The Scale of Demographic Heterogeneity in a Population of Peromyscus leucopus

International Association for Ecology

The Scale of Demographic Heterogeneity in a Population of Peromyscus leucopusAuthor(s): D. T. Krohne and A. B. BurginSource: Oecologia, Vol. 82, No. 1 (1990), pp. 97-101Published by: Springer in cooperation with International Association for EcologyStable URL: http://www.jstor.org/stable/4219207 .

Accessed: 04/09/2013 14:05

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Springer and International Association for Ecology are collaborating with JSTOR to digitize, preserve andextend access to Oecologia.

http://www.jstor.org

This content downloaded from 147.8.204.164 on Wed, 4 Sep 2013 14:05:11 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=springer

http://www.jstor.org/action/showPublisher?publisherCode=intecol

http://www.jstor.org/stable/4219207?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Oecologia (1990) 82:97-101 prigerVerOg190 ? Springer-Verlag 1990

The scale of demographic heterogeneity in a population of Peromyscus leucopus D.T. Krohne' and A.B. Burgin2 1 Department of Biology, Wabash College, Crawfordsville, IN 47933, USA 2 Biology Department, Indiana University, Bloomington, IN 47405, USA

Summary. The patterns of spatial heterogeneity in density and demography were studied in a population of Peromys- cus leucopus inhabiting a deciduous forest in west-central Indiana. A series of 9 live-trapping grids sampled densities from 3 spatial scales: 3 ha, 80 ha and 350 km2. We found high levels of variation within all three spatial scales. There was as much variation within a single, large grid as within the entire Sugar Creek Valley (350 km2). The patterns of density variation were not temporally stable and thus represent variation in demographic processes rather than simply differences in habitat quality. We suggest that the Sugar Creek Valley represents a metapopulation of Peromyscus leucopus composed of a number of temporally unstable sub- populations. This structure provides considerable demographic variation that dispersing animals can exploit.

Key words: Peromyscus - Scaling - Demography - Popula- tion regulation

Most population ecologists would agree that densities of many species are not spatially homogeneous. Empirical ex- amples of spatial demographic heterogeneity abound (West 1979; Krohne 1982; Krebs 1966; Garrett 1982; e.g.). How- ever, while many studies describe spatial heterogeneity, few document its scale. Community ecologists have in recent years recognized the importance of understanding the spatial scale over which community phenomena are analyzed (Wiens and Rottenberry 1980; Wiens 1976; 1980). The question of scaling has received less attention from population ecologists (but see Morris 1987). We suggest that the scaling of demographic variation has considerable theoretical and practical importance in population biology for at least two reasons. First, many theoretical treatments including those of population genetics (Smith et al. 1978), the evolution of dispersal (Roff 1974), and the process of population regulation (Lidicker 1978; Anderson 1980), depend upon the existence of spatial heterogeneity. The scaling of the variation is at least implicitly important in these treatments because it determines the proximity and quality of exploitable ecological, genetic and social situations.

Second, we suggest that at least some of the continuing confusion surrounding the regulation of numbers in populations is a product of our ignorance of the scaling of spatial

demographic parameters. Not only is population regulation a product of many factors (Lidicker 1988), it is complicated by spatial variation in microhabitat and the relative importance of regulatory factors. For few populations have we identified the boundaries of demographically homogeneous regions. Conclusions drawn from studies in which the patterns of spatial demographic variation are unknown may be biased by the unwitting inclusion of more than one demographic situation in the study area. We do not know how many population studies fall into this category. How- ever, if many studies are thus affected it would explain why determination of the relative importance of various regulatory mechanisms has been a difficult process.

This study was designed to address the following questions of demographic scaling for a population of Peromys- cus leucopus: 1) over what spatial scales are populations demographically homogeneous? 2) how does density and the variance in density change as a function of area? 3) to what extent are homogeneous regions temporally stable? and 4) what are the demographic consequences of local variation in density? We addressed the questions outlined above with a nested analysis of demographic variation on three spatial scales: 350 kM2, 80 ha, and 3 ha. For each of these scales, we tested the null hypothesis that demographic variables are homogeneous over space. That is, at any given time, there are no significant differences among samples from a particular scale.

Materials and methods

Study sites and trapping protocols

The study was conducted from 1980 to 1985 in a band of continuous deciduous forest approximately 35 km long and 10 km wide along the Sugar Creek Valley in west-central Indiana. The forest in this region is dominated by American beech (Fagus grandifolia) and sugar maple (Acer saccharum). All trapping sites were in structurally similar forest habitats although there was slight variation among sites in disturbance history and the degree of dominance by beech and maple.

Demographic variation was analyzed with nine live- trapping grids distributed throughout the Sugar Creek Val- ley (Table 1). Trapping protocol was as outlined in Krohne and Baccus (1985). These grids sampled from three spatial scales (Level I: Grid 1-3 ha; Level II: Allee Woods- 80 ha; Level III: Sugar Creek Valley- 350 km2). The sizes of the Offprint requests to: D.T. Krohne



98

Table 1. Locations and sizes of live-trapping grids. Levels refer to spatial scales sampled. Maximum distance is the greatest distance between sampling points of that scale

Grids Area (ha) # Traps Location Maximum distance

Level I Grid 1 3.0 105 Allee Mem. Woods (13 km NW Rockville, Parke Co.) 230 M

Level II Grid 1 3.0 105 Allee Mem. Woods 750 m Grid 2 2.9 125 Allee Mem. Woods Grid 3 4.9 211 Allee Mem. Woods Grid 4 2.4 55 Allee Mem. Woods Grid 5 2.4 80 Allee Mem. Woods

Level III Grid 1 3.0 105 Allee Mem. Woods 35 km Grid 4 2.4 55 Allee Mem. Woods Grid 5 2.4 80 Allee Mem. Woods Grid 6 1.9 80 32 km SW Crawfordsville, Montgomery Co., IN Grid 7 0.5 35 5 km NW Crawfordsville, Montgomery Co., IN Grid 8 1.9 80 10 km SW Crawfordsville, Montgomery Co., IN Grid 9 1.9 80 Turkey Run State Park, Parke Co., IN

grids varied slightly to conform to topographic and vegeta- tion patterns (Table 1).

Spatial demographic analysis

The null hypothesis of demographic homogeneity was tested within each of the three spatial scales. We define demographic homogeneity as demographic similarity in samples taken from within a particular area (eg. within Allee Memorial Woods); heterogeneity refers to variation among sites within a particular area. For all three scales, demographic heterogeneity was analyzed as the deviation of observed local densities from that expected if the mean density for the entire area were uniform. For Grid 1, the local variation was measured by calculating demographic parameters for the entire grid and then for four 0.75 ha subdivisions of the grid (similar results were obtained for subdivision into halves and thirds).

Demographic heterogeneity within the larger scales (the Sugar Creek Valley and Allee Memorial Woods) could result from variation within local grids or variation between the grids sampling that scale. To analyze this we compared the variance in density within Grid 1 with that among the grids sampling Allee Memorial Woods or the Sugar Creek Valley. If the variance within the larger scale is greater than that within Grid 1, it implies that the demographic heterogeneity in the larger scale is due to differences among grids. If it is the same as that within Grid 1, it implies that the heterogeneity in the larger scale is due to local variation such as that within Grid 1.

Because of the large variation in density between years, comparisons of demography among scales were made only where data from both scales were available from the same year. Grid 1 was live-trapped in all six years of the study. The variance in density there was compared with that on the scale of Allee Woods in 1980, 1981 and 1982. The grids sampling the Sugar Creek Valley were trapped in 1984 and 1985. Thus, they were compared with Grid 1 in those years. Data from 1983 were omitted because removal trapping on other grids in the area may have confounded the results. Peak summer densities were used for all analyses presented here. However, similar results were obtained for densities achieved in other months.

At spatial scales smaller than one fourth of Grid 1 (0.75 ha) density and variance in density become mean-

ingless or at least assessed with very small samples. How- ever, at these scales one can analyze the dispersion of individuals. The dispersion of home ranges on Grid 1 was analyzed with a technique developed by Morisita (1959) to identify aggregations of individuals. Morisita's index (Id)

was calculated for home range centers (Hayne, 1949) for each of four scales based on quadrats increasing in size from 232 m2 to 2438 M2.

Results

The population on Grid I that was live-trapped in all six years of the study showed typical Peromyscus density changes characterized by annual density fluctuations with peak density occurring in the period July-September (Fig. 1).

Summer density varied considerably within all three spatial scales. (Figure 2). The heterogeneity within scales was significant in eight of ten comparisons for the three scales. One of the two non-significant comparisons was for variation within Grid 1 in 1983, a very low density year which resulted in very small samples sizes for that comparison. On the scale of the Sugar Creek Valley, the highest density grid had up to four times the density of the lowest density grid in some years. At the 80 ha scale (Allee Woods) two- fold differences in density among grids were observed. Sub- division of Grid 1 resulted in variation in density among

,. ..... .... ..i:: :: , , , ,-,,,.

1980 1981 1982 1983 1984 I~~~~~~~~~~~~~~~ . . , . . . . .. .. . .. ..............t1 | ........

n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 10 il . .. . . . . . . . . . .j

,.i.', '..l,.,,,, . I

111'.-....................... ! .1 !A.-!.!.: ................................. I

* -*\ 0 A ' !~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. .. .... .. ................,....'. 1 ! ~~~~~~~~~~~~~. . . . . . . -- .

Fig. 1. Density changes on Grid 1. Stippled areas represent winter months



Table 2. Variance in density on different spatial scales

Year Level I Level II Level III df F Grid I Allee Sugar

Woods Creek Valley

1980 10.75 3.10 - 3, 4 3.47 n.s. 1981 51.70 44.63 - 3, 4 1.16 n.s. 1982 7.84 7.78 - 3, 4 1.01 n.s. 1984 6.10 - 3.99 3, 6 1.58 n.s. 1985 24.50 - 36.90 6, 3 1.51 n.s.

subplots that was similar in magnitude to that on the larger spatial scales.

None of the comparisons of variances in density between Grid I and the larger scales (Allee Woods and Sugar Creek) showed significant variation (Table 2; Fmaxtest; Sokal and Rohlf, 1969). That is, there was as much variation within Grid 1 as within all of Allee Woods or the entire Sugar Creek Valley. Also note that the variance was lowest in 1984 when densities were low everywhere in the Sugar Creek Valley.

For the finer scale comparisons within Grid 1 measured with Morisita's index, centers of home ranges were randomly distributed on a quadrats from 232 m2 to 2438 M2. For no size quadrat and in no year was Id significantly different from 1.0. In a few cases (for example, 1983) there was an apparent tendency toward clumping at the 232 m2 and 930 m2 scales but small sample sizes precluded statisti- cal significance. Thus, at the small scales measured with this analysis, home ranges were randomly distributed. Giv- en the large variation in Grid 1 density with time, these data indicate no density-dependent change in the fine-scale dispersion of home ranges. Indeed, there was no correlation between density and Id (232 m2:r=-0.267, P>0.05; 536 m2:r=0.013, P>0.05; 929 m2:r= -0.245, P>0.05; 2438 m2: - = -0.01 1, P > 0.05).

For the larger scales within which there was significant heterogeneity, the patterns of variation in density were not temporally stable. For example, note in Fig. 2A that Grids I and 6 were high in 1984. The next year, Grids 6 and 8 were very high while Grid I was the lowest density grid. Changes in the location of high densities were seen at the scale of Allee Woods (Fig. 2B) and within Grid 1 (Fig. 2C). In 1980 and 1981 Area D of Grid 1 had the highest density while Areas A and C had the highest density in subsequent years. These data indicate that the spatial variation is a result of demographic processes rather than

simply microhabitat selection. If the latter were causing the variation, one would expect the highs to be located con- sistently in the same place unless microhabitat structure changes markedly between years. We assume that this is unlikely.

Since the heterogeneity was temporally unstable, dispersal and/or reproduction must have subsequently affected the high and low areas differentially. We examined the patterns of recruitment in the highs and the lows in the months of the breeding season remaining after the high or low was detected. Over the remainder of the breeding season, the proportion of juvenile recruits in the highs and lows was similar (Table 3). Notice that while the highs produced more recruits in terms of absolute numbers, the total recruitment rate (newly marked animals per resident female) was 3.6 times higher in the lows. The recruitment rates of

99

2 5-

20 -

1 5 9

5-

1231456 1245 124578

1 90 191 1 9

2 5

1 5

1 0

1980 1981 1982

theSugrCCee VAlley. B B Vaito within Ale MeoralWods

C Variation within Grid 1. Chi square values indicating significant differences between grids are denoted by * and ** for differences at the 0.05 and 0.01 levels respectively. Locations of grids are listed in Table I

Table 3. Recruitment to high and low density sections of Grid I during the portion of the breeding season remaining after the highs and lows were detected

Recruits Recruits High Juvenile Recruits Recruits Low Juvenile per adult proportion recruits per adult proportion recruits female juvenile per adult female female juvenile per adult female

recruits recruits

1980 2 0.20 0.50 0.10 3 0.75 0.00 0.00 1981 10 0.55 0.60 0.33 13 3.25 0.46 1.50 1982 13 1.44 0.46 0.67 7 1.75 0.57 0.44 1983 17 2.33 0.13 0.33 2 1.00 0.50 0.50 1984 11 1.83 0.73 1.33 7 7.00 0.86 6.00

Total 43 0.59 0.51 0.48 32 2.13 0.53 1.13



100

juveniles showed the same tendeny for increased propor- tional recruitment in the lows (Table 3). Although the relative importance of dispersal and in situ reproduction is not known, the effect must be that the lows will gradually increase in density relative to the highs.

Discussion

There was variation in the density of Peromyscus leucopus on several spatial scales in this study. In addition, the patterns and scale of the heterogeneity were not temporally stable. The fine-scale heterogeneity within Grid 1 was strik- ing. However, this fine-scale heterogeneity is in fact consis- tent with a number of other studies that show that important population parameters including density, gene fre- quency, and life history parameters often vary over relatively small distances (Adler, 1987; Krohne, 1982; Krohne and Baccus, 1985; Bowen, 1982; Patton and Feder, 1981). Ap- parently, many important population processes are local in effect.

We feel that these data may be important in helping to understand the process of population regulation. For many years, single factors were sought to explain changes in the numbers of animals. Proponents of single factor regulation accumulated evidence of the importance of factors such as weather, predation, food supply, behavior, and genetics. Recently, interest has increased in multi-factorial models of regulation (Lidicker 1988; Krebs 1985). The find- ing that there is considerable fine-scale demographic heterogeneity (this study, Cockburn and Lidicker 1983), makes sense to us in light of the current interest in multi-factorial models of regulation. It seems likely that it is actually the metapopulation (Gill 1978) that is regulated by multiple factors. The metapopulation shows demographic heterogeneity. It is composed of a number of smaller homogeneous demographic units that vary independently in spatial scale and with time. We suggest that the fine-scale homogeneity in these units is a consequence of the local effect of a single, overriding regulatory factor such as an abundance of resources or lack of predators. The density of the metapopulation thus depends on the summed trends of these smaller homogeneous units. This model can explain not only the existence of demographic variation but the ability of different investigators to obtain evidence of regulation by many different single factors. Studies that isolate primar- ily a single demographic unit reveal the effects of single factors. Also explained is the difficulty of quantitatively testing multi-factorial models given the detailed information that would be required from the entire metapopulation.

Clearly, the spatial extent of continuous habitat, the vagility of the species and the effectiveness of dispersal bar- riers determine the patterns of demographic heterogeneity comprising a metapopulation. In the case of the Sugar Creek Valley, there is a large band of continuous Peromys- cus habitat surrounded by agricultural fields. Dispersal from the forest is probably difficult. The result is a large, complex metapopulation of Peromyscus. Our data indicate that there is as much demographic variation within Grid 1 as there is between grids sampling larger scales. Thus, fine- scale local variation, such as that within Grid 1, plays an important role in the heterogeneity we observed within the metapopulation of the Sugar Creek Valley. This suggests that extrinsic or intrinsic regulatory factors operate locally. For example, a mast year for a single canopy oak may

be responsible for local demographic differences such as those observed within Grid 1.

These data are also relevant to the problem of dispersal. Virtually no juvenile males remain in their natal area (Wolff et al. 1988). Several theoretical as well as empirical treatments of dispersal suggest that some individuals disperse not because they are forced out of the local population, but because they realize genetic or economic advantages from emigration (Bekoff, 1977; Lidicker, 1975; Wolff et al. 1988). These juveniles must seek a demographic or social situation into which they can integrate. Previous work at Allee Memorial Woods has shown that Peromyscus are opportunistic dispersers that exploit any available low density areas (Krohne and Miner, 1985). Young Peromyscus appar- ently have considerable demographic heterogeneity to exploit. Moreover, our data show that the heterogeneity oc- curs on a spatial scale such that exploitable variation exists within the dispersal distance of this species. Krohne et al. (1984) showed that the average dispersal distance in this population is 330 m. Thus, Grid I animals can easily reach low density areas. We still do not know how dispersers integrate into a new social system nor what demographic or social conditions permit immigration. Thus, we do not know if the variation within Grid I is sufficient to accomo- date opportunistic dispersers. Nevertheless, in every year of the study, there was variation that dispersers could at least attempt to exploit.

We found that the low density regions gradually increase relative to the high density areas. Although our data do not allow us to identify the relative importance of dispersal and in situ reproduction in homogenizing densities, given the opportunistic nature of dispersal in this species, we suspect that movement plays an important role. Dispers- al reduces local demographic heterogeneity. New density variation is constantly emerging, as evidenced by the temporal instability of local high density areas and the regular appearance of new ones. We thus envision a process in which locally acting processes regularly create density variation while dispersal gradually reduces it.

Of course, the sampling schemes for the different spatial scales may have biased the results. At Level I (Grid 1), the entire area was sampled. For Level III (Sugar Creek Valley), the area in question is some 350 km2. A complete census of the area was not possible. Thus, we sampled the area with several grids. If unsampled areas had greater differences in demography, our methods would underestimate the variation in the larger scales. While this may have been the case, we do not feel that the effect was large or altered our basic conclusions. We included study sites that encom- passed a considerable range of forest habitat variation. Some of the sites were separated by up to 35 km. Thus, we expect that we measured a large proportion of the ex- isting variation on this scale. Also, the more surprising result was the large variation within Grid 1, not a lack of variation within the Sugar Creek Valley.

There are also practical implications of these data for censusing populations. Most studies of small mammal populations, including previous studies at Allee Memorial Woods (Krohne and Baccus 1985), assume that live-trapping grids sample from a relatively homogeneous population within which demographic features and regulatory processes are uniform. If as shown above, the grid contains demographically heterogeneous units, a large grid may ob- fuscate, with a less meaningful average, the critical local



101

changes that are occurring. Consider Grid 1 in the present study. Its size and placement determines the qualitative and quantitative demographic changes. From 1980 to 1981 the grid showed a density increase, then decreased in 1982 (Fig. 1). However, if the grid had been designed so as to include just sections A and B (Fig. 2C), we would have shown a decrease in density from 1980 to 1981 and an increase in 1982. First, as we have suggested, the fine-scale dynamics may be a key to understanding large-scale population trends and thus may be of inherent interest. Second, sections A and B of Grid 1 (1.5 ha) are larger than some typical Peromyscus grids. If fine-scale demographic heterogeneity is common, the assumption that grids sample demographically uniform populations must be regularly tested.

Acknowledgement. We thank Mark Miner, Chris Krupp, Brad Dubbs and Jerry Noll for their assistance in the field. Ken Gregory facilitated our work at Turkey Run State Park. Bob Petty kindly allowed us to trap on his property. David Maharry and Alice Moore solved computer problems. Gregory H. Adler and Jerry 0. Wolff made many helpful comments on an earlier draft. This work was supported by grants from the National Science Founda- tion (No. SP1-80256687) and The Indiana Academy of Sciences. Financial support of the Treves Fund is also gratefully acknowl- edged. The McLain-McTurnan-Arnold Fund supported a leave of absence for DTK to pursue this work in 1982.

References

Adler GH (1987) Influence of habitat structure on demography of two rodent species in eastern Massachusetts. Can J Zool 65:903-912

Anderson PK (1980) Evolutionary implications of microtine behavioral systems on the ecological stage. The Biologist 62: 70-88

Batzli GO (1977) Population dynamics of the white-footed mouse in flood-plain and upland forest. Am Midl Nat 97:18-32

Bekoff M (1977) Mammalian dispersal and the ontogeny of indi- vidual behavioral phenotypes. Am Nat 111:715-732

Bowen BS (1982) Temporal dynamics of microgeographic structure of genetic variation in Microtus californicus. J Mammal 63:625-638

Cockburn A, Lidicker WZ Jr (1983) Microhabitat heterogeneity and population ecology of an herbivorous rodent, Microtus californicus. Oecologia 59:167-177

Garrett MG (1982) Demographic differences between an old and new colony of black-tailed prairie dogs. Am Nat 108:51-59

Gill DE (1978) The metapopulation ecology of the red-spotted newt, Notophthalmus viridescens (Rafinesque). Ecol Monogr 48:145-166

Hayne DW (1949) Calculation of size of home range. J Mammal 30:1-18

Krebs CJ (1966) Demographic changes in fluctuating populations of Microtus californicus. Ecol Monogr 36:239-273

Krebs CJ (1985) Ecology: the experimental analysis of distribution and abundance. 3rd ed. Harper and Row, NY

Krohne DT (1982) The demography of low litter size populations of Microtus californicus. Can J Zool 60:368-374

Krohne DT, Baccus R (1985) Genetic and ecological structure of a population of Peromyscus leucopus. J Mammal 6:529-537

Krohne DT, Miner MS (1985) Removal trapping studies of dispersal in Peromyscus leucopus. Can J Zool 63:71-75

Krohne DT, Baccus R, Dubbs, BA (1984) An analysis of dispersal in an unmanipulated population of Peromyscus leucopus. Am Midl Nat 112:146-156

Lidicker WZ Jr (1975) The role of dispersal in the demography of small mammals. In: Golley FB, Petrusewicz K, Ryskowski L (eds). Small Mammals: Their Productivity and Population Dynameis. Cambridge University Press, Cambridge, pp 103- 128

Lidicker WZ Jr (1978) Regulation of numbers in small mammal populations- historical reflections and a synthesis. In: Snyder DP (ed) Populations of Small Mammals Under Natural Condi- tions. Spec. Publ. Ser. No. 5, Pymatuning Lab Ecol., University of Pittsburgh. pp 122-166

Lidicker WZ Jr (1988) Solving the enigma of microtine "cycles". J Mammal 69:225-235

Morisita M (1959) Measuring the dispersion of individuals and analysis of the distributional patterns. Mem Faculty Sci Kuyushu Univ Ser E 2:213-235

Morris DW (1987) Ecological scale and habitat use. Ecology 68:362-370

Patton JL, Feder J (1981) Microspatial heterogeneity in pocket gophers: non-random breeding and drift. Evolution 35:912-920

Roff DA (1974) The analysis of a population model demonstrating the importance of dispersal in a heterogeneous environment. Oecologia 15:259-275

Smith MH, Manlove MN, Joule J (1978) Genetic organization in space and time. In: Snyder DP (ed) Populations of Small Mammals Under Natural conditions. Spec. Publ. Pymatuning Lab of Ecology No. 5, University of Pittsburgh. pp 99-111

Sokal RR, Rohlf FJ (1969) Biometry, 1st edition, W.H. Freeman, San Francisco

West SD (1979) Dynamics of colonization and abundance in central Alaskan populations of the northern red-backed vole, C. rutilus. J Mammal 63:128-143

Wiens JA (1976) Population responses to patchy environments. Ann Rev Ecol Syst 7:81-120

Wiens JA (1980) Concluding comments: are bird communities real? Acta XVII Congressus Int. Ornith., Berlin, pp 1088-1089

Wiens JA, Rottenberry JT (1980) Bird community structure in cold shrub deserts: competition or chaos? Acta XVII Congres- sus Int. Ornith., Berlin, pp 1063-1070

Wolff JO, Lundy KI, Baccus R (1988) Dispersal, inbreeding avoid- ance and reproductive success in white-footed mice. Anim Be- hav 36:456-465

Received November 1, 1988 / Accepted August 21, 1989



Documents

The Scale of Demographic Heterogeneity in a Population of Peromyscus leucopus