TY - JOUR
AU - Miller, Richard, S.
AB - The wood mouse, Apodemus sylvaticus, is common throughout the British Isles and Europe and eastward into Asia (Matthews, 1952). Ecologically it is one of the most adaptable and important of the European small mammals but, like most common species, it has attracted very little attention from ecologists. The wood mouse has been studied far less than its American counterpart, Peromyscus maniculatus, although detailed comparisons between the two would be extremely valuable as a guide to the nature of the population dynamics and community relationships of small mammals. This study was done in a five-acre copse known as “The Pasticks,” 4½ miles northwest of Oxford in Wytham Woods, Berkshire. The Pasticks is a mixed-wood copse; it is surrounded by arable fields and was chosen for its relative isolation from adjacent woodlands. The dominant vegetation of the copse is an oak-ash-sycamore association (Quercus robur—Fraxinus excelsior—Acer pseudo-pUtanus) modified by plantings of elm (Ulmus campestris), birch (Betula alba), beech (Fagus sylvatica) and sweet chestnut (Castanea sativa). A long history of natural development without management has allowed the formation of a relatively primitive woodland. The dense canopy and deep shade of this woods is reflected in the simplicity of the shrub and field layers of vegetation. Hazel (Corylus avellana) and elder (Sambucus nigra) occur scattered through the woods and the field layer is composed almost entirely of Dog's Mercury (Mercurialis perennis) and nettle (Urtica dioica). The edge of the copse is bordered by an unkept hawthorne hedge (Crataegus oxyacanthoides) which has grown to heights of 10 and 12 feet in places. Most of the seed foods produced in the Pasticks are palatable to wood mice (Miller, 1954) and 1950 seemed to be an exceptionally good year for seed production. The seeds of oak, beech and chestnut seemed to be selected by the mice and disappeared rather early from the floor of the woods, but sycamore seeds were abundant on the ground for several months and hawthorne fruits were still present at the start of the 1951 growing season. It can safely be assumed that food supplies for the mice were abundant, even during the winter of 1950–51. Methods A grid of 41 squares, each 25 yards on a side, was laid out as shown in Fig. 1. Each of the 41 squares consisted in turn of 25 smaller squares (1×1 yd. ) in which traps were set according to a schedule of random trap placements predetermined from random number tables. The entire area was trapped once each month with two Longworth live traps (Chitty and Kempson, 1949) placed at random within each of the 41 squares. The traps were prebaited with oats for two days, set at about 5 pm on the second day, and visited at about 9 am on the third day. The animals were marked with monel rings (Chitty, 1937) and released, after appropriate data on sex, weight and reproductive condition were recorded. Fig. 1 Open in new tabDownload slide Pasticks trapping grid. Arable fields surrounding the copse are indicated by cross-hatching. Fig. 1 Open in new tabDownload slide Pasticks trapping grid. Arable fields surrounding the copse are indicated by cross-hatching. Although the technique of prebaiting is designed to overcome “trap-shyness,” many small mammals, and especially Apodemus, may also become “trapaddicted.” If the mice become accustomed to finding food in traps which are left in place for several days, they are not only stopped before they can reveal the extent of their home range, as pointed out by Chitty (1937), but certain individuals may also exclude others from capture ( Chitty and Kempson, 1949; Tanaka, 1951). Although Brown (1954) found that a large percentage of the mice he trapped did not return to traps that were left in place for several days, preliminary experiments during this study showed that some individuals do indeed return repeatedly to the same trap site. To overcome sampling errors of this sort, two traps were placed at each position, the trap sites were selected randomly, trapping periods were short and the trapping campaigns were widely spaced. Results 1. Relative numbers of mice, voles and shrews. The numbers of mice, voles and shrews trapped each month are shown in Table 1. Voles (Clethrionomys glareolus) and shrews (Sorex araneus) seemed to be especially abundant during the months of May, June, July and August, partly because of a real increase in their numbers but perhaps also because of the nature of their activity. Wood mice (Apodemus sylvaticus) are more nocturnal than either voles or shrews (Crowcroft, 1954; Miller, 1955) and their movements are less affected by the presence or absence of protective cover. During the winter, when there is no ground layer of vegetation and protective cover is sparse, voles, and probably shrews as well, are seldom trapped any distance from logs, bramble patches, or stands of bracken (Evans, 1942; Miller, 1955); but during the spring and summer, when the ground is covered with dense stands of Dog's Mercury and nettle, they are able to range over greater areas under the protective cover of vegetation. Moreover, as mice do not seem to take advantage of cover as voles do, their activity rhythms are more sensitive to changes in day length and their activity is more restricted by short summer nights (Miller, 1955). Consequently, traps set at random in an area such as the Pasticks will have a higher probability of capturing voles and shrews in summer than in winter, and the results shown in Table 1 probably reflect changes in activity pattern as well as seasonal trends of abundance. Table 1 Numbers of Apodemus sylvaticus, Clethrionomys glareolus and Sorex araneus trapped from January, 1950 to March, 1951 MONTH . Apodemus . Clethrionomys . Sorex . TOTAL . 1950 Jan. 28 6 1 35 Feb. no sample Mar. 20 6 0 26 Apr. 18 8 0 26 May 19 10 6 35 June 22 12 10 44 July 2 14 6 22 Aug. 22 18 3 43 Sept. 48 4 1 53 Oct. 48 4 6 58 Nov. 48 3 0 51 Dec. 8 2 0 10 1951 Jan. 40 4 4 48 Feb. no sample Mar. 30 3 0 33 Total 352 94 37 483 MONTH . Apodemus . Clethrionomys . Sorex . TOTAL . 1950 Jan. 28 6 1 35 Feb. no sample Mar. 20 6 0 26 Apr. 18 8 0 26 May 19 10 6 35 June 22 12 10 44 July 2 14 6 22 Aug. 22 18 3 43 Sept. 48 4 1 53 Oct. 48 4 6 58 Nov. 48 3 0 51 Dec. 8 2 0 10 1951 Jan. 40 4 4 48 Feb. no sample Mar. 30 3 0 33 Total 352 94 37 483 Open in new tab Table 1 Numbers of Apodemus sylvaticus, Clethrionomys glareolus and Sorex araneus trapped from January, 1950 to March, 1951 MONTH . Apodemus . Clethrionomys . Sorex . TOTAL . 1950 Jan. 28 6 1 35 Feb. no sample Mar. 20 6 0 26 Apr. 18 8 0 26 May 19 10 6 35 June 22 12 10 44 July 2 14 6 22 Aug. 22 18 3 43 Sept. 48 4 1 53 Oct. 48 4 6 58 Nov. 48 3 0 51 Dec. 8 2 0 10 1951 Jan. 40 4 4 48 Feb. no sample Mar. 30 3 0 33 Total 352 94 37 483 MONTH . Apodemus . Clethrionomys . Sorex . TOTAL . 1950 Jan. 28 6 1 35 Feb. no sample Mar. 20 6 0 26 Apr. 18 8 0 26 May 19 10 6 35 June 22 12 10 44 July 2 14 6 22 Aug. 22 18 3 43 Sept. 48 4 1 53 Oct. 48 4 6 58 Nov. 48 3 0 51 Dec. 8 2 0 10 1951 Jan. 40 4 4 48 Feb. no sample Mar. 30 3 0 33 Total 352 94 37 483 Open in new tab 2. Seasonal trends in the Apodemus population. a. Sex ratio.—Elton et al. (1931) observed remarkably consistent variations in the sex ratio of Apodemus during two consecutive seasons; the proportion of males rose to its highest in late summer and early autumn and dropped to equality with the females during winter and spring. Evans (1942), on the other hand, found only slight but irregular variations in sex ratio and his data show no consistent trends. The number of males and females captured each month during this study are shown in column 2 of Table 2. When tested for chi-square the relative numbers of males and females were found to diverge from unity only in the September sample, although the values for October are nearly significant and should be considered part of the same trend (Snedecor, 1956). The preponderance of males during these two months is similar to the situation reported by Elton et al. (1931), although it occurred slightly later in the season than was the case in their study. Elton et al. (1931) considered that this phenomenon might be associated with breeding activity and a tendency for the males to wander more and therefore be more liable to capture at this time of the year. Males were more active and moved greater distances than females during September and October and may have been more liable to capture, but they were also more active than females during March and April when the sex ratios were equal. Table 2 Seasonal changes in abundance and activity of Apodemus sylvaticus from January, 1950 to March, 1951 MONTH . TOTAL CAPTURES . number/100 trap nights . KNOWN POPULATION . NUMBER /ACRE . PER CENT RECAPTURES . PER CENT FEMALES REPRODUCTIVE . ♂♂ . ♀♀ . 1950 Jan. 14 14 37.3 28 5.3 0.0 0.0 Feb. no sample Mar. 10 10 26.3 22 4.2 65.0 44.4 Apr. 9 9 24.3 21 4.0 88.9 75.0 May 8 11 28.8 19 3.6 78.9 90.0 June 8 13 31.4 22 4.2 28.6 84.6 July 1 1 3.2 7 1.3 0.0 100.0 Aug. 8 14 36.1 24 4.5 10.4 85.7 Sept. 31 17 62.3 51 9.6 12.2 70.6 Oct. 30 18 66.7 54 10.2 29.2 11.8 Nov. 21 27 60.8 58 10.9 41.7 0.0 Dec. 5 3 10.0 39 7.4 75.0 0.0 1951 Jan. 23 17 54.0 44 8.3 72.5 5.3 Feb. no sample Mar. 19 11 38.0 30 5.7 83.3 61.5 MONTH . TOTAL CAPTURES . number/100 trap nights . KNOWN POPULATION . NUMBER /ACRE . PER CENT RECAPTURES . PER CENT FEMALES REPRODUCTIVE . ♂♂ . ♀♀ . 1950 Jan. 14 14 37.3 28 5.3 0.0 0.0 Feb. no sample Mar. 10 10 26.3 22 4.2 65.0 44.4 Apr. 9 9 24.3 21 4.0 88.9 75.0 May 8 11 28.8 19 3.6 78.9 90.0 June 8 13 31.4 22 4.2 28.6 84.6 July 1 1 3.2 7 1.3 0.0 100.0 Aug. 8 14 36.1 24 4.5 10.4 85.7 Sept. 31 17 62.3 51 9.6 12.2 70.6 Oct. 30 18 66.7 54 10.2 29.2 11.8 Nov. 21 27 60.8 58 10.9 41.7 0.0 Dec. 5 3 10.0 39 7.4 75.0 0.0 1951 Jan. 23 17 54.0 44 8.3 72.5 5.3 Feb. no sample Mar. 19 11 38.0 30 5.7 83.3 61.5 Open in new tab Table 2 Seasonal changes in abundance and activity of Apodemus sylvaticus from January, 1950 to March, 1951 MONTH . TOTAL CAPTURES . number/100 trap nights . KNOWN POPULATION . NUMBER /ACRE . PER CENT RECAPTURES . PER CENT FEMALES REPRODUCTIVE . ♂♂ . ♀♀ . 1950 Jan. 14 14 37.3 28 5.3 0.0 0.0 Feb. no sample Mar. 10 10 26.3 22 4.2 65.0 44.4 Apr. 9 9 24.3 21 4.0 88.9 75.0 May 8 11 28.8 19 3.6 78.9 90.0 June 8 13 31.4 22 4.2 28.6 84.6 July 1 1 3.2 7 1.3 0.0 100.0 Aug. 8 14 36.1 24 4.5 10.4 85.7 Sept. 31 17 62.3 51 9.6 12.2 70.6 Oct. 30 18 66.7 54 10.2 29.2 11.8 Nov. 21 27 60.8 58 10.9 41.7 0.0 Dec. 5 3 10.0 39 7.4 75.0 0.0 1951 Jan. 23 17 54.0 44 8.3 72.5 5.3 Feb. no sample Mar. 19 11 38.0 30 5.7 83.3 61.5 MONTH . TOTAL CAPTURES . number/100 trap nights . KNOWN POPULATION . NUMBER /ACRE . PER CENT RECAPTURES . PER CENT FEMALES REPRODUCTIVE . ♂♂ . ♀♀ . 1950 Jan. 14 14 37.3 28 5.3 0.0 0.0 Feb. no sample Mar. 10 10 26.3 22 4.2 65.0 44.4 Apr. 9 9 24.3 21 4.0 88.9 75.0 May 8 11 28.8 19 3.6 78.9 90.0 June 8 13 31.4 22 4.2 28.6 84.6 July 1 1 3.2 7 1.3 0.0 100.0 Aug. 8 14 36.1 24 4.5 10.4 85.7 Sept. 31 17 62.3 51 9.6 12.2 70.6 Oct. 30 18 66.7 54 10.2 29.2 11.8 Nov. 21 27 60.8 58 10.9 41.7 0.0 Dec. 5 3 10.0 39 7.4 75.0 0.0 1951 Jan. 23 17 54.0 44 8.3 72.5 5.3 Feb. no sample Mar. 19 11 38.0 30 5.7 83.3 61.5 Open in new tab b. Breeding.—Column 7 of Table 2 shows the percentage of the females in each month's sample that showed external evidence of reproductive activity. The criteria used were a swollen vulva, perforate vagina, lactation or visible pregnancy. These criteria have certain rather obvious shortcomings, but are sufficient to establish a rough index of breeding activity and the extent of the breeding season. The breeding season in wood mouse populations usually begins in March, reaches a peak in June and July, and ends by October or November, although cessation of breeding through the winter is not an invariable rule. During a three-year study in Bagley Wood near Oxford, Baker (1930) and Elton et al (1931) found that breeding ceased for six months during the first winter, during two months of the second winter and only during one month of the third winter. There was no apparent correlation between winter breeding and environmental conditions, but their data do suggest that it might have accompanied a yearly increase in population density. The data shown in Table 2 are similar to those reported by Baker (1930) for the second year of his study. None of the animals trapped in January, 1950, was in breeding condition but by March the testes of all of the males were scrotal and 44.4 per cent of the females were reproductive. By April, 75.0 per cent of the females had perforate vaginas or swollen vulvas, although none was lactating or visibly pregnant, and by May, when the first juvenile was trapped, 90.0 per cent of the females were breeding. By this time many were lactating or pregnant. Breeding continued into October and had apparently ended by November, but by January 5.3 per cent of the females showed signs of reproductive activity and the proportion had increased to 61.5 per cent by March, 1951. Baker and Ranson (1932)a, b) have shown that light, temperature and nutrition control breeding in Microtus agrestis and we may logically assume that this combination of factors is also important in the breeding cycle of Apodemus. Of these factors, nutrition seems most likely to have been an important variable responsible, in part at least, for the extended breeding season of 1950–51. A second factor should also be considered, however; small mammal numbers were relatively high in Wytham Woods in 1950 and 1951 and it is quite possible that the same correlation between population density and breeding that was noted by Elton et al. (1931) existed during this study. c. Age structure and mortality.—Apodemus seldom live more than a year in the field and the overwintering population that is established by late fall consists almost entirely of young born that year (Elton et al., 1931; Evans, 1942). This fact is evident in the recapture data in column 6 of Table 2. After the initial marking in January, 1950, the percentage of recaptures increased to 88.9 per cent by April but declined from May onwards as the population was diluted by young of the breeding season. The unusually low number of captures in July was apparently due to a mass migration into the grain fields adjacent to the north edge of the copse. During this period the mice moved as far as 50 yards away from the woods and into the fields, harvesting the ripe grain. Five of the mice captured in August had been marked during the preceding months, however, showing that the July exodus was to some extent temporary, even though the period between June and August was apparently one of high mortality. Recaptures became more frequent in October and in subsequent months as breeding declined and the overwintering population became established. Of the mice marked between January and July, 1950, only five were recovered as late as August, two were still present in September, and none was taken in October or subsequent months and the October population consisted, therefore, of animals born in 1950, probably during the latter part of the breeding season. Changes in weight distribution from August, 1950 to March, 1951 are shown in Table 3. and Fig. 2. Baker (1930) concluded that mice weighing 15 grams or more may be classed as adults, although it does not necessarily follow that those weighing less than 15 grams are sexually immature. Wood mice may breed when only a few months old and it was common in the present study to find pregnant or lactating females that weighed 15 grams or less; the smallest lactating female weighed 12.0 grams and one 7.5-gram animal had a perforate vagina. Nevertheless, 15.0 grams seems to be the most convenient dividing point between adult and immature mice, and the data in Table 3 and Fig. 2 describe the trends in age distribution that occurred in the overwintering mouse population. Fig. 2 Open in new tabDownload slide Weight distributions of male and female Apodemus from August, 1950 to March, 1951. Fig. 2 Open in new tabDownload slide Weight distributions of male and female Apodemus from August, 1950 to March, 1951. Table 3 Weight distribution and age structure of population from August, 1950 to March, 1951 MONTH . N . − 10 gm. . 11–15 . 16–20 . 21–25 . 26+ . Males: 1950 Aug. 7 28.6% 28.6% 42.8% Sept. 31 25.8 58.1 12.9 3.2% Oct. 27 33.3 63.0 3.7 Nov. 20 90.0 10.0 Dec. 5 100.0 1951 Jan. 20 80.0 20.0 Feb. Mar. 16 6.2 12.5 43.8 37.5% Females: 1950 Aug. 13 15.4 15.4 23.1 46.1 Sept. 17 17.6 35.3 29.4 11.8 5.9 Oct. 17 41.2 47.1 11.8 Nov. 25 32.0 64.0 4.0 Dec. 3 66.7 33.3 1951 Jan. 19 100.0 Feb. Mar. 13 7.8 92.9 MONTH . N . − 10 gm. . 11–15 . 16–20 . 21–25 . 26+ . Males: 1950 Aug. 7 28.6% 28.6% 42.8% Sept. 31 25.8 58.1 12.9 3.2% Oct. 27 33.3 63.0 3.7 Nov. 20 90.0 10.0 Dec. 5 100.0 1951 Jan. 20 80.0 20.0 Feb. Mar. 16 6.2 12.5 43.8 37.5% Females: 1950 Aug. 13 15.4 15.4 23.1 46.1 Sept. 17 17.6 35.3 29.4 11.8 5.9 Oct. 17 41.2 47.1 11.8 Nov. 25 32.0 64.0 4.0 Dec. 3 66.7 33.3 1951 Jan. 19 100.0 Feb. Mar. 13 7.8 92.9 Open in new tab Table 3 Weight distribution and age structure of population from August, 1950 to March, 1951 MONTH . N . − 10 gm. . 11–15 . 16–20 . 21–25 . 26+ . Males: 1950 Aug. 7 28.6% 28.6% 42.8% Sept. 31 25.8 58.1 12.9 3.2% Oct. 27 33.3 63.0 3.7 Nov. 20 90.0 10.0 Dec. 5 100.0 1951 Jan. 20 80.0 20.0 Feb. Mar. 16 6.2 12.5 43.8 37.5% Females: 1950 Aug. 13 15.4 15.4 23.1 46.1 Sept. 17 17.6 35.3 29.4 11.8 5.9 Oct. 17 41.2 47.1 11.8 Nov. 25 32.0 64.0 4.0 Dec. 3 66.7 33.3 1951 Jan. 19 100.0 Feb. Mar. 13 7.8 92.9 MONTH . N . − 10 gm. . 11–15 . 16–20 . 21–25 . 26+ . Males: 1950 Aug. 7 28.6% 28.6% 42.8% Sept. 31 25.8 58.1 12.9 3.2% Oct. 27 33.3 63.0 3.7 Nov. 20 90.0 10.0 Dec. 5 100.0 1951 Jan. 20 80.0 20.0 Feb. Mar. 16 6.2 12.5 43.8 37.5% Females: 1950 Aug. 13 15.4 15.4 23.1 46.1 Sept. 17 17.6 35.3 29.4 11.8 5.9 Oct. 17 41.2 47.1 11.8 Nov. 25 32.0 64.0 4.0 Dec. 3 66.7 33.3 1951 Jan. 19 100.0 Feb. Mar. 13 7.8 92.9 Open in new tab The August and September populations were composed of animals from a wide range of age groups, but with a greater proportion of females than males in the adult age classes. By October the proportion of older mice had decreased until there were no males or females in the 21–25 gram class and very few in the 16–20 gram class. From November onwards there was a gradual shift in weight distributions to successively heavier weight classes; all of the mice captured in December weighed more than 11 grams and all of those captured in January weighed over 16 grams, although the weights of the males were distributed over a wider range of classes than were those of the females. d. Seasonal abundance.—The values for the number of mice per 100 trap-nights in column 3 of Table 2 were obtained by first subtracting the number of traps occupied by voles or shrews from the total of 82 trapnights and using the remaining number of available trapnights to calculate values for Apodemus. It was assumed, in other words, that the voles and shrews pre-empted traps that might otherwise have been occupied by mice, thereby reducing the number of possible trapnights for Apodemus. The values for “known population” were obtained by adding the number of mice trapped in a sample to the number also known, through subsequent captures, to be in the study area. For example, if a mouse was marked in March and was subsequently captured in May but not in April, it was included in the known population for all three months. This presumes of course that the animal did not leave the trapping area, or that immigration and emigration were balanced. The values for density in column 5 are based on the known population relative to the 5.3 acres covered by the trapping grid, rather than the actual area of the copse. The high turnover of individuals from May through September made it difficult to estimate mortality rates or population density during that period; estimates obtained with the Lincoln Index method were considered unreliable and not particularly useful. It is felt, however, that the values for known population are a reasonably reliable index to abundance, in spite of the fact that they are minimum values. The known population density varied from 3.6 mice per acre in May to 10.9 per acre in November, with a mean of 6.1 per acre for the total period of study, if we disregard the density of 1.3 per acre in July, since this was apparently an unusual situation. Brown (1954) reported an average density of 8.5 per acre during an 18-month period with a maximum of 17.2 per acre in November, in a comparable study of Apodemus populations. His results showed a sharp increase between September and November and an overwintering decline from November until the following breeding season. The data from the present study show a similar trend; the lowest population density was reached at a time when the breeding season had begun and juveniles were replacing adults in the population. The rate of breeding and replacement was sufficient to maintain a fairly steady density of about four mice per acre until August. Between August and November there was a sharp increase in density, followed by an overwintering decline to approximately the same density that was observed in the preceding spring. The critical density for small mammal populations is usually reached in spring when the adult population must be large enough to provide the replacement rate that is necessary in order for the population to survive a high summer mortality and to establish a new and vigorous overwintering population (Leslie and Hanson, 1940; Elton, 1942). Apodemus populations are not especially erratic in their density variations from year to year and it would appear that they have achieved a remarkably close adjustment to the forces of natural control. They are capable of a high rate of increase, and yet overpopulation does not seem to be a feature of their ecology. 3. Population activity patterns. As Elton, et al (1931) point out, most census techniques measure combined changes in both the density and activity of a population. Changes in the activity rhythm and in the spatial patterns of activity of a species are invariably reflected in trapping data, and it is essential that these changes and their effects be accounted for. Moreover, it is also important to establish which of the observed activity changes are regular, seasonal phenomena and which are unpredictable in the sense that they are not correlated with normal, seasonal variations in physical and biotic factors. a. Recapture rate.—The frequency of captures of males and females is shown in Table 4. A comparison of the percentages of males and females captured from one to six times shows no significant differences between the sexes; thus 40.8 per cent of the males and 40.7 per cent of the females were recaptured and the maximum number of captures was six for one male and five for three females, even though the males usually moved greater distances during the intervals between samples. Table 4 Frequency of recaptures . NUMBER OF TIMES CAPTURED . . POPULATION . 1 . 2 . 3 . 4 . 5 . 6 . TOTAL . Males: Number 61 19 13 5 4 1 103 Per cent 59.2 18.4 12.6 4.8 3.9 1.0 Females: Number 54 17 11 6 3 0 91 Per cent 59.3 18.7 12.1 6.6 3.3 . NUMBER OF TIMES CAPTURED . . POPULATION . 1 . 2 . 3 . 4 . 5 . 6 . TOTAL . Males: Number 61 19 13 5 4 1 103 Per cent 59.2 18.4 12.6 4.8 3.9 1.0 Females: Number 54 17 11 6 3 0 91 Per cent 59.3 18.7 12.1 6.6 3.3 Open in new tab Table 4 Frequency of recaptures . NUMBER OF TIMES CAPTURED . . POPULATION . 1 . 2 . 3 . 4 . 5 . 6 . TOTAL . Males: Number 61 19 13 5 4 1 103 Per cent 59.2 18.4 12.6 4.8 3.9 1.0 Females: Number 54 17 11 6 3 0 91 Per cent 59.3 18.7 12.1 6.6 3.3 . NUMBER OF TIMES CAPTURED . . POPULATION . 1 . 2 . 3 . 4 . 5 . 6 . TOTAL . Males: Number 61 19 13 5 4 1 103 Per cent 59.2 18.4 12.6 4.8 3.9 1.0 Females: Number 54 17 11 6 3 0 91 Per cent 59.3 18.7 12.1 6.6 3.3 Open in new tab b. Seasonal variations in movement—Table 5 shows the average distances moved during the year; these values are averages of the distance moved by each individual between the point of capture for the month shown and the last previous point of capture. In most cases the interval between captures was one month but in others the interval may have been as much as three months. Although the varying length of these intervals is a possible source of error, there was no correlation between the distance moved by an individual and the length of the interval between captures. These data are illustrated in Fig. 3. Fig. 3 Open in new tabDownload slide Average distance (yards) moved between the month shown and the last previous capture. Males, solid Une; females, dotted line. Fig. 3 Open in new tabDownload slide Average distance (yards) moved between the month shown and the last previous capture. Males, solid Une; females, dotted line. Table 5 Average monthly distances moved, based on the distance between the point of capture for the month shown and the last previous point of capture MONTH . MALES . FEMALES . Number of individuals . Average distance (yards) . Number of individuals . Average distance ( yards ) . 1950 Jan. Feb. Mar. 6 34.5 6 46.2 Apr. 7 55.1 8 31.2 May 5 88.0 10 29.8 June 2 81.5 4 30.8 July Aug. 3 15.0 Sept. 3 96.0 2 25.0 Oct. 7 47.9 7 17.0 Nov. 12 20.8 8 20.0 Dec. 4 28.2 2 46.0 1951 Jan. 17 33.5 12 15.2 Feb. Mar. 14 31.1 10 24.8 MONTH . MALES . FEMALES . Number of individuals . Average distance (yards) . Number of individuals . Average distance ( yards ) . 1950 Jan. Feb. Mar. 6 34.5 6 46.2 Apr. 7 55.1 8 31.2 May 5 88.0 10 29.8 June 2 81.5 4 30.8 July Aug. 3 15.0 Sept. 3 96.0 2 25.0 Oct. 7 47.9 7 17.0 Nov. 12 20.8 8 20.0 Dec. 4 28.2 2 46.0 1951 Jan. 17 33.5 12 15.2 Feb. Mar. 14 31.1 10 24.8 Open in new tab Table 5 Average monthly distances moved, based on the distance between the point of capture for the month shown and the last previous point of capture MONTH . MALES . FEMALES . Number of individuals . Average distance (yards) . Number of individuals . Average distance ( yards ) . 1950 Jan. Feb. Mar. 6 34.5 6 46.2 Apr. 7 55.1 8 31.2 May 5 88.0 10 29.8 June 2 81.5 4 30.8 July Aug. 3 15.0 Sept. 3 96.0 2 25.0 Oct. 7 47.9 7 17.0 Nov. 12 20.8 8 20.0 Dec. 4 28.2 2 46.0 1951 Jan. 17 33.5 12 15.2 Feb. Mar. 14 31.1 10 24.8 MONTH . MALES . FEMALES . Number of individuals . Average distance (yards) . Number of individuals . Average distance ( yards ) . 1950 Jan. Feb. Mar. 6 34.5 6 46.2 Apr. 7 55.1 8 31.2 May 5 88.0 10 29.8 June 2 81.5 4 30.8 July Aug. 3 15.0 Sept. 3 96.0 2 25.0 Oct. 7 47.9 7 17.0 Nov. 12 20.8 8 20.0 Dec. 4 28.2 2 46.0 1951 Jan. 17 33.5 12 15.2 Feb. Mar. 14 31.1 10 24.8 Open in new tab The average distance moved by females did not fluctuate greatly during the year; there seemed to be a slight decrease in activity as the breeding season progressed, but this trend was not pronounced. The increase shown for December was based on the movements of only two animals and is not reliably significant. The activity of the males, however, showed striking and rather consistent changes. The average distance moved was more than doubled between March and May, 1950, and long movements continued from May until September. Following September there was a decrease to an average distance of about 20 yards per individual in November and a slight increase from November to January and March, 1951, when the average distances were nearly equal to those recorded for March, 1950. Even though these values are based on rather small samples, the observed changes are so pronounced that they suggest a definite, seasonal trend. The changes in the activity patterns of the males occurred during the breeding season but did not coincide with the breeding condition of the males; by March, 1950, all the males had apparently begun reproductive activity, although the height of the female season did not occur until May. The most obvious correlation is with the breeding condition of the females; when the average distance moved by males increased from 34 yards in March to 55 yards in April, the percentage of females in breeding condition increased from 44.4 to 75.0 per cent. By May, 90 per cent of the females were breeding and the distances moved by males had increased to an average of 88 yards. During the period when male activity remained high the reproductive activity of the females was also high and as the percentage of reproductive females decreased from about 70 per cent in September to 12 per cent in October, and breeding finally ceased in November, the average distances moved by the males decreasd from 96 yards in September to 48 yards in October, and finally to a low value of about 21 yards in November. While further research is needed before this apparent corelation can be established as a causal relationship, it does suggest that the breeding condition of the females might possibly control the activity patterns of the males as well as the reproductive cycle of the population. c. Home range and observed dispersal—Of the various methods for measuring and calculating home range that have been described, the “Observed Range Length” method is the most applicable to these data. This method assumes that the distance between the most widely separated capture sites is equivalent to the diameter of the home range. Although Stickel (1954) has shown that the “Adjusted Range Length” method, in which half the distance to the next trap site is added to the observed range length, is more accurate, it and many other available methods cannot be used with these data because of the practice of using random rather than fixed trap positions. The average observed range lengths and calculated home ranges for mice captured three or more times are shown in Table 6. Brown (1956) studied the movements of Apodemus and reported average observed range lengths of 64.6 yards for males and 57.6 yards for females; corresponding values from the present study are 66.5 ± 9.75 yards for males and 38.4 ± 5.24 yards for females. Using Manville's (1949) method for calculating trap-revealed range, Brown obtained values of 3092 yards square for male and 2628 yards square for female home ranges; home ranges calculated from observed range lengths in the present study were approximately 3472 yards square for males and 1158 for females. Thus the data obtained by Brown (1956) for male Apodemus are roughly comparable to these, but his values for females are considerably higher. Table 6 Observed range lengths and home ranges of A. sylvaticus captured three or more times OBSERVATION OF . MALES . FEMALES . Range length (yards): Mean 66.5 ± 9.75 38.4 ± 5.24 Range (11.0–187.0) (5.0–88.0) Home range (yards2): Mean 3471.5 1157.5 Range (95.0–27450.7) (19.6–6079.0) Home range (acres): Mean 0.72 0.24 Range (0.02–5.67) (0.004–1.26) OBSERVATION OF . MALES . FEMALES . Range length (yards): Mean 66.5 ± 9.75 38.4 ± 5.24 Range (11.0–187.0) (5.0–88.0) Home range (yards2): Mean 3471.5 1157.5 Range (95.0–27450.7) (19.6–6079.0) Home range (acres): Mean 0.72 0.24 Range (0.02–5.67) (0.004–1.26) Open in new tab Table 6 Observed range lengths and home ranges of A. sylvaticus captured three or more times OBSERVATION OF . MALES . FEMALES . Range length (yards): Mean 66.5 ± 9.75 38.4 ± 5.24 Range (11.0–187.0) (5.0–88.0) Home range (yards2): Mean 3471.5 1157.5 Range (95.0–27450.7) (19.6–6079.0) Home range (acres): Mean 0.72 0.24 Range (0.02–5.67) (0.004–1.26) OBSERVATION OF . MALES . FEMALES . Range length (yards): Mean 66.5 ± 9.75 38.4 ± 5.24 Range (11.0–187.0) (5.0–88.0) Home range (yards2): Mean 3471.5 1157.5 Range (95.0–27450.7) (19.6–6079.0) Home range (acres): Mean 0.72 0.24 Range (0.02–5.67) (0.004–1.26) Open in new tab Davis (1953) has discussed several reasons for using the frequency distribution of distances between captures in the analysis of small mammal movements. While this method cannot be used to establish the home range of an individual and is perhaps less useful than most other methods for the analysis of home range as such, it.does provide a useful index to the nature of the movements of a population and avoids some of the difficulties inherent in other methods. Table 7 shows the observed values for frequency of capture at different distances and an adjusted frequency distribution based on the probability of capture at each interval of distance. As shown in column 6, there is a curve of increasing and decreasing probability of capture with increasing distance; in most cases this will be a normal curve (Davis, 1953), but the curve for these data is skewed and slightly bimodal because of the shape of the trapping grid. There was a marked increase in the probability of capture between the intervals of 0 to 20 yards and 21 to 40 yards, and a gradual decrease in probability at distances greater than 100 yards. There were no opportunities for capture at distances of 200 yards or more. Table 7 Frequency of capture at different distances DISTANCE (yards) . OBSERVED FREQUENCY . PROBABILITY OF CAPTURE, % . ADJUSTED FREQUENCY . Males . Females . N . % . N . % . Males, % . Females, % . 0–20 22 28.6 35 48.6 4.1 53.3 73.4 21–40 28 36.4 24 33.3 10.8 25.7 19.1 41–60 11 14.3 7 9.7 15.2 7.2 4.0 61–80 7 9.1 4 5.6 14.0 5.0 2.5 81–100 4 5.2 2 2.8 16.6 2.4 1.0 101–120 2 2.6 13.1 1.5 121–140 1 1.3 11.6 .8 141–160 7.7 161–180 2 2.6 4.8 4.1 181–200 2.1 201+ 0.0 DISTANCE (yards) . OBSERVED FREQUENCY . PROBABILITY OF CAPTURE, % . ADJUSTED FREQUENCY . Males . Females . N . % . N . % . Males, % . Females, % . 0–20 22 28.6 35 48.6 4.1 53.3 73.4 21–40 28 36.4 24 33.3 10.8 25.7 19.1 41–60 11 14.3 7 9.7 15.2 7.2 4.0 61–80 7 9.1 4 5.6 14.0 5.0 2.5 81–100 4 5.2 2 2.8 16.6 2.4 1.0 101–120 2 2.6 13.1 1.5 121–140 1 1.3 11.6 .8 141–160 7.7 161–180 2 2.6 4.8 4.1 181–200 2.1 201+ 0.0 Open in new tab Table 7 Frequency of capture at different distances DISTANCE (yards) . OBSERVED FREQUENCY . PROBABILITY OF CAPTURE, % . ADJUSTED FREQUENCY . Males . Females . N . % . N . % . Males, % . Females, % . 0–20 22 28.6 35 48.6 4.1 53.3 73.4 21–40 28 36.4 24 33.3 10.8 25.7 19.1 41–60 11 14.3 7 9.7 15.2 7.2 4.0 61–80 7 9.1 4 5.6 14.0 5.0 2.5 81–100 4 5.2 2 2.8 16.6 2.4 1.0 101–120 2 2.6 13.1 1.5 121–140 1 1.3 11.6 .8 141–160 7.7 161–180 2 2.6 4.8 4.1 181–200 2.1 201+ 0.0 DISTANCE (yards) . OBSERVED FREQUENCY . PROBABILITY OF CAPTURE, % . ADJUSTED FREQUENCY . Males . Females . N . % . N . % . Males, % . Females, % . 0–20 22 28.6 35 48.6 4.1 53.3 73.4 21–40 28 36.4 24 33.3 10.8 25.7 19.1 41–60 11 14.3 7 9.7 15.2 7.2 4.0 61–80 7 9.1 4 5.6 14.0 5.0 2.5 81–100 4 5.2 2 2.8 16.6 2.4 1.0 101–120 2 2.6 13.1 1.5 121–140 1 1.3 11.6 .8 141–160 7.7 161–180 2 2.6 4.8 4.1 181–200 2.1 201+ 0.0 Open in new tab The adjusted frequency distributions shown in Table 7 and Fig. 4 show a strong tendency for females to restrict their movements to short distances. Males were captured with a frequency of 53.3 per cent and females 73.4 per cent at distances between 0 and 20 yards and 86.2 per cent of the movements by males and 96.5 per cent of the movements by females were restricted to distances of 60 yards or less. In each case there is a strongly descending frequency curve with increasing distance, showing that a very large proportion of the movements of both sexes were confined to rather short distances; but none of the females moved more than 100 yards whereas more than 6 per cent of the movements recorded for males were of distances greater than this. Fig. 4 Open in new tabDownload slide Adjusted frequency distribution of number of movements at each interval of distance shown. Fig. 4 Open in new tabDownload slide Adjusted frequency distribution of number of movements at each interval of distance shown. In order to compare the results of different studies, using the frequency distribution method of analysis, the data from each study must be converted to adjusted frequencies based on probability of capture. Since the data reported for comparable species have not included this adjustment such comparisons cannot be made at present. It should perhaps be emphasized that an observed frequency which is not adjusted for probability of capture has very little meaning; it is especially biased for movements of short distances (Davis, 1953) and does not account for the effects of different distances between trap sites or the size and shape of the trapping grids. The same is true for home range values; differences in method affect the results, and the significance of a value for home range depends on the degree of which it represents the behavior of the total population. Davis (1953) noted that his data included dispersal as well as home range movements and the same is undoubtedly true of the data shown in Table 7. This problem has also been noted by other workers. Brown (1956), for example, states that “stay at homes” as well as “wanderers” were present in the populations he studied and that both contributed to the wide range in the types of movement he examined. He also found, however, that 73 per cent of the mice moved less than 30 yards, 92 per cent less than 60 yards and that 96 per cent moved less than 90 yards from one day's trapping to the next. He found no evidence that any of the mice had moved to new positions within their home ranges or had adopted new home ranges during any single trapping period. Evans (1942) recorded the movements of 194 Apodemus that survived for a month or more on his trapping plots in Bagley Woods and found that only 75 per cent confined their movements to distances of 100 yards or less and that movements of over 400 yards occasionally occurred. He recorded one case in which a distance of 240 yards was covered in one day. It would appear from the latter study that rather long movements are more common than might be supposed from the data presented by Brown and the present study, although the data do not show which of the movements were dispersal and which were home range. If we consider the shape of the curves shown in Fig. 4, and the fact that 86.2 per cent of the males and 96.5 per cent of the females moved less than 60 yards during the entire period of study, it becomes obvious that the occasional long movements that were recorded had a disproportionate effect on the size of the calculated home ranges. This fact is also evident in the results presented by Brown ( 1956); 92 per cent of the population moved less than 60 yards between trappings, yet the average home ranges were calculated to be 64.6 yards square for males and 57.6 yards square for females. It appears, therefore, that the home ranges given in both of these studies are overestimates of the true home range, since they are based on dispersal as well as home range movements. At least three typical patterns of movement can be distinguished in recapture data, as illustrated by selected examples in Fig. 5. Typical home range patterns were shown by two mice (No. 419♂ and 498♀ ) that confined all of their movements to well-defined areas. A second pattern is shown in the movements of two mice (No. 342♂ and 12♀ ) which were initially captured at points some distance from the areas of most of their captures; these mice apparently dispersed to new home ranges between their first and second captures and subsequently stayed within their new home ranges during the remainder of the observations. The male (No. 342) moved 163 yards between August and September and then moved 40,15,17 and 15 yards between subsequent captures. The female (No. 12) moved 100 yards between the time it was marked in January and its second capture in March; it then moved 43 yards between March and April, 32 yards between April and May, and by September when it was last captured it had only moved 22 yards from where it was taken in May. This pattern, with several possible variations, would appear to represent a combination of dispersal and home range movements; it is the sort of pattern that might be shown when animals are crowded out of one area and into another or when they are attracted to a new home range by more favorable factors such as food supply or shelter. A third pattern is illustrated by two males (No. 11 and 34) which invariably moved rather long distances between each capture. Number 11 was first trapped in the Pasticks in March, 1950, and was captured again in April, about 170 yards from its initial capture-site, in a copse south of the study area. By May it had returned to the Pasticks, 90 yards from its second point of capture, having moved across a plowed field and back again in its travels. Number 34 likewise showed a pattern of relatively long movements of 61,45,132 and 102 yards between its inital capture in January, 1950, and its last capture in June. This third pattern of apparently aimless wanderings seems to represent a continued series of dispersal movements. In this study it was only shown by adult males during the breeding season and is perhaps best explained as a search by unmated males for breeding females and a nest site. Fig. 5 Open in new tabDownload slide Selected examples of patterns of movement. The ring number of each individual is shown beside a circle indicating its initial point of capture; dotted lines connect subsequent points of capture. Fig. 5 Open in new tabDownload slide Selected examples of patterns of movement. The ring number of each individual is shown beside a circle indicating its initial point of capture; dotted lines connect subsequent points of capture. Thus the following types of movement seem to be included in these data: (1) home range, (2) home range + dispersal and (3) dispersal. An example of the error that can be introduced into home range values by including dispersal movements in the calculations is shown by a comparison between the calculated home ranges and range lengths of animals 32 and 12 before and after eliminating dispersal movements from the data. If we consider that the first movements of each animal were dispersal, the following values for all movements (a) and home range movements only (b) are obtained: ANIMAI. NO. . AVERAGE DISTANCE MOVED . OBSERVED HANGE LENGTH . HOME RANGE (yds.2) . HOME RANGE ( acres ) . 32 (a) 50.0 147.5 17101.8 3.53 (b) 21.8 21.5 350.4 0.07 12 (a) 49.2 102.5 8247.5 1.70 (b) 32.3 40.0 1256.0 0.26 ANIMAI. NO. . AVERAGE DISTANCE MOVED . OBSERVED HANGE LENGTH . HOME RANGE (yds.2) . HOME RANGE ( acres ) . 32 (a) 50.0 147.5 17101.8 3.53 (b) 21.8 21.5 350.4 0.07 12 (a) 49.2 102.5 8247.5 1.70 (b) 32.3 40.0 1256.0 0.26 Open in new tab Although no attempt has been made to treat all of the data in this way, it is obvious that the results would be far different if such treatment could be consistently applied. As yet there is no objective method for distinguishing between home range and dispersal in recapture data. Davis (1953) noted an empirical relationship between frequency of capture and distance, such that a straight line results when this relationship is plotted on double logarithm paper. “A mathematical model that represents this relation is frx=c, or log f + x log r= log c, which is a straight line of negative slope x″(Davis, 1953). In this expression, f represents frequency of capture, r represents distance and c is a constant. This line descends and bends sharply toward the r axis (f=0), so that the value of r at the bend might be related in a constant manner to the size of the home range. The data necessary to generalize this relationship have not been collected, but there is an obvious need for further research on this problem and for establishing an objective method for distinguishing between home range and dispersal in recapture data. Summary A population of Apodemus sylvaticus was studied by live-trapping from January, 1950, to March, 1951, in a five-acre copse in Wytham Woods, Berkshire. The 1950 breeding season extended from March to October with a peak in May, and the 1951 breeding season began in January. The early start of the 1951 breeding season is attributed to abundant food supplies and, possibly, to a relatively high population density in the fall of 1950. Except for a temporary migration of mice out of the study area in July, the population reached its lowest density of 3.6 per acre in May and a maximum density of 10.9 per acre in November, and the overwintering population that was established by November consisted entirely of mice born during the 1950 breeding season. The average home range of males captured three or more times was 3742 yards square, or 0.72 acres, and of females 1158 yards square, or 0.24 acres. The average distances moved each month by females were relatively constant, but males moved far greater distances during the period from May to September than at other times of the year. This increased activity seemed to be correlated with the breeding condition of the female population. Three types of movement were noted in the recapture data: (1) home range, (2) home range + dispersal and (3) dispersal. Although there is, as yet, no objective method for distinguishing between home range and dispersal from recapture data, the latter is extremely important in the activity patterns of the population and influences estimates of home range considerably. Acknowledgments The author gratefully acknowledges the kind assistance of Mr. Charles Elton who supervised this research. The research was done at Oxford University in partial fulfillment of the requirements of the D.Phil. degree and was made possible by a Fulbright Award. Literature Cited Baker J. R. 1930 . The breeding season in British wild mice . Proc. Zool. Soc. Lond. , 96 : 113 – 126 . Google Scholar OpenURL Placeholder Text WorldCat Baker J. R. Ranson R. M. . 1932a . Factors affecting the breeding of the field-mouse (Microtus agrestis). Part 1—Light . Proc. Roy. Soc. Lond., HOB : 313 – 332 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Baker J. R. Ranson R. M. 1932b . Factors affecting the breeding of the field-mouse (Microtus agrestis). Part II—Temperature and food . Proc. Roy. Soc. Lond. , 112B : 39 – 46 . Google Scholar Crossref Search ADS WorldCat Brown L. E. 1954 . Small mammal populations at Silwood Park Field Center, Berkshire, England . Jour. Mamm. , 35 : 161 – 176 . Google Scholar Crossref Search ADS WorldCat Brown L. E. 1956 . Movements of some British small mammals . Jour. Anim. Ecol. , 25 : 54 – 71 . Google Scholar Crossref Search ADS WorldCat Chitty D. 1937 . A ringing technique for small mammals . Jour. Anim. Ecol. , 6 : 36 – 53 . Google Scholar Crossref Search ADS WorldCat Chitty D. Kempson D. A. . 1949 . Prebaiting small mammals and a new design of live trap . Ecology , 30 : 536 – 542 . Google Scholar Crossref Search ADS WorldCat Crowcroft P. 1954 . The daily cycle of activity in British shrews . Proc. Zool. Soc. Lond. , 123 : 715 – 729 . Google Scholar Crossref Search ADS WorldCat Davis D. E. 1953 . Analysis of home range from recapture data . Jour. Mamm. , 34 : 352 – 358 . Google Scholar Crossref Search ADS WorldCat Elton C. 1942 . Voles, mice and lemmings . Clarendon Press , Oxford , 496 pp. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Elton C. Baker J. R. Ford E. B. Gardner A. D. . 1931 . The health and parasites of a wild mouse population . Proc. Zool. Soc. Lond. , 95 : 657 – 721 . Google Scholar OpenURL Placeholder Text WorldCat Evans F. C. 1942 . Studies of a small mammal population in Bagley Wood, Berkshire . Jour. Anim. Ecol. , 11 : 182 – 197 . Google Scholar Crossref Search ADS WorldCat Leslie P. H. Ranson R. M. . 1940 . The mortality, fertility, and rate of natural increase of the vole (Microtus agrestis) as observed in the laboratory . Jour. Anim. Ecol. , 9 : 27 – 52 . Google Scholar Crossref Search ADS WorldCat Manville R. H. 1949 . A study of small mammal populations in northern Michigan . Misc. Publ. Mus. Zool., Univ. Mich. , No. 73 , pp. 1 – 83 . Google Scholar OpenURL Placeholder Text WorldCat Matthews L. H. 1952 . British mammals . Collins , London , 410 pp. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Miller R. S. 1954 . Food habits of the wood-mouse (Apodemus sylvaticus) and the bank vole (Clethrionomys glareolus) in Wytham Woods, Berkshire . Saugetier-kundliche Mitteilungen , 2 : 109 – 114 . Google Scholar OpenURL Placeholder Text WorldCat Miller R. S. 1955 . Activity rhythms in the wood mouse, Apodemus sylvaticus and the bank vole, Clethrionomys glareolus . Proc. Zool. Soc. Lond. , 125 : 505 – 519 . Google Scholar Crossref Search ADS WorldCat Snedecor G. W. 1956 . Statistical methods . The Iowa State College Press , Ames, Iowa . 5th ed., 534 pp. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Stickel L. F. 1954 . A comparison of certain methods for measuring ranges of small mammals . Jour. Mamm. , 35 : 1 – 15 . Google Scholar Crossref Search ADS WorldCat Tanaka R. 1951 . Estimation of vole and mouse populations on Mt. Ishizuchi and on the uplands of southern Shikoku . Jour. Mamm. , 32 : 450 – 458 . Google Scholar Crossref Search ADS WorldCat 1958 American Society of Mammalogists
TI - A Study of a Wood Mouse Population in Wytham Woods, Berkshire
JF - Journal of Mammalogy
DO - 10.2307/1376784
DA - 1958-11-20
UR - https://www.deepdyve.com/lp/oxford-university-press/a-study-of-a-wood-mouse-population-in-wytham-woods-berkshire-fWvsAxuPjq
SP - 477
EP - 493
VL - 39
IS - 4
DP - DeepDyve
ER -