Proposal (616) to South
American Classification Committee
Split Cacicus leucoramphus from Cacicus chrysonotus
The paper by Alexis F.L.A. Powell, F. Keith Barker, Scott M. Lanyon, Kevin J.
Burns, John Klicka, Irby J. Lovette (2013 “2014”) A comprehensive species-level
molecular phylogeny of the New World blackbirds (Icteridae) Molecular Phylogenetics and Evolution, 71 (Dec.13): 94-112 offers
radically new insights into the phylogeny of all four subfamilies included in
the Icteridae: Sturnellinae: Meadowlarks; Cacicinae: Caciques and Oropendolas;
Icterinae: Orioles; Agelaiinae: Blackbirds, Cowbirds and Grackles.
To quote from their abstract: “Using mitochondrial gene sequences from all ~108 currently recognized
species 7 and six additional distinct lineages, together with strategic
sampling of four nuclear loci and 8 whole mitochondrial genomes, we were able
to resolve most relationships with high confidence. Our phylogeny is consistent with the
strongly-supported results of past studies, but it also contains many novel
inferences of relationship, including unexpected placement of some newly
sampled taxa, resolution of relationships among major clades within Icteridae,
and resolution of genus-level relationships within the largest of those clades,
the grackles and allies. “
Below
I have inserted part of Figure 4 from their paper focusing of the
Cacicinae. The node uniting Cacicus chrysonotus and Cacicus leucoramphus is the second
deepest of the all of the species nodes in the Cacicinae. It is much deeper
than the node joining Cacicus sclateri
and Cacicus koepckeae and marginally
deeper than the node joining the two species just mentioned and Cacicus chrysopterus. It is also much deeper than the node joining Cacicus haemorrhous and Cacicus (was Ocyalus) latirostris and Cacicus (was Clypicterus) oseryi.
The
range of Cacicus leucoramphus
(northwest Venezuela to Ecuador) is distinct from that Cacicus chrysonotus pacificus (north Peru) and Cacicus chrysonotus chrysonotus of south Peru and north Bolivia.
Accordingly,
I propose that Cacicus leucoramphus.be
recognised as a species distinct form Cacicus
chrysonotus.
References:
Alexis F.L.A. Powell, F. Keith Barker, Scott M. Lanyon, Kevin J.
Burns, John Klicka, Irby J. Lovette (2013 “2014”) A comprehensive species-level
molecular phylogeny of the New World blackbirds (Icteridae) Molecular Phylogenetics and Evolution, 71 (Dec.13): 94-112
John Penhallurick
______________________________________________________________________________
Comments
from Remsen: “NO. I appreciate the point that these two
differ in the same ways that C. sclateri
and C. koepckeae do in terms of
plumage but are roughly 3X more differentiated at the neutral loci
sampled. However, the design of this
study is inadequate for addressing differences at this level of taxonomy
because the two samples of leucoramphus
were taken from N. Ecuador (Imbabura), roughly 1800 km from the nearest
population of chrysonotus in southern
Peru. Not only would
isolation-by-distance come into play dramatically here, but those 1800 km cross
multiple major biogeographic boundaries, especially the Marañon, that would
likely create strong genetic subdivisions in any montane cloud-forest
bird. To use genetic data to address
species limits in these allopatric taxa, sampling would have to be directed at
the potential contact zone in southern Peru.
Hellmayr (1938, Catalogue of Birds of the Americas) noted that there are
plumage hints of possible gene flow between these two. [The Bond (1953) reference in our footnote
also evidently mentions this also but I can’t find my copy to check.] Note also that Powell et al., almost
certainly aware of the limitations of their data with respect to species limits
in these taxa, did not mention it in their discussion of classification and
treat the two as subspecies in their tree.
There is no relevant evidence in Powell et al. that chrysonotus and leucoramphus
should be treated as separate species.”
Comments
from Cadena: “NO. Sure,
genetic divergence is strong, but this may well reflect variation within a
single biological species. Using genetic distance as a yardstick to assess
species status is extremely problematic for various reasons (in brief, the
correlation between time and reproductive isolation is rather messy), and I see
no strong published evidence based on other traits (voices, behavior, plumage
variation) to support this proposed split.”
Comments
from Dan Lane: I agree with Van that the Powell et al.
(2013) paper can hardly be used to make taxonomic changes of this nature given
the limited sampling of individuals, the geographic distance between the
individuals sampled, and the genes used. But in addition to that, I have
further concerns about the above proposal. First, the suggestion that C.
leucoramphus be separate from C. c. “pacificus” (sic, should be peruvianus.
Pacificus is part of the C. uropygialis/microramphus group) and C.
c. chrysonotus is a novel taxonomy that has not been suggested anywhere
before that I can find, and as such would require considerably more background
to support. Cacicus leucoramphus, when split from C. chrysonotus,
has always included peruvianus as both have the yellow wing patches.
Assuming that this change to the organization of taxa within the complex was
unintentional, an additional issue arises as there are, as Van also stated,
specimens suggesting introgression between peruvianus and leucoramphus.
I myself have collected such an individual in the Mantaro valley, Junin, which
is near the supposed distributional break between peruvianus and leucoramphus.
A more densely sampled study including all three named taxa, including
populations such as that from the Mantaro valley, would be necessary to show
conclusively whether or not chrysonotus, peruvianus, and leucoramphus
are acting as good biological species or not. Until then, I would err on
the side of caution and maintain them as one species.”
Comments
from Stiles: “NO, for
the reasons advanced by Van and Daniel. An 1800 km gap between two small
samples makes the interpretation of the genetic differences equivocal. Without
statistically and geographically meaningful samples of genetics, plumages and
voice, the evidence for this split is insufficient to justify it.”
Comments from Nores: “YES. Although the data presented
here are not sufficient to justify the split, the two taxa are too different
(especially the lack of yellow on wings in chrysonotus).
Note that Schulenberg et al. (Birds of Peru) and Hilty (Birds of Venezuela)
treat chrysonotus and leucoramphus as separate species.”
Comments
from Zimmer: “NO, for
reasons already elucidated by Van, Daniel, Gary & Dan.”
Comments
from Jaramillo: “YES – this
is a poorly constructed proposal, but let’s not throw out the baby with the
bathwater here. In our Icteridae book we had some information on this problem.
First of all, Blake (1968), the Peter’s Checklist, actually considered chrysonotus separate from leucoramphus. To clarify as Dan has
done, there is no pacificus involved
in this issue (that is another issue), the three taxa fall into two groups with
peruvianus being the southern form of
leucoramphus. The two (chrysonotus with leucoramphus) were merged later based on intermediate looking
specimens, as Dan confirms. However, it was not clear to me who first came up
with the idea of the merger, it is too long since I looked into the issue, but
I do not think it is based on anything other than the few specimens of southern
birds with yellow feathers on the wing! The issue here is that chrysonotus with some yellow on the wing
occur way south in the range, in Bolivia. It appears to me that these birds
with some yellow on the shoulder are a variation that can pop out anywhere, not
intermediates. This is similar to C.
sclateri specimens that show some yellow on the rump both from Peru and
Ecuador! It is just something that shows up, perhaps a “throwback” to an
ancestral plumage state that is still in the bird’s genes?
“I would argue
that these two taxa were lumped based on poor analysis of the data, birds that
were thought to be intermediate are not, and I say we go back to Blake’s idea
and treat the two as separate species given that coloration, voice, and
genetics point to two sisters taxa that in all of those characteristics are as,
or more, different than other Cacique species we currently recognize.
At the time of
the book I had only heard the northern birds, and eventually in Bolivia I was
able to hear and see the southern birds. Calls are consistently different, and
call notes are important in flock cohesion in these social birds. Now we have
some recordings of their songs and they are extremely different to my ears:
Song of leucoramphus: http://www.xeno-canto.org/63204
Song of chrysonotus: http://www.xeno-canto.org/1996
“Furthermore my
limited personal experience with both of these birds is that the southern birds
(in Bolivia) are much shyer and skulking, difficult to see. While northern
birds (Ecuador) are much bolder and obvious. Of course that is a distance
apart, and there could be clines in behavior.
“Below are
specific parts from our book that deal with the problem:
… Ridgely and Tudor (1989) state that the vocalisations of the northern
leucoramphus group and the southern chrysonotus group to be similar. However,
recordings of the primary call (the ‘wheehk’ or ‘whak’) call appear to show a
difference. The southern birds sound higher pitched, metallic and quick; while
the northern birds are lower pitched, harsher and more powerful sounding. More
research needs to be done to elucidate how similar the vocalisations of
northern and southern forms are. ….. [we now have a lot more
information on xeno-canto to compare and confirm that these differences are
real and throughout the distribution]
GEOGRAPHIC VARIATION Three subspecies that fall into
two groups: leucoramphus [Northern Mountain Cacique] and chrysonotus [Southern
Mountain Cacique]. These two forms were listed as separate species in Blake
1968, but more recently the two have been considered conspecific (Ridgely and
Tudor 1989) mainly due to the presence of intermediate looking specimens.
Specimens of peruvianus from Auquimarca, Peru show black fringes to the yellow
wing coverts (Bond 1953). However, the vocalisations do not appear to be
similar, as has been stated (Ridgely and Tudor 1989). The two are certainly
good phylogenetic species, and perhaps even good biological species, more work
needs to be done before a decision as to the systematic placement can be made.
The
Northern Mountain Cacique is composed of two races. C .c. leucoramphus lives
from NW Venezuela, south along the E Colombian Andes to E Ecuador. It is
described above. C. c. peruvianus inhabits the east slope of the Andes from
Amazonas south to Junín. It is similar to leucoramphus, but has a heavier bill,
with the culmen more noticeably arched. The bluish base to the bill is less
extensive. As well, the concealed white collar is thinner in this form than in
leucoramphus.
The Southern Mountain
Cacique, C. c. chrysonotus, is found south of Junín, Peru to Cochabamba,
Bolivia. It lacks yellow on the wing coverts, however some individuals may show
one or two yellow covert feathers or small yellow tips to several coverts. This
is particularly true in the northern part of their range. However, some
Bolivian chrysonotus specimens, far from where intergradation could be
occurring, show yellow fringes to some of the wing coverts (Bond 1953).”
Comments from Dan Lane:
“Reading
the further responses in this proposal, I feel there are some clarifications of
the situation that I am in a position to make.
“First of all, in response to the comment above by Nores,
Birds of Peru (Schulenberg et al. 2007, 2010) did *not* separate C.
chrysonotus from leucoramphus at the species level, we maintained
them as conspecifics.
“Although I appreciate Alvaro’s comments on the subject, I
think they oversimplify the situation regarding this complex. Hand-picking two
recordings of what may or may not be homologous vocalizations from the broad
repertoire of an oscine with a linear distribution isn’t a very satisfying
comparison to me. Yes, the two recordings are both of songs, but are they the
same song types given under the same circumstances? Do we really even know what
kind of song types C. chrysonotus/leucoramphus has? Al’s comments from
Proposal 624 suggest that he believes there are several song types within any
given population of C. cela—an idea with which I agree—so why would this
congener be any different? And as we know is true in icterids, dialect
formation will be a serious issue to consider when dealing with geographic
variation among populations within this complex. Take a look at all the
recordings now available online:
http://www.xeno-canto.org/species/Cacicus-chrysonotus
http://www.xeno-canto.org/species/Cacicus-leucoramphus
“The Macaulay recordings include a selection which fills the
noticeable gap from central Peru (La Libertad, Huánuco, Pasco depts..) among
the Xeno-canto cuts. Listening to these cuts from San Martin to Bolivia, I hear
no obvious, discrete break in vocalization types between peruvianus and chrysonotus
to correspond to the turnover of plumage types. I have uploaded my own
recordings of the complex, including several from the Mantaro valley (both
sides), where birds with intermediate plumage are found (see a photo here of an
intermediate bird that was collected from a flock that I had recorded):
https://www.flickr.com/photos/8013969@N03/13547056163/
“I should point out that I used recordings of C. c.
peruvianus successfully to draw in the birds of intermediate plumage in
Junin.
“Checking LSU specimens, we have 13 C. chrysonotus
from Puno, Peru, and Bolivia. Of these, none show the yellow on the shoulder
visible in the Junin photo linked above. Colombian birds in our collection
appear smaller, with Bolivian birds appearing largest, but this could be
Bergman’s Rule at work, as the increase in size appears clinal. Again, I’m
really not impressed by the variation in voice among the recordings, especially
when one compares recordings along the length of the distribution of the
complex, rather than listening to extremes at distant points within it. The
lack of a clear change in voice when one compares recordings from the
Junin/Huancavelica/Cusco region of Peru, where the two 'species' turn over is
particularly unconvincing as evidence that there are indeed two species,
and added to this the presence of intermediate birds at this same area, it
almost appears as if this is one of the few cases among Andean birds where
allopatry does not stymie our application of the Biological Species Concept!
“Regardless, I think that I
would ‘throw the baby out’ with the present proposal. As it is worded, it does
a very poor job of clarifying the situation. A new proposal, with real
comparative data, would be needed (in my opinion) to make a strong case for the
separation of the complex into two species. Those data should involve a much
more directed study, including tissues from specimens from the turnover zone in
Pasco/Junin/Huancavelica/Cusco, as well as from all along the distribution of
the complex, as this is really necessary to make further sense of the situation
here.”
Comments from Robbins: “NO. After reading Dan Lane’s comments (especially
those in response to Alvaro’s remarks), I feel that much more information is
need in the region between the two extreme genetic samples.”
Comments from Pacheco: “NO. A partir das lúcidas colocações de Dan Lane, penso que
é muito prematuro decidir algo sem a existência de uma revisão consistente da
relação entre estes dois táxons.”