NOTES
Abstract: According to central place foraging theory, adult Patagonian gray foxes (Pseudoalopex ,griseus) behaving efficiently should consume on the spot those food items too large to be carried and those too small to warrant being carried to the den. Consequently, fox cubs should be provisioned with intermediate-sized items. We tested this prediction by analyzing 96 fecal samples from adult foxes and 82 fecal samples from cubs belonging to three different dens collected near Junin de los Andes in the province of Neuqudn, Argentina, from November 1994 to January 1995. As predicted, adult foxes consumed carrion of large ungulates (large items) and arthropods and berries (small items) significantly more often than cubs 416. Cubs ate more intemiediate-sized vertebrates, from lizards and birds to rodents and hares. Medium-sized armadillos, which are difficult to handle and transport, were consumed only by adults. Cubs ate some insects, which were probably captured by the cubs themselves near the den.
Résumé : D’après l’hypothdse de la quête de nourriture autour d’un point central, on peut penser que, pour étre efficaces, les renards de Patagonie (Pseudoalopex griseus) consommeront sur place les alimenis trop gros pour être transportés ou trop petits pour qu’il vaille Ia peine de les transporter. En poursuivant ce raisonnement, il est permis de croire que la nourriture apportée aux jeunes renards dans Ia taniére sera constituée d’aliments de taille intermddiaire. Nous avons éprouvé cette hypothèse en analysant 96 échantillons de feces d’adultes et 82 échantilLons de feces de jeunes provenant de trois tanidres diffdrentes, prés de Junin de los Andes, province de Neuqudn, en Argentine, de novembre 1994 a janvier 1995. Tel que prévu, les adultes ont consommé significativement plus souvent des charognes de gros ongulés (aliments de grande taille) ainsi que des arthropodes et petits fruits (petits aliments) que les jeunes. Les jeunes ont consommé plus de vertdbrés de taille intermédiaire, ldzards, oiseaux, rongeurs et liCvres. Lcs tatous de taille moyeime, difficiles a manipuler et a transporter, ont dté consommds exclusivemer,t par des adultes. Les jeunes ont mange des insectes qu’ils ont probablement attrapés eux-mémes aux abords de leur taniCre.
[Traduit par la Redaction]
Introduction
Central place foraging (CPF) theory predicts that a single- prey carrier behaving optimally should cosis’arne osi the spot food items that it could not economically bring to its feeding place (Orians and Pearson 1979; Schoener 1979). This the- my, frequently used to analyze patterns of food collection by nesting birds (e.g., Kacelnik 1984), may also be applied to breeding foxes, which usually bring individual food items to the den to feed their cubs (Lovari and Parigi 1995).
Since Murie (1936, 1944), several authors (Englund 1969;
Harris 1981; Rau et al. 1985) have claimed that adult red
Received December 15, 1997. Accepted July 31, 1998. S.C. Zapata, A. Travaini, and M. Delibes? EstaciOn
Bioldgica de Doflana, Consejo Superior de Investigaciones
Cientificas, Pabelh5n de Peni, Avenida Ma. Luisa s.n., 41013
Sevilla, Spain.
‘Author to whom all correspondence should be addressed
(e-mail: ).
foxes (Vulpes vulpes) seem to prefer to carry to their cubs bulkier prey items than those they consume themselves. Likewise, it may be expected that very large food items,
such as ungulates taken as carrion, should also be consumed on the spot, as they cannot be carried away. Hence, we pre dict that during the rearing period, adult foxes should con sume either larger or smaller food items than their cubs, which should be provisioned with intermediate-sized food items.
To test our prediction, we studied the diets of adults and
cubs of the Patagonian gray fox (Pseudoalopex grireus). This is a small fox (approximately 3 kg in the study area; unpublished data) inhabiting plains, pampas, deserts, and low mountains (Ginsberg and Macdonald 1990) from Atacama (northern Chile) and Santiago del Estero (north eastern Argentina) to Tierra del Fuego (Cabrera 1958; Wil son and Reeder 1993). Its feeding habits have been studied chiefly in Chile, where it is considered an opportunistic feeder that hunts prey, mainly rodents, according to their availability (Medel and Jaksic 1988). Up to now, detailed in formation on the diet of gray fox cubs in the den has been
Table 1. Frequency of occurrences of prey in the feces of adults and cubs of the Patagonian gray fox (Pseudoalopex griseus).
Adults Cubs
(a = 96 fecal (it = 82 fecal
Food item samples) samples)
Mammals
Order Xenarthra
those of adults by their very small size (moreover, incorrect assign ment of adult feces to cubs should be conservative, as our nul.t hypothesis is that adults and cubs consume the same types of food items).
The components of the feces were separated by Land. Mammals were identified to the species level, based on hair and bone re mains. Food items were arbitrarily assigned to three groups: large and (or) difficult to carry (ungulates, armadillos), intermediate- sized (hares, rodents, birds, reptiles), and small (insects, berries).
Zaidyus pichiy (1.2)
Order Artiodactyla Cervus elaphus (>20) Ovis aries (>20)
Order Lagornorpha
Lepus europaeus (<3)
Order Rodeatia
Oryzomys longicaudarus
(0.03)
Auliscomys mkropus (0.07)
Eliginodontia typus (0.02)
!renornys tarsalis (0115) Ph9llotis danvini (0.06) Rèithrodon auritus (0.08) Akodon sp. (0.03)
Geoxus valdivio.nus (0.03) Galea musteloides (0.25) Ctenomys haigi (0.15) Unknown
Birds (<0.5) Lizards (<0.05) Arthropods (<0.01)
Berries (<0.01)
8
6
10 24
1
2 7
2 7
2 1
4 12
9 7
20 18
2
8 1
5
8 21
5 —
45 24
39 —
To reduce biases, diet results were expressed in two ways: (1) per cent occurrence and (2) percent dry mass of each food type in fe ces. We assume that this provides a good representation (Corbett
1989) uf the food items consumed directly by adults or brought to
the den and consumed by the cubs.
Results
Food items consumed by adult gray foxes in Catdn-Lil in cluded mammals (at least 14 species), birds, reptiles, arthro pods, and berries (Table 1). Rodents were the best represented group (31.1% of all occurrences and 33% of dry mass in fe ces), followed by berries (22 and 22.3%, respectively) and asthropods (25.4 and 13.7%, respectively). Carrion was rep resented by the remains of sheep, deer, and perhaps arma dillos (8.5% of occurrences and 19.9% of dry mass overall). Lagomorphs, birds, and lizards were of less importance (be low 6% of occurrences and 7% of dry mass each).
The list of food items brought for cubs was less diverse,
including mammals (nine species), birds, reptiles, and arthro pods (Table 1). As with adults, rodents were the main prey of cubs (46.1% of all occurrences and 57.1% of dry mass in feces). Lagomorphs (18.8 and 22%, respectively), birds (16.4 and 13.2%, respectively), and arthropods (18.8 and 7.7%, re
Total 177 128
Note: The numbers in parentheses show the approximate mass (kg) of
food items.
lacking, although Johnson and Franklin (1994) reported that both parents provisioned pups with mice, hares, birds, and duck eggs.
Study area and methods
‘The study was carried out from November 1994 to January 1995 on Catán-Lil ranch, 50 km northeast of Junin de Los Andes in the province of Neuquén, Argentina (40°S, 71°W). The area belongs to the Patagonian phytogeographic province Western District (Cabrera 1976), and is characterized by great plains 800—900 m high, dissected by steep, rugged valleys. The vegetation is a naixed steppe of grass and shrubs dominated by Mullium spin osurn, Senecio sp., Stipa sp., and Faa sp. The mammalian fauna is char ac.teristic of the ch-y Patagonian shrub—steppe (Pearson and Pearson
1982), but introduced European brown hares (Lepus europaeus) and, to a lesser extent, European red deer (Cervus elaphus) and wild boar (Sus scrofa), are common. The annual mean temperature is 17.4’C, and the annual mean rainfall is 360 mm, mostly falling during winter (June—August).
Ninety-six fecal samples from adult foxes were collected in an area of approximately 4000 ha, while 82 fecal samples from cubs were collected in three different dens separated by more than 1 km within the same area. Cubs were born in October, and in January they were still living in the den or its surroundings (Johnson and Franklin 1994), where their feces can be easily distinguished from
spectively) were the next most prevalent items in the diet.
No carrion or berries were detected in cubs.
Percent occurrence and percent dry mass in feces of small,
intermediate-sized, and large food items (as previously de
fined) differed in the diets of adults and cubs (0 test, p
0.001) (Fig. 1). Cubs consumed more intermediate-sized food items, including birds, rodents, aM lagomorphs, and fewer large (ungulates) and small food items (berries and arthropods). Interestingly, the medium-sized armadillo (Z2e- dyus pichiy), which is difficult to handle and carry away, was eaten only by adults.
Discussion
We did not attempt to test CPF theory, but rather a spe cific prediction derived from it. An experimental testing of CPF theory should consider the “currency” to be optimized (for instance, the energetic value or the nutritious compo nents of the food item as benefit, and the time (distance) or energy needed to bring it to the feeding place as cost). We have assumed that all types of food items had the same probability of being found at every distance from the feed ing place and that the size of a food item is a good estimator of its. nutritional value and the energy needed to take it away (Lindsironi 1994). Under these conditions, intermediate- sized items should be brought to the den more often than
large or small items.
The results seem to confirm our prediction. However, the
predominance of intermediate-sized food items in the cubs’
diets may arise from a sampling bias because a prey items that yield more remains (intermediate-sized rather than small prey) can be handled for a longer time around the den (Lloyd 1980). Nevertheless, such a bias should overestimate the frequency of occurrence of large prey, contrary to our re sults. Lindstrom (1994) specifically addressed this point by comparing the relationship between the frequencies of oc currence of large and small prey in the diets of adults and cubs of the Swedish red fox over 3 years, concluding that proportions were not consistent through the years and there fore the hypothesis of sampling bias was not supported.
Resolving at least two other methodological questions could modify the interpretation of our results. On one hand, medium-sized portions of large items, such as deer and sheep, could be brought to the den. Again, ignoring this pos sibility is conservative, as it favours the null hypothesis of similar diets in adult foxes and cubs. On the other hand, cubs could be fed raw meat from large food items, leaving no remains in seats. While we have confidence in our results (usually it is possible to recognize significant amounts of in gested soft food in the feces; also, it is unlikely that no hair will be found), this is a real possibility, although it is im probable,.
Undoubtedly the small Patagonian gray fox cannot carry ungulates to the den (but see the previous paragraph) and probably finds it difficult to carry hard-shelled armadillos in its mouth. Also, it would be unprofitable to take small ber ries and insects to provision the cubs unless large amounts of these foods were eaten and subsequently regurgitated, but this behaviour is unfrequent in foxes (Lovari and Parigi
1995). However, insects were well represented in the feces of the cubs, probably because they were able to capture in sects themselves close to the den (Murie 1936).
Several authors have found differences in the size of food items eaten by adults and cubs of the Holaretic red fox Englund (1969) observed in Sweden that adult foxes carried hares for feeding cubs at the den, while they themselves fed on rodents (rodents are relatively small prey for the red fox, which weighs about 6—7 kg, while they would be considered medium-sized prey for the smaller- gray fox). In southern Spain, adult red foxes consumed berries and insects but provisioned their cubs with rabbits (Rau et al. 1985). Also, in suburban London, adult foxes fed selectively on the smaller food items obtained, carrying the larger ones to the cubs (Harris 1981). In all these cases, adult foxes consumed the smaller and less rewarding prey on the spot and fed their cubs the rulatively bulky items that could be carried, which is in accord with CPF theory.
However, Lovari and Parigi (1995) found that adult foxes consumed large pheasants, but selected for their cubs small and medium-sized passerines, probably because they are easier to digest. This observation opens up the possibility of alternative hypotheses (e.g., protein may be more important than gross energy for the growing pups, or adults could se lect suitable partially succumbed prey for bringing to the den in order to stimulate cubs to chase them).
In a review of the feeding ecology of the gray fox in Chile, Medel and Jaksic (1988) found that rodents were the principal component of the diet of this canid, followed by birds and insects. Fruits, lagomorphs, and reptiles were
Fig. 1. Percent occurrence and percent dry mass of of large, intermediate-sized, and small food items in the feces of adults and cubs of the Patagcmian gray fox (Psendra1opex grisecs). Armadillos are included in the first category because of the difficulty of carrying them for a small fox.
food item
poorly represented, depending on region and season. During the rearing period in our study area, the gray fox followed this general pattern, although berries occurred in the feces of adults of this species more often than in those of Chilean gray foxes. In any case, dietary changes according to season and geographic location must be expected in an opportunis tic forager such as the gray fox.
Acknowledgements
We are grateful to 0. Ceballos, 13. Castillo, LA. Donázar, M. Funes, F. Hiraldo, and 0. Monsalvo for help with the fieldwork. 0. Alefiá permitted us to work on his land. A. del Valle and the Centro de Ecologla Aplicada del Neuqudn (Ar gentina) provided us with logistic support and kind assis tance. M.C. Blázquez, P. Ferreras, S. Lovari, F. Palomares, E. Revilla, and J. Weaver improved earlier versions of the