of this keel, developed as described above, so that it seems proper
to record here certain field observations made by the writer that
indicate the use of this structure. As might be expected it serves in
securing and preparing certain parts of the food. In December 1917,
near Stuttgart in eastern Arkansas, during a time when the ground was
covered by a light fall of snow, flocks of Bronzed Grackles were found
feeding among small groves of a pin oak (_Quercus pagodaefolia_). The
ground under these trees was nearly bare and the birds were working
about searching for the small acorns that had fallen and were partly
concealed under leaves and low plant growth beneath the oaks. The
Grackles were tame and with a pair of binoculars it was an easy
matter to watch them at close range. The acorns were picked up, held
in the bill and pressed firmly against the keel on the palate, then
released, turned slightly by means of mandibles and tongue, and then
again gripped strongly. In this way the acorn was rotated until a line
had been impressed entirely around the shell. With a little further
manipulation the shell dropped off in two halves and the kernel was
swallowed entire without further preparation, though frequently it was
gulped down only after some effort. After watching one feeding flock
for some time I clapped my hands sharply to startle them and then
examined the ground where they had been at work. Scattered among the
leaves were many acorn shells, most of which had been cut in two in
a line transverse to the longitudinal axis. Some had fairly smooth,
clean-cut margins, while others were roughened and jagged. In searching
through the leaves I picked up one acorn still intact that had been
dropped by one of the birds, perhaps when the flock was frightened up,
in which a line had been impressed entirely around the center. In this
the impressions of the palatal keel were distinctly visible.
When attention was once attracted to this manner of feeding other
incidents were noted in which the palatal keel was brought in play.
On one occasion on the streets of Washington a Purple Grackle was
observed attempting to split open a kernel of corn dropped from some
passing dray. The bird held this grain in the slight notch near the
center of the bill and pressed it against the angular keel. The grain
proved refractory, as it snapped out several times, dropping 8 or 10
inches away, to be seized and again compressed. Watching until it had
been dropped I frightened the bird and secured the kernel of corn. On
one side four grooves impressed in the hard outer surface were visible
showing where, and with what force, the sharp keel had been applied.
Apparently the palatal ridge develops with the gradual growth of the
bill, and becomes fully functional shortly after the immature bird is
left by its parents to its own resources in securing food. It seems
to be fully grown in all by the middle of September. In many adult
specimens the ridge shows signs of heavy wear from the nearly constant
use to which it is put. In some the cutting angle was well rounded in
front from constant abrasion, while in others the anterior margin had
become irregular and broken. In one specimen the thin lower margin
of the compressed keel was entirely worn away, leaving a low rounded
projection in which the two sides of the fold by which the keel had
been formed were clearly visible, with a line of separation between
them. It was interesting to note that the palatal ridge was usually
well worn in old adults, taken in late fall or early spring, belonging
to the northern races (_Quiscalus q. quiscula_ and _Q. q. æneus_) while
little or no wear was apparent in similar specimens of the southern
form (_Q. q. aglæus_) from South Carolina and Florida. The data
available from the examination of a small number of stomachs of this
form from Florida show a preponderance of insects and fruits with very
little mast or grain, a fact of interest, but one that is not fully
substantiated as the material available is small.
Among near relatives of _Quiscalus quiscula_ a slightly developed
palatal ridge was encountered in _Megaquiscalus macrourus_, where the
projection was broad and well rounded posteriorly, and narrow in front
with the lower margin acute, forming a sharp keel. In some specimens
seen this keel was slowly reduced until it merged smoothly with the
palatal surface in front. In others the anterior margin was obtusely
declivous. The obtuse anterior cutting angle projected below the
margins of the tomia for nearly a millimeter in a few individuals, and
in these occasional specimens the resemblance was striking to those
bills of _Quiscalus_ in which the ridge was most poorly developed.
Juvenile specimens of _Megaquiscalus m. macrourus_ from Fort Clark,
Texas, that had been collected just after they had left the nest, had
the palatal ridge already well indicated though only about one-half
developed. In the slender-billed forms known as _Megaquiscalus
tenuirostris_ and _M. nicaraguensis_ the palatal keel was much as in
_M. major_ though slighter and less pronounced.
In Blackbirds belonging to the West Indian group known as
_Holoquiscalus_ a raised line was also more or less developed. In
general the growth was similar to that in _Megaquiscalus_ as the
posterior portion was broad and rounded, while anteriorly the ridge was
narrowed and the lower margin became acute. There is some variation
in the size of this anterior portion; in a few the crest is obtusely
declivous in front, approaching the condition found in _Quiscalus_,
but never with the keel produced so that it projects below the plane
subtended by the cutting edges of the tomia.
The discovery of a peculiar knoblike process on the palate of the
mexican orioles belonging to the species _Icterus gularis_ was one of
the really surprising discoveries made during a more or less cursory
examination of the palate in various species and genera of _Icteridæ_
picked out at random, and it was the finding of this structure in
an Oriole that led to a detailed examination of all of the material
available. In _Icterus gularis_ the palatal ridge is from 1.2 to 1.5
millimeters high at its anterior end (Fig. 2). The entire structure is
broad and somewhat flattened. The ventral surface is slightly rounded,
the sides slightly sloping, the sides and lower surface joining at a
sharp angle. In front the ridge is abruptly truncated at its ventral
margin where there is sometimes a slight tooth or projection. Below
this point the anterior surface slopes abruptly, and then passes over
into the roof of the palate. The ridge is about two millimeters broad,
and there is a slightly indicated raised line on the ventral surface
for three-quarters of its length behind. From this description it may
be seen that this blunt projection is entirely different from the
sharply keeled ridge found in _Quiscalus_.
[Illustration: Fig. 2. Head of _Icterus gularis yucatanensis_.
_b._ Palatal knob (about natural size.)]
Examination of other orioles shows that _Icterus gularis_ stands alone
in respect to this development as there is nothing found in other
species that approaches it save for a broad, low, rounded projection,
slight but distinct, that is found on the palate in _Icterus
xanthornus_. In _Icterus laudabilis_ and _I. prosthemelas_ there is
a very slightly raised median ridge developed on the posterior part
of the roof of the mouth. In twenty-eight other species belonging to
this genus the palate exhibits no peculiarities worthy of mention.
This structure in the bill in _Icterus gularis_ is constant in its
presence, and serves as a trenchant character distinguishing it from
other orioles, or in fact from any other members of the _Icteridæ_ that
have been available for examination. The differences pointed out above,
together with others of lesser importance, seem to be of generic value.
It is therefore proposed to recognize for this species the genus name
_Andriopsar_ Cassin.[16]
_Type._—_Ps[arocolius] gularis_ Wagler, Isis, 1829, p. 754 (type
locality, Tehuantepec, Oaxaca).
_Diagnosis._—Medium-sized _Icteridæ_ with short, heavy bill; a
prominent knoblike projection on the posterior median portion of the
palate, broad and somewhat flattened in general form, with abrupt
sides, truncated in front, sometimes with a tooth or notch at the
anterior ventral angle, about 2 millimeters broad and from 1.2 to
1.5 millimeters high in front; depth of culmen at base nearly equal
to one-half length of culmen (varying from slightly more to slightly
less); tarsus slightly longer than culmen from base; middle toe with
claw equal to two-thirds, or slightly more, of length of tarsus.
One species in which three subspecies have been described is at present
known to belong in this genus. These will stand as follows:
_Andriopsar gularis gularis_ (Wagler)
_Andriopsar gularis tamaulipensis_ (Ridgway)
_Andriopsar gularis yucatanensis_ (Berlepsch)
At present there is no information on the feeding habits of these
orioles available but it seems certain that they will show some
striking peculiarity in choice of food or in manner of securing and
handling it when the life history of the species is better known.
In conclusion I desire to give a brief summary of the condition of the
palate in other _Icteridæ_ where comment is necessary. In _Euphagus
carolinus_ and _E. cyanocephalus_ there is a slight elongate ridge
of low elevation, rounded posteriorly more acute in front, and not
projecting as far as the level of the tomia. This raised line is
slightly more pronounced in _E. carolinus_ than in _E. cyanocephalus_
in spite of the fact that the latter has a heavier, stronger bill. The
species known as _Ptiloxena atroviolacea_ has an elongate, narrow,
slightly elevated ridge on the posterior portion of the palate, rounded
behind and more or less acute in front, but with too low an elevation
to be considered a highly specialized structure. Sumichrast’s Blackbird
(_Dives dives_) has a palatal structure somewhat resembling that of the
genus _Holoquiscalus_ save that the entire ridge is shorter.
With regard to others, _Tangavius æneus_ has a slight ridge, that
becomes stronger behind, extending for two-thirds the length of the
palate. A similar ridge in _Molothrus badius_ is less developed at its
anterior end than in the preceding genus. In _Molothrus fringillarius_
(one specimen only examined) this ridge is still less in development.
In _Molothrus ater_, the cutting edges of the tomia do not extend
below the level of the palate, and there is a rounded swelling behind
the center; in _Molothrus atronitens_ only a very slight ridge is
present, and finally in _M. rufo-axillaris_ there is no peculiarity
worthy of mention. _Nesopsar nigerrimus_ shows a well marked rounded
ridge on the posterior part of the palate that merges into the anterior
surface without becoming produced as an angle. _Xanthopsar imthurmi_
shows a slightly developed posterior ridge, while in _Agelaius
phœniceus_ (including _gubernator_) there is a very faint swelling at
the posterior end of the palate, that becomes much more pronounced
in _A. tricolor_. _Agelaius thilius_ and _A. icterocephalus_ show a
faintly raised median line, that in the latter species is broadened
and rounded posteriorly. _Amblyrhamphus holosericeus_ has a long, low,
keeled median ridge, and in the three species of _Sturnella_ there is
an elongate keel, that is rounded behind and acute in front. In _Curæus
aterrimus_ the palate is on a level with the edge of the tomia, and has
a low rounded bulge on its posterior surface. _Trupialis militaris_ and
_T. falklandicus_ have a slight rounded posterior ridge, that is absent
in _T. bellicosa_ and _T. defillipi_, and finally in _Gymnomystax
melanicterus_ there is a low, narrow, keeled ridge on the posterior
part of the palate, that merges gradually into the surrounding level in
front. None of the other species seen present any marked peculiarities.
FOOTNOTE:
[16] Proc. Acad. Nat. Sci. Philadelphia, Vol. XIX, 1867, p. 49.
THE CROW IN COLORADO.
BY W. H. BERGTOLD.
A study of the technical status, and distribution of the Crow
in Colorado discloses, at once, an interesting, and a peculiar
situation.[17]
The Crow was first recorded in Colorado, so far as I am able to learn,
by Aiken (1), who reported it in this State in 1872 under the name
_Corvus americanus_; thereafter several other writers mentioned the
bird, as having been found in Colorado:—Ridgway in 1877 (2), Stephens
in 1878 (3), and Drew in 1881 and 1885 (4), all using the same name
employed by Aiken.
Ridgway (5) erected the subspecies _hesperis_ in 1887, at that time
giving its range substantially as outlined today by the A. O. U.
‘Check-List’; the validity of this subspecies was not admitted by the
A. O. U. Committee until July, 1908 (6). In his original description
of the new subspecies (_hesperis_) Ridgway did not state how many
skins he examined nor whence they came, but gave as the eastern limit
of the new subspecies “east to the Rocky Mountains,” while in his
later account (7) of _hesperis_, for which he utilized twenty-three
skins for study purposes, he carefully qualifies the eastern limit by
adding “from the Eastern portion of the arid region?” It is to be noted
that he did not definitely mention Colorado as being included within
the _hesperis_ area; in his coincidental review of the literature
possibly related to the new subspecies, however, all citations of
previous records of Colorado Crows are grouped under the literature of
subspecies _hesperis_. This probably was done because he did not have
time to sift out the records relating to the eastern slope from those
of the western slope so as to place them under the literature relating
to the individual subspecies. So far as Colorado is concerned in this
question, Ridgway probably did not take this matter up in detail
because there is not a single Crow skin in either the National Museum,
or in the Biological Survey Collections, which came from Colorado.
Most, if not all, of the writers who thereafter, directly or
indirectly, touched on the Crow’s position in Colorado, made their
diagnoses as to subspecies on regional grounds alone.
In the interval between Ridgway’s erection of subspecies _hesperis_,
and its admittance to the A. O. U. ‘Check List’ (1887 to 1908)
Morrison (8) and Drew (19) were, so far as I know, the only writers to
record the Crow in Colorado, Morrison mentioning it first, as _Corvus
frugivorus_ and the second time (9) as _Corvus americanus_, while Drew
entered his record under the latter name.
Cooke’s ‘List of the Birds of Colorado’ was published in March
1897, and in it he grouped all of the previous Colorado Crow
records, regardless of region, under the name _Corvus americanus_;
notwithstanding that Ridgway had ten years previously separated the
eastern and the western Crows, Cooke (22) logically disregarded this
action, because he followed the A. O. U. ‘Check-List’ in assembling his
‘List of Colorado Birds.’ In all the various supplements which Cooke
published to his list (the last being in ‘The Auk’ of October 1909) he
did not change his early naming of the Colorado Crows, allowing them
to stand as _Corvus americanus_ or its synonym. I am confident that he
recognized the probability of there being two subspecies in the State,
but wisely refrained from opening the question because of lack of
material available for definite determination. Furthermore I am given
to understand that there are no Crow skins in the collections of the
State Agriculture College at Fort Collins, where Cooke was located when
he compiled his ‘List,’ which fact would lend support to the idea that
his omission to mention the possibility of both the Eastern and the
Western Crows being found in Colorado was due to his unwillingness to
pass judgment on a question without the support of definite material or
data.
In his ‘The Present Status of the Colorado Check-List of Birds’
(10), Cooke again was silent as to the presence of subspecies
_brachyrhynchos_ or of _hesperis_ or of both within the confines
of Colorado, though at least three writers (11), (12), (13), had
previously mentioned the Colorado Crow in their respective papers,
as being _hesperis_; Cooke was too careful and experienced an
ornithologist to have overlooked these records and I am sure his
silence was judiciously intentional and premeditated.
It thus appears that between 1887 and 1912 the Crows of Colorado had
been recorded by some observers, so far as subspecies were concerned,
as _brachyrhynchos_, and by others as _hesperis_, but so far as I
know and am able to learn, none suggested or recorded that these two
subspecies coexisted in the State.
I am inclined to believe that Sclater’s (13) designating the Colorado
Crow as _hesperis_ was made on purely geographical grounds, because the
collection then at his command, (that at Colorado College, Colorado
Springs) contains but one crow skin, a partial albino, which proves to
be, under examination, subspecies _brachyrhynchos_. E. R. Warren allows
me to state that he has no Crow skins in his collection, and that he
made his subspecific diagnosis of _hesperis_, for the birds seen near
Bulah, Colorado, on geographic grounds only. In later records Warren
(14) wisely refrains from trying to decide as to the subspecies, when
listing the Crows seen in Montrose County, and in northern Colorado,
mentioning the birds merely as _Corvus brachyrhynchos_, and Henderson
(18) did likewise in his Boulder County List.
I do not know on what grounds Hersey and Rockwell (11) made their
statement that subspecies _hesperis_ was to be found on the eastern
slope of the Rockies.
Since Cooke’s last word on our Colorado avifauna, two more writers have
given the Crow as a species found within the State, each listing it as
_hesperis_, and both records are for the Atlantic slope. I am permitted
by F. C. Lincoln (15), the first of these two writers, to say that he
did not take any Crows in Yuma County, and that he made his subspecific
diagnosis on geographic grounds alone. It is now, unhappily, impossible
to determine what led Betts (16), the second of these two writers, to
conclude that the Boulder County Crow was _hesperis_. I do not know
whether he collected specimens in Boulder County; but Junius Henderson
informs me that Betts sent crow _eggs_ to the National Museum. But he
probably did not send skins for, as has already been said, there is
not a Crow skin in the National Museum collection, from Colorado. The
internal evidence (18) points to the belief that Betts too, recorded
the Boulder County Crow as _hesperis_, on geographic grounds alone.
Crows seen by Warren (17 and 20) in other parts of the State are given
as subspecies _brachyrhynchos_, but again named on regional grounds
only.
From the foregoing it appears that the Crows of Colorado were
listed, principally as _Corvus americanus_ up to the acceptance of
subspecies _hesperis_ in the A. O. U. ‘Check-List,’ and since then
variously listed as _Corvus brachyrhynchos_, _Corvus brachyrhynchos
brachyrhynchos_, or _Corvus brachyrhynchos hesperis_, but, to repeat,
so far as I can learn, in no instance have any of the last two kinds
of records been made on skin determinations. This statement is based
on a study of the published records, and on a considerable relevant
correspondence with my associates throughout the State; if I err the
statement is open and subject to correction.
The western third of Colorado lies on the Pacific slope, and the
eastern two-thirds on the Atlantic and on both of these slopes the
Crow has been detected, and variously recorded as to subspecies. The
A. O. U. ‘Check-List’ does not speak of _hesperis_ actually extending
eastward to the Rocky Mountains, but Mr. Ridgway, in a recent
communication said to me “I feel quite sure that any Crow found west
of the Divide in Colorado would be _C. b. hesperis_. On the other
hand, those found on the eastern side would almost certainly be _C. b.
brachyrhynchos_.”
I am fortunate, not only in having material in my own collection, which
substantiates Ridgway’s belief, but in also having had access, thanks
to my obliging friends, to specimens and data which also show that his
belief is essentially correct.
I have been able to study fourteen Crow skins from the eastern side
of the Rockies in Colorado, six males and eight females; of the males
three are typical _brachyrhynchos_, two are clearly _hesperis_, and
the last is mainly _brachyrhynchos_, but with weaker bill and tarsus
than is ordinarily found with that subspecies. It is of interest to
note that this last specimen was taken in Weld County close to the
locality whence came the two previously mentioned _hesperis_ skins. It
is much more difficult to allocate the females of this group of skins;
however four are more typically subspecies _brachyrhynchos_ than is
another female in my collection which I collected many years ago in
New York, and another female is also of this subspecies, but with a
weak bill, while the remaining three are too near the dividing line
to be definitely located as to subspecies, all showing characters of
one or of the other of the two forms under study, in varying degrees of
intensity.
I have been able to study but one Crow skin from the western slope in
Colorado, to-wit, a skin in my collection, which was taken at Ignacio,
Colorado, in October, 1917, by my friend and colleague, Dr. Walter
L. Mattick; fortunately it is the skin of a male, and is typical
_hesperis_.
We are now on firm ground; those skins from the eastern slope which are
most likely to be characteristic of a given subspecies, to-wit, males,
show that both _brachyrhynchos_ and _hesperis_ are to be found on
that slope, and the Ignacio skin proves that _hesperis_ occurs on the
western slope.
Hence one can say now that both _Corvus brachyrhynchos brachyrhynchos_
and _Corvus brachyrhynchos hesperis_ are to be included in future lists
of Colorado birds.
The common Crow is normally a bird of moderately large and fairly dense
timber, a growth found in Colorado only along the larger streams and
in the mountains; if one plot the Crow stations of Colorado on a map,
it at once becomes patent that most, if not all, of these stations are
to be found along the courses and headwaters of the State’s larger
streams. This fact seems to lend color and support to the idea that
subspecies _brachyrhynchos_ probably penetrated Colorado from the
east by following the larger streams towards the mountains, for it is
along these rivers that one finds trees to the Crow’s liking, and too,
Crows are increasingly more common as one travels eastward along these
watercourses. It would seem reasonable to believe that along similar
natural “crow” highways _hesperis_ would find its way eastward from the
Pacific side into Colorado.
The smaller size, alone, of _hesperis_, often makes it distinguishable
in the field, a fact which first came to my attention while in the
“hills” on the Gila River in New Mexico, in 1906. During the same year
I saw a considerable flock of Crows immediately south of Antonito,
Colorado; I was then again impressed by the smaller size of these
southern Colorado and New Mexico Crows. I now believe these Antonito
Crows were subspecies _hesperis_; Antonito is on (or very close) to
the Rio Grande River, which drains part of the Atlantic-Gulf of
Mexico watershed, part of which watershed forms the western portion
of Texas, an area included in the present known range of _hesperis_.
It does not seem unreasonable to believe that _hesperis_ works its
way from western Texas, up along the Rio Grande, finally reaching the
vicinity of Antonito, and also the San Luis Valley. In support of this
latter view I am permitted to say that Mrs. Jesse Stevenson of Monte
Vista, Colorado, recently saw a Crow for the first time in twenty-five
years in this valley, and was at once impressed with its small size as
compared with those she formerly studied in the East.
As mentioned above, it is clear that _hesperis_ occurs on both sides
of the Rocky Mountains in Colorado. Now one must ask if subspecies
_brachyrhynchos_ occurs on the western slope as well as on the eastern
slope.