AMEGHINIANA (Rev. Asoc. Paleontol. Argent.) - 40 (3): 379-385.

Buenos Aires, 30-09-2003 ISSN 0002-7014

Ecological and reproductive constraints of body

size in the gigantic Argentavis magnificens (Aves,
Theratornithidae) from the Miocene of Argentina
Paul PALMQVIST1 and Sergio F. VIZCANO2

Abstract. Several estimates of ecological and reproductive parameters are offered for the giant Argentavis
magnificens (Teratornithidae), from upper Miocene localities in Argentina. Using the allometric relation-
ship between home range and body mass observed in modern birds of prey, the home range of this extinct
raptor is estimated in 542 km2. This implies that an adult individual soaring at an altitude of 120 m would
travel around 2,168 km searching for its prey, a task that needs nearly three days flying 12 hours per day
at 68 km/h. Such estimate suggests it behaved as vultures rather than eagles. The amount of meat ingest-
ed daily by an adult individual is estimated in 5-10 kg. Clutch size is predicted in 0.78 eggs per year for a
bird this large, with an estimate for egg mass above 1000 g, suggesting one hatchling every two years. The
scaling of incubation, nestling and post-fledging periods in modern raptors provide for A. magnificens fig-
ures of 64, 230 and 190 days, respectively; the estimate of the time needed for acquiring adult plumage is
12.6 years. Annual mortality rate of this species was probably less than 2%. All these figures provide an
emerging picture about A. magnificens as a scavenger with a very low population density and a strikingly
slow turnover rate, whose individuals had very high metabolic requirements, thus needing to scan a large
home range searching for food, and invested too much energy cost and time in reproduction.
Resumen. TAMAO CORPORAL Y LIMITACIONES ECOLGICAS Y REPRODUCTIVAS DE ARGENTAVIS MAGNIFICENS (AVES ,
TERATORNITHIDAE) DEL MIOCENO DE ARGENTINA. Se realizan estimaciones de parmetros ecolgicos y repro-
ductivos de Argentavis magnificens (Teratornithidae), del Mioceno superior de Argentina. Usando rela-
ciones alomtricas en aves modernas, se estim el tamao del territorio en 542 km2. Esto implica que un
individuo solitario planeando a una altitud de 120 m debera recorrer 2168 km para encontrar una presa,
lo que le demandara aproximadamente 3 das volando a 68 km/h. Tales valores sugieren un compor-
tamiento ms similar al de los buitres que al de las guilas modernas. Se estima en 5 a 10 kg la cantidad
de carne ingerida diariamente por un adulto. El tamao de la puesta se infiere en 0,78 huevos por ao, la
masa de cada huevo en 1000 g y la frecuencia de puesta cada dos aos. Los tiempos de incubacin,
anidamiento y crianza se estiman en 64, 230 y 190 das, respectivamente, y el perodo necesario para
adquirir el plumaje adulto en 12,6 aos. La tasa de mortalidad fue probablemente inferior al 2%. La con-
sideracin de estos valores en conjunto sugiere que A. magnificens fue un ave carroera, con una baja den-
sidad poblacional, una remarcablemente baja tasa de renovacin y unos requerimientos metablicos muy
altos, por lo que necesitaba prospectar un territorio bastante amplio en busca de su alimento, invirtiendo
un alto costo energtico y tiempo en la reproduccin.

Key words. Palaeoecology. Body size. Aves. Teratornithidae. Argentavis magnificens. Miocene. Argentina.
Palabras clave. Paleoecologa. Tamao corporal. Aves. Teratornithidae. Argentavis magnificens. Mioceno. Argentina.

Introduction from three Upper Miocene localities in central and

northwestern Argentina. Although there is some
In an article recently published, Vizcano and variability in the estimates of body mass offered for
Faria (1999) reanalyse the fossil evidence available A. magnificens, with a range of values comprised be-
of the giant bird Argentavis magnificens Campbell and tween 65 and 120 kg (Campbell and Tonni, 1980,
Tonni (Aves: Ciconiiformes: Teratornithidae), known 1981, 1983; Campbell and Marcus, 1992; Campbell,
1995; Vizcano and Faria, 1999), most authors agree
that the mean size of this species would have been
1
Departamento de Ecologa y Geologa (rea de Paleontologa),
Facultad de Ciencias, Universidad de Mlaga. Campus around 80 kg, with a wingspan of 6.5-7.0 m (the
Universitario de Teatinos. 29071 Mlaga, Spain. largest extant volant birds range in mass 8-14 kg,
[email protected] with wingspans of 3.0-3.5 m). Given that A. magnifi-
2
Departamento Cientfico Paleontologa de Vertebrados, Museo
cens was the worlds largest known flying bird, with
de La Plata, Paseo del Bosque s/n, 1900 La Plata, Argentina
(CONICET). a body mass at least five times heavier than that of
[email protected] the related California condor, Gymnogyps californi-
Asociacin Paleontolgica Argentina AMGHB2-0002-7014/03$00.00+.50
380 P. Palmqvist and S.F. Vizcano

anus (Shaw), which is the largest New World vulture, clutch size, egg size, incubation, nestling and fled-
Vizcano and Faria (1999) concentrate on the flight ging periods, and time required for acquiring adult
capabilities and palaeogeographic distribution of this plumage.
extinct bird.
According to the calculations of Vizcano and Territory size and feeding in Argentavis
Faria (1999), A. magnificens had to reach a speed of
40 km/h to take off. Campbell and Tonni (1983) have Birds of prey hold home ranges whose surface is
argued that the anatomical features of the pelvis and adjusted for providing them their feeding require-
the hindlimb of teratorns, including A. magnificens, ments, which is largely determined by prey avail-
indicate that they probably were not good runners. ability. A part of this home range, from several hun-
However, given that the critical speed to take off is dred meters to a few kilometers around the nest, is
wing speed relative to the air, it could be attained by used as the territory, and most birds of prey defend
running against the wind, as many large extant birds them against their own kind (Newton, 1979). Data
do. Vizcano and Faria (1999) also suggested that compiled for extant birds of prey by Palmqvist et al.
the pampas were only part of a wide home range or (1996a) from different sources in Cramp and Sim-
territory of A. magnificens that included areas with mons (1980) showed that territory size (T, in km2) is
mountain ranges to nest and were taking off would interspecifically related in Falconiformes to mean
have been much easier. The wing shape was inferred adult body mass (M, in g) by the 0.787 power (Table
as more similar to that of large extant birds that soar 1); according to the standard error of the slope, the
relatively slowly on thermals over land than to that latter value does not differ significantly from 0.75.
of large pelagic birds that soar over water (Campbell Given that territory size is the reciprocal of optimal
and Tonni, 1983; Vizcano and Faria, 1999). This population density (i.e., number of pairs per km2),
high wing loading would have allowed it to fly in this relationship is then very close to the one expect-
moderate to strong winds that must have been pro- ed according to Damuths rule, which predicts that
hibitive for the largest known contemporary thermal population density is allometrically related to body
soarers. The high wing loading would have been ill size by the power -0.75 (Damuth, 1981, 1987).
suited to flight under poor thermal conditions, but it Using the equation shown in Table 1, adjusted for
would have been useful in slope soaring on uprising modern hawks and eagles, an estimate of 542 km2 is
air current against hillsides. Vizcano and Faria obtained for the size of the territory held by A. mag-
(1999) also showed that flapping flight was not im- nificens (i.e. approximately 23 x 23 km). Although
possible for the wing muscles of such a 80-kg bird, some caution must be observed with this estimate,
not especially for long term flights but to be used in obtained by extrapolations, because teratornithids
short bursts during take off and landing, as well as belong to the order Ciconiiformes and the data used
for sudden manoeuvres in the air. in the analysis are derived from species of a different
Apart from biomechanical constraints on flap- avian order, such figure is clearly very large, since it
ping/soaring flight in a bird this large, which are dis- implies that an adult individual could not easily
cussed in depth by Alexander (1992, 1998), Penny- roam daily a territory this large in search of its prey:
cuick (1996), Rayner (1996) and Vizcano and Faria the widest territory recorded in a living bird of prey
(1999), body size must have also involved severe lim- is that of the golden eagle, Aquila chrysaetos
itations in the ecology and reproductive behaviour of (Linnaeus) (mean adult body mass of 4.3 kg), in xeric
A. magnificens. A huge amount of empirical evidence localities of southern Spain, with occasional esti-
has been collected during the last decades indicating mates in excess of 250 km2 (Cramp and Simmons,
that body size is closely related to the autecology and 1980), although the average territory size of this
function of organisms, including many aspects of species is 61 km2 (Palmqvist et al., 1996a).
their physiology, metabolism, growth and behaviour, The largest eagles, whose visual acuity is equiva-
plus a plethora of related phenomena in their life his- lent to that of a person equipped with a 8x binocu-
tories, evolution, and several synecological proper- lars, fly searching for their prey at heights comprised
ties, such as population density, community size between 60 and 120 m, while raptors of smaller size
structure and predator-prey relationships (Peters, usually fly at lower altitudes, typically 20-60 m
1983; Calder, 1984; Damuth, 1981, 1982, 1987, 1991; (Cramp and Simmons, 1980). Soaring at 120 m above
Palmqvist et al., 1996a). Due to these reasons, body the ground A. magnificens could scan a maximum
size is probably the best predictive tool for unravel- land strip of around 250 m, which implies that it
ling many issues related to the palaeoautecology of would require four passings to cover each km2 of the
an extinct species like A. magnificens, such as its home territory, thus needing to fly a distance of approxi-
range, daily food requirements and annual mortality, mately 2,168 km per day for scanning the whole sur-
as well as several breeding parameters, including face of its territory. The soaring velocity of the largest
AMEGHINIANA 40 (3), 2003
 Ecology and reproduction of Argentavis 381

modern raptors is 30 to 50 km per hour. Such speed It is worthwhile to address that these are still con-
is maintained with minimum power requirements, servative hypotheses. In our analysis we have as-
and is allometrically scaled to body mass by the 0.15 sumed that the animal flew indefinitely at the same
power (Peters, 1983). Given that a 10 kg vulture level and did not loose time gaining height soaring in
reaches an optimal velocity of 50 km/h, this implies circles in ascending warm air columns (i.e. thermals).
that A. magnificens could soar at around 68 km/h, We reckon that this is not true, but we cannot know
thus needing approximately 32 hours for the task of the speed of the rising air within the thermal, how
searching the whole surface of its territory (i.e., near- frequent they were, and other aerodynamic features
ly three days for travelling a foraging distance of of A. magnificens needed to make this calculations.
more than two thousand kilometers, considering 12 Hence, the overall time needed to go over the whole
hours per day devoted to soaring flight, which is the territory searching for preys must have been close to
maximum period observed in modern raptors). On (or even in excess of) four days.
the other hand, the energy expenditure of A. magnifi- The relationship between the amount of meat in-
cens would have been greater than that of the largest gested daily by captive raptors (which is an indicator
living birds of prey, since transport costs rise with of their basal metabolic requirements; data from
body mass (Peters, 1983). Therefore, the strikingly Kendeigh, 1970; Kirkwood, 1981) and body mass is al-
large estimate obtained for the territory size that lometric and positive, with a value of 0.721 for the ex-
should be hold by a bird of prey this large allows to ponent (i.e., close to 0.75, as predicted by Kleibers
discard the possibility that the predatory behaviour law for metabolic requirements and size; Kleiber,
of A. magnificens was similar to that of modern ea- 1932, 1975). After the equation provided in Table 1, a
gles, which catch and consume their own prey bird the size of A. magnificens would have needed
whole, as it has been proposed previously (Campbell 2,404 g of meat per day to survive. Given that the
and Tonni, 1981). basal metabolic requirements are ca. 25-50% of the
The speed mentioned above is the air speed, but daily field expenditure (Peters, 1983), this estimate
this can be greatly affected by the wind, changing the implies that an adult individual must have consumed
final time to cover some distance on the ground. around 5-10 kg of meat per day; such quantity would
Campbell and Tonni (1983) argued that the constant have even increased substantially during the breed-
westerlies blowing over that part of South America ing season, because estimates for Rppells vultures
during the Miocene would have been important for (Houston, 1976) indicate that food requirements rise
A. magnificens to take off. With the available data, the up to three times during the breeding season. Despite
influence of wind on the final time to scan an area this important constraint on the palaeobiology of A.
can be estimated in the way an aircraft pilot can do it magnificens, there are however important metabolic
using the following equation: advantages derived of the large body size reached by
this bird of prey: given that mass-specific metabolic
Vg= Va.cos{asin[Vw/Va.sin(2..Aw/360)]}+Vw.cos (Aw/360) rates decline as the inverse of the fourth root of body
in which Vg is the ground speed, Va is the air speed, mass (Damuth, 1981, 1987; Peters 1983), this raptor
Vw is the wind speed, and A w is the wind angle. presumably could survive much longer on its energy
stores than could smaller ones, thus taking longer to
We have recalculated the time to cover an squared starve. In addition, it would have been less adversely
territory, assuming a wind speed of 40 km/h blowing affected by low temperatures in metabolic-cost terms,
constantly from different angles in relation to the its body temperature probably changed more slowly
flight direction. This speed, the minimum speed than that of smaller species experiencing the same
needed to generate lift, is normal for Southern thermal gradient, and a given energy supply would
Patagonia today, an environment comparable to that have supported a much greater biomass of A. magnif-
proposed by Campbell and Tonni (1983) as the habi- icens than of smaller raptors.
tat of A. magnificens. The figures obtained range from
38 h 35 min 8.5 s (i.e. 3 days and 2 h 30 min 20.5 s)
Breeding behaviour of Argentavis
when the wind blows completely laterally (90) to 46
h 37 min 26 s (i.e. 3 days and 10 h 35 min 2 s) with the There are three main trends in the breeding be-
wind blowing in the direction of flight most of the haviour of birds of prey related to body size
time (i.e directly from the back in one way and from (Newton, 1979): the larger the species, the later the
the front in the other). That implies that it would be age at which breeding begins, the longer each suc-
strategically more convenient to fly most of the time cessful attempt takes, and the fewer young produced
on a South-North than on an East-West axis. In this with each attempt. As a result, large, long-lived
way the animal would save almost one day of work species have relatively small eggs in relation to body
to cover the same area. size, single-egg clutches, protracted breeding cycles,
AMEGHINIANA 40 (3), 2003
382 P. Palmqvist and S.F. Vizcano

and deferred maturity. Such trends reflect the greater figures suggest that the adults elapsed 484 days (i.e.
advantage that large species gain from reducing the approximately 16 months) between laying the eggs
risk and energy expended in any one breeding at- and emancipation of the young, which probably im-
tempt in the interests of improved survival for future plies that the egg was incubated during the winter
attempts, and a generally greater difficulty that large season and blooming happened in spring, being the
species have in obtaining food and/or in digesting chick nourished in the nest until mid autumn. Such a
and processing it, because metabolic rate slows with prolonged reproductive season indicates clearly that
increase in body size. All these limitations must have breeding took place every two years, in agreement
posed several constraints on the reproductive behav- with the low estimate obtained for clutch size.
iour of A. magnificens. Death is caused in modern birds of prey by star-
Usual clutch size (N, in number of eggs per year) vation, diseases, parasites, predation (mostly by oth-
is allometrically related to body mass in raptors by er raptors or owls) or accidents. Annual mortality
the power -0.273 (Table 1), which provides for A. rates scale allometrically with adult body mass by
magnificens an estimate of 0.78 clutched eggs per year, the -0.565 power (Table 1), which gives an estimate of
a figure suggesting one (or occasionally two) hatch- only 1.91% for a bird as large as A. magnificens.
ling every two years. Egg mass is scaled to female
weight in modern birds of prey by the 0.662 expo- Discussion
nent, thus predicting a figure slightly larger than
1000 g in the case of A. magnificens. The mass of each Our data suggest that A. magnificens was a bird of
individual egg, expressed as a percentage of female prey with both low population density and mortality
weight, is negatively correlated to body size of adults turnover rate, whose individuals had very high feed-
by the power -0.332, which gives an estimate of only ing requirements and thus required a strikingly large
1.36% for A. magnificens. Incubation and nestling pe- territory, investing too much energy cost and time in
riods are positively related to body mass by the 0.137 reproduction.
and 0.354 powers, which provide estimates of 64 and Larger animals tend to have fewer individuals in
230 days, respectively; post-fledging period of chicks each species, which represents a risk for the long
is scaled to female size by the exponent 0.370, indi- term viability of their populations (Alexander, 1998).
cating 190 days for a bird as large as A. magnificens. As indicated above, larger extant raptors have long
Finally, the estimate of the time needed for acquiring breeding cycles and small clutches, due to three rea-
adult plumage, which is related in extant raptors to sons (Newton, 1979): (i) large, long-lived species gain
body mass by the 0.360 power, is 12.6 years. These comparatively less by investing heavily in any one
Table 1. Least squares allometric curves describing the relationship between adult body mass (M, in g) and several ecological/breeding
parameters (Y) measured in extant raptors (Y = aMb)1. / Curvas alomtricas por mnimos cuadrados que describen la relacin entre la masa cor-
poral de los ejemplares adultos (M, en g) y diversos parmetros ecolgicos y reproductivos (Y) estimados en aves de presa actuales (Y = aMb)1.

Prediction for
Parameter (Y) log(a) (s.e.) b (s.e.) s.e.e. r F p N Argentavis
(p < 0.05 interval)
territory size (km2 ) -2.590 (+0.929) 0.787 (+0.133) 0.616 0.820 34.9 <0.001 19 542 (162, 1812)
meat ingested
daily (g) -0.359 (+0.072) 0.721 (+0.230) 0.282 0.931 304.4 <0.001 48 2,404 (1,917, 3,287)
usual clutch
size (N) 2.839 (+0.254) -0.273 (+0.037) 0.335 0.678 55.4 <0.001 67 0.78 (0.41, 1.51)
egg mass(g) -0.515 (+0.113) 0.662 (+0.016) 0.149 0.981 1659.0 <0.001 67 1,052 (786, 1,409)
Egg mass as %
of female mass 4.058 (+0.115) -0.332 (+0.017) 0.151 0.928 402.2 <0.001 67 1.36 (1.01, 1.83)
Incubation period
(days) 2.619 (+0.084) 0.137 (+0.012) 0.104 0.826 131.0 <0.001 63 64.4 (55.5, 79.0)
Nestling period
(days) 1.442 (+0.149) 0.354 (+0.021) 0.186 0.904 273.0 <0.001 63 230.1 (164.6, 331.3)
post-fledging period
(days) 1.070 (+0.396) 0.370 (+0.058) 0.407 0.707 41.0 <0.001 43 190.0 (85.6, 422.0)
Adult plumage
(years) -1.528 (+0.242) 0.360 (+0.035) 0.292 0.821 107.5 <0.001 54 12.6 (9.4, 16.9)
annual adult
mortality (%) 7.024 (+0.617) -0.565 (+0.093) 0.295 0.870 32.3 <0.001 14 1.91 (1.42, 2.56)
1
Data obtained from Kendeigh (1970), Newton (1979), Cramp and Simmons (1980), Kirkwood (1981) and Palmqvist et al. (1996b). The
equation for territory size was obtained with data from predator species; vultures and insectivores were excluded, because these species
show comparatively smaller territories than those of birds of prey of similar size (Palmqvist et al., 1996a). Abbreviations: log(a): Y-inter-
cept, b: slope, s.e.e.: standard error of estimate, r: Pearson correlation coefficient, F: F-test, p: probability, N: number of cases.

AMEGHINIANA 40 (3), 2003

 Ecology and reproduction of Argentavis 383

Figura 1. Reconstruction of kleptoparasitism (i.e. aggressive scavenging) by the giant bird Argentavis magnificens on the prey (a large un-
gulate carcass) of thylacosmilid sabre-tooths. Although modern vultures do not steal the prey from large carnivores, A. magnificens might
have used its large size for intimidating these marsupial predators and taking their food, as often do spotted hyaenas (Crocuta crocuta
Kaup) with the prey of wild dogs (Lycaon pictus Brookes). Given the high energetic cost of hunting (twenty-five times the basal metabo-
lic rate for wild dogs; Gorman et al., 1998), the loss of prey to kleptoparasites would have had a large impact on the amount of time that
thylacosmilids devoted to hunting activities in order to achieve their energy balance. Drawn by Nstor Toledo. / Reconstruccin del com-
portamiento cleptoparsito (carroeo agresivo) del ave gigante Argentavis magnificens sobre las presas (cadveres de grandes ungulados) de los ti-
lacosmlidos con dientes de sable. Aunque los buitres modernos no roban las presas de los grandes carnvoros, A. magnificens pudo haber usado su
gran tamao para intimidar a estos predadores marsupiales y tomar su comida, como suelen hacer las hienas manchadas (Crocuta crocuta Kaup) con
las presas de los perros salvajes (Lycaon pictus Brookes). Dado el alto costo energtico de la caza (unas veinticinco veces la tasa metablica basal en
los perros salvajes; Gorman et al., 1998), la prdida de las presas robadas por los cleptoparsitos pudo tener un gran impacto sobre la cantidad de tiem-
po que los tilacosmlidos dedicaran a las actividades de caza con vistas a satisfacer su balance energtico. Dibujo de Nstor Toledo.

breeding attempt, and natural selection favours a competing successfully with others of their species
low reproductive effort in any one season in the in- (i.e. k-selection). Due to these reasons, population
terests of better chances to breed in several future turnover is generally low in large species, with more
seasons, which also favours deferment of maturity, overlap between generations and a more stable age
giving time for useful experience to be gained; (ii) the structure, all of which tend to dampen fluctuations in
metabolic rate in birds slows with increase in body numbers. There are also trends to be a relatively large
size, and so the lower breeding rates of the largest non-breeding population, consisting mainly or en-
raptors are due partly to their slower metabolism, a tirely of immatures, and less than half the total pop-
given amount of food taking longer to digest and ulation breeds in any one year, producing only a
convert to egg or body tissue, which accounts for the small number of young. All these trends seem to
relatively smaller eggs, longer laying intervals, and have been carried to an extreme in A. magnificens.
longer growth periods of the larger species; and, fi- In certain large condors and tropical eagles, each
nally, (iii) large size confers not only a longer poten- breeding cycle lasts more than one year, and success-
tial lifespan, but also a greater immunity from preda- ful pairs breed no more than once in two years. These
tion and an increased ability to survive temporary raptors resemble in their population dynamics cer-
food shortages. Hence, other things being equal, the tain seabirds with small clutches, long breeding cy-
larger the bird, the more consistently is its population cles, and deferred maturity. In both groups, single-
likely to remain close to the level that the environ- egg clutches are frequent and, when two eggs are
ment will support (i.e. the carrying capacity of the laid, often only one young is raised. Long post-fledg-
ecosystem); under these conditions, reproduction ing periods, in which the juveniles are fed near the
generally will be difficult, and vacancies in the breed- nest, occur in some tropical seabirds; moreover, the
ing population will be few at any given time. For only other flying birds whose complete breeding cy-
large species, then, selection pressure will favour the cles are known to last more than one year are some
production of well-nurtured juveniles released from albatrosses and frigate birds, with periods of up to
prolonged parental care with the greatest chance of ten years to reach maturity. This evidence offers reli-
AMEGHINIANA 40 (3), 2003
384 P. Palmqvist and S.F. Vizcano

ability for the figures presented above concerning the modern puma, Puma concolor (Linnaeus), showing a
breeding parameters of A. magnificens. highly specialised craniodental design, similar to that
The estimates of territory size and metabolic re- of the most advanced machairodontines (e.g.
quirements of A. magnificens are more realistic if we Smilodon and Megantereon) and nimravines (e.g.
envisage it as a scavenger rather than as a predator. It Barbourofelis), which includes: (i) a well developed,
is difficult to obtain reliable estimates of territory curved backwards symphyseal region in the
sizes in vultures, since many species show gregarious mandible, which protected the extremely elongated
behaviour, loosely colonial nesting, and use commu- upper canines from lateral bending; (ii) a lowered
nal home ranges, usually with no evidence of re- glenoid fossa, a reduced height of the coronoid
stricted feeding areas used; all this implies that no process, a laterally shifted angular process, and a
obvious territory is held (and defended) in most cas- shortened zygomatic arch, features which allow a
es. New World vultures and condors usually nest in wide gape; and (iii) a lowered and ventrally extend-
cliffs well apart from one another, but all they roost ed mastoid process, which is enormous relative to
communally and feed in groups. However, data on modern felids, thus indicating that the neck muscles
population densities and the scarce evidence avail- were correspondingly large, which suggests that a
able for those species which maintain individual head-depressing motion was involved in the pene-
home ranges (Cramp and Simmons, 1980), such as tration of the canines inside the body of the prey.
the lapped-facet vulture (Torgos tracheliotus Forster; Sabre-toothed predators were able to hunt very
mass of adults around 7 kg, territory size of 43 km2) large prey relative to their own size, and they left on
and the Egyptian vulture (Neophron percnopterus the carcasses of the ungulates hunted large amounts
Linnaeus; 2 kg, 12 km2), indicate that vultures hold, of flesh and all nutrients within bone, which could be
on average, a territory size two or three times small- subsequently scavenged by other carnivores (Arribas
er than that of an eagle of the same size (Palmqvist et and Palmqvist, 1999; Palmqvist et al., 2003).
al., 1996a). This provides an estimate of approximate- Therefore, those marsupial predators were also like-
ly 180-260 km2 for a bird the size of A. magnificens, a ly to have left only partially eaten carcasses of large
figure that can be considered more realistic than the herbivores, as the notoungulate Pisanodon nazari
previous one, as it is actually covered by modern (Cabrera and Kraglievich), the litoptern Proma-
birds of prey. On the other hand, vultures often fly at crauchenia sp., the ground sloths Plesiomegatherium
larger altitudes than eagles when searching for food, sp. and Elassotherium altirostre Cabrera, and even the
typically at 100-200 m in the case of griffon vultures glyptodonts Aspidocalyptus castroi Cabrera, Hoplo-
(Gyps fulvus Hablizl), but up to some thousand me- phractus tapinocephalus Cabrera, Coscinocercus mar-
tres in other species (Cramp and Simmons, 1980). calaini Cabrera, Coscinocercus brachyurus Cabrera and
Vultures can see objects 4-8 cm diameter from heights Plohophorus araucanus Ameghino. Given the extreme-
of 1000 m. Although they are usually attracted to car- ly slicing condition of thylacosmilid teeth, the car-
casses by sight, most vultures often watch the move- casses of such prey must have provided rather huge
ments of other neighbouring birds on ground or in amounts of flesh and intact, marrow-rich bones to
air, which facilitates the search for food at higher al- scavengers, thus opening a new ecological niche that
titudes, thus allowing them to scan a wider surface allowed the evolution of a vulture as large as A. mag-
than that of eagles and travelling their territory in a nificens in South America during late Miocene times
shorter time. (figure 1).
There is a potential candidate for supplying large
enough food as that required by a 80 kg vulture dur- Acknowledgments
ing late Miocene times in Argentina. Palmqvist et al.
(1996b), Arribas and Palmqvist (1999) and Palmqvist We are grateful to Richard A. Faria, Walter Bock, Jorge I.
and Arribas (2001) have proposed for a fossil fauna in Noriega and David Steadman for helpful comments which im-
proved the original manuscript. Ernesto Fasciolo helped us in
the Old World (the lower Pleistocene assemblage of some calculations. This research has been financed in part by pro-
large mammals from Venta Micena, Orce, Spain) a jects PB97-1082, BOS2001-3888 and Pict 7-6348.
similar ecological relationship between the sabre-
toothed felids Homotherium latidens (Owen) and
Megantereon whitei (Broom), and the giant, strictly References
scavenging hyaena Pachycrocuta brevirostris Alexander, R. McN. 1992. Exploring Biomechanics: Animals in
(Aymard). Indeed, in the Miocene South American Motion. Scientific American Library, New York, 247 pp.
fauna to which A. magnificens belonged, the marsupi- Alexander, R. McN. 1998. All-time giants: the largest animals and
their problems. Palaeontology 41: 1231-1245.
al sabre-tooth Achlysictis lelongii Ameghino (Bo- Arribas, A. and Palmqvist, P. 1999. On the ecological connection
rhyaenoidea, Thylacosmilidae) is found. This top between sabre-tooths and hominids: Faunal dispersal events
predator reached approximately the body size of a in the lower Pleistocene and a review of the evidence for the

AMEGHINIANA 40 (3), 2003

 Ecology and reproduction of Argentavis 385
first human arrival in Europe. Journal of Archaeological Science Kirkwood, J. K. 1981. Manteinance energy requirements and rate
26: 571-585. of weight loss during starvation in birds of prey. In: J.E. Cooper
Calder, W. A. 1984. Size, Function, and Life History. Harvard and A.G. Greenwoods (Eds.), Recent Advances in the Study of
University Press, Cambridge, Massachussetts, 431 pp. Raptor Diseases, Chiron Publications, Keighley, pp. 153-157.
Campbell, K. E. 1995. Additional specimens of the giant tera- Kleiber, M. 1932. Body size and metabolism. Hilgardia 6: 315-353.
torn, Argentavis magnificens, from Argentina (Aves: Terator- Kleiber, M. 1975. The Fire of Life: An Introduction to Animal
nithidae). Courier Forschungsinstitut Senckenberg 181: 199- Energetics. Krieger Publishing Company, New York, 453 pp.
201. Newton, I. 1979. Population Ecology of Raptors. T. & A. Poiser,
Campbell, K. E. and Marcus, L. 1992. The relationship of hindlimb Hertfordshire (Reino Unido), 399 pp.
bone dimensions to body weight in birds. In: K.E. Campbell Palmqvist, P. and Arribas, A. 2001. Taphonomic decoding of the
(ed.), Papers in Avian Paleontology Honoring Pierce Brodkorb, paleobiological information locked in a lower Pleistocene as-
Science Series Natural History Museum of Los Angeles City semblage of large mammals. Paleobiology 27: 512-530.
36, pp. 395-412 Palmqvist, P., Palomo, L. J., Prez-Claros, J. A and Vargas, J. M.
Campbell, K.E. Jr. and Tonni, E. P. 1980. A new genus of teratorn 1996a. Relacin entre peso corporal, tamao del territorio,
from the Huayquerian of Argentina (Aves: Teratornithidae). tamao de puesta y tiempo de desarrollo en algunas rapaces
Contributions in Science of the Natural History Museum of Los del Palertico occidental. Boletn de la Real Sociedad Espaola de
Angeles County 330: 59-68. Historia Natural (Seccin Biolgica) 92: 47-54.
Campbell, K.E. Jr. and Tonni, E. P. 1981. Preliminary observations Palmqvist, P., Martnez-Navarro, B. and Arribas, A. 1996b. Prey
on the paleobiology and evolution of teratorns (Aves: selection by terrestrial carnivores in a Lower Pleistocene
Teratornithidae). Journal of Vertebrate Paleontology 1: 265-272. paleocommunity. Paleobiology 22: 514-534.
Campbell, K.E. Jr. and Tonni, E. P. 1983. Size and locomotion in Palmqvist, P., Grcke, D.R., Arribas, A. and Faria, R. 2003.
Teratorns (Aves: Terathornitidae). The Auk 100: 390-403. Paleoecological reconstruction of a lower Pleistocene large
Cramp, P. and Simmons, K. E. L. (eds.) 1980. The Birds of the mammals community using biogeochemical (13 C, 15N, 1 8O,
Western Palearctic, Vol. II: Hawks to Bustards. Oxford University Sr:Zn) and ecomorphological approaches. Paleobiology 29: 205-
Press, Oxford, 695 pp. 229.
Damuth, J. 1981. Population density and body size in mammals. Pennycuick, C. J. 1996. Stress and strain in the flight muscles as
Nature 290: 699-700. constraints on the evolution of flying animals. Journal of
Damuth, J. 1982. Analysis of the preservation of community struc- Biomechanics 29: 577-581.
ture in assemblages of fossil mammals. Paleobiology 8: 434-446. Peters, R. H. 1983. The Ecological Implications of Body Size.
Damuth, J. 1987. Interspecific allometry of population density in Cambridge University Press, Cambridge.
mammals and other animals: the independence of body mass Rayner, J. M. B. 1996. Biomechanical constraints on size in flying
and population energy-use. Biological Journal of the Linnean vertebrates. Symposium of the Zoological Society of London 69: 83-
Society 31: 193-246. 109.
Damuth, J. 1991. Of size and abundance. Nature 351: 268-269. Vizcano, S. F. and Faria, R. A. 1999. On the flight capabilities and
Gorman, M.L., Mills, M.G., Raath, J.P. and Speakman, J.R. 1998. distribution of the giant Miocene bird Argentavis magnificens
High hunting costs make African wild dogs vulnerable to (Teratornithidae). Lethaia 32: 271-278.
kleptoparasitism by hyaenas. Nature 391: 479-481.
Houston, D. C. 1976. Breeding of the white-backed and Rppels
griffon vultures Gyps africanus and G. rueppellii. Ibis 118: 474-
488.
Kendeigh, S. C. 1970. Energy requirements for existence in rela- Recibido: 26 de junio de 2002.
tion to size of bird. Condor 72: 60-65. Aceptado: 14 de noviembre de 2002.

AMEGHINIANA 40 (3), 2003

 Reunin Anual de Comunicaciones de la
Asociacin Paleontolgica Argentina
y
Simposio de Tafonoma y Paleoecologa

Santa Rosa 27, 28 y 29 de noviembre de 2003

Comisin organizadora
Silvio Casado
Edsel Brussa
Claudia Montalvo
Ana Parras
Miguel Griffin
Ricardo Melchor
Marcelo Zrate

Edificio Central de la Universidad Nacional de La Pampa

Saln Azul, primer piso
Coronel Gil 353, Santa Rosa, La Pampa
Telfono: 02954-436787, int. 13
[email protected]

AMEGHINIANA 40 (3), 2003