Ecology of the Feral Cat, Felis catus (L.), in South-Eastern Australia 111." Home Ranges and Population Ecology in Semiarid North-West Victoria

Transcription

1 Aust. Wildl. Res., 1982, 9, Ecology of the Feral Cat, Felis catus (L.), in South-Eastern Australia 111." Home Ranges and Population Ecology in Semiarid North-West Victoria Evan ones^^ and Brian J. comana *Department of Crown Lands and Survey, Keith Turnbull Research Institute, Frankston, Vic BPresent address: Regional Office, c/- Department of Agriculture, P.O. Box 125, Bendigo, Vic Abstract This paper reports home range sizes and population ecology of feral cats in a ha study area situated in the Victorian Mallee. Movements of six cats were monitored by radio-tracking for 8-21 months. Adults maintained discrete home ranges; areas varied from 3.3 to 9.9 (mean 6.2) km2 for males and from 0.7 to 2.7 (mean 1,7) km2 for females. Rabbit warrens, hollow logs and dense thickets were favoured daytime refuges. Mean daily straight-line distances moved bet-veen daytime refuges varied from 0.06 km for a female with juveniles to 1.67 km for an adult male. Relative abundance of cats over four years showed seasonal fluctuations, with summer maxima and winter or spring minima; the calculated mean summer and winter densities were 2.4 and 0.74 cats per km2 respectively. Summer maxima were composed of adults, adolescents and juveniles; winter minima were usually composed only of adults. Mortality, presumably caused by a nutritional stress acting particularly on subadults, maintained the adult population at a relatively stable level. Introduction The study was undertaken from June 1977 to May 1981 in the Hattah-Kulkyne National Park, situated in the Victorian Mallee. It is not known when feral cats Felis catus (L.) first became established in the Mallee, but they were probably introduced during early European settlement, which commenced in the 1840s. By the 1870s European rabbits Oryctolagus cuniculus were abundant (Anon. 1974) and their colonization would have aided the spread of feral cats by providing an important source of food; feral cats are now common in the area. Aspects of the population ecology of feral cats inhabiting subantarctic islands have been reported (Derenne 1976; Derenne and Mougin 1976; Jones 1977; van Aarde 1978; Pascal 1980), as have home ranges and social behaviour of cats living in some association with man (e.g. Dards 1978; MacDonald and Apps 1978; Liberg 1980), but there is a paucity of data on the home ranges of feral cats living in the wild without any contact with man. The Habitat The Hattah-Kulkyne National Park is semiarid and average rainfall (300 mm) is not reliable; summers are hot (often >34OC) and dry, and winters mild (Anon. 1974). The Park has an area of ha and is bounded on the east by the Murray River and on the west by mallee scrub. A series of lakes are fed by the Murray River through a system of overflow creeks and flood plains. The soils are grey clays and *Part 11, Aust. Wildl. Res., 9;

2 E. Jones and B. J. Coman yellow sands; land forms consist of plains, hummocks, dunes and lunettes. Sandy hummocks make up a large area of the park but their original vegetation of slender cypress pine Callitris preissii and mallee has been heavily thinned or removed. They now form open grasslands seasonally covered with poached-egg daisy Myriocephalus stuartii. Black box Eucalyptus largflorens woodland is widespread and found on the drainage areas and flood plains, while the margins of the river, creeks and lakes support stands of red gum E. camaldulensis. Large areas of mallee are present, and there are lesser stands of moonah Melaleuca pubescens, scattered belar Casuarina cristata and buloke C. leuhmanii. The chosen study site for transect counts and home range determinations encompassed some ha of mainly open grassland, but included all the habitat types described including small areas of mallee. Methods Home Range Cats were captured in cage traps and padded-jaw rabbit leghold traps, or treed by a dog and subsequently anaesthetized with ketamine hydrochloride by means of a projectile syringe and blowpipe (Reddacliff 1979). Captured cats were fitted with radiotracking collars transmitting on 151 MHz (AVM Corp., Champaign, Illinois, U.S.A.) and released. Instrumented cats were usually located once a week or once a fortnight durlng daylight, by a telemetry receiver (AVM) and hand-held ZL aerial (Anon. 1976) for distances up to approximately 1 km, and a pole-mounted long Yagi aerial for distances up to approximately 3 km. In addition, during September and October 1980 five instrumented cats were located for up to five consecutive days each week, for eight consecutive weeks. Exact positions were obtained either by locating the cats' refuges, or by visual contact when cats were outside refuges; care being taken not to cause them undue disturbance. When located, the cats' positions were marked on aerial photographs. Instrumented cats were recaptured each 6-8 months for collar replacement. Burt (1943) defined home range as that area traversed by the individual in its normal activities of food gathering, mating and caring for young. The definition was accepted for this study, and home ranges were determined for each cat by joining its outermost location points plotted on aerial photographs (scale 1:25000) to form a convex polygon (Mohr 1947). but ignoring uncharacteristic movements. Activity Pattern A folded colinear omnidirectional aerial (Judd 1979) mounted on a stayed support, was erected in the known home ranges of instrumented cats. A chart recorder (Rustrak, Gulton Industries, U.S.A.) connected to the radiotracklng receiver gave a continuous record of the signal received; continuous variations in the recorded signal strength represented periods of activity. Cats were classified as active or passive, for each hour of the day throughout the monitored periods, by subjective evaluation of the chart trace. However, this method of measuring activity could not discriminate between resting and a passive hunting strategy. Seasonal Abundance The seasonal abundance of feral cats was monitored by transect counts carried out at night with spotlights; cats were detected by their eyeshine. The counts, and low-intensity collections of cats, were made at irregular intervals from a vehicle driven at 15 km h-i along tracks chosen at random through the study area. Either one or two observers with spotlights took part; each observer spotlighted only one side of the track and nightly transect lengths varied from 20 to 80 km. Because the1.e was an overlap of eyeshine colours between cats and foxes Vulpes vulpes (cats' eyeshine usually green, foxes' usually orange), sightings were confirmed where possible by observation of body characteristics with binoculars. Collected cats were classified into three size categories; adults (males of body weight g, females >2400 g*), adolescents (males g, females g), and juveniles (males <2200 g, females < 1900 gt). The physical condition of collected cats was also assessed (Coman et al. 1981) and peritoneal depot fat scored visually on a scale of O a corresponding to an absence *Corresponding to initial attainment of sexual maturity at c. 12 months old (Jones and Coman 1982). tcorresponding to final breakup of family or sibling group at c. 7 months old.

3 Ecology of the Feral Cat. I11 of, to very heavy, deposits of fat. The size of those cats sighted but not collected was estimated. However, since observed cats could not be sexed, all cats judged to have been equivalent in size to adult females were classified as adults. This procedure would have underestimated the number of adolescent males seen. For juveniles, accurate size estimates were easier since at least one specimen was collected from most litters seen. Ages of subadult (adolescent and juvenile) cats were estimated from the growth equations of Rosenstein and Berman (1973); see Jones and Coman (1982) for details. Fox sightings were also recorded, and when a positive distinction between cats and foxes could not be made the sightings were classified as queries. The total number of cats observed per night was obtained by adding together the number of cats observed, and the number of queries multiplied by the ratio of adult cats to adult cats plus foxes observed. This procedure often resulted in the numbers of cats observed being expressed in fractions rather than whole numbers. Transect counts were grouped by season, and relative abundance (A) was expressed as the total number (C) of cats observed per 100 single spotlight-kilometres: A = (100C),L, where L is the transect length in kilometres. The distance at which cats could be seen by spotlight varied with the type of terrain and ground cover. However, the mean transect width for most of the tracks; determined with a rangefinder, was 140 m per side (n = 200 readings). For a given transect count the observed density of cats per square kilometre is given by the expression: 1000CIL W where W is the transect width in metres. Thus observed density is: 10Ail40. Due to their relatively small size and elusive nature, not all cats present in the transect areas would have been detected by spotlight. For estimation of sighting efficiency, use was made of the instrumented cats. During normal transect counts instrumented cats were located (by radio signal strength or foot search) within the transect areas 22 times and were observed by spotlight 11 times, giving a sighting efficiency of 50%. Thus calculated density equals twice the observed density. During the final year of the study nine rabbit transect counts were carried out at approximately 5- week intervals. Four widely separated portions of tracks (total distance 10 km) were selected for the rabbit counts. Spotlight counts of rabbits on both sides of the tracks were commenced immediately after dark, and the four counts were always carried out in the same order. Table 1. Home ranges of instrumented cats in the Hattah-Kulkyne study area Cat Status Monitored period Days No. of Home h-0. when separate range area captured locations (km2) 1 Adolescent male 30.i ix Mature male 21.ii x Mature male 10.iv iv Adolescent male 10.iv xii. I Mature female 19.v i Mature female 19.v iii Results Home Range The home ranges of six cats monitored for 8-21 months are summarized in Table 1. The mean home range areas ( 2 so) were 6.2 f 2.7 km2 for adult males and 1.7 i 1.4 km2 for adult females; the difference was not significant (t = 2.62, P >0.05). Cat 1, a male captured as an adolescent (estimated age 8 months), remained near the capture site for 6 weeks, disappeared after being handled and was relocated 5 weeks later outside the Park, 24 km south of the capture point. It subsequently moved 13 km north into the Park, and then 5 km to its final home range site 27 weeks after capture, after one earlier journey there (Fig. 1). It remained in the area until routine observations ceased in September 1980, and was collected there in March 1981.

4 E. Jones and B. J. Coman

5 Ecology of the Feral Cat Cat 3, a mature male, was never detected outside its home range (Fig. 2) before locations on it ceased in October 1980, but could not be located there in March However cat 4, also a mature male occupying an adjacent home range (Fig. 2), made one excursion of short duration to the east side of cat 3's home range and returned. After recapture in April 1980 cat 4 moved deep into mallee scrub and was not relocated = No. 5 Fig. 3. Daily activity patterns of three cats during the summer in the Hattah-Kulkyne study area, averaged over 15, 7 and 17 days respectively. Arrows indicate the times of sunrise (left) and sunset (right). Cat 5 (Fig. 1) was a male captured as an adolescent (estimated age 7 months). Eight weeks later it made an excursion of 4 weeks' duration to a location 11 km south of the capture point and returned. It also made three trips of 2-12 weeks' duration into mallee scrub adjacent to its home range, two of the trips immediately after recapture. It was still in its home range when collected in March Cats 6 and 7 were both mature females with adjacent home ranges (Fig. I), and each gave birth to a litter during the monitoring period. Cat 6 remained within its home range until January 1981, then disappeared. Cat 7 remained within its home range until collected in March The mean daily straight-line distances ( fs~) between daytime locations during September and October 1980 of five of the instrumented cats were: 0.55 f 0.63 km (cat 1): 1.67 i 0.71 km (cat 3); 0.25 i 0.44 km (cat 5); km (cat 6); 0.06 f 0.11 km (cat 7). The differences in the mean daily distances moved by the cats were significant (F = 52.6, P < 0. Ol), but a regression analysis showed no significant difference based upon sex (P > 0.05).

6 E. Jones and B. J. Coman Activity Pattern Daily summer ( ) activity of two mature males (Nos 1 and 5) and one mature female (No. 7) are illustrated in Fig. 3. Activity was not monitored in other Table 2. Seasonal changes in the relative abundance and population density of feral cats in the Hattah-Kulkyne study area See text for methods used in determining these values. C, number of cats seen; L, transect length in single spotlight-kilometres; A; relative abundance of cats; D; observed density per square kilometre; D,, calculated density (= 20) Year Season C L A D Dc Winter Spring Summer Autumn Winter Spring Summer Autumn Winter Spring Summer Autumn Winter Spring Summer Autumn I I I I I I I I I, W Sp Su A W Sp Su A W Sp Su A W Sp Su A Fig. 4. Seasonal changes in the relative abundance of adult, adolescent (coarse stipple) and juvenile (light stipple) cats over four years, in the Hattah-Kulkyne study area. seasons. The cats were generally more active at night, with periods of greatest activity near the times of sunrise and sunset. Cat 7, a female with two juveniles, had a relatively higher level of daytime activity than the males. Seasonal Abundance and Population Structure Seasonal changes in relative abundance, observed density and calculated density of feral cats over a 4-year period are presented in Table 2. Relative abundance

7 Ecology of the Feral Cat. I showed seasonal fluctuations with summer maxima and winter or spring minima. The calculated density varied from a minimum of 0.34 cats per square kilometre in the 1977 winter to a maximum of 3.5 per square kilometre in the following ( ) summer. The calculated mean density was 0.74 kmp2 in winter and 2.4 kmp2 in summer. Seasonal changes in the population structure also occurred Table 3. Group structure and numbers of cats observed in each size of group in the Hattah-Kulkyne study area N = 367 cats Group structure No. of groups of size: All adult All adolescent All juvenile One adult +adolescents I 0 One adult +juveniles Total number of cats Percentage frequency (Fig. 4). During summer peaks all three size groups were at their maxima, and the mean summer peak comprised 57% adults, 11% adolescents and 33% juveniles. However, few adolescents and juveniles persisted past the autumn, and the winter minima were usually composed only of adults. A shifting phase analysis correlation (Horton and West 1977) between abundance in one season and that in the corresponding season 12 months later showed no significant seasonal pattern for adult cats, but it did for subadults (r = 0.70, P < 0.01). Table 4. Depot fat scores for 93 feral cats collected in the Hattah-Kulkyne Park Depot No. of No. of No. of fat adults non-lactating subadults score adults Total Group Size, Diet and Condition The group sizes and group structure of cats sighted during transect counts are presented in Table 3. The most frequently observed group size was one; 71% of all cats sighted and 90% of all adults sighted were solitary. Only 18% of cats sighted were in groups of three or more, and none of these groups contained more than one

8 416 E. Jones and B. J. Coman adult. Groups of two adults (when confirmed by collection) comprised a male and a female. Family groups (when confirmed by collection) comprised an adult female and subadults; sampled groups of subadults were never older than 7 months (estimated). One female in the company of four adolescents (estimated age 7 months) was lactating, and probably suckling a second litter. A total of 94 cats was collected in the Park, from inside and outside the study area. For adults, 27 were male and 23 were female, for subadults 20 were male and 24 were female; the overall sex ratio was 1 : 1. The diet of 60 adults and adolescents (juveniles were excluded) was obtained as a subsample from a previous dietary study (Jones and Coman 1981). Food intake by weight comprised 82.7% rabbit, 12.5% bird and 2.8% house mouse Mus musculus. Depot fat scores for adult and subadult cats are presented in Table 4. There was no significant difference in mean depot fat scores between all adults and subadults (X2 = 7.13, 3 d.f., P >0.05) but when lactating females were excluded from the adult sample the difference was significant (X2 = 14.88, 3 d.f., P <0.01). Discussion Home Range Various factors affected the accuracy of the home range determinations. Since uncharacteristic movements and migrations were ignored, subjective judgements were made on which locations to include in these determinations, and which to exclude. Also, the cats were more active at night but, because the determinations were based upon daytime locations when they were often in refuges, night-time movements outside their derived home ranges would have gone undetected. This factor would have been most important in underestimating the home range area for cat 7 and least important for cat 3. Additional movements would have been missed because there were often intervals of several weeks between successive locations. The final factor was the variable tracking period for different individuals; in general, longer tracking periods tend to give higher values for home range areas. All instrumented cats maintained home ranges and within each home range particular areas were favoured as daytime refuges. These were invariably areas which provided good shelter: rabbit warrens, hollow trees or logs, or dense thickets. However, all of the cats apart from No. 7 undertook longer exploratory or migratory movements from time to time or shifted their home ranges, and three (Nos 1,4 and 5) disappeared from their home ranges immediately after recapture. In undertaking these movements, the cats displayed good navigational ability; it seems that they were prepared to shift home ranges when necessary, and that an unpleasant experience may be one triggering mechanism. Individual differences in home range sizes were apparent; in general the larger the home range, the greater was the mean daily distance moved between refuges. The two extremes were shown by cat 3, which had the largest home range and the greatest mean daily distance between refuges, and cat 7, which had the smallest and least. This does not necessarily mean that cat 7 had a shorter nightly foraging distance than cat 3, but it does mean that she returned much more frequently to the same or nearby refuges. This difference in her behaviour may have been caused at least partly by the necessity of returning after each trip to her litter (estimated birth month September 1980). The less sedentary, second female (No. 6) lost her litter (estimated birth month October 1980) early.

9 Ecology of the Feral Cat. I11 Population Density The densities of feral cats in the Hattah-Kulkyne Park are broadly comparable to those reported by some other authors. In an agricultural area of New Zealand an observed density of cats per square kilometre (mean values over 2 years) was estimated from spotlight transect counts, and another 9-month survey gave an observed density of 0.74 km-2 (Collins and Charleston 1979). In a New Zealand forest a density of 1.1 km-2 was estimated, based upon sightings over 3 years (Fitzgerald and Karl 1979) and on an American waterfowl refuge 7.8 km-2 based upon trapping (Hubbs 1951). For subantarctic islands higher densities have been recorded: on Kerguelen I. estimated densities varied from 0.5 to 6.7 km-2, depending upon habitat (Derenne 1976); on Hog I. from 2.5 kmp2 in winter to 8 or 9 kmp2 in summer (Derenne and Mougin 1976); on Macquarie I. from 2 to 7 adults per square kilometre (Jones 1977). Although direct comparisons are difficult to make, these densities may be linked to prey availability, because cats were most abundant on those islands which contained large populations of burrow-nesting petrels (a favoured and easily caught prey of cats). On Marion I., where cats are recent colonizers and burrow-nesting petrels still abundant (Anderson and Condy 1974), cats reached densities of 16 km-2 (van Aarde 1981). In contrast, on Campbell I., where the avifauna has been impoverished by cats and rats Rattus norvegicus, but alternative major prey such as rabbits are absent, cats are now rare; only two were seen and 20 scats found in 20 weeks of field work (Dilks 1979). Population Limiting Mechanism Although seasonal fluctuations occurred, the abundance of adult cats remained relatively stable during the study, compared to the much larger fluctuations of subadults. For adults, summer maxima were caused by immigration andlor recruitment of young into the population, and winter or spring minima by emigration or mortality. For subadults, summer maxima were caused by reproduction (Jones and Coman 1982) and autumn or winter minima by emigration, recruitment into the adult population or mortality. During the study the Park rabbit population was characterized by large seasonal fluctuations. Our observations indicated that, usually, numbers were highest in early summer and lowest in early winter. In the final 12 months of the study rabbit abundance varied from a maximum of 30 km-' in November 1980, to minima of 13 km-' in July 1980 and 12 km-' in May 1981 when observations ceased, and because breeding was linked to seasonal conditions the early winter populations consisted mainly of adults. In a concurrent study on the diet of Mallee cats, the intake of rabbit (the major dietary item) was highest during summer and lowest during winter, but the intake of scavenged and other infrequently eaten food peaked in winter (Jones and Coman 1981); this suggested that the availability of rabbit was least during winter. Also, in an experimental rabbit enclosure in New Zealand most rabbits eaten by cats were young and the cats had more difficulty catching and killing adults (Gibb et al. 1978), and on Macquarie I. 81% of rabbits eaten by cats weighed <600 g (Jones 1977). Thus the size structure of rabbit populations, as well as the abundance of rabbits, affects their availability as prey for cats. Fluctuations in the size of the cat population over the 4-year study indicated that the carrying capacity of the environment had been reached, and that a population-

10 418 E. Jones and B. J. Coman limiting mechanism was operating. The annual rates of increase of feral cat populations on two subantarctic islands were 43% for Kerguelen I. (Derenne 1976) and 23% for Marion I. (van Aarde 1978). The natalities of these two island populations were similar to that of the feral cats in this study (Jones and Coman 1982). Since the reproductive capacity existed for a rapid increase in the population, the population-limiting mechanism could have been either emigration, or mortality caused by disease or nutritional stress. Because the surrounding land was either mallee scrub or farmland and carried fewer rabbits than the Park, emigration would have been disadvantageous and balanced at least by immigration. Disease was not likely to have been a major cause of adult mortality, but its effect on juveniles was not known (Coman et al. 1981). Because juveniles were not usually seen outside their lairs until 2 months old, mortality factors operating before this time would have gone undetected. Cat 6 lost her litter of five within 2 months of their birth; disease may have been the cause, but no conclusive data was obtained. However, some evidence was found of nutritional stress in cats. The physical condition of most adults sighted appeared good, but several subadults in poor condition were seen. Collected cats which showed signs of a nutritional stress were either lactating females or subadults. Instrumented adult cats maintained good health and showed little variation in weight during the study, and the two caught as adolescents displayed steady weight gains. Cat 1 grew from 2500 g at capture to 6050 g 25 months later, and cat 5 grew from 2300 g at capture to 5100 g 20 months later. However, another cat instrumented as a juvenile (estimated age 5 months) died after 3 weeks. Since rabbit was the staple food and was usually less available during winter, we suggest that the major mechanism limiting the cat population was mortality caused by an autumn-winter nutritional stress acting particularly on subadults. On Hog I. and Macquarie I. the feral cat populations were also limited by the winter availability of rabbit (Derenne and Mougin 1976; Jones 1977). Social Organization The elusive nature, low population density and solitary existence of the cats prevented any detailed observations on their behaviour, but the survival rate of instrumented cats and the persistence of their home ranges indicated a stable social organization for the population at large, and a mechanism which effectively spaced out individuals. Presumably this involved some form of territorial behaviour but no information on its operation was obtained. No overlap of home ranges occurred between two adjacent males (Nos 3 and 4) or between the two adjacent females (Nos 6 and 7), but insufficient animals were tracked to establish whether or not this was the general situation. Other adult cats were occasionally seen in the home ranges of instrumented cats during spotlight transects, but their sexes and status (resident or transient) could not be determined. However, some overlap of home ranges was indicated by the population density; although the mean home range area was 4 km2 per cat (with a 1 : 1 sex ratio), the calculated mean winter density gave an exclusive area per cat of 1.4 km2. The social organization of some other solitary felids has been studied; both mountain lions Felis concolor and bobcats Lynx rufus maintain stable land-tenure systems based upon prior right, marking and avoidance mechanisms. Male mountain lions have almost exclusive home areas, while those of females overlap those of males and each other (Seidensticker et al. 1973); the home ranges of female bobcats are almost exclusive while those of males overlap those of females and each other (Bailey 1974).

11 Ecology of the Feral Cat. I11 Cats may either be completely feral, solitary and at low densities, as in this study, or live at much higher densities and in various associations with man. In a group of four semi-dependant farm cats the home range areas of three females varied between 2 and 7 ha and that of the male was 60 ha; the animals functioned as a social group with various stable relationships between individuals (Macdonald and Apps 1978). For a group of urban feral cats living in the Portsmouth Dockyards, the density was 200 per km2 and there was a plentiful supply of food (mainly human food scraps) and shelter. Three-quarters of the females lived in social groups with at least one other female, and their home ranges varied from 0.03 to 4.2 ha. Males had larger, overlapping home ranges which varied in size from 0.8 to 24 ha (Dards 1978). Liberg (1980) made a detailed study of spacing patterns in a population of rural freeroaming house cats, which also contained some feral cats. Male feral cats maintained evenly distributed, partly overlapping home ranges 24 km across and were dominant. Domestic males had smaller home ranges and were subordinate to the feral males, while the more sedentary domestic females often lived in social groups and shared smaller joint home ranges. Territorial behaviour was suggested by marking (urine spraying, defaecating, pole-clawing and cheek rubbing) and by the stability of the home ranges. Acknowledgments We wish to acknowledge the technical assistance of Mrs M. Driesen and Mr D. Franklin; Mr J. Edmonds assisted in aerial design and Rangers of the Victorian National Parks and Wildlife Service collected some cats. Mr C. Ernst determined the daily movements of cats, Dr B. Horton carried out statistical tests, and constructive comments were made by the referee. The study was partly financed from funds provided by the Australian National Parks and Wildlife Service. References Anderson, G. D., and Condy, P. R. (1974). A note on the feral house cat and house mouse on Marion Island. S. Afr. J. Antarct. Res. 4: Anon. (1974). Report on the Mallee study area. Land Conservation Council of Victoria. (Government Printer: Melbourne.) Anon. (1976). 'Radio Communications Handbook.' Vol. 2, pp (Radio Society of Great Britain: London.) Bailey, T. N. (1974). Social organizations in a bobcat population. J. Wildl. Manage. 38, Burt, W. H. (1943). Territoriality and home range concepts as applied to mammals. J. Mammal. 24, Collins, G. H., and Charleston, W. A. G. (1979). Studies on sarcocystis species: 1. Feral cats as definitive hosts for sporozoa. N.Z. Vet. J. 27, 804. Coman, B. J., Jones, E. H., and Westbury, H. A. (1981). Protozoan and viral infections of feral cats. Aust. Vet. J. 57, Dards, J. L. (1978). Home ranges of feral cats in Portsmouth Dockyard. Carniv. Genet. Newsl. 3, Derenne, Ph. (1976). Notes sur la biologie du chat haret de Kerguelen. Mammalia 40, Derenne, Ph., and Mougin, J. L. (1976). Donnees ecologiques sur les mammiferes introduits de I'ile aux Cochons, Archipel Crozet (46'06'S, 50 14'E). Mammalia 40; Dilks, P. J. (1979). Observations on the food of feral cats on Campbell Island. N.Z. J. Ecol. 2, Fitzgerald, B. M., and Karl, B. J. (1979). Foods of feral house cats (Felis catus L.) in forest of the Orongorongo Valley; Wellington. N.Z. J. Zool. 6, Gibb, J. A., Ward, C. P., and Ward, G. D. (1978). Frequency, food and behaviour of predators. In 'Natural control of a population of rabbits, Oryctolagus cuniculus (L.), for ten years in the Kourarau enclosure'. N.Z. DSIR Bull. No. 223, pp

12 E. Jones and B. J. Coman Horton, B. J., and West, C. E. (1977). Shifting phase analysis correlation in the determination of reproducibility of biological rhythms. Its use in the study of patterns of rabbit and rat feeding. J. Interdiscipl. Cycle Res. 8, Hubbs, E. L. (1951). Food habits of feral house cats in the Sacramento Valley. Calif Fish Game Jones, E. (1977). Ecology of the feral cat, Felis catus (L.), (Carnivora: Felidae) on Macquarie Island. Aust. Wildl. Res. 4, Jones, E., and Coman, B. J. (1981). Ecology of the feral cat Felis catus (L.) in south-eastern Australia. I. Diet. Aust. Wildl. Res. 8, Jones, E., and Coman, B. J. (1982). Ecology of the feral cat Felis catus (L.) in south-eastern Australia. 11. Reproduction. Aust. Wildl. Res. 9, Judd, F. (1979). VHFRJHF folded colinear aerial array. Pract. Wireless Apr. 1979, pp Liberg, 0. (1980). Spacing patterns in a population of rural free roaming domestic cats. Oikos Macdonald, D. W., and Apps, P. J. (1978). The social behaviour of a group of semi-dependant farm cats. Felis carus: A progress report. Carniv. Genet. Neusl. 3, Mohr, C. 0. (1947). Table of equivalent populations of North American small mammals. Am. Midl. Nut. 37, Pascal, M. (1980). Structure et dynamique de la population de chats harets de l'archipel des Kerguelen. Mammalia Reddacliff, G. L. (1979). Home-made projectile syringes. N.Z. Vet. J. 27, Rosenstein, L., and Berman, E. (1973). Postnatal body weight changes of domestic cats maintained in an outdoor colony. Am. J. Vet. Res. 34, Seidensticker, J. C., IV., Hornocker, M. G., Wiles, W. V., and Messick, J. P. (1973). Mountain lion social organization in the Idaho Primitive Area. Wildl. Monogr. 35, Van Aarde, R. J. (1978). Reproduction and population ecology in the feral house cat Felis catus on Marion Island. Carniv. Genet. hreusl Van Aarde, R. J. (1981). Distribution and density of the feral house cat Felis catus at Marion Island. S. Afr. J. Antarct. Res. 9, Manuscript received 8 September 198 1; accepted 28 January 1982