'9^ o '">,

HARVARD UNIVERSITY

Library of the

Museum of

Comparative Zoology

UNIVERSITY OF KANSAS

MUSEUM OF NATURAL HISTORY

PUBLICATIONS

The University of Kansas Publications, Museum of Natural History,
beginning with volume 1 in 1946, was discontinued with volume 20 in
1971. Shorter research papers formerly published in the above series are
now published as The University of Kansas Museum of Natural History
Occasional Papers. The University of Kansas Museum of Natural
History Miscellaneous Publications began with number 1 in 1946.
Longer research papers are published in that series. Monographs of the
Museum of Natural History were initiated in 1970. Authors should
contact the editor regarding style and submission procedures before
manuscript submission. All manuscripts are subjected to critical review
by intra- and extramural specialists; final acceptance is at the discretion
of the Director.

This publication is printed on acid-free paper. Occasional Papers and
Miscellaneous Publications are typeset using Microsoft " Word and Aldus
PageMaker" on a Macintosh computer. - Museum of Natural History,
The University of Kansas, Lawrence.

Institutional libraries interested in exchanging publications may obtain
the Occasional Papers and Miscellaneous Publications by addressing the
Exchange Librarian, The University of Kansas Library, Lawrence,
Kansas 66045-2800. USA. Individuals may purchase separate numbers
from the Office of Publications, Museum of Natural History, The
University of Kansas, Lawrence, Kansas 66045-2454, USA.

Editor: Linda Trueb

Managing Editor: Joseph T. Collins

Assistant Managing Editor: Kate Shaw

Printed by

University of Kansas Printing Service

Lawrence. Kansas

OCCASIONAL PAPERS

MCZ
LIBRARY

FEB 1 1 1993

of the HAR\/,5.Pn

MUSEUM OF NATURAL HISTORY
The University of Kansas
Lawrence, Kansas

number 154. pages 1-37 4 february 1993

Avifauna of Three Holocene Cave Deposits

IN Southern Chile "^

Philip S. Humphrey, Jaime E. Pefaur
AND Pamela C. Rasmussen

Museum of Natural History and Department of Systematics and Ecology. The

University of Kansas. Lawrence. Kansas 66045-2454. USA (P.S.H.):
Departamento de Biologi'a. Facultad de Ciencias. Universidad de los Andes,

Merida. Venezuela (J.E.P.);

and Room 336 NHB MRC 114. National Museum of Natural History.

Smithsonian Institution. Washington. D. C. 20560. USA (RC.R.I

ABSTRACT A collection of 139 bird bones and 50 feathers from FelLs Cave,
Mylodon Cave, and Alero del Diablo, southern Magallanes Province. Chile was
found to represent a minimum of 84 individuals of at least 25 species of 10 orders
of birds. These included a rhea. a grebe, a cormorant, an ibis, at least six species of
waterfowl, three hawks, three caracaras, two rallids. a seedsnipe. two owls, and four
passerines. Juvenile birds of two species were present at Fell's Cave, and the
presence of several migratory breeding species and one wintering species confirms
the use of this cave both during warmer and colder seasons. No such inferences
were possible for Mylodon Cave and Alero del Diablo, as all species present there
are largely nonmigratory. The abundance of waterbirds at Fell's Cave confirms that
aquatic habitats were present nearby more or less continuously from about 1 1,000
ybp to the present, and three moorland/steppe specialists but no obligate forest
species were deposited there during the same period. At Mylodon Cave, four forest
specialists and no steppe species were present in dated layers. All birds found at
Alero del Diablo are habitat generalists. Compared to mammals, birds were not a
major part of the food economy of the humans that inhabited Fell's Cave. Mylodon
Cave, and Alero del Diablo. In contrast, birds were an important constituent of
archaeological sites on the Beagle Channel.

^&'

Key words: Magallanes, Chile. Archeoavifauna. Zooarcheology

© Museum of Natural Hislory. The Universilv of Kansas. Lawrence. ISSN:009 1-7958

UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

RESUMEN Una coleccion de 139 huesos y 50 plumas de aves conteniendo tin
mi'niino de 84 individuos pertenecientes al menos a 25 especies de 10 ordenes de
aves es estudiada de la Cueva de Fell, Cueva del Milodon y Alero del Diablo, en la
Provincia de Magallanes, Chile. La misma contiene el iiandii (Pterocnemia
pennata), el blanquillo (Podiceps cf. P. occipitalis), una especie de cormoran
{Phalacrocorci.x atriceps). la bandurria (Theristicus melcinopis). dos especies de
gansos (Chloephaga picta y C. ruhicllccps), una especie de quetru (Tachyeres
patachonicus), al menos tres otras especies de patos {Anas sihilafrix. A. cf. A.
i>eorgica,A. cf. A. fiavirostris). el aguila {Geranoaetus melanoleucits), dos especies
de aguiluchos (Buteo polyosoma y B. ventralis). dos especies de tiuques
{Phalcohocniis aUx)giilaiis y Milva(;o chii)Hini;o). el traro (Polyhorns plancus). un
especie de tagua (Eiilica armillala), una especie indeterminada de la familia
Rallidae, una especie de perdicita (Attagis malouinus), la lechuza (Tyto alha). el
concon {Strix riifipes), el rayadito {Aphrastiira spiuicaiula), el zorzal (Turdiis
falcklanclii), una especie de cometocino (Phrygiliis patagoniciis) y una especie de
jilguero (Cardiielis harhata). Juveniles de un pato y un aguilucho, varias especies
migratorias nidificando (mayonnente Anserifomies) y una perdicita invemando
estan presentes en la Cueva de Fell. Este hecho continna que la cueva fue usada
durante las estaciones calidas y frias. Esta afirmacion no puede extenderse a la
Cueva del Milodon y Alero del Diablo debido a que todos las especies presentes en
estas son mas o menos sedentarias. La abundancia de restos de aves acuaticas en la
Cueva de Fell contirma la existencia de ambientes acuaticas en los alrededores
desde aproximadamente 1 1 .000 aiios atras hasta la actualidad. En este cueva fueron
hallados restos de tres especialistas de praderas magallanicas y estepas, pero
ninguna de ellas correspondiente a las especialistas del bosque. Restos de cuatro
especies de aves restringidas a bosques se encontraron en los niveles datados con
radiocarbono en la Cueva del Milodon; no fueron encontrados aquellas especies
restringidas a las estepas o praderas. Todas las aves encontradas en Alero del Diablo
son generalistas ambientales. Comparadas con los mami'feros, las aves no
representaban un elemento importante en la econoniia de los habitantes de la Cueva
de Fell, Cueva del Milodon y Alero del Diablo; este hecho contrasta con los sitios
arqueologicos de los indios del Canal Beagle, que contienen un alto porcentaje de
aves.

Paiabras chives: Magallanes, Chile, Arqueoavifauna, Zooarqueologfa

Anthropogenic deposits of bones and other remains in caves may pro-
vide documentation of the prehistoric existence, abundance, and breeding
of animal species in an area, as well as abundant information concerning
the environments and economic strategies of ancient peoples. Archaeologi-
cal investigations of caves in southern South America have a long history
(Hauthal, 1899; Lehmann-Nitsche, 1899; Roth, 1899; Lothrop, 1928; Bird,
1938; Emperaire and Laming, 1954; Salmi, 1955; Emperaire et al., 1963;
Laming-Emperaire. 1972; Wellman. 1972; Ortiz-Troncoso. 1973; Massone,
1981; Sanguinetti de Bormida and Borrero, 1983; Figuerero Torres. 1986;
Cardich et al., 1987; Nami, 1987), and Patagonian sites have been particu-

AVIFAUNA OF CHILEAN CAVES 3

larly informative (Dillehay, 1984, 1989). Remains of mammals excavated
from caves in southern continental Chile have been well documented
(Emperaire et al.. 1963; Saxon, 1976; Rau and Yanez, 1980; Simonetti and
Rau, 1989), but only a preliminary listing of the birds from a single site
(Fell's Cave) has been published (Saxon. 1979), and only a few other
studies have dealt with Holocene birds from southernmost South America
(Tambussi and Tonni, 1984, 1985; Lefevre, 1989; Siegel-Causey and
Lefevre, 1989; Salemme, 1990; Rasmussen and Humphrey, 1991).

Bird remains were excavated from three caves in southern Patagonia
(Fell's Cave, Mylodon Cave, and Alero del Diablo) from 1970-1976;
radiocarbon dates are known for deposits from Fell's and Mylodon caves.
These dated remains allow for examination of changes that may have
occurred in the local avifauna in the past 12,000 years and the possible
correlation of changes in species composition or abundance to climatic and
habitat fluctuation in southern Patagonia. In this paper we document the
avifaunas preserved in each cave as a basis for making inferences on
seasonality of cave use, past abundances of certain species, and the rela-
tionship of species composition and abundance to climate and habitat.

MATERIALS AND METHODS

Study Areas

Fell's Cave. — Fell's Cave (50 4'S, 69 7 'W) is on the property of Estancia
Brazo Norte, Rio Chico, Chile. 75 km SW of the Monte Aymond frontier
post on the road to Rio Gallegos, Argentina (Fig. 1 ). The site provides the
oldest (11.000 ybp) evidence of human occupation in the region (Saxon,
1976). The Rio Chico at one time undercut the sandstone of the canyon
wall, creating a cave 9 m deep, 1 1.5 m wide, and about 3.5 m high (Bird,
1938; Table 1 ). A detailed account of the history, excavation, stratigraphy,
human occupation, and mammalian fauna of Fell's Cave is presented in
Emperaire et al. (1963). Their list of vertebrate remains includes
Neomyloclon, Ouohippidium, Lama, Hippocamelus, Conepatus, Ctenomys,
and Oryzomys, remains of recent sheep, and 26 unidentified bird bones.

Bird remains were excavated from Fell's Cave (Fig. 1) in 1970 by J.
Bird, and were provided for our study by E. Saxon. Results of excavations
of Fell's Cave undertaken in 1970 and 1975, including a summary of the
distribution of mammalian remains found in the cave by radiocarbon date,
were presented by Saxon (1976). A list of birds found at Fell's Cave was
presented in Saxon (1979); discrepancies between Saxon's list and that in
this paper are due to the preliminary nature of the former. The stratigraphic
sequence and associated radiocarbon dates for Fell's Cave (Saxon, 1976; in
litt.) are shown in Table 2.

Mylodon Cave. — Mylodon Cave or Cueva del Milodon (51 35'S,

UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 1.^4

ATLANTIC
OCEAN

PACIFIC
OCEAN

Kilometers

Fig. 1 . Map of southern Chile and Argentina, showing the locations of the
three caves in this study and of other archaeological sites from which birds are
known. Abbreviations: E, Englefield; BC, Bahia Colorada: PB. Punta Baja; BB,
Bahia Buena; PSA, Punta Santa Ana: PM. Punta Maria; SP, San Pablo (Lefevre,
1989): LP, Lancha Packewaia (Rasmussen et al.. MS); T, Tiinel (Humphrey and
Rasmussen, unpubl. data). Los Toldos, in Santa Cruz Province, Argentina (Tambussi
and Tonni. 1985), is north of the area covered by this map.

72 38'W) is located 20 km NW of Puerto Natales. about 6 km NE of Puerto
Consuelo, Seno de Ultima Esperanza, Chile (Fig. 1 ). The cave is approxi-
mately 200 m above sea level in the south slope of a 600 m hill (Hauthal,
1899) and consists of a single cavity 170 in wide. 270 m long and approxi-
mately 40 m high (Emperaire and Laining. 1954; Table 1). The deposits in
the cave are well known because of the presence of remains of extinct
ground sloth (Neomylodon) and horse (Onoluppidiiim). Soine of the cave
deposits contain feathers and owl pellets; these and a few avian bones were
excavated by E. Saxon in 1975 and 1976. Lehmann-Nitsche (1899) noted

AVIFAUNA OF CHILEAN CAVES 5

that remains of owls were found in Mylodon Cave and Roth (1899), in his
discussion of animal remains encountered in Mylodon Cave, mentioned
that feathers and a rhea tarsometatarsus were collected in the cave. A
detailed account of the excavation, stratigraphy, and history of Mylodon
Cave is presented in Emperaire and Laming (1954). The stratigraphic
sequence and associated radiocarbon dates for Mylodon Cave (Saxon,
1976. in litt.) are shown in Table 3.

Alero del Diablo. — Alero del Diablo is a small rock shelter about 2400
m SE of Mylodon Cave (Fig. 1 ). It faces west under a rocky overhang in a
promontory that is part of the same conglomerate bluff in which Mylodon
Cave and the rock shelter Dos Herraduras are located. Alero del Diablo is
1 1 m wide and 5 m deep; the height of the roof at the entrance is about 3 m
(Borrero et al.. 1976; Table 1 ). Remains of animals excavated from Alero
del Diablo include guanaco (Lama guanicoe), tuco-tuco {Ctenomys
mageUanicus). fox [Dusicyon sp.). seal {Arctocephaliis australis), mussels
(Mytilus chilensis), and unidentified fish (Borrero et al., 1976). Bird re-
mains were excavated by E. Saxon in 1975-1976. To our knowledge,
radiocarbon dates have not been determined for remains excavated from
Alero del Diablo.

Specimens

One hundred and thirty-nine bird bones and 15 packets of feathers
comprise the total avian material available to us for study. Nearly all bones
and packets of specimens were numbered individually with the same
numbering system as that used in the various excavations, such that speci-
mens are automatically identified to cave, trench, and layer (two bones
lacked numbers, probably owing to breakage). For specimens from
Mylodon Cave, apostrophes around a number or letter in this paper are
equivalent to a circle in the original excavation number. Upon completion
of this study, all specimens from Fell's Cave were returned to the personal
collection of Mrs. Peggy Fell McKay, Casilla 2D, Punta Arenas, Chile, and
specimens from Mylodon Cave and Alero del Diablo were sent to the
Instituto de la Patagonia. Casilla 102D. Punta Arenas, Chile.

Methods

In identifying the bones, we utilized the extensive series of skeletons of
Patagonian birds in the collections of The University of Kansas Museum of
Natural History and the National Museum of Natural History, as well as
material borrowed from other collections. Osteological terminology fol-
lows Howard (1929). Species names follow Narosky and Yzurieta (1987)
except in the cases of the King Shag, Phalacrocorax "albiventer": which
we consider to be a morph of the Imperial Shag. P. atriceps (Rasmussen,
1991). and the Neotropic Cormorant, P. hrasilianus: a name that has

6 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. LS4

priority over P. olivaceus (Browning, 1989). In addition, Phalcohoenus
species are not considered congeneric with Polyborus in the present treat-
ment, and the Black-faced Ibis, Theristicus melanopls. is considered spe-
cifically distinct from the Buff-necked Ibis. T. caudatus (following Fjeldsa
andKrabbe. 1990).

Feathers with distinctive markings, colors, or moiphologies were iden-
tified macroscopically by Rasmussen. Relatively well preserved, but mac-
roscopically less distinctive, feathers were identified by R. Laybourne, C.
Dove. D. Deedrick, and Rasmussen, following methods developed by
Laybourne (pers. comm.). Poorly preserved feathers and those lacking
down, diagnostic markings or colors could not be identified.

Bone measurements (mm) were taken with dial calipers and follow von
den Driesch ( 1976); if standard measurements provided in that guide were
not possible for a given bone owing to breakage or wear, no measurements
are listed in the present paper.

Order of presentation of remains is as follows: for sites, those from
Fell's Cave are listed first, followed by those from Mylodon Cave and
Alero del Diablo (1 ). (2). (3). (mixed layers); for type of material, feathers
are listed first, followed by bones.

The minimum number of individuals was calculated in each case as the
number of the most abundant element of one side for each radiocarbon-
dated layer at each site, considering whether proximal and distal fragments
overlap or are of similar sizes and whether right and left elements are of
similar sizes ("matching"; Klein and Cruz-Uribe, 1984).

SYSTEMATIC ACCOUNTS

Ptewcnemia pennata (d'Orbigny), Lesser Rhea

Material. — Fell's Cave: tibiotarsi, two distal fragments, right and left
(204/01 [2|); phalanges, two ungual (204/02[2]).

Identification. — The tibiotarsi are indistinguishable from those of
Ptcrocucinia pennata, and differ from those of the Greater Rhea. Rhea
americana. in relative stoutness of the shaft and in the presence of a
prominence on the ventral surface slightly anterior to the condylar fossa. A
large ungual phalanx is indistinguishable from that of the third digit of
Ptevocnemia and differs from that oi R. americana in being slightly shorter
and stouter. A second, more worn specimen lacking the distal end is
probably from the fourth digit. It differs from a comparable specimen of R.
americana in shape of the articular surface and in being stouter and longer.
Rhea americana does not occur south of Rio Negro Province. Argentina
(Blake. 1977). whereas Pterocnemia is common throughout open habitats
of Patagonia. Material refened to P. pennata has been recorded from the
Los Toldos archaeological site in southern Argentina (Tambussi and Tonni.
1985).

AVIFAUNA OF CHILEAN CAVES 7

Table 1 . Characteristics of the three caves, x signifies nearby presence of
habitat type.

Characteristic

Feirs

Mylodon

Alero del Diablo

Dimensions of cavity (m)

Width

11.5

170.0

11.0

Length

9.0

270.0

5.0

Height

3.5

40.0

3.0

Type of cavity

Chambered

Simple

Simple

Associated habitats

Steppe

—

X

Freshwater

X

Forest

X

X

—

Marine

X

X

—

Podiceps cf. P. occipitalis Gamot. Silvery Grebe

Material. — Fell's Cave: femora, one complete right (204/24). one left
with proximal end missing (204/30): tibiotarsus. one right distal end (204/
24).

Measurements. — Femora: length, 35.7; proximal breadth. 9.3: distal
breadth, 9.0, 9.6. Tibiotarsus: distal breadth, 6.6.

Remarks. — Two of these specimens were listed by Saxon (1979) as
"'Podilymhiis cf. Podiceps [sic] antarcticus.'"' However, the two femora
(right and left) differ from those of North American Pied-billed Grebes,
Podilymhiis podiceps, and an immature female P. podiceps antarcticus
(KU 79850) — which in any case has not been recorded in Magallanes
Province, Chile (Venegas and Jory, 1979) — in the following characteris-
tics: the external condyle is more distally located: the internal proximal
edge of the shaft curves more gradually to the head; the trochanteric ridge
is thicker and longer: the proximal end is stouter; and the head is smaller
and not rotated dorsally. The cave femora are slightly larger than those of
Podiceps occipitalis and much larger than in the White-tufted Grebe, P.
rollaiul, from continental South America. The Hooded Grebe, P. gallardoi,
is now only casual in Magallanes Province (Fjeldsa and Krabbe. 1990), and
the three available comparative specimens of P. gallardoi are considerably
larger and heavier than the cave femora (R. W. Storer, in litt.), the latter
probably being within the range of size variation of P. occipitalis. The
tibiotarsus is too worn for specific identification (R. W. Storer, in litt.), and
herein is referred to P. occipitalis because of its association and size. This
tibiotarsus differs from that of Podilymhiis podiceps antarcticus (KU
79850) in that the condyles are much less twisted internally. Podiceps
occipitalis is widespread and cominon throughout Patagonia; the species

8 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 134

Table 2. Layer code and radiocarbon dates in years before present (ybp) for
Fell's Cave.

Layer code Radiocarbon date

01-04 —

05-06 685 ± 90

07-08 ca. 3000

09-12 —

13-17 6485 ±115 through 6560 ±115

18 —

19 8180 ±135

20 ca. 8000
21-22 8480 ±135

33 9100 ±150

23-26 9030 ± 230

34-39 —

27-31 . 10.080 ± 160

32 10,720 ± 300 through 1 1 ,000 ± 1 70

occurs mainly on fresh water, but in Llanquihue Province, Chile, it is also
found on salt water (Rasmussen and Lopez, 1988).

Phalacrocorax atriceps (King), Imperial Shag

Material. — Fell's Cave: ulna, one left distal end (unnumbered).

Measurement. — Distal diagonal, 11.1.

Remarks. — Except as affected by wear between the carpal tuberosity
and the internal condyle, this specimen agrees with male Phalacrocorax
atriceps from the Beagle Channel, which are at the large end of the range of
size variation in this species (Rasmussen, 1990); thus, it cannot be con-
fused with other species of this family found in Patagonia, such as the much
smaller f. hrasiliamis, which is common on fresh water. The presence off.
atriceps in the cave avifauna seems suiprising. but a few populations of this
species exist in lakes (localities summarized in Rasmussen, 1990), al-
though no freshwater populations are known in southern continental
Magallanes Province. Also, P. atriceps was a cominon item in the diet of
coastal inhabitants of southern Chile and Tierra del Fuego (Orquera et al.,
1977; Lefevre, 1989; Rasmussen et al., MS; Humphrey and Rasmussen,
unpubl. data), and were sometimes presented as gifts (Bridges, 1949), so
carcasses might have been brought to the cave by humans.

Theristicus rnelanopis (Gmelin), Black-faced Ibis

Material. — Fell's Cave: humerus, one right partial shaft (203/16); ra-
dius, one right shaft (204/26); caipometacarpus, one right proximal half
lacking most of the trochleae and process of metacarpal I (204/26). Mylodon

AVIFAUNA OF CHILEAN CAVES

T3

C

o

a
o

o

■3
a:

c
o

2

C

■a
o

Z

o

c

rT

o

c

(U

c

rt

c

o

lO

m

tT

+1

o

+1

+1

+1

o

vC

m

so

'^

1/-,

s

IT-,

V

vO

oc

DC

U-,

r<~,

vO

t^

fN

>r-j

U-^1

t^

oo

+1

C
00

oc

1^

oc

oc
oc

+1

+1

+1

ri
ir,

OC

ri

4J
^ /— ^ /^ - — ^

.:; .z: .:i .=: '-^ o

(L)
O

■o
■a

x:

i/5

13

C

o
o

E 6 °

-■ -^ >^

■73 a 7i

&. D. D.

3 3 3

o o

o o

o o

O 'J —

o o —

O O •-':

c c c c

3 3 3 3 iJ >

x i: I I g ^

c
o
X;

73

3

o

CQ

CM

SO

o

<

1

Z

O

ri

r--'

i±.

<

CQ

r)

r-i

0)

— o o

O E O

c — — j3 _2J

C SB

3 C

c
o

3

c

OJ

a-

OJ

3 " -

3 •"= 5

^ ^ O

Q

J

3 3

13 13

r- r-

5 5

13 13

_ _2 _o

s s s

T

r^ ZI —

I I I I

r '^' 0^ '^'

9 9 m '<

f^ r^i ^- -

'^' <^-

r<-,

I

oc

^ 7

^ ^ <

r*-, r<-j m

■It-

c
o

13
O

O

o
a.

<u

13

1)

>

1 = 1
lie

O ^ .r_
C ^ "_)
. , ~ '-^

_ o

3

-w ^ >v
S C/J OJ

o t: >

C/J X

■£

~

OX)

>

J

•o

"O

OJ

O

r3

r3

JO

o

'^

y:

o

^

■U

./;

y.

o

./:

y.

C3

r3

"d.

y;

y!i

P

O

O

**"

c::

•^

■-T.

•u

u.

u-

~

._o

■^

■'■'

^

^

_^

-^

E

E

c

3

3

3

DC

re

X

10 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

Cave: scapula, one right proximal end (2A"5\A2); coracoid. one proximal
end(2A'5\Al).

Measurement. — Scapula: proximal diagonal, 13.6.

Remarks. — Although fragmentary, these specimens are highly diag-
nostic and cannot be confused with those from any other species found in
Patagonia. Theristicus melanopis is a common breeding species typical of
Magellanic moorland and aquatic habitats, including small openings in
forests (Vuilleumier, 1985); it winters on the pampas (Fjeldsa and Krabbe,
1990). but a few overwinter as far south as Tierra del Fuego (Humphrey et
al., 1970). Despite its abundance, the species has not been found in ar-
chaeological sites of the Beagle Channel (Rasmussen et al., MS; Humphrey
and Rasmussen, unpubl. data), or from those of the Brunswick Peninsula,
Seno Otway, or the Atlantic coast (Lefevre, 1989), showing it was not a
favored prey of coastal humans.

Chloephaga picta (Gmelin). Upland Goose

Material. — Fell's Cave: coracoids, one nearly complete right (204/05),
one left proximal half lacking furcular facet (204/01 ); humeri, three com-
plete (right, 204/05 [2], and left, 204/01), two proximal fragments (right,
204/10, and left, 204/14 [both referred]); eight distal fragments (right, 203/
20, 204/14, 204/20, and left, 20J05, 204/13, 204/14,^204/15 [refen-ed],
204/17); ulnae, three partial shafts (right, 204/26 [2], left, 204/29 [all
referred]); carpometacarpus, one right shaft (204/04); synsacrum, one in-
complete (203/03); tibiotarsi, two shafts (right, 203/19 [referred], and left,
203/02), two distal fragments (right, 203/11, and left. 204/19 [both re-
ferred]). Mylodon Cave: feathers, two contour, one underwing covert
(2" 1 ■& 7/2' 1 '.F,-Fn, [3 referred]); skull, one fragment of right orbital rim
and skull roof (Nordenskjold 'A'). Alero del Diablo (mixed): one vertebra,
incomplete (referred).

Measurements. — Humeri: length, 141.8; proximal breadth, 30.4; distal
breadth, 19.9, 21.0, 21.2. 22.2. 23.4. 24.6. Tibiotarsus: distal breadth, 15.9.

Remarks. — Three feathers from Mylodon Cave were identified micro-
scopically as being from a species of Chloephaga and macroscopically
they best match female C. picta: thus, they are referred to this species.

A skull fragment from Mylodon Cave is identified as Chloephaga picta
because the nasal gland impression is more rounded than in the Ruddy-
headed Goose, C. ruhidiceps, and the Ashy-headed Goose, C. poliocephala.
Also, the impressions are not contiguous medially as in the Kelp Goose, C.
hyhrida.

Two coracoids are longer and have thicker shafts than in Chloephaga
hyhrida and the more nearly complete specimen differs from that of the
Magellanic Flightless Steamer-Duck, Tachyeres pteneres. in that the ven-
tral lip of the sternal facet is wide and prominent; both coracoids are

AVIFAUNA OF CHILEAN CAVES 1 1

heavier than are those of Tachyeres.

Nine specimens of humeri are clearly those oi Chloephaga picta, either
because they are larger than those of C. hyhrida or they lack the distinct
anconally produced process on the external condyle found in C. hybrida.
Three other humeri are referred to C. picta, but are too fragmentary to allow
certain identification.

Three ulnar fragments may represent Chloephaga picta but cannot be
distinguished with certainty from those of other species in the genus.
Species of the genus Tachyeres are ruled out as they have much shorter and
more curved ulnae. A carpometacarpus is larger and longer than those of
any other related species in the region, as is a synsacral specimen. A partial
tibiotarsus is larger than those of any other species of Chloephaga or
Tachyeres. Three other tibiotarsi are referred to C. picta but are not diag-
nostic to species.

Chloephaga picta is now abundant in a variety of habitats, especially
Magellanic moorland, and large numbers are killed as agricultural pests. It
is migratory in the southern part of its range (Madge and Burn, 1988),
although some winter in the Magellanic zone (Fjeldsa and Krabbe, 1990).
For the Beagle Channel, very few bones of this species were found in the
Tiinel site (Humphrey and Rasmussen, unpubl. data) and none was found at
Lancha Packewaia (Rasmussen et al., MS): bones of at least 15 were found
by Lefevre (1989) at the Punta Maria site (300 ± 100 ybp) on the Atlantic
coast of Tierra del Fuego and at least three at the Punta Santa Ana site
(about 5000-6000 ybp) on Peninsula Brunswick.

Chloephaga ruhidiceps Sclater. Ruddy-headed Goose

Material. — Fell's Cave: humeri, two distal fragments (right, 203/21,
and left, 204/37): ulna, one incomplete left (204/26): tibiotarsus, one right
shaft (204/26).

Remarks. — These specimens are indistinguishable from female
Chloephaga ruhidiceps: the left humerus differs from that of C.
poliocephala in having the internal edge of the brachial depression more
nearly parallel with the shaft. The cave specimens are smaller than females
of C. poliocephala, the only other small species in the genus. The right
humerus is referred to C. ruhidiceps on the basis of its small size.

Chloephaga ruhidiceps was formerly abundant (Crawshay, 1907: Scott,
1954), but this grassland species is now rare on the continent (Collar and
Andrew, 1988), possibly owing to predation by introduced foxes (Rumboll,
1975), lack of protection from hunting (Christie. 1984), and overgrazing
(Fjeldsa, 1988). It is a strongly migratory breeding visitor to Magallanes
Province (Venegas, 1986), and is semi-aquatic when the young are small
(Johnson, 1965). Unlike other waterfowl, this species apparently has no
flightless molting period (Summers, 1986), presumably making it less

12 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

vulnerable to capture. Lefevre (1989) reported a minimum of three indi-
viduals of this species from a site dated at around 5000-6000 ybp at Punta
Santa Ana. on the Strait of Magellan coast of Peninsula Brunswick, as well
as at least two individuals from the San Pablo site (ca. 300 ybp) on the
Atlantic coast of Tierra del Fuego (Fig. 1 ).

Tachyeres patachonicus (King), Flying Steamer-Duck

Material. — Fell's Cave: humerus, one right distal fragment (204/29).

Remarks. — This fragment consists of only the condyles and
ectepicondyle. It differs from comparably sized Chloephaga ruhidiceps
and agrees with Tachyeres patachonicus in the rounded, rather than squared,
outline of the external condyle in external view, the shallowness of the
distal end of the external tricipital groove, the relative depth of the olecra-
non fossa, and the prominently ridged ectepicondyle. The humerus is the
size of that of female T. patachonicus; in this highly dimorphic species
(Livezey and Humphrey. 1984). females do not overlap in size with either
sex of the larger T. pteneres.

Tachyeres patachonicus is common both in fresh and saltwater environ-
ments throughout Patagonia. Some large males are permanently flightless
(Humphrey and Livezey. 1982); this, together with the usual flightless
molting period, may have facilitated the capture of this species by Fuegians.
who used several methods (Bridges. 1949). The species was represented by
a few elements in the archaeological sites of humans of the Beagle Channel
(Rasmussen et al.. MS; Humphrey and Rasmussen, unpubl. data), but was
absent in middens on Seno Otway, Peninsula Brunswick, and the Atlantic
coast of Tierra del Fuego (Lefevre. 1989).

Anas sihilatrix Poeppig, SoutheiTi Wigeon

Material. — Fell's Cave: humerus, one left lacking distal end (204/37);
carpometacarpus. one left lacking most of both ends (204/26).

Measurement. — Humerus: proximal breadth. 18.6.

Remarks. — These specimens are the size of bones irova Anas sihilatrix
and are notably stouter than those of the Brown Pintail. ^4. georgica, which
are of comparable length and shorter than those of the similarly stout
Bronze-winged Duck. A. specularis. The subgenus Mareca. to which A.
sihilatrix belongs (Livezey, 1991). is typified by heavy bones, but is
otherwise similar to other species of Anas (Woolfenden. 1961 ).

Anas sihilatrix, which migrates north for the austral winter (Johnsgard.
1978). is a common species in freshwater habitats of Patagonia. It also
occurs in estuarine habitats, at least at Peninsula Lacuy. Chiloe Island
(Humphrey and Rasmussen. pers. obs.). A minimum of five individuals of
this species were found in the Bahia Buena midden (Fig. 1 ) on Peninsula
Brunswick (Lefevre, 1989).

AVIFAUNA OF CHILEAN CAVES 1 3

Anas cf. A. georgica Gmelin, Brown Pintail

Material. — Fell's Cave: humerus, one left distal third (204/01).

Measurement. — Distal breadth. 11.2.

Remarks. — This humerus is referable to Anas georgica and is larger
than humeri of the Speckled Teal, A. flavirostris. the Silver Teal, A.
versicolor, and the Cinnamon Teal, A. cyanoptera. It is approximately the
same size as a specimen of the Red Shoveler, A. platalea. but differs from
it in that the shaft does not naiTow to the same degree immediately proxi-
mal to the impression of the brachialis amicus. This specimen is smaller
and has a thinner shaft than do humeri of A. sihilatrix and A. specular is.

Anas georgica is widespread and abundant in Patagonia in freshwater,
estuarine, and coastal habitats, and is migratory in the southern portion of
its range (Madge and Burn, 1988). Lefevre (1989) recorded this species
from the Bahia Buena site (dated at about 5000-6000 ybp) on Peninsula
Brunswick and the Punta Maria site (300 ± 100) on the Atlantic coast of
Tierra del Fuego (Fig. 1 ).

Anas cf. A. flavirostris Vieillot, Speckled Teal

Material. — Fell's Cave: humerus, one left lacking proximal end (204/
19).

Measurement. — Distal breadth, 9.4.

Remarks. — The preserved portion of this humerus is identical to that of
Anas flavirostris; it is slightly smaller than those of .4. versicolor and much
smaller than those of other species of Anas in the region. We cannot rule out
the possibility that this specimen pertains to A. versicolor Anas flavirostris
is common in freshwater to coastal habitats throughout Patagonia and
southern populations migrate north for the austral winter (Fjeldsa and
Krabbe. 1990). Anas versicolor is also an abundant summer visitor to
southern Chile. Anas flavirostris was recorded from the Punta Maria site
(dated at 300 ± 100 ybp) on the Atlantic coast of Tierra del Fuego (Lefevre,
1989; Fig. 1).

Anas species indetemiinate

Material. — Fell's Cave: humeri, two distal fragments (right. 204/21,
and left. 204/38); tarsometatarsus, one left lacking most of hypotarsus
(204/37).

Measurements. — Tarsometatarsus: length, 38.4: distal breadth, 7.5.

Remarks. — The humeri are the size of those of Anas platalea and A.
georgica, but are too fragmentary to allow further identification. The
tarsometatarsus. which is from a largely ossified juvenile, may pertain to
either of the above species ox A. sihilatrix. adults of which have tarsometa-
tarsi with heavier shafts. The cave tarsometatarsus is too small to be that of

14 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

A. specularis and too large to be that of other species oi Anas known from
the region.

Buteo ventralis Gould, Rufous-tailed Hawk

Material. — FelFs Cave: coracoid. one right incomplete (204/20); hu-
meri, one left incomplete (204/37), two proximal fragments (right, 204/20,
and left, 204/37), four distal fragments (right. 204/20, 204/37 [2], and left,
204/20); ulnae, one complete left (204/01), one left shaft (204/26);
synsacrum, one incomplete (204/37); femora, four right proximal frag-
ments (204/20 [2], 204/37 [2]), three distal fragments (right. 204/37. and
left. 204/19. 204/20). Alero del Diablo (mixed): scapula, one left proximal
fragment; radius, one left shaft; tibiotarsus, one left distal fragment.

Measurements. — Humeri: length, 101.8; distal breadth, 17.1, 17.2,
19.7. Ulna: length. 119.7; proximal breadth. 11.7; distal diagonal. 10.4.
Femora: proximal breadth, 16.9, 18.3; distal breadth, 17.3. Tibiotarsus:
distal breadth. 13.7.

Remarks. — Because no skeletal specimens of the Rufous-tailed Hawk,
Buteo ventralis, are available (Wood and Schnell, 1986), a humerus lacking
the proximal end and a complete ulna were removed from a study skin of a
male B. ventralis (USNM 400056) by staff of the Division of Birds,
National Museuin of Natural History. Buteo ventralis is member of the B.
jamaicensis superspecies (Hellmayr and Conover, 1949; Amadon, 1964;
Mayr and Short, 1970; Clark, 1986; Sibley and Monroe. 1990); therefore,
for elements unobtainable from study skins, skeletons of eight male B.
jamaicensis were used for comparison. Like B. jamaicensis, B. ventralis is
a larger-bodied, relatively shorter-winged bird than B. polyosoma. The ulna
and partial humerus of B. ventralis are very similar to those of B.
jamaicensis.

The scapula from Alero del Diablo is slightly larger than it is in female
Buteo polyosoma, and the blade is distally more slender; it is similar to that
of male B. jamaicensis. The coracoid has a heavier shaft both in medial and
dorsal views than it does in B. polyosoma; in this regard, the coracoid is
similar to that oi B. jamaicensis.

The assigned humeri are like those of B. jamaicensis and USNM 400056
in being relatively short, having a heavier shaft than in those of B.
polyosoma, and having a wider pneumatic foramen.

Like Buteo jamaicensis and B. ventralis, the cave ulnae are shorter with
relatively thicker shafts than are those of a female B. polyosoma. The radius
matches that of B. jamaicensis in being heavier than that of B. polyosoma.
The sacrum is larger than are those of any of nine B. polyosoma.

The femora listed resemble those of Buteo jamaicensis and differ from
those of B. polyosoma as described in the account of the latter species. The
distal tibiotarsus is larger than is that ofB. polyosoma, has the supratendinal

AVIFAUNA OF CHILEAN CAVES 15

bridge more vertically oriented, and has a distinct furrow along the anterior
center of the shaft, as in that of B. janmicensis.

Buteo veiitralis is a poorly known, rare species throughout its range
(Humphrey and Bridge. 1970; Collar and Andrew, 1988) and there is
conflicting information on its habitat preferences. Some authors consider it
to be restricted to forest (Narosky and Yzurieta, 1987), but it has been
recorded from the forest-steppe ecotone (Vuilleumier, 1985; Rasmussen et
al., 1992) and from open country (Philippi et al., 1954; Bernath, 1965;
Clark. 1986). Fjeldsa and Krabbe (1990) stated that the species occurs in
"forest edge, open forest, and parkland . . . Patagonian brush-steppe," and
Rasmussen found an individual in heavy forest on the shore of Lago
General Vintter, Chubut Province, Argentina (Rasmussen et al., 1992).

Buteo polyosoma (Quoy and Gaimard), Red-backed Hawk

Material. — FelTs Cave: humeri, one left proximal fragment (204/37),
four distal fragments (three right, one left, 204/20 [4]); synsacrum, one
partial (204/37); femora, three nearly complete (right, 204/37 [2], left, 204/
20), five proximal fragments (right, 204/20 [2], 204/21, and left, 204/21,
204/37), one right distal end (204/20); tibiotarsus, one right distal end (203/
20); tarsometatarsus. one left distal third (204/26). Mylodon Cave: feather,
one mantle (Nordenskjold 'A." refened). Alero del Diablo (2): tarsometa-
tarsus, one left shaft. Alero del Diablo (mixed): tarsometatarsi, two com-
plete, right and left.

Measurements. — Humeri: distal breadth, 16.7. 19.2. Femora: length.
77.8; proximal breadth, 14.2. 14.2. 15.9, 16.2, 16.7; distal breadth. 15.4.
Tibiotarsus: distal breadth, 13.0. Tarsometatarsi: length, 85.7, 85.7; proxi-
mal breadth, 14.5, 14.6; distal breadth, 15.1, 15.3.

Remarks. — A feather from Mylodon Cave is referred to this species; its
microscopic structure is consistent with its being a Buteo species and it is
a dark reddish brown, as is the mantle of female B. polyosoma.

The humerus oi Buteo polyosoma is relatively longer and slimmer than
are those of B.jamaicensis and B. ventralis. The humerus of B. polyosoma
also has a narrower pneumatic foramen than does that of B. jamaicensis.
One of the distal humeri is the size of an exceptionally small male B.
polyosoma.

The femora were identified as pertaining to Buteo polyosoma based on
their having, in comparison to those of male B. jamaicensis, more slender
shafts and necks, a more laterally directed fibular condyle, a deeper and
narrower furrow on the external side of the fibular condyle, and a more
distinct obdurator ridge. Two femoral fragments (204/21 and 204/20) are
from juvenile birds, probably nearly fully grown. The distal part of a
tibiotarsus is referred to B. polyosoma based on the less longitudinally
oriented supratendinal bridge and the shallow furrow proximal to the

16 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

tendinal groove in comparison with male B. jamaicensis.

Two complete and one partial tarsometatarsi from Alero del Diablo and
a charred distal third of a tarsometatarsus from Fell's Cave agree with
Buteo polyosoma and differ from B. jamaicensis in having thin shafts, and
a less posteriorly flaring and more distally located wing of the trochlea for
Digit II.

Buteo polyosoma is a common breeding species both in open country
and forest (Vuilleumier. 1985) that migrates to the pampas (Fjeldsa and
Krabbe. 1990). This species is represented by very little material in the
Tiinel site on the Beagle Channel (Humphrey and Rasmussen, unpubl.
data).

Buteo species

Material. — Fell's Cave: humeri, one right proximal fragment (204/20),
two left distal fragments (204/20, 204/37); femur, one left proximal frag-
ment (204/30); claw, one (204/18). Alero del Diablo (mixed): pes phalanx,
one; claw, one.

Measurement. — Humerus: distal breadth, 16.6.

Remarks. — This material is not diagnostic and may pertain either to
Buteo ventralis or B. polyosoma.

Geranoaetus melanoleucus (Vieillot), Black-chested Buzzard-
Eagle

Material. — Fell's Cave: humerus, one left proximal fragment (204/20).
Mylodon Cave: vertebrae, seven nearly complete (3'5A'.2E.A1).

Measurement. — Humerus: proximal breadth, 28.8.

Remarks. — The seven vertebrae are slightly larger than those from a
male specimen of Geranoaetus melanoleucus, suggesting that they were
from a female of the species.

The humerus of Geranoaetus melanoleucus is much larger than those of
other accipitrids in the region; the cave specimen is indistinguishable from
the humerus of G. melanoleucus. a relatively common species in a wide
range of habitats, from closed forest to dry steppe, in the southern part of its
range (Vuilleumier, 1985).

Polyborus plancus (J. F. Miller), Crested Caracara

Material. — Alero del Diablo ( 1 ): coracoid, one right nearly complete.
Alero del Diablo (mixed): furcula, one incomplete; humerus, one left shaft
with portion of proximal end; tibiotarsus, one complete left.

Measurements. — Furcula: length, 54.6. Tibiotarsus: length, 116.8;
proximal breadth, 13.9; distal breadth, 15.2.

Remarks. — A furcula with the distal two-thirds of the left clavicle
missing is much larger than the furcula of the Chimango Caracara, Milvago

AVIFAUNA OF CHILEAN CAVES 17

chimango and that of the White-throated Caracara. Phalcohoenus
alhogiilaris: it is the same size as those of the Striated Caracara,
Phalcohoenus australis and the Crested Caracara, Polyhorus plancus. The
furcula resembles that of P. plancus and differs from that of P. australis in
having the symphysis distinctly ridged, the anterior margin of the clavicle
thin for its distal three-fourths; the distal two-thirds of the clavicle consid-
erably broader, and the clavicles relatively broad at the symphysis.

Acoracoid of Polyborus plancus from Alero del Diablo is indistinguish-
able from an unsexed specimen of that species. The coracoid of
Phalcohoenus australis is about the same length as that of Polyborus, but
differs in having a much wider shaft.

The shaft and distal part of the head of a humerus of Polyhorus plancus
from Alero del Diablo (mixed) appears identical to that of an unsexed
specimen and differs from that of Phalcohoenus australis in the curvature
of the shaft and greater length of the shaft between the distal end of the
deltoid crest and the proximal end of the impression of the brachialis
anticus.

A complete tibiotarsus differs from that of Phalcohoenus australis and
resembles that of Polyhorus in having a much narrower shaft (especially
distally), a shorter, shallower tendinal groove, and a different conformation
of the supratendinal bridge.

In Patagonia, Polyhorus plancus is a common species in open habitats
and forest edges (Vuilleumier, 1985). A very few were recovered from
middens along the Beagle Channel (Rasmussen et al., MS; Humphrey and
Rasmussen, unpubl. data) and the Punta Maria site on the Atlantic coast of
Tierra del Fuego (Fig. 1; Lefevre, 1989), and a furcula referred to the genus
Polyhorus was found in a cave deposit in southern Santa Cruz Province.
Tonni (1984) suggested that humans may have caught this species in traps.

Phalcohoenus albogularis Gould, White-throated Caracara

Material. — FelFs Cave: humerus, one left lacking most of proximal
portion (204/04); femora, two complete (right. 204/04, and left, 204/05).
one distal right end (number broken off); tibiotarsus, one right shaft frag-
ment (203/11). Mylodon Cave: coracoid, one left proximal third
(2B'3\A1).

Measurements. — Humerus: distal breadth, 15.3. Femora; length, 63.6,
65.1; proximal breadth, 14.9, 15.1; distal breadth, 13.6, 14.3, 14.6.

Remarks. — These elements are assigned to Phalcohoenus albogularis,
which is much larger than Milvago chimango and much smaller than
Polyhorus plancus and Phalcohoenus australis.

The distal end of a femur pertains to Phalcohoenus species, as the ridge
on the internal edge of the popliteal area is narrower, the fibular condyle is
less-produced than in Polyhorus, and the fibular groove does not extend

18 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

proximally beyond the proximal edge of the external condyle. The femur is
considerably larger than that of a male PJialcohoemis alhogidaris, but
because it is much smaller than the femora of either sex off. australis. it
probably is from a female P. alhogularis.

The tibiotarsus has a much thicker shaft than it does in male P.
albogularis and is probably from a female of the species, because it is much
smaller than the tibiotarsi of either sex of P. australis or of Polybonis
planciis from Tierra del Fuego.

Phalcoboenus alhogularis is an uncommon species that typically is
associated with forest, but sometimes is found in nearby open areas
(Vuilleumier, 1985), even breeding in steppe (Fjeldsa and Krabbe, 1990).
This species was found in the Tiinel site (Humphrey and Rasmussen,
unpubl. data).

Milvago chimango (Vieillot), Chimango Caracara

Material. — Mylodon Cave: feathers, two contour, one underwing co-
vert (Nordenskjold 'A").

Remarks. — These feathers are in very good condition; they have re-
tained their diagnostic color pattern and are indistinguishable microscopi-
cally from those of Milvago species. Milvago chimango now is a common
to abundant species throughout Patagonia in virtually all terrestrial habitats
(except where persecuted; Jehl and RumboU, 1976). Specimens of this
species were uncommon in kitchen middens in the Beagle Channel
(Rasmussen et al., MS; Humphrey and Rasmussen. unpubl. data) and other
sites on the Brunswick Peninsula (Bahia Buena). Seno Otway (Bahia
Colorada, Punta Baja), and the Atlantic coast of Tierra del Fuego (Punta
Maria, San Pablo; Lefevre, 1989).

Fulica armillata Vieillot. Red-gartered Coot

Material. — Fell's Cave: femur, one left distal half (204/19); tarsometa-
tarsus. one right lacking proximal end (203/20).

Measurements. — Femur: distal breadth, 12.1. Tarsometatarsus: distal
breadth, 10.7.

Remarks. — The tarsometatarsus is indistinguishable from that of a
male specimen of Fulica armillata, the largest and most common species
of Fulica in Patagonia. The allocation of the femur, however, is less certain
because it has an especially heavy shaft and is considerably larger than the
reference material. There is minimal size variation in skin measurements
among a small sample of male F. armillata, and females are considerably
smaller than males (Ripley. 1977). Fulica armillata is abundant and wide-
spread in wetlands (lakes or large ponds) throughout Patagonia (Ripley,
1977).

AVIFAUNA OF CHILEAN CAVES 19

Rallidae, genus and species indetenninate

Material. — Mylodon Cave: feathers, three contour (3'6'.3).
Remarks. — These feathers are not diagnostic to genus or species within
the family Rallidae, although they clearly are not from the genus Fiilica.

Attagis malouiuus (Boddaert), White-bellied Seedsnipe

Material. — Fell's Cave: synsacrum, one partial (204/37).

Remarks. — This synsacrum agrees in size and characters with that of a
specimen of Attagis maloiiimis. It is somewhat smaller than that of the
Rufous-bellied Seedsnipe, A. gayi and larger than synsacra of both species
of Thinocorus. The low ridge on the proximoventral surface of the fused
synsacral vertebrae is distinctive in the Thinocoridae.

Several authors cited A. gayi as occurring throughout the continental
Andes (e.g., Hayman et al., 1986; Narosky and Yzurieta. 1987: Fjeldsa and
Krabbe. 1990), but its present status in Magallanes Province is unknown
(Venegas, 1986). Attagis malouiuus is a rather rare, sparsely distributed,
alpine-breeding species that winters in small flocks in the moorlands and
steppe of northern Tierra del Fuego (Crawshay, 1907; Humphrey et al.,
1970; Clark. 1986) and Magallanes Province (Venegas, 1986): its presence
in FelFs Cave shows that this shelter was used in autumn or winter. This
genus was found in the Punta Maria site (dated at 300 ± 100 ybp) on the
Atlantic coast of Tierra del Fuego (Fig. 1; Lefevre, 1989).

Tyto alba (Scopoli), Bam Owl

Material. — Mylodon Cave: feathers, one primary, one contour
(Nordenskjold "A'): one facial disk, five contour (2'r& 7/2'r.F,-F,J; one
contour (3'6M): one upper wing covert (3'6\3): one facial disk, one
secondary, one contour (3'6\4): one probable rectrix (3*6*. 7W); one flight
feather (3"6'.8W): one contour (3'6\9W); one facial disk, one contour
(5'3DM); two contour (5"7A'); one upper wing covert, two secondaries,
two contour (5'7M); sternum, right half with carina (5'7B\A1). Alero del
Diablo ( 1 ): coracoid, one right nearly complete; humerus, one right nearly
complete: ulna, one right complete except for olecranon: radius, one left
proximal half. Alero del Diablo (2): scapula, one left proximal half; cora-
coid, one left nearly complete: humerus, one left partial shaft. Alero del
Diablo (3): ulna, one left lacking proximal end. Alero del Diablo (mixed):
humeri, one left nearly complete, one left proximal half: radius, one right
complete: femur, one right shaft.

Measurements. — Scapula: proximal diagonal, 8.2. Coracoid: medial
length, 38.3. Humeri: length, 90.5, 92.5; proximal breadth, 15.5, 15.5;
distal breadth, 13.6. 14.4. Ulnae: proximal breadth, 8.8; distal diagonal, 7.2,
7.3. Radius: length, 97.6.

20 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

Remarks. — Most feathers and skeletal specimens recovered of Tyto
cannot be confused with any other species in the region. A number of
additional feathers and feather fragments not listed here but assigned to this
taxon are in poor condition and cannot be certainly identified microscopi-
cally because they lack down.

Tyto alha is uncommon in the southern part of its range and there, as
elsewhere, is catholic in its choice of habitat, although it requires the
presence of caves or crevices (Fjeldsa and Krabbe. 1990). It is also repre-
sented in kitchen middens of the Beagle Channel (Rasmussen et al., MS;
Humphrey and Rasmussen, unpubl. data).

Stvix rufipes King, Rufous-legged Owl
Material. — Mylodon Cave: feathers, four contour {2'\'&. 7/2'!". F-

Fn).

Remarks. — These feathers are distinctively marked; no other owl in
Patagonia has such boldly banded contour feathers. So far as is known, the
southern subspecies. Strix rufipes rufipes, is strictly associated with forest
or forest/steppe ecotone (Fjeldsa and Krabbe, 1990), and is rare in
Magallanes Province (Venegas, 1986), although it may often be over-
looked. One bone of this species was found in the Tiinel site from the
Beagle Channel (Humphrey and Rasmussen, unpubl. data).

Aphrastura spinicauda (Gmelin), Thorn-tailed Rayadito

Material. — Mylodon Cave: feathers, one inner right primary (3'6M);
one right rectrix 4 (3'6\4); one right rectrix 3 (2A'2".F); two rectrices
(2T&7/2'r.F-F„).

Remarks. — These feathers are readily diagnosible by the pattern of
rufous and dark brown and the spines on the rectrices. Aphrastura
spinicauda is most abundant in forest, but also occurs on treeless islands in
bunchgrass and bushes (FjeldsS and Krabbe, 1990; Vuilleumier, 1991 ) and
in tussock grass along the Beagle Channel. Isla Grande de Tierra del Fuego
(Rasmussen and Lopez, pers. obs.). One bone of this species was found in
the Tiinel site (Humphrey and Rasmussen, unpubl. data).

Tiirdus falcklandii Quoy and Gaimard, Austral Thrush

Material. — Mylodon Cave: feather, one secondary (2'\'& 1/2' ]'.¥.-
F„); humerus, one right proximal end (7CS'7\A1). Alero del Diablo (1):
humerus, one right complete. Alero del Diablo (mixed): sternum, one
nearly complete; humerus, one left complete.

Measurements. — Sternum: length, 31.5. Humeri: length, 28.5, 29.5;
proximal breadth, 9.0, 9.4, 9.4; distal breadth. 7.2, 7.6.

Remarks. — No other species of Turdus occurs in southern Patagonia.
Turdus falcklandii occurs in a wide range of habitats, from herbaceous

AVIFAUNA OF CHILEAN CAVES 2 1

grasslands through forest (Ralph, 1985). It also is represented In low
numbers in the Tiinel site (Humphrey and Rasmussen, unpubl. data).

Phrygilus patagoniciis Lowe, Patagonian Sien^a-Finch

Material. — Mylodon Cave: feathers, two contour (Nordenskjold "A'),
one contour (5'7A'). two down (2'r& 7/2'r); maxilla, one (3'6'.2);
synsacrum, one partial (3'6'.7W).

Remarks. — Two distal tips of contour feathers match the mantle color
of male Phrygilus patagonicus, which is unique among small birds in
Patagonia. One of these feathers was indistinguishable microscopically
from a modern specimen of this species. The two down feathers were
determined microscopically to be from this genus. The other feather is dark
gray, as in the head oi P. patagonicus. and is referred to this species on the
basis of this and its association with the dorsal feathers.

The maxilla is indistinguishable from those of two male Phrygilus
patagonicus. The Gray-hooded Sierra-Finch. P. gayi. is a slightly larger
bird, with a less conical bill (Fjeldsa and Krabbe, 1990). In the Mourning
Sierra-Finch, P.fruticeti, the ventral strut of the nasal bone is narrower, and
the dorsal strut of the nasal and premaxilla is longer and more convex
dorsally. No other fringillid of the region shares the maxillary conforma-
tion of Phrygilus.

The synsacrum is the same size and shape as that of male Phrygilus
patagonicus. It differs from the synsacrum of the similarly sized Yellow-
bridled Finch, Melanodera melanodera. in being more ventrally curved at
the distal end. The synsacrum is larger than that of the Rufous-collared
Sparrow, Zonotrichia capensis. It can be distinguished from the synsacrum
of the Common Diuca-Finch, Diuca cliuca, because the proximal edge of
the ilium is more laterally oriented and the fused sacral vertebrae are wider
in ventral aspect.

Phrygilus patagonicus is primarily a forest species, whereas its close
relative, P. gayi, is associated with drier habitats; the two species hybridize
(Vuilleumier, 1991) and flock together during the non-breeding season
(Ridgely and Tudor, 1989). In contrast to Fjeldsa and Krabbe (1990), we
found P. patagonicus abundant in heavy beech forest, such as that found
around Lago General Vintter. Chubut (Rasmussen et al., 1992).

Carduelis harhata (Vieillot), Black-chinned Siskin

Material. — Mylodon Cave: feather, one left rectrix, probably rectrix 5
(3-5AM.F1.F2).

Remarks. — This feather retains the distinctive pattern of light and dark
of the rectrices of this species, although the yellow areas have faded to
whitish. Carduelis barbata now is one of the most common birds in forest
and forest edge in Patagonia.

22 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

DISCUSSION

The avian remains identified in this study differ in species content and
abundance among the three caves (Table 4). These bird remains comprise
96 bones from Fell's Cave (Table 5). 15 bones and 50 feathers from
Mylodon Cave (Table 6), and 28 bones from Alero del Diablo (Table 7), for
a total of 139 bones and 50 feathers, representing at least 84 individuals of
25 species and 10 orders of birds. This tally does not include eight uniden-
tified bone fragments and a number of unidentified feathers (many frag-
mentary).

Of the three caves. Fell's Cave is the richest in numbers of individuals
and species of birds (Table 4), although in comparison to mammals, rela-
tively few birds were found at any of the sites (cf. Saxon, 1976, 1979). The
avifauna of these caves differs markedly from those found in other sites of
southern South America (Lefevre, 1989: Rasmussen and Humphrey, 1991;
Rasmussen et al., MS; Humphrey and Rasmussen, unpubl. data), which
were predominantly comprised of penguins and especially cormorants. (At
one site, Engleheld, all of the avian remains were from cormorants [Lefevre,
1989].) Both Mylodon Cave and Alero del Diablo are very near salt water
and Alero del Diablo contained many marine vertebrates; thus, it is surpris-
ing that no marine birds were recovered from these caves. The abundance
of diurnal raptors in the cave deposits is much greater than in the coastal
middens, which only contained the hawks Buteo polyosoma and Acclpiter
bicolor (Lefevre, 1989; Humphrey and Rasmussen, unpubl. data).

The use of Fell's Cave by humans in both the austral breeding and non-
breeding seasons is indicated by the presence of several migratory breeding
bird species, as well as Attagis malouinus, which is expected at low
elevations only in winter. The juvenile hawk femora and duck tarsometatar-
sus are from birds that may have been post-fledging; this suggests that they
were deposited in late summer or fall. No definite evidence of seasonality
of use was detected at either Mylodon Cave or Alero del Diablo, although
the presence of Theristicus melanopis, Chloephaga picta, and Buteo
polyosoma strongly suggests use during the austral summer. Most species
of Patagonian birds are known to be residents or only partial migrants, or
lack reliable data on winter ranges. Middens on the Beagle Channel were
utilized throughout the year, based on the presence of juvenile bones and
bird species that today are migratory breeding species or winter visitors
(Lefevre, 1989; Rasmussen and Humphrey, 1991).

Climatic and Habitat Change and Relationship to Avifauna

The bird remains in the two dated caves span a period of approximately
12,000 years and corresponding post-glacial climatic and vegetational
changes summarized in Heusser (1966, 1984), Paskoff (1977), Moore

AVIFAUNA OF CHILEAN CAVES 23

Table 4. Bird species from the three caves in this study. First number in each
column is total number of bones of that species from that cave; second is minimum
number of individuals (MNI). "F" following a number signifies number of feathers
identified.

Species

Fell's

Mylodon

Alero del Diablo Totals

Pterocneinia pennata

4/2

4/2

Podiceps cf. P. occipitalis

3/2

3/2

Phalacrocorax atriceps

1/1

1/1

Theristicus melanopis

3/2

2/1

5/3

Chloephaga picta

24/14

3F. 1/2

1/1

3F, 26/17

Chloephaga ruliidiceps

4/3

4/3

Tachyeres patachoiiiciis

1/1

1/1

Anas sihilatrix

2/2

2/2 y

Anas cf. A. georgica

1/1

1/1

Anas cf. A. flavirosths

1/1

1/1

Anas sp.

3/2

3/2

Buteo polyosoma

17/8

lF/1

3/1

IF. 20/10

Buteo ventralis

18/7

3/1

21/8

Buteo sp.

5/2

2/1

7/3

Geranoaetiis melanoleucus

1/1

7/1

—

8/2

Polyhorus plancus

4/1

4/1

Phalcoboenus alliogularis

5/2

1/1

6/3

Milvago chimango

—

3F/1

3F/1

Rallidae gen. et sp. indet.

—

3F/1

—

3F/1

Fulica armillata

2/2

—

2/2

Attagis malouinus

1/1

—

1/1

Tyto alba

24F, 1/5

12/3

24F. 13/8

Strix rujipes

4F/1

4F/1

Aphrastnra spinicauda

5F/2

—

5F/2

Turdus falcklandii

—

IF. 1/1

3/1

IF. 4/3

PhrygUus patagonicits

5F, 2/2

5F, 2/2

Carduelis harbata

lF/1

lF/1

Totals

96754

50F. 15/20

28/9

50F. 139/84

Number of species

16-

13

6

25

' Not including eight unidentified bones.

-Not including remains identified only to genus.

(1 978), Markgraf and Bradbury ( 1 982), and references listed in the ensuing
discussion. The forest apparently never extended as far east as Fell's Cave
(Auer, 1958). Saxon (1979) suggested that the faunal remains in Fell's
Cave "show five phases which parallel the climatological evidence" con-
cerning the post-glacial history of the region. However, Saxon (1979)
apparently assumed that the shift to patches of Nothofagus forest that
occurred at La Mision. Tierra del Fuego (Markgraf. 1980) around 8000 ybp

24 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

also occurred at Fell's Cave. Markgraf s later work (1988), however, shows
only a shift in type of grassland present at Fell's Cave during the last 12,000
years. The vegetation around Fell's Cave changed between 11,000 and
10,000 ybp from mesic grassland (<400 mm mean annual precipitation;
Pisano, 1977) with a high Gramineae component (Markgraf, 1988) and a
cool, wet climate, to an increasingly wami and dry climate interpreted as
xeric grassland (<200 mm; Pisano, 1977), with a higher proportion of
Compositae (Markgraf, 1988); the Fell's Cave area remains xeric today
(Pisano, 1977).

There is considerable disagreement on the habitat preferences of many
species of Patagonian birds (e.g., Ralph, 1985; Vuilleumier, 1985; Fjeldsa
and Krabbe, 1990). In fact, most species of birds in Patagonia are found in
a variety of habitats in the breeding season, and very few data are available
on the migratory and wintering habitats of most species. For this reason,
most species must be considered generalists and only a few species can be
taken as indicators of particular habitat types. As expected, based on the
reconstructed habitat of the area (Markgraf, 1988), none of the avian
species found at Fell's Cave is strictly associated with forest (Fig. 2). Biiteo
ventralis and Phalcohoenus alhogitkiris seem to be most closely associated
with Nothofagus forests throughout their ranges, although there are records
of both from transitional and open habitats (Clark, 1986). Nevertheless, the
apparent prehistoric abundance of 5. ventralis around Fell's Cave is unex-
pected because of its present scarcity and affinity for forests.

Most species from Fell's Cave are habitat generalists within the region
(Fig. 2), although Pterocnemia pennata. Chloephaga ruhidiceps, and
Attagis malouinus are typical of mesic to xeric grassland and probably are
never found in forests. The most dramatic shift in grassland type predates
avian remains from Fell's Cave. Relatively large numbers oi Chloephaga
picta, which is most abundant in moorland but is also found in grassy
openings within forest, were found in Fell's Cave (Fig. 2). but few were
found in Mylodon Cave and Alero del Diablo.

Although mesic to xeric steppe conditions have prevailed in the vicinity
of Fell's Cave for at least the past 1 1 ,000 years, a stream now flows through
the valley below the cave and aquatic environments were present during
the prehistoric period. This is reflected in the presence of small numbers of
several species of aquatic birds {Podiceps, Tachyeres, 3 or more species of
Anas, and Fiilica) in most of the strata in Fell's Cave from 10,720 ± 300
ybp to modern. Most of these aquatic species are associated only with
freshwater or estuarine habitats, with the exceptions of Tachyeres
patachonicus and Phalacrocorax atriceps, which are also found on marine
coasts. Both species of Chloephaga are semi-aquatic when accompanied
by young.

According to Saxon (1979), Mylodon Cave and Alero del Diablo were
covered with ice during the height of the last major glaciation (approxi-

AVIFAUNA OF CHILEAN CAVES

25

#

a
in

^

100-
90-
80-
70-
60-
50-
40-
30-
20-
10-

100—
90—

SO-
TO—
60—
50—
40—
30—
20—
10-

[T] = Steppe □ = Forest

P3 = Aquatic 0 = Generalists

Feirs Cave

a

Fell's Cave

JE

i.

Mylodon Cave Alero del Diablo

Mylodon Cave Alero del Diablo

Fig. 2. Habitat preferences of birds represented in southern Chilean caves. The
"generalists" category includes Tlierisriciis melatwpis, Chloephaga pitta. Biiteo
veiitralis. and Phalcoboemis alhogularis: see text for discussion of habitat preferences.

mately 14,000 ybp) and. following glacial retreat, were ice-free by approxi-
mately 13,500 ybp. This estimate for glacial retreat following maximal
glaciation is approximately 1000 years earlier than Mercer's (1976) esti-
mate for Seno Otway, a locality approximately 1 .5 degrees latitude south of
Mylodon Cave and slightly to the east. At any rate, the areas around both
Mylodon Cave and Alero del Diablo were ice-free well before the avian

26

UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

x:

C

O

c

3

3

-o

C

a.

-a

c
o

O

o
■5

X3
>

U

o

C/3

c
o

X)

s

3
C

C/5

1/5

ID

OJ

C

o
x>

■a

c
2

o

X3

>

ui

<

03

_o

■a

i/~,

73

_aj

%^

X5

^

.=^

c3

H

<u

c

ID

JD

t/j

f2

o

O

ON

A

o o

in m

— r-i ^

+1 +1 S

C3

-a

*"' !
5

"^

2
■a

o

O

ON ON

o
o

ir, ir,

+1 +i ^

in o

oo oo

nO 00

On --
+1 >

00

lU

4—*

T3

C

_4>
O

00

ri — r<-, ^-n — ^ — ri

r<-i "* ^ — ri — — r<-,

oc r--

t~- 00

n — n ri
m, — m, n

NO
ON

-^ ri r<-i — I
r^ <j~, 00 ri

r-j

— r> —
oi r-i n

— 00 —

— 00 —

— ~ 00 r^.

ri

r<-,

ri

5

0, ii.

r I

■>2

2 S;

ir,

oo

2 ^
5

2
S

in

:5
5

5 ~ cc ^

$D CO O

C/5

^

o

^ ^ ^ s^j -^ -^ ;_^ ^ ^ ^ ^ ■*... *-,

03 O

-c

Q

■^

Q

^

'Si

ri

'w

l;::;

2

4—1

;s

■4^

o

S:

>*,

^^c:

H

AVIFAUNA OF CHILEAN CAVES

27

r-

u

5

^

X)

rt

"3

^

o

•«

F

c

C3

00

■^

—

+1

^

ON

^^

(N

^

(N

00

+1

cn

o

00

u

r-

«

-a

O ^'
+1

'^

00

in

C3
O

o

o
o
m
m

V

+1
m

E

O

c

'o

CO

-- r-~ — t^i m

in

ri
^ in (N r^ — vTj

n

ri

ri
ri

(N

fN

ri

n

ON

nO "* r^i — ri

oo

n — (-<-, —

rn

S Q

-2

-o

t; u

&

;s 5 Q

•^

■^

a

2 Cc J^

a a

*u w ^

•-J^ ^ -^ =: ^

~ ^ >■ Ojd

t^ioc^oS^^Qit^'

c^ '^t: h. Q, U

t2

(U

>

U

c
o
•o
o

C3

o

c
a

C

o

u

XI

u

C

x:
o

X

OJ

C

T3

n

=5

(U
73

_3

c

o

c

o

C3

28 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

remains were deposited (except for the Barn Owl remains from ca. 12,500
ybp), and forest would have had time to regenerate.

At Mylodon Cave and the closely adjacent Alero del Diablo, lower
temperatures, higher rainfall, and Magellanic moorland predominated from
at least 12,500 ybp until 8000 ybp (Markgraf, 1983). By 8000 ybp,
Nothofagus forest had expanded through the area, which since has experi-
enced only minor fluctuations in the relative abundances of the three
species oi Nothofagus (Markgraf, 1983). All species found in these caves
are habitat generalists or forest/forest edge specialists, the latter being Strix
nifipes, Aphrastiira spiuicauda, Carduelis harhafa. and Phrygiliis
patagoniciis, in addition to Buteo ventralis and Phalcohocmis alhogularis,
whose habitat preferences are discussed above. Tyto alha is reputed not to
occur in deep forests (Fjeldsa and Krabbe, 1990), but was present both at
Mylodon Cave and Alero del Diablo, as well as in forested areas of the
Beagle Channel (Rasmussen et al., MS; Humphrey and Rasmussen, unpubl.
data). The only dated, identifiable bird remains from Mylodon Cave prior
to the major habitat shift there around 8000 ybp are those of Tyto alha, a
cosmopolitan habitat generalist. The rhea bone reported by Roth (1899)
would be predicted to have been deposited during the pre-8000 ybp moor-
land phase, but we do not know the provenience or whereabouts of this
specimen or the authenticity of the identification.

Because all species found at the undated Alero del Diablo are habitat
generalists (Fig. 2), they are not useful in correlation of avifauna with
habitat.

Presumed Origin of Bird Remains in the Three Caves

The absence of passerine remains at Fell's Cave may be related to the
lack of owls there, because at both of the other caves owls and passerines
occur together (Table 8), presumably as predator and prey. The remains of
waterfowl, other aquatic birds, rheas, and Attagis at Fell's Cave represent
63% of all individuals and 75% of the species found there (Table 8);
presumably these birds were hunted by humans for food. Hawks and
caracaras comprise another 37% of the bird individuals and 25% of the
species (Table 8).

Bam Owls inhabited Mylodon Cave at least intermittently from 12,496
± 148 ybp to near the present (Table 6). For part or all of this time, humans
and owls coexisted in the cave, which is large enough to have afforded
some protection for the owls. The owl bones and feathers from the cave
deposits might be accounted for either by natural mortality or by an
occasional owl being killed by humans. Although mainland Barn Owls
prey principally on small mammals, they also take small birds (Andrews,
1990); this probably accounts for the passerine remains in Mylodon Cave.
Little is known of the habits of Strix nifipes; Johnson (1965) mentioned

AVIFAUNA OF CHILEAN CAVES 29

Table 7. Distribution of avian remains at Alero del Diablo by layer (total
number of bones/MNI).

Chloephaga picta

—

Biiteo polyosonia

—

Buteo rentralis

Buteo sp.

Polyboriis plane us

l/I

Tyto alba

4/1

Turdus falcklandii

1/1

Totals

6/3

1/1

1/1

2/1

3/1

3/1

3/1

2/1

2/1

3/1

4/1

4/1

12/3

2/1

3/1

,7/3'

28/9'

Components (after Borrero et al., 1976)
A B

Species 1 2 3 Mixed' Totals'

1/1 —

3/1 1/1

4/2 1/1

' Remains in mixed layer not used in calculation of total MNIs except when they
were the only representation of a species at Alero del Diablo.

reports of its nesting in trees, but it is apparently not established whether
this species also roosts or nests in caves. The presence of a rhea (reported
by Roth, 1899) and an indeterminate rallid species in a cave is less easily
explained unless they were brought in by humans.

Prehistoric Distribution of Avifauna

The avifauna of these three caves in southern Chile suggests that, unlike
the mammalian fauna (Borrero, 1984: Markgraf, 1985), avian species
composition has changed very little in the past 12,000 years in southern
Patagonia. No extinct species were found and no range extensions were
discovered. This situation is comparable to that found in kitchen middens
along the Beagle Channel, which have a very similar avifaunal composi-
tion to that of the present time, although the middens are biased in favor of
easily captured prey species (Rasmussen et al., MS; Humphrey and
Rasmussen, unpubl. data). The stability of the avifauna evidenced by the
cave and Beagle Channel deposits, despite the predation by humans, differs
markedly from the high rates of human-caused extinction found on small,
isolated islands (Olson and James, 1982; Steadman, 1986, 1989; Balouet
and Olson, 1989; Olson, 1990). As far as is known, no avian species of
southern Patagonia have become extinct in the Holocene, but no vulner-
able, narrowly endemic species are known to have occurred in areas of high
human density. The apparent decline of the Austral Rail (Ralliis antarcticus)
is a possible exception — human-caused habitat loss has been blamed for its
rarity (Collar and Andrew, 1988), although it may simply be overlooked
(Ripley, 1977).

30

UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

-a

3

c

>

o

-a
c

a

t/j

00

t2

C3

T3
O

a.

o
Z

^

O

z

>

U

c
o
■a
o

ex

d

Z

ex

O

z

z

ex

>

u

1)

ex
ex

d
Z

c

o o o

o o o

o —
o ^

o
o

IT)

o
o

vd \d
o ^ — 1

O ro ^
rn '— <

o o o
o o o

O "

Q R " c

o o m — ■

o

o o

00

o

00

o

o

o o

Sc

U-,

O

00

o

^

f^i

-*■

O^

o'

^

00

o o
o o

R ~ C

O — ri

^ —

ri

5^

o

ir,

—

Z

ri

ri

ri

r-i

r^

>n

n

M

^

id

id

id

id

r<-,

ri

id

id

c c

1^

r~;

00

r^

^

q

r^

00

c o

^^

r^i

r^,

— '■

r*-]

^"

r-^

r^,

— '.

z

^

r<"j

s

^

(N r^i ~ ri

C/3

(U

(^1 CI

— — o

C/3

(U

o
o

,o £

■a
ex

o

'■^ .o

C3

= ■- O '*-
id '5

c
o

-a

en

a

'an

■c ^

U < [i. O U c/5

C3

AVIFAUNA OF CHILEAN CAVES 31

Most of the bird species reported from these caves have poor or nonexist-
ent fossil records. The Rheidae occur from the Pliocene through the
Pleistocene (Pascual et al., 1966; Tambussi and Tonni. 1984, 1985; Cardich et
al.. 1987; O. Odreman. pers. comm.). Very few bird bones (all unidentified)
have been excavated at Monte Verde, in south-central Chile (Dillehay. 1989)
and Las Buitreras Cave, Santa Cruz Province, Argentina (Sanguinetti de
Bormida and Borrero, 1983), and only five bird species were found at Los
Toldos, Santa Cruz (Tambussi and Tonni, 1985). Birds were not found in more
northern caves of South America, such as Huargo and Lauricocha Caves from
Peru, and La Maneta, in the Andes of Venezuela (Cardich. 1973; Pascual and
Odreman, 1973; Armand, 1985), with the exception oi Thehsticiis sp. and an
emberizid, genus indeterminate, found in Huargo (Pascual and Odreman,
1973).

Birds were apparently less important to humans of southern continental
Patagonia than they were to those of the Beagle Channel (Rasmussen et al.,
MS; Humphrey and Rasmussen, unpubl. data). This can probably best be
explained by the prehistoric abundance of large terrestrial mammals in
continental Patagonia and the lack of colonial, easily collected large birds in
this region as compared to the Beagle Channel.

Acknowledgments: E. Saxon, formerly at the University of Durham,
England, provided us the opportunity to study the specimens. We thank
authorities of the following museums for access to skeletal specimens:
American Museum of Natural History (AMNH); Museum of Vertebrate
Zoology. University of California at Berkeley (MVZ); University of Michi-
gan Museum of Zoology (UMMZ); National Museum of Natural History,
Smithsonian Institution (USNM); Peabody Museum, Yale University ( YPM).
G. E. Woolfenden loaned osteological specimens, and staff of the Division of
Birds, The University of Kansas Museum of Natural History arranged loans.
L. D. Martin, Division of Vertebrate Paleontology, The University of Kansas
Museum of Natural History, and O. Odreman. Ministerio de Energia. Merida,
Venezuela, provided valuable advice and infomiation. C. Evans. Smithsonian
Institution, and V Markgraf, INSTAAR. sent us relevant reprints. R. W.
Storer, The University of Michigan, kindly identified the grebe specimens,
and K. W. Campbell. Los Angeles County Museum of Natural History,
examined two hawk specimens. J. P. Angle, National Museum of Natural
History, arranged for C. Ross to remove wing elements from a hawk specimen.
R. Layboume and C. Dove. National Museum of Natural History, and D.
Deedrick. F.B.I. , graciously identified many of the feathers, using a
microscope provided by the U.S. Air Force. M. D. Gottfried, Calvert Marine
Museum, and H. F. James, National Museum of Natural History, improved
the manuscript, and the resumen was greatly improved by J. Casciotta, The
University of Kansas. This study was undertaken with the assistance of an
award to Humphrey from the General Research Fund of The University of
Kansas.

32 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

LITERATURE CITED

Amadon, D. 1964. Taxonomic notes on birds of prey. Am. Mus. Novit. 2166:1-24.
Andrews, P. 1990. Owls, Caves and Fossils. Chicago: Univ. Chicago Press.
Armand, J. 1985. La Maneta. Vestigios Esqueleticos de un Nino del Siglo XVH con

Manas de Posihle Antropofagia. Monogvafias 3. Merida: Consejo de

Publicaciones, Universidad de los Andes.
AuER, V. 1958. The Pleistocene of Fuego-Patagonia. Part II. The history of flora and

vegetation. Ann. Acad. Scient. Fenn. A (III) 50:1-239.
Balouet, J. C, AND S. L. Olson. 1989. Fossil birds from late Quaternary deposits in

New Caledonia. Smithsonian Contr. Zool. 469:1-38.
Bernath, E. L. 1965. Observations in southern Chile in the southern hemisphere

autumn. Auk 82:95-101.
Bird, J. 1938. Antiquity and migrations of the early inhabitants of Patagonia.

Geogr. Rev. 28:250-275.
Blake, E. R. 1977. Manual of Neotropical Birds. Vol. 1 . Spheniscidae (Penguins) to

Laridae [Gulls and Allies). Chicago: Univ. Chicago Press.
Borrero. L. a. 1984. Pleistocene extinctions in South America. Pp. 115-125 in

Rabassa. J. (ed). Quaternary of South America and Antarctic Peninsula. Vol. 2.

Rotterdam: A. A. Balkema.
Borrero, L. A., E. A. Crivelli, and G. Mengoni. 1976. Investigaciones

arqueologicas en el sitio "Alero del Diablo," Seno de Ultima Esperanza (Chile).

Ans. Inst. Pat.. Punta Arenas (Chile) 7:75-85.
Bridges, E. L. 1949. Uttermost Part of the Earth. New York: E. P. Dutton.
Browning, M.R.I 989. The correct name for the Olivaceous Cormorant, "Maiague"

of Piso( 1658). Wilson Bull. 101:101-106.
Cardich, a. 1973. Excavaciones en la Cavema de Huargo, Peru. Rev. Mus. Nac.

(Lima) 39:11-17.
Cardich, A. R.,L. L. Miotti, and A. S.Castro. 1987. Paleoindians in the Patagonian

region: Los Toldos archaeological locality, Santa Cruz Province, Argentina.

Curr. Res. Pleistocene 4:145-147.
Christie, M. I. 1984. Determinacion de prioridades conservacionistas para la fauna

de vertebrados patagonicos. Rev. Mus. Arg. Cienc. Nat. Zool. 13(56):535-544.
Clark, R. 1986. Aves de Tierra del Fuego v Caho de Homos. Gui'a de Campo.

Buenos Aires: Literature of Latin America.
Clark, W. S. 1986. What is Buteo ventralis'l Birds of Prey Bull. No. 3:115-118.
Collar, N. J., and P Andrew. 1988. Birds to Watch. The ICBP World Check-list of

Threatened Birds. ICBP Tech. Publ. No. 8. Washington, D.C.: Smithsonian

Institution Press.
Crawshay, R. 1907. The Birds of Tierra del Fuego. London: Bernard Quaritch.
DiLLEHAY, T. D. 1984. A late ice-age settlement in southern Chile. Scientific Ameri-
can 251(4): 100-109.
Dillehay, T. D. 1989. Monte Verde, a Late Pleistocene Settlement in Chile. Vol. I .

Palaeoenvironnient and Site Conte.xt. Washington, D.C.: Smithsonian Institu-
tion Press.

AVIFAUNA OF CHILEAN CAVES 33

VON DEN Dreisch. A. 1976. A Guide to the Measurement of Animal Bones from

Archaeological Sites. Camhridge. Massachusetts: Peabody Museum of Archae-
ology and Ethnology Bull. 1:1-137 pp.
Emperaire, J., AND A. Laming. 1954. La Grotte du Mylodon. J. Soc. Amer. Paris

43(1-2): 173-205.
Emperaire. J., A. Laming-Emperaire, and H. Reichlen. 1963. La grotte Fell et autres

sites de la region volcanique de la Patagonie chilienne. J. Soc. Amer. Paris

52:169-255. ^
Figuerero Torres. M. J. 1986. Biological and archaeological infomiation in copro-

lites from an early site in Patagonia. Curr. Res. Pleistocene 3:74-75.
FjeldsA, J. 1988. Status of birds of steppe habitats of the Andean zone and Patagonia.

Pp. 81-95 in Goriup, P. D. (ed). Ecology and Conservation of Grassland Birds.

ICBPTech. Publ. No. 7. Washington. D.C.: Smithsonian Institution Press.
FjeldsA. J., and N. Krabbe. 1990. Birds of the High Andes. Svendborg. Denmark:

Zoological Museum. University of Copenhagen and Apollo Books.
Hauthal, R. 1899. Resefia de los hallazgos en las cavernas de Ultima Esperanza.

Rev. Mus. La Plata 9:41 1^20.
Hayman. p.. J. Marchant. and T. Prater. 1986. Shorebirds. An Identification Guide

to the Waders of the World. Boston: Houghton Mifflin Company.
Hellmayr. C. E., and B. Conover. 1949. Catalogue of birds of the Americas. Part V.

Field Mus. Nat. Hist.. Zool. Sen 13:1-517.
Heusser. C. J. 1966. Late-Pleistocene pollen diagrams from the Province of

Llanquihue. southern Chile. Proc. Am. Philos. Soc. 110(4):269-305.
Heusser. C. J. 1984. Southernmost land-based pollen sequence from the southern

hemisphere. Abstr.. Am. Quat. Assoc. 1984:59.
Howard. H. 1929. The avifauna of Emeryville Shellmound. Univ. Calif. Publ. Zool.

32(2):301-394.
Humphrey, P. S.. and D. Bridge. 1970. Apuntes sobre distribucion de aves en la

Tierra del Fuego y la Patagonia Argentina. Rev. Mus. Arg. Cienc. Nat. Zool.

10(17):25 1-265.
Humphrey, P. S.. D. Bridge, P. W. Reynolds, and R. T. Peterson. 1970. Birds oflsla

Grande {Tierra del Fuego). Preliminary Smithsonian Manual. Washington. D.

C: Smithsonian Institution Press.
Humphrey, P. S.. and B. C. Livezey. 1982. Flightlessness in Flying Steamer-Ducks.

Auk 99:368-372.
Jehl. Jr.. J. R., and M. A. E. Rumboll. 1976. Notes on the avifauna of Isla Grande

and Patagonia, Argentina. Trans. San Diego Soc. Nat. Hist. 18:145-154.
Johnsgard. p. a. 1978. Ducks. Geese and Swans of the World. Lincoln. Nebraska:

Univ. Nebraska Press.
Johnson. A. W. 1965. The Birds of Chile and Adjacent Regions of Argentina.

Bolivia and Peru. Vols. I and II. Buenos Aires: Platt Establecimientos Graficos.
Klein, R. G., and K. Cruz-Uribe. 1984. The Analysis of Animal Bones from

Archaeological Sites. Chicago: Univ. Chicago Press.
Laming-Emperaire, A. 1972. Los sitios arqueologicos de los archipielagos de

Patagonia occidental. Ans. Inst. Pat. 3:87-96.

34 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

Lefevre, C. 1 989. L'Avifaune de Pata^onie Australe et ses Relations avec V Homme

an Cows des Six Derniers Millenaiies. Doctoral dissertation. Paris: Univ. Paris-

Pantheon-Sorbonne.
Lehmann-Nitsche, R. 1899. Coexistencia del hombre con un gran desdentado y un

equino en las cavernas patagonicas. Rev. Mus. La Plata 9:455-473.
LivEZEY, B. C. 1991. A phylogenetic analysis and classification of Recent dabbling

ducks (Tribe Anatini) based on comparative moiphology. Auk 108:471-507.
LivEZEY, B. C, AND P. S. HUMPHREY. 1984. Scxual dimoiphism in continental

steamer-ducks. Condor 86:368-377.
LoTHROP, S.K.I 928. The Indians ofTierra del Fuego. New York: Heye Foundation,

Museum of the American Indian.
Madge, S., and H. Burn. 1988. Waterfowl. An Identification Guide to the Ducks.

Geese and Swans of the World. Boston: Houghton Mifflin.
Markgraf, V. 1980. New data on the Late and Postglacial vegetational history of

"La Mision", Tierra del Fuego, Argentina. IV Int. Palynol. Conf., Lucknow

(1976-77) 3:68-74.
Markgraf, V. 1983. Late and postglacial vegetational and paleoclimatic changes in

subantarctic, temperate, and arid environments in Argentina. Palynology 7:43-

70.
Markgraf, V. 1985. Late Pleistocene faunal extinctions in southern Patagonia.

Science 228(4703): 11 10-1 11 2.
Markgraf, V. 1988. Fells Cave: 11,000 years of changes in paleoenvironments,

fauna, and human occupation. //; Bird, J. B., M. Bird (J. Hyslop) (eds). Travels

and Archaeology in South Chile. Iowa City: Univ. Iowa Press.
Markgraf, V., and J. P. Bradbury. 1982. Holocene climatic history of South

America. Striae 16:40^5.
Massone, M. 1981. Arqueologfa de la region volcanica de Pali Aike (Patagonia

meridional chilena). Ans. Inst. Pat., Punta Arenas (Chile) 12:95-124.
Mayr, E., and L. L. Short. 1970. Species Ta.xa of North American Birds. Cam-
bridge, Massachusetts: Publ. Nuttall Omith. Club No. 9:1-127.
Mercer, J. H. 1976. Glacial history of southernmost South America. Quat. Res.

6:125-166.
Moore, D. M. 1978. Post-glacial vegetation in the south Patagonian territory of the

giant ground sloth, Mylodon. Bot. J. Linn. Soc. 77:177-202.
Nami, H. G. 1987. Cueva del Medio: a significant paleoindian site in southern South

America. Curr. Res. Pleistocene 4:157-159.
Narosky, T., and D. Yzurieta. 1987. Guia Para la Identificacion de las Aves de

Argentina y Uruguay. Buenos Aires: Asociacion Ornitologica del Plata.
Olson, S. L. 1990. The prehistoric impact of man on biogeographical patterns of

insular birds. Pp. 45-5 1 //; Biogeographical Aspects of Insularity. Atti dei

Convegni Lincei 85. Rome: Accademia Nazionale dei Lincei.
Olson, S. L., and H. F. James. 1982. Fossil birds from the Hawaiian Islands:

evidence for wholesale extinction by man before western contact. Science

217:633-635.
Orquera, L. a., a. E. Sala, E. L. Piana, and A. H.Tapia. \911 .Lcmcha Packewaia:

Arqueologfa de los Canales Fueguinos. Buenos Aires: Editorial Huemul.

AVIFAUNA OF CHILEAN CAVES 35

Ortiz-Troncoso, O. 1973. Aspectos arqueologicos de la Peni'nsuia de Brunswick.
Ans. Inst. Pat., Punta Arenas (Chile) 4( 1-3): 109-130.

Pascual, R.. N. V. Cattoi, J. C. Francis, D. Gondar, E. Ortega Hinojosa, J. A.
PiSANO, A. B. DE RiNGUELET, E. ToNNi, AND J. Zetti. 1966. Vertebrata. Pp. 1-202
in Borrello. A. V. (ed), Paleontografia Bonaerense. Fasc. IV. La Plata. Argen-
tina: Comision de Investigacion Cientitica, Gobemacion de la Provincia de
Buenos Aires.

Pascual, R., and O. Odreman. 1973. Estudio del material osteologico extraido de la
Cavema de Huargo, Departamento de Huanuco, Peru. Pp. 3 1-39 /// Cardich, .A.
(ed), Excavaciones en la Cavema de Huari>(>. Peru. ApenJice II. Rev. Mus.
Nacional (Lima) 39.

Paskoff, R. p. 1977. Quaternary of Chile: the state of research. Quat. Res. 8:2-31.

Philippi B., R. a., a. W. Johnson, J. D. Goodall. and F. Behn. 1954. Notas sobre
aves de Magallanes y Tierra del Fuego. Bol. Mus. Nac. Hist. Nat. Chile 26(3): 1-,
65. -^

PiSANO, V. E. 1977. Fitogeografia de Fuego-Patagonia chilena. I. Comunidades
vegetales entre las latitudes 52° y 56° S. Ans. Inst. Patagonia, Punta Arenas
(Chile) 8:121-250.

Ralph, C. J. 1985. Habitat association patterns of forest and steppe birds of northern
Patagonia, Argentina. Condor 87:471-483.

Rasmussen, P. C. 1990. Geographic Variation and Evolutionary History of Blue-
eyed Shags of South America {Phalacrocoracidae: Phalacrocorax [Notocarho]).
Doctoral dissertation. Lawrence, Kansas: University of Kansas.

Rasmussen. P. C. 1991. Relationships between coastal South American King and
Blue-eyed Shags. Condor 93:825-839.

Rasmussen. P. C. and P. S. Humphrey. 1991. A postglacial record of the avifauna of
the Beagle Channel from Indian middens. Abstract. Joint Cooper- Wilson Orni-
thological societies meetings. Norman. Oklahoma.

Rasmussen. P C. and N. Lopez H. 1988. The status of some birds of Region X,
Chile. Bull. Brit. Ornith. CI. 108:154-159.

Rasmussen. P. C. P. S. Humphrey, and J. Muniz-Saavedra. 1992. Imperial Shags
and other birds of the Lago General Vintter area, Chubut Province. Argentina.
Univ. Kansas Mus. Nat. Hist. Occ. Pap.. 146.

Rasmussen, P. C. P S. Humphrey, and J. E. Pefaur. MS. Avifauna of a Beagle
Channel archaeological site.

Rau, J., AND J. Yanez. 1980. Cricetidos fosiles de la Cueva del Milodon. Chile
(Mammalia, Cricetidae). Not. Mens. Mus. Nac. Hist. Nat. Chile 285:9-10.

Ridgely, R. S.,and G. Tudor. 1989. The Birds of South America. Vol. I. The Oscine
Passerines. Austin: Univ. Texas Press.

Ripley. S. D. 1977. Rails of the World. A Monograph of the Family Rallidae.
Boston: David R. Godine.

Roth. S. 1899. Descripcion de los restos encontrados en la Cavema de Ultima
Esperanza. Rev. Mus. La Plata 9:421^53.

Rumboll. M. a. E. 1975. El Cauquen de Cabeza Colorada {Chloephaga rubidiceps):
una nota de alarma. Homero 1 1:315-316.

Salemme. M. C. 1990. Zooarchaeological studies in the humid pampas. Argentina.

36 UNIV. KANSAS MUS. NAT. HIST. OCC. PAP. No. 154

Pp. 309-335 in Rabassa, J. (ed), Qiiaterncny of South America and Antarctic

Peninsula. Vol. 6. Rotterdam: A. A. Balkema.
Salmi, M. 1955. Additional information on the findings in the Mylodon Cave at

Ultima Esperanza. Acta Geogr. Helsinki 14{ 19):314-333.
Sanguinetti de Bormida. a. C, and L. A. Borrero. 1983. Las Buitreras Cave and

the palaeoenvironments of the Rio Gallegos valley, Province of Santa Cruz,

Argentina. Pp. 151-156 in Rabassa, J. (ed). Quaternary of Soutf} America and

Antarctic Peninsula. Vol. 2. Rotterdam: A. A. Balkema.
Saxon, E. C. 1976. La prehistoria de Fuego-Patagonia: colonizacion de un habitat

marginal. Ans. Inst. Pat., Punta Arenas (Chile) 7:63-73.
Saxon, E. C. 1979. Natural prehistory: the archaeology of Fuego-Patagonian ecol-
ogy. Quaternaria 21:329-356.
Scott, R 1954. South America— 1953. Ann. Rept. Wildfowl Trust 6:54-69.
Sibley, C. G., and B. L. Monroe, Jr. 1990. Distribution and Ta.xonomy of the Birds

of the World. New Haven, Connecticut: Yale Univ. Press.
Siegel-Causey, D., and C. Lefevre. 1989. Holocene records of the Antarctic Shag

(Phalacrocorax [Notocarho] hransfieldensis) in Fuegian waters. Condor

91:408-^15.
Simonetti, J. A., AND J. R. Ral. 1 989. Roedores de Holoceno Temprano de la Cueva

del Milodon, Magallanes, Chile. Not. Mens. Mus. Nac. Hist. Nat. Chile 315:3-

5.
Steadman, D. W. 1986. Holocene vertebrate fossils from Isla Floreana, Galapagos.

Smithsonian Contrib. Zool. 413:1-103.
Steadman, D. W. 1989. Extinction of birds in Eastern Polynesia: a review of the

record, and comparisons with other Pacific island groups. J. Archaeo. Sci.

16:177-205.
Summers, R. W. 1986. The absence of flightless moult in the Ruddy-headed Goose

in Argentina and Chile. Wildfowl 33:5-6.
Tambussi, C. p., AND E. P. Tonni. 1984. La distribucion del genero Rhea (Aves:

Rheiformes) en el Pleistocene tardfo-Holoceno de la Region Patagonica.

Primeras Jomadas Arg. Paleontol. Vert. Resiim. p. 1 1.
Tambussi, C. P., and E. P. Tonni. 1985. Aves del Sitio Arqueologico Los Toldos,

Canadon de las Cuevas, Provincia de Santa Cruz (Republica Argentina).

Ameghiniana (Rev. Asoc. Paleontol. Argent.) 22(l-2):69-74.
Tonni, E. P. 1984. La arqueologi'a biologica el la Argentina: el estudio de los

vertebrados. ADENA, Asoc. Est. Hist.-Arq. Reg. Pamp. (Buenos Aires) 6: 11

pp. (mimeo).
Venegas C, C. 1986. Aves de Patagonia y Tierra del Fuego Chileno-Argentina.

Punta Arenas, Chile: Edic. Univ. Magallanes.
Venegas C, C, and J. Jor^ H. 1 979. Guia de Campo Para las Aves de Magallanes.

Punta Arenas, Chile: Inst. Patagonia.
Vuilleumier, F. 1985. Forest birds of Patagonia: ecological geography, speciation,

endemism, and faunal history. Pp. 255-304 //; Buckley, P. A., M. S. Foster, E. S.

Morton, R. S. Ridgely, and F. G. Buckley (eds). Neotropical Ornithology.

Ornithological Monographs No. 36. Washington. D.C.: American Ornitholo-
gists' Union.

AVIFAUNA OF CHILEAN CAVES 37

VuiLLEUMiER, F. 1991. A quantitative survey of speciation phenomena in Patagonian

birds. Omit. Neotr. 2:5-28.
Wellman, R. W. 1 972. Origen de la Cueva del Mylodon en Ultima Esperanza. Ans.

Inst. Pat., Punta Arenas (Chile) 3(l-2):97-101.
Wood, D. S., and G. D. Schnell. 1986. Revised World Inventoiy of Avian Skeletal

Specimens. 19H6. Norman, Oklahoma: American Ornithologists' Union and

Oklahoma Biological Survey.
WooLreNDEN, G. E. 1961. Postcranial Osteology of the Waterfowl. Florida State

Mas. Bull. 6:1-129.

715