Proposal
(316) to South American
Classification Committee
Recognize
the genus Dendroplex Swainson 1827 (Dendrocolaptidae) as valid
Effect on South American CL: this would transfer polytypic Xiphorhynchus
picus and monotypic X. kienerii to Dendroplex.
Background: Since 1951, authors (Peters 1951, Clements
2000, Marantz et al. 2003, Dickinson 2003) have placed the
Straight-billed (X. picus) and Zimmer's (X. kienerii) woodcreepers
in the genus Xiphorhynchus, even though earlier authors classified them
in the genus Dendroplex (Sclater 1890, Hellmayr 1925, Zimmer 1934, Todd
1948). The original characterization of Dendroplex (1827: 354)
provided only a brief diagnosis of the new taxon, and no reference to a type
species. Ten years later, the same author (Swainson 1837: 313-314) provided
essentially the same diagnosis of the original description, but this time it
was accompanied by an illustration showing the straight culmen and lateral
compression of the type species. However, at the end of the characterization,
Swainson added: "The scansorial type D. guttatus Spix i, 91,
f. 1", which refers to figure 1 of plate 91 in Spix (1824), thereby
satisfying the requirements of ICZN for type species designation by subsequent
monotypy (ICZN 1999). Subsequently, Hellmayr (1925: 288) pointed out that
Swainson's diagnoses of 1827 and 1837 and bill outline correspond to the
characters of the Straight-billed Woodcreeper (originally described as Oriolus
picus), although the only species mentioned (D. guttatus Spix i, 91,
f. 1), "belongs to the genus Xiphorhynchus Swainson".
Following Hellmayr (1925), Peters (1951: 36) recognized that "D.
guttatus Spix i, 91, f. 1" depicts in fact a bird now known as Xiphorhynchus
ocellatus (Spix 1824), and stressed that under Opinion 65 (Schenk &
McMasters 1948: 54) the case of misidentification had to be formally presented
to the ICZN for ruling, and that until a decision was reached, Xiphorhynchus
ocellatus ocellatus = Dendrocolaptes ocellatus Spix 1824 continued
to be the type of Dendroplex.
New information: Aleixo (2002) showed with high
statistical support that the genus Xiphorhynchus (sensu Peters
1951) is paraphyletic, and that the sister taxa X. picus and X.
kienerii are the only species which do not belong in a clade with the
remaining Xiphorhynchus species; actually, all phylogenetic
hypotheses obtained strongly indicated that X. picus and X. kienerii belong
to a separate clade not nested within any of its closely related and apparently
monophyletic genera Campylorhamphus, Lepidocolaptes, or Xiphorhynchus
(Aleixo 2002). Thus, he suggested that they might be grouped in a different
genus, in which case the name Dendroplex Swainson, 1827, would be
available if problems with its type species designation were resolved.
Because the latest (fourth) edition of the International Code of
Zoological Nomenclature (ICZN 1999) now allows a misidentified type species to
be set aside without the requirement of a ruling from the Commission, Aleixo et
al. (2007) proposed the conservation of Dendroplex Swainson, 1827, as a
valid taxon. According to them, the following lines of evidence support
Hellmayr's (1925) interpretation that Swainson's (1837) identification of
"D. guttatus Spix, i, 91, f. 1" as the type was a case of
misidentification, and that D. picus = Oriolus picus J. F.
Gmelin 1788 was the taxon upon which Swainson actually based Dendroplex:
1) Swainson (1827: 354) himself was unsure whether the taxon on
which he based Dendroplex had been described or not. Ten years later,
when he next cited Dendroplex (Swainson 1837: 313-314), the original
diagnosis was maintained and even illustrated in detail (see Fig. 1), but
"D. guttatus Spix, i, [pl.] 91. f. 1." was mentioned as
belonging to the genus apparently in accordance with Lesson (1830: 313), who a
few years before explicitly equated "D. guttatus Spix, pl. 91"
with "Oriolus picus Gm"
2). There is a significant discrepancy between the levels of detail
and resolution of the bill outline presented in figure 281e of Swainson (1837:
313) as diagnostic of Dendroplex (see Fig. 1) and that of D. guttatus
as illustrated in Spix's plate, chosen by Swainson (1837) as the type of Dendroplex.
While the latter illustration is poor in resolution and depicts a bird which in
fact resembles several taxa currently classified in the genus Xiphorhynchus,
figure 281e of Swainson (1837), on the other hand, is very well resolved and
refers unambiguously to the only species in the entire family Dendrocolaptidae
known to this day to possess such a bill shape: X. picus = Oriolus
picus J. F. Gmelin, 1788 (Marantz et al. 2003).
Overall recommendation: Those taxa originally described
or classified as Dendroplex according to Gray (1840), Sclater
(1890), Hellmayr (1925), Zimmer (1934), and Todd (1948), but later transferred
to Xiphorhynchus by Peters (1951), should be returned to Dendroplex,
which will contain only two distinct sister biological species: the polytypic Dendroplex
picus and the monotypic Dendroplex kienerii, as delineated by
Marantz et al. (2003).
Literature cited
Aleixo, A.,
S. M. S. Gregory & J. Penhallurick. 2007. Fixation of the type species and
revalidation of the genus Dendroplex Swainson, 1827 (Dendrocolaptidae). Bull.
B. O. C. 127: 242-246.
Aleixo, A.
2002. Molecular systematics and the role of the
"várzea"-"terra-firme" ecotone in the diversification of Xiphorhynchus
Woodcreepers (Aves: Dendrocolaptidae). Auk 119: 621-640.
Clements,
J. F. 2000. Birds of the World, A Checklist. Pica Press,
Robertbridge, East Sussex.
Dickinson,
E. C. 2003. The Howard & Moore Complete Checklist of the Birds of
the World. Christopher Helm, London.
Gray, G. R.
1840. A List of the Genera of Birds, with an indication of the typical
species of each genus. Compiled from various sources. Richard and John E.
Taylor, London.
Hellmayr,
C. E. 1925. Catalogue of Birds of the Americas and the adjacent Islands,
part 4. Furnariidae-Dendrocolaptidae. Field Museum of Natural History, Chicago.
I.C.Z.N.
1999. International Code of Zoological Nomenclature. 4th.
Edition. The International Trust for Zoological Nomenclature, London.
Lesson, R.
P. 1830-31. Traité d'Ornithologie ou Tableau Méthodique. F. G.
Levrault, Paris.
Marantz,
C., Aleixo, A., Bevier, L. R. & Patten, M. A. 2003. Family Dendrocolaptidae
(Woodcreepers). Pp. 358-447 in J. del Hoyo, A. Elliott &
Christie, D., (eds.) Handbook of the Birds of the World, Vol. 8,
Broadbills to Tapaculos. Lynx Edicions, Barcelona, Spain.
Peters, J.
L. 1951. Check-list of Birds of the World, vol. 7. Harvard University
Press, Cambridge, MA.
Schenk, E.
T. & McMasters, J. H. 1948. Procedure in taxonomy. Revised ed.
Stanford University Press, Stanford California.
Sclater, P.
L. 1890. Catalogue of the Birds in the British Museum, vol. 15.
Catalogue of the Passeriformes or perching birds, in the collection of the
British Museum. Tracheophonæ, or the families Dendrocolaptidae, Formicariidae,
Conopophagidae, and Pteroptochidæ. Trustees of the British Museum, London.
Spix, J. B.
1824. Avium species novae, quas in itinere per Brasiliam. vol.
1. Hübschmanni, Munich.
Swainson,
W. 1827. Several new groups in Ornithology. Zoological Journal 3:
158-363.
Swainson,
W. 1837. The Natural History and Classification of Birds, vol. 2.
Longman, Rees etc., London.
Todd, W. E
.C. 1948. Critical Remarks on the Wood-hewers. Ann. Carnegie Mus. 31(2):
5-18.
Zimmer, J.
T. 1934. Studies of Peruvian Birds, part 14. Am. Mus. Nov. 753:
1-26.
Alex
Aleixo, October 2007
______________________________________________________________________________________________________
Comments from Nores: "YES. Tanto el análisis
molecular que muestra a picus y a kienerii como
un género diferente de Xiphorhynchus, como las razones para
resurgir el género Dendroplex, están para mi muy bien
fundamentadas."
Comments from Cadena: NO. The proposal states that
Aleixo (2002) demonstrated that Xiphorhynchus is paraphyletic and that X.
picus and X. kienerii belong to a separate clade with "high
statistical support", an assertion that I think is somewhat inaccurate. It
is true that the analyses convincingly showed Xiphorhynchus to be
paraphyletic (in the sense that X. fuscus, then included in Lepidocolaptes,
is nested within it), but support for the finding that picus and kienerii are
more closely related to other woodcreeper genera is far from strong. Figure 1
in Aleixo (2002) is the strict consensus of the most parsimonious trees; here,
the clade formed by picus and kienerii occupies an
unresolved position, which implies that one cannot reject the hypothesis that
these two taxa form a clade that is sister to the rest of Xiphorhynchus,
a situation that would not require a taxonomic change. Figure 2 is a maximum
likelihood tree in which picus and kienerii form a clade that
seems to be allied with other woodcreeper genera and not with the remainder of Xiphorhynchus.
However, the relevant node here has less than 50% bootstrap support, so again
one cannot reject the hypothesis that these two species are sister to the rest
of Xiphorhynchus. Figure 4 is a Bayesian tree, and here again picus
and kienerii form a clade that does not appear to be sister to the rest
of Xiphorhynchus, but the posterior probability of the relevant node is
only 64%, which again is far from significant. In sum, the data are suggestive
but not conclusive, so I'd rather wait until the new comprehensive studies of
Furnariid phylogeny based on more genes (i.e. Brumfield et al., Moyle et al.)
are completed to decide whether this proposed change is warranted."
Additional comments from Aleixo: "Cadena is right in his interpretation that the clade picus/kienerii cannot
be probabilistically or parsimoniously completely ruled out as the sister clade
of all remaining Xiphorhynchus species plus ex-Lepidocolaptes fuscus in
which case, according to his opinion, the recognition of Dendroplex as
a separate genus from Xiphorhynchus would not be justified.
“Even though none of the phylogenies presented
in Aleixo (2002) show picus/kienerii and the remaining Xiphorhynchus
species as sister taxa (in fact ML and Bayesian estimates suggest a completely
different topology), strong statistical support falsifying this relationship is
also lacking. Nonetheless, what all phylogenies presented in Aleixo (2002) did
show with strong statistical support, was that the picus/kienerii clade
cannot be grouped probabilistically or parsimoniously within the remaining Xiphorhynchus
species. Considering that this study sampled all known species of Xiphorhynchus
(by far the largest woodcreeper genus), those results strongly support the
evolutionary independence of the picus/kienerii clade
irrespective of the resolution concerning its sister clade. After all, is there
a completely objective way to delimit genera using phylogenies? For example, it
could be also argued that the phylogenies presented in Aleixo (2002) support
the recognition of two genera within the "true Xiphorhynchus",
since two well-supported clades were consistently recovered by all phylogeny
estimates.
“When this is contrasted with the taxonomic
history involving taxa of the picus/kienerii clade, it becomes
more apparent that the results showed in Aleixo (2002) finally provide a
phylogenetic basis for the distinct treatment those taxa received originally
when grouped under the genus Dendroplex by taxonomists such as C.
E. Hellmayr and J. T. Zimmer. In fact, the history of how Dendroplex was
suppressed makes it evident that in fact no hard evidence has ever been
presented to lump it with Xiphorhynchus in the first place! The genus Dendroplex
was used for both picus and kienerii (back then necopinus)
until 1948 when Todd (1948) moved kienerii (but not picus) to Xiphorhynchus,
and this was done in just a single sentence (!), which I transcribe below:
"I am
convinced that the resemblance, close as it is, between necopinus and Dendroplex
is superficial and fortuitous, so to speak, and not indicative of congeneric
affinity."
“In Peters (1951), picus was finally moved
to Xiphorhynchus solely on the basis of the nomenclatural
issues surrounding the name Dendroplex, which Aleixo et al. (2007)
finally helped to solve. Thus, the only "evidence" conflicting with
the independent taxonomic treatment which picus/kienerii have
historically always received in the genus Dendroplex was Todd's sentence
arguing for the paraphyly of Dendroplex, which as Aleixo (2002)
showed, has absolutely no phylogenetic basis.
“In sum, what seems really key to the discussion
is that the phylogenies presented in Aleixo (2002) strongly support the
evolutionary independence of picus/kienerii from Xiphorhynchus,
as recognized early on by their assignment to a different genus. Therefore, the
bulk of all available evidence (traditional taxonomy + molecular phylogeny)
favors the recognition of Dendroplex rather than their lumping with
Xiphorhynchus."
Additional comments from Cadena: "I appreciate all the
points made by Alex and he is absolutely right that one cannot positively (i.e.
statistically) demonstrate that picus and kienerii are not
more closely allied to other woodcreepers than to other taxa currently placed
in Xiphorhynchus. That said, the important issue here is that there
is a hypothesis about relationships that is implicit in our current taxonomy
(i.e. that picus and kienerii belong in a clade with the
rest of Xiphorhynchus), and the data in Aleixo (2002) do not allow
rejecting such hypothesis. Therefore, because we strive to maintain a stable
classification, I argue that this change is not (yet) well-substantiated
because it is not based on evidence that allows falsifying our current
hypothesis, so I see no need to change. Again, it is true that one cannot
"prove" that our current hypothesis is correct (which is true for
essentially all scientific hypotheses), but unless there is evidence to reject
it, I maintain that we should not modify our classification. If we were
starting to classify woodcreepers from scratch, I would probably go for Alex's
suggestion, but given that we need to work based on an existing classification,
I believe we need to be conservative and accept only changes that are strongly
supported by published evidence. And, with all due respect, I insist that the
assertion that "the phylogenies presented in Aleixo (2002) strongly
support the evolutionary independence of picus/kienerii from Xiphorhynchus"
is not correct. The standard ways to assess “strong” support for phylogenetic
relationships are clade support values (bootstrap proportions and posterior
probabilities) and statistical tests of topologies. Clade support values in
this case are not strong, and test of topologies have not been conducted (and
would very likely not reject the monophyly of Xiphorhynchus as currently
defined). In sum, lets wait for the detailed phylogenetic analyses that are
underway; I predict Alex will be right at the end, but the evidence to
substantiate his proposed change is simply not out there yet."
Additional comments from
Aleixo: "I am not arguing that the phylogenetic
position of picus / kienerii is well resolved, but again,
nodal support indices indicate with a high degree of confidence that they are
not nested within a clade grouping all remaining species of Xiphorhynchus (all
phylogenies shown in Aleixo 2002 clearly show that). Whether or not you think that
this degree of evolutionary independence is enough to delimit different genera
ends up being a matter of taste. My main point is actually that the current
hypothesis about relationships implicit in our current taxonomy (i.e.
that picus and kienerii belong in a clade with the
rest of Xiphorhynchus), to use Cadena´s words, results from bad taxonomy
practice in the first place, and that in a way is totally independent from the
phylogenetic evidence available nowadays. In fact, if it were not for Peters´ (1951)
comments on the nomenclatural issues preventing the use of Dendroplex,
this would be exactly the "null-hypothesis" that would be implied by
current taxonomy. In other words: Dendroplex was submerged into Xiphorhynchus due
to a nomenclatural issue, not as the result of a taxonomic review. As far as
the only "hard-data" reviews to date on picus/kienerii are
concerned (Hellmayr 1925, Zimmer 1934, Aleixo 2002), they all point towards the
evolutionary / taxonomic independence between those two biological species and
the remaining species of the genus Xiphorhynchus. Thus, when Aleixo et
al. (2007) "rescued" the nomenclatural validity of Dendroplex,
they restated the original null-hypothesis of evolutionary relationships
implied by taxonomy before the nomenclatural impediment discussed by Peters
(1951) was put into effect. Furthermore, having picus / kienerii grouped
under Dendroplex is more consistent with nomenclatural stability
than have them grouped under Xiphorhynchus, since the former
treatment had been in place for nearly 150 years before its replacement. With
that said, the molecular data shown in Aleixo (2002) is not at odds with the
proper taxonomic null hypothesis that should be evaluated when
considering picus/kienerii, i.e., their placement in a separate
genus from Xiphorhynchus."
Comments from Stiles: "YES, despite the caveats of
Daniel. It clearly doesn't belong in Xiphorhynchus, and whether or not
to split the latter is not an issue at present."
Comments from Robbins: "YES. This really comes down
to one's choice in whether the uniqueness of picus and kienerii merit
generic distinction from other species currently treated as Xiphorhynchus.
I could go either way on this, but given that these taxa were originally placed
in Dendroplex and the subsequent transfer of them into Xiphorhynchus
was based solely on superficial plumage similarities, I vote "yes" on
this proposal."
Subsequent Comments from Robbins: "Change to NO. I was
on the fence on whether to recognize Dendroplex. I'm quite copasetic
with following Curtis's suggestion of waiting on making a change."
Comments from Stotz: "YES. While I
recognize Daniel's argument that the data presented by Aleixo and company does not
absolutely demonstrate that Xiphorhynchus would be paraphyletic if it
included, it does suggest that would in fact be the case. Given that the
lumping of Dendroplex into Xiphorhynchus was done on such
weak grounds in the first case, I think splitting them off, even though they
might be sister to Xiphorhynchus makes sense."
Comments from Remsen: "NO. Tough decision.
Although I appreciate Alex's excellent point that the only reason that the
status quo keeps these species in Xiphorhynchus is from an
error in nomenclature, I think that it is necessary to demonstrate that keeping
the two Dendroplex in Xiphorhynchus would beyond a doubt
make this a paraphyletic genus. Existing genetic data are suggestive but not
conclusive. Until additional data fortify the hypothesis that Dendroplex is
closer to other genera, I see no reason to place them in a separate genus. I am
unimpressed with any subjective rationale as to why this group of birds differs
in any meaningful way from Xiphorhynchus in terms of plumage,
voice, and morphology; bill shape by itself, one of the most plastic characters
in bird morphology, is not a valid delimiter of generic boundaries."
Comments from Zimmer: "YES, for reasons stated by
Alex in the proposal, and by Mark, Gary and Doug."
Comments from Pacheco: "YES. Os argumentos apresentados pelo Alex, são ao meu ver,
apesar das opiniões respeitáveis de Cadena e Van, suficientes e consistentes
para o reconhecimento de Dendroplex no arranjo proposto -- como
válido."
Comments solicited from Curtis Marantz: "This is a difficult situation for two reasons. First,
as Daniel pointed out, the genetic evidence published to date is equivocal
regarding the need for recognizing the two Dendroplex as separate
from Xiphorhynchus. I would probably be more strongly swayed by the evidence
already published had there not been a project underway looking in depth at
Furnariid phylogeny (including the Dendrocolaptidae / Dendrocolaptinae).
Under the assumption that this complex will be studied more extensively in the
near future I would recommend waiting to see if the situation is better resolved
before making any changes.
“Secondly, I have looked into the nomenclatural
issues discussed by Aleixo et al. (2007) and must admit that they are most
complex. It indeed appears that Swainson (1837) erred when he chose D.
guttatus as the type for Dendroplex despite the fact that his figure
281e indeed depicts X. picus and only X. picus. Although I
am far from an expert on nomenclatural issues I would think that common sense
would dictate that, if at all possible, the best course of action to take, if
indeed the recognition a separate genus is warranted, may be to abandon the
name Dendroplex altogether and propose a new name. This would
seem reasonable given not only the confused past surrounding Dendroplex but
also because the bill characters first used by Swainson to diagnose Dendroplex do
not really apply to kieneri, which has a somewhat different bill
shape."
Comments from Jaramillo: "YES - I am swayed by Alex's
reasoning, and going a bit on gut feeling here. There is room for a certain
amount of subjectivity in these decisions on genera, and I will be honest about
it."