Proposal (775) to South American Classification Committee
Split Urochroa bougueri into two species
The genus Urochroa
has been considered monotypic by virtually all authors in the past, despite
the fact that notable differences exist between the two subspecies of bougueri: nominate bougueri and leucura. Ridgely
& Greenfield (2001) were the first to suggest that two species might be
involved, although they did not implement this suggestion, which was formally
taken up by del Hoyo & Collar (2014) and endorsed by Donegan et al. (2015).
Here I consider this question in more detail.
The previous authors based their decision mainly upon details of plumage
coloration and distribution, although Donegan et al. did find appreciable
differences between the call notes of the two -the songs of these taxa (if
indeed they do sing) have yet to be recorded.
Urochroa. b. bougueri is
restricted to the very wet Pacific slope of the Western Andes from Colombia to
W Ecuador, typical of Chocó endemics, whereas leucura occupies the Eastern slope of the Andes from southern Colombia
south to central Ecuador; both occupy middle elevations between ca. 1000 and
2000 m. Here I examine this question in rather more detail.
The plumage characters used by previous authors to
justify this split are summarized in the following table (with some amendments
I have added).
Character |
Sex |
bougueri |
leucura |
Upperparts |
♂ |
Very dark: blackish
greenish-bronze, brighter bronze on upper tail-coverts |
Much paler and
brighter green throughout, including upper tail-coverts |
Central rectrices |
“ |
Black, more or less
tinged bluish green |
Greenish bronze, in
some with bluish tinge distally |
Lateral rectrices |
“ |
R2-4 extensively
white, especially on inner webs; outer webs with black borders and along
shafts; outer web r5 mostly black, white only at base: inner web also with
broad black border |
R2-4 almost entirely white
with at most narrow black borders toward tip; r5 similar but black border of
outer web longer and somewhat broader. |
Sides of head |
“ |
Blackish, but with a
conspicuous short, broad rufous malar stripe extending back from gape; a
small but definite cinnamon postocular spot |
Greenish-black, the
rufous malar marking absent or limited to an inconspicuous small, dark spot
at gape; cinnamon postocular spot usually lacking, at most very small and
inconspicuous. |
Throat to upper
breast |
“ |
Dark violet-blue,
sharply defined posteriorly |
Similar or slightly
paler, brighter blue, may be less sharply defined |
Lower breast |
“ |
Usually some dark
greenish-bronze feathers, especially laterally |
Bright green feathers
more extensive laterally, but may form a continuous band across breast |
Abdomen |
“ |
Dark brownish-gray
with some dark greenish-bronze feathers on sides and flanks |
Decidedly paler gray,
with more extensive green on sides and flanks |
Throat and chest |
♀ |
Similar to male but
somewhat less extensive posteriorly, the blue feathers with dark
grayish-brown fringes giving a scaly appearance |
Also similar to male
but the blue slightly paler and brighter; at most very narrow pale gray
fringes broader posteriorly, giving way to a green-spotted effect on extreme
lower breast |
Abdomen |
“ |
Similar to that of
male |
Similar to or
slightly paler than male |
Again, previous authors have failed to mention that bougueri is most conspicuously the
darker bird, especially on the upperparts and abdomen, again obvious from
several meters away. Also previously
overlooked is the fact that bougueri is
a much larger bird, sex for sex: see the following table. Overlap in bill, wing, and tail lengths is
limited in males, and nonexistent in females, as well as in other parameters in
males, but small sample sizes preclude meaningful statistical analyses. In
particular, note the much lower body mass, narrower wings and lower wing loading
in both sexes of leucura, all of
which suggest differences in foraging and aggressive behavior between the two
taxa.
Measurement |
bougueri ♂♂ (n=7) |
leucura ♂♂ (n=2, except where *, n=1) |
bougueri ♀♀ (n=3) |
leucura ♀♀
(n=2) |
Total culmen |
34.36 ±1.50 |
32.80 ± 1.13 |
36.27 ± 1.29 |
33.25 ± 0.21 |
Closed wing length |
78.63 ± 2.43 |
76.70 ± 0.57 |
73.97 ± 0.85 |
71.20 ± 1.27 |
Tail length |
47.76 ± 2.81 |
45.87 ± 1.16 |
45.25 ± 0.35 |
43.05 ± 0.49 |
Tarsus length |
7,57 ± 0.35 |
6.75 ± 0.07 |
7.13 ± 0.22 |
6.65 ± 0.07 |
Body mass |
11.71 ± 0.60 |
8.90 ± 0.57 |
10.30 ± 0.53 |
7.75 ± 0.21 |
Wing shape |
2.973 ± 0.054 |
0.311* |
2.836 ± 0.081 |
3.036 ± 0.042 |
Wing loading |
0.304 ± 0.020 |
0.275* |
0.238 ± 0.010 |
0.245 ± 0.030 |
Specimen photos:
With regard to possible species status, the two most
important differences in color between the two, the present vs. absent rufous
malar and the more striking differences in tail colors, could well be important
in species recognition and mate choice.
The large size difference is unusual among undoubted subspecies; an
example of two species differing less in plumage but similarly in size are the
narrowly parapatric Campylopterus
excellens and curvipennis of
southwestern Mexico. The differences in
color at least match some of those in the closely related genera Heliodoxa and Boissoneaua but size differences, both between species and between
sexes of the same species, are less extreme than in the two taxa treated here;
the same goes for the also related genus Coeligena. It is perhaps worth mentioning that on
the same slope as leucura, two
species of Eutoxeres differing
slightly less in plumage but virtually identical in size overlap rather broadly
without interbreeding. Vocal
differences, as evidenced by several sonograms, also exist, based on the
available evidence– but both forms are notably quiet and perhaps do not sing at
all.
Although genetic evidence is lacking, all of the
aforementioned considerations lead me to recommend a YES on this split also. The English epithets “Rufous-gaped” and
“White-tailed” Hillstars suggested by Donegan et al. for bougueri and leucura respectively
are descriptive and accurate. However, restricting the latter name to leucura rather than to the nominate goes
against current practice. Although I prefer these names, if one must bow to
convention the more aseptic names Western (or Pacific or Chocó) and Eastern
could serve, as the two taxa also show widely disjunct distributions.
F.
Gary Stiles
References
Del
Hoyo, J. & N.J. Collar. 2014. Illustrated checklist of Birds of the world,
vol. 1 (non-Passerines). Lynx Edicions, Barcelona, Spain and BirdLife
International, Cambridge, UK.
Donegan,
T., A. Quevedo, J.C. Verhelst, O. Cortés-Herrera, T.
Ellery & P. Salaman. 2015. Revision of the status of Bird species occurring
in Colombia, with discussion of BirdLife International’s new taxonomy.
Conservación Colombiana 23:3-48.
__________________________________________________________
Comments
from Jaramillo:
“YES. I find the argument convincing,
plumage differences, call differences and size differences to be beyond the
level that would be expected in subspecies.”
Comments
from Robbins:
“YES, for elevating Ieucura to species status based on Gary’s
detailed analyses of the morphology.”
Comments from Pacheco: “YES. The information presented seems to me
sufficient to corroborate the proposal.”
Comments from Areta: “YES. I would like to see more thorough
comparative studies on plumage variation and vocalizations, but published evidence
tips the balance towards recognition of two species.”
Comments from Remsen: ”YES.
Using a comparative approach within hummingbirds in terms of degree of
phenotypic differentiation indicates that Gary’s tabulation of differences
strongly favors species rank, and Donegan et al.’s call note differences,
although slim on sample size, are consistent with that treatment. Assigning taxon rank to allotaxa is always
problematic, but in my opinion of available evidence suggests burden-of-proof
would be on continued treatment as conspecific.
I continue to point out that DNA sequence data for neutral loci in these
allotaxa provide little if any relevant information for a taxonomic decision –
we already know they differ, assuming that the suite of phenotypic differences
is genetically based; therefore, unless genetic distance was at one of the far
extremes, how would this metric be useful?
What if there were NO differences evident in sequence data of neutral
loci? All that would say, in my opinion,
is that rates of phenotypic characters has been extremely rapid relative to
that at neutral loci; the former are potentially directly relevant to whether
these populations have crossed the threshold of “no return” in terms of
separate evolutionary trajectories, whereas the relevance of the neutral loci
to this is uncertain at best.”
“We will need a separate proposal on English names. Cory (1918) used “Pied-tailed Hillstar” for bougueri and “White-tailed Hillstar” for
leucura. Use of White-tailed for the daughter species,
as advocated by Donegan and favored by Gary, is inadvisable for the reasons
mentioned by Gary. In fact, in this case
it is unusually bad because U. bougueri
sensu lato has always been known as “White-tailed Hillstar” and using “White-tailed
Hillstar” for the split species leucura
would create obvious problems, despite the match of English and scientific
name. In addition, both have “white
tails” as evidenced for the use of that name for U. bougueri sensu. I would
rank this as by far the worst decision we’ve ever made on English names.”