The Ecology of Urban Red Foxes (_Vulpes vulpes_) written in partial fulfillment of the requirements of the degree of Master of Sciences in Zoology for the University of Minnesota, December 1994, by Timothy Susman Introduction Increasing human population density has affected land-use patterns worldwide over the past century. Urban habitats have become more and more common, yet only in the past two decades has there been significant interest in the effect of urbanization on non-human species such as the red fox (Vulpes vulpes). Several recent studies have examined the home ranges and diets of foxes in urban areas compared to foxes in rural areas (Harris, 1980; Harris, 1981b; Harris and Rayner, 1986; Harris and Smith, 1987; Doncaster et al., 1990; Coman et al., 1991; Doncaster and MacDonald, 1991). The idea that ecological interactions of animals may differ in different habitats is not new (Lessells, 1991), with the differences depending on the animal's life history and the nature of its habitat interactions. An urban environment is very different from our concept of a _natural_ habitat, and we might expect animals living in urban areas to have different behaviors and life histories compared to conspecifics in _natural_ habitats. This appears to be true for mammals such as the gray squirrel (Sciurus carolinensis) and eastern chipmunk (Tamias striatus), which are strongly affected by urban variables such as proximity to artificial feeders, houses, or other physical structures (Flyger, 1970; Ryan and Larson, 1976). For small mammals (Apodemus sylvaticus, Clethrionomys glareolus, Microtus agrestis, and Sorex araneus) in Oxford, however, vegetation is more important than urban variables in predicting population density, because urban areas are a patchwork of distinct habitat types (i.e., woodland, scrub, detached house garden), and individuals of those four species remain in a single habitat patch for most of their lives (Dickman and Doncaster, 1987). The red fox is a territorial animal with a home range usually covering at least several hectares. In those urban areas it has invaded successfully, the home ranges of individuals include patches of several different habitat types. Therefore, it is not surprising to see differences between the behavior and diet of urban foxes and the behavior and diet of rural foxes. I studied the parental behavior and habitat interactions of urban and rural foxes in the Minneapolis-St. Paul area to see if any differences existed in those behaviors between urban and rural foxes. This paper will present the results of that study and examine the ecology of urban foxes compared to the ecology of rural foxes. Some characteristics of urban environments that favor the behavioral changes observed in urban foxes will be described. Methods and Materials My goal was to locate five urban den sites and five rural den sites. Den sites were located by direct investigation of sites recommended by personal communications with residents of the metropolitan area and surrounding rural areas. To pinpoint den sites and areas frequented by foxes, two field assistants and I searched for signs of foxes such as tracks, scat, and prey remains as described in Henry (1993). The classification of den sites as "urban" is somewhat subjective. Dickman and Doncaster (1987) arbitrarily designated any site within city limits as urban; Harris and Rayner (1986) used a much more restrictive definition that excluded corridors of natural habitat such as a river valley. In Minneapolis-St. Paul, there is a great variety of urbanization within the city limits; the central downtown areas have been in existence for over 100 years, but certain areas within the city limits are much less disturbed than some more remote suburbs. In addition, there is an extensive park system in the city. For the purposes of this study, a den site was considered part of an urban habitat if it was within a quarter mile of at least five occupied buildings and one paved road. I planned to make observations weekly at each site for 10 weeks, from early May until late July. This covers the time period from the first appearance of the cubs outside the den at about 4-5 weeks of age (Storm et al., 1976; Henry, 1993) until the cubs are 14-15 weeks old. The adults no longer provide food for the cubs after this time, although the cubs remain in the territory of their parents for another 3 months (Storm et al., 1976). Observation periods were 4 hours long. Prior to the study, I planned to observe one urban den and one rural den twice each for 24 consecutive hours. Based on these observations, the observation periods for the study would be set to cover the widest possible range of fox activity and interactions. Observations were made by a single observer from a location that provided a clear view of the den, but was not close enough to alarm the foxes. Use of a blind was considered, but Henry (1993) asserted that because foxes have a sensitive sense of smell, a blind is not very effective in hiding a human observer. He found that it is more important to determine the distance at which a human will be tolerated from each fox or group of foxes and to remain at that distance. The distance differs for each fox and must be determined carefully, starting at a large distance and gradually moving forward over several hours. Data recorded during each observation were to be: maximum number of cubs seen at once, number of times that adults return with food for the cubs, the type of food brought back by the adults for the cubs, any food the cubs take for themselves (not provided by the adults), all intraspecific and interspecific interactions, and all behavior toward man-made structures. All data collection was to be made using the cubs as the focal group for observations and also recording all occurrences of the above behavior in the adults, as described in Altmann (1974). The data for the urban sites were to be compared to the data for the non-urban sites, and the differences tested for significance by the Chi-squared test. Results In seven weeks of searching the Minneapolis-St. Paul area, my field assistants and I were only able to locate five fox dens, two urban and three rural. We found many signs of fox, such as scat, trails, and prey remains, but were unable to trace these signs to a den in most cases. Three possible reasons can explain our inability to locate many urban fox dens. The first is that we were all inexperienced at tracking foxes and finding their dens. The second is that urban foxes must take extra care to conceal their dens, because of the high risk of encounters with humans and dogs. The third is that because of the variety of habitats available in an urban area, urban fox dens look different than rural fox dens. Although I am certain that inexperience was a contributing factor, I believe that extra concealment of urban fox dens was also important. My assistants and I spent most of our search time in urban areas, where we did not find any dens that had not been pointed out to us. In rural areas, by contrast, we were shown the locations of two dens, and found a third one in only a couple hours upon further investigation of the area. The biggest problem in rural areas was that many more species inhabit underground dens in those areas than inhabit dens in urban areas, so we could not always be sure that a den was inhabited by foxes. Despite having found only half the number of dens I had hoped for, I prepared to carry out the study on the dens I had located. My assistants began observations on the rural dens, which were located in Maple Plain, Minnesota (about 45 minutes west of Minneapolis). I began observations on the urban dens, both located on land owned by the U. S. Army, formerly used for the Twin Cities Army Ammunition Plant (commonly called the Arsenal). In our initial observations, neither my assistants nor myself saw any foxes at the dens, although we had seen fresh sign there recently, and in one case we had seen two fox cubs playing. I ended my observation of one of the den sites after seeing a striped skunk (Mephitis mephitis) enter the den; that was the only wild mammal I saw in several observations of the urban den sites. We concluded that the foxes had abandoned the dens. Foxes rearing cubs often move the cubs to different dens, but in these cases abandonment might have been precipitated by the disturbance my assistants and I caused while investigating the den sites. By the time we discovered that none of the sites we had located were still inhabited, it was too late to locate more den sites. By the middle of June, most fox cubs in the Minneapolis- St. Paul area are between six and twelve weeks old, so even in the best case, I would be starting my observations in the middle of the period I was most interested in. I decided to review the literature on urban foxes again, and see where the observations I did make fit into other studies. I also spent some time during the summer working with a woman who rehabilitates injured foxes and releases them to the wild. Many of the foxes she receives for treatment come from the Minneapolis-St. Paul area, so I was able to get a little more data from observing some of her patients. The most interesting observation I made at this location was of a young cub, about six to eight weeks old on 5 August. She must therefore have been born in early June, which is very late in the year relative to parturition dates listed in the literature. I will discuss this and other observations I made in the rest of this paper. First, I will discuss the general life history of red foxes as described in the literature, noting differences between urban and rural foxes. Next, I will examine in more detail the aspects of fox life history that differ most between urban and rural areas: diet, home ranges and dens (which are partially determined by diet), and mortality and life expectancy. Last, I will examine the scope of fox-human interactions in urban areas. Life History Reproduction of red foxes is described by Storm et al. (1976) and Henry (1993). The gestation period and time of weaning are constant over different populations. Timing of mating and birth, by contrast, may vary geographically and is even subject to variation within a small area. Dispersal of juvenile foxes takes place every year, and because of variation in the time of birth, one might suspect that the time at which dispersal begins would also reflect strong geographic variation. In fact, the time of year at which dispersal is initiated seemed to be constant over all the populations studied. These and other aspects of the life of an urban fox over a single year are detailed in the following paragraphs. Courtship for foxes begins in mid to late December. Spermatogenesis begins in October (even in first-year males), and continues through March (Storm et al., 1976; Henry, 1993). When courtship starts, previously solitary foxes travel in pairs, or sometimes in groups of three. Mating usually takes place in January and February, although it may be earlier in more southern areas and later in northern areas. In the city of Minneapolis, I observed fox cubs that were about six weeks old as early as 7 May, and as late as 5 August. Therefore, there is at least a three-month window during which mating can take place and result in production of viable offspring. Given a gestation period of 52-53 days, which is consistent through the literature for all fox populations, these observations indicate that mating takes place in the Minneapolis area between late January and late April, and possibly earlier or later. Although Storm et al. (1976) claimed that breeding can take place as late as April, they noted that 94% of the rural foxes they studied bred between mid-December and mid-February, with a peak during the third week of January. Only 1% of their sample had bred after mid-February. Harris and Smith (1987) reported that for foxes in both London and Bristol the peak birth period was March, indicating a peak breeding period in January. They noted that in those cities matings can also take place in early December (about 10%), or after the first week of February (about 25%). No study has been conducted directly comparing the lengths of breeding seasons in urban areas as opposed to nearby rural areas, but it seems that the breeding season may be extended somewhat in urban areas. If this is true, the most likely explanation is the protection an urban environment provides from the extremes of weather. Most importantly, food is not scarce in urban areas in winter, as it often is in rural areas (Harris, 1977; Storm et al., 1976). The availability of food also affects the composition of fox family groups (von Schantz, 1984; Zabel and Taggart, 1989). A typical family group consists of the dog fox and vixen, plus up to four females that may or may not breed. Macdonald (1979) referred to these females as "helpers," and observed a definite dominance hierarchy in the family group, wherein the "helpers" were subordinate to the breeding female. Harris (1981a) also noted that some family groups in Bristol produced more than one litter of cubs in a given year, indicating that some of the "helpers" may also breed under certain conditions, presumably when resources are plentiful. Space is an important resource for foxes. Each family needs several safe den sites in which to raise their cubs in addition to safe resting sites outside of the den. Den space seems to be readily available for urban and rural foxes alike. Foxes maintain territories, and it is common for them to have several different dens within their territory. Prior to parturition, the mated pair selects a den, called the natal den, in which to bear the cubs. This den is usually larger than other available dens, and has several exits (Henry, 1993). The other dens can be as simple as a single entrance leading to a hole where the foxes can take shelter. In urban areas, a great variety of den sites is available to foxes, but places to rest during the day are scarcer. Denning, home ranges, and daytime resting sites are covered in a separate section. The natal den must be safe, because the fox family is most vulnerable to predators during the first two weeks after parturition. Fox cubs are blind and helpless for about ten days after birth. The vixen stays with them constantly for up to two weeks (longer in colder climates than in warmer climates), during which time she is dependent on the dog fox and any "helpers" for food. After this time, she may leave to hunt for herself, but will return regularly to nurse the cubs. The parents may first move the cubs to an alternate den when they are about two weeks old. The cubs usually are moved at least once, and may be moved several times during their first summer. For example, all the dens I located were abandoned in late spring. In two of the rural dens, human disturbance might have led to the abandonment, but humans frequented the immediate areas of the urban dens at least every other day, so it is unlikely that human disturbance was the only cause of abandonment of the urban dens. This suggests that urban foxes, like rural foxes, move frequently between dens. Movement may help protect the fox family from humans and other predators when the cubs are most vulnerable. When the cubs are about five weeks old, they begin to venture outside the den. Weaning takes place at this time, so the parents bring back prey for the cubs to eat. Later in the summer the cubs learn to hunt for themselves, and their parents will provide less food and spend less time at the den. The cubs range farther from the den as they grow older, hunting for themselves but remaining in the territory of their parents. October is the month during which dispersal of foxes is most likely to begin. Not all cubs disperse, and not all of those that disperse leave in their first year. In a study of dispersal, Harris (1988) found that only 3.5% of female cubs and 10.8% of male cubs dispersed before 1 October. By the end of December, Harris (1988) found that 32.5% of tagged females and 58.5% of tagged males had dispersed. Most of the animals that dispersed did so by the end of their second year, the final proportions being 37.8% for females and 75.8% for males. This corresponds closely with data gathered by Storm et al. (1976) on rural foxes, which indicated that most of them also disperse in October. The proportion of foxes dispersing in the rural population they studied was somewhat higher, reaching 58% for females and 96% for males. Population density is probably the most important factor affecting dispersal. In urban areas, for example, Harris (1988) found that male cubs from large litters were more likely to disperse than male cubs from small litters. He also found that animals from areas of low fox density (indicating poor fox habitat) dispersed farther than animals from areas of high or medium fox density. Another factor affecting dispersal is the food supply in the territory. Cubs might remain in their natal territory longer in urban areas, where food is plentiful, than in rural areas. However, the sample sizes and geographical areas of the studies by Harris (1988) and Storm et al. (1976) are too small to draw a definite conclusion. Once the cubs have dispersed, the adults and remaining cubs stay on the same territory, but continue to hunt alone. They may rest with other foxes in areas where resting habitats are limited, but do not associate in regular pairs until courtship begins again in December. Feeding and Diet Foxes are omnivorous, feeding on fruit, birds, small mammals, insects, and other invertebrates, especially earthworms. In rural areas, up to 60% of their diet may come from only a few species, the particular species varying by region (Henry, 1986). In two major studies of the food of urban and suburban foxes, earthworms (Lumbricus terrestris) were found to comprise from 5 to 25 percent of the diet of London foxes and from 15 to 37 percent of the diet of foxes in Oxford, the exact percentage depending on the season. No single species was more important, but the general category of scavenged items formed from 20 to 50 percent of the diet of both populations of foxes (Harris, 1981b; Doncaster et al., 1990). Scavenged items is the only type of food found in much greater abundance in an urban area than in a rural area. This category includes edible refuse discarded by humans, as well as some inedible refuse and food provided by humans for foxes or other wildlife. Harris (1981b) noted that foxes often preyed on the other wildlife the food was provided for, although those prey would not be included in the scavenged items category. Table 3 presents a list of edible scavenged items found in the stomachs of London foxes by Harris (1981b), which shows the variety of food available to urban foxes. Scavenged food is important in the diet of urban foxes because it requires less energy to catch than live prey and it is available year-round. Figure 1 shows the seasonal variation in food contents of stomachs of London foxes examined by Harris (1981b). Scavenged meat and bones shows the least seasonal variation of all the food categories, increasing noticeably in importance only in winter. This may be because other prey was scarce, or because people perceive a greater need to feed wildlife in the winter and leave more food for them (Doncaster et al., 1990). Seasonal variation is more pronounced in all other food categories because they include animals or plants which follow a yearly life cycle. Doncaster et al. (1990) found that the abundance of food types in the feces of foxes in Oxford followed roughly the same seasonal trends as the measured availability of those food types. Both Harris (1981b) and Doncaster et al. (1990) conclude that scavenged items are readily available and important to the survival of urban foxes, and may even create a food surplus. Harris (1981) remarks that the body weight of the foxes he studied was constant through all times of the year, and that he found no evidence of food shortage at any time. Doncaster et al. (1990) observed some sites where food was provided by residents for foxes, and found that scavenging foxes were highly selective, discarding some edible scavenged items in favor of others. These studies support Harris's assertion that shelter is a more important limiting resource for urban fox populations than is food (Harris, 1977); food availability does not seem to be limiting at all. Dens and Home Ranges The most distinctive feature of an urban environment is the abundance of man-made structures that offer shelter for many non- human animals. Foxes take advantage of man-made structures, sometimes even setting up their natal dens in occupied houses (Harris, 1981a; Harris, 1980). Harris (1977) and Storm et al. (1976) determined that the most important factor governing the distribution of breeding dens for both urban and rural foxes was human disturbance. Harris (1977) located 97 dens in London, 46 of which were under residential buildings (36 of these under garden sheds). Of the remaining 51, more than half (29) were in banks of earth, such as garden banks or railway embankments. Some foxes raised their cubs under the floorboards of occupied houses, often gaining entrance through a pet door. In one case, a domestic cat and dog, both of which tolerated the foxes, were present in the household where a fox den was located (Harris, 1980). Usually, foxes avoid dogs; the distribution of foxes in Bristol showed a significant negative correlation to the distribution of feral dogs (Harris, 1981a). Fox distribution in urban areas is also strongly influenced by the local habitat. The number and type of habitats available to foxes in an urban area are highly variable. Harris (1977) listed some of the habitats found in London, and noted that most of the foxes he studied were recovered from residential habitats, industrial habitats, and vacant land with no public access. Surprisingly, despite reports that open public parks were important for daytime shelter of foxes, Harris recovered only 8.73% of his foxes from parks, although parks comprised 10.53% of suburban land. By contrast, nearly 60% of his foxes were recovered from residential habitats, which comprised just over 40% of suburban land. Harris and Rayner (1986) found a significant positive correlation between areas of high fox density and residential areas where the housing was owner- occupied. Harris (1977) speculated that space for daytime shelter is an important limiting resource in London, and that the availability of suitable daytime resting habitat has a strong influence on the distribution of foxes. He found that between autumn and spring, several foxes will often share a daytime retreat. This clumping is not unknown in rural fox populations, but it is more common and takes place for a longer period of time in urban and suburban fox populations (Harris, 1977). Clumping is to be expected in urban and suburban populations not only because daytime resting sites are rare, but also because fox densities are generally higher in urban areas than in rural areas. Rural foxes are known to prefer regions of diverse habitat, such as edge habitats (Henry, 1993). The abundance of such diverse regions in an urban area may account for the unusually high density of foxes found in cities -- Harris (1977) recorded up to five family groups per km2 in several of his studies. By contrast, two families of foxes studied by Storm et al. (1976) in Minnesota stayed within separate well-defined areas of 9.6 km2 each. The average home range of an urban fox is significantly smaller than the average home range of a rural fox (Harris, 1980; Coman et al., 1991; Doncaster and MacDonald, 1991). In a study of foxes in Australia, Coman et al. (1991) found that two family groups of foxes living on farmland had home ranges of 5.0 and 7.2 km2, while three family groups of foxes on the fringe of an urban area had home ranges of 0.6, 1.3, and 0.8 km2. Doncaster and Macdonald (1991) found a similar difference between foxes living in central Oxford and foxes living in suburban and rural areas around Oxford, and also found that the home ranges of urban foxes changed over time. "City ranges were not spatially stable over months or even weeks," Doncaster and MacDonald (1991) wrote. "They moved in step-wise extensions to encompass new areas whilst at the same time contracting other parts of the range to expel old areas." This movement of territories is unique among foxes. Doncaster and Macdonald speculated that movement of home ranges is a behavioral adaptation developed since foxes invaded urban areas in England in the 1930's. The average amount of food available over the area of a city is higher than in a similarly sized rural area, but the variance in food availability is also much greater. For foxes to make the best use of the resources in urban environments, they must constantly explore new areas and re- explore old ones. This activity would require far too much energy to be viable in a large rural home range, but the high density of different patches in a city permit foxes following this strategy to survive and prosper. Life Expectancy and Mortality Harris (1977) studied a suburban fox population in London, and constructed an abbreviated life table from examination of specimens either found dead (mostly on roads) or recovered from control operations (Table 1). Harris and Smith (1987) compared the demography of London foxes to that of foxes living in Bristol, using specimens from the same sources. A life table for rural fox populations was constructed by Storm et al. (1976) by studying foxes tagged as juveniles and recovered when killed (Table 2). In all studies, juvenile mortality was found to be significantly higher than adult mortality. Adult mortality remained constant after the first year, and varied little over the six-year study in London and Bristol (Harris and Smith, 1987). By contrast, Storm et al. (1976) found that the mortality of foxes in rural areas of the upper midwestern United States sometimes varied greatly from year to year. They attributed some of this variation to climate (for instance, in one winter, snowfall was much greater than in other winters, and fox mortality was higher). Foxes in urban areas may not be affected as much by climatic variation, due to the presence of abundant shelters and human activities such as snow plowing. The data from these studies, however, are insufficient for definite conclusions. Some human activities, such as fox control operations, obviously can have a great impact on fox mortality. Harris and Smith (1987) investigated the effect of control on the fox population of London, and concluded that the productivity (total cubs produced/total population) of a population subjected to the stress of human control is higher than that of an unstressed population. Through examination of the fox population, they found that the increase in productivity did not come from increased litter size. Instead, the proportions of non-breeding vixens and of dog foxes were lower in stressed populations than in unstressed populations, and thus a greater percentage of the population produced cubs. Urban populations also experience stresses from non-human sources. In Minneapolis-St. Paul, the fox population is subject to epizootics of sarcoptic mange, which can have a great impact on its numbers (Roger Johnson, pers. comm.). Rabies can also reduce fox populations, although it was not an important factor in any of the studies cited in this report, nor was it considered to be a problem in Minneapolis-St. Paul during the period of this study. The greatest contributor to fox mortality in both Harris and Smith’s and Storm et al.’s studies was human action. In 82% of the cases of mortality in London foxes, the fox had been shot, snared, trapped, or dug out of a den (Harris and Smith, 1987). Storm et al. (1976) reported that in the rural areas where his foxes were recovered, 78% of juveniles and subadults and 90% of adults had been shot or trapped. In Bristol, where no control measures were taken and foxes were not hunted for fur, the main cause of death for foxes studied by Harris and Smith (1987) was road accidents (63% over all age and sex categories); only 11% were trapped or shot. In a study conducted on the Cedar Creek Natural History Area in Minnesota, a rural area in which foxes are not subject to control, Sargeant et al. (unpublished) studied the movements of foxes via radio telemetry. They found that the greatest cause of fox mortality was human action, both road accidents and shooting or trapping. These results probably reflect inflated percentages, because foxes killed by human action are the most likely to be recovered for study. Sargeant et al. noted the difficulty of estimating fox mortality from natural causes. Like many animals, sick foxes often will select a secluded area in which to die. Sargeant et al. radio-collared several foxes, and found a few that had died in dens or other secluded places. Many of the foxes in their study were not recovered, and could have been lost to natural death, unreported road accidents or shooting, or migration. However, Sargeant et al. reported more foxes lost to human action than to unknown causes. Although one might initially suppose that living in close proximity to a large number of people would greatly increase fox mortality, or at least human-induced mortality, Storm et al.’s and Harris and Smith’s studies show that this is not necessarily the case. In fact, the rural foxes studied by Storm et al. were also subject to human-induced mortality, whether they were killed as pests or trapped for fur. In general, foxes rarely live beyond five years in the wild, in both urban and rural areas (Harris and Smith, 1987; Storm et al., 1976). The major variation in fox mortality in different areas is in the causes of death and in differences in annual mortality rate. Contact with humans The densities of foxes and of people are greater in urban than in rural environments. With such a concentrated human population, interactions between humans and foxes are inevitable in an urban area. Foxes avoid people whenever they can. They certainly use human structures for their dens, and human refuse for their food, but they prefer to stay away from the humans themselves. Their frequent movement to different den sites may help them avoid human notice, and their mostly nocturnal life minimizes encounters with humans. Humans vary greatly in their attitude towards foxes. Some consider them pests or trap them for their fur, but others enjoy their presence and will even set out food for them. One source of problems between humans and foxes is fox predation on domestic pets. In his study of food preference in suburban foxes, Harris (1981b) questioned 5,191 households in Bristol to find out how many had lost pets to foxes. Only 2.7% of households who owned cats had lost a cat to foxes, and most of the cats lost were kittens. Of families who kept pets other than cats or dogs in an outdoor garden accessible to foxes, 8.0% had lost a pet to foxes within the past year, and 51.1% had lost a pet more than a year prior to the survey. However, many people in Bristol provided food for foxes, sometimes regularly, and sometimes when they saw a fox on or near their property. Conclusion An urban habitat is distinguished from other habitats by its higher density of people and man-made structures. It is characterized by a number of sub-habitats, densely packed together in small patches. In this environment, many non-human animals have been able to survive and maintain secure populations. Among these is the red fox. Foxes are able to take advantage of the year-round shelter offered by urban buildings, and the year-round food supply of human refuse. The extremely patchy and variable nature of the environment does not pose a problem for these animals; unlike their rural counterparts, they do not keep stable territories, but have constantly shifting home ranges that drift as they abandon old areas and explore new ones. There are many differences in behavior between urban foxes and rural foxes besides the stability of their home ranges. Urban foxes have a different diet than rural foxes, and there are some indications that their breeding season may be longer. The shelter and food provided in an urban environment may allow foxes born very early or very late in the breeding season to survive, whereas in a rural environment these foxes might perish. Although urban foxes rarely starve, they are subject to many other causes of mortality. There are more roads in an urban area than in a rural area, and in urban areas where foxes are not controlled by humans, road accidents are the primary cause of death. Even in areas where foxes are controlled, they maintain population densities comparable to fox densities in uncontrolled areas. In conclusion, although urban environments are highly modified, they offer opportunities and risks to their inhabitants, just as do other habitats. The red fox, a generalist known to be very adaptable, has invaded urban areas and established viable permanent populations there. Its behavior is similar to that of rural foxes, but differs where the features of the urban environment have favored behavioral changes. The fox is an excellent example of an animal whose adaptability permits it to survive even in the highly modified habitat created by and for humans. Literature Cited Altmann, J. 1974. Observational study of behavior: sampling methods. Behavior, 49: 227-265. Coman, B. J., J. Robinson, and C. Beaumont. 1991. Home range, dispersal, and density of red foxes (Vulpes vulpes) in Central Victoria. Wildlife Research, 18: 215-223. Dickman, C. R., and C. P. Doncaster. 1987. The ecology of small mammals in urban habitats. I. Populations in a patchy environment. Journal of Animal Ecology, 56: 629-640. Doncaster, C. P., C. R. Dickman, and D. W. MacDonald. 1990. Feeding ecology of red foxes (Vulpes vulpes) in the city of Oxford, England. Journal of Mammalogy, 71: 188-194. Doncaster, C. P., and D. W. MacDonald. 1991. Drifting territoriality in the red fox Vulpes vulpes. Journal of Animal Ecology, 60: 423-439. Flyger, V. 1970. Urban gray squirrels -- problems, management, and comparisons with forest populations. Transactions of the Northeast Fish and Wildlife Conference, 27: 107-113. Harris, S. 1981a. An estimation of the number of foxes in the city of Bristol and some possible factors affecting their distribution. Journal of Applied Ecology, 18: 455-465. -----. 1981b. The food of suburban foxes (Vulpes vulpes) with special reference to London. Mammal Review, 11: 151-168. -----. 1980. Home ranges and patterns of distribution of foxes (Vulpes vulpes) in an urban area, as revealed by radio tracking. Pp 685-690 in A handbook of biotelemetry and radio tracking (C.J. Amlaner, Jr. and D.W. MacDonald, eds.) Pergamon Press, Oxford, 804 pp. Harris, S., and J. M. Rayner. 1986. Urban fox (Vulpes vulpes) population estimates and habitat requirements in several British cities. Journal of Animal Ecology, 55: 575-591. Harris, S., and G. Smith. 1987. Demography of two urban fox (Vulpes vulpes) populations. Journal of Applied Ecology, 24: 75-86. Henry, J. D. 1993. How to spot a fox. Chapters Publishing, Shelburne, Vermont, 110 pages. -----. 1986. Red fox: the catlike canine. Smithsonian Institution Press, Washington, D.C., 174 pages. Lessells, C. 1991. The evolution of life histories. Pp 32-68, in Behavioural ecology (J.R. Krebs and N.B. Davies, eds.). Blackwell Scientific Publication, Oxford, 482 pages. MacDonald, D. W. 1977. On food preference in the red fox. Mammal Review, 7: 7-23. -----. 1979. "Helpers" in fox society. Nature, 282: 69-71. Ryan, D. A., and J. S. Larson. 1976. Chipmunks in residential communities. Urban Ecology, 2:17-38. von Schantz, T. 1984. "Non-breeders" in the red fox Vulpes vulpes: a case of resource surplus. Oikos, 49: 59-65. Sargeant, A. B., D. B. Siniff, and D. W. Warner. (unpublished). Ecology of the red fox in the upper midwest. Storm, G. L., R. D. Andrews, R. L. Phillips, R. A. Bishop, D. B. Siniff, and J. R. Tester. 1976. Morphology, reproduction, dispersal, and mortality of midwestern red fox populations. Wildlife Monographs, 49: 1-82. Zabel, C. J., and S. J. Taggart. 1989. Shift in red fox, Vulpes vulpes, mating system associated with El Niño in the Bering Sea. Animal Behaviour, 38: 830-838. Table 1. A partial life table of London foxes. Slightly modified from Harris (1977) to include only post-natal age classes and to combine "Birth to six weeks" and "Juveniles" (less than one year). Mortality rates from Harris's original table were used to create the lx and dx columns in this table. They reflect his original data. Number Number Mortality Life alive dying rate per expectancy, in at during 1000 years, for start age alive at those of age class start of attaining age class age class class Age class lx dx qx ex Juveniles 1000 696 .696 1.463 (0-1 yr) Adults: 1-2 years 404 218 .539 1.486 2-3 years 186 82 .443 1.635 3-4 years 104 44 .419 1.538 4-5 years 60 30 .500 1.278 5-6 years 30 20 .667 1.037 6-7 years 10 4 .444 1.111 7-8 years 6 6 1.000 0.500 Table 2. A partial life table of rural foxes in the northern midwestern United States. From Storm et al. (1976) Age class lx dx qx ex Juveniles 498.2 325.6 .65 1.42 (0-1 yr) Adults: 1-2 years 172.6 80.9 .47 2.17 2-3 years 91.7 18.1 .20 2.64 3-4 years 73.6 10.2 .14 2.16 4-5 years 63.4 4.6 .07 1.48 5-6 years 58.8 58.8 1.00 0.50 Table 3. Edible scavenged items found in the stomachs of 571 London foxes. From Harris (1981). Food type Number of % occurrence % of diet occurrences Meat and 278 48.7 18.4 bones from poultry carcasses and meat joints Bread 61 10.7 3.1 Dried fruit 58 10.2 2.4 Potato peelings 48 8.4 1.6 Bird seed 11 1.9 0.4 Bacon rind 10 1.8 0.4 Orange peel 10 1.8 0.3 Fish bones 8 1.4 0.4 Carrot peel 6 1.1 0.2 Cooked peas 6 1.1 0.2 Eggshell 5 0.9 0.1 Nuts 5 0.9 0.2 Apple peel 3 0.5 0.1 Tomatoes 3 0.5 0.1 Bananas 2 0.4 0.1 Runner beans 2 0.4 0.1 Cloves 1 0.2 < 0.1 Chinese dinner 1 0.2 0.1 Figure 1: Figure 1 is not included with the paper. It is available as a separate file, fig1.jpg