Division of Birds, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, USA
On 18 November 1894, Frank B. Armstrong col-
lected an unusual woodcreeper near Altamira, Ta-
maulipas, Mexico. This specimen was sent with others
for identification to the United States National Mu-
seum during the winter of 1894-1895 (Richmond 1896),
and was eventually described as a new species, Den-
drornis [=Xiphorhynchus] striatigularis, based on its dis-
tinct plumage (Richmond 1899; USNM 135157; 18 No-
vember 1894, Alta Mira, Tamaulipas; female).
Despite the fact that this region can now be con-
sidered rather well collected, this taxon continues to
be represented only by the unique type, suggesting
that it is: (A) a very rare and now probably extinct
(but valid) species; (B) the hybrid product of two valid
species; or (C) an aberrant individual of a valid species
(cf. Graves 1990). Xiphorhynchus striatigularis has re-
ceived little attention since its description. Some au-
thors simply list the taxon with little or no discussion
(e.g. Ridgway 1911, Cory and Hellmayr 1925, Peters
1951, Blake 1953, Miller et al. 1957), while others
either exclude it altogether (Edwards 1972, 1989, Pe-
terson and Chalif 1973), or consider it an aberrant X.
fiavigaster (AOU 1983, Sibley and Monroe 1990). No
evidence has been presented to favor any of the above
hypotheses or treatments. My examination attempts
to resolve the nature of the unique type.
Methods.--It is assumed that woodcreeper species
occupying the region around Altamira (lower ele-
vations of northeastern Mexico) will be morpholog-
ically distinct and that, if valid, X. striatigularis would
exhibit a size and shape different from that of same-
sex individuals of other species in this area. This as-
sumption would not be robust for all taxonomic groups
or geographic areas. It seems robust here, however,
because the center of species diversity in Xiphorhyn-
chus occurs in South America (Sibley and Monroe
1990), and no cryptic species of Dendrocolaptidae are
known in northern Middle America. Armstrong's
handwriting is apparent on the original label of X.
striatigularis, suggesting that the data there are correct
(see Oberholser 1974:8).
The hypothesis of hybrid origin does not make
morphological predictions. Although it would seem
likely, intermediacy in body form might not occur,
even if the individual is an F hybrid (G. Graves pets.
comm.). If X. striatigularis is an aberrant individual of
a valid species and the aberrancy is restricted to pig-
ment deposition in the plumage (the latter is the basis
for this taxon; Richmond 1899), its external morpho-
metrics should match one of the species occurring in
the region.
Diagnosis of this unique type is aided by the fact
that southern Tamaulipas is near the northerly limits
of the Dendrocolaptidae, and only two species (of two
genera) currently occur in the region where the unique
type was taken: X. fiavigaster and Lepidocolaptes affinis.
A third species, L. souleyetii, reaches central Veracruz,
and is ,ncluded in this study. Three other dendro-
colaptids occur in northern Mexico, but were not in-
cluded in this analysis. Xiphocolaptes promeropirhyn-
chus and Sittasomus griseicapillus were excluded be-
cause their sizes, ranges, and plumage characteristics
preclude their involvement in the latter two hypoth-
eses presented above. I excluded Xiphorhynchus ery-
thropygius because of its range, its montane elevational
preference in Mexico, and a lack of evidence in the
plumage of X. striatigularis of a contribution from er-
ythropygius.
The unique type of X. striatigularis is in basic plum-
age, and its distinct plumage characteristics are not
TABIE 1. Univariate morphometric characteristics of female Xiphorhynchus striatigularis, X. fiavigaster, Lepi-
docolaptes affinis, and L. souleyetti (œ ñ SD, with range in parentheses; measurements in mm).
X. stria- X. fiavigaster L. affinis L. souleyetii
tigularis (n = 46) (n = 26) (n = 30)
Wing chord 104.5 107.6 ñ 3.49 (98.9-114.9) 106.0 ñ 2.59 (101.7-111.0) 92.5 ñ 2.66 (86.9-97.5)
Tail length 80.5 85.9 ñ 5.95 (73.0-98.0) 91.0 ñ 3.86 (82.1-97.4) 78.5 ñ 3.66 (69.7-85.0)
Tarsometatarsus
length 22.6 21.6 ñ 0.55 (20.6-22.8) 19.5 ñ 0.58 (18.3-20.8) 18.3 ñ 0.64 (16.9-19.6)
Bill length 30.1 30.8 ñ 1.42 (27.2-34.0) 22.7 ñ 1.00 (20.2-24.5) 22.2 ñ 0.89 (19.8-23.8)
Bill height 8.0 7.7 ñ 0.54 (6.7-9.1) 5.3 ñ 0.33 (4.8-6.1) 5.4 ñ 0.28 (4.9-6.0)
Bill width 6.4 6.2 ñ 0.81 (4.1-7.1) 4.4 ñ 0.60 (3.6-5.7) 4.9 ñ 0.74 (3.8-6.2)
Length of
primary 8 80.0 83.8 ñ 2.95 (77.0-92.0) 84.1 ñ 2.30 (79.0-88.0) 72.3 ñ 2.61 (67.0-77.5)
Length of
primary 9 75.0 79.5 ñ 2.98 (72.0-87.5) 80.5 ñ 2.44 (76.0-85.0) 69.6 ñ 2.38 (63.5-73.5)
Length of
primary 10 59.5 62.5 ñ 3.18 (50.5-69.0) 62.8 ñ 2.87 (55.0-67.5) 54.2 + 1.74 (50.0-58.5)
caused by the retention of juvenal plumage charac-
teristics; the juvenal plumage of woodcreepers tends
to closely match the plumage of the adult (pers. obs.).
For morphometric study, I examined a broad sam-
pie of female specimens of X. fiavigaster, L. affinis, and
L. souleyetii from the Atlantic slope of northern Mid-
dle America. Postulations of sedentariness in Neo-
tropical species can be invalid (see Winker et al. 1992:
714, Winker 1995). Also, in studies like this, it is useful
to have a good sample of the morphological variation
occurring within the taxa examined, and large spec-
imen samples from restricted geographic areas in the
Neotropics are rare.
Measurements to the nearest 0.1 mm were taken of
the lengths of wing chord, tail, tarsometatarsus, and
bill, as well as of bill height and bill width, using
vernier calipers (following Baldwin et al. 1931). Also,
the lengths of primaries 8, 9, and 10 were measured
to the nearest 0.5 mm (following Jenni and Winkler
1989). Specimens with missing values were not in-
cluded. Patterns among the measured characters were
examined using principal component analysis. The
first two principal components were extracted from
the correlation matrix of log-transformed mensural
data (Tables 1 and 2). Unstandardized principal com-
ponent scores were generated for each individual.
Results and discussion.--Except for X. striatigularis,
the taxa examined occupy discrete regions of mor-
phological space (Fig. 1). Xiphorhynchus striatigularis is
morphometrically inseparable from the common X.
fiavigaster (Table 1, Fig. 1). Given the propensity for
the species of Dendrocolaptidae in the lowlands of
northeastern Mexico to be morphologically distinct,
and under the assumptions upon which the study was
based, X. striatigularis is not a valid species. Its taxo-
nomic status, therefore, is either nomen hybridurn or
nomen monstrositatum (Lincoln et al. 1982).
It is more difficult to distinguish between the two
remaining hypotheses of origin: hybridization or ab-
errancy. The position of X. striatigularis in morpho-
logical space (Fig. 1) does not approach either of the
Lepidocolaptes species. It does, however, closely match
other female X. fiavigaster. Morphometrically, the hy-
pothesis of aberrancy is supported. The morphomet-
ric data are neutral with respect to the hypothesis of
hybrid origin (see Methods). However, hybrid origin
seems highly unlikely from several perspectives. First,
hybrids among the woodcreepers and their nearest
suboscine relatives are decidedly rare; only one has
been described (Graves 1992). Also, in this case, hy-
bridization would most likely be intergeneric, a less
common source of hybrids than congeneric crossings
(Gray 1958).
The geographically most probable second parent
taxon for a hybrid woodcreeper at Altamira would be
L. affinis, which occupies higher elevations than the
other two species considered here (> 550 m; Miller et
al. 1957). Armstrong did not obtain L. affinis at Alta-
mira (Richmond 1896, Phillips 1911; extant specimen
record), which is not surprising, given that it does
not usually occur in the lowlands. Only one species
TABLE 2. Eigenvectors and eigenvalues for first two
principal components based on correlation matrix
of log-transformed measurements of female wood-
creepers of four taxa.
Character PC 1 PC2
Wing chord 0.422 - 0.093
Tail length 0.282 -0.325
Tarsometatarsus length 0.384 0.225
Bill length 0.352 0.318
Bill height 0.324 0.371
Bill width 0.170 0.528
Length of primary 8 0.365 -0.319
Length of primary 9 0.230 -0.437
Length of primary 10 0.389 -0.156
Eigenvalue 5.097 2.744
Percent of total
variance explained 56.6 30.5
Fig. 1.
creepers.
4
Xiphorhynchus striatigularis
.. Xiphorhynchus fiavigaster
' n .++ + +
Lepidocolaptes souleyetii ++
. ;'
Lepidocolaptes affinis
-+ +
+ +
+ +
-4 [ ] [ [ '
-6 -4 -2 0 2 4 6
PC1
Bivariate plot of individual scores on first two principal components of 101 adult female wood-
of woodcreeper, X. fiavigaster, is known to occur at
Altamira. Topographic maps show that the nearest
locality with an elevation suitable for L. affinis occurs
about 60 km from Altamira (in the Sierra de Tamau-
lipas, where the species is known to occur; Miller et
al. 1957). It is approximately 430 km from Altamira
to the range of L. souleyetii (using ranges from Miller
et al. 1957). Banding studies in southern Veracruz
suggest that X. fiavigaster is sedentary there (unpubl.
data). Morphometric studies of L. affinis and long-term
observations in southern Veracruz (unpubl. data) sug-
gest that this species is also sedentary in eastern Mex-
ico.
Given the inadequacy of morphometrics to resolve
the two remaining hypotheses of origin, plumage ex-
amination is useful. Unlike any other member of the
family in northern Middle America, the throat of X.
striatigularis shows heavy streaking (Fig. 2). These
streaks are caused by feathers having a black margin
around a buff center, and the nature of these feathers
is also unusual, in that the black/buff interface is
ragged, rather than smooth. This ragged interface,
which characterizes streaked ventral feathers to the
abdomen, is not found in other members of this ge-
nus, nor in the two Lepidocolaptes species examined.
It appears that genetic control of black pigment de-
position on the venter went awry, causing aberrancy
in how black pigment was deposited in ventral feath-
ers, as well as an exaggeration in the occurrence of
these feathers anteriorly. The density of pigment also
appears greater than in other woodcreepers (X. fia-
vigaster) taken at the same time and locality. Except
for these characteristics, restricted largely to the an-
terior venter, all aspects of the plumage of X. striati-
gularis can be found represented in female X. fiavi-
gaster from eastern Mexico (Fig. 2).
Although the hypothesis of hybrid origin cannot
be rejected, it is very improbable for several reasons:
(1) rarity of hybrids in woodcreepers and their rela-
tives (Graves 1992); (2) only one species of wood-
creeper occurs at the type locality; and (3) the type,
clearly different from other woodcreepers, neverthe-
less shows only localized plumage differences that
lack intermediacy between two possible parent taxa.
The data are more consistent with origin through
aberrancy. Unlike hybridization, genetic mutations
occur regularly and, although odd melanistic plum-
ages seem uncommon, aberrancies in feather pigment
deposition are relatively frequent (e.g. partial albi-
nism or leucism). Based upon this evidence, I con-
clude that X. striatigularis is an aberrant X. fiavigaster.
Markedly aberrant individuals are monstrosities, and
a name based upon such an individual has the taxo-
nomic standing of nomen monstrositatum (Lincoln et
al. 1982).
The day after Armstrong collected the unique X.
striatigularis, he collected another bird in the same area
that Wetmore (1942) later designated as the type of
X. fiavigaster saltuarius (USNM 135161; Alta Mira, Ta-
maulipas, 19 November 1894; male). Xiphorhynchus
Fig. 2. Type of Xiphorhynchus striatigularis.
striatigularis should henceforth be considered a syn-
onym of X. fiavigaster saltuarius.
Acknowledgments.--1 dedicate this paper to Fred-
erick Loetscher, Jr., whose kindness stimulated this
examination. 1 thank Gary Graves, Ralph Browning,
John Rappole, Scott Lanyon, and an anonymous re-
viewer for thoughtful discussion and comments, Carl
Hansen for photographs, and the following for loans
of comparative material: Scott Lanyon and David Wil-
lard (Field Museum of Natural History), Raymond
Paynter, Jr. (Museum of Comparative Zoology, Har-
vard University), Mary LeCroy (American Museum
of Natural History), J. Van Remsen, Jr. (Louisiana
State University Museum of Natural Science), John
Hafner and James Northern (Moore Laboratory of
Zoology, Occidental College), and Robert Payne (Uni-
versity of Michigan Museum of Zoology).
LITERATURE CTED
AMERICAN ORNITHOLOGISTS' UNION. 1983. Check-list
of North American birds, 6th ed. Am. Ornithol.
Union, Washington, D.C.
B^LDWIN, S. P., H. C. OBERHOLSER, AND L. G. WORL3/4.
1931. Measurements of birds. Sci. Publ. Cleve-
land Mus. Nat. Hist. 2:1-165.
BLAKE, E.R. 1953. Birds of Mexico. Univ. Chicago
Press, Chicago.
CORY, C. g., AND C. E. HELLMAYR. 1925. Catalogue
of birds of the Americas, part 4. Field Mus. Nat.
Hist. Zool. Ser. 13(4):1-390.
EDWARDS, E. P. 1972. A field guide to the birds of
Mexico. Privately published, Sweet Briar, Virgin-
ia.
EDWARDS, E. P. 1989. A field guide to the birds of
Mexico, 2nd ed. Privately published, Sweet Briar,
Virginia.
G,AVES, G.R. 1990. Systematics of the "green-throat-
ed sunangles" (Aves: Trochilidae): Valid taxa or
hybrids? Proc. Biol. Soc. Wash. 103:6-25.
GIAVES, G. R. 1992. Diagnosis of a hybrid antbird
(Phlegopsis nigromaculata x Phlegopsis erythroptera )
and the rarity of hybridization among subos-
cines. Proc. Biol. Soc. Wash. 105:834-840.
GIA3/4, A. P. 1958. Bird hybrids. Commonwealth Ag-
ricultural Bureau, Bucks, England.
JENNI, L., AND R. WINKLER. 1989. The feather-length
of small passerines: A measurement for wing-
length in live birds and museum skins. Bird Study
36:1-15.
LNCOLN, R. J., G. A. BOXSHALL, AND P. F. CEARC. 1982.
A dictionary of ecology, evolution and system-
atics. Cambridge Univ. Press, Cambridge, Great
Britain.
MILLER, A. H., H. FPaEDMANN, L. GPaSCOM, AND R. T.
MOOKE. 1957. Distributional check-list of the
birds of Mexico, part 2. Pacific Coast Avifauna
33.
OBERHOI, SER, H.C. 1974. The bird life of Texas. Univ.
Texas Press, Austin.
PETERS, J. L. 1951. Check-list of birds of the world,
vol. 7. Museum of Comparative Zoology, Har-
vard Univ., Cambridge, Massachusetts.
PETERSON, R. T., AND E. L. CHALIF. 1973. A field guide
to Mexican birds. Houghton Mifflin Co., Boston,
Massachusetts.
PHILLIPS, J. C. 1911. A year's collecting in the state
of Tamaulipas, Mexico. Auk 28:67-89.
RICHMOND, C.W. 1896. Partial list of birds collected
at Alta Mira, Mexico, by Mr. Frank B. Armstrong.
Proc. U.S. Natl. Mus. 18:627-632.
PaCHMOND, C. W. 1899 [1900]. Description of a new
bird of the genus Dendrornis. Proc. U.S. Natl. Mus.
22:317-318.
RIDGW^Y, R. 1911. The birds of North and Middle
America, part 5. Bull. U.S. Natl. Mus. 50(5):1-859.
SIBLEY, C. G., AND B. L. MONROE, JR. 1990. Distri-
bution and taxonomy of birds of the world. Yale
Univ. Press, New Haven, Connecticut.
WETMORE, A. 1942. New forms of birds from Mexico
and Colombia. Auk 59:265-268.
WINKER, K. 1995. Neotropical stopover sites and
Middle American migrations: The view from
southern Mxico. Pages 150-163 in Conservation
of migratory birds in Mxico (M. Wilson and S.
Sader, Eds). Maine Agric. For. Exp. Stat. Misc.
Tech. Publ. 727.
WINKER, K., R. J. OEHENSCHAGER, M. A. RAMOS, R.
M. ZINK, J. H. RAPPOLE, AND D. W. WARNER. 1992.
Bird distribution and abundance records for the
Sierra de los Tuxtlas, Veracruz, M6xico. Wilson
Bull. 104:699-718.
Received 13 February 1995, accepted 18 March 1995.