Circumscription of a monophyletic family for the tapaculos (aves rhinocryptidae) psiloramphus in and melanopareia out
1. J Ornithol (2010) 151:337–345
DOI 10.1007/s10336-009-0460-9
ORIGINAL ARTICLE
Circumscription of a monophyletic family for the tapaculos
(Aves: Rhinocryptidae): Psiloramphus in and Melanopareia out
Per G. P. Ericson • Storrs L. Olson •
Martin Irestedt • Herculano Alvarenga •
˚
Jon Fjeldsa
Received: 22 February 2009 / Revised: 3 August 2009 / Accepted: 21 September 2009 / Published online: 14 October 2009
Ó Dt. Ornithologen-Gesellschaft e.V. 2009
Abstract The tapaculos (Rhinocryptidae) are tracheo- within the Rhinocryptidae, Melanopareia falls far outside
phone, suboscine birds restricted to South and Central that clade. A new family is erected for Melanopareia.
America. Most tapaculos share a number of internal and
external characteristics that have been used to define the Keywords Melanopareia Á Psiloramphus Á
family taxonomically. The genera Melanopareia and Rhinocryptidae Á Tapaculos Á Molecular systematics Á
Psiloramphus do not fully fit this pattern and have caused Taxonomy Á South America
considerable dispute among taxonomists since they were
first described. In this paper we delimit the systematic
boundaries of the tapaculos and assess their generic rela- Introduction
tionships by analysis of molecular sequence data. The
results show that whereas Psiloramphus is nested well The impudently named tapaculos (Rhinocryptidae) are a
small group of tracheophone, suboscine passerines whose
greatest generic diversity is in southern South America.
Most are large-footed, strong-legged ground birds remi-
Communicated by M. Wink. niscent of some of the ground-dwelling antthrushes (For-
micariidae), with which they were often associated. The
P. G. P. Ericson (&)
family is generally well defined by the presence of oper-
Department of Vertebrate Zoology,
Swedish Museum of Natural History, culate nostrils, a tracheophone syrinx, a somewhat curved
P.O. Box 50007, 10405 Stockholm, Sweden humerus, and a four-notched sternum (Ames 1971;
e-mail: per.ericson@nrm.se ´
Feduccia and Olson 1982; Maurıcio et al. 2008). Thus,
Krabbe and Schulenberg (2003), p. 748, considered that the
S. L. Olson
Division of Birds, Department of Vertebrate Zoology, ‘‘tapaculos constitute a well-knit group the members of
Smithsonian Institution, National Museum of Natural History, which are united by several derived characters. Only the
P.O. Box 37012, Washington, DC 20013-7012, USA genera Melanopareia and Psiloramphus differ to such a
degree that their systematic position as tapaculos could be
M. Irestedt
Molecular Systematics Laboratorium, disputed.’’ Furthermore, the phylogenetic relationships
Swedish Museum of Natural History, within the family Rhinocryptidae are poorly known, par-
P.O. Box 50007, 10405 Stockholm, Sweden ticularly regarding the placement of Liosceles and
Acropternis, and the large austral species of Pteroptochos
H. Alvarenga
´ ´
Museu de Historia Natural de Taubate, and Scelorchilus.
´
Rua Juvenal Dias de Carvalho, 111, Taubate, The early taxonomic history of the group was ably
SP 12070-640, Brazil summarized by Sclater (1874). d’Orbigny (1837) first
˚
erected a family ‘‘Rhinomyidaeae’’ (sic., p. 192) for Pter-
J. Fjeldsa
Zoological Museum, University of Copenhagen, optochos and his Rhinomya (= Rhinocrypta) using the
Universitetsparken 15, 2100 Copenhagen, Denmark operculate nostril to separate them from the Formicariidae.
123
2. 338 J Ornithol (2010) 151:337–345
However, there was no coherent understanding of the su-
boscine groups until their distinction from the oscines was
¨
established by the pioneering work of Muller (1847) on the
¨
syrinx. Muller showed that Scytalopus was a tracheophone
suboscine and not a wren (Troglodytidae) and also that
Scytalopus and Pteroptochus differed from other known
passerines in having a four-notched sternum.
Sclater’s (1874) ‘‘Pteroptochidae’’ comprised Scytal-
opus (including the type species of what later became
Myornis Chapman 1915), Merulaxis, Liosceles, Pteropto-
chos (including the type species of what later became
Teledromas Wetmore and Peters 1922, and Scelorchilus
Oberholser 1923), Rhinocrypta, Hylactes (now included in
Pteroptochos), Acropternis, and Triptorhinus (= Eugralla).
Except for the problem genera Psiloramphus and Mela-
nopareia, this composition of the group was essentially
maintained until Peters (1951).
The three or four species of Melanopareia differ from
other tapaculos by their rather slender build and boldly and
attractively patterned plumage, and by sharing a semi-
concealed white dorsal patch with various true antbirds
(Thamnophilidae). They were originally described in the
genus Synallaxis (Furnariidae), in which Sclater (1890)
later submerged the genus. Salvin (1876) described a new
species from Ecuador as Formicivora speciosa, duly rec-
ognized in that combination by Sclater (1890), p. 251, and Fig. 1 The bamboo-wren Psilorhamphus guttatus bears little external
resemblance to typical members of the tapaculo family. With its grey
others until Hellmayr (1906), p. 334, showed that this was a iris, facial expression, bill shape, and wing-coverts with white dots
synonym of Synallaxis elegans Lesson, in which genus Psilorhamphus instead resembles some antbirds (Dysithamnus, Myr-
Hellmayr continued to place it while regarding Salvin’s motherula) with which early ornithologists consequently placed it.
allocation of it to Formicivora with incredulity. Ridgway Unlike other tapaculos Psilorhamphus spends most of the time above
the ground. Photo: Edson Endrigo
(1909) seems to have overlooked this when he created a
new genus Rhoporchilis for Formicivora speciosa. It was
Hellmayr (1921) who eventually established the modern comment that these genera ‘‘might perhaps be more natu-
concept of the genus in showing that Synallaxis elegans, S. rally placed as a distinct subfamily of Pteroptochidae
torquata, and S. maximiliani were congeneric and would [= Rhinocryptidae]’’ despite the fact that ‘‘there is little
all fall under Reichenbach’s earlier generic name Mela- external difference between the appearance of these birds
nopareia and ‘‘find their natural place in the Formicarii- and the true Wrens [Troglodytidae]’’. We are not aware,
dae’’, where they stayed for only a few years (Cory and however, of any instance in which Psilorhamphus was
Hellmayr 1924). Next came the observation of W.D.W. placed in either the Troglodytidae or in a family with only
Miller that the sternum of Melanopareia was four-not- sylviid-like genera, as might be inferred from Krabbe and
ched—information that was conveyed to and presented by Schulenberg (2003). Psilorhamphus continued to be asso-
Wetmore (1926), p. 292. On this basis, Peters (1951) ciated with Ramphocaenus in the Formicariidae—e.g.
included Melanopareia in the Rhinocryptidae, where it has Sclater (1890) and Cory and Hellmayr (1924), p. 205—
resided since. although in the latter reference it was noted that W.D.W.
The other problem species, the Bamboo-wren Psilor- Miller would show Psilorhamphus and Ramphocaenus to
hamphus guttatus, is a small bamboo specialist with a ‘‘constitute a separate family’’ in ‘‘a paper shortly to be
rather long, slender bill, a long tail, and relatively weak published.’’ Peters (1951), p. 213, later explained that
feet, so it bears little resemblance to large-footed terrestrial Miller’s death prevented publication of his results but that
´ ´ ´
tapaculos (Fig. 1). From the beginning (Menetries 1835) it Wetmore (1943), p. 306, had shown Ramphocaenus to
was placed with the antbirds in the Myiotherinae (= For- have an oscine syrinx, and had told Peters that Microbates
micariidae). Sclater (1858), p. 243, associated Psilorham- likewise was oscine and that he believed, on the basis of
phus with Ramphocaenus (which is now in the oscine external morphology, that Psilorhamphus was also proba-
family Polioptilidae) in the Formicariidae with the bly oscine. Therefore, Peters postponed his treatment of
123
3. J Ornithol (2010) 151:337–345 339
those genera for a future volume treating Sylviidae. Sick 1994; Krabbe and Schulenberg 2003), including represen-
(1954) placed Ramphocaenus in the Sylviidae while pro- tatives of the large genus Scytalopus, including one rep-
visionally referring Psilorhamphus to the Formicariidae. resentative (indigoticus) of the ‘‘blue’’ species, which were
´
Then, Plotnick (1958) revealed that Psilorhamphus had a recently placed in a separate genus Eleoscytalopus
four-notched sternum, a tracheophone syrinx, and had other ´
(Maurıcio et al. 2008). Three of the authors have significant
characters, including an operculate nostril, indicating that it field experience of the biology and vocalizations of
should be placed in the Rhinocryptidae. Thus, in Peters’ tapaculos, and this was supplemented with comments and
Checklist Psilorhamphus appears as an addendum to the analyses of sound archives by Niels Krabbe (personal
Rhinocryptidae that appeared in the volume on Sylviidae communication). Representatives of the main lineages
(Paynter 1964). within the tracheophone radiation serve as outgroups
Heimerdinger and Ames (1967) confirmed that the rhi- (Ridgely and Tudor 1994; Irestedt et al. 2002; Krabbe and
nocryptids they examined all had a four-notched sternum Schulenberg 2003; Chesser 2004).
but also showed that this condition obtained in at least two Three nuclear gene regions, myoglobin intron 2, orni-
genera of grallarine Formicariidae, which was confirmed thine decarboxylase (ODC) introns 6 to 7, and glycer-
by Feduccia and Olson (1982). aldehyde-3-phosphodehydrogenase (G3PDH) intron 11,
Ames (1971) made a thorough study of the anatomy of were sequenced and used to estimate phylogenetic rela-
the syrinx in passerine birds and examined Melanopareia tionships. For each gene and taxon, multiple sequence
first-hand but had to rely on the description of Plotnick´ fragments were obtained by sequencing with different
(1958) for Psilorhamphus. Although he noted that the primers. These sequences were assembled to complete
cartilaginous elements of Melanopareia differed from sequences with SEQMAN II (DNASTAR). Positions where
those of all other tapaculos examined, he found no grounds the nucleotide could not be determined with certainty were
for excluding either Melanopareia or Psilorhamphus from coded with the appropriate IUPAC code. GenBank acces-
the Rhinocryptidae. sion numbers are given in Table 1. See Irestedt et al.
Feduccia and Olson (1982) made the much unexpected ˚
(2002); Allen and Omland (2003); and Fjeldsa et al. (2003)
discovery that the stapes in Melanopareia was of the for extractions, amplifications, and sequencing procedures
primitive oscine type with a flattened footplate, rather than for fresh tissue/blood samples. Corresponding laboratory
having an expanded, bulbous, fenestrate footplate as in all procedures for study skins are detailed in Irestedt et al.
other suboscine birds. They went on to show other mor- (2006).
phological similarities (which is all they ever claimed they
were) between some of the Rhinocryptidae and the oscine Phylogenetic inference and model selection
Menurae (Menura and Atrichornis) of Australia. Although
this observation was dismissed on the grounds that the Because of the rather low number of insertions in the
characters involved are either primitive or convergent introns, the combined sequences could easily be aligned by
(Krabbe and Schulenberg 2003), the fact remains that the eye. All gaps have been treated as missing data in the
oscines and suboscines had to share a common ancestor analyses. Bayesian inference (Holder and Lewis 2003;
and that the ancestor was very likely to have looked like Huelsenbeck et al. 2001) was used to estimate the phylo-
Atrichornis, Melanopareia, or one of the Rhinocryptidae, genetic relationships. The models for nucleotide substitu-
for example Scelorchilus. tions used in the analyses were selected for each gene
This paper aims to delimit the boundaries of the tapac- individually by applying the Akaike information criterion
ulos and assess generic relationships within the group, (AIC; Akaike 1973) and the software MrModeltest 2.2
supplementing the very detailed ongoing studies of rela- (Nylander 2005) in conjunction with PAUP* (Swofford
tionships and speciation of small tapaculos (notably Scy- 1998).
´
talopus; Maurıcio et al. 2008, Cadena et al. unpublished) Posterior probabilities of trees and parameters in the
by providing a broader phylogenetic framework for the substitution models were approximated with MCMC and
family. Metropolis coupling using the software MrBayes 3.1.1
(Ronquist and Huelsenbeck 2003). Analyses were per-
formed for both the individual gene partitions and the
Materials and methods combined data set. In the analysis of the combined data set,
the models selected for the individual gene partition were
Taxon sampling, amplification and sequencing used. The chains for the individual gene partitions were run
for five million generations while the chains for the com-
This study includes representatives of all genera tradi- bined data set were run for ten million generations. Trees
tionally recognized in Rhinocryptidae (Ridgely and Tudor were sampled every 100th generation, and the trees
123
4. 340 J Ornithol (2010) 151:337–345
Table 1 Specimen data and GenBank accession numbers for samples used in the study
Species Family: Subfamily Voucher/Sample No. Myoglobin G3PDH ODC
Acropternis orthonyx Rhinocryptidae ZMUC 125695* GQ925894 GQ925879 GQ925860
Merulaxis ater Rhinocryptidae ZMUC 128820 GQ925895 GQ925880 GQ925861
Pteroptochos tarnii Rhinocryptidae AMNH RTC467* AY065774a AY590096e GQ925862
Rhinocrypta lanceolata Rhinocryptidae NRM 966793 AY065775a DQ438953f DQ435499f
Teledromas fuscus Rhinocryptidae USNM BKS3703* GQ925896 GQ925881 GQ925863
Psilorhamphus guttatus Rhinocryptidae MHNT-4812 GQ925897 GQ925882 GQ925864
Myornis senilis Rhinocryptidae ZMUC 134967* GQ925898 GQ925883 GQ925865
Scytalopus parvirostris Rhinocryptidae ZMUC 128441 GQ925899 GQ925884 GQ925866
Scytalopus speluncae Rhinocryptidae ZMUC 128818 GQ925900 GQ925885 GQ925867
Scytalopus spillmannii Rhinocryptidae ZMUC 125091* AY065773a AY590097e GQ925868
Scytalopus zimmeri Rhinocryptidae ZMUC 126278* GQ925901 GQ925886 GQ925869
Scytalopus superciliaris Rhinocryptidae USNM BKS 3592* GQ925902 GQ925887 GQ925870
Eugralla paradoxa Rhinocryptidae NRM 570026*, TP GQ925903 GQ925888 GQ925871
Scelorchilus rubecula Rhinocryptidae NRM 570029*, TP GQ925904 GQ925889 GQ925872
Liosceles thoracicus Rhinocryptidae NRM 570027*, TP GQ925905 GQ925890 GQ925873
Eleoscytalopus indigoticus Rhinocryptidae NRM 570028*, TP GQ925906 GQ925891 GQ925874
Melanopareia maximiliani Rhinocryptidae ZMUC 125045* AY065785a GQ925892 GQ925875
Furnarius cristatus Furnariidae: Furnariinae NRM 966772* AY064255b AY590066e DQ435482f
a e
Philydor atricapillus Furnariidae: Furnariinae NRM 937334* AY065758 AY590076 EF212110h
a e
Synallaxis ruficapilla Furnariidae: Furnariinae NRM 956643* AY065763 AY590068 EF212119h
a g
Lepidocolaptes angoustirostris Furnariidae: Dendrocolaptinae NRM 937184* AY065767 AY336576 DQ435486f
c e
Dendrocincla tyrannina Furnariidae: Dendrocolaptinae ZMUC 125661* AY442959 AY590087 EF212098h
a e
Chamaeza meruloides Formicariidae ZMUC 126604* AY065776 AY590095 GQ140036i
a
Formicarius nigricapillus Formicariidae ZMUC 125987* AY065777 GQ925893 GQ925876
Grallaria squamigera Grallariidae ZMUC 124629* AY065778a AY677078c GQ140073i
Dysithamnus mentalis Thamnophilidae NRM 956629* AY676995c AY677042c GQ925877
Terenura humeralis Thamnophilidae FMNH 389941 AY677004c AY677051c GQ925878
Thamnophilus caerulescens Thamnophilidae NRM 967007* AY065783a AY336587g DQ435504f
a
Conopophaga aurita Conopophagidae ZMUC 125796* AY065784 DQ435478f
g
Conopophaga lineata Conopophagidae NRM 956653* AY336577
Pipra fasciicauda Pipridae NRM 947271* AY065787a AY336583g DQ435495f
a f
Corythopsis delalandi Tyrannidae NRM 937282* AY065788 DQ435463 DQ435479f
d f
Elaenia flavogaster Tyrannidae NRM 966970* AY228295 DQ435464 DQ435480f
Samples vouchered with a study skin are indicated by an asterisk. ‘‘TP’’ indicates that the sample is obtained from toe-pads of an old study skin. Acronyms
are AMNH, American Museum of Natural History; FMNH, Field Museum of Natural History, Chicago, USA; MHNT, Museu de Historia Natural de ´
´
Taubate, Sao Paulo, Brazil; NRM, Swedish Museum of Natural History; USNM, National Museum of Natural History, Washington, USA; ZMUC
Zoological Museum, University of Copenhagen, Denmark
a
Irestedt et al. (2002)
b
Ericson et al. (2002)
c
Irestedt et al. (2004)
d
Ericson and Johansson (2003)
e
˚
Fjeldsa et al. (2005)
f
Ericson et al. (2006)
g
˚
Fjeldsa et al. (2003)
h
˚
Fjeldsa et al. (2007)
i
Irestedt et al. (in press)
sampled during the burn-in phase (i.e., before the chain had Sequence lengths and alignments
reached its apparent target distribution) were then dis-
carded after checking for convergence; final inference was We were able to sequence all three gene regions almost
made from the concatenated outputs. completely for all included taxa (a few sequences miss
123
5. J Ornithol (2010) 151:337–345 341
some base pairs in the 30 or 50 ends in the myoglobin or the contains Formicariidae sensu stricto and Furnariidae, in
ODC regions, and in the ODC region all sequences agreement with previous molecular studies of tracheo-
obtained from study skins lack a short fragment of 22 bp in phone suboscines (Irestedt et al. 2002; Chesser 2004).
exon 7). Taking into account the missing base pairs, the Within the radiation of tapaculos there is also good
sequences obtained varied in length between 667 and support for two major clades. Clade 1 includes Teledromas,
701 bp for the myoglobin intron 2, and between 313 and Acropternis, Rhinocrypta, Liosceles, and Psilorhamphus
363 bp for the G3PDH intron 11, except for the two ant- and Clade 2 includes Scytalopus, Eugralla, Myornis,
thrushes Chamaeza and Formicarius which contain two Merulaxis, and Eleoscytalopus. The only conflict within
large deletions in the G3PDH intron 11 which makes these the rhinocryptid radiation supported by posterior proba-
sequences 251–252 bp long. In the ODC region all Rhi- bilities above 0.95 involves determining to which of the
nocryptidae and Furnariidae taxa have a large deletion in previous two clades Pteroptochos and Scelorchilus belong.
intron 7 and the sequences from these taxa range between In the ODC tree they group with Clade 1 (0.97), whereas
403 and 500 bp, while the sequences for all other taxa the myoglobin tree indicates that these two taxa are sister
range between 586 and 624 bp. to Clade 2 (1.00). In the G3PDH tree this relationship is
Most indels observed in the introns were autapomorphic unresolved. On the basis of the overall congruence of the
and mainly found in certain variable regions. Some indels individual gene trees we believe that the tree obtained from
vary in length between taxa, which makes it difficult to the combined analysis (Fig. 2) represents the best estimate
know if these indels are homologous or represent inde- of the phylogenetic relationship of the tapaculos and in this
pendent evolutionary events. Several apparently synapo- Pteroptochos and Scelorchilus fall out as sister to Clade 2.
morphic indels were also observed when mapping the data This tree is fully congruent with the results of studies using
on to the tree topology obtained from the Bayesian anal- other genetic markers, but with focus on detailed rela-
yses of the combined data set. A few indels were also found ´
tionships within Clade 2 (Maurıcio et al. 2008 and Cadena
to be incongruent with the phylogenetic tree obtained from et al. unpublished).
analysis of the combined data set. These were generally
found in the most variable regions and some of the single
base pair insertions actually consist of different bases. Discussion
Models for nucleotide substitutions Melanopareia resembles members of the Rhinocryptidae in
having the lacrimal bones partly fused with the ectethmoid
The prior selection of nucleotide substitution models sug- (but the lacrimals are lacking in Conopophagidae, Tham-
gested that the GTR ? C model had the best fit for all three nophilidae, Grallaridae, and Formicariidae) and in having a
gene regions, but as the nucleotide state frequencies and four-notched sternum. The significance of these characters
gamma distribution differed between the partitions we is uncertain because of the weak cranial ossification in
applied a partitioned analysis of the combined data set. these groups and substantial flexibility (including varying
After discarding the burn-in phase the inferences for degrees of developmental asymmetry) in the degree of
myoglobin and G3PDH were based on a total of 45,000 ossification of the membranes serving as attachment of
samples from the posterior, the inferences for ODC were pectoral muscles (Heimerdinger and Ames 1967). The
based on a total of 40,000 samples, and the inferences for molecular data are not suggestive of a close relationship of
the combined data set were based on a total of 95,000 Melanopareia to the Rhinocryptidae. With four closely
samples. The posterior distribution of topologies is pre- related extant species, Melanopareia is a long, unbroken
sented as a majority-rule consensus tree from the combined phylogenetic branch, and it may be difficult to tell with
analysis in Fig. 2. confidence whether this clade is nested within the Conop-
ophagidae–Thamnophilidae complex or is a relictual, basal
tracheophone type of bird.
Results Irestedt et al. (2002) associated Teledromas with Mel-
anopareia primarily because their vocalizations are con-
The trees obtained from the Bayesian analyses of the fusingly similar, and Teledromas was considered to
individual gene partitions (Fig. 3) are overall topologically resemble a robust and pale version of Melanopareia. Both
congruent. Melanopareia clusters with Thamnophilidae genera are reported to share a peculiarity of the pterylog-
and Conopophagidae and we can therefore reject it as a raphy of the flank region, and details of the nasal opercu-
member of the family Rhinocryptidae with high confi- lum and tarsal scutellation, and an X-ray photo suggested
dence. Apart from this, all traditional tapaculo genera form almost straight humeri (approaching those of Melano-
a monophyletic clade within a broader group which also pareia; Irestedt et al. 2002). However, our DNA data and
123
6. 342 J Ornithol (2010) 151:337–345
60 Acropternis orthonyx
94 Rhinocrypta lanceolata
100 Teledromas fuscus
100 Liosceles thoracicus
Psilorhamphus guttatus
56 Scytalopus parvirostris
Scytalopus spillmannii
100 100 100 Scytalopus superciliaris
98 Scytalopus zimmeri
Scytalopus speluncae
100
Eugralla paradoxa
100 Myornis senilis
100 Merulaxis ater
100 Eleoscytalopus indigoticus
100 100 Pteroptochos tarnii
Scelorchilus rubecula
100 Chamaeza meruloides
Formicarius nigricapillus
100 Dendrocincla tyrannina
100
Lepidocolaptes angustirostri
100
100 Furnarius cristatus
100 Synallaxis ruficapilla
100 Philydor atricapillus
Grallaria squamigera
55 Conopophaga aurita
Melanopareia maximiliani
91 100 Dysithamnus mentalis
100 Thamnophilus caerulescens
Terenura humeralis
99 Corythopsis delalandi
Elaenia flavogaster
Pipra fasciicauda
Fig. 2 Majority rule consensus tree obtained from Bayesian analyses of the combined data set (myoglobin intron 2, ODC introns 6 and 7, and
G3PDH intron 11). Posterior probability values are indicated at the node
further examination of skeletal characters suggest rejection the pace and quality of the song notes of Psilorhamphus are
of a closer relationship between them and places Tele- most like the songs of larger species of Pteroptochos,
dromas centrally in Clade 1 of the Rhinocryptidae. which are even lower pitched (0.5–0.6 kHz). Interestingly,
Psilorhamphus was placed with Polioptiline oscines Psilorhamphus shares with Liosceles barred posterior un-
based on the acutiplantar tarsus, but a similar tarsal sca- derparts and distinctive whitish subterminal spots with a
lation is also found in some antbirds, and osteology and black outline on the middle and greater wing-coverts. Apart
syringeal morphology suggested placement with Rhino- from this, it is difficult to see any external features sup-
cryptidae. We were able to confirm from examination of porting the suggested relationships within Clade 1.
skeletal specimens that Psilorhamphus—and Teledro- The possible association of the large Chilean tapac-
mas—have the expanded footplate of the stapes typical of ulos (Pteroptochos, Scelorchilus) with Clade 2 receives
other suboscines (except Melanopareia). some morphological support, as Pteroptochos has 14
The distinctive appearance of Psilorhamphus may result rectrices, something that is also found in some species
from its divergent habits (albeit shared with Myornis, or individuals of Scytalopus, although other representa-
N. Krabbe personal communication), because it generally tives of this genus have a reduced number of rectrices or
feeds by climbing in tangles of vine-like bamboo, occa- asymmetrical tails (Krabbe and Schulenberg 2003).
sionally up to 7 m, and it rarely feeds on the ground. This Other suboscine birds typically have 12 rectrices,
species is also known for its unbelievably loud and low- although there are many cases of reduction. Molecular
pitched vocalizations (for its small size, 13.5 cm): a fast relationships within Pteroptochos have been analyzed by
series of hollow whistles at 0.9–1 kHz. Its sister taxon in Chesser (1999).
Fig. 2, Liosceles, also gives hollow whistles, but they are Within Clade 2, the Myornis–Eugralla–Scytalopus
higher pitched (1.3 kHz) and are given at a slower pace; group is particularly well defined morphologically by small
123
8. 344 J Ornithol (2010) 151:337–345
size and sooty-grey to blackish plumage and atrophied Acknowledgments Our thanks go especially to those researchers
clavicles, which do not form a fused furcula (Maurıcio ´ who, for years, and often with very small economic means, have
collected the specimens and tissue samples without which studies like
et al. 2008). This may reflect reduced flying ability of these this could not be made. N. Krabbe is thanked for comments on the
specialized birds, which tunnel through the densest parts of manuscript. E. Endrigo kindly provided the photograph of Psilor-
the forest understorey. Myornis bears some resemblance to hamphus. The following institutions provided samples that have been
Merulaxis, in shape and juvenile plumage (Krabbe and included in the study: Zoological Museum of Copenhagen University,
Swedish Museum of Natural History, National Museum of Natural
Schulenberg 2003), but our result does not support such an History, American Museum of Natural History and Field Museum of
association. On the other hand, Myiornis and Eugralla Natural History. The Swedish Research Council provided financial
clearly fall outside the group of species of Scytalopus that support for the laboratory work (grant no. 621–2007–5280 to PE).
we studied (except with G3PDH) and we therefore support
keeping them in separate monotypic genera.
Under ‘‘Incertae Sedis’’, Irestedt et al. (2002) introduced References
a family ‘‘Melanopareiidae (new family, incl. Melano-
pareia and Teledromas)’’, which, however, was invalid Akaike H (1973) Information theory as an extension of the maximum
likelihood principle. In: Petrov BN, Csaki F (eds) Second
because it did not meet the requirements of the Interna-
international symposium on information theory. Akademini
tional Code of Zoological Nomenclature (ICZN 1999) in Kiado, Budapest, pp 267–281
lacking any description purporting to differentiate the Allen ES, Omland KE (2003) Novel intron phylogeny (ODC)
taxon from other taxa. The need still remains for separate supports plumage convergence in orioles (Icterus). Auk 120:
961–969
family status for the genus, which we re-propose here:
Ames PL (1971) The morphology of the syrinx in passerine birds.
Melanopareiidae, new family Bull Peabody Mus Nat Hist 37:1–194
Chapman FL (1915) The more northern species of the genus
Type and only included genus: Melanopareia Reichen- Scytalopus Gould. Auk 32:406–423
bach 1853. Chesser RT (1999) Molecular systematics of the rhinocryptid genus
Diagnosis: Tracheophone suboscine passeriformes dif- Pteroptochus. Condor 101:439–446
fering from other Tracheophonae except Rhinocryptidae Chesser RT (2004) Molecular systematics of New World Suboscine
birds. Mol Phyl Evol 32:11–24
(and a few other taxa: Myrmothera, Hylopezus, Pittaso- Cory CB, Hellmayr CE (1924) Catalogue of birds of the Americas and
ma, Conopophaga) in having a four-notched sternum adjacent islands. Part III. Field Mus Nat Hist Zool Ser 13:1–369
and differing from the Rhinocryptidae and all other ´ ´
d’Orbigny AD (1837) Voyage dans l’Amerique Septentrionale. vol 4.
suboscines in retaining the primitive morphology of the part 3, Ve. Levrault, Oiseaux, Strasbourg
Ericson PGP, Johansson US (2003) Phylogeny of Passerida (Aves:
stapes, with a flat, rather than inflated, footplate. Passeriformes) based on nuclear and mitochondrial sequence
data. Mol Phyl Evol 29:126–138
Ericson PGP, Christidis L, Irestedt M, Norman JA (2002) Systematic
affinities of the lyrebirds (Passeriformes: Menura), with a novel
Zusammenfassung classification of the major groups of passerine birds. Mol Phyl
Evol 25:53–62
Ericson PGP, Zuccon D, Ohlson JI, Johansson US, Alvarenga H,
¨
Umschreibung einer monophyletischen Familie fur die
Prum RO (2006) Higher level phylogeny and morphological
¨
Burzelstelzer (Aves: Rhinocryptidae): Psiloramphus evolution of tyrant flycatchers, cotingas, manakins and their
rein und Melanopareia raus allies (Aves: Tyrannida). Mol Phyl Evol 40:471–483
Feduccia A, Olson SL (1982) Morphological similarities between the
Menurae and the Rhinocryptidae, relict passerine birds of the
¨
Die Burzelstelzer (Rhinocryptidae) sind tracheophone,
Southern Hemisphere. Smithson Contr Zool 366:1–22
¨ ¨
suboscine Vogel, deren Vorkommen auf Sud- und Mittel- ˚
Fjeldsa J, Zuccon D, Irestedt M, Johansson US, Ericson PG (2003)
¨ ¨
amerika beschrankt ist. Die meisten Burzelstelzer haben Sapayoa aenigma: a New World representative of ‘Old World
¨
eine Reihe innerer und außerer Merkmale gemeinsam, die suboscines’. Proc R Soc Lond B [Suppl] 270:238–241
˚
Fjeldsa J, Irestedt M, Ericson PGP (2005) Molecular data reveal some
benutzt worden sind, um die Familie taxonomisch zu
major adaptational shifts in the early evolution of the most
definieren. Die Gattungen Melanopareia and Psiloramphus diverse avian family, the Furnariidae. J Ornithol 146:1–13
¨
passen nicht vollstandig in dieses Schema und haben zu ˚ ¨
Fjeldsa J, Irestedt M, Jonsson KA, Ohlson JI, Ericson PGP (2007)
¨ ¨
betrachtlichen Disputen unter Taxonomen gefuhrt, seit Phylogeny of the ovenbird genus Upucerthia: a case of
independent adaptations for terrestrial life. Zool Scr 36:133–141
sie erstmals beschrieben worden sind. In diesem Artikel
Heimerdinger MA, Ames PL (1967) Variation in the sternal notches
¨
stecken wir die systematischen Grenzen der Burzelstelzer of suboscine passeriform birds. Postilla 105:1–44
ab und bewerten ihre Gattungsbeziehungen mit Hilfe Hellmayr CE (1906) Critical notes on the types of little-known
einer Analyse molekularer Sequenzdaten. Die Ergebnisse species of Neotropical birds. Novitates Zoologicae 13:305–352
Hellmayr CE (1921) Review of the birds collected by Alcide
¨
zeigen, dass Melanopareia weit aus dieser Klade herausfallt,
d’Orbigny. Novitates Zoologicae 28:230–276
¨
wahrend Psiloramphus gut in die Rhinocryptidae hineinpasst. Holder M, Lewis PO (2003) Phylogeny estimation: traditional and
¨
Eine neue Familie wird fur Melanopareia eingerichtet. Bayesian approaches. Nat Gen 4:275–284
123
9. J Ornithol (2010) 151:337–345 345
Huelsenbeck JP, Ronquist F, Nielsen R, Bollback JP (2001) Bayesian Oberholser HC (1923) Note on the generic name Pteroptochos
inference of phylogeny and its impact on evolutionary biology. Kittlitz. Auk 40:326–327
Science 294:2310–2314 Paynter RA Jr (1964) Family Rhinocryptidae (part). In: Mayr E,
International Commission on Zoological Nomenclature (1999) Inter- Paynter RA Jr (eds) Check-list of Birds of the World. Museum of
national Code of Zoological Nomenclature, 4th edn. Interna- Comparative Zoology, Cambridge, Massachusetts, p 456
tional Trust for Zoological Nomenclature, British Museum Peters JL (1951) Check-list of birds of the world, vol VII. Museum of
(Natural History), University of California Press Comparative Zoology, Cambridge, Massachusetts
˚ ´
Irestedt M, Fjeldsa J, Dalen L, Ericson PGP (in press) Convergent ´ ´ ´
Plotnick R (1958) Posicion [sic] sistematica del genero ‘‘Psiloram-
evolution, habitat shifts and variable diversification rates in the phus’’ (Rhinocryptidae, Passeriformes). Physis 21:130–136
ovenbird-woodcreeper family (Furnariidae). BMC Evol Biol Reichenbach HGL (1853) Handbuch der speciellen Ornithologie, Die
˚
Irestedt M, Fjeldsa J, Johansson US, Ericson PGP (2002) Systematic ¨
Vogel, part 3. Dresden
relationships and biogeography of the tracheophone suboscines Ridgely RS, Tudor G (1994) The birds of South America, vol II.
(Aves: Passeriformes). Mol Phyl Evol 23:499–512 Oxford University Press, Oxford
˚
Irestedt M, Fjeldsa J, Ericson PGP (2004) Phylogeny and morpho- Ridgway R (1909) New genera, species, and subspecies of Formi-
logical plasticity in woodcreepers (Dendrocolaptinae). J Avian cariidae, Furnariidae, and Dendrocolaptidae. Proc Biol Soc
Biol 35:280–288 Washington 22:69–74
¨ ¨
Irestedt M, Ohlson JI, Zuccon D, Kallersjo M, Ericson PGP (2006) Ronquist F, Huelsenbeck JP (2003) MrBAYES 3: Bayesian phylo-
Nuclear DNA from old collections of avian study skins reveals genetic inference under mixed models. Bioinformatics 19:1572–
the evolutionary history of the Old World suboscines (Aves, 1574
Passeriformes). Zool Scr 35:567–580 Salvin O (1876) On some new species of birds from western Ecuador.
Krabbe NK, Schulenberg TS (2003) Family Rhinocryptidae (Tapac- Ibis 6:493–496
ulos). In: del Hoyo J, Elliott A, Christie DA (eds) Handbook of Sclater PL (1858) Synopsis of the American ant-birds (Formicarii-
birds of the world, vol 8. Lynx Edicions, Barcelona, pp 748–787 dae). Proc Zool Soc Lond 1858:232–254
´
Maurıcio GN, Mata H, Bornschein MR, Cadena CD, Alvarenga H, Sclater PL (1874) On the Neotropical species of the family
Bonatto SL (2008) Hidden genetic diversity in Neotropical birds: Pteroptochidae. Ibis 4:189–206
molecular and anatomical data support a new genus for the Sclater PL (1890) Catalogue of the birds in the British Museum, vol
‘‘Scytalopus’’ indigoticus species-group (Aves: Rhinocryptidae). 15. British Museum, London
Mol Phyl Evol 49:125–135 Sick H (1954) Zur Kenntnis von Ramphocaenus (Sylviidae) und
´ ´ ´
Menetries E (1835) Monographie de la famille des Myiotherinae ou ` Psilorhamphus (Formicariidae). Bonner Zool Beitr 5:179–190
´ ` ´
sont decrites les especes qui ornent le Musee de l’Academie ´ Swofford DL (1998) Paup*. Phylogenetic analysis using parsimony
´ ´ ´ ´
Imperiale des Sciences. Memoires de l’Academie Imperiale des (* and other methods), v.4. Sinauer, Sunderland
Sciences de Saint-Petersburg. Ser. 6, 3 (pt 2):443–543 Wetmore A (1926) Observations on the birds of Argentina, Paraguay,
¨ ¨
Muller J (1847) Uber die bisher unbekannten typischen Vers- Uruguay, and Chile. Bull United States Nat Mus 133:1–448
chiedenheiten der Stimmorgane der Passerinen. Abhandlung Wetmore A (1943) The birds of southern Veracruz, Mexico. Proc
¨
der Koniglichen. Akademie Wissenschaft zu Berlin, pp 321–391, United States Nat Mus 93:215–340
405–406 Wetmore A, Peters JL (1922) A new genus and four new subspecies
Nylander JAA (2005) MrModeltest v.2.2. Program distributed by the of American birds. Proc Biol Soc Washington 35:41–46
author. Department of Systematic Zoology, Uppsala University,
Uppsala
123
![338 J Ornithol (2010) 151:337–345
However, there was no coherent understanding of the su-
boscine groups until their distinction from the oscines was
¨
established by the pioneering work of Muller (1847) on the
¨
syrinx. Muller showed that Scytalopus was a tracheophone
suboscine and not a wren (Troglodytidae) and also that
Scytalopus and Pteroptochus differed from other known
passerines in having a four-notched sternum.
Sclater’s (1874) ‘‘Pteroptochidae’’ comprised Scytal-
opus (including the type species of what later became
Myornis Chapman 1915), Merulaxis, Liosceles, Pteropto-
chos (including the type species of what later became
Teledromas Wetmore and Peters 1922, and Scelorchilus
Oberholser 1923), Rhinocrypta, Hylactes (now included in
Pteroptochos), Acropternis, and Triptorhinus (= Eugralla).
Except for the problem genera Psiloramphus and Mela-
nopareia, this composition of the group was essentially
maintained until Peters (1951).
The three or four species of Melanopareia differ from
other tapaculos by their rather slender build and boldly and
attractively patterned plumage, and by sharing a semi-
concealed white dorsal patch with various true antbirds
(Thamnophilidae). They were originally described in the
genus Synallaxis (Furnariidae), in which Sclater (1890)
later submerged the genus. Salvin (1876) described a new
species from Ecuador as Formicivora speciosa, duly rec-
ognized in that combination by Sclater (1890), p. 251, and Fig. 1 The bamboo-wren Psilorhamphus guttatus bears little external
resemblance to typical members of the tapaculo family. With its grey
others until Hellmayr (1906), p. 334, showed that this was a iris, facial expression, bill shape, and wing-coverts with white dots
synonym of Synallaxis elegans Lesson, in which genus Psilorhamphus instead resembles some antbirds (Dysithamnus, Myr-
Hellmayr continued to place it while regarding Salvin’s motherula) with which early ornithologists consequently placed it.
allocation of it to Formicivora with incredulity. Ridgway Unlike other tapaculos Psilorhamphus spends most of the time above
the ground. Photo: Edson Endrigo
(1909) seems to have overlooked this when he created a
new genus Rhoporchilis for Formicivora speciosa. It was
Hellmayr (1921) who eventually established the modern comment that these genera ‘‘might perhaps be more natu-
concept of the genus in showing that Synallaxis elegans, S. rally placed as a distinct subfamily of Pteroptochidae
torquata, and S. maximiliani were congeneric and would [= Rhinocryptidae]’’ despite the fact that ‘‘there is little
all fall under Reichenbach’s earlier generic name Mela- external difference between the appearance of these birds
nopareia and ‘‘find their natural place in the Formicarii- and the true Wrens [Troglodytidae]’’. We are not aware,
dae’’, where they stayed for only a few years (Cory and however, of any instance in which Psilorhamphus was
Hellmayr 1924). Next came the observation of W.D.W. placed in either the Troglodytidae or in a family with only
Miller that the sternum of Melanopareia was four-not- sylviid-like genera, as might be inferred from Krabbe and
ched—information that was conveyed to and presented by Schulenberg (2003). Psilorhamphus continued to be asso-
Wetmore (1926), p. 292. On this basis, Peters (1951) ciated with Ramphocaenus in the Formicariidae—e.g.
included Melanopareia in the Rhinocryptidae, where it has Sclater (1890) and Cory and Hellmayr (1924), p. 205—
resided since. although in the latter reference it was noted that W.D.W.
The other problem species, the Bamboo-wren Psilor- Miller would show Psilorhamphus and Ramphocaenus to
hamphus guttatus, is a small bamboo specialist with a ‘‘constitute a separate family’’ in ‘‘a paper shortly to be
rather long, slender bill, a long tail, and relatively weak published.’’ Peters (1951), p. 213, later explained that
feet, so it bears little resemblance to large-footed terrestrial Miller’s death prevented publication of his results but that
´ ´ ´
tapaculos (Fig. 1). From the beginning (Menetries 1835) it Wetmore (1943), p. 306, had shown Ramphocaenus to
was placed with the antbirds in the Myiotherinae (= For- have an oscine syrinx, and had told Peters that Microbates
micariidae). Sclater (1858), p. 243, associated Psilorham- likewise was oscine and that he believed, on the basis of
phus with Ramphocaenus (which is now in the oscine external morphology, that Psilorhamphus was also proba-
family Polioptilidae) in the Formicariidae with the bly oscine. Therefore, Peters postponed his treatment of
123](data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7)