Of all the major rivers in western Amazonia, the Putumayo is arguably the least known ornithologically. Approximately 1,600 km in length, the Putumayo—known as the Içá in Brazil—is the tenth longest tributary of the Amazon (Goulding et al. 2003). Rising in the southern Colombian Andes, it marks the boundary between Colombia and Peru for much of its length; its basin is sparsely populated, and for the most part difficult to access. Several recent surveys by the Field Museum of Natural History (FMNH), Chicago, have shed light on avian distributions immediately south of the Putumayo in Peru (Pitman et al. 2004, Alverson et al. 2008, Gilmore et al. 2010, Pitman et al. 2011, 2013, 2016), but the avifauna on the Colombian side is essentially unknown. Ornithological exploration of the Putumayo–Caquetá interfluvium has commenced only very recently, with much of the field work in the vicinity of Puerto Leguizamo (Bonilla-Castillo et al. 2017); the Colombian side of the Putumayo downstream of Puerto Leguizamo has been almost completely neglected by ornithologists and birders. Here we report on several avian range extensions from a site on the Colombian bank of the Putumayo c.240 km downriver of Puerto Leguizamo, and discuss their biogeographical implications.

Study sites and Methods

On 1–9 February 2017, we undertook ornithological field work around the indigenous Muiri communities of El Encanto (01°44′40″S, 73°12′24″W), San Rafael del Caraparaná (01°41′15″S, 73°13′55″W) and Tercera India (01°42′32″S, 73°13′41″W) in dpto. Amazonas, Colombia. El Encanto is located at the confluence of the Putumayo and Caraparaná Rivers, while San Rafael and Tercera India are both a few kilometres upstream along the Caraparaná. Elevation at all three sites is c.140 m. Rainfall data from Puerto Alegría, c.120 km to the north-west along the Putumayo River, indicates that February is the second wettest month after March, with the driest period in June–August (OEA 1993); we experienced some rain on most days, with several very heavy afternoon downpours. To our knowledge, the only previous bird survey of this area was by J. Beckers in March 2015. We concentrated our efforts on several trails through terra firme forest on clay soils in the vicinity of the three villages; less time was spent in pasture and second growth in the immediate environs of settlements. We walked trails from dawn to early afternoon conducting qualitative observations and attempting to document interesting sightings using digital photography and sound-recordings using an Olympus LS-11 recorder with Sennheiser ME66 microphone.

Results

We recorded a total of 197 species during our field work (Appendix 1). All of our records, including 116 photos of 45 species, are archived in eBird ( www.ebird.org); we documented 40 species with sound-recordings, the more significant of which—nine recordings of seven species—are deposited at  www.xeno-canto.org. Major range extensions and other noteworthy records are detailed below, with supporting documentation indicated by its archive number in xeno-canto (XC) or the Macaulay Library (ML). Taxonomy and nomenclature follow Remsen et al. (2018).

COMMON NIGHTHAWK Chordeiles minor

A massive dusk flight over the Caraparaná River at San Rafael on 5 February was conservatively estimated to number 250 individuals (ML 59418331, 59418331, 59418481). Data on the species' winter range are limited (Brigham et al. 2011) and, while it occurs regularly in western Amazonia, this appears to be an exceptionally high count. There are no published counts of more than ten in Ecuador (Freile et al. 2017) and in Loreto, Peru, where the species is known to winter, the only eBird counts in double digits are from October, and presumably refer to migrants. Further north in the Colombian Amazon—where the species was recorded for the first time by Stiles (2010)—counts from the vicinity of Mitú, Vaupés, have reached 50 in January and up to 150 in late February (eBird data), the latter possibly involving early spring migrants. Our count from early February should represent wintering birds and suggests that Common Nighthawks may be more numerous in winter than previously thought in this part of Amazonia. Interestingly, for a species generally associated with open habitats, our record is from a heavily forested area.

Figure 1.

Green-tailed Goldenthroat Polytmus theresiae, San Rafael del Caraparaná, Amazonas, Colombia, 8 February 2017 (Ottavio Janni)

WHITE-SHOULDERED ANTSHRIKE Thamnophilus aethiops

One heard singing in terra firme forest near El Encanto on 3 February, and at least three were seen, photographed (male ML 49753351; female ML 49753331, 49753341; Fig. 2) and soundrecorded (XC 357343) in a treefall gap in terra firme forest near Tercera India on 7 February. T. aethiops is polytypic and a recent genetic study recovered nine reciprocally monophyletic lineages equating to evolutionary species, but not fully matching current subspecies limits (Thom & Aleixo 2015). Our records are from an area where T. aethiops was not previously known, and as such their taxonomic affiliation is of special interest. Four different subspecies occur within 500 km of El Encanto. Race kapouni, which is unknown north of the Amazon, occurs as close as c.210 km south of El Encanto on the south bank of the Amazon opposite the mouth of the Napo River (Robbins et al. 1991). North of the Amazon, the closest records of T. aethiops to our study site are from the Serranía de Chiribiquete c.215 km to the north (Álvarez et al. 2003). Examination of these specimens by J. E. Avendaño (in litt. 2018) suggests they best match the poorly known race wetmorei, for which the nearest otherwise confirmed records to our study site are from the foothills of the Serranía de los Churumbelos, dpto. Cauca, c.450 km to the north-west (Salaman et al. 1999). Records from the Zona Reservada Pucaruro, dpto. Loreto, Peru (Diaz-Alván et al. 2017) approximately 265 km to the west-southwest, and Yasuní National Park, Orellana province, Ecuador (XC86523) c.375 km west-southwest, have been assigned to the nominate race, of which the closest specimen records are from Santa Cecilia, Sucumbíos province, Ecuador, c.460 km west-northwest of our study site (Ridgely & Greenfield 2001). The status of the nominate race in Colombia is somewhat confused, as Hilty & Brown (1986) reported it to occur in south-eastern dpto. Nariño, and it was included in a recent effort to list all bird taxa known from Colombia (Verhelst-Montenegro & Salaman 2015)—presumably based on Meyer de Schauensee (1952), who reported two males and a female collected along the San Miguel and Churuyacu Rivers in extreme south-east Nariño, c.510 km west-northwest of our study site—yet it is not listed by recent field guides (McMullan & Donegan 2014, Ayerbe-Quiñones 2018, McMullan 2018) and taxonomic references (del Hoyo & Collar 2016), perhaps due to the earlier reference being overlooked. There are also records of T. aethiops from Estación Biológica Caparú, dpto. Vaupés, approximately 415 km to the east-northeast (Bennet-Defler 1994); the race concerned is unknown but is presumed—on range—to be polionotus, the closest confirmed records of which come from the lowermost Caquetá / Japurá River in Brazil, c.800 km east of our study site (Thom & Aleixo 2015). Assigning our records to race is problematic without specimens. Plumage perhaps best matches polionotus: the female was essentially uniformly rufous, whereas in female wetmorei the mantle should be much paler, which makes the rufous cap stand out more clearly from the rest of the upper parts’ (Meyer de Schauensee 1945); the male was rather uniformly blackish grey, with a darker crown and white spots on the wing-coverts, coincident with polionotus / wetmorei, and should rule out the nominate race, which is mostly black. However, we cannot eliminate wetmorei and, given how distant our records are from other known localities, the possibility of an undescribed subspecies also cannot be discarded (A. Aleixo in litt. 2018). In Peru, T. aethiops is unknown north of the Amazon and it has not been recorded on any of the rapid biological inventories conducted between the Napo and Putumayo Rivers. Given that in the T. aethiops complex, all splits involving the recognized lineages coincide with at least part of the course of a major Amazonian river’ (Thom & Aleixo 2015), it might be speculated that the Putumayo River is also one such barrier.

Figure 2.

Female White-shouldered Antshrike Thamnophilus aethiops, Tercera India, Amazonas, Colombia, 7 February 2017 (Ottavio Janni)

BLACK-HEADED ANTBIRD Percnostola rufifrons minor

This species, first recorded by J. Beckers in March 2015, was fairly common in the area, with 1–5 encountered daily at all three sites. As the area between the Putumayo and Caquetá Rivers lacked any records of P. rufifrons, our finding raises questions as to which taxon of the species occurs there. While P. r. jensoni is fairly widespread on the south bank of the Putumayo in Peru, including just 30 km south of our study site along the lower Ere River (Pitman et al. 2013), the nearest published records of minor appear to be from Chiribiquete National Park, Caquetá, c.250 km north-east (Álvarez et al. 2003) and Estación Biológica Caparú, Vaupés, 415 km to the east-northeast (Isler et al. 2001); closer to our study site are two unpublished specimens from La Chorrera (IavH-A 1563, 5756; J. E. Avendaño in litt. 2018). We sound-recorded several individuals (XC 357329, 357334 and 357339; Fig. 6) and photographed a male (ML 49753111; Fig. 3) and female (ML 49752771, 49752761), which we identified as race minor based on plumage (dark reddish-brown crown and ferruginous head-sides in the female vs. blackish-grey crown and greyish head-sides in jensoni) and vocal features (song slower paced than in jensoni; the loudsong pace in our recordings—3.03–3.52 notes / second—matches and in one case slightly exceeds the pace of loudsong reported in Isler et al. 2001 for minor—2.67–3.31 notes / second—while they are well outside that reported for jensoni—4.09–4.50 notes / second; M. L. Isler in litt. 2018; compare our recording in Fig. 6 to XC 56750 in Fig. 8). Our findings confirm the suggestion by Capparella et al. (1997) that the Putumayo / Içá River separates these two taxa. Genetic work on species limits in the P. rufifrons complex is ongoing (G. A. Bravo in litt. 2017) and our findings shed light on the geographic limits of these taxa. Interestingly, the birds responded strongly to playback of the song of P. r. jensoni from Loreto, Peru (XC 56750), albeit during unstructured playback experiments and with the caveat that assessing lack of response is likely to be more taxonomically significant. More formal playback trials along the lines of those conducted for other allopatric, closely related taxa in the Neotropics (Freeman & Montgomery 2017) and including calls, could provide additional data by which to assess the taxonomic rank of these two taxa.

Figure 3.

Male Black-headed Antbird Percnostola rufifrons minor, El Encanto, Amazonas, 4 February 2017 (Ottavio Janni)

Discussion

Although our survey was brief, our findings add to the limited but growing body of data on the birds of the Putumayo basin. Previous surveys have concentrated on the Peruvian side of the river, and our records highlight some interesting similarities and differences between the avifaunas on the north and south banks. One similarity is the presence of a suite of birds associated with poor-soil habitats: in addition to Polytmus theresiae and Percnostola rufifrons, other poor-soil specialists we found included Pearly Antshrike Megastictus margaritatus (ML 58147011, XC 357335), Cinnamon Manakin-Tyrant Neopipo cinnamomea (ML 59871381, 59871431, 59871491, 59871511; Fig. 4) and Citron-bellied Attila Attila citriniventris, despite the absence of white-sand formations. Several poor-soil specialists were also found at some of the FMNH study sites south of the Putumayo, suggesting that such species are widespread in the basin.

Figure 4.

Cinnamon Manakin-Tyrant Neopipo cinnamomea, Tercera India, Amazonas, Colombia, 7 February 2017 (Ottavio Janni)

Figure 5.

Imeri Warbling Antbird Hypocnemis flavescens loudsong, recorded at San Rafael del Caraparaná, Amazonas, Colombia, on 9 February 2017; sonogram produced by Michele Viganò from recording XC 357330 ( www.xeno-canto.org) by Ottavio Janni.

Figure 6.

Black-headed Antbird Percnostola rufifrons minor loudsong, recorded at San Rafael del Caraparaná, Amazonas, Colombia, on 6 February 2017; sonogram produced by Michele Viganò from recording XC 357334 ( www.xeno-canto.org) by Ottavio Janni.

Figure 7.

Peruvian Warbling-Antbird Hypocnemis peruviana loudsong, recorded at El Para, Napo, Ecuador, on 24 January 2011; sonogram produced by Michele Viganò from recording XC 72568 ( www.xeno-canto.org) by Andrew Spencer.

Figure 8.

Black-headed Antbird Percnostola rufifrons jensoni loudsong, recorded at Otorongo Lodge, Loreto, Peru, on 27 May 2010; sonogram produced by Michele Viganò from recording XC 56750 ( www. xeno-canto.org) by David Geale.

Perhaps more interestingly, our findings evidence that the Putumayo River is a biogeographic barrier. Although the role of rivers in shaping the distribution and speciation of Amazonian fauna has long been known (Wallace 1852) and extensively studied (Haffer 1969, Weir et al. 2015), the role of the Putumayo has received virtually no attention in the ornithological literature, despite evidence that some vertebrates, such as Red Titi Monkey Callicebus discolor, are bound by it (van Roosmalen et al. 2002). Our data begin to address this gap: we found that Hypocnemis flavescens and H. peruviana replace one another across the Putumayo, contrary to published maps, and confirmed the suspicion that the Putumayo also separates Percnostola rufifrons minor and P. r. jensoni. The lack of records of Thamnophilus aethiops from the Peruvian side of the Putumayo despite a number of surveys there strongly suggests that the river serves as a barrier also to this taxon. Several species not yet recorded in Colombia—including an undescribed antwren Herpsilochmus sp., Variegated Antpitta Grallaria varia and Orange-crowned Manakin Heterocercus aurantiivertex—have recently been found across the Putumayo in Peru (Pitman et al. 2004, Gilmore et al. 2010, Pitman et al. 2011), but ornithological field work on the Colombian side of the river—and indeed in the Colombian Amazon as a whole (Avendaño et al. 2017)—has been so limited that it is difficult to gauge whether the lack of records provides additional evidence of the Putumayo's role as a barrier, or is merely due to these species remaining undetected to date.

The role of rivers in defining areas of bird endemism in the Amazon basin is also under debate (Naka et al. 2010, Fernandes et al. 2014), with a recent review finding that rivers mark steep changes in species composition, but do not limit areas of endemism’ (Oliveira et al. 2017). Traditionally, the Putumayo basin has been thought to belong to the North Amazon / Napo centre of endemism (Cracraft 1985), with the Negro / Vaupés River generally mapped as separating it from the Imeri centre to the north (Silva et al. 2005, Naka et al. 2012, Oliveira et al. 2017). However, several taxa we found at our study site— especially Hypocnemis flavescens and Percnostola rufifrons minor, and potentially Thamnophilus aethiops if racial assignment to polionotus is confirmed—are limited to the Imeri centre of endemism, and in the latter two cases they replace sister taxa from the Napo centre across the Putumayo. On the other hand, the only Rhegmatorhina we found at our study site was Hairy-crested Antbird R. melanosticta (XC 357344), whose range encompasses the Napo centre of endemism, and not the Imeri centre representative, Chestnut-crested Antbird R. cristata, for which species the Caquetá / Japurá River was recently confirmed as a distributional limit (Ribas et al. 2018). Our records of Selenidera toucanets were heard only, and we could not determine if they referred to Golden-collared Toucanet S. reinwardtii or the Imeri endemic Tawny-tufted Toucanet S. nattereri. The picture that emerges is of the area between the Putumayo and Vaupés Rivers being a zone of transition between the Napo and Imeri centres of endemism, with taxa typical of the latter having been found well to the south-west of the centre as traditionally defined, even beyond the geological boundaries of the Guiana Shield (Osmonson et al. 2000). Similar transitional areas have been found, for example, in the Branco / Negro interfluvium in Brazil, where two distinct avifaunas from west of the Negro and east of the Branco meet (Naka et al. 2010). Recent records of Yellow-throated Antwren Myrmotherula ambigua and Rose-breasted Chat Granatellus pelzelni from the Puerto Leguízamo area (D. Calderón-Franco et al. in prep.) further underscore the Guianan / Imeri flavour’ to the avifauna on the Colombian bank of the Putumayo, and hint at the complexity of distributional patterns in the poorly explored Colombian Amazon.

Additional field work on the Colombian side of the Putumayo is certain to provide useful data to further unravel these patterns, and to address distributional questions at a finer scale, including the extent to which the Caquetá River, which runs almost parallel to the Putumayo to the north and is even longer (c.2,800 km, making it the fourth-largest tributary of the Amazon, Goulding et al. 2003), also constitutes a biogeographical barrier. Genetic studies of avian populations either side of the Putumayo would also shed light on the river's role as a barrier, especially given increasing awareness of cryptic diversity in Amazonian birds (Whitney & Cohn-Haft 2013), and of the fine-scale nature of endemism in Amazonia (Fernandes 2013). Recent field work on the Colombian side of the Putumayo has already resulted in one new species for the national list—Cocha Antshrike Thamnophilus praecox (Williams 2016, Janni et al. in prep.)—and more such discoveries surely await. Of the species mentioned above as occurring on the Peruvian side of the Putumayo, Heterocercus aurantiivertex seems especially likely to occur in Colombia as well, as its range is very similar to that of Thamnophilus praecox. One obstacle to field work along the Colombian Putumayo is difficulty of access: the area downstream of Puerto Asís lacks roads and transport is by boat along the main waterways, with Puerto Leguízamo the only town easily accessed by air. From the latter, a weekly boat visits San Antonio del Estrecho on the Peruvian bank, a journey of c.12 hours; El Encanto is about three-quarters of the way downstream and is one of the few settlements where some accommodation can be secured. The small settlement of La Chorrera, c.55 km north-east of our study site, is reached by a weekly flight from Leticia, and could also serve as a base for exploration. In part due to these logistical challenges, the Colombian part of the Putumayo basin is one of the frontiers of Amazonian ornithology, and we strongly encourage its continued exploration.