Fork-tailed Woodnymph Thalurania furcata is widely distributed in South America, but there are comparatively few data on its breeding biology. A nest found with two eggs in an urban forest fragment in Acre, south-west Amazonia, was monitored until the nestlings fledged. The use of fibres from rhizomes of the fern Phlebodium decumanum as the main material used in the nest is reported for the first time in this species. Similarly, daily variation in the mass and growth of the nestlings is presented for the first time. Nest and egg dimensions, as well as the nestling period, were similar to those reported previously.
The genus Thalurania comprises seven species distributed from Mexico to Argentina (Stiles et al. 2020). Fork-tailed Woodnymph T. furcata is the geographically most widespread species of the genus, occurring across most of tropical South America, including the Guianas, Paraguay, much of Brazil, eastern Venezuela, Colombia, Ecuador, Peru and Bolivia (Stiles et al. 2020). At least 13 subspecies of T. furcata are generally accepted (Stiles et al. 2020), 12 of which occur in Brazil (Piacentini et al. 2015) and at least three in the state of Acre (T. f. jelskii, T. f. simoni and T. f. boliviana) (Guilherme 2016). The species inhabits the edge and interior of forest, occasionally plantations and more open areas (Stiles et al. 2020). Although it occurs in a wide geographic range, information on reproduction of this species is quite limited (Stiles et al. 2020). Formal descriptions of Fork-tailed Woodnymph nests in the literature are available only from north-east Brazilian Amazonia (T. f. furcata) and Amazonian Ecuador (T. f. viridipectus) (Ruschi 1986, Greeney & Gelis 2008). We present here for the first time descriptions of the nest, eggs and nestling development of T. furcata in south-west Brazilian Amazonia.
Study area.—We studied a nest of T. furcata in the Zoobotanical Park of the Universidade Federal do Acre (UFAC) (09°57′03.22″S, 67°52′30.65″W), in the city of Rio Branco, capital of Acre state, in the south-western Brazilian Amazon. Descriptions of the ecosystems and avifauna of the area can be found in Guilherme (2001).
Measurements, monitoring and marking.—We measured the nest and eggs using digital callipers. Eggs and nestlings were weighed using a digital scale (0.05 g precision). We visited the nest every two days during the observation period. We banded the nestlings using numbered metal rings supplied by CEMAVE (Centro Nacional de Pesquisa e Conservação de Aves Silvestres), under the scope of project 1099, coordinated by EG (senior bird bander, reg. no. 324654). Once the nest was no longer active, we collected it and deposited it in the nest collection of the UFAC ornithology laboratory (reg. no. AC 0043).
On 4 January 2016 we found a nest of Fork-tailed Woodnymph in the forest understorey at the Zoobotanical Park, when it contained two eggs (Fig. 1A). The nest was sited in the two-way fork of a small understorey sapling (cf. Ocotea sp.; Lauraceae), 1.55 m above ground. The branches where the nest was built were 4.05 mm (main) and 2.41 mm (secondary) in diameter, respectively. The low cup / fork nest (sensu Simon & Pacheco 2005) was constructed primarily of fern ramenta and lined internally with pale brown kapok seed down (Ceiba sp.; Bombacaceae) (Fig. 1A–B). Comparing the ramenta of Cyathea sp. and Phlebodium sp., both common in the Zoobotanical Park, with those used in the nest, it was clear that the material in the latter was the Phlebodium decumanum rhizome (Fig. 2). There were few lichens adorning the outer wall (Fig. 1A) and spider webs on the rim of the cup (Fig. 1B). The nest measured as follows: height of the outer wall 47.0 mm; depth of cup 19.0 mm; external diameter 53.2 mm; internal diameter of cup 37.6 mm, and wall thickness 7.8 mm (the latter an average of three different measurements).
On 11 January we measured and weighed the two white eggs (Fig. 1B): 13 × 9 mm, 0.8 g; and 13 × 9 mm, 0.5 g. The nestlings hatched sometime between 12 and 13 January but were weighed for the first time only on 14 January (Fig. 3). The only traces of plumage on hatching are beige-coloured feathers on the dorsal spinal tract which are pushed up and out by the developing contour feathers. By 26 January (c.14 days after hatching) the body was wholly covered in well-developed feather papillae on the remiges and other tracts (Fig. 1C). On 29 January the nestlings were completely feathered (Fig. 1D).
On 14 January the nestlings weighed 1.2 and 1.3 g, respectively (Fig. 3). Over the next 14 days (until 28 January) the nestlings gained on average 0.47 ± 0.37 and 0.53 ± 0.36 g per day reaching the asymptote at 4.5 and 5.0 g, respectively (Fig. 3). Over the next four days their mass declined by c.0.025 and 0.12 g per day, to 4.4 and 4.5 g, respectively, on 1 February (Fig. 3) when they were last weighed. We ringed the nestlings the same day (ring codes A64412 and A64414). On 2 February just one nestling remained in the nest (Fig. 1E) and it was not present the next day. The max. duration of the nestling period was 21 days (assuming that the first nestling hatched on 12 January and fledged on 2 February 2016).
Fork-tailed Woodnymph is common in south-west Amazonia, especially in Acre (Guilherme 2016), but it is poorly known ecologically. Although three races may occur in Acre, in the east of the state, where the Zoobotanical Park is situated, the subspecies is T. f. boliviana (Guilherme 2009). The species constructs its forked nest low above ground as reported elsewhere (Stiles et al. 2020). Almost all nest biometrics reported here were similar to those reported from Ecuador by Greeney & Gelis (2008), except internal diameter. In Acre and the nests from north-east Amazonian Brazil reported by Ruschi (1986), the internal diameter of the cup was relatively larger (almost double) than that of the Ecuador nests. This difference is perhaps explained by the time when the nests were measured: in our case, these were made when the nest had already been collected, whereas the measurements of the Ecuadorian nest were taken in the field during the incubation period (Greeney & Gelis 2008). The description of the eggs, and their sizes, were practically identical to those reported by Greeney & Gelis (2008) from Ecuador, and Brazil (Stiles et al. 2020). Unlike other hummingbird nests that are extensively decorated with lichens on their outer walls for camouflage (McCormac & Showman 2009), the Acre nest had practically no external decoration (Fig. 1), somewhat like that described by Greeney & Gelis (2008) in Ecuador.
The type of material used in the cup differed slightly from that reported in Ecuador, which contained much seed down in the external wall (Greeney & Gelis 2008). Ruschi (1986) stated that Fork-tailed Woodnymph constructs the outer walls with ramenta of arborescent ferns, and lines the nest with fibres of Bromeliaceae, Gramineae and Brobacaceae. This description is quite different from the Acre nest. Comparison of the material used to construct the cup of the latter nest with ramenta of the unique arborescent ferns (Cyathea sp.) in the Zoobotanical Park was not compatible, but did match rhizome fibres of the epiphytic fern Phlebodium decumanum (Fig. 2). Use of fern ramenta of P. decumanum is reported here for the first time for T. furcata, although this type of material (not identified to species) is often reported for other hummingbirds and passerines (Greeney et al. 2010, Fjeldså et al. 2020). Indeed, the material described in the Ecuador nest as ‘brown plant fibres’ (Greeney & Gelis 2008) was clearly fern ramenta.
Development of the nestling is also presented here for the first time. Nestlings gain weight until the 16th day and then lose mass slowly until they fledge, which is similar to growth patterns in other families of birds (Seixas & Mourão 2003, Guilherme & Souza 2013) including hummingbirds (Muir & Butler 1925, Verea 2016).
We are grateful to the director of the UFAC Zoobotanical Park for permission to conduct research in the study area, and to all the members of the UFAC ornithology laboratory for their logistical support during the field work. We are grateful to CNPq for providing JML with an undergraduate research stipend and to the Centro Nacional de Pesquisa e Conservação de Aves Silvestres (CEMAVE / ICMBio) for providing the bands used in project 1099. We also thank Prof. Marcos Silveira and Martin Acosta Oliveira for their help in identifying plant species, and Harold F. Greeney for refereeing the submitted manuscript.