• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Ecological relationships of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor)

ProQuest Dissertations and Theses, 2009
Dissertation
Author: Brian D Carver
Abstract:
Two ecologically similar mammalian predators, Virginia opossums ( Didelphis virginiana ) and raccoons ( Procyon lotor ), were examined at a macrohabitat scale to assess the interspecific relationships of the taxa spatially and temporally and the relatedness of coefficients of association with estimates of abundance. Association coefficients were calculated from mark-recapture data from 6 sites in western Tennessee over 5 years. Results suggested neutral relationships spatially and temporally. Population size had no correlation with association values in 9 of 10 analyses. The lone correlation between abundance and association strength was positive, indicating a lack of competitive interactions. This study presents a novel approach for studying interactions among mesopredators, and supports the assumption that the presence of one species is independent of the other at the macrohabitat scale. Relationships in abundance of Virginia opossums and raccoons were investigated at 5 sites in western Tennessee. Using mark-recapture techniques over a 5-year period, relative abundance was derived for site and seasonal comparisons. Data were assessed utilizing Pearson's correlation coefficient and chi-square analysis. No associations between abundance of Virginia opossums and raccoons were observed for individual sites, seasons, or for pooled data. Additionally, abundance was not related to body mass of the species. The general trend was that neither species was consistently most abundant. Results provide support for the neutral theory of community organization. Temporal partitioning in foraging patterns of Virginia opossums and raccoons was investigated at 3 sites in western Tennessee through the use of timers attached to live traps on trapping grids and through time-stamped remote photography using infrared- triggered cameras at bait stations. Data were analyzed using Watson's U2 test, chi-square analyses, and probability tests. Results revealed no intraspecific differences in time of foraging for either species, as well as, a lack of interspecific differentiation. A high degree of overlap in foraging times of the species was observed. Virginia opossums and raccoons apparently did not partition their times of foraging. Foraging patterns of each species seem to be independent of the other (neutral association).

TABLE OF CONTENTS PAGE LIST OF TABLES x LIST OF FIGURES xv CHAPTER I. Interspecific associations of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor): Do they vary with space, time, and abundance? 2 Abstract 2 Introduction 3 Materials and Methods 5 Results 10 Discussion 16 Conclusions 20 Acknowledgments 20 Literature Cited 21 II. Abundance of co-occurring populations of Virginia opossums {Didelphis virginiana) and raccoons {Procyon lotor): Is there a relationship? 29 Abstract 29 Introduction 30 Materials and Methods 31 Results 36 Discussion 40 Conclusions 44 Acknowledgments 44 Literature Cited 45 III. An assessment of temporal partitioning in foraging patterns of Virginia opossums {Didelphis virginiana) and raccoons {Procyon lotor). 51 Abstract 51 Introduction 52 Materials and Methods 53 Results 58 Discussion 64 Conclusions 68 Acknowledgments 68 viii

Literature Cited 69 APPENDICES Appendix I 74 Appendix II 94 Appendix III 115 Appendix IV 194 Appendix V 247 Appendix VI 262 Appendix VII 270 Appendix VIII (IACUC Protocol) 273 IX

LIST OF TABLES TABLE PAGE CHAPTER I: Interspecific associations of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor): Do they vary with space, time, and abundance? TABLE 1.—Cole's coefficient of association (C7) for Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor) captured in live-traps in western Tennessee during 2000-2005. See text for descriptions of sites. 11 TABLE 2.—Relative abundance (catch/unit effort: number of animals captured including recaptures/number of trap-nights x 100) of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor) at 6 sites in western Tennessee during 2000-2005. See text for descriptions of sites. 13 TABLE 3.—Spearman's rank correlation coefficients (r) between relative abundance of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor) and Cole's coefficient of association (C7) at 6 sites in western Tennessee during 2000-2005. See text for descriptions of sites. Values represented by a dash (—) could not be calculated due to small sample size. C7 and abundance values are given in Tables 1 and 2, respectively. 15 CHAPTER II: Abundance of co-occurring populations of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor): Is there a relationship? TABLE 1.—Minimum number of animals known alive (MNKA) for Virginia opossums and raccoons at 5 sites in western Tennessee, 2000-2005. The total number of captures for each species (including recaptures) is also given. Descriptions of sites and seasons are presented in the text. 37 TABLE 2.—Pearson's correlation coefficients of the abundance of Virginia opossums to raccoons at 5 sites in western Tennessee, 2000-2005. Abundance values are given in Table 1. Site descriptions are presented in the text. 38 TABLE 3.—Chi-square analyses (Yates-corrected) of the distribution of abundance between Virginia opossums and raccoons at 5 sites in western Tennessee, 2000-2005. Abundance values were pooled across sites and examined to determine if the number of occasions where one species was more abundant was different from random. Abundance values are given in Table 1. Descriptions of sites and seasons are presented in the text. 39

CHAPTER III: An assessment of temporal partitioning in foraging patterns of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor). TABLE 1.—Intraspecific comparisons of foraging activity between age classes for raccoons from animals captured in live-traps at the Ames Plantation, Tennessee, 2000-2003. Comparisons were made using Watson's if tests. A false discovery rate (FDR) control was applied because of multiple comparisons. The calculated P-value was significant only if smaller than the FDR control. See text for description of age classes. 60 TABLE 2.—Mean time of capture (h) and mean time of activity (h) from photographs of Virginia opossums and raccoons at the Ames Plantation, Tennessee, 2000-2004. Times have been standardized to a day of equal day and night length (sunrise at 0600 and sunset at 1800 h). 61 APPENDIX I TABLE 1.—Universal Transverse Mercator coordinates for trap locations at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. These locations were utilized in a study assessing intra- and interspecific relationships of sympatric Virginia opossums and raccoons during 2000-2005. 83 APPENDIX II TABLE 1.—Trapping dates for winter 2000 at the Ames Plantation, Tennessee. 94 TABLE 2.—Trapping dates for spring 2001 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 95 TABLE 3.—Trapping dates for summer 2001 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 97 TABLE 4.—Trapping dates for winter 2001 at the Ames Plantation, Tennessee. 99 TABLE 5.—Trapping dates for spring 2002 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 100 TABLE 6.—Trapping dates for summer 2002 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 102 TABLE 7.—Trapping dates for winter 2002 at the Ames Plantation, Tennessee. 104 TABLE 8.—Trapping dates for spring 2003 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 105 XI

TABLE 9.—Trapping dates for summer 2003 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 107 TABLE 10.—Trapping dates for winter 2003 at the Ames Plantation, Tennessee. 109 TABLE 11.—Trapping dates for spring 2004 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 110 TABLE 12.—Trapping dates for winter 2004 at the Ames Plantation, Tennessee. 112 TABLE 13.—Trapping dates for spring 2005 at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee. 113 APPENDIX III TABLE 1.—Capture history and external-morphometric data of Virginia opossums captured at the Edward J. Meeman Biological Station, Tennessee, 2001-2005. Right and left ID are the numbers of the ear tags attached to the right and left ears, respectively. Age is according to Gardner (1972). Weight (WT) is given in kg. Measurements are total length (TL), length of tail (LT), length of hind foot (LHF), and length of ear (LE). Site is the trap number. Site abbreviation of DE refers to an individual found dead, RK to individual killed by automobile, and an asterisk (*) to trap-induced mortality. See Appendix I for locations of trap sites. 115 TABLE 1.—Capture history and external-morphometric data of Virginia opossums captured at the Ames Plantation, Tennessee, 2000-2005. Right and left ID are the numbers of the ear tags attached to the right and left ears, respectively. Age is according to Gardner (1972). Weight (WT) is given in kg. Measurements are total length (TL), length of tail (LT), length of hind foot (LHF), and length of ear (LE). Site is the trap number. Site abbreviation of DE refers to an individual found dead, RK to individual killed by automobile, and an asterisk (*) to trap-induced mortality. See Appendix I for locations of trap sites. 146 APPENDIX IV TABLE 1.—Capture history and external-morphometric data of raccoons captured at the Edward J. Meeman Biological Station, Tennessee, 2001-2005. Right and left ID are numbers of the ear tags attached to the right and left ears, respectively. Age is according to Grau et al. (1970). Weight (WT) is given in kg. Measurements are total length (TL), length of tail (LT), length of hind foot (LHF), and length of ear (LE). Site is the trap number. See Appendix I for locations of trap sites. 194 TABLE 2.—Capture history and external-morphometric data of raccoons captured at the Ames Plantation, Tennessee, 2000-2005. Right and left ID are numbers of the ear tags attached to the right and left ears, respectively. Age is according to Grau et al. xn

(1970). Weight (WT) is given in kg. Measurements are total length (TL), length of tail (LT), length of hind foot (LHF), and length of ear (LE). Site is the trap number. Site abbreviation of DE refers to animal found dead, RK to individual killed by automobile, and an asterisk (*) to trap-induced mortality. See Appendix I for locations of trap sites. 213 APPENDIX V TABLE 1.—Capture times of Virginia opossums in live-traps at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee, 2000-2005. R and L are right and left ear tag numbers, respectively. Age was defined as juvenile (J) or adult-(A) following Gardner (1972). Time has been standardized according to the methods in Chapter 3. Site definitions can be found in Appendix I. 247 TABLE 2.—Foraging times of Virginia opossums as determined from infrared- triggered photographs at the Ames Plantation, Tennessee, 2000-2004. 251 TABLE 3.—Capture times of raccoons in live-traps at the Edward J. Meeman Biological Station and the Ames Plantation, Tennessee, 2000-2005. R and L are right and left ear tag numbers, respectively. Age classes are according to Grau et al. (1970). Time has been standardized according to the methods in Chapter 3. Site definitions can be found in Appendix I. 252 TABLE 4.—Foraging times of raccoons as determined from infrared-triggered photographs taken at the Ames Plantation, Tennessee, 2000-2004. 256 APPENDIX VI TABLE 1.—Maximum likelihood estimates of density (±SE) for Virginia opossums and raccoons at the Edward J. Meeman Biological Station, Tennessee, 2001-2005. Density estimates are presented as ha per 1 animal. Site description is given in Appendix I. 265 TABLE 2.—Maximum likelihood estimates of density (±SE) for Virginia opossums and raccoons on site 2, Ames Plantation, Tennessee, 2000-2004. Estimates followed by an asterisk (*) were based on the boundary strip method because no animals were recaptured. Density estimates are presented as ha per 1 animal. Site description is given in Appendix I. 266 TABLE 3.—Maximum likelihood estimates of density (±SE) for Virginia opossums and raccoons on site 3, Ames Plantation, Tennessee, 2001-2005. Density estimates are presented as ha per 1 animal. Site description is given in Appendix I. 267 TABLE 4.—Maximum likelihood estimates of density (±SE) for Virginia opossums and raccoons on site 4, Ames Plantation, Tennessee, 2001-2005. Estimates followed xm

by an asterisk (*) were generated by inverse prediction. Estimates followed by a double asterisk (**) were generated by the boundary strip method because no animals were recaptured. Density estimates are presented as ha per 1 animal. Site description is given in Appendix I. 268 TABLE 5.—Maximum likelihood estimates of density (±SE) for Virginia opossums and raccoons on site 5, Ames Plantation, Tennessee, 2000-2004. Estimates followed by an asterisk (*) were based on the boundary strip method because no animals were recaptured. Density estimates are presented as ha per 1 animal. Site description is given in Appendix I. 269 TABLE 6.—Maximum likelihood estimates of density (±SE) for Virginia opossums and raccoons on site 6, Ames Plantation, Tennessee, 2000-2004. Density estimates are presented as ha per 1 animal. Site description is given in Appendix I. 269 APPENDIX VII TABLE 1.—Comparisons of time-of-capture between years for raccoons captured in live traps at the Ames Plantation, Tennessee, 2001-2003. Comparisons were made using Watson's U2 tests. 272 xiv

LIST OF FIGURES FIGURE PAGE CHAPTER I: Interspecific associations of Virginia opossums {Didelphis virginiand) and raccoons {Procyon lotor): Do they vary with space, time, and abundance? FiG. 1.—Location of study areas (Edward J. Meeman Biological Station and Ames Plantation) in western Tennessee utilized in an investigation of interspecific associations of Virginia opossums {Didelphis virginiand) and raccoons (Procyon lotor) during 2000-2005. 6 CHAPTER II: Abundance of co-occurring populations of Virginia opossums {Didelphis virginiand) and raccoons {Procyon lotor): Is there a relationship? FIG. 1.—Map showing the location of the Edward J. Meeman Biological Station and the Ames Plantation in western Tennessee. These sites were utilized in a study assessing the relationship in abundance of co-occurring populations of Virginia opossums and raccoons during 2000-2005. 32 CHAPTER III: An assessment of temporal partitioning in foraging patterns of Virginia opossums {Didelphis virginiand) and raccoons {Procyon lotor). FIG. 1.—Times of capture of Virginia opossums and raccoons in live traps at the Ames Plantation, Tennessee, 2000-2003. Times have been standardized to a day of equal day and night length (sunrise at 06:00 and sunset at 18:00 h). Category width is 1 h. Numbers on concentric circles represent sample sizes. Bold lines represent mean vectors (opossums = 22:43 h, raccoons = 22:20 h). Arcs outside the circles represent 95% CI of the mean vectors. 59 FIG. 2.—Times of foraging activity of Virginia opossums and raccoons at the Ames Plantation, Tennessee, 2002-2004 as calculated from remotely-triggered photographs. Times have been standardized to a day of equal day and night length (sunrise at 06:00 and sunset at 18:00 h). Category width is 1 h. Numbers on concentric circles represent sample sizes. Bold lines represent mean vectors (opossums = 01:31 h, raccoons = 00:57 h). Arcs outside the circles represent 95% CI of the mean vectors. 63 APPENDIX I FiG. 1.—Map showing the location of the Edward J. Meeman Biological Station and the Ames Plantation in western Tennessee. These sites were utilized in a study assessing intra- and interspecific relationships of co-occurring populations of Virginia opossums and raccoons during 2000-2005. 75 xv

FIG. 2.—Digital-orthophoto map showing the locations of traps on the Edward J. Meeman Biological Station, Tennessee (site 1). Data from animals captured at these trap sites were used in a study of the intra- and interspecific relationships of Virginia opossums and raccoons, 2000-2005. Site definitions are given in the text. Trap numbering follows that given in Table 1. 76 FIG. 3.—Map showing the relationship of 5 sites at the Ames Plantation in western Tennessee. A trapping grid was established at each site for use in a study assessing intra- and interspecific relationships of co-occurring populations of Virginia opossums and raccoons during 2000-2005. Site 1 was at the Edward J. Meeman Biological Station, Tennessee, and is shown in Fig. 2. 77 FIG. 4.—Digital-orthophoto map showing the locations of traps on the Ames Plantation, Tennessee (site 2). Data from animals captured at these trap sites were used in a study of the intra- and interspecific relationships of Virginia opossums and raccoons, 2000-2005. Site definitions are given in the text. Trap numbering follows that given in Table 1. 78 FIG. 5.—Digital-orthophoto map showing the locations of traps on the Ames Plantation, Tennessee (site 3). Data from animals captured at these trap sites were used in a study of the intra- and interspecific relationships of Virginia opossums and raccoons, 2000-2005. Site definitions are given in the text. Trap numbering follows that given in Table 1. 79 FIG. 6.—Digital-orthophoto map showing the locations of traps on the Ames Plantation, Tennessee (site 4). Data from animals captured at these trap sites were used in a study of the intra- and interspecific relationships of Virginia opossums and raccoons, 2000-2005. Site definitions are given in the text. Trap numbering follows that given in Table 1. 80 FIG. 7.—Digital-orthophoto map showing the locations of traps on the Ames Plantation, Tennessee (site 5). Data from animals captured at these trap sites were used in a study of the intra- and interspecific relationships of Virginia opossums and raccoons, 2000-2005. Site definitions are given in the text. Trap numbering follows that given in Table 1. 81 FIG. 8.—Digital-orthophoto map showing the locations of traps on the Ames Plantation, Tennessee (site 6). Data from animals captured at these trap sites were used in a study of the intra- and interspecific relationships of Virginia opossums and raccoons, 2000-2005. Site definitions are given in the text. Trap numbering follows that given in Table 1. 82 xvi

CHAPTER I

Brian D. Carver Department of Biology Freed-Hardeman University 158 E. Main Street Henderson, TN 38340 Email :bcarver@fhu. edu Running head: Association of Virginia opossums and raccoons INTERSPECIFIC ASSOCIATIONS OF VIRGINIA OPOSSUMS (DIDELPHIS VIRGINIANA) AND RACCOONS (PROCYON LOTOR): DO THEY VARY WITH SPACE, TIME, AND ABUNDANCE? Brian D. Carver Department of Biology, The University of Memphis, Memphis, TN 38152 Present Address: Department of Biology, Freed-Hardeman University, Henderson, TN 38340 Two ecologically similar mammalian predators, Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor), were examined at a macrohabitat scale to assess the interspecific relationships of the taxa spatially and temporally and the relatedness of coefficients of association with estimates of abundance. Animals were captured using live traps, and association coefficients were calculated from mark- recapture data from 6 sites in western Tennessee over 5 years. Results suggested neutral relationships spatially and temporally. Population size had no correlation with association values in 9 of 10 analyses. The lone correlation between abundance and association strength was positive, indicating a lack of competitive interactions. This 2

study presents a novel approach for studying interactions among mesopredators, and supports the assumption that the presence of one species is independent of the other at the macrohabitat scale. Keywords: coefficient of association, co-occurrence, Didelphis virginiana, mammals, population size, Procyon lotor, raccoon, spatial variability, temporal variability, Virginia opossum Leibold et al. (2004) pointed out the importance of understanding species relationships because of the role they may have in shaping community structure. Other investigations (Henke and Bryant 1999; Holling 1992; Naiman et al. 1986; Soule et al. 2003) have shown experimentally the central role such associations play in community processes. However, interspecific relationships are complex (Mills et al. 1993) and have been the subject of many biological investigations (see Begon et al. 2006; Connell 1983; Schoener 1983). Ecological factors such as space, time, and population size influence interspecific associations (Banks and Dickman 2000; Danielson and Gaines 1987; Heske et al. 1984; Levin 1992). A complete understanding of interspecific interactions must necessarily be formed at the large-scale on which the species actually exist (Linnell and Strand 2000); at present, such studies are lacking for many associations. Virginia opossums {Didelphis virginiana, hereafter = opossums) and raccoons {Procyon lotor) provide an interesting model for studying interspecific relationships between co-occurring taxa in relation to ecological factors. These mesopredators have distributions that overlap extensively, feed on many of the same foods, and utilize similar habitats (Gardner and Sunquist 2003; Gehrt 2003; Llewellyn and Uhler 1952; Zeveloff 3

2002). Therefore, numerous opportunities for interspecific relationships exist. However, at present, the linkages are not clearly understood, and coefficients of association, which have been used to examine such relationships (e.g., Cole 1949; Hurlbert 1969; Rehg 2006), have not been employed to assess the relatedness of interspecific associations spatially and temporally, including the relationships between population estimates and interspecific association of species. For opossums and raccoons, previous investigations (Ginger et al. 2003; Kissell and Kennedy 1992; Ladine 1995) conducted at a macrohabitat (trapping grid) level at local sites but different geographic areas all reported neutral associations in occurrence. These findings suggest a lack of geographic or temporal variability in the association of these species. However, at present, such conclusions have not been tested. The purpose of this investigation was to assess the interspecific relationship of opossums and raccoons spatially and temporally and to examine the relatedness of coefficients of association with estimates of abundance. Three predictions were tested. First, interspecific association of opossums and raccoons varies across geographic space. Second, interspecific association of opossums and raccoons differs temporally at individual sites. Third, interspecific association of opossums and raccoons is correlated to abundance of the species. Examining interspecific associations between mammalian mesopredators in relation to spatial and temporal variability and to estimates of population size represents a novel approach to investigating interactions among species. Therefore, the results of this study should provide new and interesting insights into understanding interspecific relationships among mammals. 4

MATERIALS AND METHODS Study area.—The study was conducted at 6 trapping sites in western Tennessee. Site 1 was at the 252-ha Edward J. Meeman Biological Station (Meeman) located ca. 20 km north of Memphis, Tennessee (UTM Zone 16S, Northing value: 3,917,202, Easting value: 225,383). Sites 2 through 6 were located on the 7,462-ha Ames Plantation (Ames) located approximately 5 km northwest of Grand Junction in Fayette and Hardeman counties, Tennessee (UTM Zone 16S, Northing value: 3,888,064, Easting value: 298,044). Meeman was owned by the State of Tennessee and operated by The University of Memphis, while Ames was owned and operated by Trustees of the Hobart Ames Foundation cooperatively with the University of Tennessee. Fig. 1 shows the location of Meeman and Ames. Habitat was typical of the region. In general, agricultural crops included soybeans {Glycine max), corn (Zea mays), and cotton {Gossypium sp.). Typically, upland tree species were loblolly pine {Pinus taeda), oaks {Quercus spp.), hickories {Carya spp.), and yellow poplar {Liriodendron tulipifera); bottomland species primarily included oaks, maples {Acer spp.), and sweetgum {Liquidambar stryaciflua). Old-fields on sites 5 and 6 included native warm-season grasses such as broomsedge {Andropogon virginicus), big blue stem {Andropogon gerardii), Indian grass {Sorgastrum nutans), and switch grass {Panicum virgatum) that were maintained by periodic burning. Topography of upland forest was characterized by gently rolling slopes and eroded gullies; whereas, bottomland forest, old-field, and agricultural areas occupied a flatter topography. Water resources included small ephemeral streams, and numerous small ponds at all sites, and the North Fork of the Wolf River at sites 2-4. A more detailed description of site 1 is given in 5

FIG. 1.—Location of study areas (Edward J. Meeman Biological Station and Ames Plantation) in western Tennessee utilized in an investigation of interspecific associations of Virginia opossums (Didelphis virginiana) and raccoons (Procyon lotor) during 2000- 2005. 6

Ladine (1997), while more in-depth descriptions of the sites at Ames are given in Baldwin (2003). Trapping procedure.—A trapping grid was established at each site. Traps followed a 5x10 configuration at site 1 with traps spaced approximately 150 m apart for a total of 50 traps and a grid area of 81 ha. For sites 2 through 6, traps were placed in an 8 x 8 configuration with spacing between traps of approximately 230 m for a total of 64 traps per site and a grid area of 259 ha. Trapping grids at Ames (sites 2-6) were spaced a minimum of 1.6 km apart in an effort to maintain grid independence during each trapping season. Radiotelemetry data (Jennings 2007) and capture data (Carver 2009) showed that sites were spatially independent. Traps on site 1 were operated for 40 nights, and those on sites 2-6 were operated for 32 nights each per season trapped, yielding approximately 2,000 trap nights (1 trap night = 1 trap set for 1 night) per site per trapping session during 2000-2005. Each site was trapped selected nights during either the fall/winter (hereafter = winter) season from 15 October-15 January or the winter/spring (hereafter = spring) season from 16 January-5 May. Site 2 was trapped during both the winter and spring season each year with the exception of spring 2005. Sites 1,3, and 4 were trapped during the spring season each year, while sites 5 and 6 were trapped during the winter season each year. Additionally, sites 1,2, and 4 were trapped during the summer season from 15 June-15 September during 2001-2003. Animals were captured in raccoon-size Tomahawk (Tomahawk Live Trap Co., Tomahawk, Wisconsin) or Havahart (Woodstream Corporation, Lititz, Pennsylvania) live traps. Traps were baited with canned cat food during winter and spring seasons and with dry dog food or doughnuts during the summer season. Upon initial capture, raccoons 7

were anesthetized with a mixture of ketamine hydrochloride (Ketaset; Bristol Laboratories, Syracuse, New York) and acepromazine maleate (PromAce; Ayerst Laboratories, New York, New York) at a 5:1 ratio with 0.1 cc of ketamine hydrochloride used per estimated kg of body weight. Captured opossums were generally not anesthetized. Opossums requiring anesthetization were treated in the same manner as raccoons. Captured opossums and raccoons were tagged in both ears with numbered rabbit tags or No. 3 Monel tags (National Band and Tag Co., Newport, Kentucky), respectively. Data collected from animals at the time of capture included species and capture site. All animals were released at the site of capture. Animals that had been anesthetized were placed in nearby cover and concealed with leaves for protection until recovery. Methods followed the guidelines for the capture, handling, and care of mammals as approved by the American Society of Mammalogists (Gannon et al. 2007). Population estimation.—The relationship between coefficients of association and density was evaluated through estimates of relative abundance. The number of captures of each species was divided by the number of trap-nights per sampling period to generate a measure of catch/unit effort. Relative abundance, including catch/unit effort, has been shown to be proportional to measures of absolute abundance generated from statistical models (Hopkins and Kennedy 2004). See Carver (2009) for the total number of animals captured per site, including recaptures. Co-occurrence estimation.—Co-occurrence was examined for each site over the course of an individual trapping session. Values were obtained by examining each individual trap site and determining whether that site caught only opossums, only raccoons, both species, or neither species over the duration of the trapping season. 8

Co-occurrence was, therefore, defined as both species being captured at the same trap location (trap site) within a season. Cole's C7 coefficient with the correction proposed by Ratliff (1982) was used for analyses of co-occurrence (Cole 1949). Cole's coefficient of association values range from -1.0 to 1.0. Strong coexistence is suggested by highly positive values, and competition by highly negative values, while values close to zero indicate co-occurrence to the degree that would be expected by chance (Cole 1949). Calculations of C7 were performed within Microsoft Excel 2003. Estimations involved the creation of 2 x 2 contingency tables. The recommendation that no cell within contingency tables have an expected frequency less than 1.0 was followed (see Cox 2002). Significance of association coefficients was examined through Fisher's exact tests (Zar 1999). Spatial and temporal analyses.—Coefficients of association were considered to vary if samples contained both significant (P>0.05) and non-significant (P<0.05) coefficients. Samples containing all non-significant or all significant coefficients were considered to lack variation. Coefficients of association were considered to vary if subsets of data contained significant associations in some samples and non-significant coefficients at others. Non-significant coefficients of association were deemed to be neutral for the purposes of the, study (they were not significantly different from a random association). Significant coefficients of association were deemed to represent positive or negative association based on the sign of the coefficient. Correlation analyses.—To determine if the abundance of one or both species influenced the coefficients of association, abundance values were examined for correlations to C7 values. Population sizes of raccoons and opossums were examined for 9

Full document contains 300 pages
Abstract: Two ecologically similar mammalian predators, Virginia opossums ( Didelphis virginiana ) and raccoons ( Procyon lotor ), were examined at a macrohabitat scale to assess the interspecific relationships of the taxa spatially and temporally and the relatedness of coefficients of association with estimates of abundance. Association coefficients were calculated from mark-recapture data from 6 sites in western Tennessee over 5 years. Results suggested neutral relationships spatially and temporally. Population size had no correlation with association values in 9 of 10 analyses. The lone correlation between abundance and association strength was positive, indicating a lack of competitive interactions. This study presents a novel approach for studying interactions among mesopredators, and supports the assumption that the presence of one species is independent of the other at the macrohabitat scale. Relationships in abundance of Virginia opossums and raccoons were investigated at 5 sites in western Tennessee. Using mark-recapture techniques over a 5-year period, relative abundance was derived for site and seasonal comparisons. Data were assessed utilizing Pearson's correlation coefficient and chi-square analysis. No associations between abundance of Virginia opossums and raccoons were observed for individual sites, seasons, or for pooled data. Additionally, abundance was not related to body mass of the species. The general trend was that neither species was consistently most abundant. Results provide support for the neutral theory of community organization. Temporal partitioning in foraging patterns of Virginia opossums and raccoons was investigated at 3 sites in western Tennessee through the use of timers attached to live traps on trapping grids and through time-stamped remote photography using infrared- triggered cameras at bait stations. Data were analyzed using Watson's U2 test, chi-square analyses, and probability tests. Results revealed no intraspecific differences in time of foraging for either species, as well as, a lack of interspecific differentiation. A high degree of overlap in foraging times of the species was observed. Virginia opossums and raccoons apparently did not partition their times of foraging. Foraging patterns of each species seem to be independent of the other (neutral association).