Scholar

31.

Tobias, J. A., Seddon, N., Spottiswoode, C. N., Pilgrim, J. D., Fishpool, L. D. C. & Collar, N. J. 2010. Quantitative criteria for species delimitation. Ibis 152: 724–746. Google Scholar

32.

Torchiano, M. 2016. Effsize - a package for efficient effect size computation. https://doi.org/10.5281/zenodo.1480624 Google Scholar

33.

White, C. M. N. 1965. A revised check list of African non-passerine birds. Govt. Printer, Lusaka. Google Scholar
© 2021 The Authors;
Alexander N. G. Kirschel, Michaella Moysi, Sifiso M. Lukhele, Matteo Sebastianelli, Tsyon Asfaw, Louis Hadjioannou, Kim G. Mortega, Ara Monadjem, and Robert G. Moyle "Taxonomic revision of the Red-fronted Tinkerbird Pogoniulus pusillus (Dumont, 1816) based on molecular and phenotypic analyses," Bulletin of the British Ornithologists’ Club 141(4), 428-442, (10 December 2021). https://doi.org/10.25226/bboc.v141i4.2021.a6
Received: 15 May 2021; Published: 10 December 2021
JOURNAL ARTICLE
15 PAGES


SHARE
ARTICLE IMPACT
Back to Top