56 Accepted by T. Chesser: 11 Jan. 2010; published: 9 Feb. 2010
ISSN 1175-5326 (print edition)
ISSN 1175-5334 (online edition)
Copyright © 2010 · Magnolia Press
Zootaxa 2354: 56–66 (2010)
A taxonomic revision of the Paramo Tapaculo Scytalopus canus Chapman (Aves:
Rhinocryptidae), with description of a new subspecies from Ecuador and Peru
NIELS KRABBE1 & CARLOS DANIEL CADENA2
1Zoological Museum, University of Copenhagen, Universitetsparken 15, DK-2100, Copenhagen, Denmark.
2Laboratorio de Biología Evolutiva de Vertebrados, Departamento de Ciencias Biológicas, Universidad de los Andes, Apartado 4976,
Bogotá, Colombia. E-mail: firstname.lastname@example.org
The nominate subspecies of the Paramo Tapaculo (Scytalopus canus Chapman), which is restricted to high elevations of
the Western Andes of Colombia, differs substantially in song from S. c. opacus from the Central Andes of Colombia,
Ecuador, and northern Peru. In accordance with current taxonomy of the genus, which is primarily based on differences
in song, we assign species rank to both taxa. Within opacus, birds from southernmost Ecuador and northernmost Peru
sing like northern opacus but call differently. They are indistinguishable morphologically except for a white patch in the
wing present in 10 out of 12 of the adult males of the southern population. We describe this population as a new taxon
and rank it as a subspecies of S. opacus. Genetically, S. canus, nominate S. opacus, and the new taxon are strongly
differentiated (>5% divergence in mtDNA); the first two appear to be sister taxa.
Key words: Scytalopus, tapaculo, Andes, Colombia, species limits, suboscines
[Una revisión taxonómica del Tapaculo de Páramo Scytalopus canus (Aves: Rhinocryptidae), con una descripción de una
nueva subespecie de Ecuador y Perú]
La taxonomía a nivel de especies de muchos tapaculos del género Scytalopus (Aves: Rhinocryptidae) no ha sido aclarada.
La subespecie nominal del Tapaculo de Páramo (Scytalopus canus Chapman), que se encuentra restringida a las
elevaciones altas de la Cordillera Occidental de los Andes de Colombia, difiere sustancialmente en el canto en
comparación con S. c. opacus de la Cordillera Central de Colombia, Ecuador y el norte de Perú. De acuerdo con la
taxonomía actual del género, la cual se basa principalmente en diferencias en el canto, le asignamos el rango de especie a
ambos taxones. Las aves del extremo sur de Ecuador y el extremo norte de Perú cantan como opacus pero tienen
reclamos diferentes. Éstas son indistinguibles de opacus en su morfología, excepto porque 10 de 12 machos adultos
presentan un parche blanco en cada ala. Describimos esta población como un nuevo taxón, que tratamos como una
subespecie de S. opacus debido a la similitud de sus cantos y morfología. Sin embargo, análisis adicionales podrían
indicar que merecen tratarse como especies diferentes. Genéticamente, S. canus, la forma nominal de S. opacus y el
taxón nuevo están marcadamente diferenciados (>5% de divergencia en ADN mitocondrial); las primeras dos formas
parecen ser hermanas.
Palabras clave: Andes, Colombia, límites de especies, suboscines
Scytalopus tapaculos constitute one of the greatest challenges to avian taxonomy. Many species look so
similar that some specimens cannot be identified with certainty. Voice is believed to be the primary means of
Zootaxa 2354 © 2010 Magnolia Press · 57
A TAXONOMIC REVISION OF SCYTALOPUS CANUS
communication, and individuals with similar vocalizations have been found to be genetically similar
(Arctander & Fjeldså 1994). Accordingly, vocal characters have been instrumental in defining species limits
in the genus, and in recent years the number of recognized species has risen from ten to nearly forty (Krabbe
& Schulenberg 1997; Remsen et al. 2009). The Paramo Tapaculo (Scytalopus canus Chapman) is common in
treeline scrub in the Northern Andes of Colombia, Ecuador, and northern Peru. It was described by Chapman
(1915) on the basis of ten specimens collected at an elevation of 12500 feet (3810 m) in the Páramo de
Paramillo, northwestern Antioquia Department, Western Andes of Colombia. Nearby Páramo de Frontino was
later added to its known range by Hilty & Brown (1986), who mentioned two immatures collected there. Cory
& Hellmayr (1924) referred a specimen collected at La Leonera, Central Andes, Caldas Department, to canus,
but Krabbe & Schulenberg (1997) treated this allocation with some doubt. Zimmer (1939) described two
specimens from Tambillo on the upper Río Upano in Morona-Santiago Province, Ecuador, as a new taxon
(obscurus, now known as opacus; Zimmer 1941), diagnosed from canus by its longer wings and tail and by a
small amount of brownish color with dark bars on the posterior underparts; these parts are uniform gray in
canus. Zimmer considered canus and opacus each other’s closest relatives, but ranked them as subspecies of a
broad S. magellanicus (Gmelin).
Krabbe & Schulenberg (1997) afforded species rank to all vocally known taxa in Zimmer’s S.
magellanicus. They showed that all, including opacus, have distinctive songs, but were unable to assess the
taxonomic status of canus because it was vocally unknown. Therefore, they left opacus, its presumed closest
relative, as a subspecies of canus. They referred to S. c. opacus two distinct populations, one north, the other
south of the Río Zamora, with similar songs but drastically different calls.
We here provide evidence that although canus and opacus are indeed each other's closest relatives, the
two have distinct songs that are as different as or more different than those of other species in the genus and
are highly differentiated genetically, suggesting they should be considered different species, with canus
genuinely restricted to the Western Andes of Colombia. We also name the southern population of opacus as a
new taxon, which we conservatively treat as a subspecies because it and the northern population have similar
On 6–12 August 2004, NK made tape recordings of ca.15 individuals of nominate S. canus at the treeline on
Páramo de Frontino at elevations ranging from 3300–3500 m (see details on study site in Krabbe et al. 2005).
A single adult male was collected; the skin specimen was deposited in the Instituto de Ciencias Naturales,
Universidad Nacional de Colombia (ICN 35020) and a tissue sample in the collection of Instituto Alexander
von Humboldt (IAvH-CT 5178). The recordings were compared to recordings of the two populations of
opacus (Table 1), all publicly accessible at www.xeno-canto.org, Macaulay Library of Natural Sounds,
Cornell University, or in Banco de Sonidos Animales, Instituto Alexander von Humboldt, Colombia.
Specimens of opacus in the Zoological Museum, University of Copenhagen (ZMUC) and data on other
museum specimens obtained previously (Krabbe & Schulenberg 1997) were reexamined. Measurements and
notes on plumage of the specimen from Depto. Caldas in the Central Andes were obtained. Plumage
descriptions of four additional specimens in the Museum of Natural Science, Louisiana State University
(LSUMZ) of southern opacus from Zamora-Chinchipe, Ecuador (83373, adult male) and Cajamarca, Peru
(169860-62, immature male, adult female, juv male, respectively) were obtained.
TABLE 1. Number of tape-recordings of different individuals examined.
Taxon Song Calls
S. canus canus 14 10
S. canus opacus northern 27 33
S. canus opacus southern 11 10
KRABBE & CADENA58 · Zootaxa 2354 © 2010 Magnolia Press
CDC examined phylogenetic relationships and patterns of genetic variation within the S. canus complex
using DNA sequences of the mitochondrial gene ND2 (c. 1000 base pairs) obtained following methods
described in a recent study (Maurício et al. 2008). Analyses were based on sequence data from that study
(including sequences of the specimen of S. c. canus collected by NK at Páramo de Frontino and of Scytalopus
vicinior, S. stilesi, S. pachecoi, S. magellanicus, and as outgroup, Eugralla paradoxa), supplemented with new
data obtained for three specimens of S. canus opacus. These include one specimen from 100 km north of the
Río Zamora (ZMUC 125150, from Zapote Najda Mts., Morona-Santiago, Ecuador, 3°01’S, 78°39’W, 3450
m), one from 100 km south of the Río Zamora (ZMUC 125688, from the upper Río Isimanchi, Zamora-
Chinchipe, Ecuador 4°45’S, 79°25’W, 3150 m), and one from 100 km further south, in northern Peru
(LSUMZ B31950, from Quebrada Lanchal above Sallique, Cajamarca, 5°41’S, 79°15’W, 2900 m; see
acknowledgements for museum acronyms). In addition, our data set included new sequences for two
specimens representing species that belong in lineages that are closely related to the canus (sensu lato) group
according to preliminary analyses that considered nearly all species in the genus (C.D. Cadena et al., unpubl.
data): S. affinis Zimmer (ZMUC 125154) and S. superciliaris Cabanis (MBM 8242). New sequences were
deposited in GenBank (accession numbers GU325813-GU325817).
Phylogenetic relationships among the selected taxa were reconstructed using maximum-likelihood,
Bayesian inference, and maximum parsimony methods. The maximum-likelihood analysis was conducted
using default settings in the program RAxML (Stamatakis 2006) run on the CIPRES Portal (http://
www.phylo.org); nodal support was estimated using 1000 bootstrap replicates (Stamatakis et al. 2008). The
program BEAST (Drummond & Rambaut 2007) was employed for Bayesian inference using the HKY+G
model (selected as the best fit to the data based on the Akaike Information Criterion in Modeltest; Posada &
Crandall 1998) and an uncorrelated relaxed lognormal clock model with a Yule process tree prior; the analysis
was run twice independently for 5 x 107 generations, of which the first 1 x 107 were discarded as burn-in in
each case. Results from both runs were entirely consistent and both resulted in large effective sample sizes for
parameters, suggesting they adequately sampled the posterior distributions. As a complement to these model-
based analyses, we performed a parsimony bootstrap analysis with 1000 replicates in the program PAUP*
The male taken at the Páramo de Frontino is the first adult specimen of S. c. canus collected away from the
type locality, Páramo de Paramillo. These two localities are separated by c. 70 km with no pass lower than
2150 m elevation between them (passes 300 m lower within the range of northern opacus have not caused
discernible differentiation in that form). Additionally, the specimen shows no trace of dark barring or brown in
the flanks, and its wings and tail measure 52 and 35 mm, respectively (Table 2). Thus, it agrees with the
description of the type series of canus (Chapman 1915, additional measurements given by Cory & Hellmayr
1924), leaving little doubt that it represents this taxon. Its body mass was 14.5 g, the first recorded for
nominate S. canus. The male specimen from La Leonera in the Central Andes (Carnegie Museum 70683)
shows traces of barring on the upper tail coverts, flanks and thighs, and has a wing of 54.5 mm and a tail of
40.2 mm (Table 2). This suggests that it belongs with opacus, as suspected by Krabbe & Schulenberg (1997),
an allocation further established by song and call notes tape-recorded nearby in the Central Andes (see
below). Thus, canus appears to be restricted to the Western Andes.
The song of canus is of about the same duration as the song of opacus (4–12 s) but differs in five aspects
(Table 3; Figs. 1–2): 1) it is much slower paced, 7–11 instead of 30–40 notes per s (Fig. 1); 2) the notes are
churring rather than simple down- or up-down-strokes (Fig. 1); 3) the pace changes through the song,
accelerating 14–43% instead of being constant or more commonly decelerating 4–27%; 4) the pitch falls by
three to seven half notes on a diatonic scale, most drastically during the beginning of the song, instead of
being constant or falling more gradually and at the most by two half notes (Fig. 2); 5) the pitch is higher, 3.5–
5.6 instead of 3.1–4.1 kHz (Figs. 1–2). Both song and call notes of birds from the northern part of the Central
Andes of Colombia are similar to those of S. c. opacus.
Zootaxa 2354 © 2010 Magnolia Press · 59
A TAXONOMIC REVISION OF SCYTALOPUS CANUS
TABLE 2. Wing and tail measurements for S. c. canus, populations of S. c. opacus from north and south of Río Zamora,
and two specimens from Páramo de Frontino, Antioquia (ICN 35020) and La Leonera, Caldas (CM 70683).
Measurements of the Antioquia specimen agree with S. c. canus, those of the Caldas specimen with S. c. opacus.
Measurements given of nominate S. canus are from the type series, those of S. c. opacus from Krabbe & Schulenberg
TABLE 3. Some properties of songs of Scytalopus c. canus and northern and southern populations of S. c. opacus.
Positive values for Pace are indicated with the prefix ‘acc’ (accelerating), negative values with ‘dec’ (decelerating).
The pitch given is that of the second harmonic. Pitch is negative for all three and is measured in half notes on the
FIGURE 1. Half second sequences of songs of (A) Scytalopus c. canus from Frontino, Antioquia, Colombia, (B)
northern S. c. opacus from Azuay, Ecuador, and (C) southern S. c. opacus from Zamora-Chinchipe, Ecuador. Notice the
churring, higher pitched and much slower paced notes of S. c. canus. Differences in note quality between northern and
southern S. c. opacus are not constant.
The calls of canus are much like those of northern populations of opacus (Fig. 3). They resemble them in
quality and pattern, but are higher pitched, 4.5–5.6 instead of 3.1–4.2 kHz. They also tend to be composed of
fewer notes, usually 3–5 instead of 5–8, that are given at a slower pace (4–6 instead of 5–10 per s). Calls of
Taxon Sex Wing flat (mm): mean (range) n Tail (mm): mean (range) n
S. c. canus M52.5 (50–55) n=2 35 n=2
S. c. canus (ICN 35020) M52 35
S. c. opacus (CM 70683) M54.5 40.2
S. c. opacus (northern) M57.2 (52–63) n=22 42.8 (35.4–46.7) n=19
S. c. opacus (southern) M57.8 (55–59) n=8 43.4 (39–47) n=8
S. c. canus F52.6 (52–53) n=3 33.8 (32–35) n=3
S. c. opacus (northern) F55.3 (54–56) n=4 39.8 (39–42) n=4
S. c. opacus (southern) F55.5 (55–56) n=2 40.4 (39–41.7) n=2
Taxon NDuration (s) Pace
(notes/s) Pace (%)
mean and range Pitch (kHz) Pitch
S. c. canus 9 4–12 7–11 acc 27 (14–43) 3.5–5.5 3–7
S. c. opacus northern 21 4–12 29–40 dec 7 (0–11) 3.2–4.0 0 (-2)
S. c. opacus southern 9 4–12 (–120) 29–43 dec 16 (4–27) 3.1–4.1 0 (-2)
KRABBE & CADENA60 · Zootaxa 2354 © 2010 Magnolia Press
birds from the Central Andes of Colombia (Huila, Quindío) are composed of 5–12 notes, paced 7–9 per s, and
pitched 3.1–3.4 kHz, and are in all other aspects typical of opacus. The calls of birds from southern Ecuador
and northernmost Peru, however, differ strikingly from calls of more northerly populations. They are like one
second long reminiscences of song, with notes given at 19–27 per s at a slightly decreasing pace, with pitch
constant or falling up to two half notes (Fig. 3).
FIGURE 2. Complete songs of (A) Scytalopus c. canus from Frontino, Antioquia, Colombia, (B) northern S. c. opacus
from Azuay, Ecuador, and (C) southern S. c. opacus from Zamora-Chinchipe, Ecuador. Notice that in S. c. canus the
pitch falls rapidly in the beginning of the song. As in the example shown, most songs of northern S. c. opacus have one to
three introductory notes. The song length of all three ranges from 4 to 12 s.
Zootaxa 2354 © 2010 Magnolia Press · 61
A TAXONOMIC REVISION OF SCYTALOPUS CANUS
FIGURE 3. Calls of (A–B) Scytalopus canus canus from Frontino, Antioquia, Colombia, (C–D) northern S. c. opacus
from Tungurahua and Carchi, Ecuador, and southern S. c. opacus from (E) Piura, Peru and (F–G) Zamora-Chinchipe,
Molecular data indicated the existence of three well-differentiated genetic lineages within the S. canus
complex: S. c. canus, the northern form of S. c. opacus and the southern form of S. c. opacus. Pairwise
mtDNA sequence divergence between members of these three forms exceeded 5%, whereas divergence
between individuals of the southern form of S. c. opacus (the only one for which we had data for more than
one individual) were negligible (Table 4). The three phylogenetic methods employed (maximum-likelihood,
Bayesian inference, and maximum parsimony) indicated that S. canus canus is sister to the northern form of S.
c. opacus, a relationship that received strong support in maximum-likelihood and maximum parsimony
analyses (94% and 83% bootstrap), but was weakly supported in the Bayesian analysis (posterior probability
0.74 ; Fig. 4).
TABLE 4. Uncorrected sequence divergence (p distance) in the ND2 mitochondrial gene between specimens in the
Scytalopus canus complex. Note the strong differentiation between nominate S. canus (1), the northern form of S. c.
opacus (2), and the southern form of S. c. opacus (3–4).
Our analyses reveal the existence of pronounced geographic variation in songs, calls, and mitochondrial DNA
sequences within the Scytalopus canus complex. This variation is also matched by some degree of
morphological variation. The differences in song between canus and opacus are as great as or greater than the
differences between several other forms in Zimmer’s broad S. magellanicus, which were elevated to
biological species rank by Krabbe & Schulenberg (1997). Accordingly, we propose to rank these taxa as two
separate biological species, Paramillo Tapaculo (S. canus) and Paramo Tapaculo (S. opacus). Being the better-
Taxon 1 2 3
1. S. canus canus ICN 35020 -
2. S. canus opacus ZMUC 125150 0.05604 -
3. S. canus opacus ZMUC 125688 0.05320 0.05319 -
4. S. canus opacus LSUMZ B31950 0.05237 0.05242 0.00097
KRABBE & CADENA62 · Zootaxa 2354 © 2010 Magnolia Press
known species, we suggest retaining the English name Paramo Tapaculo for S. opacus and propose a new
name for the heretofore virtually unknown S. canus.
FIGURE 4. Phylogram showing relationships among members of the Scytalopus canus complex and near relatives
estimated based on maximum-likelihood (ML) analysis of sequences of the mitochondrial gene ND2. Values on the
nodes indicate ML bootstrap support, MP bootstrap support, and Bayesian posterior probabilities.
Apart from their distinctive calls, populations of S. opacus from south of the Río Zamora can be
distinguished from those north of this river by the white greater primary coverts (see illustration in Krabbe &
Schulenberg 2003) present in 7 out of 9 adult males (and three more observed in the field). This includes
specimens of males collected 50 km apart, but none of 5 female specimens. Thus, at least one distinctive
plumage feature allows the recognition of southern populations as a separate taxon. Additionally, the southern
population differs genetically from the northern populations by more than 5% sequence divergence in the
ND2 gene. Despite these differences, which are sufficient grounds to consider both forms of S. opacus as
different phylogenetic (Cracraft 1983) and evolutionary (Wiley 1978) species, these two populations do not
differ appreciably in songs. Current species-level taxonomy of this group follows the biological species
concept (sensu Johnson et al. 1999) and is based largely on differences in songs (Krabbe & Schulenberg 2003;
Remsen et al. 2009).
Correspondence and geographic cohesiveness between distinctive song and genetic differentiation has
been shown for a number of Scytalopus taxa (Arctander & Fjeldså 1994; Cuervo et al. 2005; Bornschein et al.
2007; CDC et al. unpublished data). Likewise, some forms with similar songs but different calls, i.e., S.
pachecoi Maurício and S. diamantinensis Bornschein et al. (Bornschein et al. 2007) and southern and
Zootaxa 2354 © 2010 Magnolia Press · 63
A TAXONOMIC REVISION OF SCYTALOPUS CANUS
northern populations of S. opacus, have been isolated genetically for extensive periods. Another case is that of
populations of S. spillmanni Stresemann from the Western Andes of Colombia, which have fairly similar
songs but distinctly different calls and were noted by Krabbe and coworkers (2006) to be genetically similar to
birds from Ecuador, which actually differ by 2–3% sequence divergence from individuals from the Eastern
Andes of Colombia and from Ecuador (CDC unpubl. data). However, because all these pairs of taxa are
allopatric, ascribing their genetic divergence to reproductive isolation resulting from variation in calls is not
possible. Although in at least one genus of suboscine birds (Hypocnemis antbirds) calls appear to play a
crucial role in reproductive isolation (Isler et al. 2007), the importance of calls as mating barriers between
Scytalopus species remains to be shown. Thus, unless future studies examining the function of songs and calls
suggest otherwise, we conservatively apply species rank only to Scytalopus populations with distinctive
songs, the approach employed by Krabbe & Schulenberg (1997, 2003). Therefore, despite the long genetic
isolation of southern from northern populations of S. opacus and their distinct calls, the two have such similar
songs and are morphologically so similar that we prefer to rank them as conspecific based on the available
Considering our phylogenetic analyses, treating the northern and southern populations of S. opacus sensu
stricto as conspecific implies the recognition of a paraphyletic biological species, within which S. canus
appears to be nested. Because the biological species concept emphasizes ability to interbreed and not common
ancestry, and because paraphyletic biological species often result from differentiation of populations in the
periphery of ranges (a likely biogeographic scenario for the differentiation of S. canus), the recognition of
such entities is not at odds with the philosophy of this species concept (Donoghue 1985; Funk & Omland
2003). In addition, we note that paraphyly of S. canus is not strongly supported in all of our analyses, and
recognize that phylogenetic patterns (but not degree of mtDNA differentiation) might change if one considers
additional individuals or loci. In any event, we emphasize that conducting studies on the function of calls and
songs in mate recognition in Scytalopus is an important priority in order to determine whether biological
species rank for the southern population of S. opacus might be warranted. For the time being, however, we
conservatively name this population as a new subspecies:
Scytalopus opacus androstictus ssp. nov. Krabbe & Cadena
HOLOTYPE: An adult male, deposited in the Zoological Museum, University of Copenhagen (ZMUC 80126).
Collected by N. Krabbe (collector’s number NK12-9.11.92) on 9 November 1992 along the Jimbura-Zumba
road in Cordillera de Las Lagunillas, Loja Province, Ecuador (4°43'S, 79°26'W; 3450 m). Tissue sample in
ZMUC (number 125728). A recording of the song of this individual was deposited on XenoCanto (number
Diagnosis. Similar in measurements and plumage coloration to S. opacus opacus from northern Ecuador
and central Colombia, but differs vocally by its song-like call, and genetically. Additionally, in seven of nine
adult males (and three more observed in the field) most or all of the greater primary coverts in both wings are
albinistic, a trait not reported in nominate S. opacus.
Description of holotype. Capitalized color names and numbers from Smithe (1975): Above Blackish
Neutral Gray (82), wings and tail very faintly washed with Brussels Brown (121B), rump and tips of tertials
with faint, 1 mm wide bars of this color and blackish. Greater primary coverts white. Underparts including
underwing uniform, between Dark Neutral Gray (83) and Medium Neutral Gray (84), tips of feathers of lower
flanks and under tail coverts with 1 mm wide blackish and Tawny Olive (223D) bars. Weight 16.8 g. Wing
chord 53 mm, wing flat 56 mm, tail length 39 mm, tarsus length 22.3 mm. Bill from fore edge of operculum
5.4 mm, from skull 10.3 mm, from mandible fork 6.4 mm. Tail composed of ten rectrices. Irides dark brown,
maxilla blackish, mandible blackish brown, feet (tarsi and toes) fuscous. Testes medium enlarged (5 x 2.5
mm), seminal vesicle enlarged. Stomach contents: remains of small insects.
Variation among males. The nine adult males examined are remarkably uniform. Seven individuals vary
slightly in the coloration of the primary coverts, which range from being mostly to entirely white; three
KRABBE & CADENA64 · Zootaxa 2354 © 2010 Magnolia Press
additional individuals observed in the field possessed this spot, but two adult male specimens from
southernmost Zamora-Chinchipe (ZMUC 80128, LSUMZ 83373) show no white and thus do not differ from
nominate S. opacus morphologically.
FEMALE: Two adult and two subadult females appear indistinguishable from corresponding plumages of
nominate S. opacus.
Description of juvenile. A recently fledged male (ZMUC 80131) with tail still in pin and crown feathers
tipped with long down is Prout's Brown (121A) above, lightly barred blackish and with narrow buff tips on the
secondaries. The throat and breast are Pale Pinkish Buff (121D) lightly mottled with blackish, belly dirty
white and unbarred, sides, flanks and vent dirty white grading to Mikado Brown (121C) on sides of lower
breast, and barred blackish. The upperparts are slightly darker brown and the belly paler than in four
specimens of nominate S. opacus, but this difference might not be constant, considering the broad range of
plumage variation seen in juveniles of nominate S. o. opacus, even within a locality (Krabbe & Schulenberg
Distribution and habitat. S. o. androstictus inhabits the East Andes south of the Río Zamora in southern
Ecuador, and the Andes of northernmost Peru north of the Río Marañón, on the eastern flank of the
Huancabamba valley. It occurs in treeline scrub at elevations ranging from 3000 to 3650 m, locally down to
2600 m along exposed ridges. On the eastern slopes it is replaced below and in taller and more bamboo-
dominated vegetation by S. parkeri Krabbe & Schulenberg and also meets the (as yet unnamed) eastern form
of S. latrans Hellmayr, which locally ascends the slopes through disturbed humid forest. On the western
slopes it is replaced below in drier, more open scrub by S. latrans subcinereus Zimmer.
Vocalizations. The distinctive calls of S. o. androstictus are described above and shown in Fig. 3. In spite
of this taxon’s long genetic isolation and markedly different call, its song does not differ appreciably from that
of S. o. opacus (Table 3; Figs. 1–2), except by lacking the 1–3 churring introductory notes found in 16 of 21
recordings of opacus (Fig. 5) and by sometimes being delivered without pause for two minutes or more, much
longer than any recorded song of S. o. opacus. Usually, however, the song is 4–12 s long, as in nominate S.
opacus. When S. o. androstictus in rare cases gives an introductory note, i.e. pauses after the first note, the
note is of the same pitch and quality as the following notes. In vocalizations of both forms of S. opacus as well
as in S. canus, the second, occasionally the first harmonic is loudest, and several more harmonics are often
Etymology. The name refers to the white spot on the primary coverts found in most males, setting it apart
from a nearby (Cordillera Colán, Amazonas, Peru) population, vocally unknown but presently referred to S.
parvirostris Zimmer, in which a similar spot is found but in females only (Krabbe & Schulenberg 1997).
Conservation. Our revised classification of the Scytalopus canus complex has implications for
conservation because a taxon that was formerly considered to represent a single species actually consists of
three differentiated lineages. However, although they range over fairly small areas, there seems to be no
immediate threat to the survival of any of them. The entire Ecuadorian part of the range of S. opacus
androstictus lies within two large and continuous protected areas: Podocarpus National Park and Bosque
Protector Colambo-Yacuri. Much of the range of S. o. opacus also lies within large national parks in Ecuador
(e.g. Sangay, Llanganates, Cayambe-Coca) and Colombia (e.g. Los Nevados, Puracé). In the Western Andes
of Colombia, S. canus is so far known to occur on Páramo de Paramillo, which is declared a national park, and
Páramo de Frontino, of which only a small part is protected. It would be of interest to determine if it occurs on
other paramos in the Western Andes, several of which are protected. The first one south of Frontino is entirely
encompassed by a nature reserve, Farallones de Citará. It is separated from Frontino by a pass of 2200 m,
higher than the lowest pass between Frontino and Paramillo (2150 m), so S. canus presumably also occurs
here. Further south, the paramos are separated by passes lower than 1800 m. It would be of conservational
interest to determine to what degree the range of S. canus coincides with that of Coeligena orina Wetmore,
another bird confined to humid treeline scrub in the Western Andes and so far only recorded from Farallones
de Citará and Frontino (Krabbe et al. 2005).
Zootaxa 2354 © 2010 Magnolia Press · 65
A TAXONOMIC REVISION OF SCYTALOPUS CANUS
FIGURE 5. Beginning of song of northern S. c. opacus from Azuay, Ecuador, including two introductory notes. In
southern S. c. opacus there is usually no introductory note and if present, it is of the same pitch and quality as the
Biogeography. It remains to be determined if S. canus evolved in situ or arrived to the northern end of
Western Andes from the south, or through dispersal from the Central Andes across the Cauca Valley (see
discussion in Krabbe et al. 2006). If it came from the south, it seems likely that all the highest paramos in this
cordillera hold populations.
The restricted range of S. o. androstictus in the Andes in northernmost Peru and southernmost Ecuador
closely matches that of Metallura odomae Graves, which is also replaced by a close relative (M. williami
atrigularis Salvin) north of the Río Zamora. Three other forms, Grallaria ridgelyi Krabbe et al., Thryothorus
euophrys atriceps (Chapman), and a yet undescribed subspecies of Synallaxis unirufa Lafresnaye (J. V.
Remsen & NK, unpubl.data) appear to have similar ranges, except that they also occur in forest down to ca.
2200 m and might range into the southern Cordillera del Cóndor (known to be the case for Grallaria ridgelyi;
T. Mark, unpubl. data). The role of the ca. 10 km wide, dry Río Zamora valley as a dispersal barrier for
treeline birds was first recognized by Robbins et al. (1994) and further discussed by Krabbe (2008). Three
species, Grallaricula lineifrons (Chapman), Anairetes agilis (Sclater) and Urothraupis stolzmanni
Taczanowski & Berlepsch, all confined to humid treeline scrub, have never been reported south of the Río
Zamora despite presence of seemingly suitable habitat to the south. In contrast, some species occupying this
habitat range across both sides of the Río Zamora (e.g. Buthraupis wetmorei R.T. Moore and eximia
Boissonneau, evidence that different species respond differently to landscape features and historical events in
We thank Steve Rogers of the Carnegie Museum kindly for providing measurements and notes on plumage of
the specimen from La Leonera; J. Van Remsen for providing notes on plumage of four Louisiana State
University Museum of Natural Science (LSUMZ) specimens of southern opacus. Paul Salaman for providing
BioMap data on museum location of specimens; the Julie von Müllens Foundation, Denmark for funding the
Frontino expedition; Paul Salaman and Pablo Flórez for coordinating the logistics; the staff of Las Orquídeas
National Park for guidance and logistical support; Ministerio del Ambiente, Ecuador and Corpourabá,
Colombia for issuing collecting permits; and Pablo Flórez, Gustavo Suárez, José Castaño, Juan David
KRABBE & CADENA66 · Zootaxa 2354 © 2010 Magnolia Press
Arango, and Arley Duque for good companionship during field work at Frontino; Oscar Laverde and Andrés
M. Cuervo for sending recordings from the Central Andes of Colombia. Laboratory work at Universidad de
los Andes was supported by the Facultad de Ciencias. We thank Robb Brumfield and Donna Dittmann
(LSUMZ) and John Klicka (Marjorie Barrick Museum, UNLV; MBM) for providing tissue samples for this
study; and three anonymous reviewers for useful comments on the manuscript.
Arctander, P. & Fjeldså, J. (1994) Andean tapaculos of the genus Scytalopus (Aves, Rhinocryptidae): A study of
speciation using DNA sequence data. In: Loeschcke, V., Tomiuk, J. & Jain, S. K. (Eds) Conservation Genetics.
Birkhäuser Verlag, Basel, Switzerland, pp. 205–225.
Bornschein, M.R., Maurício, G.N., Belmonte-Lopes, R., Mata, H. & Bonatto, S.L. (2007) Diamantina Tapaculo, a new
Scytalopus endemic to the Chapada Diamantina, northeastern Brazil (Passeriformes: Rhinocryptidae). Revista
Brasileira de Ornitologia, 15, 151–174.
Chapman, F.M. (1915) The more northern species of the genus Scytalopus Gould. Auk, 32, 406–423.
Cracraft, J. (1983) Species concepts and speciation analysis. Current Ornithology, 1, 159–187.
Cuervo, A.M., Cadena, C.D., Krabbe, N. & Renjifo, L.M. (2005) Scytalopus stilesi, a new species of tapaculo
(Rhinocryptidae) from the Cordillera Central of Colombia. Auk, 122, 445–463.
Donoghue, M.J. (1985) A critique of the biological species concept and recommendations for a phylogenetic alternative.
Bryologist, 88, 172–181.
Drummond, A.J. & Rambaut, A. (2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary
Biology, 7, 214.
Funk, D.J. & Omland, K.E. (2003) Species-level paraphyly and polyphyly: frequency, causes, and consequences, with
insights from animal mitochondrial DNA. Annual Review of Ecology, Evolution and Systematics, 34, 397–423.
Isler, M.L., Isler, P.R. & Whitney, B.M. (2007) Species limits in antbirds (Thamnophilidae) the Warbling Antbird
(Hypocnemis cantator) complex. Auk, 124, 11–28.
Johnson, N.K., Remsen, Jr., J.V. & Cicero, C. (1999) Resolution of the debate over species concepts in ornithology: a
new comprehensive biologic species concept. Proceedings of the 22 International Ornithological Congress,
Krabbe, N. (2008) Arid valleys as dispersal barriers to high-Andean forest birds in Ecuador. Cotinga, 29, 28–30.
Krabbe, N. & Schulenberg, T.S. (1997) Species limits and natural history of Scytalopus tapaculos (Rhinocryptidae), with
descriptions of the Ecuadorian taxa, including three new species. Ornithological Monographs, 48, 46–88.
Krabbe, N., Florez, P., Suárez, G., Castaño, J., Arango, J.D., Pulgarín, P., Múnera, W.A. Stiles, F.G. & Salaman, P. (2005)
Rediscovery of the Dusky Starfrontlet Coeligena orina, with a description of the adult plumages and a reassessment
of its taxonomic status. Ornitología Colombiana, 3, 28–35.
Krabbe, N., Florez, P., Suárez, G., Castaño, J., Arango, D. & Duque, A. (2006) The birds of Páramo de Frontino, Western
Andes of Colombia. Ornitología Colombiana, 4, 37–48.
Krabbe, N.K. & Schulenberg, T.S. (2003) Family Rhinocryptidae (tapaculos). In: J. del Hoyo, J., Elliott, A. & Christie,
D. (Eds) Handbook of the Birds of the World, Vol. 8. Barcelona, Spain: Lynx Edicions, pp. 748–787.
Maurício, G.N., Mata, H., Bornschein, M.R., Cadena, C.D., Alvarenga, H. & Bonatto, S.L. (2008) Hidden generic
diversity in Neotropical birds: molecular and anatomical data support a new genus for the "Scytalopus" indigoticus
species-group (Aves: Rhinocryptidae). Molecular Phylogenetics and Evolution, 49, 125–135.
Posada, D. & Crandall, K.A. (1998) Modeltest: testing the model of DNA substitution. Bioinformatics, 14, 817–818.
Remsen, J.V., Jr., Cadena, C.D., Jaramillo, A., Nores, M., Pacheco, J.F., Robbins, M.B., Schulenberg, T.S., Stiles, F.G.,
Stotz, D.F. & Zimmer, K.J. (2009) A classification of the bird species of South America (version 24 June 2009).
American Ornithologists' Union. http://www.museum.lsu.edu/~Remsen/SACCBaseline.html.
Robbins, M.B., Krabbe, N., Ridgely, R.S. & Sornoza Molina, F. (1994) Notes on the natural history of the Crescent-faced
Antpitta. Wilson Bulletin, 106, 169–173.
Smithe, F.B. (1975) Naturalist’s Color Guide. American Museum of Natural History, New York.
Stamatakis, A. (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and
mixed models. Bioinformatics, 22, 2688–2690.
Stamatakis, A., Hoover, P. & Rougemont, J. (2008) A rapid bootstrap algorithm for the RAxML web servers. Systematic
Biology, 57, 758–771.
Swofford, D.L. (2002) PAUP* Version 4.0b10: Phylogenetic analysis using parsimony (and other methods). Sinauer
Associates, Sunderland, Massachusetts.
Wiley, E.O. (1978) The evolutionary species concept reconsidered. Systematic Zoology, 27,17–26.