<%BANNER%>

Phylogenetics, Niche Modeling, and Biogeography of Mycotrupes (Coleoptera

Permanent Link: http://ufdc.ufl.edu/UFE0024876/00001

Material Information

Title: Phylogenetics, Niche Modeling, and Biogeography of Mycotrupes (Coleoptera Geotrupidae).
Physical Description: 1 online resource (150 p.)
Language: english
Creator: Beucke, Kyle
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: beetle, burrow, distribution, evolution, gis, insect, marine, pars, scarab, soil, stridulation, stridulatory
Entomology and Nematology -- Dissertations, Academic -- UF
Genre: Entomology and Nematology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: I conducted a study of the molecular phylogenetics, niche modeling, and biogeography of Mycotrupes (Coleoptera: Scarabaeoidea: Geotrupidae). I collected specimens of all five known species of Mycotrupes from across the known distributions. From these specimens, I sequenced a 481 base pair fragment of the mitochondrial gene Cytochrome Oxidase I, providing data for phylogenetic analyses done with both Parsimony as well as Bayesian methods. The resulting phylogenetic hypotheses provided a test of the species boundaries of Mycotrupes, as well as providing the basis for a test of an a priori biogeographic hypothesis of the species of Mycotrupes. I performed a niche modeling study with all Mycotrupes species in order to test the importance of various environmental layers (climate, soil, etc.) in describing the distributions of the five species. The phylogenetic analyses provided support for four of the five species of Mycotrupes. The fifth species, M. cartwrighti Olson and Hubbell, was polyphyletic. Pairwise distances suggested that the haplotype that caused this species to be non-monophyletic may be a cryptic species. Except for the two specimens represented by this haplotype, M. cartwrighti is monophyletic as are the other species of Mycotrupes in the analysis. The phylogenetic hypothesis does not support the a priori hypothesis of evolution of the species of Mycotrupes. As a result, the pre-existing biogeographical hypothesis is also not supported, and I proposed a new hypothesis. In the niche modeling study, I confirmed that well-drained soil is an important factor in the distribution of Mycotrupes species.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kyle Beucke.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Branham, Marc A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024876:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024876/00001

Material Information

Title: Phylogenetics, Niche Modeling, and Biogeography of Mycotrupes (Coleoptera Geotrupidae).
Physical Description: 1 online resource (150 p.)
Language: english
Creator: Beucke, Kyle
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: beetle, burrow, distribution, evolution, gis, insect, marine, pars, scarab, soil, stridulation, stridulatory
Entomology and Nematology -- Dissertations, Academic -- UF
Genre: Entomology and Nematology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: I conducted a study of the molecular phylogenetics, niche modeling, and biogeography of Mycotrupes (Coleoptera: Scarabaeoidea: Geotrupidae). I collected specimens of all five known species of Mycotrupes from across the known distributions. From these specimens, I sequenced a 481 base pair fragment of the mitochondrial gene Cytochrome Oxidase I, providing data for phylogenetic analyses done with both Parsimony as well as Bayesian methods. The resulting phylogenetic hypotheses provided a test of the species boundaries of Mycotrupes, as well as providing the basis for a test of an a priori biogeographic hypothesis of the species of Mycotrupes. I performed a niche modeling study with all Mycotrupes species in order to test the importance of various environmental layers (climate, soil, etc.) in describing the distributions of the five species. The phylogenetic analyses provided support for four of the five species of Mycotrupes. The fifth species, M. cartwrighti Olson and Hubbell, was polyphyletic. Pairwise distances suggested that the haplotype that caused this species to be non-monophyletic may be a cryptic species. Except for the two specimens represented by this haplotype, M. cartwrighti is monophyletic as are the other species of Mycotrupes in the analysis. The phylogenetic hypothesis does not support the a priori hypothesis of evolution of the species of Mycotrupes. As a result, the pre-existing biogeographical hypothesis is also not supported, and I proposed a new hypothesis. In the niche modeling study, I confirmed that well-drained soil is an important factor in the distribution of Mycotrupes species.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kyle Beucke.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Branham, Marc A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024876:00001


This item has the following downloads:


Full Text

PAGE 1

PHYLOGENETICS, NICHE MODELING, AND BIOGEOGRAPHY OF MYCOTRUPES (COLEOPTERA: GEOTRUPIDAE) By KYLE A. BEUCKE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Kyle A. Beucke 2

PAGE 3

To Mom and Dad 3

PAGE 4

ACKNOWLEDGMENTS This work has benefited greatly from the generous assistance of many individuals and institutions. First, I thank my parents, Dan and Diane, for th eir love and support throughout my life. My advisor, Dr. Marc Branham, has been a mentor and a friend du ring the course of my education and research, and gave me the opportun ity to do this work. My committee, composed of Drs. Marc Branham, David R eed, Paul Skelley, and Robert Woodr uff, gave willingly of their knowledge and perspectives. Dr. Woodruff, Dr Skelley, and Dr. Paul Choate were great company in the field. Dr. Choate collected num erous specimens and location information. Dr. Woodruff's 1973 book on Florida scarabs contributed to my interest in scarab beetles from a young age, and he is partly responsible for my decision to study this group. I am grateful to have had the opportunity to interact on a regular basis with the other members of the Branham Lab, both past and presen t. I thank Dr. Seth Bybee and Mr. Bradley Smith for helping make behavioral observati ons on captive specimens and Dr. Jennifer Zaspel for instructing me in the technique of the Polymerase Chain Reaction (PCR). Dr. Dulce Bustamante (University of Florida) provided assistance with statistical analysis. Dr. Thomas Walker (University of Florida) assi sted me with the operatio n of acoustic recording equipment. Mr. David Almquist and Dr. Jim Surdick (both with the Florida Natural Areas Inventory, Tallahassee, FL) helped me gain access to several interesting locations which yielded Mycotrupes. Dr. Thomas Scott, formerly of the Florida State Geological Survey, discussed with me the geology of the southeastern United States. Mr. Lyle Buss photographed Mycotrupes specimens. Drs. Alejandra and James Maruniak (University of Florida) provided primers for the polymerase chain reaction. Institutions and indi viduals generously lent me specimens of Mycotrupes. Drs. Barry O'Connor and Mark O'Brien (Museum of Zoology, Un iversity of Michigan), Drs. Max Barclay 4

PAGE 5

and Malcolm Kerley (Natural History Museum, London), Dr. John Morse and Ian Stocks (Clemson University), Drs. Yves Bousquet and Serge Laplante (Canadian National Collection), Dr. Bob Blinn (North Carolina State Universi ty), Dr. David Furth (United States National Museum of Natural History), Dr. Zachary Falin (U niversity of Kansas Natural History Museum), Dr. Darren Mann (Oxford University Museum of Natural History), Dr. Lee Herman (American Museum of Natural History), Dr. Keith Phillips (Western Kentucky University), and Dr. Paul Skelley (Florida State Collecti on of Arthropods). Mr. S. Full erton and Mr. P. Harpootlian contributed Mycotrupes location data. Individuals and organizations allowed me access to collect Mycotrupes on their land. Cheri Taylor (Summerfield, FL), Ms. Fergus on (Thomasville, GA), O'Leno State Park (High Springs, FL), Tall Timbers Research Station (Tallahassee, FL), and Elinor Klapp-Phipps Park (Tallahassee, FL) are all thanked. A Theodore Roosevelt Memorial Grant from the American Museum of Natural History supported my field collecting efforts. A gran t from the Center for Systematic Entomology provided funds for the sequencing of Mycotrupes specimens. I apologize if I have neglected to mention any individual who has helped me during this study. 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................9 LIST OF FIGURES .......................................................................................................................10 ABSTRACT ...................................................................................................................................12 CHAPTER 1 INTRODUCTION................................................................................................................. .14 2 LITERATURE REVIEW OF MYCOTRUPES ......................................................................17 Taxonomic History of Mycotrupes LeConte..........................................................................17 Distribution and Habitat....................................................................................................... ..18 Morphology............................................................................................................................21 Biology...................................................................................................................................22 Associated Life Forms.......................................................................................................... ..23 Phylogenetics..........................................................................................................................24 Biogeography..........................................................................................................................24 Conservation Status................................................................................................................25 3 BEHAVIOR..................................................................................................................... .......30 Introduction................................................................................................................... ..........30 Field and Laboratory Observations on Mycotrupes Behavior................................................32 Adult Feeding Behavior..................................................................................................32 Mycotrupes lethroides ..............................................................................................32 Mycotrupes cartwrighti ............................................................................................34 Mycotrupes gaigei ....................................................................................................34 Burrowing........................................................................................................................35 Mycotrupes lethroides ..............................................................................................35 Mycotrupes cartwrighti ............................................................................................35 Mycotrupes gaigei ....................................................................................................36 Larval Food.....................................................................................................................37 Mycotrupes gaigei ...........................................................................................................37 Discussion...............................................................................................................................37 4 VARIATION IN THE PARS STRIDENS OF THE STRIDULATORY APPARATUS OF MYCOTRUPES .................................................................................................................46 Introduction................................................................................................................... ..........46 Materials and Methods...........................................................................................................48 6

PAGE 7

Results.....................................................................................................................................49 Discussion...............................................................................................................................50 5 DELIMITING SPECIES BOUNDARIES AND DIAGNOSING POSSIBLE CRYPTIC SPECIES IN MYCOTRUPES .................................................................................................54 Introduction................................................................................................................... ..........54 Materials and Methods...........................................................................................................56 Sampling..........................................................................................................................56 DNA extraction, PCR Amplification, Sequencing and Nucleotide Alignments.............57 Phylogenetic Analyses.....................................................................................................57 Parsimony analysis...................................................................................................58 Bayesian analysis.....................................................................................................58 Nucleotide Divergence....................................................................................................60 Testing Alternative Phylogenetic Hypotheses.................................................................60 Results.....................................................................................................................................60 Discussion...............................................................................................................................61 Phylogenetics...................................................................................................................61 Pairwise Distances...........................................................................................................65 Testing Alternative Phylogenetic Hypotheses.................................................................66 Species Delimitation........................................................................................................66 Recommendation.............................................................................................................67 6 BAYESIAN PHYLOGENETIC INFE RENCE AND BIOGEOGRAPHY OF MYCOTRUPES .......................................................................................................................74 Introduction................................................................................................................... ..........74 Materials and Methods...........................................................................................................78 Results.....................................................................................................................................78 Nucleotide Substitution Rates and Divergence Time Estimates.....................................78 Discussion...............................................................................................................................79 Tree Topologies...............................................................................................................79 Species-Level Biogeography in Mycotrupes ...................................................................79 Comparison of divergence times of two calibration methods..................................79 Dispersal may complicat e taxon-area biogeography................................................80 A vicariance-based biogeogr aphical interpretation..................................................81 Comparison with studies of othe r taxa in the same region......................................84 Future Research...............................................................................................................85 7 EMPLOYING ECOLOGICAL NICHE MODELING TO PREDICT SPECIES DISTRIBUTIONS IN MYCOTRUPES ..................................................................................88 Introduction................................................................................................................... ..........88 Materials and Methods...........................................................................................................90 Location Data..................................................................................................................90 Environmental Layers.....................................................................................................90 Maximum Entropy...........................................................................................................91 7

PAGE 8

Results.....................................................................................................................................91 Discussion...............................................................................................................................92 Areas Predicted by MaxEnt.............................................................................................92 Mycotrupes lethroides ..............................................................................................92 Mycotrupes retusus ..................................................................................................92 Mycotrupes cartwrighti ............................................................................................93 Mycotrupes gaigei ....................................................................................................94 Mycotrupes pedester ................................................................................................95 Effect of Bioclim Data on Predicted Distributions..........................................................95 Relative Importance of Environmental Layers Across Species of Mycotrupes ..............97 Soil Types Associated with Mycotrupes Species............................................................98 Detailed Discussion of the Distribution of Mycotrupes gaigei .......................................99 Conclusion............................................................................................................................104 8 CONCLUSION................................................................................................................... ..115 APPENDIX A CYTOCHROME OXIDASE I SEQUENCE DATA...........................................................119 B SPECIMENS EXAMINED..................................................................................................130 C MYCOTRUPES LOCALITY RECORDS FROM WHICH NO SPECIMENS WERE STUDIED.............................................................................................................................138 LITERATURE CITED................................................................................................................139 BIOGRAPHICAL SKETCH.......................................................................................................150 8

PAGE 9

LIST OF TABLES Table page 3-1 Seasonality of specime n data from Appendix B................................................................45 4-1 Results of the analysis of covariance.................................................................................53 5-1 Mycotrupes Map of collecting local itions of.sequenced Mycotrupes specimens..............68 5-2 Settings for BEAUTi .xml files.........................................................................................69 5-3 Mean pairwise distances in Mycotrupes............................................................................72 6-1 Selected divergence time estimates...................................................................................87 7-1 Mycotrupes presence data fo r niche modeling.................................................................107 7-2 Environmental layers used...............................................................................................108 7-3 Diagnostics and result s from niche modeling..................................................................109 7-4 Taxonomic classification of soil at each site...................................................................110 9

PAGE 10

LIST OF FIGURES Figure page 2-1 Distribution map for species of Mycotrupes......................................................................27 2-2 Mycotrupes species............................................................................................................28 2-3 Hubbell's hypothesis of evolution in Mycotrupes..............................................................28 2-4 Mycotrupes speciation events as hypot hesized by Hubbell (1954)...................................29 3-1 Female M. lethroides pulling a piece of dog food.............................................................42 3-2 M. lethroides "perching" at the top of a burrow................................................................43 3-3 M. cartwrighti habitat at Thomas ville, Georgia.................................................................43 3-4 M. cartwrighti burrow with "pushup."...............................................................................44 3-5 M. cartwrighti burrow without "pushup."..........................................................................44 4-1 Pars stridens.............................................................................................................. .........52 4-2 Relationship between rib counts of the pars stridens and body size in Mycotrupes..........52 5-1 Map of collecting lo cations of sequenced Mycotrupes specimens....................................69 5-2 The single most parsimonious tree r ecovered in the Parsimony analysis..........................70 5-3 BEAST maximum clade credibility tree, pre-set substitution rate....................................71 5-4 BEAST maximum clade credibility tree, biogeographical calibration..............................72 5-5 Mean pairwise distances in Mycotrupes............................................................................73 6-1 Collecting locations of Mycotrupes specimens with numbered vicariance events............87 7-1 Likelihood of occurrence map for M. lethroides.............................................................111 7-2 Likelihood of occurrence map for M. retusus ..................................................................111 7-3 Likelihood of occurrence map for M. cartwrighti ...........................................................112 7-4 Likelihood of occurrence map for M. gaigei ...................................................................112 7-5 Likelihood of occurrence map for M. pedester................................................................113 7-6 Well-drained soil and M. gaigei collecting localities......................................................113 10

PAGE 11

7-7 Marine terraces in Florida................................................................................................1 14 7-8 The Ocala Platform and the Sanford High.......................................................................114 11

PAGE 12

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PHYLOGENETICS, NICHE MODELING, AND BIOGEOGRAPHY OF MYCOTRUPES (COLEOPTERA: GEOTRUPIDAE) By Kyle A. Beucke December 2009 Chair: Marc A. Branham Major: Entomology and Nematology I conducted a study of the molecular phylogene tics, niche modeling, and biogeography of Mycotrupes (Coleoptera: Scarabaeoidea: Geotrupidae). I collected specimens of all five known species of Mycotrupes from across the known distributions. From these specimens, I sequenced a 481 base pair fragment of the mitochondrial ge ne Cytochrome Oxidase I, providing data for phylogenetic analyses done with both Parsimony as well as Bayesian methods. The resulting phylogenetic hypotheses provided a test of the species boundaries of Mycotrupes as well as providing the basis for a test of an a priori biogeographic hypothesis of the species of Mycotrupes. I performed a niche modeling study with all Mycotrupes species in order to test the importance of various environmental layers (climate soil, etc.) in describing the distributions of the five species. The phylogenetic analyses provided support for four of the five species of Mycotrupes. The fifth species, M. cartwrighti Olson and Hubbell, was polyphyletic. Pairwise distances suggested that the haplotype th at caused this species to be non-monophyletic may be a cryptic species. Except for the two specimens represented by this haplotype, M. cartwrighti is monophyletic as are the other species of Mycotrupes in the analysis. The phylogenetic hypothesis does not support the a priori hypot hesis of evolution of the species of Mycotrupes 12

PAGE 13

As a result, the pre-existing biogeographical hyp othesis is also not supported, and I proposed a new hypothesis. In the niche m odeling study, I confirmed that well-drained soil is an important factor in the distribution of Mycotrupes species. 13

PAGE 14

CHAPTER 1 INTRODUCTION The beetle genus Mycotrupes LeConte is restricted in dist ribution to the Atlantic Coastal Plain of the Southeastern United States. The fi ve known species are allopatric (non-overlapping) in distribution. Two species, M. lethroides (Westwood) and M. retusus (LeConte), occur along the Fall Line (a region which separates the Piedm ont from the Atlantic Coastal Plain) in Georgia and South Carolina, with their distributions being separated by the Savannah River. Mycotrupes cartwrighti Olson and Hubbell is found in several ar eas in southern Georgia and northern Florida. Mycotrupes gaigei Olson and Hubbell is known from an extensive portion of northern Peninsular Florida. Mycotrupes pedester Howden occurs only in southwestern Florida. Species of Mycotrupes are flightless and are known to di g deep burrows, up to six feet in depth for M. gaigei Data on feeding is fragmentary although adults can be caught in pitfall traps baited with dung and/or fermenting malt. The larval stage is passed underground in burrows excavated by adults. The larva of only one species, M. gaigei has been described. The larval food for this species was appa rently cattle dung (Howden 1954). Mycotrupes species are found most co mmonly in well-drained areas such as sandhill, in Florida and the Fall Line of Geor gia and South Carolina. This ha bitat is being developed at a rapid rate for housing and agricultu re (Enge et al. 1986; Kautz et al 2007). This, combined with their restricted geographic distributions, makes Mycotrupes quite vulnerable to habitat loss. Some of these sandy environments are associated with ancient seashores which are the product of a long history of fluctuating climate and sea levels during the Cenozoic Era. In the monograph of the genus Mycotrupes (Olson, Hubbell, and Howden 1954), Hubbell (1954) proposed a hypothesis for the evolution of the ge nus, which he based upon changes in sea level and assumptions of the evolution of certain morphological features. Until now, this hypothesis 14

PAGE 15

has not been tested with modern phyloge netic and biogeographic methods. Multiple characteristics associated with Mycotrupes, such as allopatric distri butions, flightlessness, and distribution of the species across a region that has been extensively affected by changes in sea level, suggest that this genus may be well suited for studying patte rns of speciation and biogeography. The first objective of my study was to test the hypothesis, pro posed by Hubbell (1954) for the evolution of Mycotrupes species, using modern phylogenetic methods. This component also provided a phylogenetic test of species bou ndaries and the possible detection of cryptic species. The second objective was to test th e biogeographic hypothesis of Hubbell (1954). The third objective was to determine what environm ental factors might be associated with, and perhaps limiting the distributions of different Mycotrupes species, using Niche Modeling techniques. This information was later used to further develop a ne w biogeographic hypothesis and suggest areas that may cont ain yet undiscovered populations of Mycotrupes The development of a character matrix is necessary in order to construct a phylogenetic hypothesis. It is extremely difficu lt to find morphological features in this genus that consistently vary between species, but which show little or no variation within a species. For this reason, molecular data was used. A 481 base pair fr agment of the mitochondrial gene Cytochrome Oxidase I was sequenced from all five Mycotrupes species and analyzed within a phylogenetic analysis. The phylogenetic an alysis was conducted by using two methods of phylogenetic inference: 1) Parsimony and 2) Bayesian. Both of these analyses resulted in similar patterns of evolution in Mycotrupes, and both did not support the hypothesis of Hubbe ll (1954). In addition, both models provided strong evidence for the existen ce of a cryptic species. The use of Bayesian methods in conjunction with the program BEAS T also provided a method for estimating how 15

PAGE 16

long ago each of these beetle speci es diverged from the next cl osely related species (e.g., time since divergence). By combining informa tion from species dist ributions, phylogenetic hypotheses and divergence time estimates, new hypotheses for the evolution and biogeography for the genus Mycotrupes have been generated. I conducted a niche modeling analysis us ing the program Maximum Entropy (Schapire 2008), which used collection locality, in conjunc tion with multiple environmental layers to produce a modeled niche and likelihood of occurrence map for each species of Mycotrupes. The environmental data layers included soil characte ristics, elevation, and c limate (i.e., temperature and precipitation). The results s howed drainage to be an important factor in the distribution for all Mycotrupes species. This result corr oborates data on the known distribution of these species and supports the hypothesis that the dispersal of Mycotrupes species is hindered by poor draining soils. This study represents the firs t molecular phylogenetic study of Mycotrupes. These findings contribute to the study of the historical biogeography of th e Southeastern United States by providing useful comparative data for studyi ng other taxa found in this region. In addition, data on the environmental variable s associated with the distribution of these beetles may help in future conservation efforts targeted at protecting Mycotrupes and their habitat. 16

PAGE 17

CHAPTER 2 LITERATURE REVIEW OF MYCOTRUPES Taxonomic History of Mycotrupes LeConte The genus Mycotrupes (Coleoptera: Geotrupi dae: Geotrupinae) is composed of five species: Mycotrupes lethroides Mycotrupes retusus, Mycotrupes gaigei Mycotrupes cartwrighti and Mycotrupes pedester (Jameson 2002; Olson and Hubbell 1954). The name Mycotrupes was erected by LeConte in 1866 as a subgenus of Geotrupes for Geotrupes retusus which he attributed to MacLeay, although no description was published by the later author (Horn 1868). Horn (1868) published an a dditional description of G. retusus and Blanchard (1888) referred to Horn as the author of the species. Le Conte was recognized as the author of G. retusus by Olson and Hubbell (1954). The type series of G. retusus consisting of five specimens, was later determined by Olson and Hubbell (1954) to includ e a specimen each from two other species, M. lethroides and M. cartwrighti Westwood (1837) described Geotrupes lethroides which was later transferred to Mycotrupes by Boucomont (1911). Although Westwood recognized the dis tinctive morphology of his species, stating that it displayed characters that seemed to be ...of higher value than those indicating a species, he did not propose a new genus. Westwood described the collecting locality of G. lethroides as America Meridionali, a confusin g description that may have been interpreted by later authors (including LeConte) as indicating that the sp ecies occurs in South America (Olson and Hubbell 1954). Thus the spec ies went unrecognized as part of the North American fauna until Boucomont (1911). In 1902, Boucomont placed Geotrupes retusus in the genus Thorectes subgenus Mycotrupes, and later (1911) moved the subgenus Mycotrupes to the genus Geotrupes. Felsche (1909) described Geotrupes aeneus from what was actually a specimen of Mycotrupes 17

PAGE 18

mistakenly labeled as collected from Senegal. Later, finding much similarity between his specimen and LeContes description of Geotrupes retusus, Felsche synonymized G. aeneus under M. retusus (Felsche 1910). In 1911, Boucomont synonymized M. retusus under M. lethroides after noting the similarity of the descripti ons of the two species and recognizing that the locality information given in Westwoods 1837 description of Geotrupes lethroides was incorrect. From this 1911 synonymy until the gene ric revision of Ols on and Hubbell (1954), all authors referred to species of Mycotrupes as M. lethroides Olson and Hubbell (1954) examined the available material of Mycotrupes and concluded that multiple species were represented. They recognized the species status of M. lethroides, resurrected M. retusus as a valid species, and described two additional species: M. gaigei and M. cartwrighti. Howden (in Olson and Hubbell 1954) described M. pedester Distribution and Habitat The genus Mycotrupes is restricted to the southeastern United States. All five species are known to be allopatric. Alt hough habitat type varies by species and geographic location, welldrained soil and an open understo ry appear to characterize most occurrences (Woodruff 1973). Many deposits of these well-drained soils, especially in the case of the sand ridges in Florida and the sandhill region of Georgia and South Carolina, are likely associated with ancient shorelines (Hubbell 1954). An account of the distribution and habitat of each sp ecies follows. See Figure 2-1 for a map of Mycotrupes species distributions; location points represent data from collections made during this study as well as data from borrowed specimens and literature records. Mycotrupes lethroides (see Figure 2-2 a and b) has only been collected from a small number of locations in Richmond and Burke counties, in Georgia (Beucke and Choate 2009; Harpootlian 1995; Olson and H ubbell 1954). Its habitat was de scribed by Olson and Hubbell 18

PAGE 19

(1954) as open forests of pine and oak on sandy so il. These sandy areas in clude part of the Fall Line sandhills, which extend into South Carolina. Mycotrupes retusus (see Figure 2-2 c and d) is found along much of the sandhills region of South Carolina from Aiken to Kers haw County (Harpootlian 1995, 2001, 2006; Olson and Hubbell 1954). Olson and Hubbell (1954) described the habitat of M. retusus as open pine and oak forest on sandy soil. The distributions of M. retusus and M. lethroides are apparently associated with similar habitat but are sepa rated by the Savannah River. Hubbell (1954) proposed the enlargement of the Savannah River as a vicariant event resulting in th e isolation of the ancestors of these two species. Mycotrupes cartwrighti (see Figure 2-2 e and f) is probably the most widely distributed species of the genus and occurs in Georgia a nd Florida (Hebard 1903 [who referred to it as M. retusus]; Olson and Hubbell 1954; Peck and Th omas 1998; Woodruff 1973; Woodruff and Deyrup 1994a, 1994b, 1994c). Most records of M. cartwrighti are from a large area extending north from Tallahassee, Florida to Fort Valley, Georgia (Olson and Hubbell 1954). Mycotrupes cartwrighti has also been collected from the widely separated areas of Jack sonville and Atlantic Beach, Florida and Hinesville, Georgia (P. Ha rpootlian, pers. comm.; Olson and Hubbell 1954; K. Beucke, unpublished data). The type locality of M. cartwrighti is 6.5 miles East of Tallahassee and was described by Olson and H ubbell (1954) as hardwood forest with little undergrowth, on a slope. The soil was describe d as Orangeburg sandy loam over clay. Mycotrupes gaigei (see Figure 2-2 g and h) is restricted to Florida and its range extends in a broad swath from Madison County in the No rth to Sumter County in the South (Olson and Hubbell 1954; Peck and Thomas 1998; W oodruff 1973; Woodruff and Deyrup 1994a, 1994b, 1994c; P. Choate, pers. comm.). This area corr esponds to the Peninsular Lime Sink Region, an 19

PAGE 20

extensive area of karst with litt le surface water and soils predom inantly of the Norfolk series (Harper 1921; Olson and Hubbell 1954). Mycotrupes gaigei also occurs in Seminole County, Florida (Woodruff 1973). The habitat of M. gaigei was described by Olson and Hubbell (1954) as pine and oak forest on well-drained sands. Mycotrupes pedester (see Figure 2-2 i and j), which is also rest ricted to Florida, has been collected at a limited number of locations in De Soto, Charlotte, and Lee counties (Olson and Hubbell 1954; Peck and Thomas 1998; W oodruff 1973; Woodruff and Deyrup 1994a, 1994b, 1994c). Woodruff (1973) collected M. pedester at two sites in Tice and Estero, Florida. One of these sites was located in a cattle pasture at the edge of a drainage canal; the other site had been recently burned and had "Caribbean pine with some scattered large live oaks and a dense mat of saw palmetto in places." The habitat of M. pedester may be more fragmented and isolated compared to that of other Mycotrupes species. All of the known collecting localities for M. pedester are contained within an area labeled as flatwoods vegetation by Harper (1927), indicating that it is genera lly a poorly drained region. Several published lo cality records of Mycotrupes require clarifica tion. LeConte (1866) listed M. retusus (as Geotrupes retusus ) from North Carolina to Loui siana. Brimley (1938) and Leng (1920) listed Geotrupes lethroides from North Carolina, pres umably based on LeConte's distribution information. There are no specific collecting data availabl e that lend credence to Mycotrupes occurring this far from its known range However, Olson and Hubbell (1954) thought that M. retusus might occur in North Carolina, becau se the Fall Line belt of sandhill habitat in which this species is found in S outh Carolina extends into that state. In his paper on Florida scar abs, Blatchley (1928) listed M. lethroides from St. Augustine and Enterprise. These records probably re fer to the currently recognized species M. cartwrighti 20

PAGE 21

and M. gaigei ; M. cartwrighti has been collected from Jacks onville and Atlantic Beach, Florida (both are near St. Augustine) by H. Klages a nd A.T. Slosson (Olson and Hubbell 1954), and I collected Mycotrupes gaigei from Geneva, Florida, which is approximately 12 miles from Enterprise. Two M. cartwrighti collecting localities appear to be doubtful. Three specimens of M. cartwrighti are recorded as having been collected in Miami, Dade County, Florida by H.M. Klages (Olson and Hubbell 1954). One of these specimens, on loan from the University of Michigan Museum of Zoology, was exam ined by the author. A specimen of M. cartwrighti also collected by H.M. Klages, is labeled as having be en collected in Comfort, Texas. These two localities (Miami, Florida, and Comfort, Texas) are far from the rest of the known distribution of Mycotrupes. In addition, these localities ha ve not produced ot her specimens of Mycotrupes. The H.M. Klages collection is reputed to have numerous mislabeled specimens (Woodruff 1973). For these reasons, I have decided to ignore the Mi ami, Florida, and Comfort, Texas records in this study. Morphology Mycotrupes are moderately sized beetles and measure between 10 and 20 millimeters in length. The genus may be separated from ot her geotrupid genera by several distinctive characters. The mesothoracic wings (elytra) ar e fused, and the metathoracic wings are entirely absent, making Mycotrupes flightless. In lateral profile both the pronotum and elytra are strongly convex and are broadly no tched at their junction. In male s of all species and in females of M. lethroides the pronotum is anteriorly excavated and this excavation is often bordered posteriorly by a pair of polished crests. Th e body surface is finely sculpted, giving it a dull appearance, and it is covered w ith rounded granules. The body color is generally black although a metallic blue or coppery sheen is sometimes apparent (Olson and Hubbell 1954). 21

PAGE 22

The species of Mycotrupes have a limited number of apparent morphological differences among them. Some characters used by Olson and Hubbell (1954) to distinguish between species include the shape of the phallic theca, presence or absence of elytral st riae, pattern of dorsal granulation (whether the granules are separate or confluent), shape of the pronotal depression, shape and position of the cephalomedian pronotal nodul e, and shape of the frontal clypeal suture. Mycotrupes gaigei which has evidence of el ytral striae as well as confluent (as opposed to distinct) dorsal granules, appears markedly di stinct from the remaining four species. Hubbell (1954) noted some mo rphological similarities between Mycotrupes and the geotrupid genera Thorectes Mulsant and Typhoeus Linnaeus subgenus Chelotrupes Jekel. He believed that these similarities were the result of parallel evolution becaus e of the presence of other seemingly important morphological differences between the genera. The larva of Mycotrupes gaigei was described by Howden (1954). The larvae of the other four species remain undescribed. Gross mo rphological differences that separate the larva of Mycotrupes from those of other Geotrupidae include the broadly truncate configuration of the endoskeleton below the anal opening, shape of the tormae of the epipharynx, and presence of a small sclerotized area on the glossa (Howden 1954; Woodruff 1973). Biology The feeding habits of adult and larval Mycotrupes are poorly known. Adult Mycotrupes appear to be attracted to a va riety of substances, including dung and fermenting malt, and they have been observed feeding on a variety of foods in the field and laboratory (Beucke and Choate 2009; Harpootlian 1995; Olson et al. 1954; Woodruff 1973). Larval food has been observed only once, and in that instance, it appeared to be ol d cow dung (Howden 1954). See Chapter 3 for additional information on feeding, burrowing, and other Mycotrupes behavior. 22

PAGE 23

Associated Life Forms Geotrupids are known to have a variety of commensal associates and parasites, including mites and nematodes (Thodorids 1952). Two i nvertebrate species, a m ite and a fly, have a close biological relationship with Mycotrupes. The mite Macrocheles mycotrupetes Krantz and Mellott (Acari: Macroche lidae) is phoretic on Mycotrupes gaigei and is known from sandhill habitat in Alachua, Levy, and Columbia counties, Florida (Krantz and Mellott 1968). This mite is predaceous, and it presumably feeds on nematode s or other invertebrates present in the food of Mycotrupes. A dihydroxy wax, present on the cuticle of Mycotrupes is attractive to Macrocheles (Krantz et al. 1991). Interestingly, 2,3dihydroxy alcohol esters found in the uropygial glands of chickens were found to be attractive to M. mycotrupetes, suggesting that the attractant in Mycotrupes may be of a similar composition (Kra ntz et al. 1991). Krantz and Royce (1994) documented, in laboratory experiments, the movement of M. mycotrupetes through more than three inches of sand to reach Mycotrupes gaigei As Mycotrupes burrows often appear to be densely aggregated, underground dispersal of mites be tween beetle burrows may be possible. The fly, Ceroptera longicauda Marshall (Diptera: Sphaeroceridae), is also a phoretic associate of Mycotrupes gaigei Other Ceroptera species apparently de velop in the underground larval food supplies of dung feeding scarab beetles, and C. longicauda presumably has a similar lifestyle (Marshall a nd Montagnes 1988; Sivins ki et al. 1999). Neither M. mycotrupetes nor C. longicauda are restricted to M. gaigei Macrocheles mycotrupetes is known from the geotrupid Geotrupes egeriei Germar (Krantz and Royce 1994), and C. longicauda is known from the geotrupid Peltotrupes profundus Howden (Sivinski et al. 1999). 23

PAGE 24

Phylogenetics Hubbell (1954) used several morphological ch aracters that he deemed significant to produce a hypothesis of evolution for members of the genus ( Figure 2-3 ). He assumed M. gaigei to be the most basal species, as it shares the presence of elyt ral striae, as well as a punctate pronotum and simple phallic spine, with othe r Geotrupidae. Hubbe ll (1954) considered M. lethroides and M. retusus to form a group based on the bird head-shaped phallic spine that they share, and M. cartwrighti and M. pedester to form another group based on their simple phallic spine and other morphological similarities. He thought that the similarities between M. cartwrighti and M. pedester were so great that their differences might not warrant their recognition as distinct species. Zunino (1984) published diagrams depicti ng his hypotheses of the relationships of geotrupid genera including Mycotrupes Zuninos methods were not clear, and it appears that he simply drew conclusions based on a few morphologi cal characters that he considered important, in a fashion similar to that of Hubbell (1954). Verdu et al. (2004), and Grebennikov and Scholtz (2004) conducted morphological phylogenetic analyses of the Geotrupidae and the Scarabaeoidea, respectively. In both analyses, Mycotrupes gaigei was included as a terminal taxon. Unfortunately, the relationship of Mycotrupes to other genera within Geotrupidae was poorly resolved. Biogeography Hubbell (1954) speculated on the biogeographic relationships of Mycotrupes and formulated an explanation that took into acc ount the hypothetical evolutionary relationships depicted in Figure 2-3 and the present-day distribution of the species. A key assumption made by Hubbell was that dispersal in Mycotrupes is limited to walking, restricting them to contiguous habitat and suggesting that th eir present-day distributions are suggestive of their past 24

PAGE 25

distributions. Hubbell used the Pleistocene te rraces studied by Cooke (1945) to formulate a sequence of divergences in Mycotrupes in response to repeated submergence and emergence of the land because of changes in sea level. H ubbell (1954) illustrated the sequence of these hypothesized events (see Figure 2-4 ). An important condition of his hypothesis is the presence of exposed land in Central Florida (G 2 in Figure 2-4 ) during some of the sea level fluctuations, as M. gaigei is assumed to have split from the basal stock early, before the differentiation of the ancestor of the remaining four species of Mycotrupes. Complete submergence of this Central Florida area would likel y have wiped out any Mycotrupes present (Hubbell 1956). Howden (1963) proposed the idea that the seasonality of Mycotrupes activity, with most activity being observed during the Fall, Winter and Spring, may be an adaptive shift from summer activity during a cooler (than present) Pleistocene climate. Howden noted that many widespread species of Geotrupidae are active during th e cooler months in more southerly, warmer climates, such as Florida, whereas these species are active during the summer farther north. Howden also noted that the deep burrowing habit of Mycotrupes could be an additiona l adaptation to higher temperatures. Conservation Status The recognition of the unique and threatened nature of the sandy, we ll-drained habitat of Mycotrupes, coupled with the restricted geographic di stributions of the species, has promoted some interest in the conservation of species in this genus. However, the poorly known biology of Mycotrupes, and an incomplete knowledge of their distributions are impediments in any attempt to assess their vulnerabil ity to human disturbance. Mycotrupes gaigei has been classified as Imperile d (NatureServe 2009a), based on its specific habitat requirements and esti mates of its distribution. It has also been classified as Rare by the Florida Natural Areas Inventory (2009 ) and by the Florida Committee on Rare and 25

PAGE 26

Endangered Plants and Animals (F CREPA) (Woodruff and Deyrup 1994b). Mycotrupes cartwrighti has been classified as Rare in Florida by FCREPA, based on its restricted distribution in that state (Woodruff and Deyrup 1994a). Mycotrupes pedester appears to have a relatively limited distribution (see Distribution) and has been classified as Imperiled (Florida Natural Areas Inventory 2009), Critically Imperiled (Nat ureServe 2009b), and Threatened (Woodruff and Deyrup 1994c). While the assignment of a th reat level to a taxon as poorly known as Mycotrupes may be somewhat subjec tive, the limited number of known occurrences of Mycotrupes pedester along with the high rate of human development in Florida, should certainly warrant concern (Woodru ff and Deyrup 1994c). 26

PAGE 27

Figure 2-1. Distributi on map for species of Mycotrupes. Locations are approximate and based on location data from Appendix B Species color code: Red= M. retusus, green=M. lethroides yellow= M. cartwrighti maroon= M. gaigei blue= M. pedester, pink=possible cryptic Mycotrupes species. 27

PAGE 28

Figure 2-2. Mycotrupes species. a, M. lethroides male. b, M. lethroides female. c, M. retusus male. d, M. retusus female. e, M. cartwrighti male. f, M. cartwrighti female. g, M. gaigei male. h, M. gaigei female. i, M. pedester male. j, M. pedester female. Figure 2-3. Hubbell's hypothe sis of evolution in Mycotrupes (Hubbell 1954, p. 43). 28

PAGE 29

Figure 2-4. Mycotrupes speciation events as hypothesized by Hubbell (1954) (Figure 3, p. 49). Shoreline history (Pliocene to recent) is s hown with the supposed distributions of the species of Mycotrupes, including common ancestors For example, "C-P 2 represents the common ancestor of M. cartwrighti and M. pedester. 29

PAGE 30

CHAPTER 3 BEHAVIOR Introduction The behavior of Mycotrupes is poorly known, partly because they are rarely observed except in pitfall traps, which pr ovides little information besides habitat and relative abundance. In addition, Mycotrupes may spend a considerable portion of their lives underground, where their behavior would not be observed ex cept by excavation of their burrows. Diel patterns of adult activity are usually no t indicated by collecting data, because pitfall traps are often left out for 24 hours or more. Limited field observations by Howden (1954) indicate that M. gaigei adults are active duri ng the day and early eveni ng. Howden collected (in pitfall traps) adults of M. gaigei in High Springs, and near Arch er, Florida between the hours of 12:00PM and 8:30PM. The activity of adult Mycotrupes, as indicated by collecting records, appears to be concentrated in the cooler months (fall, winter, and spring) (Howden 1963). Howden hypothesized that this seasonality represents a shift from summer activity during an earlier (last glacial maximum) cooler period, to fall, winter, and spring activ ity as the climate warmed. Howden reasoned that a taxon adapted to a cooler climate could respond to a warming climate by either 1) shifting its distribution to a cooler latitude, or 2) shift its activity to a cooler season. Mycotrupes is flightless, and apparently restricted in distribution to patchy habitat, so the later possibility seems the more plausible of the two. Little is known of th e feeding habits of Mycotrupes. In several instances, adult Mycotrupes have been observed feeding on f ungus. LeConte (1866) mentioned that M. retusus was found under decomposing fungi. This prompted him to propose the name Mycotrupes. Horn (1868) mentioned that LeConte distributed specimens to some major European 30

PAGE 31

collections...as fungivorous. Harpootlian (1995) collected M. lethroides from under earthball mushrooms ( Scleroderma sp). Mycotrupes, like many other geotrupids, are known to dig deep burrows. Burrows as deep as 91 cm ( M. retusus ) and 208 cm ( M. gaigei ) have been recorded (Howden 1954). Mycotrupes burrows are often indicated at the surface by mounds of soil ("pushups") up to several inches high. Burrows usually open directly underneath these pushups and extend vertically downwards (Howden 1954; pers. obs.). Burrow diameter appears to be approximately as wide as the beetle (pers. obs.). Other geotrupid beetles dig simila r burrows, but with experience it is possible to recognize Mycotrupes burrows. For example, beetles of the genus Peltotrupes occur in many of the localities where M. gaigei is found, and this genus also digs very deep burrows (Woodruff 1973). The greater diameter, a nd larger soil "pushups," of Peltotrupes burrows help distinguish them from the burrows of M. gaigei Burrowing activity in Mycotrupes often appears to be concentrated in a given area. Howden (1954) ex cavated an area three feet wide by six feet long that contained nine burrows of M. gaigei Adult geotrupids typically provision cells (in burrows) with food for larval development (Howden 1955). The larval food of Mycotrupes has been observed only once. On March 20, 1953, H.F. Howden and B.K. Dozier excavated a deep (208 cm) burrow of M. gaigei in High Springs, Florida. Several larval cells were f ound at various depths, but most of the larvae had completed development, and little identifiable food matter remained. In one cell, where the resident larva had not yet completed development, there was material that appeared to be old cow dung. The area was being used as cattle past ure at the time, and cattle dung was abundant. The pupal cells constructed by the M. gaigei larvae consisted of an extremely thin (1 mm thick) layer of larval feces (Howden 1954). 31

PAGE 32

The following observations of Mycotrupes biology were made during my field and laboratory work. The next section combines these observations with information obtained through personal communications. Field and Laboratory Observations on Mycotrupes Behavior Adult Feeding Behavior Mycotrupes lethroides On November 3, 2007, I traveled to the Yu chi Wildlife Management Area (YWMA) near Girard, Burke County, Georgia. The area in whic h I made observations was generally pine and oak forest with much open sand. It was soon apparent that M. lethroides was active at this location in large numbers. At approximately 3: 00 PM, beetles were seen crawling on the surface of a sandy road in an area shaded by oak trees. On this road, I observed a male feeding on an acorn in a shallow, cup-shaped depression (not a burrow), which was apparently excavated by the beetle. The acorn was standi ng upright, and the beetle was stra ddling it with its legs wrapped around the acorn. The beetle was eating the soft tissue that was exposed through the top of the acorn. In the process of excavat ing burrows of beetles in the same area, acorns and oak leaves were found in several of the burrows with single adults (Paul C hoate, pers. comm.). At the YWMA, on November 6, 2007, pig dung-ba ited pitfall traps, which had been set on November 3, 2007, were picked up. In one of the pitfall traps, I obser ved a beetle feeding on a caterpillar (Lepidoptera) that had apparently fallen into the trap with the beetle. A live series (5 males; 5 females) was collected at the YWMA on November 6, 2007, and maintained in the laboratory at the Department of Entomology & Nematology, University of Florida, Gainesville, Florida. The beetles were kept in a glass terrarium (approximate dimensions: 40 cm long, 20 cm wide and 20 cm hi gh), which was mostly filled with sand. The sand was periodically moistened with water. I offered a variety of potential foods to these 32

PAGE 33

beetles. The foods were chosen on their similarity to foods Mycotrupes have been observed feeding on in nature. Although it is not naturall y occurring, dog food was also offered, because other species of Mycotrupes have fed on this material in the laboratory (P. Skelley, pers. comm.). On November 7, 2007, several peanuts and cashe ws were soaked in water for 30 minutes and placed in the cage. I observed one female f eeding on a cashew. Later, on the same day, this female attempted to drag the cashew backwards. She used her fore-tibial spurs to grasp the cashew and used her middle and hind legs to wa lk backwards. Although this beetle abandoned the cashew after approximately 10 minutes, all peanuts and cashews were buried by the next day (November 8, 2007). On December 6, 2007, I observed a female feeding on a cashew. She excavated a burrow next to the cashew, and pulled th e cashew down into the burrow. On November 13, 2007, at 2:20PM, three fres h Portobello mushrooms were placed in the cage; two of these were "planted" with their stems in the sand. Ten minutes later, at 2:30PM, a female started to feed on one of the mushrooms; she excavated a small hole in the sand next to the mushroom, and continued to feed from the hole. On November 27, 2007, I placed four pieces of dog food (moistened with water) on the sand in the cage. I observed a single female atte mpting to feed on a piece. Minutes later, she pulled the dog food down into a burrow, holding the dog food with her front legs and walking backwards with her middle and hind legs ( Figure 3-1 ). Other behavior observed incl uded the following. A pair wa s observed crawling out of a burrow en copula (S. Bybee and B. Smith, pers. comm.). On several o ccasions, beetles were observed sitting, or "perching," at the top of a burrow, waving their antennae as if "smelling" the air ( Figure 3-2 ). When disturbed, they would often retreat quickly into the burrow. 33

PAGE 34

Mycotrupes cartwrighti On March 7, 2007, in upland pine vegetation at the Tall Timbers Research Station, Leon County, Florida, several burrows were excavated and found to contain M. cartwrighti At the bottom of one burrow that did not contain any beetles, I found a ma ss of material that appeared to be hair. I identified it as mammal hair, probably opossum ( Didelphis virginiana Kerr), after studying the mammal collection at th e Florida Museum of Natural History (Gainesville Florida). In several instances, adults were found a ssociated with dung. In February, 2007, D. Almquist observed deer dung pellets in a beetle burrow in a field at Elinor Klapp-Phipps Park (Tallahassee, Leon County, Florida). Almquist also found a specimen in a pile of deer dung at Tall Timbers Research Station (Leon County, Florida). On December 13, 2007, in a pine-oakhickory forest at Miccosukee Canopy Greenwa y (Leon County, Florida), Almquist found a specimen in a 20 cm burrow under what appeared to be old horse dung (D. Almquist, pers. comm.). Mycotrupes gaigei Mycotrupes gaigei adults were observed feeding on, or in association with, several substances, in the vicinity of an alpaca ranc h and a residential area in Summerfield, Marion County, Florida. One beetle was collect ed as it was apparently feeding on an Opuntia pad at the alpaca ranch. The beetle was found on the unders ide of the pad, which had evidence of feeding damage. Beetles were observed feeding on cat fo od which had been left in a dish on a porch in the front of a house in the residential area. Burrows, apparently of M. gaigei were observed among alpaca dung at the alpaca ranch in June, 2006 (R.E. Woodruff, pers. comm.). This suggests the possibility that the beetles were using the alpaca dung as food for adults and/or larvae. 34

PAGE 35

On two occasions, adults were found associated with dung of an unknown mammal in sandhill vegetation approximately three miles west of Archer, Levy County, Florida. The dung contained hair, and could have been from a dog, coyote, or bobcat. On the first occasion, one beetle was found feeding on the dung. On the second occasion, one beetle was excavated, along with several dung beetles ( Phanaeus igneus floridanus d'Olsoufieff), from the soil underneath a pile of dung (P. Skelley, pers. comm.). Burrowing Mycotrupes lethroides At the YWMA, on November 3, 2007, I excavated burrows of M. lethroides in a sandy road in an oak forest. These burrows appeared to be concentrated in areas with packed, moist soil, whereas looser, drier sand covered most of the road. Upon excavation, the burrows were found to be plugged with sand for part of their dept h. Four burrows were ex cavated with a single specimen in each (3 males, 1 female). The beet les were found at depths ranging from 8 to 12 cm. At another location in the YWMA, a single specimen was excavated at a depth of 25 cm. Mycotrupes cartwrighti On March 7, 2007, I excavated, with the he lp of D. Almquist and P. Skelley, two M. cartwrighti burrows in a longleaf pine forest at Elea nor Klapp-Phipps Park in Tallahassee, Leon County, Florida. The area was hilly, and the soil had an orange-red color. Neither burrow had a pronounced "push up" of excavated soil, but instead had a raised rim of clay at the surface. At the bottom of each burrow was a single male, at a depth of 20 cm in one burrow and 24 cm in the other. On March 15, 2007, I excavated nine burrows at Thomasville, Thomas County, Georgia. The burrows were in a recently burned pine fore st with an open understory and scant leaf litter ( Figure 3-3 ). These burrows were similar in depth to those at Eleanor Klapp-Phipps Park 35

PAGE 36

(approximately 20 cm), and the soil here also had an orange color. Some were covered with small (approximately 5 cm in diam eter) pushups of excavated soil ( Figure 3-4 ), while others were open at the top ( Figure 3-5 ). A single beetle was found in eac h burrow; in total, 6 males and 3 females were collected. The beetles were not found in the open space of the burrow itself, but were found a short distance (~3 cm) away in packed cl ay that appeared to form a distinct layer in the soil. Mycotrupes gaigei On March 6, 2008, I excavated a single M. gaigei burrow at O'Leno State Park, High Springs, Columbia County, Florida. The s ite was surrounded by typical sandhill vegetation, including Longleaf Pine ( Pinus palustris Mill.) and wire grass (possibly Aristida sp.). In the burrow, I found one beetle at a depth of appr oximately 120 cm. The beetle was located in packed sand approximately 5 cm away in a latera l direction from the apparent termination of the burrow. At the same location in O'Leno State Par k, on March 6, 2008 at 12:00PM, P. Choate and I placed seven pitfall traps baited with pig dung and fermenting malt on the ground in a sandy road in the sandhill area. By 1:40PM, five beet les had been trapped. By 4:15PM (when the traps were removed), two additional beetles had been trapped. On February 17, 2009, a burrow was excavated in O'Leno State Park. A single female was found at a depth of approximately 160 cm. Mo st of the burrows that appeared to be of M. gaigei at this site were covered with a pushup of excavated soil approximately 7 cm in diameter and 7 cm high. These pushups were sometimes obscured, possibly as a result of rain. 36

PAGE 37

Larval Food Mycotrupes gaigei In the Spring of 2007, P. Skelley placed five M. gaigei collected from a location west of Newberry, Gilchrist County, Florida, in a glass terrarium. This terrarium was filled to a depth of approximately 25 cm with sand. Dry dog food, whic h was periodically placed in the cage, was buried by the beetles. In August 2007, rotting l eaf litter was placed on top of the sand in the terrarium. This material was also buried by th e beetles. On October 26, 2007, P. Skelley and I carefully excavated the burrows in the terrarium. Approximately 5 cm below the surface of the sand, a mass of leaf litter measuri ng approximately two inches in length was found. As the rest of the sand was excavated, several burrows were f ound to be partially filled with leaf litter at different depths. At the bottom of the terrarium, a large mass of leaf litter was found which measured approximately 7 cm in length. In this mass, a small (probably first instar) Mycotrupes larva was found. On the other side of the terrariu m, also at the bottom, another similar larva was found. I placed the two larvae in individual vials with rotting leaf litter and sand for the purpose of rearing them to the adult stage. On N ovember 28, 2007, one larva was found dead. The other larva was still alive, and appeared to be slightly larger than be fore. It was seen manipulating, with its mouthparts, light brown pl ant matter, probably leaf litter. It may have been feeding. Unfortunately, this larva died soon after that observation. Discussion Field and laboratory observ ations suggest that adult Mycotrupes are generalist feeders. Among the foods that M. lethroides was observed to feed upon in the field was an acorn, and acorns were found in burrows of th is species. Prez-Ramos et al (2007) and Verd et al (2007) reported acorn feeding by the geotrupid Thorectes lusitanicus Jeckel in Spain, so this habit may be common in the Geotrupidae. The fact that M. lethroides fed on cashews and dog food, foods 37

PAGE 38

which do not occur naturally over the distribution of any Mycotrupes species, further suggests that adult Mycotrupes will feed on a wide variety of foods. The ability of adult Mycotrupes to exploit a wide variety of foods would be advantageous in a heterogeneous and unpredictable environment. This is especially true because Mycotrupes are flightless, and cannot cover as much area in search of food. A dependence on one particular food, such as dung, would be a risky strategy for a flightless sp ecies. Known foods of adult Mycotrupes, including fungus, acorns, and other materials, vary in relative a bundance throughout the year (and across years, for acorns [Whitney et al. 2004, p. 99]), and as a result, the diet of Mycotrupes probably varies over time as well. The larval diet of Mycotrupes remains unknown except for a single observation by Howden (1954), in which case the larval food appeared to be ol d cow dung. In this study, adult M. gaigei provisioned larval chambers with decomposi ng leaf litter in the laboratory. The use of a material by adult M. gaigei in the provisioning of larval ch ambers in the laboratory is not evidence that this material is used in this way in nature. The failure of the two larvae to develop could have been caused by a variety of factors, among them, the possibility that the leaf litter was an inadequate food. However, records of ot her geotrupids using leaf litter as a larval food suggests the possibility of similar behavior in Mycotrupes Howden et al. (2007) recorded the use of humus as food for larvae in the Australian geotrupid genus Bolborhachium In addition, leaf litter is used as la rval food (along with othe r materials, such as dung) in the geotrupid genera Geotrupes and Peltotrupes (Howden 1952; Howden 1964; Young et al. 1955). Peltotrupes profundus Howden and M. gaigei are sometimes found together, and both species dig deep burrows. One possible function of deep burrows in Mycotrupes could be to allow, in a more humid underground environment, the decomposition of refractory leaf litter into a microbe and 38

PAGE 39

fungus-rich food of high er nutritional quality. The ability of Mycotrupes to utilize leaf litter as a larval diet would be advantageous as this material is widely distributed. Field observations made over the co urse of my study indicate that M. gaigei and M. lethroides are active during the daytime. This supports Howden's observations of daytime activity in M. gaigei It is therefore likely that the entire genus Mycotrupes is diurnal, although the possibility of night-time activity cannot be excluded. The seasonality of adult Mycotrupes activity that was noted by Howden (1963) is supported further by Table 3-1 which is based on the specimens available to me in this study ( Appendix B ). There are several caveats in the interpreta tion of this data. It is possible that this apparent seasonality is partially due to seasona l bias in collecting effort. Collectors who are attempting to find Mycotrupes would likely collect during seasons that are perceived as being the most productive. The analys is of the large number of M. cartwrighti collected in unbaited pitfall traps at Tall Timbers Research Station by D.L. Harris, W.H. Whitcomb, and W.W. Baker, which is currently stored at the Flor ida State Collection of Arthropods (Gainesville, Florida), might provide valuable information of seasonality. Ev en barring seasonal collec ting bias, it should be remembered that most Mycotrupes collecting records are the resu lt of pitfall trapping. This means that, at best, these numbers are indicativ e of adult activity above ground. There could be seasonal differences in the attractiveness of the baits used in pitfall traps; this might be expected if there are seasonal differences in food pref erence (e.g., feeding during a period of adult maturation versus gathering of food fo r the provisioning of larval cells). Laboratory observations suggest the possibility that mating in Mycotrupes takes place underground. Mycotrupes being flightless and a pparently diurnal, are probably quite vulnerable to predation when above ground. The restricti on of as many activities as possible, including 39

PAGE 40

mating, to underground burrows, may reduce the risk of predation. When Mycotrupes were excavated from burrows, they were often located some distance away (several centimeters) from the open burrow and were packed in soil. By ensconcing themselves in soil, Mycotrupes may protect themselves from predators that enter their burrows. Adult Mycotrupes were excavated from burrows that ranged in depth from 8 to 160 cm. The limited observational data available suggest that burrow depth may depend on the species of Mycotrupes, and possibly on the soil type as well. Mycotrupes gaigei dig the deepest burrows known for the genus. Howden (1954) found larvae of M. gaigei at depths ranging from 140 to 208 cm. In this study, the excavated burrows of M. cartwrighti and M. lethroides were much shallower (8-25 cm). Moisture may be important in determining the depth of Mycotrupes burrows. This possibility can be illustrate d with a comparison between the burrows of M. gaigei and those of M. cartwrighti The soil at the bottom of excavated M. gaigei burrows often appeared to be moister in relation to th e soil at the surface. In the case of M. cartwrighti the depth at which adult beetles were found was often marked with a distinct layer of clay. This suggests two possibilities: 1) the clay impe ded burrowing, or 2) the clay provided some favorable condition, such as resistance to de siccation. In addition, the soils in which M. cartwrighti occurs in may be more resistant to desi ccation, and this species may, as a result, not need to burrow as deeply as M. gaigei The soil present at the type locality of M. cartwrighti (6.5 mi E. of Tallahassee, Florida) was desc ribed as an Orangeburg sandy loam (Olson and Hubbell 1954). Such soils, with greater clay content, have a higher water retention capacity compared to well-drained sands (which are characteristic of the distribution of M. gaigei ) (Bouma et al. 1982). Howden et al. (2007), who found the brood cells of the Australian bolboceratine geotrupid Bolborhachium to be located just above a layer of clay, proposed the 40

PAGE 41

idea that such a position would be ideal for preserving moisture. Desiccation would be a serious problem for Mycotrupes larvae or adults, which are found in well-drained soils. Henderson (1939) states that soils of the Norfolk seri es, which Olson and Hubbell (1954) note as being present in many of the localities where M. gaigei occurs, have a layer of friable sandy clay at a depth of 6-8 feet. This further support s the idea that the deep burrows of M. gaigei may be a result of a need to access a so il horizon with a higher capacity to resist desiccation. However, M. lethroides which lives in very sandy soil that appears similar to that inhabited by M. gaigei was found in relatively shallow burrows. A careful study of the different soils associated with Mycotrupes burrows might yield much useful inform ation. Differences in soil morphology may be responsible for the apparent absence of Mycotrupes from scrub habitat in the central ridge of Florida. It is possible that Mycotrupe s excavate distinct burrows for different purposes. For example, burrows constructed for adult feeding and resting could differ in depth (and possibly other characteristics) from those excavated for nidification. This is an other reason, besides the fact that so few burrows have been excavated, that it may be premature to compare burrows across species until more is known about the biology of each species. It would also be useful to excavate burrows containing la rvae for species other than M. gaigei which is the only species for which a burrow containing larvae has been de scribed. Such fortuitous discoveries (in my experience, excavating burrows has only yielded a dults, and it is hard work and time consuming) would allow the study of variation (within a nd across species) of burrow depth, larval food, design of the larval chamber a nd developmental phenology. This kind of biological information could aid in the understandi ng of the biogeography of Mycotrupes as well as augment conservation measures aimed at protecting these species. 41

PAGE 42

Although knowledge of the biology of Mycotrupes is fragmentary, it is possible to speculate on the timing of the life cycle in the ge nus. As mentioned earlier, H.F. Howden and B.K. Dozier excavated a mature larva of M. gaigei along with the empty cells of larvae which had presumably recently completed development, in March (Howden 1954). Teneral specimens of M. gaigei have been collected in March in Florida (P. Choate, pers. comm .). Monthly totals of Mycotrupes collections suggest that a larger number of specimens are collected in the spring, compared to the fall ( Table 3-1 ). Taken together, th ese data suggest that Mycotrupes emerge as fresh adults in the spring. There is no data on development time in Mycotrupes ; this process could take more than a year. Howd en (1954) kept an adult female of M. retusus alive in captivity for 13 months, suggesting a lengthy life. The possible use by Mycotrupes of such abundant material as leaf litter for larval food, combined with a long reproductive life, would allow for the provisioning of a large number of larval cells, po ssibly over an extended length of time. Figure 3-1. Female M. lethroides pulling a piece of dog food. 42

PAGE 43

Figure 3-2. M. lethroides "perching" at the top of a burrow. Figure 3-3. M. cartwrighti habitat at Thomasville, Ge orgia. Photo by P. Choate. 43

PAGE 44

Figure 3-4. M. cartwrighti burrow with "pushup." Photo by P. Choate. Figure 3-5. M. cartwrighti burrow without "pushup." Photo by P. Choate. 44

PAGE 45

Table 3-1. Seasonality of sp ecimen data from Appendix B. Jan. Feb. Mar. April May June July Aug. Sept. Oct. Nov. Dec. M. lethroides 4 29 4 32 6 M. retusus 5 49 2 3 13 25 10 M. cartwrighti 42 4 33 20 9 7 2 31 22 11 64 33 M. gaigei 154 125 448 58 58 29 2 7 2 41 1 15 M. pedester 8 5 21 178 1 1 1 Cryptic species? 2 2 45

PAGE 46

CHAPTER 4 VARIATION IN THE PARS STRIDENS OF THE STRIDULATORY APPARATUS OF MYCOTRUPES Introduction Stridulation, defined as the production of a "shrill creaking noise by rubbing together special bodily structures" (Merriam-Webster 2009), is common in insects. Although it is perhaps most often associated with the Orthoptera, stridula tion is known to occur in at least seven insect orders (Ewing 1989). Stridulation is known to occur during mating, and as a response to disturbance, and it has been shown to have a dete rrent effect against predators. Stridulations produced in response to disturbance are termed "d istress" stridulations (Bauer 1976; Buchler et al. 1981; Hirschberger 2001; Masters 1979; Wink ing-Nikolay 1975) and can function to repel predators; the effect on the pred ator may be auditory or tactil e (Dumortier 1963). In insects, stridulation appears to re pel predators as diverse as wolf spiders, mice, and sand pipers (a bird) (Bauer 1976; Masters 1979). Stridulation is known in at least 30 families of Coleoptera, and this behavior has evolved numerous times in the order (Wessel 2006). In beetles, stridulation is accomplished through a wide variety of sound producing structures, perh aps due to the highly sclerotized beetle body (Arrow 1942; Wessel 2006). In the beetle family Geotrupidae, stridulation occurs in adults, and probably in larvae as well, as indicated by the presence of stridulatory structures (Howden 1954; Woodruff 1973). In adults, stridulation has been shown to play a role in competition and mating (Winking-Nikolay 1975). At least two different sound-producing mechanisms are known in adult geotrupids: 1) the coxo-abdomin al apparatus, and 2) the thorax -elytral apparatus (Carisio et al. 2004). Although they are composed of diffe rent structures, both mechanisms produce sound through rhythmic movements of the abdomen. 46

PAGE 47

The coxo-abdominal apparatus is composed of a pars stridens (file) and a plectrum (scraper). The pars stridens is a raised and ribbed area on the posterior surface of the hind coxa, and is mostly hidden within the coxal cavity. Th e plectrum is located on the inner face of the coxal cavity. The morphology of the plectrum is poorly understood; it has been described as a sclerified ridge on the posterior border of the abdominal sternite (the "hinterrand" of WinkingNikolay 1975) and as a field of dentiform processe s in the area of the coxal cavity (Palestrini et al. 1988; Palestrini & Pavan 1995; Carisio et al. 2004). Sound is produced through the scraping of the pars stridens by the plectrum. The m ovement of the plectrum is accomplished through shortening and lengthening of the abdomen; this movement may have been derived from the preflight "pumping" movement in Coleoptera (Winking-Nikolay 1975). The coxo-abdominal stridulatory apparatus is common in the Geotrupidae and appears to be best developed in (and possibly present in all members of) the subf amily Geotrupinae (Arrow 1904). Zunino and Ferrero (1988) studied the pars stridens in 17 genera of Geotr upidae and found differences in the density of ribs across speci es and in some cases sexua l differences as well. Mycotrupes species are morphologically similar, and a limited number of characters separate them. Mycotrupes stridulate when handled; this may be a defensive behavior. My observations indicated variation in the number of ribs of the pa rs stridens across species of Mycotrupes. The pars stridens had not been previously studied in this genus, and the focus of my study was to investigate whether morphological di fferences in this stru cture could be useful for delimiting species. In order to quantify variation in the number of ribs across different species, 38 specimens of Mycotrupes were examined (5-10 specimens per species). For each specimen, body size and 47

PAGE 48

number of ribs per pars stridens were recorde d. The effects of species, sex, and body size on the rib count of the pars stridens in Mycotrupes are described below. Materials and Methods Dead, pinned specimens of Mycotrupes were studied from several collections (American Museum of Natural History, Univ ersity of Michigan, United Stat es National Museum) as well as self-collected specimens from across the known ra nge of these species (Florida, Georgia, and South Carolina). Sample sizes of specimens used for the morphological study were as follows: M. cartwrighti Olson and Hubbell (N=5; 4 1 ), M. gaigei Olson and Hubbell (N=9; 5 4), M. lethroides (Westwood) (N=5; 3 males, 2 ), M. pedester Olson and Hubbell (N=9; 4 5) and M. retusus (LeConte) (N=10; 7 3 ). All morphological observations were made with a Leica MZ16 dissecting microscope fitted with an ocular micrometer. Because of the possible effect of body flexure on resulting measurements of body length, body size was measured as the width of the pronotum, viewed dorsally, at its widest point. A ll observed ribs on the pars stride ns were counted. It was often necessary to place specimens in hot water for seve ral minutes in order to facilitate flexing the coxa and expose the pars stridens. The following statistics were used to analyze data: The Mann-Whitney test was used to test the effect of sex on rib count (Avery 2008). An Analysis of Covariance (ANCOVA), implemented in R (R Development Core Team 2008), was used to test two null hypotheses: 1) there are no differences between species in the number of ribs in the pars stridens, and 2) there is no effect of body size (width of pr onotum) on the number of ribs in the pars stridens. The small sample sizes precluded the use of ANCOVA to test the effect of sex on rib count. Assistance with the ANCOVA analysis was provided by D. Bustamante. 48

PAGE 49

Results The general form of the pars stridens in Mycotrupes conforms to that present in Trypocopris and other genera of Geotrupinae (Carisio et al. 2004). Ribs along the main portion of the pars stridens are raised and distinctly pronounced ( Figure 4-1 ). The pars stridens has less relief and the ribs become less pronounced (but still are demarcated by stri ae) at each end of the pars stridens toward the margins of the coxa. There were no apparent morphological differences between the pars stridens in different species of Mycotrupes, other than the number of ribs. The number of ribs in the pars stridens varies within and across species of Mycotrupes ( Figure 4-2 ). Mean rib counts were as follows: M. lethroides 224.6; M. retusus, 169.3; M. gaigei 120.9; M. cartwrighti 118.2; M. pedester, 104.2. Within all species of Mycotrupes there were no sexual differences in rib number (Man n-Whitney test) at the 0.05 confidence level: ( M. lethroides U=3, p=1; M. retusus U=14.5, p=0.38; M. gaigei U=15, p=0.29; M. cartwrighti U=2, p=1; M. pedester, U=9.5, p=0.91). After fitting an ANCOVA model to th e data including all species of Mycotrupes a visual assessment of the residual plots (standardized residuals versus predicted values) indicated a strong, non-random pattern that invalidated the a ssumption of homoscedasticity; in other words, the variance appeared to differ across sp ecies. This pattern was due to the M. lethroides data. Data transformation failed to improve model diagnostics. Thus, a second model, excluding M. lethroides was fitted. This model conformed better w ith the assumptions of homoscedasticity and normality of residuals, which increased confidence in testing the two proposed null hypotheses. The results of this new model i ndicated that the value of the slope for the relationship between pronotum si ze and rib counts was not signifi cantly different among species (F=0.75, p=0.53), however at least one of the inte rcepts was significantly different from the intercept of other species (F =131.10, p=2.02e-15), indicating that at least one species had a 49

PAGE 50

significantly different number of ribs. There was a significant positive relationship between size (pronotum width) and rib count (slope=4.37, F=2.02, P=0.03). Pair wise comparisons revealed the following significant differences in intercepts: M. retusus > M. cartwrighti, M. gaigei, M. pedester and M. gaigei > M. pedester ( Table 4-1 ). The results of the ANCOVA support rejecting the null hypotheses proposed, indicating that there are differences in the rib counts among species (suggested by significant intercept differences), and that within species, the number of ribs has a moderate positive rela tionship with body size. Although M. lethroides was excluded from this model, visual inspecti on of the rib counts of five indi viduals indicates a difference in rib count between this and the other species ( Figure 4-2 ). Discussion There are differences in the number of ri bs on the pars stride ns across species of Mycotrupes, and there is a positive relationship betw een body size and number of ribs. No sexual differences in rib count were found in this study. The behavioral significance of stridulation in Mycotrupes remains unknown. Stridulation may be used as a means of defense against predators, as stridulation has been shown to have a deterrent effect against predators in other be etles. In Australia, Howden et al. (2007) occasionally found flightless ground beetles (Coleopt era: Carabidae) feeding on adults of the geotrupid Blackburnium "at the bottom of th eir burrows." As Mycotrupes are not able to fly and presumably spend a considerable portion of th eir time underground, it might be important for them to have an effective means of repelling any predator that might gain access to their burrows. Young et al. (1955) repor ted stridulation by the geotrupid Peltotrupes profundus upon excavation of their burrows. Though little is known regard ing the life history of Mycotrupes, it is possible that communication between adults and/or larvae oc curs in underground burrows. Stridulation may 50

PAGE 51

be involved in mating. A mating pair of Mycotrupes lethroides was observed leaving a burrow in a laboratory setting, sugges ting that mating may occur underg round (see Chapter 3) (S. Bybee and B. Smith, pers. comm.). Because Mycotrupes species are presently understood to be completely allopatric in distribution, there is no reason to expect continued selection for specific differences in acoustic signaling related to mating. Aggression is another possible function for stridulation in Mycotrupes The apparently aggregated nature of Mycotrupes burrows in the field may promot e aggressive interactions, and specimens of Mycotrupes lethroides have been observed in the laboratory pushing each other out of burrows (S. Bybee, pers. comm.). The geotrupid Thorectes intermedius stridulates aggressively when defending a burrow from invadi ng beetles (Palestrini a nd Pavan 1995). It is possible that similar aggressi ve stridulation is used by Mycotrupes. Behavioral experiments with Mycotrupes, similar to those conducted by WinkingNikolay (1975), might yield important information regarding the significanc e of stridulation in this genus. The stridulations of Mycotrupes are clearly audible to the human ear. Recordings were made of stridulations of four species of Mycotrupes in the hope of finding possible differences in stridulation resulting from the differences in pars stridens rib c ount. The recorded stridulations were variable within species as well as within individuals. Unfortunatel y, the poor quality of the recordings made it impossible to study them further. 51

PAGE 52

Figure 4-1. Pars stridens on the posterior face of the right metacoxa of a male Mycotrupes pedester Figure 4-2. Relationship betw een rib counts of the pars stridens and body size in Mycotrupes. 52

PAGE 53

Table 4-1. Results of the anal ysis of covariance: pairwise comparisons of the differences between intercepts for th e four species model ( M. cartwrighti M. gaigei M. pedester, and M. retusus). D is the value of the difference between the intercepts, t (se) denote the t value and standard error for the test of significant differences between the intercepts, and p is the probability of obtaining a value larger than |t|. M. cartwrighti M. gaigei M. pedester D t (se) P D t (se) P D t (se) P M. gaigei 8.25 1.68 (4.91) 0.10 M. pedester -9.43 -2.01 (4.69) 0.05 -17.69 -4.92 (3.60) 3.46E-05 M. retusus 55.62 12.03 (4.62) 1.39E-12 47.37 13.51 (3.51) 8.72E-14 65.06 18.72 (3.48) <2.00E-16 53

PAGE 54

CHAPTER 5 DELIMITING SPECIES BOUNDARIES AN D DIAGNOSING POSSIBLE CRYPTIC SPECIES IN MYCOTRUPES Introduction Hubbell (1954) hypothesized that Mycotrupes speciated allopatrical ly in response to sea level changes. Modern phylogenetic methods pe rmit a test of this hypot hesis with nucleotide sequence data. But before such a study can be do ne, it is necessary to have some confidence in the known species limits of Mycotrupes Although molecular data is becoming more and more popular within insect phylogeneti cs, the identification of inse ct species is still largely accomplished with morphological characters. Distinguishing Mycotrupes species is often difficult, as there are few mo rphological differences between the species. While Olson and Hubbell (1954) presented ch aracters that facilitate their identification, the geographic origin of a specimen is often the primary criterion us ed for determining its identity. Given the conservative morphology, patc hy habitat, and flig htless habits of Mycotrupes (leading to little gene flow be tween populations), I would suspect and others have suggested, that there is an increased potential for the existence of cryptic ( unrecognized) species in Mycotrupes (Woodruff 1973). The problem then become s how to define a species when species concepts abound (Wheeler and Meier 2000). Species concepts vary in thei r focus on the process of species formation (i.e., theo retical concepts), and in their ability to diagnose species (i.e., operational concepts). Two of the more recent species concepts, both of which are operational concepts, are the Phylogenetic Species Concept (PSC) of Mish ler and Theriot (2000), and the PSC of Wheeler and Platnick (2000). The PSC of Mishler and Theriot define species as monophyletic groups that can be supported by a phylogenetic analysis; they leave the decision for the "cut-off point" (the sma llest monophyletic group to be rec ognized) to the systematist. Wheeler and Platnick define a sp ecies as "the smalle st aggregation of (s exual) populations or 54

PAGE 55

(asexual) lineages diagnosable by a unique combination of character states." They do not require that a species be monophyletic or recoverable in a phyl ogenetic analysis; it mu st be defined prior to an analysis based on a unique combination of characters. Their (Wheeler and Platnick) species are testable in that the unique characters must be consistent across all members of the species, and unique to that specie s. For my study, I decided to test the species boundaries of Mycotrupes with the PSC of Mishler and Theriot w ith the criterion that a species will be recovered as monophyletic. Because of the morphologically conservative nature of Mycotrupes, I determined that the acquisition of molecular sequence data would probably be the most efficient means to obtain a phylogenetic data set. Because Hubbell (1954) hypothesized that Mycotrupes speciated in response to geologically recent sea level changes, I was inclined to select a more quickly evolving gene with an increased potential to offer differences across closely related species. Mitochondrial DNA is a relatively rapidly evolving a ssemblage of genes, and is a popular choice for low-level phylogenetics and population genetics study in insects (Avise et al. 1987; Funk and Omland 2003). In the case of Mycotrupes, the use of the mitochondrial Cytochrome Oxidase I gene (COI) would allow extracti on of a potentially rich source of phylogenetic character data from a group that is morphologically conser vative and possibly recently speciated. DNA barcoding is a recent development that seeks to use relatively short DNA fragments for the identification of species. Another possible application of such a short barcoding fragment is the separation of groups of sequences into probable distinct species based on a cutoff value of sequence divergence (Hebert et al. 2 003). This latter application provides a method of discovering possible cryptic species (which are often, at least initially, morphologically indistinct from other species) based on the disc overy of a high degree of sequence divergence 55

PAGE 56

between sequences. This method could be applied to Mycotrupes to assess the amount of divergence between species and possibly de termine the presence of cryptic species. The goal of my study was to test "traditional" species limits in Mycotrupes using phylogenetics and pairwise dist ances, and to obtain a phylogenetic hypothesis of the relationships between the species for use in a subsequent biogeographi c study (see Chapter 6). Materials and Methods Sampling I attempted to use fresh specimens from as many different collecting localities as possible ( Table 5-1 ). All five species of Mycotrupes were collected across much of the known distribution of the genus ( Figure 5-1 ). All, except for M. lethroides were represented by material collected from multiple locations. Historic records for M. cartwrighti from the vicinities of Jacksonville, Flor ida and Americus, Georgia could not be substantiated by my recent collecting efforts. Thus, no samples were available for DNA analys is. Most specimens were collected in pitfall tr aps baited with a combinati on of pig dung and fermenting malt solution and were brought back alive, if possibl e, to maximally preserve DNA. Field collected Mycotrupes specimens were stored in 95% ethanol at -80C in th e Branham Laboratory at the University of Florida, Gainesville, Florida. The use of outgroup taxa is currently the most widely accepted method for the polarization of characters in phylogenetic studies (Maddison et al. 1984; Nixon and Carpenter 1993). It was considered that the ideal outgroup for this study woul d be one that was as closely related to Mycotrupes as possible. Unfortunately, the phyl ogenetics of the Geotrupidae have received little study, and it was not possible to apply such a cr iterion for outgroup selection. Peltotrupes youngi Howden (Coleoptera: Geotrupidae) was chosen as an outgroup taxon. A fresh specimen was collected from the Ocala National Forest in Florida. 56

PAGE 57

DNA extraction, PCR Amplification, Se quencing and Nucleotide Alignments DNA was extracted, using the DNeasy blood and tissue kit (Qiagen Inc., Valencia, California), from one or two legs of each indivi dual. Amplifications were carried out using 1 L of extracted DNA template and 1 L of each DNA primer in a 25 L polymerase chain reaction (PCR) containing 12.5 L Accuzyme mix (Bioline, Taunton, Massachusetts), 0.4 L Taq polymerase (Bioline) and 9.1 L autoclaved H 2 O. To obtain the 481 bp (length after editing) fragment of the mitochondrial gene Cytochrome Oxidase I, the primers Tonya (5'GAAGTTTATATTTTAA TTTTACCGGG -3') and Hobbes (5'AAATGTTGNGGRAAAAAT GTTA -3') were used (Rand et al. 2000). PCR reactions were done in an Eppendorf Mastercycler EP Gradient Thermocycler (Eppendorf International) with the following cycle parameters: 3 minutes at 95 29 cycles of 1 minute at 94C, 1 minute at 48C and 1 minute at 72C, follo wed by 5 minutes at 72C and 15C for infinity. The PCR product was visualized for fragment length on agarose gels (1.6 grams agarose, 20ml TBE buffer, 180ml H 2 O and 200 L ethidium bromide) in TBE buffe r solution. Verified PCR product was purified with a QIAquick PCR purification kit (Qiagen Inc.). The purified product was sequenced in both directions with an App lied Biosystems Model 3130 Genetic Analyzer (Applied Biosystems, Foster City, California) at the Interdisciplinary Center for Biotechnology Research (ICBR), University of Florida. Seque nces were edited manually in Sequencher (Gene Codes 1998) or by the ICBR, and then aligned by eye in Sequencher. See Appendix A for sequence data. Phylogenetic Analyses Two methods were used for phylogenetic r econstructions. A Parsimony analysis was implemented in PAUP (Swofford 2002), and a Ba yesian analysis was implemented in BEAST (Drummond and Rambaut 2006). Because these phylogenetic hypotheses would be used later 57

PAGE 58

for the biogeographic portion of the study, two diffe rent Bayesian analyses were conducted with different "prior" information to give two different estimates of divergence times. Even though the time since divergence aspect of these Bayesi an analyses is irreleva nt for the purpose of Mycotrupes phylogenetics and species delimitation, the methods are described here. Parsimony analysis The Parsimony analysis was implemented in PAUP 4.0b10. A heuristic search was performed with the following settings: starting tree(s) obtained via stepwise addition; addition sequence = random; number of replicates = 10,000; number of trees held at each step during stepwise addition = 1; branch swapping algorith m = tree-bisection-reconnection (TBR); steepest descent option = not in effect; initial 'MaxTrees setting = 100; 'MulTrees option = in effect. Bootstrap support values were calculated in PAUP using 10,000 replicates and 10 randomaddition sequence replicates. Bayesian analysis BEAST ver. 1.4.8 was used to infer phylogeni es and obtain divergence time estimates. The aligned COI data were imported into BE AUTi (Rambaut and Drummond 2008a), and an xml file was created for each of the two analyses. The monophyly of the ingroup ( Mycotrupes ) was constrained. In order to obtain a temporal scale fo r dating nodes on the phylogeny, two approaches were used: a pre-set rate of nucleotide substitution and a biogeographic calibration. This required two xml files with different settings ( Table 5-2 ). For the pre-set rate, I used Brower's (1994) estimate of the sequence divergence rate (2.3% per million years). This pairwise rate was halved to obtain a linea ge divergence rate of 1.15% per million years, entered into BEAUTi as 0.0115 (mean rate). This resulted in the node ages indicated on the Bayesian trees as millions of years before present. 58

PAGE 59

The biogeographical calibration was done as follows: The topology supported by the BEAST analysis, using the pre-set ra te, indicated a clade made up of M. gaigei and M. pedester, the known distributions of which ar e less than approximately 50 mete rs above sea level. The sea level history of the past 100 million years was st udied by Miller et al. (2005). Approximately 5 mya the sea level was 50 meters above contemporar y sea level. After this point, high stands of sea level were successively lower. Assuming that M. gaigei and M. pedester never occurred at elevations higher than 50 meters above contemporary sea level, we could assume the ancestor of M. gaigei and M. pedester arose at a time later than the last time the sea level was at this elevation because such an ancestor would have been inundated if it were present before this time. Even with such an assumption, the 5 mya estim ate would only be a maximum age limit of the clade made up by M. gaigei and M. pedester, meaning that all of the node ages of the resulting tree would be maximum limits. A prior was set for the time since the most recent common ancestor (tMRCA) of the clade ( M. gaigei + M. pedester ) as normally distributed with a mean of 5 million years with a standard deviation of 0.1 million years. The use of a normally distributed prior age estimate accounts for uncertainty wh en incorporating bioge ographical events in divergence time estimation (Ho 2007). The model GTR+I+G was found to be the most appropriate according to a likelihood ratio test in Model Test v3.7 (Posada and Crandall 2005) impl emented in PAUP. A likelihood ratio test implemented in PAUP failed to reject a molecular clock so a strict molecular clock approach was used for both the pre-set rate as well as the biogeographical calibration runs. A Monte Carlo Markov Chain with a length of 10,000,000 was used for the pre-set rate runs; a chain length of 20,000,000 was used for th e biogeographical calibra tion, due to the low Effective Sample Size (ESS) values obtained with a shorter chain. Two chains were run for each 59

PAGE 60

analysis. Parameters and trees were logged every 1000 genera tions. Tracer (Drummond and Rambaut 2007) was used to check on the trace files of the chain. ESS values over 200 were obtained. LogCombiner (Drummond and Rambaut 2008b) was used to combine the trees sampled in the two chains with a burn-in of 1000 generations. TreeAnnotator (Drummond and Rambaut 2008c) was used to produce a maximum clade credibility tr ee with a posterior probability limit of 0.5, with a bur n-in of 1000 generations. The re sulting tree viewed in FigTree (Rambaut 2008) ver. 1.1.2 was checked for pos terior clade probabilit ies and divergence time estimates (with 95% highest probability density bars). Nucleotide Divergence A matrix of uncorrected p-distances wa s calculated in PAUP 4.10 from the sequence data. I calculated the mean pairwise divergence within each species of Mycotrupes, and the mean divergences across all Mycotrupes Testing Alternative Phylogenetic Hypotheses I used a method from Wuster et al. (2008) to test the support for alternative phylogenetic hypotheses. The following operation was performe d when the maximum clade credibility tree did not contain a monophyletic clade for one or mo re presumed species. I constructed a target tree constraining only the relationship of interest in this case, a monophyl etic clade containing all of the sequences of the presumed species. This target tree was used to filter, in PAUP, the post-burn-in tree files from the molecular clock BE AST run. The percenta ge of the total trees that had this constrained monophyl etic relationship was used as a measure of support. The alternative hypothesis was reject ed if it was supported by less than 5% of the total trees. Results The Parsimony and Bayesian topologies all support M. lethroides M. retusus M. gaigei and M. pedester as monophyletic, but not M. cartwrighti (Figures 5-2 5-3 and 5-4 ). In the 60

PAGE 61

SumTrees analysis, the clade comprised of all of the M. cartwrighti sequences (including the Fenholloway sequence) was recove red in 23% of the molecular clock trees and 25% of the biogeographical calibration trees. Pairwise distances show sma ller within-species distances compared to across-species distances. The mean pairwise distance between the Fenholloway haplotype of M. cartwrighti (I-1 and I-2) and M. cartwrighti is closer to the mean pairwise distance across Mycotrupes species than the mean within species distance ( Table 5-3 and Figure 5-5 ). Discussion Phylogenetics With the exception of the M. cartwrighti haplotype from Fenholloway (I-1 and I-2), all known species of Mycotrupes were recovered as monophyletic in both the Parsimony as well as the Bayesian analyses (Figures 5-2 5-3 and 5-4 ). The M. cartwrighti haplotype from Fenholloway (I-1 and I-2) fell out basal to the clade ( M. cartwrighti ( M. gaigei + M. pedester)), and would therefore make the species, presently defined as M. cartwrighti paraphyletic. If M. cartwrighti I-1 and I-2 are not considered part of M. cartwrighti the following support values are assigned to the known species of Mycotrupes. The Bayesian analyses gave high posterior probabilities to a ll species (0.9997-1.0). There was more variation in the bootstrap values obtained in the Parsimony analysis. Bootstrap values were high for the species M. lethroides M. retusus and M. pedester (99-100) and lower for M. cartwrighti and M. gaigei (8485). Posterior probabilities are often higher than bootstrap values for a given clade (Erixon et al. 2003). It is possible that the use of additional sequence data would result in higher bootstrap support for M. cartwrighti and M. gaigei The paraphyletic nature of M. cartwrighti suggests that M. cartwrighti I-1 and I-2 may be a cryptic species. The recovery, in both th e Parsimony and Bayesian analyses, of all other 61

PAGE 62

sequences as monophyletic clades conf orming to the present concept of Mycotrupes species boundaries, further supports that th e Fenholloway haplotype represents a distinct species. Olson and Hubbell (1954) did not examine specimens from Fenholloway. Since these specimens appear similar to M. cartwrighti and were collected from an area close to known populations of M. cartwrighti they would likely have been considered to be M. cartwrighti by most taxonomists. I examined two male Mycotrupes from Fenholloway, and compared them to male M. cartwrighti that were collected from Tall Timbers Research Station (Florida) and several localities in Georgia. Specifica lly, I examined the head, mouthpart s (as this required dissection, I compared one specimen of Mycotrupes from Fenhollway to two male M. cartwrighti from Tall Timbers), thorax, and male genitalia. I was not able to find consistent morphological differences between the Fenholloway Mycotrupes and the M. cartwrighti from other locations. Further study may result in the discovery of such differences, however. Recovery of traditionally circumscribed sp ecies as non-monophyletic is actually quite common in molecular phylogenetic analyses Crisp and Chandler (1996) argued that "paraspecies" (a term which they use to refer to polyor paraphyl etic species) are expected to occur with phylogenetic evolution. Crisp a nd Chandler expect th ese paraspecies would eventually become monophyletic through the fixation of mutati ons, though a certain amount of time would elapse before this occurred. Funk and Omland (2003), who reviewed the phenomenon of polyphyletic animal species and disc ussed the possible causes for such patterns, found 23% of species to be polyphyletic in the studies they sampled. Funk and Omland recognized that sampling protocol often makes it effectively impossible for workers to recognize polyphyletic species because phylogeneticists of ten use only one sequence from each species 62

PAGE 63

under study, whereas population biologists sample many sequences within only one species. The result in either case is that monophyly is effec tively not tested. With a greater degree of sampling within "species," polyphyly may be fo und to be more common than is currently recognized. There are a variety of possible expl anations for apparent polyphyly of M. cartwrighti (if the Fenholloway haplotype is considered to be part of this species). First, it must be remembered that these phylogenies are gene tr ees rather than species trees, a nd they represent the evolution of a particular fragment of the Cytochrome Oxidas e I gene. Ideally, the gene tree also should reflect the evolution of the speci es. If this were the case in this study, we are faced with the paraphyly of a traditional species. Strict adhe rance to a species con cept that species are monophyletic requires one of two actions to be u ndertaken. The first would be to combine the entire clade subtended by the Fenholloway M. cartwrighti haplotype into one species, resulting in lumping M. cartwrighti M. gaigei and M. pedester into one species. The other would be to recognize the Fenholloway M. cartwrighti specimens as a distinct species. Considering the strong phylogenetic and morphologica l support for the distinctness of M. cartwrighti M. gaigei and M. pedester it would seem preferable to keep th ese species separate and recognize the Fenholloway specimens as a new and distinct species. This later action would also preserve the greatest amount of taxonomic stability, an impor tant consideration when making nomenclatural changes. There also remains the possibility that the gene tree does not reflect the species tree. One plausible explanation for this s ituation could be that there is insufficient signal in the gene sequence used in the analysis (Funk and Omland 2003) For this reason, it is always preferable to use a larger data set, and if possible, multiple genes. Funk and Omland suggested the use of 63

PAGE 64

bootstrap support as a measure of confidence in the polyphyly of a species; they used the largest bootstrap value that grouped any haplotype of th e apparently polyphyletic species with one or more haplotypes of another specie s to the phylogenetic exclusion of one or more haplotypes of the apparently polyphyletic species. The clade that is composed of M. cartwrighti (excluding the Fenholloway haplotype), M. gaigei and M. pedester is supported by a bootstrap value of 56%. A collapse of this clade, however, would result in a tritomy compos ed of the Fenholloway haplotype, the M. cartwrighti clade, and the clade composed of M. gaigei and M. pedester. The Fenholloway haplotype would remain distinct from M. cartwrighti and so the low bootstrap support may not indicate a low measure of support for the distinctness of the Fenholloway haplotype. The collecting and sequencing of additional Mycotrupes material from Fenholloway may provide additional sequence data and greate r haplotype diversity with which to more strongly assess the distinctness of this popula tion with phylogenetic methods. With such additional data, support for the reciprocal monophyly of the possible cryptic species and M. cartwrighti could be calculated. There also remain additional possibilities as to why a gene tree might not represent the true species tree, such as mitochondrial introgress ion and incomplete line age sorting. Both of these situations might be recogn ized through the addition of sequenc e data from a different gene, which would make possible the recovery of a different phylogenetic pattern as a result. Once again, a larger data set would give greater confidence in the results. The phylogenetic hypotheses of Mycotrupes depicted in Figures 5-2 5-3 and 5-4 refute the hypothesis of Hubbell (1954), which was ba sed on his assumptions on morphological evolution and biogeography in the genus Mycotrupes The results of the current study are preferred over the hypothesis of Hubbell, because it is based on a modern phylogenetic analysis 64

PAGE 65

that produces a pattern of relationship based on the contribution of all data points (481 base pairs of COI) in the analysis. Cons equently, this analysis does no t support the hypothesis of Hubbell that M. gaigei is a "primitive" Mycotrupes and that M. cartwrighti and M. pedester are sister taxa. Pairwise Distances The mean pairwise uncorrected p-distances within species of Mycotrupes ranged from 0.8 (in M. lethroides ) to 4.1% (in M. gaigei ). The mean intraspecific divergence values for M. cartwrighti, M. retusus and M. gaigei are all higher than the mean divergences between conspecifics of taxa studied by Hebert et al. (2003). High intraspecific divergences are often observed across geographically isolated populations (Hebert et al. 2003). The exclusion of the anomalous Fenholloway haplotype from M. cartwrighti reduces the mean intrapsecific divergence of that species from 4.6 to 3.5%. The mean pairwise divergence across Mycotrupes species was 9%. This corresponds to a difference of approximately 43 base pairs in the COI fragment used he re. This level of interspecific divergence was atta ined by the majority of the c ongeneric species pairs studied by Hebert et al. (2003). Nucleotide divergence above a certain thre shold is often considered to indicate a species-level difference. However, rather than blindly applying a pre-determined "cutoff" value based on molecular divergence found in other studies, it is important to consider the divergence values calculated here within the context of Mycotrupes. Because Mycotrupes are flightless, and its populations are probably is olated geographically, we could expect higher molecular divergence values within species, compared to a taxon with stronger dispersal capabilities that occurs in more contiguous habitat. The mean pairwise molecular divergence across Mycotrupes species is 9%, which is twice the value for the greatest observed mean intraspecific divergence (4.1%, in M. gaigei ). The mean pairwise divergence between the M. 65

PAGE 66

cartwrighti haplotype from Fenholloway and sequences contained within the monophyletic M. cartwrighti clade is 7.8%, which is closer to a level of nucleotide di vergence seen across Mycotrupes species, and would suggest that the Fenhollo way haplotype is a di stinct species. Testing Alternative Phylogenetic Hypotheses Of the 20,001 trees produced by the two mol ecular clock BEAST runs, 4,642, or 23%, had the monophyletic clade composed of all M. cartwrighti sequences (including the Fenholloway haplotype). This would s uggest that it is not possible to reject the hypothesis of M. cartwrighti (including the Fenholloway haplotype) as a monophylet ic clade, and thus, a species. It must be remembered, however, that the other 77% of the trees produced in the BEAST run did not have a monophyletic M. cartwrighti The results may thus be considered somewhat ambiguous. Species Delimitation Two lines of evidence, th e phylogenetic topology and the hi gh nucleotide divergence, suggest that the specimens from Fenholloway represent a cryptic sp ecies distinct from M. cartwrighti. The phylogenetic support a ppears to be somewhat wea k, however. The choice of species concept can dramatically alter the interpretation of a phylogene tic test of species boundaries, including this one. Wheeler and Plat nick (2000), who (in their own words) provided a species concept that gives "...the finest level of resolution of kinds of organisms that can be justified on the basis of constantly distributed, observable attributes," state that phylogenetic species "...are the smallest groups of organi sms among which historical patterns of common ancestry may potentially be retrieved and which may not be divided into smaller units with similar properties." Alth ough all of the species of Mycotrupes, with the exception of the Fenholloway haplotype, are supported as monophylet ic, there appears to be some degree of geographical structure below the species level. 66

PAGE 67

Recommendation Based on molecular evidence, there may be a cryptic species in what has been considered M. cartwrighti To more rigorously test species boundaries, a larger series of specimens from Fenholloway is needed to obtain more sequence data and further assess morphological variation in the population. If analyses using additio nal data support the Fenholloway population as a separate clade, there would be st ronger evidence for a cryptic speci es. The 481 base pair data set used in this study is small by modern standards of molecular systematics, and the collection of additional nucleotide data would be a valuable contribution to the phyl ogenetic analysis of Mycotrupes. At the very least, sequencing one additional gene across taxa, possibly a nuclear gene, would increase the amount of data a nd allow a comparison be tween the phylogenetic "signals." The presently weak support, in terms of the number of Bayesian trees showing the non-monophyly of M. cartwrighti could change in either direc tion with the addition of more sequence data. A larger series of specimens from Fenhollo way would provide bett er insight into the morphological variation present in that population, which could provi de increased confidence in any discovered morphological differences between Fenholloway and the remaining M. cartwrighti populations. 67

PAGE 68

Table 5-1. Mycotrupes collecting localities of specimens used in molecular analysis. Species Code Specimens State County Location Latitude, Longitude M. lethroides A 2 GA Burke Yuchi Wildlife Management Area N33 05.017' W81 46.535' M. retusus A 2 SC Lexington Near Gaston N33 49.406' W81 11.853' B 1 SC Richland Near Sesquicentennial State Park N34 06.120' W80 54.672' C 1 SC Aiken Hwy 78 and Oak Club Road N33 30.678' W81 33.539' D 2 SC Aiken Webb Pond Road N33 27.672' W81 25.987' M. cartwrighti A 1 FL Leon Eleanor Klapp-Phipps Park N30 32.294' W84 17.340' B 1 FL Leon Eleanor Klapp-Phipps Park N30 32.310' W84 17.363' C 2 FL Jefferson Avalon conservati on easement N30 23.660' W83 53.763' E 1 FL Leon Tall Timbers Research Station N30 40.279' W84 14.154' F 1 FL Leon Tall Timbers Research Station N30 40.317' W84 14.175' H 2 GA Thomas Thomasville N30 49.693' W84 00.690' (cryptic species?) I 2 FL Taylor Fenholloway N30 04.974' W83 30.476' J 2 GA Liberty Hinesville N31 52.479' W81 34.452' L 1 FL Liberty Torreya State Park N30 33.536' W84 57.016' M. gaigei A 2 FL Lafayette Mayo N30 03.706' W83 11.427' C 2 FL Suwannee Hildreth N29 59.626' W82 48.633' D 2 FL Madison State Road 53, N of County Line N30 16.459' W83 17.320' E 1 FL Marion Summerfield N29 00.618' W82 08.058' F 1 FL Marion Marion Oaks N29 01.382'; W82 14.770' M 2 FL Lafayette W. of Mayo N30 08.964' W83 19.719' N 2 FL Seminole Geneva N28 44.754' W081 07.801' O 2 FL Gilchrist W. of Newberry N29 37.879' W082 42.313' P 2 FL Columbia O'Leno State Park N29 55.001' W82 35.057' M. pedester A 2 FL Lee Estero N26 28.472' W81 50.165' B 1 FL Lee Babcock Ranch N26 45.665' W81 40.848' C 1 FL Lee Babcock Ranch N26 45.578' W81 40.869' 68

PAGE 69

Figure 5-1. Map of collec ting locations (colored circles) of sequenced Mycotrupes specimens. Species color code: Red= M. retusus, green=M. lethroides yellow=M. cartwrighti maroon= M. gaigei blue= M. pedester Table 5-2. Settings fo r BEAUTi .xml files. Pre-set rate Biogeographical calibration Substitution model GTR GTR Base frequencies Estimated Estimated Site heterogeneity model I+G I+G Gamma categories 4 4 Partition into codon positions? No No Fix mean substitution rate? Yes (0.0115) No Molecular clock model Stri ct clock Strict clock 69

PAGE 70

Figure 5-2. The single most pars imonious tree (L = 319 steps, CI = 0.56, RI = 0.85) recovered in the Parsimony analysis. Bootstrap values (>70%) are reported below the branches. Note that M. cartwrighti specimens and the possible cryptic species from Fenholloway (" M. cartwrighti I-1 and I-2) are highlighted in red. 70

PAGE 71

3.0 my M. retusus D-2 M. cartwrighti C-2 M. cartwrighti H-2 M. cartwrighti I-1 M. cartwrighti C-1 M. cartwrighti H-1 M. cartwrighti E-1 M. cartwrighti J-1 M. cartwrighti F-1 M. cartwrighti J-2 M. cartwrighti B-1 M. cartwrighti I-2 M. cartwrighti A-1 M. cartwrighti L-1 [5 [4.7 0.6337 0.8473 0.8128 M. lethroides A-1 M. gaigei C-1 M. lethroides A-2 M. gaigei O-1 M. pedester A-2 M. pedester A-1 M. gaigei D-2M. gaigei O-2M. gaigei P-2 M. pedester C-1M. gaigei A-1M. retusus C-1 M. gaigeiD-1M. gaigei M-1 M. pedester B-1 M. gaigei F-1M. retusus D-1M. gaigei E-1 M. retusus A-1 M. retusus B-1 M. retusus A-2 M. gaigei M-2 M. gaigei C-2 P. youngi M. gaigei A-2 M. gaigei N-2 M. gaigei P-1M. gaigei N-1[6.3506,13.5979] [11.9565,28.3933] .5714,11.3205] [8.8446,20.7558] 613,9.7168] [15.5404,42.4986] 0.9965 1 1 0.9997 1 1 11 Figure 5-3. BEAST maximum clad e credibility tree, pre-set s ubstitution rate. 95% HPD (in millions of years ago) are in brackets; Posterior probabilities of clades are on branches below clades (red=<0.95). Note that M. cartwrighti specimens and the possible cryptic species from Fenholloway (" M. cartwrighti I-1 and I-2) are highlighted in red. 71

PAGE 72

2.0 my M. retusus D-2 M. gaigei D-2 M. cartwrighti H-1 M. cartwrighti B-1 M. cartwrighti L-1 M. cartwrighti I-1 M. cartwrighti C-1 M. cartwrighti H-2 M. carwrighti A-1 M. cartwright C-2 M. cartwrighti E-1 M. cartwrighti I-2 M. cartwrighti J-2 M. cartwrighti J-1 M. cartwrighti F-1[4. 0.8578 0.6748 M. lethroides A-2 M. retusus B-1 M. gaigei O-2 M. lethroides A-1 M. pedester A-1 M. pedester A-2 M. gaigei C-1 M. gaigei P-2 M. gaigei F-1 M. gaigei M-1 M. gaigei D-1 M. gaigei P-1 M. pedester B-1 M. pedester C-1 M. gaigei C-2 M. retusus D-1 M. gaigei E-1 M. gaigei O-1 M. gaigei A-2 M. retusus C-1 M. retusus A-1 M. gaigei N-1 M. gaigei N-2 M. gaigei M-2 M. retusus A-2 P. youngi M. gaigei A-1 [4.7751,6.6807] 8021,5.1948] [7.928,16.8077] [5.7551,10.8354] [6.9656,13.7816][4.8032,7.6114] 0.5245 1 1 1 1 1 1 0.9997 1 0.9956 Figure 5-4. BEAST maximum clade credibility tree, biogeographical calibration. 95% HPD (in millions of years ago) are in brackets; Posterior probabilities of clades are on branches below clades (red=<0.95). Note that M. cartwrighti specimens and the possible cryptic species from Fenholloway (" M. cartwrighti I-1 and I-2) are highlighted in red. Table 5-3. Mean pairwise distances in Mycotrupes. P-distances (uncorrected) Within Species: M. lethroides 0.0083 M. retusus 0.0270 M. pedester 0.0076 M. cartwrighti (incl. Fenholloway) 0.0459 M. cartwrighti (excl. Fenholloway) 0.0345 M. gaigei 0.0406 Across species: 0.0904 Fenholloway (site I) and all other M. cartwrighti 0.0779 72

PAGE 73

Figure 5-5. Mean pairwise distances in Mycotrupes. Mean across Mycotrupes species, mean within Mycotrupes species and pairwise distance between M. cartwrighti I-1, I-2 and other M. cartwrighti 73

PAGE 74

CHAPTER 6 BAYESIAN PHYLOGENETIC INFERE NCE AND BIOGEOGRAPHY OF MYCOTRUPES Introduction The evolutionary hypothesis of Mycotrupes by Hubbell (1954) ( Figure 2-1 ) was based on assumptions of morphological evol ution in the genus, and it was inextricably linked with his biogeographical hypothesis as well. Hubbell proposed that speciation in Mycotrupes was coincident with, and the result of, the Quaternary fluctuations in sea level that affected the southeastern Atlantic Coastal Plain. Although historic changes in sea level, largely attributed to the growth and melting of continental ice sheets, are taken for granted now, it was not always the case (Miller et al. 2005). Charles Schuchert (1910), the great palaeogeographer, was one of the early proponents of the concept of eustatic changes in sea leve l. Schuchert developed maps showing various inundations of North America by th e sea from the Cambrian to the Pleistocene. His map of the Eocene shows inundation to the poin t of the Fall Line of the southeastern United States. Biogeography is, simply stated, the study of how organisms how come to be where they are today. Neill (1957) studied the biogeography of Florida. He focused on patterns of distribution common to multiple taxa and speculated on the mechanisms that may have been responsible for these distributions. None of the patterns he descri bed appear to have any special similarity to distributions of Mycotrupes species. In attempting to explain some of these patterns, Neill proposed that the broad embayments along rivers that woul d have resulted from flooding in Pleistocene interglaci als may have been a more important cause of the present-day distributional patterns than the ri vers as they are today. Thus, the Savannah River may, at a past time with much higher sea level, have been an even broader (and more significant) barrier between M. lethroides and M. retusus In this way, even if the Savannah River has existed since 74

PAGE 75

before the ancestor of Mycotrupes, such an ancestor could have b een distributed on both sides of the river, and isolation (followed by speciation) may have been a resu lt of a rise in sea level and the consequent formation of a broad embaymen t along the river's course Also of possible relevance to this study of Mycotrupes Neill proposed that "a number of distinctive west coast organisms may once have ranged over an ex tensive territory now largely submerged." Very little work appears to have been done on insect biogeography in the southeastern United States. Lamb and Justice (2005) studi ed insect phylogeography across ridges of scrub habitat in Florida. This work was at the popul ation (below species) level. Lamb and Justice found that genetic diversity was partitioned primarily by North-South tren ding ridges, and they presented evidence that the olde r ridges (those more inland, su ch as the Lake Wales Ridge) provided a source of colonists for younge r ridges closer to the coast. Marine terraces are a prominent feature of the Atlantic Coasta l Plain of the United States, and they represent prolonged high sea stands. It is generally ag reed that the higher terraces are older than the lower ones, and that during fluctuations of sea level, the high stands (interglacials) have been progressively lower in elevati on over time until the present (Alt and Brooks 1965; Colquhoun et al. 1991). While marine terraces have been mapped in Florida and Georgia (Healy 1975; MacNeil 1949), such work has not b een completed for South Carolina. Not long after the work of Hubbell (1954), cladistic phylogenetic methods became available that would have enab led inference for the sequence of speciation in the genus, if significant morphological charact er data had been availabl e. However, to study the biogeography of Mycotrupes in relation to particular events su ch as changes in sea level, it is important to not only resolve the phylogenetic rela tionships within the genus, but to obtain an 75

PAGE 76

estimate of the divergence times as well. At pr esent, the estimation of divergence times requires the use of molecular sequence data. To impose a time scale on a phylogeny based on nucleotide change, it is necessary to assume a molecular clock. The molecular cloc k, proposed by Zuckerkandl and Pauling (1965), has been applied to a wide vari ety of evolutionary questions. The central assumption is that nucleotide base changes are accumulated at a more-or-less regular rate, and as a result, the amount of divergence between two sequences can be converted to a time scale. Such a tool has obvious applications to biogeograp hy, especially when fossil da ta are lacking. Two species separated by an obvious barrier may have a di vergence estimate congruen t with the estimated age of the barrier (for example, a river or mountain range). Such a temporal congruence would provide support for the hypothesis that the formati on of the barrier resulted in a vicariant event that separated the two taxa under study. Alternatively, a divergence date later than the age of the barrier could suggest a dispersal event that occu rred subsequent to the ag e of the barrier. Some concerns, regarding application of the molecular clock to biogeography, include the potentially unrealistic assumpti on of rate homogeneity across lineages and time, and wide confidence intervals, which make it difficult to correlate a diverg ence to a particular biogeographic event (furthermore, such events may themselves have wide confidence intervals). These issues and others are discussed in Arbogast et al. (2002) and Ho (2007). Another, potentially more problematic, issu e with the molecular clock is the clock calibration method. Use of dateable fossils may be the least assumption-laden method of calibrating a molecular clock. Unfortunately, there are no known fossils of Mycotrupes. In this study, two other methods were used to calibrate th e molecular clock and apply a time scale to the phylogenetic hypothesis: a pre-set nucleotide subs titution rate and biogeographical calibration. 76

PAGE 77

When applying a molecular clock with a pre-set nucleotide substitution rate to sequence data, the researcher imports a rate of nucleotide substitution obtained from a taxonomic group that is more or less related to those being studied. Brower's (1994) 2.3% pairwise divergence per million years is probably the most widely used molecular clock rate in insect studies. Brower calculated this rate from a study of exemplar taxa of four in sect orders (Cole optera, Diptera, Hemiptera, and Orthoptera) and one species of decapod crustacean ( Alpheus sp.); calibration was accomplished through assumed dates of divergence, including geological and paleoclimatological events. Estimated times of biogeographical events ha ve also been used to calibrate molecular clocks (Shoo et al. 2008; Weir and Schluter 2008 ; Zhang et al. 2008). Renner (2005) noted that "...constraining nodes in a phyloge netic tree by geological even ts risks circularity in biogeographic analyses because it already assume s that those events caused the divergence, rather than testing temporal congruence." While this is indisputable, it might be argued that some information can be gained through this me thod by testing for geological events other than the one(s) used for calibration. By assuming that nucleotide changes are more-or-less clock like, but not wishing to assume a certain value for the rate of change (which has often been calibrated in other taxonomic groups), we can examine th e relationship between speciation events and geologic events, calibrating the analysis with the "safest" biogeographical event. The program Bayesian Inference Samp ling Trees (BEAST) (Drummond and Rambaut 2006) implements Bayesian inference in order to sample tree space, allo wing the user to impose a great variety of molecular models. Especially si gnificant is the users ability to apply "priors" to the model. Priors are any constraints that the user wishes to impose on the evolutionary model, and can include dated ca libration points such as fossils and biogeographical events. 77

PAGE 78

Because divergence times can be calculated within trees, BEAST has seen much recent application to questions of bi ogeography and coevolution (Mathe ny et al. 2009; Light and Reed 2009). The goal of this study was to use the phy logenetic hypotheses an d divergence date estimates for Mycotrupes produced with COI data (see Chapter 5), attempting to link divergences within Mycotrupes with biogeographic events. The two BEAST analyses used different models of nucleotide substitution rate s (pre-set substitution rate and biogeographical calibration) under a strict molecu lar clock. The biogeographical im plications of these new data are discussed. Materials and Methods I used the maximum clade credibility trees resulting from the BEAST analyses for the biogeographical study. One of these BEAST analys es was done with a pre-set molecular clock rate (Figure 5-3 ); the other was done with a bi ogeographical calibration (Figure 5-4 ). Further details on the BEAST analyses can be found in Chapter 5. Results Nucleotide Substitution Rates a nd Divergence Time Estimates The nucleotide substitution rate for the preset rate analysis was specified at 1.15% per million years. The mean rate obtained in the biogeographical calibration runs was 0.058 (5.8%), with a standard devia tion of 0.00013 (0.013%). The divergence times (in terms of range of 95% highest posterior de nsity) indicated by the resulting topologies are given in Table 6-1 The two methods produced different confidence intervals for the divergence times. 78

PAGE 79

Discussion Tree Topologies There were few differences in BEAST tr ee topology between the pre-set rate and biogeographical calibration analyses. The mean rate obtained in the biogeographical calibration analysis was 5.8% per million years, which is much higher than the rate used for the pre-set runs (1.15%). This means that the assumption of the maximum age constraint imposed in the biogeographical calibration on the divergence of the M. gaigei + M. pedester clade at 5 mya is inconsistent with an assumed nuc leotide substitution rate of 1.15%. The use of a divergence date younger than 5 mya would have resulted in an even more rapid substitution rate Species-Level Biogeography in Mycotrupes Comparison of divergence times of two calibration methods The biogeographical calibration anal ysis assumed that the ancestor of M. gaigei and M. pedester could not have diverged from M. cartwrighti earlier than 5 mya, and this assumption was associated with a nucleotide substitution ra te of around 5.8%. If this divergence in fact occurred later than 5 mya, this would require an even higher rate of substitution. This rate (5.8%) was much higher than the pre-set rate used in our companion analysis, which then produced an earlier divergence time for M. gaigei and M. pedester from M. cartwrighti (excl. Fenholloway haplotype) between 5.6-11.3 mya. The 95% highest probability distribution (H PD) values of divergence times were broad ( Table 6-1 ). Such broad confidence intervals make it very difficult to correlate estimated divergence times with events such as a specific shift in sea level, because these changes have occurred frequently over the past 10 million years. In one of the more recent studies, Miller et al. (2005: Figure 4) show hundreds of shifts in sea level over this time. Although a limited number of marine terraces have been mapped in the southeastern Atlantic Coastal Plain, they 79

PAGE 80

represent only the rela tively prolonged high stands. The actua l history of sea level was more complex, and an inundation of brief duration woul d be sufficient to extirpate a population of Mycotrupes. Alt and Brooks (1965) state that "it is possible to find local topographic evidence somewhere in Florida for an abandoned shore line at almost any elevation." Divergence time confidence intervals of Mycotrupes could only be narrowed with the addition of more information, such as dated fossils. Dispersal may complicate taxon-area biogeography Due to their flightlessness a nd the conception of their habita t as remnants of old shorelines, it is tempting to consider the distributions of Mycotrupes as static, minimizing the role of dispersal relative to that of vicariance. The possible role of disp ersal is especially problematic when attempting to relate the evolution of Mycotrupes to changes in sea level, long thought to be the primary driving force for sp eciation within the genus. The relative importance of dispersa l in the resulting distribution of Mycotrupes species is not known, but circumstantial evid ence reveals that it may have been quite important. For example, the known coll ecting localities of M. pedester (including imprecise label localities of Arcadia and Punta Gorda) are between >1 and 17 me ters above sea level (Arcadia is the highest in elevation, at 17 meters asl) (Google 2006). The most recent high-stand in sea level, as indicated in Miller et al. ( 2005), was almost 30 meters above sea level at approximately 200,000 years ago. Such a high sea level would have co mpletely inundated the known distribution of M. pedester Olson and Hubbell (1954) hypothesized that M. pedester might occur farther north, possibly in the central ridge of Florida, but no species of Mycotrupes is known from the central ridge. Another example suggests that dispersal may have been important in M. gaigei Most of the collecting localities for this species range in elevation from 11 to 30.5 meters above sea level 80

PAGE 81

(Google 2006). The Old Town site is at a mu ch lower elevation, 7.6 meters above sea level (Google 2006). The Old Town site is near th e Suwannee River, which may have influenced drainage locally through re-working sa nd deposits. It is possible that M. gaigei was able to disperse southwards along the Suwannee River af ter the ocean last receded from a level that would have inundated this site. A nested clade analysis of M. gaigei with greatly increased sampling, might reveal such a historical pattern. Mycotrupes cartwrighti has been collected from sites that range in elevation from 10 (Hinesville, Georgia) to 80 (Thomasville, Georgi a) meters above sea level (Google 2006). This distribution cannot be convenientl y related to a given sea level. The lower, more coastal occurrences (such as Hinesville) could be a resu lt of dispersal after a r ecession of the sea from such areas. A vicariance-based biogeographical interpretation Figure 6-1 is a depiction of the hypothesized vi cariance events supported by the BEAST topologies. Backed up with the phylogeny fr om the phylogenetic analysis, the following scenario is proposed to explain the biogeogr aphy of this genus. Numbers correspond to vicariance events depicted on Figure 6-1 Mycotrupes originated on the Fall Line, or in th e vicinity, possibly some time during the Tertiary. The ancestor of ( M. retusus (cryptic Mycotrupes species from Fenholloway( M. cartwrighti( M. gaigei + M. pedester)))) was isolated from M. lethroides possibly by the Savannah River (1). At some point after the sea level ha d dropped below the Fall Line sandhills for the last time, a southward dispersal event from M. retusus gave rise to the ancestor of (cryptic Mycotrupes species from Fenholloway( M. cartwrighti ( M. gaigei + M. pedester ))) after an event such as a rise in sea level cut off this area from M. retusus (2). An event such as a rise in sea level isolated the ancestor of the cryptic Fenholloway Mycotrupes from M. cartwrighti (3). 81

PAGE 82

Mycotrupes cartwrighti was separated from the ancestor of ( M. gaigei + M. pedester) possibly in northern Florida (4). Later, M. pedester and M. gaigei were separated (5). The distributions of M. gaigei and M. pedester are separated by more than 100 miles. The present land exposed in the state of Florida is only appr oximately one half of the "Florida Plateau," much of which is shallowly submerged along the west side of the peninsula (Randazzo and Jones 1997; Vaughan 1910). A portion of this now-submerged area was exposed as recently as the last glacial maximum (Cooke 1945). Considering that M. pedester is found very close to the coast, it is tempting to consider the possibility that there may have been habitat on the Florida shelf at a time with lower sea level that would have exposed this currently submerged area that allowed dispersal of the ancestor. The site near Fenholloway, Florida, wh ere the apparent cryptic species of Mycotrupes was collected, is approximately 12 m iles from Townsend, Florida, where M. gaigei has been collected. Land along this 12 mile stretch is predominan tly poorly drained pine flatwoods. This is geographically the closest that any two species of Mycotrupes are known to occur. Because all species of Mycotrupes are currently understood to be allopatric, vicariance would appear to be the most likely cause of speci ation. It might be possi ble to identify barriers to dispersal, past or present, between species distributions th at may have played a role in vicariance. Barriers for Mycotrupes could include areas that ha ve been inundated by marine transgressions in the past, bodies of water (such as rivers), and poorly drained habitat. The general sequence of speciation, as proposed by the results of the phylogenetic study, suggests that sea level played a role in the evolution of Mycotrupes. The two most basal species, M. lethroides and M. retusus occur on the highest, most inland habitat, which would suggest that an ancestral Mycotrupes was restricted to the Fall Line sand hills, when the sea covered much of 82

PAGE 83

the lower elevations of the Atlantic Coas tal Plain. The two most derived species ( M. gaigei and M. pedester) occupy lower elevations where more recent high-stands of the sea could have caused vicariance. The species M. gaigei appears to occupy a fairly restricted range in elevation. Of the 20 collecting sites used in the niche m odeling analysis (Chapter 7), 13 are within the range of elevation listed by Co lquhoun et al. (1991) for the Penholloway Terrace (12.9-23.0 m asl). If the elevations for the Wicomico Terrace is included also (12.9-30. 3 m asl), 18 sites out of 20 fall within this range. There has been deform ation of the bedrock in Florida, which would have changed the terrace eleva tions (Harper 1921 [in referenc e to the peninsular lime-sink region]; Opdyke et al. 1984 [in reference to Florida in general]). While th e degree to which this has affected terrace elevations is unknown, the us e of elevation interval s to indicate marine terraces may be problematic. The role the Savannah River might have played in the split between M. lethroides and M. retusus is an interesting problem. The Savannah River is an obvious barrier to dispersal between populations of these two species. The age of the Savannah River is not precisely known, although rivers that drain the Appa lachian Mountains, such as the Savannah, could be as old as Early to Middle Miocene (T. Scott, pers. comm.). Such an age would suggest that it was present before much of the lower Atlantic Coastal Plain habitat of M. cartwrighti M. gaigei and M. pedester was most recently exposed. Over time, ri vers have changed their course and they presumably change their flow volume as a result of changes in climate. Thus, it may not be realistic to use the present-day Savannah River as a proxy for the Savannah River of the past with regards to its possible biogeographical importance. Under th e assumption that the Savannah River was the cause of the split between M. lethroides and the ancestor to the rest of the species, there are two possibilities to the mechanism of isol ation: 1) a chance dispersal event 83

PAGE 84

across the Savannah River, or 2) a shift in the river's course, dividing a once-continuous population. There is also the possibi lity that a rise in sea level to the level of the Fall Line could have inundated the area of the river, thus isolatin g the two populations. It is possible that climate change over the past 10 million or so years may have had a significant effect on the distribution of Mycotrupes. Apart from the obvious impact of inundation with rising sea levels changes in water table, ve getation, and weather may have resulted in Mycotrupes populations moving, making it diffi cult to study their biogeography. Such changes are undoubtedly oc curring at the present time, and the distributions of Mycotrupes species are probably changing in response. Comparison with studies of other taxa in the same region Scrub and sandhill are two major vegetation assemblages that occur on well-drained uplands in the southeastern Atlantic Coastal Plai n. Sandhill (which is wide spread in the Atlantic Coastal Plain in the southeastern United States ) and scrub (restricted to Florida) are often confused, but are characterized by distinctly different plant assemblages (Laessle 1958; Myers 1985). Mycotrupes is often collected in habitat that c ould be characterized as sandhill, but the genus is not known to occur in scrub. Mycotrupes is also collected in habitat with vegetation that could not be referred to as sandhill, including grassy fields although this may have been modified by humans from previous sandhill habitat. The reasons for the apparent preference for sandhill are not known, but they may include fact ors such as a more open understory in sandhill or differences in soil type. Sand ridges in Florida, which generally tr end North-South, have been named and mapped (Brooks 1981). Although some Mycotrupes habitat may be associated with such ridges (for example Brooksville Ridge in Newberry and Arch er), sandy soil appears to be more important 84

PAGE 85

than a well-defined ridge, and other localities that host Mycotrupes are not part of an obvious ridge system. A significant amount of work has been done on the population genetics of animals inhabiting Florida scrub. Lamb and Justice (2005) studied in sect phylogeography across ridges of scrub habitat in Florida. All of this wo rk, however, was at the popul ation (below species) level, and they did not estimate tim es of divergence. In general, they found the ge netic diversity to be partitioned primarily by North-South trendi ng ridges, and they pres ented evidence that the older ridges (those more inland, such as the La ke Wales Ridge) provided a source of colonists for younger ridges closer to the coast. These resu lts would somewhat agree with those of this study, as the more derived Mycotrupes species occur, in genera l, at lower elevations. Future Research Little work has been done using modern phylogenetic tools to study the species-level biogeography of southeastern insects. More attention has been given to the molecular phylogeography (below the species level) of Flor ida scrub animals, including insects (e.g. Lamb and Justice 2005). This work on Mycotrupes will provide an interest ing comparison to future work on species-level biogeography in the sout heastern United States. One remaining significant barrier to such work that remains is th e lack of clear-cut, dateable vicariance events. It should still be possible in a historical bi ogeographic framework to compare the speciation patterns of different groups to support or reject common speciation mechanisms in this region. The study of plants that occur in similar situations may offe r some information that could aid in understanding the biogeography of Mycotrupes. Ceratiola ericoides Michaux, or Scrub Rosemary, a common shrub in sandhill and scrub habitat in the Atlantic Coastal Plain of the southeastern United States, is thought to have expanded its distribu tion during the glacial maxima (Trapnell et al. 2007). There is evidence that much drier conditions in the past, along 85

PAGE 86

with a lower water table, favored xeri c vegetation in Flor ida (Watts 1975). Mycotrupes may have expanded in distribution during these dry periods, and what today appear to be isolated populations may have then been more extensiv e with a higher degree of connection. It is likely that further Mycotrupes character data will be in the form of nucleotide sequences. As I used only a fragment of a singl e gene (481 base pairs of Cytochrome Oxidase I) in this study, additional sequence data may resu lt in significantly di fferent topologies and a different biogeographic hypothesis for Mycotrupes. 86

PAGE 87

Table 6-1. Selected divergence tim e estimates with 95% HPD. Divergence Pre-Set Rate Biogeographical Calibration PeltotrupesMycotrupes 15.5-42.5 mya 7.9-16.8 mya M. lethroides -(rest of Mycotrupes ) 12.0-28.4 mya 7.0-13.7 mya M. retusus -(rest of Mycotrupes ) 8.9-20.8 mya 5.8-10.8 mya M. gaigei M. pedester 4.8-9.7 mya 4.8-5.2 mya Figure 6-1. Collecting locations of Mycotrupes specimens used in this study with numbered (15) vicariance events. Green = M. lethroides red = M. retusus yellow = M. cartwrighti, pink = Mycotrupes from Fenholloway, maroon = M. gaigei blue = M. pedester 87

PAGE 88

CHAPTER 7 EMPLOYING ECOLOGICAL NICHE MODELING TO PREDICT SPECIES DISTRIBUTIONS IN MYCOTRUPES Introduction Because of the secretive habitats of the adults, Mycotrupes are rarely collected except when they are specifically targeted with baited pitfall traps. Hen ce, it is possible that significant portions of their distribution remain unknown. The only way to know with certainty if a species occurs in a given area is to colle ct a specimen there. However, it isn't practical or efficient to collect across a large geographic area. In addition, it would be beneficial to know why a species occurs where it does, i.e. what its habitat requir ements are. Such information can be used to target surveys for the species, guide conservation efforts, and it can help us understand the biogeography of the group in question. Niche m odeling can provide possi ble answers to these questions. The goal of niche modeling is to develop a m odel of a species' distribution, based on areas where the species is known to occur (and sometim es also incorporating areas where the species is known not to occur), using environmental data. An inherent assumption is that the biology of the organism is related to its environment, and that environmen tal data can then be used to predict the geographic distribution of the species. This is, of c ourse, only possible if biologically informative data are used in the analysis. The Geotrupidae have received little attention with regard to niche modeling. Lobo et al. (2006) studied the distribution of two speci es of the flightless geotrupid genus Jekelius in the Iberian Peninsula, by using logist ic regression. They found that bo th climate and substrate were important factors in the distributi on of these species. Barriers to dispersal (in this case, rivers) appeared to be important in further li miting the distributions of the species. 88

PAGE 89

Maximum Entropy (MaxEnt) has become a popular program for predicting species distributions. MaxEnt exhibits superior performance, comp ared to other niche modeling programs, in terms of area under receiver operati ng characteristic curv es (AUC) and omission rates (Elith et al. 2006; Phillips et al. 2006). In addition, MaxEnt appears to perform well even at small sample sizes of presence data (Hernandez et al. 2006). MaxEnt is a maximum likelihood algorithm that attempts to find the model with the highest entropy, in other words, the model explaining the distribution of a sp ecies with the fewest constrai nts. Two types of data are necessary for MaxEnt: 1) environmental layers and 2) presence data (i.e., points of known species occurrence). All environmental layers must be in raster form with identical dimensions and coverage, because MaxEnt uses a grid in which each cell is assigned a certain value for each environmental layer, as well as a presence or ab sence status for the species under study. MaxEnt does not require that known absence data points be input; a pseudo-absence method is used to approximate absence data from background pixe ls chosen at random. The MaxEnt output includes a map with a likelihood of occurrence valu e given for each grid cell. The reader is referred to Phillips et al. (2006) for a more in-depth discussion of MaxEnt. I used MaxEnt to generate predicted distributions of the five recognized species of Mycotrupes by using several types of environmental data. The Fenholloway site was excluded because of the evidence supporting a cryptic Mycotrupes species at that site, and because only one locality is known (see Chapter 5). There were four goals of this study: The first was to improve the knowledge of the distributions of Mycotrupes species. The second was to observe the effects of using different type s of environmental data (climate and soil data). Because of the burrowing habits of Mycotrupes soil was expected to be crit ical in describing the niche of Mycotrupes. Elevation was included as a layer, because changes in sea level are thought to have 89

PAGE 90

been important in speciation in Mycotrupes. The third goal was to examine how the niches suggested by MaxEnt relate to the dist ributions and habitat of each species of Mycotrupes especially the habitats as I observed and as described in the liter ature. The fourth goal was to consider how biogeography might explain the comparisons of th e likelihood of occurrence map with the known distributions. Materials and Methods Location Data Mycotrupes species presence data points were made up mostly by recent (2006-2008) field collections by the author and other individuals. A Garmin GPSMAP 60CSx (Garmin International, Inc., Olathe, Kansas) hand-held GPS unit allowed precise geographic coordinate information to be logged at collecting sites. Literature records guided much of the recent collecting. The amount of available presence data was distributed unequally ac ross the five species ( Table 7-1 ); this is a result of the restricted geogra phic distributions of the species, and regional differences in collecting effort. The reliability of a model produced from few sites is expected to be poor, thus in the case of M. lethroides (which had only three, clos ely-spaced collecting sites), several literature records were used which were precise enough to obtai n coordinates from the program GoogleEarth (Google 2006). Only th ree collecting sites were available for M. pedester and the available literature records were not pr ecise enough to use as presence data. The total numbers of presence sites used, per species, were as follows: M. pedester three sites; M. retusus four sites; M. lethroides, six sites; M. cartwrighti nine sites; M. gaigei 20 sites. Environmental Layers The following databases of environmental laye rs were obtained for this study: Elevation data, Bioclim data, and U.S. soil maps. A total of 27 layers ( Table 7-2 ) were obtained from these 90

PAGE 91

databases. Geographic data cove rage included the majority of Florida, Georgia, and South Carolina, as well as a portion of North Carolina. All layers were formatted as a raster grid (with uniform dimensions) with a 0.001 degree cell si ze, which is approximately 111 meters. Elevation data were downloaded from the National Map Seamless Server (USGS 2008). Bioclim layers, described as "biologically meani ngful variables" derived from temperature and precipitation values, were downloaded from th e WorldClim web site (WorldClim 2006). Soil data from the U.S. General Soil Maps were downloaded from the USDA Natural Resources Conservation Service web site (USDA 2008). Maximum Entropy MaxEnt (Version 3.1.0) (Schapire 2008) runs were done with the following settings: 50% of sites used for testing; rando m seed selected; 500 iterations (for M. cartwrighti 10,000 iterations were used because 500 iterations were not sufficient to converg e); jacknife variable analysis selected. A convergence thres hold of 0.00001 was used. For each species of Mycotrupes, two separate MaxEnt runs were conducted; one with all of the environmental layers included, and one excluding the BioClim data to te st the effect of clim ate data on the model. Results The algorithm converged for all species, resul ting in the maps of lik elihood of occurrence depicted in Figures 7-1 7-2 7-3 7-4 and 7-5 Maps are given for both with and without the Bioclim layers, as there may be reason to expect climate to bias the models (see below). Species-specific model characteristics are given in Table 7-2 The "most important" layers in Table 7-2 are those layers that contri buted at least 5% to the m odel. The area under the curve (AUC) is the area under a Receiver Operating Ch aracteristic curve depi cting the relationship between sensitivity (y-axis) and (1-specificity) (x-a xis) for all thresholds. Higher values for the AUC, which indicate that the mode l was sensitive enough to include all or most of the presence 91

PAGE 92

sites, while still being specific enough that a minimum of geographi c area is predicted as being of high likelihood of occurrence, ar e considered to indicate a bette r model fit. The P-value and threshold value are those for the likelihood of o ccurrence threshold that maximizes the sum of sensitivity+specificity. The P-values obtained for this threshold we re significant for all analyses (<0.05). Discussion Areas Predicted by MaxEnt Mycotrupes lethroides The area of high likelihood of occurrence for M. lethroides ( Figure 7-1 ) is focused in two major areas. The first area extends South, Southeas t and to a lesser extent Southwest of Augusta, Georgia, and includes the presently known distribution of this species. The other area straddles part of the Oconee River south of Dublin, Georgia. When Bioclim layers are omitted from the analysis, the likelihood map for M. lethroides becomes the broadest obtained in any of the analyses. Most of the Atlantic Coastal Plain region (below the Fall Line) is assigned a likelihood of occurrence of at least 0.5. The AUC is relatively low (0.930), which suggest s that there was less informati on in the presence data to build a model. Mycotrupes retusus The map of likelihood for M. retusus ( Figure 7-2 ) suggests that this species is restricted to the Fall Line sandhill region. There is an ar ea of high likelihood of occurrence northeast of Aiken, S.C. This map also shows a relatively broad area of unsuitable habitat corresponding to the upper Congaree River, which suggests that this area could act as a barrier to gene flow. The Fall Line appears to drop off in likelihood abruptly northeast of Columbia, S.C. This agrees with the known distribution of M. retusus which only extends approximately 10 miles northeast of 92

PAGE 93

Columbia to Blaney, in Kersha w County. Another large area that was assigned somewhat high likelihood is the vicinity of Ma rlboro County, SC, which lies outsid e of the known distribution of M. retusus and would be an interesting area to sample in the future. When the Bioclim layers are excluded, the map of likelihood is somewhat different. A large area in Florida, within part of the distribution for M. gaigei is assigned high likelihood of occurrence for M. retusus. Scattered patches of high likelihoo d extend along the Fall Line across Georgia. Within South Carolin a, the pattern does not change much, although there is now a small area of much higher likelihood (0.9) in the vicinity of Cheste rfield County, SC. Mycotrupes cartwrighti Several areas were assigned a high likelihood of occurrence for M. cartwrighti ( Figure 73 ). One is roughly coincident with the Tallahassee red hills region, and is well represented by collecting records, except for the area west of Torreya State Park (Harpe r 1914). Another area of high likelihood for M. cartwrighti is located in North-Central Fl orida. The Hinesville, Georgia locality was not assigned high lik elihood, although there is an ar ea with moderate likelihood of occurrence located NW of Hinesville. The known localities (not represented by collections in this study) in the vicinity of Americus and Vienna, Georgia are assigned low likelihood. When Bioclim layers are exclude d, the predicted distribution of M. cartwrighti becomes much broader. There are still moderate to high likelihood values assi gned to the Tallahassee Hills region of the Florida Panhandle and the areas in North Central Florida, but the pattern in Georgia has changed. Much of the Upper Coastal Plain region of Georgia is assigned a moderate level of likelihood, and the known localities of M. cartwrighti near Americus and Vienna are now represented. This moderate likelihood region extends North east into South Carolina, and includes the Fall Line region (where M. lethroides and M. retusus occur). The collecting site in Hinesville, Georgia still occupies a relatively isolat ed position. 93

PAGE 94

Mycotrupes gaigei Most of the area assigned high leve ls of likelihood of occurrence for M. gaigei is in Florida ( Figure 7-4 ). In addition, there are some scatte red areas with mode rate likelihood in Georgia and South Carolina. Much of the large area assigned high like lihood in Florida is represented by known occurrences. The Brooks ville Ridge and the area along the Suwannee River, both of which host M. gaigei have been assigned high lik elihoods. Several large areas that have been assigned high likelihood ar e outside of the presently known range of M. gaigei These areas include the Southern Brooksville Ri dge, the Central Ridge, and several areas near Jacksonville. The Seminole County, Florida sites are isolated in terms of distance to the nearest known occurrence (Marion County). The map of likel ihood of occurrence suggests a relatively broad swath of suitable hab itat extending west from Seminole Co unty with the likelihood becoming patchier in the vicinity of the Central Ridge. When Bioclim layers are excluded, the map of likelihood of occurr ence is apparently little changed. The area of known distribution of M. gaigei is still represented. The few differences include an area of moderate like lihood now present in the Florida Panhandle. Mycotrupes gaigei would appear to be the species with the least difference between the model produced with, and that produced without the Bi oclim layers. Two possibilities that might explain this difference (both possi bilities assume that climate is not an important factor in the distribution of the species) are: 1) Mycotrupes gaigei is represented by more collecting localities than the other species, and more data are exp ected to yield a better-fit ting niche model. 2) M. gaigei is widely distributed geograp hically, and its distribution is probably represented by more climatic variation, meaning that climate w ould be found to be less important by MaxEnt. 94

PAGE 95

Mycotrupes pedester The model for M. pedester is remarkable, because the entirety of the area assigned moderate or high likelihood is re stricted to southwest Florida ( Figure 7-5 ). This area includes the known localities, and it also suggests the exis tence of suitable habitat further to the East along the Caloosahatchee River, and South into Hendry and Collier counties. The model for M. pedester had the highest AUC value of all the analyses (0.999). These results should be interpreted with caution, because the use of a small number of presence data points with similar environmental characteristics would be expected to give the same results, and it is possible that there are presently unknown populations of M. pedester that are not includ ed in the areas of moderate to high likelihood of occurrence. The use of additional presence localities would give more confidence in the results, however, it is possible that M. pedester has an extremely limited geographic distribution, and the addi tion of presence localities may be difficult. In that case, surveying for Mycotrupes in the areas suggested above (from which M. pedester is not presently known) may be advisable. When Bioclim layers are excluded, the likelihood of occurrence map for M. pedester includes much of coastal Florida and parts of coastal Georgia and South Carolina. This coastal pattern is interesting, as it is a differe nt pattern from that seen in the other Mycotrupes species. Effect of Bioclim Data on Predicted Distributions For all Mycotrupes species, the models including Bioclim layers have higher AUC values compared to those excluding Bioclim. This i ndicates that the Bioclim layers are providing information that allows MaxEnt to build a more precise niche model for each species, at least based on the presence localities available. In addition, the areas of high likelihood of occurrence produced by the models including Bioclim show less overlap between species (for instance, when Bioclim layers ar e excluded, the likelihood of occurrence areas for M. retusus overlaps 95

PAGE 96

with that of M. gaigei ). The use of the Bioclim layers pr oduces areas of likelihood of occurrence that agree more with what is known about the distributions of Mycotrupes species. It is highly possibl e that species of Mycotrupes are adapted to the cl imate characterizing their distributions. For example, the c limate present across the distribution of M. pedester is different from the climate farthe r north, where other species of Mycotrupes are present (Mitchell and Ensign 1928). Likewise, Lobo et al. (2006) found climate data to be an important factor in the distribution of the geotrupid genus Jekelius in the Iberian Penins ula. Minimum winter temperatures can cause significant mortality in insects living unde rground, and it is possible that different species of Mycotrupes could have different tolerances for winter ground-freezes (especially considering the observed species differe nces in burrow depth) (M ail 1930). It is also possible, however, that the Bioclim layers, whil e being associated with the distri butions of species of Mycotrupes, are not important factors in limiting the distribution of Mycotrupes species. The geographic distributio ns of certain climate layers could be coincident with the distributions of Mycotrupes species. If this were true, it w ould limit the ability of the model to predict unknown populations and skew the analysis of the determinan ts of habitat suitability. Of all of the species of Mycotrupes studied, M. gaigei appears to have the least difference between the likelihood maps produced w ith, versus without, Bioclim data. Mycotrupes gaigei is also the only species in which th e most important layer in the model including Bioclim data was found to be a non-Bioclim layer (drainage). This result may be due to the wide geographic spread and large number of presence locations available for M. gaigei relative to the other species of Mycotrupes, which may have lessened the potential influence of coincidental climate layers. 96

PAGE 97

Mycotrupes are burrowing beetles, and the larvae live and feed in underground chambers (Howden 1954). These facts, combined with the apparent association of Mycotrupes with welldrained habitat (sandhills, etc. ), suggest that soil characteristics would have an important influence on the distribution of these beetles. This suggests to me that a dependence on deep well-drained soil and barriers to dispersal are the primary reasons for the restricted distributions of Mycotrupes. Because the non-Bioclim layers are more readily interpretable, further discussion will be limited to them. Relative Importance of Environm ental Layers Across Species of Mycotrupes The modeled niche of each species of Mycotrupes was different in terms of which environmental layers were found to be important, a nd the preferences within each of these layers. As discussed above, the association of different species with distinct variables across their geographic distributions does not mean that there are differences in habitat preferences between such species. The restricted geogr aphic area of distribution for many Mycotrupes species may increase the likelihood that the distribution will correlate, even if by chance, with some variable. This could apply to soil as well as climate. For example, if a species is restricted to a very small area with one soil type, this spec ies may be capable of living in other soil types, but it may not only because of an inability to disperse to those other areas. Depending on the size and location of the distribution of each species there might be different layers that could correlate with the distribution. A possible example of such an effect from the present study is as follows. The niche model for M. pedester had the highest contribution from the elevation layer (22.3%). It is possible that this is because all of the presence locations are at low elevations relative to the wide range of values in the elev ation layer (which includes m ountains in northern Georgia). Perhaps more significantly, some environmental layers appeared to be of general importance across all or most species of Mycotrupes Drainage was important for all, with most 97

PAGE 98

species preferring more well-drained soil. Low silt content was important for M. gaigei M. pedester and M. retusus Percent clay, elevation, and per cent sand were important for only one or two species. An interesting difference was ob served in the response to the soil bulk density layer between two species. Mycotrupes cartwrighti occurs in low bulk density soils, while M. retusus occurs in high bulk density soils. Bulk density is often negatively correlated with organic matter, which suggests that M. retusus prefers soil with less organic matter relative to M. cartwrighti. Soil Types Associated with Mycotrupes Species In an attempt to further characterize the soils associated with the distribution of Mycotrupes species, I obtained the ta xonomic classification of the soil at each site from the USDA General Soils Map ( Table 7-4 ). The following generalizati ons are based on these data. Proportions indicate the number of sites, of the total sites used in the niche modeling analysis, with a particular type of soil. It must be re membered that these associ ations between site and soil type are based, not on sampling at the site an d identification, but on digital soil maps derived from soil surveys, and there is the potential for inaccuracy. Not surprisingly, most of the soils at the sites are well drained. Mycotrupes gaigei soils are mainly Quartzipsamments (10/20) and Paleudults (8/20), both of which are well-drained. The Quartzipsamments are in the Entisol order, and have little or no diagnos tic horizons. They are sandy throughout their depth (USDA 2009). Quartzip samments are also present at many of the M. lethroides and M. retusus sites. Mycotrupes cartwrighti inhabits a different group of soils While these soils are also welldrained, most (8/9) sites are in Kandiudult soils, meaning they have a clay-rich horizon. The soils at six of these sites are fine-loamy, kaolinitic, thermic typic Kandiudults, which is the classification of the Orangeburg series (USDA 2009). The soil present at the type locality of M. 98

PAGE 99

cartwrighti was described as in the Or angeburg series (Olson and Hubbell 1954). This soil is widely distributed in the Tallahassee Red Hills region. The soils at the four Mycotrupes pedester localities are clas sified under the Aquod suborder, which is characterized as having a sh allow water table (USDA 2009). It is possible that the sites in which M. pedester occurs may represent small areas of relatively well-drained soil amongst larger areas of poorly drained soil. Such "islands" of better-drained soil could have gone unnoticed at the scale of soil mapping. In Flor ida, slight differences in elevation are known to be associated with dramatic differences in drainage and vegetation (P. Skelley, pers. comm.). Detailed Discussion of the Distribution of Mycotrupes gaigei There are more presence records available for M. gaigei than any of the other species of Mycotrupes, hence there is reason to place more confiden ce in the modeled niche of this species. In addition, the distribution of th is species appears to be associ ated with certain geological features. Factors that may have affected the distribution of M. gaigei are discussed in detail here. The apparent association of M. gaigei with well-drained soils is not surprising, as adults are burrowing (to a depth of six feet [Howde n 1954]) and larvae live underground, and even a single inundation event would pose an obvious problem for their surv ival. In this study, it was found that all Mycotrupes species are associated with somewhat well-drained soils. Figure 7-6 is a map of drainage in the study ar ea. Dark areas are those that have been categorized under the three highest drainage classes (from USDA soil map). Dark points represent the M. gaigei localities. Note that all of the points are co ntained within the dark areas or near them. Many of the areas in which M. gaigei has been collected could be characterized as sandhill, a habitat that is defined by well-dr ained sandy soil and a characteristic mix of vegetation including turkey oak (Quercus laevis Walt.), longleaf pine ( Pinus palustris Miller), 99

PAGE 100

and wiregrass ( Aristida spp.) (Laessle 1958; Myers 1985). The apparently strong association of M. gaigei with both a particular vegeta tion type as well as a soil ty pe supports the observations of Harper (1914), who noted the influence of soil on vegetation in Florida. Another widely distributed (in Florida) type of vegetation that occurs in we ll-drained sandy soil in Florida is scrub. Scrub and sandhill are sometimes confused as they both are dominated by pines and oaks and have sandy soil. Scrub is a distinct ha bitat type characterized by a dense canopy of "scrubby" vegetation, including sand pine ( Pinus clausa (Chapm. ex Engelm.) Vasey ex Sarg.) and various oaks ( Quercus spp.) (Laessle 1958; Myers 1985). There are no records of Mycotrupes from scrub habitat. Peltotrupes (Coleoptera: Geotrupidae), which is also a deep burrowing beetle does occur in scrub, and it is also found in sandhills with M. gaigei (Woodruff 1973; Young 1950) Scrub and sandhill ha bitat are often in close proximity, with "islands" of one habitat of ten being found within areas of the other habitat (Myers 1985). Kurz (1942) hypothe sized that there were differen ces in subsoil water retention between sandhill and scrub, and that this was the cau se of the different vege tative associations. Kurz thought that scrub soils tended to retain water at a shallower de pth than sandhill soils. Laessle (1958) also suggested soil differences as a factor in th e distribution of the two habitats. Laessle hypothesized that scrub developed preferentially on the higherenergy, better washed dune sands, whereas sandhill represented the offs hore deposits of less washed sands. Such a difference in the origin of soils between the tw o habitats would suggest soil differences as a reason for the absence of Mycotrupes from scrub. Myers (1985), however, found evidence that frequency of fire played a role in the distributions of sandhill and scrub. In his study sites at the Archbold Biological Station (Florida), Myers f ound evidence that sandhill had been replaced by scrub in the absence of fire. In some cases, sandhill burns on an almost annual basis, whereas 100

PAGE 101

scrub is thought to burn less frequently (Laessle 1958). Deep burrowing by Mycotrupes may be an adaptation not only to warmer post-Pleistocene temperatures, but to frequent burning as well. As sandhill and scrub often are close together, and even grade into each other, a pitfall trap transect across a sandhill-scrub ecotone could yield interesting data on the habitat preferences of Mycotrupes. A striking characteristic of th e likelihood of occurrence map for M. gaigei is that most of the presence localities (except those in Seminol e County) are located in a relatively large, contiguous area of high likeli hood of probability. Because M. gaigei has been collected in so many localities, and there are no records from site s outside of this apparently contiguous area (except for the Seminole County records), it w ould seem unlikely that there would be large populations in the other large areas of high likelih ood of occurrence in Flor ida, such as the area within Citrus and Hernando Counties or the Lake Wales Ridge in Central Florida. This is because of the amount of collecting that has occu rred across Florida, not because of an inability of MaxEnt to properly model the niche of this species. The predicted distribution of M. gaigei appears to be coincident with two geological features. Florida is composed of a sequence of marine terraces, remnants of past sea levels caused by climate cycles. It is assumed that the higher (in elevation) terra ces are the oldest, and that there was a progressive lowering of maximu m interglacial sea leve l. The Wicomico and Penholloway terraces, which are adjacent and cons idered by Colquhoun et al. (1991) to include 12.9-30.3 meters above sea level (bas ed on stratigraphic units in South Carolina), contain 18 of the 20 localities for M. gaigei ( Figure 7-7 ). There has been some deformation of the Florida bedrock, resulting in changes in elevation, so el evation is only a rough es timate of the terraces (Opdyke et al. 1984). In addition, all attempts to map marine terraces in Florida have relied on 101

PAGE 102

elevation, rather than mapping terrace features (T. Scott, pers. comm.). Nevertheless, the apparent relationship between the distribution of M. gaigei and elevation suggests that sea level changes may have been an important factor in th e distribution of the speci es. The distributions of the subspecies of the snake Stilosoma extenuatum Brown (a Florida endemic) are separated by areas of less than 100 feet in elevation. It is possible that the subsp ecies inhabited separate islands at a time when the sea level was at the Wicomico shoreline. Stilosoma inhabits similar habitat to that of M. gaigei (sandhill and upland hammock) (Highton 1956). The predicted distribution of M. gaigei is also mostly coinci dent with a geological structure referred to as the Ocala Platform ( Figure 7-8 ). This area in NW peninsular Florida has been previously referred to as the "Ocala Uplift," but this te rm is misleading. Although the bedrock in this area is higher than that surrounding it, this is apparently the result of the subsidence of this surrounding rock in relation to the Ocala Platform, which has remained more stable. The subsidence is thought to be related to the general Miocene subsidence of the North American Plate and the Gulf of Mexico Basin (T. Scott, pers. comm.). The Ocala Platform is one of the few areas in Florida where the aqui fer is not confined by the Hawthorne Formation, which may result in better drainage in th is area (Brinkmann et al. 2007). Although the Hawthorne Formation was deposited on the entire Fl orida Platform, these sediments were eroded from the Ocala Platform when this area was exposed above sea level during the Miocene (T. Scott, pers. comm.). Interestingly, the other ar ea of distribution (Sem inole County) is near another high, the Sanford High ( Figure 7-8 ). An area such as the Ocala Platform coul d be important for two reasons: First, M. gaigei could have originated on such a struct ure. Alternatively, the habitat for M. gaigei could have 102

PAGE 103

formerly been more extensive (possibly because of a different climate or lower sea level) and a recent contraction in ha bitat could have resulted in remnant populations in this area. The records in Seminole County are isolated from the rest of the distribution of M. gaigei both geographically (being far away from the nearest r ecords, which are in Marion County, Florida), and apparently ecologically as well, because the MaxEnt likelihood of occurrence map suggests that the favorable habitat that contains the Se minole County sites is more isolated and patchy (compared to the major area of distribution of M. gaigei ). Geneva Hill, where the two Seminole County collecting sites that were used are situated, is an isolated, sandy hill. The lower, less well-drained nature of th e surrounding area is evident when travelling north on Route 46 out of Geneva. There are sandy areas s cattered to the southwest and west of Geneva Hill, toward Orlando, and there are records of M. gaigei in nearby Oviedo approximately 7 miles SW of the sites in Geneva (S. Fullerton, in litt.). The seemingly disjun ct distribution of the Seminole County records from the rest of the known distribution begs for an explanation. There are two simple (involvi ng only one step) explanations: 1) (Dispersal) M. gaigei dispersed at some point from somewhere in the main body of distribution to the Seminole County area, or from the latter to the former area. 2. (Vicariance) There was, at a previous tim e, a relatively con tinuous distribution of M. gaigei from the main area to the Seminole County area, and some event (such as a rise in sea level) later split the two areas with no subsequent re-c olonization of this intervening area. The northern part of the distribution of M. gaigei is also coincident with an area of particularly dense records of th e Southeastern Pocket Gopher, Geomys pinetus Rafinesque (Kovarik et al. 2000). Geomys pinetus like Mycotrupes, digs deep burrows, and apparently 103

PAGE 104

requires well-drained sandy soil. The gopher, howe ver, has evidently been more successful at dispersal, as indicated by its wider geographic distribution. Conclusion Was MaxEnt successful in mapping the true distributions of Mycotrupes species, and was it successful in proposin g additional areas where Mycotrupes may be found? The answer may depend on scale. Even with a species such as M. gaigei where there are many occurrence localities, we will not have a complete idea of its distribution in space and time. Apart from extensive pitfall trapping acro ss the mapped areas, there may not be an objective method to assess the success of the niche modeling. Howeve r, the high AUC values combined with (in some cases) the likelihood of occurrence maps showing presence localities nested within contiguous areas of apparently good habitat, suggests that the niche modeling was largely effective for Mycotrupes Some large, contiguous areas were assigned a hi gh likelihood of occurrence by MaxEnt for Mycotrupes species where they are not known. These areas may warrant surveying for Mycotrupes. Two species that may have the l east completely known distributions, M. pedester and M. lethroides unfortunately have the fewest known presence localities. This m eans that the MaxEnt niche models for these species are likely to be inferior to the niche models of M. gaigei a species for which many collecting localities we re available. Of course, it is expected that a species with a smaller geographic distribution will be repres ented by a smaller number of occurrence points. In this case, questions of scale may have conseq uences for the ability of the niche of a given species to be modeled sufficientl y. More collecting in the known distribution of a species, such as M. pedester would be expected to yiel d additional occurrence locali ties, because the scale of this analysis is relati vely fine (the raster cell size is approximately 111 meters on a side). Additional presence localities would give grea ter confidence in the resulting niche model. 104

PAGE 105

A major part of my work on Mycotrupes is the biogeography of the genus. A central question to this issue is the link between the biogeography of Mycotrupes and the diversity of habitats occupied by the species. Are Mycotrupes found in different habitats because of biological adaptations to different habitats, or is the difference in habitat a product of geographic patterns in soil and vegetation and th e insular nature of such habitat? The latter would imply that Mycotrupes has simple niche requirements, such as we ll-drained soil, that are satisfied by any of the habitats it is known to occupy. This woul d also be consistent with a vicariance-based hypothesis of speciation. Fo r example: A species of Mycotrupes ("species X") may be distributed in a patch of land that has a unique geological origin (such as soils with a high clay content). Species X could just as easily live in so ils with less clay in other areas, but because it is flightless and its dispersa l is limited from those other areas by certain barriers (such as poorlydrained habitat or a river), it remains restricted in distribution to a relatively small area with a distinct soil type. The reality of the situation probably lies somewhere in between the two extremes. Progress toward an answer to the qu estions of the degree of adaptation of Mycotrupes species to their present habitat, and their ability to live in ot her habitats where they are not presently known to occur (for instance, a well-drained habitat that happens to be cut off from a Mycotrupes population only because dispersal to it is not possible), could be made in several ways. One possible approach w ould be to study the biology of Mycotrupes in a controlled setting, varying certain environmen tal factors such as soil draina ge. Comparing the fitness of Mycotrupes across a range of conditions might reveal important limiting factors. Such a study would be difficult, given the long lifecycle (probable at least one year), deep burrowing habits, and largely unknown biology of Mycotrupes A simpler approach to study habitat specialization 105

PAGE 106

106 would be to sample Mycotrupes along pitfall trap tr ansects to find respons es along gradients of habitat variables such as vegetati on cover or soil drainage. Such a transect approach might also be used to quantify barriers to dispersa l between habitat hosti ng a given species of Mycotrupes and the nearest unoccupied habitat of a type that is known to host other species of Mycotrupes in other areas. Phillips et al. (2006) stated "If the realized niche and fundamental niche do not fully coincide, we cannot hope for any modeling algorithm to characterize the species' full fundamental niche; the necessary information is si mply not present in the occurrence localities. This problem is likely exacerbated when o ccurrence records are drawn from too small a geographic area ." They add "...the departure between the fundamental niche (a theoretical construct) and realized niche (which can be observed) of a species will remain unknown." Hence, when modeling the distribu tion of species, the results must be interpreted in light of the limitations of the data. Ideally, niche modeling should utilize biol ogically meaningful information rather than simply using factors that happen to be corre lated with a species' distribution. This is difficult, however, when two things are considered: 1) That these techniques are used in order to discover biologically meaningful informati on, and 2) one has to consciously choose to include each layer of in terest in the analysis. This is a philosophical problem that might be impossible to solve.

PAGE 107

107 Table 7-1. Mycotrupes presence data for niche modeling. Species State: County Location Latitude Longitude Source M. lethroides GA: Burke Yuchi Wildlife Management Area 33.0808 -81.7845 K. Beucke, specimen data. M. lethroides GA: Burke Yuchi Wildlife Management Area 33.0836 -81.7756 K. Beucke, specimen data. M. lethroides GA: Burke Yuchi Wildlife Management Area 33.0828 -81.7798 K. Beucke, specimen data. M. lethroides GA: Richmond US Hwy. 1, 0.5 mi E. of Rich mond/Jefferson Co. line. 33. 2832 -82.2920 Harpootlian 1995 M. lethroides GA: Richmond Junction of I-520 and US Hwy. 1. 33.4235 -82.0589 P. Skelley, specimen data. M. lethroides GA: Richmond Junction of US Hwy. 25 and SR 415. 33.3851 -82.0261 K. Phillips, specimen data. M. retusus SC: Lexington Near Gaston. 33.8234 -81.1976 K. Phillips, specimen data. M. retusus SC: Aiken Junction of Oak Club Rd. and US Hwy. 78. 33.5113 -81.5590 K. Beucke, specimen data. M. retusus SC: Aiken NE of White Pond on Webb Pond Rd. 33.4612 -81.4331 K. Beucke, specimen data. M. retusus SC: Richland US Hwy. 1, Near Sesquicentennial St ate Park 34.1020 -80.9112 K. Beucke, specimen data. M. cartwrighti FL: Leon Eleanor Klapp-Phipps Park 30. 5382 -84.2890 K. Beucke, specimen data. M. cartwrighti FL: Leon Eleanor Klapp-Phipps Park 30. 5385 -84.2894 K. Beucke, specimen data. M. cartwrighti FL: Leon Tall Timbers Research Station 30. 6713 -84.2359 D. Almquist, specimen data. M. cartwrighti FL: Leon Tall Timbers Research Station 30.6720 -84.2363 K. Beucke, specimen data. M. cartwrighti FL: Jefferson Avalon conservation easement. 30.3943 -83.8961 K. Beucke, specimen data. M. cartwrighti FL: Jefferson Avalon conservation easement. 30. 3942 -83.8969 D. Almquist, specimen data. M. cartwrighti GA: Thomas Thomasville 30.8282 -84.0115 K. Beucke, specimen data. M. cartwrighti GA: Liberty Hinesville 31.8747 -81.5742 K. Beucke, specimen data. M. cartwrighti FL: Liberty Torreya State Park 30.5589 -84.9503 D. Almquist, specimen data. M. gaigei FL: Alachua Archer 29.5259 -82.5350 K. Beucke, specimen data. M. gaigei FL: Columbia O'Leno State Park 29.916 7 -82.5843 K. Beucke, specimen data. M. gaigei FL: Columbia Ft. White, US Hwy. 27 and Paisley Drive 29.9136 -82.7038 P. Choate, specimen data. M. gaigei FL: Columbia SR 47, S. of Ft. White 29.8814 -82.7339 P. Choate, specimen data. M. gaigei FL: Sumter Tillman Hammock 28.9536 -82.1333 K. Beucke, specimen data. M. gaigei FL: Seminole Geneva, Cochran Rd. 28.745 9 -81.1300 K. Beucke, specimen data. M. gaigei FL: Seminole Geneva, SR 46 and Ridge Rd. 28.7470 -81.1295 K. Beucke, specimen data. M. gaigei FL: Gilchrist SR 26 W. of Newberry 29.6313 -82.7052 K. Beucke, specimen data. M. gaigei FL: Levy SR 24, near Alachua/Levy Co. li ne. 29.5088 -82.5623 K. Beucke, specimen data. M. gaigei FL: Marion SR 484 and 76th Court. 29.0230 -82.2462 K. Beucke, specimen data. M. gaigei FL: Marion Marion Oaks Manor and SW 56th Court 28.9962 -82.2129 K. Beucke, specimen data. M. gaigei FL: Marion SW 59th Ave. Rd. and SW 158th Ln 28.9920 -82.2197 K. Beucke, specimen data. M. gaigei FL: Marion SR 474A and SE 1st Ave. Rd. 29.0093 -82.1350 K. Beucke, specimen data. M. gaigei FL: Suwannee CR 137 N. of US Hwy. 27. 29.9938 -82.8106 P. Choate, specimen data. M. gaigei FL: Dixie Old Town 29.5760 -82.9678 K. Beucke, specimen data. M. gaigei FL: Lafayette CR 425, 0.6 mi N of US Hwy. 27. 29.9905 -82.9957 P. Choate, specimen data. M. gaigei FL: Lafayette Mayo 30.0618 -83.1905 P. Choate, specimen data. M. gaigei FL: Lafayette CR 354 N. of US Hwy. 27. 30.0716 -83.0961 P. Choate, specimen data. M. gaigei FL: Lafayette Townsend 30.1494 -83.3287 P. Choate, specimen data. M. gaigei FL: Madison CR 53, N. of Madison Co. line. 30.2743 -83.2887 P. Choate, specimen data. M. pedester FL: Lee US Hwy. 41, 0.1 mi S. of Constituti on Blvd. 26.4745 81.8361 K. Beucke, specimen data. M. pedester FL: Lee Babcock Ranch 26.7611 81.6808 K. Beucke, specimen data. M. pedester FL: Lee Babcock Ranch 26.7596 81.6812 K. Beucke, specimen data.

PAGE 108

Table 7-2. Environm ental layers used. Database Layer Type Elevation National Map Seamless Server Elevation Continuous Soil U.S. General Soils Map USDA Drainage Categorical, 9 classes Hydric Class Categorical, 3 classes Bulk Density Continuous Sand Percent (top layer) Continuous Clay Percent (top layer) Continuous Silt Percent (top layer) Continuous Bioclim WorldClim 1, Annual Mean Temperature Continuous 2, Mean Diurnal Range Continuous 3, Isothermality Continuous 4, Temperature Seasonality Continuous 5, Max. Temperature Warmest Month Continuous 6, Min. Temperature Coldest Month Continuous 7, Temperature Annual Range Continuous 8, Mean Temperature Wettest Quarter Continuous 9, Mean Temperature Driest Quarter Continuous 10, Mean Temperature Warmest Quarter Continuous 11, Mean Temperature Coldest Quarter Continuous 12, Annual Precipitation Continuous 13, Precipitation Wettest Month Continuous 14, Precipitation Driest Month Continuous 15, Precipitation S easonality Continuous 16, Precipitation Wettest Quarter Continuous 17, Precipitation Driest Quarter Continuous 18, Precipitation Warmest Quarter Continuous 19, Precipitation Coldest Quarter Continuous 108

PAGE 109

109 Table 7-3. Diagnostics and results from niche modeling. With Bioclim Without Bioclim Species AUC P-value, threshold Most Important Layers AUC P-v alue, threshold Most Important Layers M. lethroides .995 1. 60E -6 .466 Bioclim 2, Bioclim 13, Clay percent, Drainage, Sand percent .930 7. 49E -4 .608 Clay percent, Drainage M. retusus .993 1. 82E -4 .301 Bioclim 7, Silt percent, Bioclim 8 .974 2. 46E -3 .121 Silt percent, Drainage, Bulk density, Elevation M. cartwrighti .845 1. 267 E -3 .091 Bioclim 19, Bioclim 8, Drainage, Bioclim 3, Bioclim 13 .685 3. 298 E -2 .588 Drainage, Hydric class, Elevation, Sand percent, Bulk density M. gaigei .969 1. 171E -11 .084 Drainage, Bioclim 8, Bioclim 2, Bioclim 14 .872 2. 551E -5 0 Silt percent, Drainage, Sand percent M. pedester .999 1. 1E -3 .691 Bioclim 15, Silt percent, Bioclim 16 .977 2. 28E -2 .534 Silt percent, Elevation, Drainage

PAGE 110

110 Table 7-4. Taxonomic classification of soil at each site used in niche modeling, acco rding to the United States General Soils Map. Species State: County Location Taxonomy M lethroides GA: Burke Yuchi Wildlife Mana g ement Area Arenic kandiudults loam y siliceous thermic M lethroides GA: Burke Yuchi Wildlife Mana g ement Area Arenic kandiudults, loa m y siliceous, thermic M lethroides GA: Burke Yuchi Wildlife Mana g ement Area Arenic kandiudults, loam y siliceous, thermic M lethroides GA: Richmond US Hw y 1, 0.5 mi E. of Siliceous, thermic t yp ic q uartzi p samments M lethroides GA: Richmond Junction of I-520 and US Hw y 1. T yp ic kanha p ludults, fine-loam y siliceous, thermic M lethroides GA: Richmond Junction of US Hw y 25 and SR 415. Thermic, coated t yp ic q uartzi p samments M retusus SC: Lexin g ton Near Gaston. T yp ic q uartzi p samments, thermic, coated M retusus SC: Aiken Junction of Oak Club Rd. and US T yp ic q uartzi p samments, thermic, coated M retusus SC: Aiken NE of White Pond on Webb Pond Rd. Grossarenic kandiudults, loam y siliceous, thermic M retusus SC: Richland US Hw y 1, Near Ses q uicentennial T yp ic q uartzi p samments, thermic, coated M cartwri g hti FL: Leon Eleanor Kla pp -Phi pp s Par k Fine-loam y kaolinitic, thermic t yp ic kandiudults M cartwri g hti FL: Leon Eleanor Kla pp -Phi pp s Par k Fine-loam y kaolinitic, thermic t yp ic kandiudults M cartwri g hti FL: Leon Tall Timbers Research Station Fine-loam y kaolinitic, thermic t yp ic kandiudults M cartwri g hti FL: Leon Tall Timbers Research Station Fine-loam y kaolinitic, thermic t yp ic kandiudults M cartwri g hti FL: Jefferson Avalon conservation easement. Fine-loam y kaolinitic, thermic t yp ic kandiudults M cartwri g hti FL: Jefferson Avalon conservation easement. Fine-loam y kaolinitic, thermic t yp ic kandiudults M cartwri g hti GA: Thomas Thomasville Fine-loam y kaolinitic, thermic p linthic kandiudults M cartwri g hti GA: Libert y Hinesville Ultic ha p la q uods, sand y siliceous, thermic M cartwri g hti FL: Libert y Torre y a State Park Loam y kaolinitic, thermic g rossarenic kandiudults M g ai g ei FL: Alachua Arche r loam y siliceous, h yp erthermic g rossarenic p aleudults M g ai g ei FL: Columbia O'Leno State Park Thermic, coated lamellic q uartzi p samments M g ai g ei FL: Columbia Ft. White, US Hw y 27 and Paisle y Loam y siliceous, subactive, thermic arenic p aleudults M g ai g ei FL: Columbia SR 47, S. of Ft. White Thermic, coated lamellic q uartzi p samments M g ai g ei FL: Sumte r Tillman Hammock H yp erthermic, uncoated t yp ic q uartzi p samments M g ai g ei FL: Seminole Geneva, Cochran Rd. H yp erthermic, uncoated t yp ic q uartzi p samments M g ai g ei FL: Seminole Geneva, SR 46 and Rid g e Rd. H yp erthermic, uncoated t yp ic q uartzi p samments M g ai g ei FL: Gilchris t SR 26 W. of Newberr y Sand y siliceous, h yp erthermic aeric ha p la q uads M g ai g ei FL: Lev y SR 24, near Alachua/Lev y Co. line. H yp erthermic, uncoated t yp ic q uartzi p samments M g ai g ei FL: Marion SR 484 and 76th Court. H yp erthermic, uncoated t yp ic q uartzi p samments M g ai g ei FL: Marion Marion Oaks Manor and SW 56th H yp erthermic, uncoated t yp ic q uartzi p samments M g ai g ei FL: Marion SW 59th Ave. Rd. and SW 158th Ln. H yp erthermic, uncoated t yp ic q uartzi p samments M g ai g ei FL: Marion SR 474A and SE 1st Ave. Rd. Loam y siliceous, h yp erthermic g rossarenic p aleudults M g ai g ei FL: Suwannee CR 137 N. of US Hw y 27. Thermic, coatedlamellic q uartzi p samments M g ai g ei FL: Dixie Old Town Loam y siliceous, subactive, thermic arenic p aleudults M g ai g ei FL: Lafa y ette CR 425, 0.6 mi N of US Hw y 27. Fine-loam y siliceous, semiactive, thermic a q uic M g ai g ei FL: Lafa y ette Ma y o Loam y siliceous, subactive, thermic arenic p aleudults M g ai g ei FL: Lafa y ette CR 354 N. of US Hw y 27. Loam y siliceous, subactive, thermic arenic p aleudults M g ai g ei FL: Lafa y ette Townsend Loam y siliceous, subactive, thermic arenic p aleudults M g ai g ei FL: Madison CR 53, N. of Madison Co. line. Loam y siliceous, subactive, thermic arenic p aleudults M p edeste r FL: Lee US Hw y 41, 0.1 mi S. of Constitution Sand y siliceous, h yp erthermic aeric ha p la q uods M p edeste r FL: Lee Babcock Ranch Sand y siliceous, h yp erthermic ultic ha p la q uods M p edeste r FL: Lee Babcock Ranch Sand y siliceous, h yp erthermic ultic ha p la q uods Cr yp tic s p ecies? FL: Ta y lo r Fenhollowa y Loam y siliceous, thermic a q uic arenic ha p ludalfs

PAGE 111

Figure 7-1. Likelihood of occurrence map for M. lethroides, with Bioclim (left) and without (right). Figure 7-2. Likelihood of occurrence map for M. retusus with Bioclim (left) and without (right). 111

PAGE 112

Figure 7-3. Likelihood of occurrence map for M. cartwrighti with Bioclim (left) and without (right). Figure 7-4. Likelihood of occurrence map for M. gaigei with Bioclim (left) and without (right). 112

PAGE 113

Figure 7-5. Likelihood of occurrence map for M. pedester, with Bioclim (left) and without (right). Figure 7-6. Well-drained soil and M. gaigei collecting localities. 113

PAGE 114

Figure 7-7. Marine terraces in Florida. The Wicomico terrace is in blue and the Penholloway is in green (Healy 1975). Figure 7-8. The Ocala Platform area (labeled). The Sanford High is the smaller area, circled in red, in Eastern Florida (USGS 2002). 114

PAGE 115

CHAPTER 8 CONCLUSION My study is the result of almo st four years of field and laboratory work on the genus Mycotrupes. This genus was selected for a variet y of reasons, among them my interest in Geotrupidae, the poorly known biology of Mycotrupes, and the desire to te st the biogeographical hypothesis of Hubbell (1954). Mycotrupes specimens were collected from much of the presently known distribution of the genus. This degree of sampling is a result of both knowledge of the distributions of the different spec ies, gained from literature reco rds, museum specimen data, and personal communication with experts, as well as persistence. Nu merous sites that did not yield Mycotrupes were visited, thus the number of successf ul collections underre presented the whole effort. Additional field work would have b een preferable, but time and money were major limiting factors. The phylogenetic hypothesis of Mycotrupes put forth in my study refutes that of Hubbell (1954), which he based on assumptions regardin g the morphological evol ution and biogeography of Mycotrupes. My study was based on newer technique s (not available to Hubbell), including nucleotide sequence data, and was conducted wi th modern phylogenetic methods. For these reasons I believe that the present phylogenetic hypothesis (see Chapter 5) is more accurate. There is little support for the hypotheses of Hubbell that M. gaigei is a "primitive" Mycotrupes, and that M. cartwrighti and M. pedester are sister taxa. This phylogenetic study has also revealed a possible cryptic species from a location near what has been considered the distribution of M. cartwrighti The possibility that this entity may be a separate species is further supported by the high mean pairwise distances between its sequence and M. cartwrighti from other locations. The collection of a larger series from this location (only four specimens, two of them se quenced, are currently known), increasing the 115

PAGE 116

potential sample size for both morphological as well as molecular study, would enable a more robust test of the distinctness of th is entity as a new species. It is therefore not formally named at this time. Because the evolutionary hypothesis of Hubbe ll (1954) was rejected in this study, it is also necessary to reject his biogeog raphic hypothesis. Hubbell proposed that M. gaigei arose through a vicariance event in which an island of ha bitat in Central Florida was isolated during a rise in sea-level, from the ancestral habitat wh ich receded to the Fall Line. In my phylogenetic study, I determined that M. gaigei is derived relative to other Mycotrupes species, so the existence of this Central Florida island habitat is not required. The all opatric distributions of Mycotrupes species, combined with the fact that the two most basal species (M. lethroides and M. retusus ) are found on the highest, most inland habitat (Fall Line), strongly suggests that sea level played a major role in the evolution of Mycotrupes. The poorly constrained divergence date estimates produced by the BEAST analysis, co mbined with the lack of fossil data and a complex history of sea level change, makes it impossible to correlated divergence times in Mycotrupes with particular changes in sea level. Many interesting biogeographic questions remain. Among them is the historical extent of Mycotrupes habitat, which was probably, at different times, more and less extensive than at present. This issue bears directly on the rela tive importance of vicarian ce versus dispersal in speciation within this group. My niche modeling study indicate d that a high degree of drainage of the soil is an important variable in determ ining the distribution of all Mycotrupes species. This may come as an obvious conclusion to entomologists who are familiar with Mycotrupes and its "typical" habitat, however, this had not been formally studied, and niche modeling provided evidence to 116

PAGE 117

support it. Other environmental layers, such as climate and elevation, were found to be important for individual Mycotrupes species, but this may have been an artifact of the restricted geographic distributions of the sp ecies, as discussed in Chapter 7. There are many other possible data layers available now, and s till others that could be construc ted by biologists, that might be used in future niche modeling studies. One interesting possibility is integrating historical information into layers that could be used to study biogeography via niche modeling. My phylogenetic study, central to this work, was based on a re latively small (by today's standards) nucleotide data set of 481 base pairs a nd a small number of specimens analyzed. It is suggested that future contributions of Mycotrupes character data will be in the form of additional nucleotide sequences and more specimens sample d per species. Although my search for a significant number of mor phological characters in Mycotrupes was unsuccessful, it is not my intention to dissuade further morphological stud y of the genus. A comparative morphological study of larvae from different species of Mycotrupes might provide morphological character data for future phylogenetic study. At present, the larva has been describe d from only one species, M. gaigei (Hubbell 1954). A rearing ex periment by P. Skelley (see Chapter 3) succeeded in yielding two larvae of M. gaigei Captive rearing or burrow ex cavation of other species of Mycotrupes may yield larvae as well. Observations made over the course of field work suggest that the depth of burrows in different species of Mycotrupes may be dependent on soil characteristics. For example, M. cartwrighti was seen to dig shallow burrows in clay-rich soil, while M. gaigei burrows were quite deep in sandy soil. With the limited data av ailable, however, it is difficult to separate the importance of species versus regional conditions. The effect of species and soil type on burrow 117

PAGE 118

depth and characteristics will require study a nd excavation of many burrows of different Mycotrupes species across a range of habitats within each species' distribution. Mycotrupes species have been listed as th reatened (see Chapter 2). Because Mycotrupes require well-drained habitat su ch as sandhill, they are uniquel y threatened by development. Well-drained uplands are develope d at a high rate in Florida for such uses as housing and agriculture (Enge et al. 1986). Any attempt to protect Mycotrupes will be hampered by an incomplete knowledge of their biology and distri bution. The distributions of certain species, such as M. pedester, may be poorly known. Further collectin g will be necessary in order to have a better idea of the co mplete distributions of Mycotrupes species and the environmental conditions that this fascinating group requires for survival. Mycotrupes is one of many taxa, including othe r flightless insects, inhabiting the southeastern Atlantic Coastal Pl ain that offer an opportunity to unravel a biogeographical story. Theodore Hubbell, who was concerned with the bi ogeography of insects in the southeastern Atlantic Coastal Plain and coauthored the 1954 monograph on Mycotrupes with Olson and Howden, recognized the potential in this re gion for the comparative study of biogeography across different insect taxa. In his words: "One may find concentrated here many kinds of evolutionary situations, the result of differences in vagility, ecological requirements, and original geographic location of the groups concerned." He also recognized the impo rtance of considering all available evidence when framing a regional explanation of biogeography: "There was, of course, only one actual regional history, and a hypot hesis that is set up to explain what happened in one group must be consistent with the eviden ce from others." (Hubbell 1956) It is my hope that my research on Mycotrupes will provide an inte resting comparison to future studies on the biogeography of life in th e Atlantic Coastal Plain. 118

PAGE 119

APPENDIX A CYTOCHROME OXIDASE I SEQUENCE DATA Peltotrupes youngi TATTATTAGACAAGAAAGA AGAAAAAAAGAAAC ATTTGGAACTTTAGGTATAATTT ATGCTATAATAGCAATTGGTTTATTAGG TTTTATTGTATGAGCAC ATCATATATTTAC TGTTGGAATAGATGTAGATACCCGAGC TTATTTCACTTCAGCAACAATAATTATTGC TGTACCTACAGGTATTAAAATTTTTAGATGATTGGCAACT CTTCATGGAACTCAAAT TAACTACTCCCCATCAATATTATGAGCT TTAGGATTTGTATTT TTATTTACAGTAGGA GGACTAACAGGCGTTATTCTTGCCAATTCAA GAATTGATATTG TTTTACATGATACG TATTATGTAGTAGCTCATTT CCATTATGTATTATCTATGGGAGCAGTATTCGCCATTA TAGGGGGATTCGTACATTGATTTCCTTT ATTTACTGGACTAAAT ATAAATAGAAAAT ATTTAAAAATTCAATTTTTTATT Mycotrupes lethroides A-1 TATTATTTGCCAAGAAAGAAGAAAAAAAGAAACTTTTGGA AGTTTAGGTATAATTTA CGCTATGATAGCGATTGGATTACTAGGTTTTA TTGTATGGGCTCATCATATATTCACA GTTGGAATAGATGTAGATACGCGAGCTTA TTTTACTTCTGCAACTATAATTATTGCTG TTCCTACAGGAATTAAAATTTTTAGTTGA TTAGCAACTTTACATGGAACACAAATAA ACTATTCACCTTCAATGCTATGAGCTTT AGGATTTGTTTTTCTATTTACAGTAGGAGG TTTAACAGGGGTAATTTTGGCTAATTCAAGA ATTGATATTGTTCTACATGATACCTAT TATGTAGTAGCTCATTTTCATTATGTCCTCT CTATAGGAGCAGTATTCGCAATTATAG CTGGTTTCATTCATTGATTTCCTTTATTTACTGGTTTAAATATAAACAGAAAATATCT AAAAATTCAATTTTCTATT M. lethroides A-2 TATTATTTGCCAAGAAAGAAGAAAAAAAGAAACTTTTGGA AGTTTAGGTATAATTTA CGCTATGATAGCGATTGGATTACTAGGTTTTA TTGTATGGGCTCATCATATATTCACA GTTGGAATAGATGTAGATACGCGAGCTTA TTTTACTTCTGCAACTATAATTATTGCTG TTCCTACAGGAATTAAAATTTTTAGTTGA TTAGCAACTTTACATGGAACACAAATAA ACTATTCACCTTCAATGCTATGAGCTTT AGGATTTGTTTTTCTATTTACAGTAGGAGG TTTAACAGGGGTAATTTTGGCTAATTCAAGA ATTGATATTGTTCTACATGATACCTAT TATGTAGTAGCTCATTTTCATTATGTCCTCT CTATAGGAGCAGTATTCGCAATTATAG CTGGTTTCGTTCATTGATTTCCCTTATTT ACTGGTTTAAATATAAACAGAAAATATTT AAAAATTCAATTTTTTATT M. retusus A-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAAC ATTTGGAACTTTAGGTATAATTT ATGCTATAATAGCAATTG GATTATTAGGGTTTATTGTAT GAGCCCATCACATATTCA CAGTAGGAATAGATGTTGACACTCGA GCTTATTTTACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTCAGATGATTAGCAACTTTACATGGAACACAAA TTAATTACTCCCCATCAAT ATTATGAGCTCTGGGATTTG TTTTTTTATTTACTGTGGG AGGTTTAACAGGTGTAATTTTAGCAAATTC AAGAATTGATATTATTTTACATGACAC ATACTATGTAGTAGCCCATTTCCATTATG TTCTTTCTATAGGAGCAGTATTTGCAATT ATAGCAGGATTTGTCCATTGATTTCCATT ATTCACAGGTTTAAATATAAATAATAAA TATTTAAAAATTCAATTTTTTATT 119

PAGE 120

M. retusus A-2 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAAC ATTTGGAACTTTAGGTATAATTT ATGCTATAATAGCAATTG GATTATTAGGGTTTATTGTAT GAGCCCATCACATATTCA CAGTAGGAATAGATGTTGACACTCGA GCTTATTTTACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTCAGATGATTAGCAACTTTACATGGAACACAAA TTAATTACTCCCCATCAAT ATTATGAGCTCTGGGATTTG TTTTTTTATTTACTGTGGG AGGTTTAACAGGTGTAATTTTAGCAAATTC AAGAATTGATATTATTTTACATGACAC ATACTATGTAGTAGCCCATTTCCATTATG TTCTTTCTATAGGAGCAGTATTTGCAATT ATAGCAGGATTTGTCCATTGATTTCCATT ATTCACAGGTTTAAATATAAATAATAAA TATTTAAAAATTCAATTTTTTATT M. retusus B-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAAC ATTTGGAACTTTAGGTATAATTT ATGCTATAATAGCAATTG GATTATTAGGGTTTATTGTAT GAGCCCATCACATATTTAC AGTAGGAATAGATGTTGATACTCGAGCTT ATTTTACTTCAGCAACTATAATTATTGCT GTCCCTACAGGAATTAAAATTTTCAGAT GATTAGCAACTTTACATGGAACACAAATT AATTACTCCCCATCAATATTATGAGCTCTGGGATTTGTTTTTTTATTTACTGTAGGAG GTTTAACAGGTGTAATTTTAGCAAATTCAAG AATTGATATTATTTTACATGACACAT ACTATGTAGTAGCCCATTTCCATTATGTTCTTTCTATAGGAGCAGTATTTGCAATTAT AGCGGGATTTGTCCATTGATT CCCATTATTCACAGGTTTA AATATAAATAATAAATA TTTAAAAATTCAATTTTTTATT M. retusus C-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACATTTG GAACCTTAGGTATAATTT ATGCTATAATAGCAATTG GATTATTAGGATTTATTGTATGAGCTCATCATATATTTAC AGTAGGAATAGATGTTGACACTCGAGCTTATTTTACTTCAGCTACTATAATTATTGCT GTTCCTACAGGAATTAAAATTTTCAGAT GATTAGCAACTTTACATGGAACACAAATT AATTATTCTCCATCAATATTATGAGCTCT GGGGTTTGTTTTTTTATTTACTGTCGGAG GTTTAACCGGTGTCATTTTAGCAAATTCAAGAATTGATA TTATTTTACATGATACATA CTATGTAGTAGCCCATTTCCATTATGTCCTCTCTATAGGAGCAGT ATTTGCAATTATA GCAGGATTTGTTCACTGATTCCCACTATTTACAGGTTTAAATATAAATAATAAATATT TAAAAATTCAATTTTTTATT M. retusus D-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACATTTG GAACCTTAGGTATAATTT ATGCTATAATAGCAATTG GATTATTAGGATTTATTGTATGAGCTCATCATATATTTAC AGTAGGAATAGATGTTGACACTCGAGCTTATTTTACTTCAGCTACTATAATTATTGCT GTTCCTACAGGAATTAAAATTTTCAGAT GATTAGCAACTTTACATGGAACACAAATT AATTATTCTCCATCAATATTATGAGCTCT GGGGTTTGTTTTTTTATTTACTGTCGGAG GTTTAACCGGTGTCATTTTAGCAAATTCAAGAATTGATA TTATTTTACATGATACATA CTATGTAGTAGCCCATTTCCATTATGTCCTCTCTATAGGAGCAGT ATTTGCAATTATA GCAGGATTTGTTCACTGATTCCCACTATTTACAGGTTTAAATATAAATAATAAATATT TAAAAATTCAATTTTTTATT 120

PAGE 121

M. retusus D-2 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACATTTG GAACCTTAGGTATAATTT ATGCTATAATAGCAATTG GATTATTAGGATTTATTGTATGAGCTCATCATATATTTAC AGTAGGAATAGATGTTGACACTCGAGCTTATTTTACTTCAGCTACTATAATTATTGCT GTTCCTACAGGAATTAAAATTTTCAGAT GATTAGCAACTTTACATGGAACACAAATT AATTATTCTCCATCAATATTATGAGCTCT GGGGTTTGTTTTTTTATTTACTGTCGGAG GTTTAACCGGTGTCATTTTAGCAAATTCAAGAATTGATA TTATTTTACATGATACATA CTATGTAGTAGCCCATTTCCATTATGTCCTCTCTATAGGAGCAGT ATTTGCAATTATA GCAGGATTTGTTCACTGATTCCCACTATTTACAGGTTTAAATATAAATAATAAATATT TAAAAATTCAATTTTTTATT M. cartwrighti A-1 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTACTAGGATTTATT GTATGAGCACATCATATATTTA CAGTAGGTATAGATGTTG ATACCCGAGCTTATTTTACTTC AGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGTTGATTAGCAACTCTACATGGAACACAAAT AAATTATTCCCCCTCTATACTATGAGCTTTGGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCTAATTCAAGTATTGATATTGTTCTTCATGATACAT ATTATGTAGTGGCTCATTTCCACTATGTT CTTTCTATAGGCGCC GTATTTGCAATTAT AGGGGGATTCGTTCATTGAT TTCCGTTATTTACAGGATT AAATATAAAT AGAAAATA TTTAAAAATTCAATTTTTAATT M. cartwrighti B-1 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTACTAGGATTTATT GTATGAGCACATCATATATTTA CAGTAGGTATAGATGTTG ATACCCGAGCTTATTTTACTTC AGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGTTGATTAGCAACTCTACATGGAACACAAAT AAATTATTCCCCCTCTATACTATGAGCTTTGGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCTAATTCAAGTATTGATATTGTTCTTCATGATACAT ATTATGTAGTGGCTCATTTCCACTATGTT CTTTCTATAGGCGCC GTATTTGCAATTAT AGGGGGATTCGTTCATTGAT TTCCGTTATTTACAGGATT AAATATAAAT AGAAAATA TTTAAAAATTCAATTTTTAATT M. cartwrighti C-1 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTATTAGGATTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTGATACCCGAGCTTATTTCACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTTAGATG ATTAGCAACTTTACATGGAACACAAA TAAATTATTCTCCCTCTATACTATGAGCT TTAGGGTTTGTTTTTTTATTCACTGTCGGA GGTTTAACGGGAGTAATTTTAGCCAATTCAAGTATTGATATCGTTCTTCATGATACAT ATTATGTAGTGGCTCATTTTCACTATGTTC TTTCTATAGGTGCTGTATTTGCAATTAT AGGAGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAACAGAAAATA TTTAAAAATCCAATTTTTAATT 121

PAGE 122

M. cartwrighti C-2 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTATTAGGATTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTGATACCCGAGCTTATTTCACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTTAGATG ATTAGCAACTTTACATGGAACACAAA TAAATTATTCTCCCTCTATACTATGAGCT TTAGGGTTTGTTTTTTTATTCACTGTCGGA GGTTTAACGGGAGTAATTTTAGCCAATTCAAGTATTGATATCGTTCTTCATGATACAT ATTATGTAGTGGCTCATTTTCACTATGTTC TTTCTATAGGTGCTGTATTTGCAATTAT AGGAGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAACAGAAAATA TTTAAAAATCCAATTTTTAATT M. cartwrighti E-1 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTATTAGGATTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTGATACCCGAGCTTATTTCACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTTAGATG ATTAGCAACTTTACATGGAACACAAA TAAATTATTCTCCCTCTATACTATGAGCT TTAGGGTTTGTTTTTTTATTCACTGTCGGA GGTTTAACGGGAGTAATTTTAGCCAATTCAAGTATTGATATCGTTCTTCATGATACAT ATTATGTAGTGGCTCATTTTCACTATGTTC TTTCTATAGGTGCTGTATTTGCAATTAT AGGAGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAACAGAAAATA TTTAAAAATCCAATTTTTAATT M. cartwrighti F-1 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTATTAGGATTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTGATACCCGAGCTTATTTCACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTTAGATG ATTAGCAACTTTACATGGAACACAAA TAAATTATTCTCCCTCTATATTATGAGCT TTAGGATTTGTTTTTTTATTCACTGTCGGA GGTTTAACGGGAGTAATTTTAGCTAATTCAAGTATTGATATTGTTCTTCATGATACAT ATTATGTAGTAGCTCATTTTCACTATGTTC TTTCTATGGGTGCTGTATTTGCAATTAT AGGAGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAACAGAAAATA TTTAAAAATCCAATTTTTAATT M. cartwrighti H-1 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTATTAGGATTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTGATACCCGAGCTTATTTCACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTTAGATG ATTAGCAACTTTACATGGAACACAAA TAAATTATTCTCCCTCTATACTATGAGCT TTAGGGTTTGTTTTTTTATTCACTGTCGGA GGTTTAACAGGAGTAATTTTAGCCAATTCAAGTATTGATATCGTTCTTCATGATACAT ATTATGTAGTGGCTCATTTTCACTATGTTC TTTCTATAGGTGCTGTATTTGCAATTAT AGGAGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAACAGAAAATA TTTAAAAATCCAATTTTTAATT 122

PAGE 123

M. cartwrighti H-2 CATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGATTATTAGGATTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTGATACCCGAGCTTATTTCACTTCAGCAACTATAATTATTG CTGTTCCTACAGGAATTAAAATTTTTAGATG ATTAGCAACTTTACATGGAACACAAA TAAATTATTCTCCCTCTATACTATGAGCT TTAGGATTTGTTTTTTTATTCACTGTCGGA GGTTTAACGGGAGTAATTTTAGCCAATTCAAGTATTGATATCGTTCTTCATGATACAT ATTATGTAGTGGCTCATTTTCACTATGTTC TTTCTATAGGTGCTGTATTTGCAATTAT AGGAGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAACAGAAAATA TTTAAAAATCCAATTTTTAATT M. cartwrighti I-1 TATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCTATAATAGCAATTG GATTATTAGGTTTCATTGTAT GAGCTCATCATATATTTAC AGTAGGAATAGATGTAGATACTCGAGCT TATTTTACTTCAGCAACTATAATTATTGC TGTTCCTACTGGAATTAA AATTTTTAGTTGATTAGCAACTTTACATGGAACACAAAT AAATTATTCTCCTTCAATATTATGAGCTCT AGGATTTGTTTTTTTATTCACTGTTGGA GGTTTAACTGGGGTAATTTTAGCTAATTCAA GAATTGATATTGTTCTCCATGACACAT ATTATGTAGTAGCCCATTTCC ATTATGTTCTTTCAATAGG AGCTGTATTTGCTATTAT AGCCGGATTTGTACATTGATTCCCATTATTTA CAGGATTAAACA TAAATAGAAAATA TTTAAAAATTCAATTTTTAATT M. cartwrighti I-2 TATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCTATAATAGCAATTG GATTATTAGGTTTCATTGTAT GAGCTCATCATATATTTAC AGTAGGAATAGATGTAGATACTCGAGCT TATTTTACTTCAGCAACTATAATTATTGC TGTTCCTACTGGAATTAA AATTTTTAGTTGATTAGCAACTTTACATGGAACACAAAT AAATTATTCTCCTTCAATATTATGAGCTCT AGGATTTGTTTTTTTATTCACTGTTGGA GGTTTAACTGGGGTAATTTTAGCTAATTCAA GAATTGATATTGTTCTCCATGACACAT ATTATGTAGTAGCCCATTTCC ATTATGTTCTTTCAATAGG AGCTGTATTTGCTATTAT AGCCGGATTTGTACATTGATTCCCATTATTTA CAGGATTAAACA TAAATAGAAAATA TTTAAAAATTCAATTTTTAATT M. cartwrighti J-1 TATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTCGGAACTTTAGGAATAATCT ATGCAATAATAGCAATCGGGT TATTAGGGTTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTG ATACCCGAGCTTATTTTACTTC AGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTGCACGGAACACAAAT AAATTATTCCCCCTCTATACTATGAGCTTTAGGGTTTGTTTTTTTATTCACTGTTGGG GGTTTAACAGGAGTAATTTTA GCTAATTCAAGAATTGATATTGTTCTTCACGATACA TATTATGTAGTAGCCCA TTTTCATTATGTTC TTTCTATAGGTGCTGTATTCGCAATTAT AGGGGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAACATAAAT AGGAAATA TTTAAAAATCCAATTTTTAGTT 123

PAGE 124

M. cartwrighti J-2 TATTATTAGACAAGAAAGAAGAAAAAAAGAAACTTTCGGAACTTTAGGAATAATCT ATGCAATAATAGCAATCGGGT TATTAGGGTTTATTGTATGAG CACATCATATATTTA CAGTAGGTATAGATGTTG ATACCCGAGCTTATTTTACTTC AGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTGCACGGAACACAAAT AAATTATTCCCCCTCTATACTATGAGCTTTAGGGTTTGTTTTTTTATTCACTGTTGGG GGTTTAACAGGAGTAATTTTA GCTAATTCAAGAATTGATATTGTTCTTCACGATACA TATTATGTAGTAGCCCA TTTTCATTATGTTC TTTCTATAGGTGCTGTATTCGCAATTAT AGGGGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAACATAAAT AGGAAATA TTTAAAAATCCAATTTTTAGTT M. cartwrighti L-1 TATTATCAGACAAGAAAGA AGAAAAAAAGAAACTTTTGGAACTTTAGGAATAATTT ATGCAATAATAGCAATTGGACTATTAGGATTTATTGTGTGAG CACATCATATATTTA CAGTAGGTATAGATGTTG ATACCCGAGCTTATTTTACTTC AGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAATTATTCTCCCTCTATGCTATGAGCTTTAGGGTTTGTTTTTTTATTTACTGTTGGAG GTTTAACAGGAGTAATTTTAGCT AATTCAAGTATTGATATTG TTCTCCATGATACATA TTATGTAGTGGCCCATTTTCACTATGTCC TTTCTATAGGCGCTGTATTTGCAATTATA GGGGGATTTGTTCATTGATTTCCATTATTTACAGGATTAAATATAAATAGAAAATAT TTAAAAATTCAATTTTTAATT M. gaigei A1 TATTATTAGCCAAGAAAGGAG AAAAAAAGAAACATTCGG AACCCTGGGTATAATTT ATGCAATAATAGCTATTGGGT TATTAGGATTTATTGTATGAG CACATCATATATTTAC CGTAGGAATAGATGTAGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTTCCTACGGGAATTAAAATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTCACTGTCGGA GGTTTAACAGGAGTAATTTTAGCAAATT CCAGAATTGACATTGTACTTCATGACACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GGGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTTCCCTTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei A-2 TATTATTAGCCAAGAAAGGAG AAAAAAAGAAACATTCGG AACCCTGGGTATAATTT ATGCAATAATAGCTATTGGGT TATTAGGATTTATTGTATGAG CACATCATATATTTAC CGTAGGAATAGATGTAGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTTCCTACGGGAATTAAAATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTCACTGTCGGA GGTTTAACAGGAGTAATTTTAGCAAATT CCAGAATTGACATTGTACTTCATGACACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GGGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTTCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT 124

PAGE 125

M. gaigei C-1 TATTGTTAGCCAAGAAAGAAG AAAAAAAGAAACATTCGG AACCCTGGGTATAATTT ATGCTATAATAGCTATTGGGTTATTAGGATTTATTGTAT GAGCACATCATATATTTAC CGTAGGTATAGATGTAGATACCCGAGCCT ATTTCACTTCAGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTACACGGAACACAAAT AAACTATTCTCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTCACTGTCGGA GGTTTAACAGGGGTAATTTTAGCCAATTCCAGAATTGATATTGTGCTTCATGATACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GGGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTCCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei C-2 TATTATTAGCCAAGAAAGAAG AAAAAAAGAAACATTCGG AACCCTAGGTATAATTT ATGCTATAATAGCTATTGGGTTATTAGGATTTATTGTATGAGCTCATCATATATTTAC CGTAGGAATAGATGTAGATACTCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTACACGGAACACAAAT AAATTATTCTCCCTCAATATTATGAGCTTTAGGGTTTGTTTTTTTATTCACTGTCGGA GGTTTAACAGGGGTAATTTTAGCCAATTCCAGAATTGATATTGTACTTCATGATACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GGGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTCCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei D-1 TATTATTAGCCAAGAAAGGAG AAAAAAAGAAACATTCGG AACCCTAGGTATAATTT ATGCAATAATAGCTATTGGGT TATTAGGATTTATTGTATGAG CACATCATATATTTAC CGTAGGAATAGATGTAGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTCCCTACGGGAATTAA AATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCCAATTCCAGAATTGACATTGTACTTCATGACACA TATTATGTAGTAGCCCATTTT CATTATGTTCTTTCAATAG GAGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTTCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei D-2 TATTATTAGCCAAGAAAGGAG AAAAAAAGAAACATTCGG AACCCTAGGTATAATTT ATGCAATAATAGCTATTGGGT TATTAGGATTTATTGTATGAG CACATCATATATTTAC CGTAGGAATAGATGTAGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTCCCTACGGGAATTAA AATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCCAATTCCAGAATTGACATTGTACTTCATGACACA TATTATGTAGTAGCCCATTTT CATTATGTTCTTTCAATAG GAGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTTCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT 125

PAGE 126

M. gaigei E-1 CATTATTAGTCAAGAAAGA AGAAAAAAAGAAACATTTG GAACCCTGGGAATAATTT ATGCTATATTAGCTATTGGATTATTAGGATTTATTGTATGAGCACATCATATATTTAC TGTAGGTATAGATGTAGATACCCGAGCCTATTTTACTTCAGCAACTATGATTATTGCT GTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACCTTGCATGGAACACAAATA AATTATTCCCCCTCAATATTATGAGCTTTAGGATTTGTTTTTTTATTTACTGTCGGAG GTTTAACGGGAGTAATTTTGG CTAATTCAAGAATTGATATTGTACTTCATGATACAT ATTATGTAGTGGCCCATTTTCACTATGTCCTTTCTATAGGAGCCGTATTTGCTATTAT AGCAGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAAT AGAAAATA TTTAAAAATTCAATTTTTAATT M. gaigei F-1 CATTATTAGTCAAGAAAGA AGAAAAAAAGAAACATTTG GAACCCTGGGAATAATTT ATGCTATATTAGCTATTGGATTATTAGGATTTATTGTATGAGCACATCATATATTTAC TGTAGGTATAGATGTAGATACCCGAGCCTATTTTACTTCAGCAACTATGATTATTGCT GTTCCTACAGGGATTAAAATTTTTAGATGATTAGCAACCTTACATGGAACACAAATA AATTATTCTCCCTCAATATTATGAGCTTTAGGATTTGTTTTTTTATTTACTGTCGGAG GTTTAACAGGAGTAATTTTGG CTAATTCAAGAATTGATATTGTACTTCATGATACAT ATTATGTAGTGGCCCATTTTCACTATGTCCTTTCTATAGGGGCCGTATTTGCCATTAT AGCGGGATTTGTTCATTGAT TTCCATTATTTACAGGATT AAATATAAAT AGAAAATA TTTAAAAATTCAATTTTTAATT M. gaigei M-1 TATTATTAGCCAAGAAAGGAG AAAAAAAGAAACATTCGG AACCCTAGGTATAATTT ATGCAATAATAGCTATTGGGT TATTAGGATTTATTGTATGAG CACATCATATATTTAC CGTGGGAATAGATGTAGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTCCCTACGGGAATTAA AATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCCAATTCCAGAATTGACATTGTACTTCATGACACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GAGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTTCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei M-2 TATTATTAGCCAAGAAAGGAG AAAAAAAGAAACATTCGG AACCCTAGGTATAATTT ATGCAATAATAGCTATTGGGT TATTAGGATTTATTGTATGAG CACATCATATATTTAC CGTGGGAATAGATGTAGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTCCCTACGGGAATTAA AATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCCAATTCCAGAATTGACATTGTACTTCATGACACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GAGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTTCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei N-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACATTTG GAACCCTAGGAATAATTT 126

PAGE 127

ATGCTATAATAGCTATTGGATTATTGGGATTTATTGTAT GGGCCCATCATATATTTAC CGTAGGTATAGATGTAGATACCCGGGCCT ATTTTACTTCAGCAACTATAATTATTGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTACACGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTA TTTACTGTTGGG GGTTTAACAGGAGTAATTTTAGCCAATTCAAGTATTGATATTGTACTCCATGACACA TATTATGTAGTAGCCCA TTTTCATTATGTTCTTTCTATAGGAGCTGTATTTGCTATTAT AGCAGGATTTGTACATTGATT CCCGTTATTCACAGGATTAAATATAAACAGAAAATA TTTAAAAATTCAATTTTTAATT M. gaigei N-2 N2TATTATTAGTCAAGAA AGAAGAAAAAAAGAAACATTTGGAACCCTAGGAATAAT TTATGCTATAATAGCTATTGGATTATTGGGATTTATTGTATGGGCCCATCATATATTT ACCGTAGGTATAGATGTAGATACCCGAG CCTATTTTACTTCAGCAACTATAATTATT GCTGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAAC TTTACACGGAACACAA ATAAATTATTCCCCCTCAATATTATGAG CTTTAGGATTTGTTTTTTTATTTACTGTTGG GGGTTTAACAGGAGTAATTTTAGCCAATTCAAGTATTGATATTGTACTCCATGACAC ATATTATGTAGTAGCCCATTTTCATTATGTTCTTTCTATAGGAGCTGTATTTGCTATT ATAGCAGGATTTGTACATTGATTCCCAT TATTCACAGGATTAAATATAAACAGAAAA TATTTAAAAATTCAATTTTTAATT M. gaigei O-1 TATTATTAGCCAAGAAAGAAG AAAAAAAGAAACATTCGG AACCCTGGGTATAATTT ATGCTATAATAGCTATTGGGTTATTAGGATTTATTGTAT GAGCACATCATATATTTAC TGTAGGAATAGATGTAGATACCCGAGCCTATTTCACTTCAGCGACTATAATTATCGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAATTATTCCCCCTCAATATTATGAGCTTTAGGATTTGTTTTTTTATTCACTGTCGGA GGTTTAACAGGAGTAATTTTAGCTAATTCTAGAATTGACATTGTACTTCATGATACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GGGCCGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTCCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei O-2 TATTATTAGCCAAGAAAGAAG AAAAAAAGAAACATTCGG AACCCTGGGTATAATTT ATGCTATAATAGCTATTGGGTTATTAGGATTTATTGTAT GAGCACATCATATATTTAC TGTAGGAATAGATGTAGATACCCGAGCCTATTTCACTTCAGCGACTATAATTATCGC TGTTCCTACAGGAATTAAAATTTTTAGATGATTAGCAACTTTACATGGAACACAAAT AAATTATTCCCCCTCAATATTATGAGCTTTAGGATTTGTTTTTTTATTCACTGTCGGA GGTTTAACAGGAGTAATTTTAGCTAATTCTAGAATTGACATTGTACTTCATGATACA TATTATGTAGTGGCCCATTTT CATTATGTTCTTTCAATAG GGGCCGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTCCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei P-1 TATTATTAGCCAAGAAAGAAG AAAAAAAGAAACATTCGG AACCCTGGGTATAATTT ATGCTATAATAGCTATTGGGTTATTAGGATTTATTGTAT GAGCACATCATATATTTAC CGTAGGAATAGATGTAGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC 127

PAGE 128

TGTTCCTACGGGAATTAAAATTTTTAGGTGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCCAATTCCAGAATTGATATTGTACTTCATGATACA TATTATGTAGTCGCCCATTTT CATTATGTTCTTTCAATAG GGGCTGTATTTGCAATTA TAGCGGGATTTGTCCATTGATTCCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. gaigei P-2 TATTATTAGCCAAGAAAGAAG AAAAAAAGAAACATTCGG AACCCTGGGTATAATTT ATGCTATAATAGCTATTGGGTTATTAGGATTTATTGTAT GAGCACATCATATATTTAC CGTAGGAATAGATGTGGATACCCGAG CCTATTTCACTTCAGCAACTATAATTATTGC TGTTCCTACGGGAATTAAAATTTTTAGGTGATTAGCAACTTTACATGGAACACAAAT AAACTATTCCCCCTCAATATTATGAGCT TTAGGATTTGTTTTTTTATTTACTGTCGGA GGTTTAACAGGAGTAATTTTAGCCAATTCCAGAATTGATATTGTACTTCATGATACA TATTATGTAGTCGCCCATTTT CATTATGTTCTTTCAATAG GGGCTGTATTTGCAATTA TAGCAGGATTTGTCCATTGATTCCCATTA TTTACAGGATTAAATATAAATAGAAAAT ATTTGAAAATTCAATTTTTAATT M. pedester A-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACCTTTGGAACTTTAGGAATAATCT ATGCTATAATAGCTATTGGATTATTAGGGTTTATTGTAT GAGCACATCATATATTTAC TGTAGGTATAGATGTTGAT ACCCGAGCTTATTTTACTTCAGC AACTATAATTATTGCT GTTCCTACAGGAATTAAAATTTTTAGATGGCTAGCAACTTTACATGGAACACAAATA AATTATTCTCCTTCAATATTATGAGCTTT AGGATTTGTTTTTTTATTTACTGTTGGAGG TTTAACGGGAGTAATTTTAGCTAATTCAAGA ATTGATATTGTTCTTCATGATACATAT TATGTAGTAGCTCACTTTCATTATGTTCT TTCCATAGGAGCTGTATTTGCAATTATAG CTGGATTTGTTCACTGATTTCCACTATTTA CAGGATTAAATATAAATAGAAAATATTT AAAAATTCAATTTTTAATT M. pedester A-2 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACCTTTGGAACTTTAGGAATAATCT ATGCTATAATAGCTATTGGATTATTAGGGTTTATTGTAT GAGCACATCATATATTTAC TGTAGGTATAGATGTTGAT ACCCGAGCTTATTTTACTTCAGC AACTATAATTATTGCT GTTCCTACAGGAATTAAAATTTTTAGATGGCTAGCAACTTTACATGGAACACAAATA AATTATTCTCCTTCAATATTATGAGCTTT AGGATTTGTTTTTTTATTTACTGTTGGAGG TTTAACGGGAGTAATTTTAGCTAATTCAAGA ATTGATATTGTTCTTCATGATACATAT TATGTAGTAGCTCACTTTCATTATGTTCT TTCCATAGGAGCTGTATTTGCAATTATAG CTGGATTTGTTCACTGATTTCCACTATTTA CAGGATTAAATATAAATAGAAAATATTT AAAAATTCAATTTTTAATT M. pedester B-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACCTTTGGAACTTTAGGAATAATCT ATGCTATAATAGCTATTGGATTATTAGGGTTTATTGTAT GAGCACATCATATATTTAC TGTAGGTATAGATGTTGAT ACCCGAGCTTATTTTACTTCAGC AACTATAATTATTGCT GTTCCTACAGGAATCAA AATTTTTAGATGGCTAGCAAC TTTACATGGAACACAAATA AATTATTCTCCTTCAATATTATGAGCTTT AGGATTTGTTTTTTTATTTACTGTCGGAGG 128

PAGE 129

TTTAACAGGAGTAATTTTAGCTAATTCAAGA ATTGACATTATTCTTCATGATACATAT TATGTAGTAGCTCACTTTCATTATGTTCT TTCCATAGGAGCTGTATTTGCAATTATAG CTGGATTTGTTCACTGATTTCCATTATTTA CAGGATTAAATATAAATAGAAAATATTT AAAAATTCAATTTTTAATT M. pedester C-1 TATTATTAGTCAAGAAAGA AGAAAAAAAGAAACCTTTGGAACTTTAGGAATAATCT ATGCTATAATAGCTATTGGATTATTAGGGTTTATTGTAT GAGCACATCATATATTTAC TGTAGGTATAGATGTTGAT ACCCGAGCTTATTTTACTTCAGC AACTATAATTATTGCT GTTCCTACAGGAATCAA AATTTTTAGATGGCTAGCAAC TTTACATGGAACACAAATA AATTATTCTCCTTCAATATTATGAGCTTT AGGATTTGTTTTTTTATTTACTGTCGGAGG TTTAACGGGAGTAATTTTAGCTAATTCAAGA ATTGACATTGTTCTTCATGATACATAT TATGTAGTAGCTCACTTTCATTATGTTCT TTCCATAGGAGCTGTATTTGCAATTATAG CTGGATTTGTTCACTGATTTCCATTATTTA CAGGATTAAATATAAATAGAAAATATTT AAAAATTCAATTTTTAATT 129

PAGE 130

APPENDIX B SPECIMENS EXAMINED Approximately 1,500 specimens from the following institutions were studied: AMNH American Museum of Natural History CNC Canadian National Collection CU Clemson University FSCA Florida State Co llection of Arthropods HMOU Hope Museum of Oxford University KBPC Kyle Beucke Private Collection KPPC Keith Philips Private Collection KU University of Kansas NCSU North Carolina State University UMMZ University of Mich igan Museum of Zoology TNHM The Natural History Museum, London USNM United States National Museum Note: All sequenced cryogenic material (indicated with a code, for example, A-1) is stored in the -80C freezer in the Branham Laboratory, University of Florida, Gainesville, Florida. M. lethroides USA: GEORGIA: No locality given other than Georgi a, "Georgia-see Abbot's drawings," 5506, retusus McLeay," 1 (TNHM); no locality given other than Georgia, 1 (Holotype) (HMOU); no locality given except Georgia, "5506a," "Figured in Abbot's dr awing of Georgian Insects," 1 (TNHM); 5 mi W of Augusta, 6-9-XII-1960, malt traps, R.E. Woodruff and E.W. Holder, 2 (FSCA); Burke Co. Yuchi Wildlife Management Area, 3-6-XI-2007, pig dung and fermenting malt pitfall, K. Beucke, 2 3 (KBPC); same except 6-XI-2007, K. Beucke, 1 (KBPC); sandy road off of Ebeneezer Church Road, turkey oak and pine with exposed sand, N33 04.848', W81 47.070', 3-6-XI-2007, pig dung and fermenti ng malt pitfalls, K. Beucke, 5 8 (KBPC); same except 6-XI-2007, excavated from burrows, K. Beucke, 1 1 (C-1) (KBPC); same except down sandy road from Ebeneezer Church Road, just past field, to right, in shaded oak forest, N33 05.017', W81 46.535', 3-XI-2007, excavated from burrows, K. Beucke, 4 1 (KBPC); same except collected on surface of road feeding on acorn, K. Beucke, 1 (KBPC); same except 130

PAGE 131

3-6-XI-2007, pig dung and fermenti ng malt pitfalls, K. Beucke, 2 (A-1, A-2), 1 (A-3) (KBPC); same except in oak-pine woods on side of sandy road, large field on other side of road, N33 04.965', W 81 46.786', 3-6-XI-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 (B-1), 1 (B-2) (KBPC); same except 8.3 mi E of St. 80 on St. 23, 6-III-1992, collected from pushups, P. Skelley, 1, 1 (FSCA); same except 7-10-III-1992, P. Skelley, 10 4 (FSCA); same except dung and malt pitfall, P. Skelley, 2 4 (FSCA); same except 14-I-1995, Skelley and Kovarik, 1 (KPPC); same except 18-20III-1996, Wappes and Turnbow, 3, 1 (FSCA); Richmond Co. Augusta, date illegible, O.L. Cartwrighti, 1 (UMMZ); same except 27-IX1930, T.H. Hubbell, 2 ("figured specimen), 1 ("figured specimen") (UMMZ), 1 (FSCA); same except 2-III-1944, O.L. Cartwrighti, 1 1 (USNM); same except Jct. I-520 and US-1, 414-I-1989, P. Skelley, 1 (FSCA); same except pig dung and malt pitfall in sand scrub, P. Skelley, 2 (FSCA); same except 13-31-XII-1989, P. Skelley, 3 (FSCA); same except 7-10III-1992, P. Skelley, 1 (FSCA); same except NW side of Hwy 25, 1.5 mi S. of Hwy 415, 18XII-1992, T.K. and T.B. Phillips, 1 (KPPC). M. pedester no locality given, determined by H.F. Strohecker, 1 (FSCA); USA: FLORIDA : Charlotte Co. Punta Gorda, 6-IV-1940, H. Ramstadt, 1 (paratype) (UMMZ); same but 15-IV-1940, H. Ramstadt, 1 (paratype) (UMMZ); same but IV-1951, M. Casselberry, 1 1 (topotype) (FSCA), 51 90 (AMNH), 1 2 (CNC), 1, 1 (USNM); same but no collector given, 1 (paratype) (UMMZ), 3 1 (all paratypes) (USNM); same but V-1953, H. Ramstadt, "fr. type series," 1 (CNC); DeSoto Co. Arcadia, X-1930, T.H. Hubbell, 1 (paratype) (UMMZ); Lee Co. Babcock Ranch, near herp. array. in open pine woods with saw palmetto, N26 45.665' W81 40.848', 30-I-6-II-2008, pig dung and fermenting malt pitfall, K. Beucke, 1 (B-1) (KBPC); same except N26 45.578' W81 40.869', K. Beucke, 2 (C-1) (KBPC); Estero, 29-III1962, malt traps, R.E. Woodruff, 2 (FSCA); same but 6-IV-1962, R.E. Woodruff, 2 (FSCA); same but 14-IV-1964, B.K. Dozier, 1 (USNM); same but 13-I-1965, B.K. Dozier, 1 (NCSU); same but malt trap 10AM-4PM, B.K. Dozier, 3 (USNM), 1 2 (FSCA); same but 10-III1965, malt trap 9AM-11AM, B.K. Dozier, 3 2 (FSCA), 3, 1 (USNM); same but malt trap 11AM-1:30PM, B.K. Dozier, 2, 2 (FSCA); same but malt trap 11AM-? (time illegible), B.K. Dozier, 1 (USNM); same but 4-IV-1965, malt trap, B.K. Dozier, 1 2 (FSCA); same but Rt. 41 just S of Constitution Blvd., at border of dense sandhill woods and parking lot for furniture store, N26 28.467' W81 50.168', 30-I-6-II-2008, pig dung and fermenting malt pitfall, K. Beucke, 1 1 (KBPC); N of Estero, 0.1 mi S of Constitution Blvd. on US-41, N26 28.472' W81 50.165', 17-19-III-2007, pitfall tra p, P. Skelley and B. Warner, 2 (FSCA), 3 (A-1, A-2, A-3) (KBPC); Ft. Myers, VI-1967, 1 (CNC); San Carlos (San Carlos Park?) on Hwy. 41, 2631-I-1992, fert. malt/dung trap, R. Morris, 1 (FSCA); 3.4 mi N of Koreshan State Historical Site, 6-9-IV-1991, M. Thomas and R. Turnbow, 11 6 (FSCA); Tice, 1.3 mi S of the Caloosahatchee River, on Ortiz Road, 28-30-III-1962, malt trap, R.E. Woodruff, 1 (FSCA, in alcohol). M. retusus USA: SOUTH CAROLINA: no locality given, with "93:" "67.45," 1 (TNHM); no locality given, from Hope Westwood collection, 3 2 (HMOU); no locality given, with "5/5"/"73," 1 131

PAGE 132

(USNM); Aiken Co. Aiken, 4-V-1939, O.L. Cartwright, 2 (USNM); Aiken, 2 mi N of Loop 118 on US-1, N33 36.101', W81 41.215', 7-10-III-1992, P. Skelley, 2 (FSCA); 2.3 mi N of Aiken, 26-III-1953, malt, "H+D," 4 (FSCA), 3 (UMMZ), 1 1 (CNC); 12.9 mi N of Aiken, 26-III-1953, malt, "H+D," 1 (FSCA); 5.3 mi E of Montmorenci post office, 7-10-III1992, pitfall, P. Skelley, 1 (FSCA); same but N33 30.466', W81 33.068', 7-III-1992, P. Skelley, 6, 4 (FSCA); same but pushups, P. Skelley, 1, 6 (FSCA); White Pond, 4-XI1932, J.E. Webb, 1, 1 (TNHM), 1, 1 (USNM); same but 5-XI-1932, J.E. Webb, 1 (TNHM); same but 15-XI-1932, J.E. Webb, 1 2 (USNM); same but 14-XII-1933, J.E. Webb, 1 (UMMZ); same but I-1934, O.L. Cartwright, 1, 3 (USNM), 1 (UMMZ); same but 28IV-1934, J.G. Watts, 1 (USNM); same but 30-III-1935, O.L. Cartwright, 1 (USNM), 1 (TNHM); same but 23-III-1937, J.G. Watts, 1 (CU); same but 25-III-1937, O.L. Cartwright, 1 (FSCA), 2, 3 (USNM), 1 (TNHM); Windsor, 25-XI1933, O.L. Cartwright, 3 1 (USNM), 1 (FSCA), 1, 1 (AMNH), 2 (one with "Figured specimen Olson + Hubbell '54 Fig 3."; other with "Figured specimen Olson + Hubbell '54 Figs 2, 10-12."), 1 (with "Described female Olson + Hubbell 1954; K.U.") (UMMZ), 1 (KU), 1 (CNC); same but David Dunavan, 2 (CU); Oak Club Road and 78, 3 mi N of 53, oak pine woods, N33 30.678' W81 33.539', 46-XI-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 (C-1) (KBPC); Webb Pond Road, oak and pine woods on side of road, N33 27.672' W81 25.987', 4-6-XI-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 (D-1), 1 (D-2) (KBPC); Kershaw Co. Blaney, 13X-1936, O.L. Cartwright, 2 1 (USNM); Lexington Co. near Hwy. 342, W of Gaston, N33 49.406'; W81 11.853', 26-XII-2006, T.K. Phillips, 1, 1 (FSCA), 4 (A-3, A-4, A-5, A-6), 2 (A-1, A-2) (KBPC); 5 mi NE of Pelion, 29-XII-1988, 1 (KPPC); Murray, 13-X-1936, O.L. Cartwright, 3 (USNM); "Hury 210"?, 1-V-1945, O.L. Cartwright, 1 (USNM); Richland Co., Columbia, 3-III-1932, O.L. Cartwright, 1 (UMMZ), 1, 1 (USNM); same but 21-III-1932, O.L. Cartwright, 1 (UMMZ), 1 (CNC), 1 (USNM), 1 (TNHM); same but 24-III-1932, O.L. Cartwright, 1 ("Figured specimen Olson + Hubbell '54; Figure 60.") (UMMZ); same but 25-X-1932, O.L. Cartwright, 2 (USNM), 1 2 (CU); same but 29-X-1933, O.L. Cartwright, 1 (TNHM), 1 (CNC); same but 23-IV-1934, O.L. Cartwright, 1 (USNM); same but 25-III1937, O.L. Cartwright, 1 (USNM); near entrance to Sesquice ntennial State Park in pine-oak woods, N34.1020 W80.9112 4-6-XI-2007, pig dung and fermenti ng malt pitfall, K. Beucke, 1 (B-1) (KBPC). M. cartwrighti USA: FLORIDA: No location given other than Florida, "From collection of Chas. Schaeffer," 1 (paratype) (USNM), 1 (AMNH); same but Hubbard and Schwarz, 1 (CNC); Dade Co., Miami, 6-IV-1919, H. Klages, 1 (paratype) (UMMZ); Duval Co. Atlantic Beach, A.T. Slosson, Ac. 26226, 1 (AMNH); Jefferson Co., Avalon conservation easement, oak/hickory forest with pine and Quercus falcata on slope, excavated from burrows approximately 18cm deep, N30 23.660' W83 53.763', 7-III-2007, K. Beucke, 2 (C-1, C-2) (KBPC); same but N30 23.652' W83 53.814', dung and fermenting malt pitfa ll, 6-7-III-2007, D. Almquist, 1 (D1) (KBPC); same but approximately 1.8 mi SE of intersection of US-27 and 19/SR-20 and Avalon Road, pine/oak/hickory forest, N30.388496 W83.886971 excavated from 20cm deep burrow in sandy loam with some clay, 26-X-2007, D. Almquist and A. Johnson, 1 (N-1) (KBPC); Monticello, 4-8-X-1914, Chas. Schaeffer, 1 (AMNH); same but Big Bend Hort. Lab, exp. plots, southwood pitf all, 14-IV-1969, Whitcomb, 1 (FSCA); same but Rainey pecans, 18132

PAGE 133

VI-1969, pitfall, Whitcomb, 1 2 (FSCA); same but trail one pitfall, 26-V-1969, Whitcomb, 1 (FSCA); same but trail two pitfall, 26-V-1969, Whitcomb, 1 (FSCA); same but trail three pitfall, beech magnolia ha mmock, 28-IV-1969, Whitcomb, 1 (FSCA); Leon Co., no location given, 10-VI-1922, J.S. Alexander, 1 (paratype) (UMMZ); same but 18-VI-1924, C.O. Handley, 1 (USNM); Bradfordville, 2-V1985, pig dung trap, S. Roman, 4 (FSCA); Elinor Klapp-Phipps Park, in field near woods, 27-II-2007, D. Almquist, 1 (G-1), 3 (G-2, G-3) (KBPC); same but recently burned upland forest, N30 32.310' W84 17.363', 7-III-2007, excavated from 24cm deep burrow, K. Beucke, 1 (B-1) (KBPC); same but N30 32.294' W84 17.340', 7-III-2007, excavated from 20cm deep burrow, K. Beucke, 1 (A-1) (KBPC); Miccosukee Canopy Greenway, NW of intersecti on of Miccosukee Road (CR-146) and Crump Road in disturbed former red oak woods (pine oak hickory forest) with shortleaf and loblolly pines in vicinity, (2 specimens with following in formation; cannot attrib ute collecting data to either specimen: 30.519153 W 84.133495 13-XII-2007, excavated from 20cm deep burrow under old horse dung on sandy trail; N 30.518187 W 84.134624 found walking on surface of sandy trail at 2:00PM), D. Almquist and A. Johnson, 1 1 (KBPC); Tallahassee, 10-12-IX1929, T.H. Hubbell, 1 (paratype) (FSCA), 1 (paratype) (UMMZ), 1 (paratype) (CNC); same but 10-13-IX-1929, T.H. Hubbell, 1 (paratype) (TNHM); sa me but 5-X-1960, crawling on ground, R.E. Woodruff, 1 (FSCA); same but 5-X-1960, dug from 1" burrow, R.E. Woodruff, 1 (FSCA); same but 5-7-X-1960, malt trap, R.E. Woodruff, 3, 1 (FSCA); same but 15-24-IV-1963, malt trap, R.E. Woodruff, 8 4 (FSCA); same but 15-24-IV-1967, R.E. Woodruff, 3 (USNM); 6 mi E of Tallahassee, 5-7-X-1960, R.E. Woodruff, 1 (NCSU); Tall Timbers Research Station, a large series of over 10,000 specimens collected on a weekly basis for 2.5 years by D.L. Harris, W.H. Whitcomb, and W.W. Baker (a small number were examined) (FSCA); same but 2-X-1972, Ross Arnett Jr., 1 1 (FSCA); same but 16-23-XI-1970, pitfall, W. Rosenberg, 4 7 (USNM); same but 27-XI-7-XII-1970, Harris, 2 (FSCA); same but 2930-IX-1989, P. Skelley, 2 (FSCA); same but 11-VIII-1969, D. Harris, 12 6 (FSCA); same but 15-IX-1969, pitfall, D. Harris, 6, 5 (FSCA); same but 19-IX-1970, D. Harris, 2 3 (FSCA); same but 27-XI-7-XII-1970, D. Harris, 3, 1 (FSCA); same but disturbed pine upland, N30.66683 W84.24455, 18-XII-2007, excavated from 12cm deep burrow in very hard packed clay soil in road, appeared to be buryi ng dense, tan, puffball-like fungi, D. Almquist and M. Paulsen, 1 (M-1) (KBPC); same but upland pine forest, N30 40.279' W84 14.154', 6-7III-2007, D. Almquist, 1 (E-1) (KBPC); same but N30 40.317' W84 14.175', 7-III-2007, excavated from 12cm deep burrow (with tiny pu shup) in packed clay soil, K. Beucke, 1 (F-1) (KBPC); same but "rep. old corn field," 9-15-XI-1971, D. Harris, 9 28 (FSCA); same but 13-20-XII-1971, D. Harris, 6 (CNC); same but 3-10-I-1972, D. Harris, 4 21 (FSCA), 3 (FSCA), 1 (CNC); same but 20-27-XII-1972, D. Harris, 4 14 (FSCA); same but "by F5A," "NB66," 7-III-1968, Wilson Baker, 1 (FSCA); same but rep. 1-A, 2-9-XI-1970, pitfall, D. Harris, 6, 6 (FSCA); same but rep. 2-E, 21-29-V-1973, pitfall, D. Harris, 1 (FSCA); same but small mammal trap B3A, 5-III-1970, Wilson Baker, 1 (FSCA); same but small mammal trap C-2, 9-III-1969, Wilson Baker, 1 (FSCA); same but in C5, Wilson Baker, 11-III-1968, 1 USNM); same but small mammal trap C5X, 9-III-1969, Wilson Baker, 1 (FSCA); same but small mammal trap D4X, 9-III-1969, Wilson Baker, 1 (FSCA); same but "by G4," 10-III-1968, Wilson Baker, 1 (FSCA); same but in G4X, 10-III-1968, Wilson Baker, 1 (USNM); Liberty Co. Torreya State Park, on dung, 26-XI-2000, Gino Nearns, 2 (FSCA); same but just inside entrance, in upland pine habitat, N30.55893 W84.95027 12-16-IV-2007, D. Almquist, 1 (L1) (KBPC); GEORGIA: Location illegible (Pelham?), 22-IV-1995, Department of Agriculture, 133

PAGE 134

"injuring (illegible) vi nes at Pelham, Ga"?, 1 (USNM); Baker Co., Newton, 30-III-1956, walking on dirt road 5:30PM, H. Howden, 1 (CNC); 6 mi N.E. Newton, 30-III-1956, walking on dirt road, 5PM, H. Howden, 1 (FSCA); Decatur Co., 2.7 mi WSW Faceville, Dist. 21, Lot 352, 22-25-III-1954, T.H. Hubbell, 1 2 (UMMZ); Dooley Co., Vienna, 24-VII-1930, "Leng No. 13300," H. Spieth, 1 (UMMZ); Grady Co., no location given, 1935 (rest of date illegible), H.S. Peters, 1 (USNM); Liberty Co., Ft. Stewart, Old Sundbury Road, N31 52.479' W81 34.452', 15-24-V-2007, pig dung and fermentin g malt pitfall, K. Beucke, 1 (J-1) (KBPC); same but N31 52.485' W81 34.440', 29-XI-3-XII-2007, K. Beucke, 1 (J-3) (KBPC); same but N31 52.571' W81 34.279', 15-24-V-2007, K. Beucke, 1 (K-1) (KBPC); same but N31 52.580' W81 34.259', 29-XI-3-XII-2007, K. Beucke, 1 (J-2) (KBPC); Hinesville, Camp Stewart, 9VII-1941, J.G. Watts, 1 (USNM); Peach Co. Fort Valley, 14-VI-1928, M.C. Swingle, 1 1 (USNM); Sumter Co. Americus, 1-V-1950, O.L. Cartwright, 1 1 (USNM); Thomas Co., 1.8 mi S of Metcalf, swine feces ba ited pitfall, 8-15-I-2006, R. Turnbow, 5 7 (FSCA); 2 mi S. of Metcalf on Metcalf Highwa y, 13-I-1995, Skelley and Kovarik, 1 (KPPC); Thomasville, no date or collector given, 1 (AMNH); same but Chas. Schaeffer, 1 1 (AMNH); same but 28-III-(year illegible), R.C. Casselberry, 1 (AMNH); same but 24-VIII-1938, P.W. Fattig, 1 1 (CNC), 1 (paratype) (FSCA), 2 (all paratypes), 4 (all paratypes) (USNM), 1 (paratype), 1 (paratype) (UMMZ); same but 28-VIII-1938, P.W. Fattig, 1 (paratype #8997) (CNC), 1 (paratype) (UMMZ); same but 3-III-1939, W.H. Thames, Jr., 2 (FSCA); Pinetree Blvd. and Lower Cairo Road, 0.3 mi W of W Thomasv ille Bypass, on north side of road in ditch and in adjacent pine plantation (burne d recently, with scant undergrowth), N30 49.693' W84 00.690', 15-III-2007, excavated from burrows, K. Beucke, 6 (H-1, H-2, H-3, H-4, H-5, H-6), 3 (H-7, H-8, H-9) (KBPC); same but pitfall trap, P. Choate, 1 (KBPC). M. gaigei No locality given, 1 (FSCA); USA: FLORIDA: "near Inverness," 28-III-1932, F.M.V and A.L.N., 1 (USNM); Alachua Co., no location given, 26-III-1959, H.V. Weems, Jr., 1 (AMNH); Archer, 21-23-II -1959, R.E. Woodruff, 1 1 (FSCA), 1 (NCSU); same but 28-III1960, R.E. Woodruff, 2 (AMNH); same but 30-III-1960, malt, Howden, 1 (FSCA), 1 1 (NCSU), 8 12 (CNC), 4, 2 (USNM); same but in malt tr ap, 8-IV-1960, R.E. Woodruff, 1 (FSCA); same but 22-III-1967, J. Mellott, USNM, 4 2 (USNM); same (Levy Co. on label) but 5-III-1977, unknow n collector, "NX 651," 1 (FSCA); W of Archer, N29 31.553' W82 32.100', 19-22-II-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 (G-1) (KBPC); 0.5 mi SW of Arch er, 21-II-1989, P. Skelley, 1 (FSCA); 2 mi W of Archer, 24-III1953, Howden and Dozier, 1 (paratype, #6526) (CNC); 2.2 mi SW of Archer, 25-III-1949, F.N. Young, 1 (paratype) (AMNH); Gainesville, 3-X-1996, P. Harpootlian, 1 (CU); High Springs, 31-III-1938, no collector given, 1 (paratype) (NCSU); same but 1-4-II-1960, R.E. Woodruff, 1 (FSCA); same but in malt trap, 4 5 (FSCA); CR-178, 0.4 mi S of CR-38, 27IV-1985, blacklight trap, K.W. Vick, 1 (FSCA); Columbia Co., no location given, 26-30-X1929, T.H. Hubbel, 1 (paratype) (TNHM); Ft. White, Pais ley Drive and Rt. 27 (just off 27), N29 54.816' W82 42.229', 23-24-I-2007, P. Choate, 5 (Q-4, Q-5), 14 (Q-1, Q-2, Q-3) (KBPC); Rt. 47, S of Ft. White, N29 52.886' W82 44.036', III-2007, pig dung and fermenting malt pitfall, P. Choate, 3 (L-1, L-2, L-3), 2 (L-4, L-5) (KBPC); High Springs, 26-X-1929, T.H. Hubbell, 1, 2 (paratypes) (FSCA), 2 1 (paratypes) (USNM); same but 26-30-X1929, T. Hubbell, 1 1 (all paratypes) (FSCA), 2 (paratypes) (USNM), 1 (paratype) 134

PAGE 135

(TNHM); 4 mi N of High Springs, 25-III-1953, Howden and Dozier, 2 (paratype #6526) (CNC); same but 19-III-1953, Howden and Dozier, 2 1 (paratypes) (FSCA), 3 (paratype) (USNM), 1 (paratype) (HMOU), 1 (paratype) (NCSU); same but 22-III-1953, malt trap, Howden and Dozier, 3, 2 (all paratypes) (FSCA), 1 (paratype) (USNM), 1 (paratype #6526) (CNC), 1, 1 (paratypes) (TNHM); same but 25-III-1953, Howden and Dozier, 3 (all paratypes) (FSCA), 4 (all paratypes) (USNM), 1 (paratype) (NCSU), 1 (paratype #6526) (CNC); Ichetucknee River, yeast and dung pitfal l trap, date illegible, L.R. Davis, Jr., 1 (FSCA); Ichetucknee River (?) ( illegible) on Rt. 27, 15-IX-1974, Lloyd R. Davis, Jr., 1 (FSCA); O'Leno State Park, on sandy road in sandhill vegetation, N29 55.216' W82 35.060', 20-22-X2006, pig dung and fermenting malt pitfall, K. Beucke, 1 (S-1) (KBPC); same but N29 55.065 W82 35.058, 18-21-I-2007, pig dung and fermenting malt pitfall, K. Beucke, 8 13 (KBPC); same but N29 55.222' W82 35.057', 18-21-I-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 1 (KBPC); same but N29 55.001' W82 35.057', 18-21-I-2007, pig dung and fermenting malt pitfall, K. Beucke, 2 (KBPC); Note: The following M. gaigei sequences are from the combined 18-21-I-2007 catch from the O'Leno sites: (P-7, P-8, P-9, P-10), (P-1, P2, P-3, P-4, P-5, P-6); Dixie Co., Old Town, edge of sandy road, N29 34.559' W82 58.069', 68-III-2007, pig dung and fermen ting malt pitfall, K. Beucke and P. Skelley, 2 (K-1, K-2) (KBPC); same but 8-14-III-2007, 5 (K-3, K-4, K-5, K-6), 4 (K-7, K-8, K-9) (KBPC); Gilchrist Co., 6.5 mi W of High Springs at US-27 in Rt. 340 (jct. Sarvis and Billy Brown Ave.), 10-15-XII-1998, Geomys burrow, P.S. Skelley, 1 (FSCA); on edge of SR-26, W of Newberry, sandhill vegetation with much exposed sand, N29 37.879' W82 42.313', 28-X-2-XI-2006, fermenting malt pitfall, K. Beucke, 1 (KBPC); same but 18-21-I-2007, pig dung and fermenting malt pitfall, K. Beucke, 17 (O-6, O-7, O-8, O-9, O-10), 19 (O-1, O-2, O-3, O-4, O-5) (KBPC); SR-26, 5 mi W of Newbe rry, 9-I-1984, pig dung trap, K.W. Vick, 3 2 (FSCA); same but 10-I-1984, K.W. Vick, 2 1 (FSCA); 6 mi W of Jc t. US-27 on Rt. 26, W of Newberry, N29 37.878' W82 42.281', 22-24-III-2007, P. Skelley, 8 1 1 larva (first instar?) obtained in rearing experiment (see Chapter 3) (FSCA); R 16E T 10S S 10, 30-III-1949, F.N. Young, 9 (all paratypes) (UMMZ); Lafayette Co., Mayo, Rt. 27 1.2 mi W of Rt. 51, N30 03.706' W83 11.427', I-2007, pig dung pitfall, P. Choate, 1 (A-1) (KBPC); same but 27-28-I2007, P. Choate, 2 (A-2), 6 (A-3, A-4) (KBPC); 4.5 mi E of Mayo, road to Convict Springs and CR-354, 1.2 mi N of Rt. 27, under live oak tree at corner of NE Rowan Rd and Convict Springs road, N30 04.295' W83 05.764', 19-22-II-2007, pig dung pitfall, P. Choate, 2 (I-1), 9 (I-2, I-3, I-4, I-5) (KBPC); 5 mi E of Ma yo on US-27, 13-18-II-1960, R.E. Woodruff and H.V. Weems, Jr., 9 6 (FSCA); same but 16-20-V-1960, R.E. Woodruff, 19 17 (FSCA); same but in pure pine stand, 16-20-V-1960, malt trap, R.E Woodruff, 4 (FSCA); Townsend, Rt. 348 off Rt. 27, N30 08.964' W83 19.719', 27-III-2007, pig dung and fermenting malt pitfall, P. Choate, 2 (M-1, M-2) (KBPC); CR-4 25, 0.6 mi N of Rt. 27, N29 59.431' W82 59.744', 23-24-I-2007, P. Choate, 4 (R-4, R-5), 8 (R-1, R-2, R-3) (KBPC); same but 25-I2007, pig dung pitfall, P. Choate, 3 7 (KBPC); same but 27-28-I-2007, P. Choate, 11 (B-6, B-7, B-8, B-9, B-10), 6 (B-1, B-2, B-3, B-4, B-5) (KBPC); Levy Co., no locality given, 31-V1956, R.A. Morse, 1 (NCSU), 1, 4 (FSCA); same but 19-21-II-1959, R.E. Woodruff, 1 (FSCA); same but in 21-23-II-1959, malt trap, R.E. Woodruff, 1 5 (FSCA), 2 (USNM); Alachua/Levy County line, 2325-II-1959, R.E. Woodruff, 2 (FSCA); Alachua/Levy County line, 1 mi N of SR-24, 19-26-II-1983, yeast a nd dung traps, M.C. Thomas and T. Zoebisch, 1 1 (FSCA); "Area 3" (believed to be on the Al achua/Levy County line, R.E. Woodruff, pers. comm.), 23-25-II-1959, R.E. Woodruff, 4 6 (CNC); 0.2 mi W of Alachua/Levy County line 135

PAGE 136

on SR-24, N29 30.605' W82 33.598', 7-9-X-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 (KBPC); close to previous site, N29 30.530' W082 33.735', 24-II-4-III-2007, pig dung and fermenting malt pitfall, K. Beucke, 2 (J-1, J-2) (KBPC); Rt. 24 at Alachua/Levy County line, 24-26-III-1978, J.D. Glaser, 1 (FSCA); W of Archer, 1-III-1987, malt trap, P. Skelley, 1 (FSCA); 2 mi W of Archer, 23III-1953, Howden and Dozier, 1 (paratype) (FSCA), 1, 2 (all paratypes) (USNM); same but 24-III-1953, malt and propionic acid trap, 3, 4 (all paratypes) (FSCA), 2 2 (AMNH) (all paratypes), 2 (paratypes) (TNHM), 1 (HMOU), 4 3 (paratypes) (USNM); 3 mi W of Ar cher, Shirley Ct., 10-VI-1986, Cicero, 1 (FSCA); same but 10-III-1987, 4 1 (FSCA); 3.8 mi SW of Ar cher, 19-22-III-1987,malt/dung pitfall, P. Skelley, 5 3 (FSCA); same but 13-16-X-1988, baited pitfall, 1 (FSCA); same but 16-II-1988, pitfall, 1 (FSCA); same but 15-20-I-1989, 1 1 (FSCA); 4 mi W of Archer, 22III-1953, malt trap, Howden and Dozier, 1 (paratype #6526) (CNC), 1 (paratype) (USNM); 4 mi W of Archer on Rt. 24, 29-III-1991, dung traps, P.E. Skelley and R.E. Woodruff, 123 101 (FSCA); Bronson, 19-II-1959, malt trap, R.E. Woodruff, 5 3 (USNM); same but 21-23-II1959, R.E. Woodruff, 3 4 (FSCA); same but 23-25-II-1959, R.E. Woodruff, 1 2 (USNM); 8 mi E of Bronson on Rt 27, 24-26-III-1978, J.D. Glaser, 1 (FSCA); Meredith, 23II-1959, malt traps, R.E. Woodruff, 3 12 (FSCA), 6, 5 (AMNH), 2 2 (USNM); Oak Ridge Estates, 4-10-XII-1990, malt trap, R. Morris, 7 7 (FSCA); T11S R17E sec 24/25, 25III-1976, pitfall traps, L.R. Davis, Jr., 17 13 (USNM); Madison Co., On CR-53, N of county line, N30 16.459' W83 17.320', 9-II-2007, pig dung pitfall, P. Choate, 1 (D-1) (KBPC); same but 10-II-2007, P. Choate, 5 (D-2, D-3, D-4) (KBPC); same but III-2007, P. Choate, 1 (D-5) (KBPC); Marion Co., Barge Canal Surv. (?), T17S R21E 4 (east central), 4-IV-1975, herp. pitfall (H-7), S. Christman, 1 1 (FSCA, in alcohol); Marion Oa ks Manor and SW 56th Court, N28.9962 W82.2129, 9-VII-2007, pig dung and fermen ting malt pitfall, K. Beucke, 1 (KBPC); Village of Rainbow Springs, 5-V-1982, M.C. Thomas, 4 7 (FSCA); same but 26III-1989, M.C. Thomas, 1 (FSCA); Summerfield, La Casta Estates, 14641 SE 1st Ave. Road, N29 00.618' W82 08.058', V-VI2005, cat food, C.E. Taylor, 3 5 (FSCA); same but 6-9VIII-2006, pig dung and fermenting malt pitfall, K. Beucke, 3 3 (E-9, E-10) (KBPC); same but 19-22-II-2007, K. Beucke, 6 (E-1, E-2, E-3, E-4, E-5), 4 (E-6, E-7, E-8) (KBPC); La Casta Subdivision, I-75 at St. Rd. 275A, 12-II-1992 (or 1995), on Opuntia R.E. Woodruff, 1 (FSCA); SW 59th Avenue Road, in pow er line cut near SW 158th Lane, N28.9920 W82.2197 9-VII-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 (KBPC); 76th Court, off of 484 East, N29 01.382' W82 14.770', 19-22-II-2007, pig dung and fe rmenting malt pitfall, K. Beucke, 2 (F-1, F-2), 2 (F-3, F-4) (KBPC); Seminole Co., Geneva, 20-IV-1960, R.E. Woodruff, 19 13 (FSCA); same but 20-21-IV-1960, in malt trap, R.E. Woodruff, 9 8 (FSCA); same but VI-1976, M. Thomas, 2 (CNC), 10 8 (USNM); same but in turkey oak, 13-20-III-1976, yeast trap, M.C. Thomas, 28 13 (FSCA); Geneva, sand pit on Cochran Road, W of SR-46, NW edge of pit in pine woods, N28 44.754 W81 07.801, 6-9-VIII-2006, fermenting malt pitfall, K. Beucke, 1 (KBPC); same but 27-III-3-IV-2007, pig dung and fermenting malt pitfall, K. Beucke, 3, 4 (N-1) (KBPC); same but 25-29-V-2007, K. Beucke, 1 (N-2) (KBPC); Geneva, on side of Ridge Road near SR-46, N28 44.817' W81 07.772', 29VIII-1-IX-2006, fermenting malt pitfall, K. Beucke, 1 (KBPC); Sumter Co., Tillman Hammock on CR-425, N28 57.218' W82 08.001', 19-22-II-2007, pig dung and fermenting malt pitfall, K. Beucke, 1 (H-1) (KBPC); Suwannee Co., Hildreth, 2.9 mi N of Rt. 27 on road to Ichetucknee Baptist Church, N29 59.626' W82 48.633', 27-28-I-2007, pig dung pitfall, P. Choate, 19 (C-6, C-7, C-8, C-9, C-10), 22 (C-1, C-2, C-3, C-4, C-5) (KBPC); 3.6 mi N of 136

PAGE 137

Obrien, 17-III-1956, malt, Howden, 1 (FSCA); on Rt. 349, 6.9 mi S. of Jct. with Rt. 252, T5S R13E section 6, 15-XII-1980, on carrion opossum, L. Davis, Jr., 1 (FSCA, in alcohol). Cryptic Mycotrupes species? USA: FLORIDA: Taylor Co. Foley, 30-III-1938, "ex: NCDA&CS 2000," 1 (NCSU); CR-356, 100 yards from Rt-27, mixed pine woods, N30 04.974' W83 30.476', 17-20-III-2007; pig dung pitfall, P. Choate, 1 (I-1) (KBPC); same but IV-2007, P. Choate, 1 1 (I-2) (KBPC). 137

PAGE 138

APPENDIX C: MYCOTRUPES LOCALITY RECORDS FROM WHIC H NO SPECIMENS WERE STUDIED M. lethroides USA: GEORGIA: Burke Co. 7 miles northwest of Girard on SR-23 (Harpootlian 1995); Jefferson Co., near Wrens on US Hwy. 1 (Harpootlian pers. comm.); Richmond Co. 0.5 mi east of the Richmond/Jefferson County line (B rier Creek) on US Hwy. 1 (Harpootlian 1995). M. retusus USA: SOUTH CAROLINA: Calhoun/Orangeburg County line on Interstate 26, (Harpootlian 2001); Aiken Co. Gopher Tortoise Herita ge Preserve (Harpootlian 2006); Aiken, Hitchcock Woods (Harpootlian 2006); picnic area 17.5 miles north of Ai ken (Olson and Hubbell 1954). M. cartwrighti USA: FLORIDA: Duval Co., Jacksonville (Olson and Hubbell 1954); Leon Co., 6.5 mi east of Tallahassee, north of U.S. Hi ghway 90 (Olson and Hubbell 1954); GEORGIA: Dooley Co., U.S. Highway 41 at Pennahatchee Creek, 2 miles north of Vienna (probably same as specimen on loan from UMMZ) (Olson and Hubbell 1954). M. gaigei USA: FLORIDA: Alachua Co., Warren's Cave, about 8 miles NW of Gainesville, in pineland near entrance (Olson and Hubbell 1954); Citrus Co. (Olson and Hubbell 1954); Levy Co., Andrews Wildlife Management Area, near "Zone C," xeric hammock dominated by Quercus geminata (D. Almquist pers. comm.); 5 miles SW of Archer on State Highway 13 (Olson and Hubbell 1954); Marion Co. Dunnellon (Olson and Hubbell 1954); Seminole Co. Oviedo, SR434 6.5 mi N of University of Central Florida (S. Fullerton, pers. comm.). 138

PAGE 139

LITERATURE CITED Alt, D., and H.K. Brooks. 1965. Age of the Florida marine terraces. Journal of Geology. 73: 406-411. Arbogast, B.S., S.V. Edwards, J. Wakely, P. Beerli, and J.B. Slowinski. 2002. Estimating divergence times from molecular data of phyl ogenetic and population genetic timescales. Annual Review of Ecology and Systematics 33:707-740. Arrow, G.J. 1904 Sound production in the lamellicorn beetles. Transactions of the Entomological Society of London 52:709-750. Arrow, G.J. 1942. The origin of stridulation in beet les. Proceedings of the Royal Entomological Society of London 17: 83-86. Avery, L. 2008. Mann-Whitney U Test. Available from: http://elegans.swmed.edu/~leon/stat s/utest.html (Accessed on 12/8/2008) Avise, J.C., J. Arnold, R.M. Ball, E. Bermingham, T. Lamb, J.E. Neigel, C.A. Reeb, and N.C. Saunders. 1987. Intraspecific phylogeography: The mitochondrial DNA bridge between population genetics and systematics. Annual Review of Ecology and Systematics 18:489-522. Bauer, V.T. 1976. Experimente zur frage der biologischen bedeutung des stridulationsverhalte ns von kafern. Zeitschrift fur Tierpsychologie 42:57-65. Beucke, K., and P. Choate. 2009. Notes on the feeding behavior of Mycotrupes lethroides (Westwood) (Coleoptera: Geotrupidae), a flightle ss North American beetle. The Coleopterists Bulletin 63(2):228-229. Blanchard, F. 1888. Some account of our species of Geotrupes. Psyche 5:103-109. Blatchley, W.S. 1928. The Scarabaeidae of Florida. Florida Entomologist 12:28-30, 44-46. Boucomont, A. 1902. Coleoptera, Lamellicornia, fam. Geotrupidae. In: P. Wytsman (editor), Genera insectorum, Fascicule 7. P. Wytsman, Brussels, Belgium. 20 pp. Boucomont, A. 1911. Contribution a la classification des Ge otrypidae. Annales de la Societe Entomologique de France 79:333-350. Boucomont, A. 1912. Scarabaeidae: Taurocerastinae, Geotr upinae. Coleopterorum Catalogus 46:1-47. Bouma, J., R.B. Brown, and P.S.C. Rao. 1982. Basics of soil-water relationships. Part II, retention of water. Soil Sc ience Factsheet SL-38. Florid a Cooperative Extension Service. Brimley, C.S. 1938. The insects of North Carolina. North Carolina Department of Agriculture, Raleigh, NC. 560 pp. 139

PAGE 140

Brinkmann, R., S. Koenig, and K. Pace-Graczyk. 2007. Introduction to the karst issues in west central Florida. Available from: http://www.karstportal.org/public/_files/ catalog/BOK_FieldGuide_2007.pdf (Accessed on 5/11/2009) Brooks, H.K. 1981. Physiographic Divisions of Flor ida (map). Institute of Food and Agricultural Sciences, Gainesville, FL. Brower, A.V.Z. 1994. Rapid morphological radiation a nd convergence among races of the butterfly Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proceedings of the National Academy of Sciences 91:6491-6495. Buchler, E.R., T.B. Wright, and E.D. Brown. 1981. On the functions of stridulation by the passalid beetle Odontotaenius disjunctus (Coleoptera: Passalidae). Animal Behavior 29:483486. Carisio, L., C. Palestrini, and A. Rolando. 2004. Stridulation variability and morphology: an examination in dung beetles of the genus Trypocopris (Coleoptera, Geotrupidae). Population Ecology 46(1):27-37. Colquhoun, D.J., G.H. Johnson, P.C. Peebles, P.F. Huddlestun, and T. Scott. 1991. Quaternary Geology of the Atlantic Coastal Pl ain [pp. 621-650]. In: R.B. Morrison (editor), Quaternary nonglacial geology: co nterminous U.S., v. K-2. Geol ogical Society of America, Boulder, CO. 672 pp. Cooke, C.W. 1945 Geology of Florida. Bulletin of the Florida Geological Society 29:1-339. Crisp, M.D., and G.T. Chandler. 1996. Paraphyletic species. Telopea 6:813-844 Drummond, A.J., and A. Rambaut. 2006. BEAST v1.4.8 Available from: http://beast.bio.ed.ac.uk (Accessed on 2/26/2009) Drummond, A.J., and A. Rambaut. 2007. Tracer v1.4. Available from: http://beast.bio.ed.ac.uk (Accessed on 2/26/2009) Dumortier, B. 1963. Ethological and physiological study of sound emissions in Arthropoda [pp. 277-345]. In: R.G. Busnel (editor), Acoustic Behavior of Animals. Elsevier Publishing Company, New York, NY. 933 pp. Elith, J., C.H. Graham, R.P. Anderson, M. Dud k, S. Ferrier, A. Guisan, R.J. Hijmans, F. Huettmann, J.R. Leathwick, A. Lehmann, J. Li L.G. Lohmann, B.A. Loiselle, G. Manion, C. Moritz, M. Nakamura, Y. Nakazawa, J.M. Overton, A.T. Peterson, S.J. Phillips, K. Richardson, R. Scachetti-Pereira, R.E. Schapi re, J. Sobern, S. Williams, M.S. Wisz, and N.E. Zimmermann. 2006. Novel methods improve prediction of species' distributions from occurrence data. Ecography 29:129-151. 140

PAGE 141

Enge, K.M., M.M. Bentzien, and H.F. Percival. 1986. Florida scrub lizard status survey. Technical Report No. 26. Florida Cooperative Fi sheries and Wildlife Research Unit, University of Florida, Gainesville, FL. 72 pp. Erixon, P., B. Svennblad, T. Britton, and B. Oxelman. 2003. Reliability of Bayesian posterior probabilities and bootstrap frequencies in phylogenetic s. Systematic Biology. 52:665673. Ewing, A.W. 1989. Arthropod Bioacoustics: Neurobiology and Behavior. Cornell University Press, Ithaca, NY. 260 pp. Felsche, C. 1909. Neue und alte coprophage scarabaeiden (Col.). Deutsche Entomologische Zeitschrift 1909:751-765. Felsche, C. 1910. Uber coprophage scarabaeiden. Deutsc he Entomologische Zeitschrift 1911:339-352. Florida Natural Area Inventory. 2009. FNAI Tracking List. Available from: http://www.fnai.org/ranks.cfm (Accessed on 6/21/2009) Funk, D.J., and K.E. Omland. 2003. Species-level paraphyly and polyphyly: frequency, causes, and consequences, with insights from animal mitochondrial DNA. Annual Review of Ecology, Evolution, and Systematics 34:397-423. Grebennikov, V.V., and C.H. Scholtz. 2004. The basal phylogeny of Scarabaeoidea (Insecta: Coleoptera) inferred from larval morphol ogy. Invertebrate Systematics 18:321-348. Gene Codes. 1998. Sequencher v3.1.1. Gene Codes, Ann Arbor, Michigan. Google. 2006. Google Earth. Version 4.0.2416. Ava ilable from: http://earth.google.com (Accessed on 10/1/2009) Harper, R.M. 1914. Geography and vegetation of northern Florida. Annual Report of the Florida State Geological Survey 6:163-437. Harper, R.M. 1921. Geography of central Florida. A nnual Report of the Florida State Geological Survey 13:71-307. Harper, R.M. 1927. Natural resources of s outhern Florida. Annual Re port of the Florida State Geological Survey 18:27-206. Harpootlian, P.J. 1995. Notes and records of Scarabaeidae from the Southeastern United States. Coleopterists Bulletin 49(3):280. Harpootlian, P.J. 2001. Scarab beetles (Coleoptera: Scarabaei dae) of South Carolina. Biota of South Carolina, Volume 2. Clemson University, Clemson, SC. 157 pp. 141

PAGE 142

Harpootlian, P.J. 2006. Sandhills earth boring scarab beetle. Available from: http://www.dnr.sc.gov/cwcs /pdf/Scarabbeetle.pdf (Accessed on 9/28/2007) Healy, H.G. 1975. Terraces and shorelines of Florida (m ap). Florida Bureau of Geology map series 71. Florida Bureau of Geology, Tallahassee, FL. Hebard, M. 1903. Notes on the collecting around Thomasville, Georgia. Entomological News 14(8):260-261. Hebert, P.D.N., S. Ratnasingham, and J.R. deWaard. 2003. Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related species. Proceedings of the Royal Society of London B (Supplement) 270:S96-S99. Henderson, J.R. 1939. The soils of Florida. University of Florida Agricultural Experiment Station Bulletin 334: 1-67. Hernandez, P.A., C.H. Graham, L.L. Master, and D.L. Albert. 2006. The effect of sample size and species characteristics on performan ce of different species distribution modeling methods. Ecography 29:773-785. Highton, R. 1956. Systematics and variation of the endemic Florida snake genus Stilosoma. Bulletin of the Florida State Museum 1:73-96. Hirschberger, P. 2001. Stridulation in Aphodius dung beetles: behavioral context and intraspecific variability of song patterns in Aphodius ater (Scarabaeidae). Journal of Insect Behavior 14(1):69-88. Ho, S.Y.W. 2007. Calibrating molecular estimates of subs titution rates and divergence times in birds. Journal of Avian Biology 38:409-414. Horn, G.E. 1868. Geotrupes of boreal America. Transactions of the American Entomological Society 1:313-322. Howden, H.F. 1952. A new name for Geotrupes ( Peltotrupes) chalybaeus LeConte, with a description of the larva a nd its biology. The Coleopterists Bulletin 6(3):41-48. Howden, H.F. 1954. Habits and life history of Mycotrupes, with a description of the larva of Mycotrupes gaigei [pp. 52-56]. In: A.L. Olson, T.H. Hubbell, and H.F. Howden, The burrowing beetles of the genus Mycotrupe s. Miscellaneous publications of the Museum of Zoology, University of Michigan, no. 84. University of Michigan Press, Ann Arbor, MI. 59 pp. Howden, H.F. 1955. Biology and taxonomy of North Amer ican beetles of the subfamily Geotrupinae with revisions of the genera Bolbocerasoma Eucanthus, Geotrupes and Peltotrupes (Scarabaeidae). Proc eedings of the United States National Museum 104(3342):151319. 142

PAGE 143

Howden, H.F. 1963. Speculations on some beetles, ba rriers, and climates during the Pleistocene and pre-Pleistocene periods in some non-glaciated portions of North America. Systematic Zoology 12(4):178-201. Howden, H.F. 1964. The Geotrupinae of North and Ce ntral America. Memoires of the Entomological Society of Canada. 39: 1-91. Howden, H., A. Howden, and G. Holloway. 2007. Digging down under: Australian Bolboceratini, their habits and a list of speci es (Coleoptera: Scarabaeoidea: Geotrupidae). Zootaxa 1499:47-59. Hubbell, T.H. 1954. Relationships and distribution of Mycotrupes [pp. 39-51]. In: A.L. Olson, T.H. Hubbell, and H.F. Howden, The Burrowing Beetles of the genus Mycotrupes. Miscellaneous publications of the Museum of Zoology, University of Michigan, no. 84. University of Michigan Pr ess, Ann Arbor, MI. 59 pp. Hubbell, T.H. 1956. Some aspects of geographic variati on in insects. Annual Review of Entomology 1:71-88. Jameson, M.L. 2002. Geotrupidae [pp. 23-27]. In: R.H. Arne tt, Jr, M.C. Thomas, P.E. Skelley, and J.H. Frank (editors), American beetles, Volume II: Polyphaga: Scarabaeoidea Curculionidae. CRC Press, Boca Raton, FL. 861 pp. Kautz, R., B. Stys, and R. Kawula. 2007. Florida vegetation 2003 and land use change between 1985-89 and 2003. Fl orida Scientist 70:12-23. Kovarik, P.W., P.E. Skelley, K.A. Milla, R.H. Turnbow, and D.L. Capers. 2000. Distribution patterns of insects i nhabiting burrows of the Southeaste rn Pocket Gopher. Available from: http://www.famu.org/gopher/in tro.html (Accessed on 10/9/2009) Krantz, G.W., and J.L. Mellott. 1968. Two new species of Macrocheles (Acarina: Macrochelidae) from Florida, with notes on thei r host-specific relations hips with geotrupine beetles (Scarabaeidae: Geotrupinae). Journal of the Kansas Entomological Society 41(1):48-56. Krantz, G.W., and L.A. Royce. 1994. Observations on the biology and behavior of Macrocheles mycotrupetes Krantz and Mellott (Acari: Macrochelid ae). International Journal of Acarology 20(2):115-121. Krantz, G.W., L.A. Royce, and R.R. Lowry. 1991. Mechanisms of phoretic specificity in Macrocheles (Acari; Macrochelidae). Modern Acarology 2:561-569. Kurz, H. 1942. Florida dunes and scrub, vegetation and geology. Florida Geological Survey Bulletin. 23:1-154. Laessle, A.M. 1958. The origin and successional relati onship of sandhill vegetation and sandpine scrub. Ecological Monographs 28(4):361-387. 143

PAGE 144

Lamb, T., and T.C. Justice. 2005. Comparative phylogeography of Florida scrub insects: implications for systematics, biogeography, and c onservation. Final report. Florida Fish and Wildlife Conservation Commission, Tallahassee, FL. v + 61 pp. LeConte, J.L. 1866. Additions to the Coleopterous fauna of the United States. Proceedings of the Academy of Natural Scien ces of Philadelphia 18:361-394. Leng, C.W. 1920. Catalogue of the Coleopt era of America, north of Mexico. John D. Sherman, Jr., Mount Vernon, NY. 470 pp. Light, J.E., and D.L. Reed. 2009. Multigene analysis of phylogenetic relationships and divergence times of primate sucking lice (Pthir aptera: Anoplura). Molecular Phylogenetics and Evolution 50:376-390. Lobo, J.M., J.R. Verd, and C. Numa. 2006. Environmental and geographical factors affecting the Iberian distribution of flightless Jekelius species (Coleopter a: Geotrupidae). Diversity and Distributions 12:179-188. MacNeil, F.J. 1949. Pleistocene shore lines in Florida and Georgia. United States Geological Survey Professional Paper 221-F:95-107. Maddison, W.P., M.J. Donoghue, and D.R. Maddison. 1984. Outgroup analysis and parsimony. Systematic Zoology 33(1):83-103. Mail, G.A. 1930. Winter soil temperatures and their rela tion to subterranean insect survival. Journal of Agricultural Research 41:571-592. Marshall, S.A., and D.J.S. Montagnes. 1988. Ceroptera longicauda, a second North American species in the kleptoparasitic genus Ceroptera Macquart (Diptera: Sphaeroceridae). Proceedings of the Entomological Society of Washington 90(2):189-192. Masters, M. 1979. Insect disturbance stridulation: its de fensive role. Beha vioral Ecology and Sociobiology 5:187-200. Matheny, P.B., M.C. Aime, N.L. Bougher, B. Buyck, D.E. Desjardin, E. Horak, B.R. Kropp, D.J. Lodge, K. Soytong, J.M. Trappe, and D.S. Hibbett. 2009. Out of the Palaeotropics? Historical bioge ography and diversification of the cosmopolitan ectomycorrhizal mushroom family Inocybaceae. Journal of Biogeography 36:577-592. Merriam-Webster. 2009. Merriam-Webster's Collegiate Dictionary, Eleventh Edition. Available from: http://www.merriam-w ebster.com (Accessed on 10/8/2009) Miller, K.G., M.A. Kominz, J.V. Browning, J.D. Wright, G.S. Mountain, M.E. Katz, P.J. Sugarman, B.S. Cramer, N. Christie-Blick, and S.F. Pekar. 2005. The Phanerozoic record of global sea-level change. Science 310:1293-1298. Mishler, B.D., and E.C. Theriot. 2000. The phylogenetic species concept ( sensu Mishler and Theriot): Monophyly, Apomorphy, and Phylogenetic Species Concepts [44-54]. In: Q.D. 144

PAGE 145

Wheeler and R. Meier (editors), Species concepts and phylogenetic theory. Columbia University Press, New York, NY. 230 pp. Mitchell, A.J., and M.R. Ensign. 1928. The climate of Florida. Bulletin of the Florida Agricultural Experiment Station 200:91-300. Myers, R.L. 1985. Fire and the dynamic relationship between Florida sandhill and sand pine scrub vegetation. Bulletin of th e Torrey Botanical Club 112:241-252. NatureServe. 2009a. Mycotrupes gaigei Olson and Hubbell, 1954. Available from: http://www.natureserve.org/ex plorer (Accessed on 6/21/2009) NatureServe. 2009b. Mycotrupes pedester Olson and Hubbell, 1954. Available from: http://www.natureserve.org/ex plorer (Accessed on 6/21/2009) Neill, W.T. 1957. Historical biogeography of present-day Florida. Bulletin of the Florida State Museum. 2:175-220. Nixon, K.C., and J.M. Carpenter. 1993. On outgroups. Cladistics 9:413-426. Olson, A.L., and T.H. Hubbell. 1954. A revision of the genus Mycotrupes [pp. 7-38]. In: A.L. Olson, T.H. Hubbell, and H.F. Howden, The burrowing beetles of the genus Mycotrupes Miscellaneous publications of the Museum of Zoology, University of Michigan, no. 84. University of Michigan Pr ess, Ann Arbor, MI. 59 pp. Olson, A.L., T.H. Hubbell, and H.F. Howden. 1954. The burrowing beetles of the genus Mycotrupes. Miscellaneous publications of the Museum of Zoology, University of Michigan, no. 84. University of Michigan Press, Ann Arbor, MI. 59 pp. Opdyke, N.D., D.P. Spangler, D.L. Smith, D.S. Jones, and R.C. Lindquist. 1984. Origin of the epeirogenic uplift of Pliocene-Pleistocene be ach ridges in Florida and development of the Florida karst. Geology 12(4):226-228. Palestrini, C., and G. Pavan. 1995. The stridulation in Thorectes intermedius (Costa, 1827) (Coleoptera: Geotrupida e). Elytron 9:115-124. Palestrini, C., R. Piazza, and M. Zunino. 1988. Segnali sonori in tre sp ecie di Geotrupini. Bolletino della Societ Entomologi ca Italiana 119(3):139-151. Peck, S.B., and M.C. Thomas. 1998. A distributional checklist of the beetles (Coleoptera) of Florida. Arthropods of Florid a and Neighboring Land Areas 16:1-180. Prez-Ramos, I. M., T. Maraon, J. M. Lobo, and J. R. Verd. 2007. Acorn removal and dispersal by the dung beetle Thorectes lusitanicus : ecological implications. Ecological Entomology 32:349-356. Phillips, S.J., R.P. Anderson, and R.E. Schapire. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modeling 190:231-259. 145

PAGE 146

Posada, D., and K.A. Crandall. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818. R Development Core Team. 2008. R: a language and environmen t for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Rambaut, A. 2008. FigTree v1.1.2. Available from: http://b east.bio.ed.ac.uk (Accessed on 2/26/2009) Rambaut, A., and A.J. Drummond. 2008a. BEAUTi v1.4.8 Available from: http://beast.bio.ed.ac.uk (Accessed on 2/26/2009) Rambaut, A., and A.J. Drummond. 2008b. LogCombiner v1.4.8 Available from: http://beast.bio.ed.ac.uk (Accessed on 2/26/2009) Rambaut, A., and A.J. Drummond. 2008c. TreeAnnotator v1.4.8 Available from: http://beast.bio.ed.ac.uk (Accessed on 2/26/2009) Rand, D.B., A. Heath, T. Suderman, and N.E. Pierce. 2000. Phylogeny and life history evolution of the genus Chrysoritis within the Aphnaeini (Lepi doptera: Lycaenidae), inferred from mitochondrial cytochrome oxi dase I sequences. Molecular Phylogenetics and Evolution. 17:85-96. Randazzo, A.F., and D.S. Jones. 1997. The geology of Florida. University Press of Florida, Gainesville, FL. 327 pp. Renner, S.S. 2005. Relaxed molecular clocks for dating hi storical plant dispersal events. TRENDS in Plant Science 10(11):550-558. Schapire, R. 2008. Maxent software for species habitat modeling. Available from: http://www.cs.princeton.edu/~schapir e/maxent (Accessed on 8/27/2008) Schuchert, C. 1910. Paleogeography of North America. Geological Society of America Bulletin. 20: 427-606. Shoo, L.P., R. Rose, P. Doughty, J.J. Austin, and J. Melville. 2008. Diversification patterns of pebble-mimic dragons are consistent with hist orical disruption of impo rtant habitat corridors in arid Australia. Molecular Phyl ogenetics and Evolution 48: 528-542. Sivinski, J., S. Marshall, and E. Petersson. 1999. Kleptoparasitism and phoresy in the Diptera. The Florida En tomologist 82(2):179-197. Swofford, D.L. 2002. PAUP. Phylogenetic analysis using parsimony. Version 4.0b10 (Altivec). Sinauer Associates, Sunderland, MA. Thodorids, J. 1952. Les parasites et commensaux des Geotrupini (Coleoptera Scarabaeidae Geotrupinae). Rivista di Parassitologia 13(4):277-298. 146

PAGE 147

Trapnell, D.W., J.P. Schmidt, P.F. Quin tana-Ascencio, and J.L. Hamrick. 2007. Genetic insights into the biogeogra phy of the southeastern No rth American endemic, Ceratiola ericoides (Empetraceae). The Journal of Heredity 98(6):587-593. USDA. 2008. Soil Data Mart. Available from: http ://soildatamart.nrcs.u sda.gov (Accessed on 8/27/2008) USDA. 2009. Soil Service Classification Database. Available from: http://soils.usda.gov/technical/classification/ scfile/index.html (Accessed on 10/10/2009) USGS. 2002. Subsurface characterization of selected water bodies in the St. Johns River Water Management District, Northeas t Florida. Available from: http://coastal.er.usgs.gov/publications/ofr/ 00-180/intro/regional-geology.html (Accessed on 6/21/2009) USGS. 2008. Seamless Server. Available from: http://seamless.usgs.gov (Accessed on 8/27/2008) Vaughan, T.W. 1910. A contribution to the geologic history of the Floridian plateau. Papers from the Marine Biological Labor atory at Tortugas. 4:99-185. Verd, J.R., E. Galante, J.P. Lumaret, and F.J. Cabrero-Saudo. 2004. Phylogenetic analysis of Geotrupidae (Coleoptera, Scarabaeoidea) based on la rvae. Systematic Entomology 29:509-523. Verd, J.R., J. M. Lobo, C. Numa, I. Prez-Ramos, E. Galante, and T. Maraon. 2007. Acorn preference by the dung beetle, Thorectes lusitanicus under laboratory and field conditions. Animal Be havior 74:1697-1704. Watts, W.A. 1975. A late Quaternary record of vegeta tion from Lake Annie, South-Central Florida. Geology 3:344-346. Weir, J.T., and D. Schluter. 2008. Calibrating the avian molecular clock. Molecular Ecology 17:2321-2328. Wheeler, Q.D., and R. Meier (editors). 2000. Species concepts and phylogenetic theory. Columbia University Press, New York, NY. 230 pp. Wheeler, Q.D., and N. I. Platnick. 2000. The phylogenetic species concept ( sensu Wheeler and Platnick) [pp. 55-69]. In: Q.D. Wheeler and R. Meier (editors), Species concepts and phylogenetic theory. Columbia University Press, New York, NY. 230 pp. Wessel, A. 2006. Stridulation in the Coleoptera an overview [pp. 397-430]. In: S. Drosopoulos, and M.F. Claridge (editors), Ins ect sounds and communication. CRC Press, Boca Raton, FL. 552 pp. Westwood, J.O. 1837. Descriptions of some new species of exotic coleopterous insects from the collection of Sir Patrick Walker. Magazine of Zoology and Botany 1:251-257. 147

PAGE 148

Whitney, E. N., D. B. Means, and A. Rudloe. 2004. Priceless Florida: Natural Ecosystems and Native Species. Pineapple Pr ess Inc., Sarasota, FL. 423 pp. Winking-Nikolay, V.A. 1975. Untersuchungen zur bio-akus tik des waldmistkafers, Geotrupes stercorosus Scriba. Zeitschrift fur Tierpsychologie 37:515-541. Woodruff, R.E. 1973. The scarab beetles of Florida. Part I. The Laparosticti. Arthropods of Florida and Neighborin g Land Areas 8:1-220. Woodruff, R.E., and M. Deyrup. 1994a. Cartwrights Mycotrupes [pp. 384-386]. In: M. Deyrup, and R. Franz (editors), Rare and endangered biota of Florida, Volume 4: Invertebrates. University Press of Florida, Gainesville, FL. xxx+798 pp. Woodruff, R.E., and M. Deyrup. 1994b. North peninsular Mycotrupes [pp. 386-388]. In: M. Deyrup, and R. Franz (editors), Rare and endangered biota of Florida, Volume 4: Invertebrates. University Press of Florida, Gainesville, FL. xxx+798 pp. Woodruff, R.E., and M. Deyrup. 1994c. Southwest Florida Mycotrupes [pp. 370-371]. In: M. Deyrup, and R. Franz (editors), Rare and endangered biota of Florida, Volume 4: Invertebrates. University Press of Florida, Gainesville, FL. xxx+798 pp. WorldClim. 2006. WorldClim. Available from: http://www.w orldclim.org (Accessed on 8/27/2008) Wuster, W., L. Peppin, C.E. Pook, and D.E. Walker. 2008. A nesting of vipers: phylogeny and historical biogeography of the Viperidae (Squ amata: Serpentes). Molecular Phylogenetics and Evolution 49:445-459. Young, F.N. 1950. Notes on the habits and habitat of Geotrupes chalybaeus LeConte in Florida. Psyche 57:88-92. Young, F.N., T.H. Hubbell, and D.W. Hayne. 1955. Further notes on the habits of Geotrupes (Coleoptera: Geotrupida e). Psyche 62: 53-54. Zhang, P., T.J. Papenfuss, M.H. Wake, L. Qu, and D.B. Wake. 2008. Phylogeny and biogeography of the family Salamandridae (A mphibia: Caudata) inferred from complete mitochondrial genomes. Molecular P hylogenetics and Evolution 49:586-597. Zuckerkandl, E. and L. Pauling. 1965. Evolutionary divergence and convergence in proteins [pp. 97-166]. In: V. Bryson and H.J. Vogel (edito rs), Evolving genes and proteins. Academic Press, New York, NY. xxi + 629 pp. Zunino, M. 1984. Sistematica generica dei Geotrupinae (Coleoptera, Scarabaeoidea: Geotrupidae), filogenesi della sottofamiglia e considerazio ni biogeografiche. Bolletino del Museo Regionale di Scienze Naturali, Torino 2(1):9-162. 148

PAGE 149

Zunino, M., and S. Ferrero. 1988. El dimorfismo sexual de los aparatos estriduladores en el genero Megatrupes (Coleoptera: Scarabaeoidea, Geotrupi dae). Folia Entomlogica Mexicana 76:65-72. 149

PAGE 150

BIOGRAPHICAL SKETCH Kyle August Beucke was born in Annaheim, CA, in 1979. He began his undergraduate studies as a freshman at the University of Ariz ona, where he was hopelessly infected with a love of the desert Southwest. He transferred to Co rnell University his sophomore year and earned a B.S. in entomology. After graduating, Beucke worked at the American Museum of Natural History in New York City for three years. In the fall of 2009, he receiv ed his Ph.D. from the University of Florida. 150