Abstract: The annual and seasonal diet of the bobcat (Lynxrufus)was determined from analysis of 188 feces in the Cape region of Baja California, Mexico, an arid zone with numerous subtropical elements in its flora and fauna. Bobcats fed mainly on lagomorphs, which reached 74% of occurrence, followed by rodents (40%), reptiles (15%), and birds (12%). No seasonal variations were observed. The results were consistent with those of studies elsewhere, indicating that bobcats still rely upon lagomorphs for much of their food in southern latitudes. This supports the hypothesis that lynx have evolved to prey on hares and rabbits. The prevalence of reptiles as prey of hohcats in our study area was the highest ever reported. They were reported as bobcat prey in only I of 20 studies from north of latitude 40°, but in 14 of the 18 studies carried out south of this latitude. With regard to feeding on reptiles, the habits of bobcats in Baja California Sur resemble those of other similar-sized felids in tropical areas, such as ocelots (Felispardalis)and servals (Leptailurusservo!). Résumé : L’analyse de 188 tbces recueillies dans la region du Cap, en Basse Californie, Mexique, une zone aride comptaot de nombreux éléments sub-tropicaux dans sa faune et sa fore, a permis de determiner les regimes alimentaires annuel et saisonnier du Lynx roux (Lynxrufus)dans la region. Les lynx se nourrissent surtout de lagomorphes, jusqu’b
74% du régime, de rongeurs (40%), de reptiles (15%) et d’oiseaux (12%). Aucune variation saisonnière n’a été enregistrée. Ces résultats concordent avee ceux obtenus en d’autres regions, ce qui semble indiquer que les lynx comptent surtout sur les lagomorphes comme proies aux latitudes australes. Ces données appuient l’hypothèse selon
lynx de cette region est Ia plus haute jamais signalée. Des reptiles ont été mentionnés dans le régime alimentairc du
Lynx roux dans seulement l’une des 20 etudes connues au nord du 40° paralléle, mais daos 14 des 18 etudes connues au sud de cette latitude. En cc qui conceroe Ia consommation de reptiles, les habitudes des lynx de la Basse Californie se rapprochent de celles d’autres félins de taille semblable des regions tropicales, l’Ocelot (Felispardalis)et le Serval (Leptoilurusserval).
[Traduit par la Redaction]
Introduction Conspecific populations of widely distributed species will encounter geographical differences in the pressures that mold their feeding behavior (Arnold 1981). Hence, we would expect the regional diets of species having wide geographical distri butions to reflect the availability of different prey resources.
The bobcat (Lynx rufus) is the most widely distributed native cat in North America, ranging from southern Canada (approximately 52°N) tosouthern Mexico (17°N) (Anderson
1987). Its food habits are well known in many parts of this
Received June 7, 1996. Accepted October 7, 1996.
M. Delibes’ and S.C. Zapata. EstaciOn Bioldgica dc Doiiana, Consejo Superior de Investigaciones Cientificas (CS IC), Apartado 1056, Scvilla 41080, Spain.
M.C. Blázquex. Estacidn BiolOgica dc DoAana, CSIC, Apartado 1056, Sevilla 41080, Spain, and Ccntro de
R. Rodriguez-Estrella. Centro de lnvcstigacioncs BiolOgicas dcl Norocstc, Apartado 128, La Paz 23000, Mexico, and EstaciOn lliológica de DoOana, CSIC, Apartado 1056, Sevilla
Author to whom all correspondence should be addressed.
range, and its diet is usually based on lagomorphs (hares and rabbits), although there is some geographical variation, deer and (or) small mammals being important prey in some areas (McCord and Cardoza 1982; Maehr and Brady 1986). Up to nt)w, no information has existed on the food habits of this felid in the southernmost part of its range, which is included in or neighboring the Neotropical region.
The cat species of northern areas seem to be quite special ized in their diet, and Holarctic lynx species (genus Lynx, following Wilson and Recdcr 1993) are thought to have evolved to prey on Iagomorphs (Kurten 1968). ‘l’ropical felids, by contrast, usually take all the vertebrate taxa they can handle, especially reptiles (Emmons 1987; Kitchcner 1991).
This paper reports the annual and seasonal diet of bobcats in the Cape region of Baja California (Mexico), an area with numerous suhtropical elements in its vegetation and fauna (Goldman and Moore 1946). Weexpected that lagomorphs were important in the diet of Baja California bobcats (as they, like other lynx species, have evolved to capture lagomorphs), but also that reptiles would be more important here than else where (owing to subtropical conditions in the area). To try to detect whether there is a latitudinal gradient in the impor tance of lagomorphs and reptiles as bobcat prey, we review
38 papers concerning the Ibod habits of this species in differ cnt areas.
Bobcats in Baja California belong to the subspecies Lynx rufuspeninsularisand are reported to be the smaNestmembers of the species (Thomas 1898; Samson 1979). Given the strong relationship between predator and prey sizes (Rosenzweig
1966), especially in feuds (Leyhausen 1965), we expected
this small size to influence the kind of prey taken.
Study area Fieldwork was conducted in an area of about 10 km2 in and around El Comitán, owned by the Centro de Investigaciones Biológicas del Noroeste. It is a coastal lowland located 17 km north of La Paz City (24°17’N, 110°20’W). Soils are sandy loam, while the xerophytic sarcocaulescent scrub vegetation (Leon de Ia Luz et a!. 1996) is dominated by cardon cacti (Pachycereuspringlel),dagger cacti (Stenocereusgu,n,nosus), mesquites (Prosopisarticulara),palo verdetrees(ercidiu,n microphyllurn),Adam’s trees (Fouquieriathguem), plum (Cyriocarpa edulis),copales(Burserospp.), lomboys (Jairophacinerea),and chollas (Opunüacholla).This kind ofvegetation (“cardonal”) is very characteristic of the best preserved habitats in Baja California Sur. Mean annual rainfall
is 150 mm, with precipiuition concentrated during the summer months, from July to September. Mean annual temperature is 23.9°C.
Methods Wecollectedbobcatfeces(n 188) during 1994 by walking peri odically along sandy paths and dry riverbeds. At least two females with kittens and some males were known to be living in the area during the study period. All of the area was covered regularly, so we considered the collected feces to be a representative sample of those produced, although some were found in easily recognizable clusters ( toilets” or fecal marking locations; Bailey 1972). Feces were separated, labeled, and dried at 60°C until they reached a constant dry mass. If possible, prey items were identified to species by comparing hairs, teeth, feathers, scales, and bones with a refer ence collection. Results are presented as frequency of occurrence (number of occurrences of each prey type x 100 divided by the number of fecal samples; this implies that the sum of frequencies may be above 100). This method does not accurately reflect the mass of ingested material (e.g., Weaver 1993), but it is usually con sidered to give a good representation of food habits (Corbett 1989). Samples were divided into four, according to season: winter (January—March), spring (April—June), summer (July—September), and autumn (October—December). Numbers of occurrences were compared among samples through a contingency-table analysis (G test; Sokal and Rohlf 1981).
Results Prey items in bobcat feces from Baja California Sur included at least 9 species of mammals, 7 birds, 5 reptiles, and 1 arthro pod (Table 1). The most important prey category was lago morphs, which occurred in 73.9% of the feces. Rodents were the second most important component of the diet, as they occurred in 40. 1 % of the feces. Reptiles ranked third in importance (15.4% in occurrence), followed by birds (12.2%). Scorpions were also found.
In the study area the predominani prey of the bobcat were the jackrahbit and cottontail rabbit, consumed in about the same proportions. The pocket mouse was the most frequently
consumed rodent, followed by the white-tooted mouse and desert wood rat. The California quail was the most frequent bird in the analyzed feces. Among reptiles, the spiny-tailed iguana was the predominant prey species. It ranked fourth in occurrence, exceeded only by the two leporids and the pocket mouse (Table 1).
No significant seasonal changes in prey composition were detected: bobcats fed primarily on rabbits and secondarily on rodents in all seasons (Table 1). This lack of seasonal change in diet suggests that prey abundance and vulnerability do not vary greatly during the year, which is characteristic of tropi cal environments.
Discussion Lagomorphs occurred in most of the seats, with little varia tion between seasons, indicating that bobcats still rely upon them for much of their food near the southern limit of their range. This is consistent with the results of most studies of bobcat diets, irrespective of latitude. Lagomorphs occurred in more than 30% of the samples (stomachs and (or) intestine and feces) in 30 (83%) of 36 studies of bobcat diets in areas ranging from 48°N to 24°N (Table 2). Lagomorph remains were present in 49% or more of the samples in almost one- half of these studies. In two cases only, they reached less than 10% occurrence, in each case being replaced by another mammal species, the mountain beaver (Aplodontiarufa;74% frequency of occurrence versus 7% for lagomorphs in 247 feces from Oregon; Witmer and deCalesta 1986) and white- tailed deer (79% of occurrence in 43 feces collected during the winter in Massachussetts, with only traces of leporids, being found; McCord 1974).
These results support Kurten‘s (1968) hypothesis concern
ing the evolution of the feeding specialization of lynx species on lagomorphs. Nevertheless, rodents seem to be more impor tant than rabbits as food of bobcats in the southern Appala chians and western United States, and deer are also a notable prey of bobcats in the northeastern United States (Machr and Brady 1986).
Besides the fact that bobcats depend upon lagomorphs even at subtropical latitudes, the most distinctive feature in the diet of bobcats in Baja California is the high frequency of occurrence of reptiles, which appeared in more than 15% of the feces. The reptiles included at least two species of iguanas and three species of snakes, one of them venomous (Table 1). Both of the iguanas are relatively large, reaching 300 g in the case of the spiny-tailed iguana. These were consumed also in winter, in spite of the fact that they have a short hibernation period in the area (Blázqucz and Ortcga-Ruhio 1996).
Predation on reptiles by bobcats has been considered unusual elsewhere. In the review by Anderson (1987), reptiles were not even referred to as an occasional prey of the spe cies, while in the reviews by McCord and Cardoza (1982) and Rolley (1987), it is stated that bobcats cat mammals and some birds, although in certain circumstances they can take almost anything available, including reptiles.
Approximately 40% of the studies reviewed reported one
or more reptiles as prey of the bobcat (Table 2), including snakes (51.4% of all identified reptiles; most were non- venomous) aiid lizards (48.6%). The frequency of occurrence
Cc) t997 NRC Canada
Can. J. Zool, Vol. 75, 1997
Table 1. Seasonal diet (numbers of prey items) of bobcats from El Comitán, Baja California Sur, through analysis
Pocket gopher (Thoniornys sp.) Pocket mouse (Chaetodipussp.) Kangaroo rat (Dipodomysmerriami)White-footed mouse (Peromyscussp.) Desert wood rat (Neotomalepida)Unknown rodent
73.9 76.9 67.3 70.8 78.0 1.919 0.7
40.1 39.0 38.7 41.4 41.4 1.778 0.7
Domestic cat (Fellscatus)
American kestrel (Falcosparverius California quail (Callipeplacalfornica) White-winged dove (Zenaidaasiatica) Ground dove (Columbinapasserina) Common flicker (Colaptesauratus) Flycatcher (Myiarchussp.)
Note:vi is the number of feces examined.
of reptiles ranged between 0 and 15.4% (this study), hut rarely exceeded 2%. A definite north to south trend was not detected, but reptiles were reported as bobcat prey in only I of 20 studies carried out north of 40N, but in 14 of 18 studies carried out south of this latitude (Table 2).
North to south differences in predation upon reptiles could he related to changes in reptile abundance and the distribu ti()n of reptile sizes. A high incidence of reptiles in the diet of predators has been reported in desert areas at low lati tudes. In a pioneer study on the ecology of Saharan birds, Valverde (1957) stated that “in hot climates, vegetarian rep tiles play the same role as prey of carnivores as rodents in
12.2 8.122.4 4.2 12.5 4.519 0.2
15.421.0 12.6 16.79.7 2.802 0.4
1.10 00 4.8
cooler climates.” Likewise, Hernández et al. (1994) reported the importance of reptiles and insects in the diet of coyotes (canislutrans) living in the arid Sonoran desert, and other authors (e.g., Bothma et al. 1984) have also reported earn iVCOSfeeding on reptiles in other deserts. This is possibly due to the high species divcrsity and abundance of reptiles in these areas, because of the increased solar radiation (Schall and Pianka 1978) and the high production effIciency of small ectothermsunder these conditions(TurnerCf al. 1976).
Moreover, reptile abundance and diversity (and possibly total reptile biomass as well) increase toward the equator, independently of aridity (Schall and Pianka 1978; Zug 1993;
Delibes et al.
Table 2. Frequency of occurrence of Iagomorphs and reptiles in the bobcat diet in different states or provinces in North
America, ordered approximately from north to south.
No. of feces and intestinal
Brittell et al. 1979
Knick et al. 1984
Trevor et al. 1989
Litvaitis et al. 1986a
Litvaitis et al. l986b
Dibello et al. 1990
Koehler and Hornocker 1991
Matlack and Evans 1992
Hamilton and Hunter 1939
Toweill and Anthony 1988
Witmer and deCalesta 1986
Litvaitis et al. 1984
Gashwiler et a!. 1960
Story et al. 1982
Sperry (in Young 1958)
Leach and Frazier 1953
Sperry (in Young 1958)
Rolley and Warde 1985
Fritts and Sealander 1978
Tumlison and McDaniel 1990
Sperry (in Young 1958)
Jones and Smith 1979
Miller and Speake 1978
Beasom and Moore 1977
Maehr and Brady 1986
Delibes and Hiraldo 1987
for North America see Currie 1991). This tendency is reflected in an analysis of the food of feral cas (Fellscatus)on a global scale (Fitzgerald 1988). That author found a signifi cant negative correlation bctween latitude and incidence of predation on reptiles: below 35°N, reptiles were usually found in more than 20% of cat stomachs and intestines, but above
40°N they were found in no more than 10%. Also, bobcat-
sized tropical Cats, such as ocelols (Fellspardalis)in South
Americaand servals (Leplailurusserval)in Africa, prey on snakes and lizards regularly (Emmons 1987; Gccrtscma 1985).
The relationship between low latitude and high incidence
of reptiles in the diets of carnivores could be related also to
differences in the availability of medium-sized and large rep
tile species. On average, reptiles reach larger sizes at low
latitudes with a greater number of hours of sunlight (Andrews
1982);hence,theywouldappearto be morerewarding in energetic terms to carnivores. The large spiny-tailed iguanas
preyed upon in our study area belong to a subtropical species whose range is restricted to below 29°N.
Some minor features in the diet of bobcats in Baja California
Sur arc also interesting. Several authors (Bailey 1972 and later works) have reported selectivity for cottontails in com parison with jackrabbits. Unfortunately, the availability of
both species is unknown in our study area, but we saw jack- rabbits and their sign much more often than cottontails and
their sign, although both were consumed in approximately equal proportions. Thus, there could be also selectivity for cottontails in Baja California.
The rather high frequency of occurrence of small rodents, such as the pocket mouse, and the lack of ungulate remains in the feces could be related to the small size of bobcats in Baja California. In the southern and southwestern United States and northern Mexico, rodents are important prey for bobcats (e.g., Beasom and Moore 1977; Miller and Speake
1978; Jones and Smith 1979; Delibes and Hiraldo 1987), but
medium-sized species (above 100 g mass), such as cotton rats
(Sigmodonspp.), wood rats (Neotorna spp.), and to a lesser
degree squirrels (Sciuridae), are consumed most often, whereas the smaller pocket mice and white-footed mice (under 30 g) are rarely captured. Maybe these small rodents are energeti cally rewarding for the small bobcats in the study area.
On the other hand, we did not find ungulates in the analyzed feces. Usually, larger individual bobcats (adult males) take deer more often than smaller ones (females and juveniles; Litvaitis et al. 1986a). Hence, the small Baja California bob cats could be limited in their ability to subdue these large prey. However, the scarcity of wild ungulates (mule deer, Odocoileushemionus)in our study area makes it impossible to resolve this issue.
Although the bobcat is considered one of the most studied of all the wild felids (Kitchener 1991), more research in the
south of its range would no doubt be very useful for a more
comprehensive understanding of its ecology.
Acknowledgements This research was partially supported by the Junta de Anda lucia, Spain, through a travel grant to M.D., by the Centro de Envestigaciones BiolOgicas del Noroeste, Mexico, and by the CSIC, Spain. M.C.B. had a fellowship granted by CSIC — Consejo Nacional de Ciencia y TdcnologIa (C2l0/393) and R.R.E. had a predoctoral fellowship from the CSIC. Special thanks are extended to P. Ferreras, J. Svenson, A. Travaini, J. Weaver, R.A. Every, and an anonymous reviewer for useful comments on the manuscript, to Avclino Cota and Andrbs Sanchez for their assistance in the field, and to Javier Juste for providing literature.
References Anderson, E.M. 1987. A critical review and annotated bibliography of literature on bobcat. Spec. Rep. No. 62, Colorado Division of Wildlife.
Andrews, R.M. 1982. Patterns of growth in reptiles. InBiology of the Reptilia. Vol. 13. Edited byC. Gans and F.H. Pough. Academic Press, London. pp. 273—320.
Arnold, 5.3. 1981. The microevolution of feeding behavior. In
Foraging behavior: ecological, ethological, and physiological approaches. EditedbyA.C. Kamil and T.D. Sargent. Garland STPM Press, New York and London. pp. 409 453.
Blázquez, M.C., and Ortega-Rubio, A. 1996. Lizard winter activity
at Baja California Sur. 3. Arid Environ. 32: 247—253.
Bothrna, i. du P., Nel, J.A.J., and Mcdonald, A. 1984. Food niche separation between four sympatrie Namib Desert carnivores. J. Zool. (t965—l984), 202: 327—340.
Brittcll, J.D., Sweeney, S.J., and Kniek, S.T. 1979. Washington bobcats: diet, population dynamics and movement. Inst. Wildl. Res. Natl. Wildi. Fed. Sei. Tech. Ser. No. 6. pp. 107—110.
Buttrey, G.W. 1979. Food habits and distribution of the bobcat (Lynxrufus)on the Catoosa Wildlife Management Area. Inst. WildI. Rcs. Natl. Wildl. Fed. Sci. Tech. Ser. No. 6. pp. 87—91. Corbett, L.K. 1989. Assessing the diet of dingoes from feces: a comparison of 3 methods. J. Wildl. Manage. 53: 343—346. Currie, D.J. 1991. Energy and large-scale patterns of animal- and
plant-species richness. Am. Nat. 137: 25—49.
Delibes, M., and Hiraldo, F. 1987. Food habits of the bobcat in two habitats of the southern Chihuahuan desert. Southwest. Nat. 32:
Emmons, L.H. 1987. Comparative feeding ecology of feuds in a Neotropical rainforest. Behav. Ecol. Sociobiol. 20: 271 —283. Fickett, SB., Jr. 1971. Food habits data for the bobcat in Florida. Florida Game and Freshwater Fish Commission, Federal Aid in
Wildlife Restoration, Project No. W-4l-18.
Fitzgerald, B.M. 1988. Diets of domestic cats and their impact on prey populations. In The domestic eat: the biology of its behaviour. EditedbyD.C. Turner and P. Bateson. Cambridge University Press, Cambridge. pp. 123—150.
Fritts, S.H., and Sealander, J.A. 1978. Diets of bobcats in Arkansas with special reference to age and sex differences. 3. WildI. Manage. 42: 533—539.
Gashwilcr, 3.5., Robinctte, W.L., and Morris, OW. 1960. Food of bobcat in Utah Sand eastern Nevada. 3. Wildl. Manage. 24:
Geertsema, A.A. 1985. Aspects of the ecology of the scrval
Leptailurusservalin the Ngorongoro Crater, Tanzania. Neth.
3. Zool. 35: 527—610.
Goldman, E.A., and Moore, R.T. 1946. Biotic provinces of Mexico.
3. Mammal. 26: 347—360.
Hamilton, W.J., and Hunter, R.P. 1939. Fall and winter food habits of Vermont bobcats. 3. Wildl. Manage. 3: 99—103. Hcrnández, L., Delibes, M., and Hiraldo, F. 1994. Role of reptiles
and arthropods in the diet of coyotes in extreme desert areas of northern Mexico. 3. Arid Environ. 26: 165—170.
Jones, J.H., and Smith, N.S. 1979. Bobcat density and prey selec tion in central Arizona. 3. WildI. Manage. 43: 666—672. Kitehener, A. 1991. The natural history of the wild cats. Christopher
Helm Ltd., London.
Kniek, S.T., Sweeney, S.J., Alldredge, J.R., and Brittell, J.D.
1984. Autumn and winter food habits of bobcats in Washington
State. Great Basin Nat. 44: 70—74.
Kochlcr, G.M., and Hornocker, M.G. 1991. Seasonal resource use among mountain lions, bobcats and coyotes. 3. Mammal. 72:
Kurtcn, II. 1968. Pleistocene mammals of Europe. Weidenfeld and
Leach, HR., and Frazier, W.H. 1953. A study on the possible extent of predation on heavy concentrations of valley quail with special reference to thehoheat. Calif. Fish Game, 39: 527—538. Leon de Ia Luz, J.L., Coria, R., and Cruz, ME. 1996. Fenologia Iloral de una eomunidad brido-tropieal de Baja California Sur,
Mexico. Aeta Bot. Mex. 35: 45—64.
Leyhausen, P. 1965. Uher die Funktion der relativen Stimmungs hierarchic (dargestellt am Beispiel der phylogenetisch und ontogenetisehen Entwieklung des Beutefangs von Rauhticrcn). Z. Tierpsychol. 22: 412—494.
Trevor, J.T., Seabloom, R.W., and Allen, S.H. 1989. Food habits in relation to sex and age of bobcats from southwestern North
Dakota. Prairie Nat. 21: 163—168.
Tumlison, R., and McDaniel, V.R. 1990. Analysis of the fall and winter diet of the bobcat in eastern Arkansas. Proc. Arkansas Acad. Sci. 44: 114—117.
Turner, PB., Medica, P.A., and Kowalewsky, B.W. 1976. Energy utilization by a desert lizard (Urnstansburiana).U.S./t.B.P. Utah University Press, Logan, Desert Biome Monogr. No. 1. Valverde, J.A. 1957. Ayes del Sahara espahol (estudio ecológico del desierto). Instituto de Estudios Africanos, Consejo Superior