Citation
Biology, Ecology, Behavior, Parasitoids and Response to Prescribed Fire of Cavity Nesting Hymenoptera in North Central Florida

Material Information

Title:
Biology, Ecology, Behavior, Parasitoids and Response to Prescribed Fire of Cavity Nesting Hymenoptera in North Central Florida
Creator:
SERRANO, DAVID
Copyright Date:
2008

Subjects

Subjects / Keywords:
Animal nesting ( jstor )
Chaos ( jstor )
Diameters ( jstor )
Ecology ( jstor )
Estimators ( jstor )
Female animals ( jstor )
Insects ( jstor )
Predators ( jstor )
Species ( jstor )
Spiders ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright David Serrano. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2006
Resource Identifier:
476802593 ( OCLC )

Downloads

This item has the following downloads:

serrano_d.pdf

serrano_d_Page_092.txt

serrano_d_Page_063.txt

serrano_d_Page_082.txt

serrano_d_Page_038.txt

serrano_d_Page_100.txt

serrano_d_Page_077.txt

serrano_d_Page_081.txt

serrano_d_Page_089.txt

serrano_d_Page_096.txt

serrano_d_Page_052.txt

serrano_d_Page_132.txt

serrano_d_Page_164.txt

serrano_d_Page_155.txt

serrano_d_Page_091.txt

serrano_d_Page_066.txt

serrano_d_Page_006.txt

serrano_d_Page_112.txt

serrano_d_Page_163.txt

serrano_d_Page_159.txt

serrano_d_Page_064.txt

serrano_d_Page_024.txt

serrano_d_Page_114.txt

serrano_d_Page_141.txt

serrano_d_Page_080.txt

serrano_d_Page_168.txt

serrano_d_Page_040.txt

serrano_d_Page_165.txt

serrano_d_Page_158.txt

serrano_d_Page_146.txt

serrano_d_Page_061.txt

serrano_d_Page_121.txt

serrano_d_Page_003.txt

serrano_d_Page_054.txt

serrano_d_Page_009.txt

serrano_d_Page_025.txt

serrano_d_Page_106.txt

serrano_d_Page_010.txt

serrano_d_Page_065.txt

serrano_d_Page_099.txt

serrano_d_Page_150.txt

serrano_d_pdf.txt

serrano_d_Page_102.txt

serrano_d_Page_086.txt

serrano_d_Page_093.txt

serrano_d_Page_160.txt

serrano_d_Page_134.txt

serrano_d_Page_005.txt

serrano_d_Page_128.txt

serrano_d_Page_087.txt

serrano_d_Page_088.txt

serrano_d_Page_079.txt

serrano_d_Page_085.txt

serrano_d_Page_029.txt

serrano_d_Page_015.txt

serrano_d_Page_047.txt

serrano_d_Page_136.txt

serrano_d_Page_032.txt

serrano_d_Page_123.txt

serrano_d_Page_084.txt

serrano_d_Page_071.txt

serrano_d_Page_118.txt

serrano_d_Page_036.txt

serrano_d_Page_147.txt

serrano_d_Page_144.txt

serrano_d_Page_011.txt

serrano_d_Page_033.txt

serrano_d_Page_153.txt

serrano_d_Page_059.txt

serrano_d_Page_131.txt

serrano_d_Page_115.txt

serrano_d_Page_124.txt

serrano_d_Page_090.txt

serrano_d_Page_041.txt

serrano_d_Page_133.txt

serrano_d_Page_020.txt

serrano_d_Page_107.txt

serrano_d_Page_119.txt

serrano_d_Page_073.txt

serrano_d_Page_116.txt

serrano_d_Page_017.txt

serrano_d_Page_117.txt

serrano_d_Page_137.txt

serrano_d_Page_148.txt

serrano_d_Page_154.txt

serrano_d_Page_075.txt

serrano_d_Page_113.txt

serrano_d_Page_166.txt

serrano_d_Page_058.txt

serrano_d_Page_098.txt

serrano_d_Page_048.txt

serrano_d_Page_095.txt

serrano_d_Page_049.txt

serrano_d_Page_021.txt

serrano_d_Page_069.txt

serrano_d_Page_046.txt

serrano_d_Page_130.txt

serrano_d_Page_170.txt

serrano_d_Page_142.txt

serrano_d_Page_094.txt

serrano_d_Page_008.txt

serrano_d_Page_030.txt

serrano_d_Page_167.txt

serrano_d_Page_109.txt

serrano_d_Page_039.txt

serrano_d_Page_129.txt

serrano_d_Page_004.txt

serrano_d_Page_013.txt

serrano_d_Page_001.txt

serrano_d_Page_149.txt

serrano_d_Page_026.txt

serrano_d_Page_057.txt

serrano_d_Page_028.txt

serrano_d_Page_097.txt

serrano_d_Page_045.txt

serrano_d_Page_067.txt

serrano_d_Page_014.txt

serrano_d_Page_120.txt

serrano_d_Page_068.txt

serrano_d_Page_138.txt

serrano_d_Page_127.txt

serrano_d_Page_042.txt

serrano_d_Page_122.txt

serrano_d_Page_037.txt

serrano_d_Page_110.txt

serrano_d_Page_105.txt

serrano_d_Page_140.txt

serrano_d_Page_031.txt

serrano_d_Page_050.txt

serrano_d_Page_053.txt

serrano_d_Page_022.txt

serrano_d_Page_125.txt

serrano_d_Page_034.txt

serrano_d_Page_157.txt

serrano_d_Page_023.txt

serrano_d_Page_156.txt

serrano_d_Page_078.txt

serrano_d_Page_111.txt

serrano_d_Page_139.txt

serrano_d_Page_035.txt

serrano_d_Page_055.txt

serrano_d_Page_062.txt

serrano_d_Page_162.txt

serrano_d_Page_019.txt

serrano_d_Page_126.txt

serrano_d_Page_043.txt

serrano_d_Page_135.txt

serrano_d_Page_051.txt

serrano_d_Page_169.txt

serrano_d_Page_016.txt

serrano_d_Page_145.txt

serrano_d_Page_152.txt

serrano_d_Page_076.txt

serrano_d_Page_044.txt

serrano_d_Page_161.txt

serrano_d_Page_083.txt

serrano_d_Page_027.txt

serrano_d_Page_012.txt

serrano_d_Page_002.txt

serrano_d_Page_074.txt

serrano_d_Page_104.txt

serrano_d_Page_056.txt

serrano_d_Page_143.txt

serrano_d_Page_018.txt

serrano_d_Page_103.txt

serrano_d_Page_101.txt

serrano_d_Page_060.txt

serrano_d_Page_007.txt

serrano_d_Page_070.txt

serrano_d_Page_108.txt

serrano_d_Page_072.txt

serrano_d_Page_151.txt


Full Text












BIOLOGY, ECOLOGY, BEHAVIOR, PARASITOIDS AND RESPONSE TO
PRESCRIBED FIRE OF CAVITY NESTING HYMENOPTERA IN NORTH
CENTRAL FLORIDA















By

DAVID SERRANO


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2006





























Copyright 2006

by

David Serrano















ACKNOWLEDGMENTS

I thank my graduate advisor, John L. Foltz, for being understanding and patient

with all of life's complications that I have had to endure throughout my program. His

insight, suggestions, advice and patience made him the best advisor I could have chosen.

I also thank my graduate committee (Drs. Lionel Stange, Robert McSorley, and Emilio

M. Bruna) for their guidance, insight and making this work substantially better. I also

thank Jim Wiley, one of the most helpful and humble men I know, for his helpful efforts

and "mastery" of Hymenoptera. I thank my parents for teaching me the values of hard

work and perseverance that were integral in obtaining my PhD. I also thank my wife,

Esther S. Serrano (DPM 2005), for her help by "pushing" me across the finish line.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF TABLES ......................... .. .............................. ............ .. viii

LIST OF FIGURES ......... ............................... ........ ............ ix

A B S T R A C T .......................................... .................................................. x iii

CHAPTER

1 INTRODUCTION AND LITERATURE REVIEW ....................................................1

Introduction ................................................................................
Species Richness And Diversity: Estimating Their Values......................................2
Overview of Trap-nesting Hymenoptera......................... ................................5
Effect Of Fire on Trap-Nesting Hymenoptera..........................................6
Hymenopterans Sampled and Summary .............................................. ...............7

2 NEST ARCHITECTURE, PREY, AND SEXUAL DIMORPHISM IN THE
GRAS S-CARRYING WASPS Isodontia (MURRA YELLA) mexicana
(SAUSSURE) AND Isodontia auripes (FERNALD) (HYMENOPTERA:
SPH E C ID A E : SPH E C IN A E ) ..................................................................................... 12

A b stra c t ..........................................................................................12
In tro d u ctio n ........................................................................................12
M methods and M materials .............................. .... ...................... .. ........ .... ............14
Tools and Trap Preparation ........................................ ........................... 14
F field Sites ..................................................................... ..........14
F ield P lacem ent .............. .................................. ................. 15
Field Collection and Laboratory Rearing ............... ..........................................15
Specimen Diagnostics and Identification .............. ............. .................... 16
Statistical A analysis ...................................... ............................ 17
Results ............... ................................ ...............17
H a b ita t ..................................................................................................1 7
Nest Architecture ........... ..................................... 17
Sex Ratio and Sexual Dim orphism .................................. ........................ 19
Prey ......... ........ ...................................... 20
Survival ............... ......... ........................21









D isc u ssio n ............................................................................................................. 2 1
H a b ita t ........................................................................................................... 2 3
N est A architecture .................. .................. ................ ...... .... ................. 23
Sex Ratio and Sexual Dim orphism ................................... ....................... 25
P re y .......................................................................................................2 6
Conclusion................... ..................28
A ck n ow ledg em ents ....................................................................................... 2 8

3 SPIDER PREY IN NESTS OF THE MUD DAUBER WASP Trypoxylon
lactitarse (HYMENOPTERA: SPHECIDAE) ........................................37

A b stract ..................................................... ............................. 37
Introdu action ....................................... ........................................................ 37
M eth od s an d M materials ......................................................................................... 39
Tools and Trap Preparation ....................................................... 39
F ie ld S ite s .................................................................3 9
F ield P lacem ent ............................................................................................. 40
Field Collection and Laboratory Rearing ............. ....... ...............40
Specim en Identifications ................ ..................................41
Statistical A nalysis.............................................41
R e su lts ....................... ... ............. ... .......................................................4 3
D discussion ............... ...................................................... 45
Acknowledgements ................................... ..... ....... ............ 49

4 EFFECTS OF PRESCRIBED FIRE ON BIODIVERSITY AND SPECIES
RICHNESS OF CAVITY NESTING HYMENOPTERA IN SUWANNEE
RIVER STATE PARK FLORIDA .................... .... ........................... ............... 60

Abstract .................. .......................... 60
In tro d u ctio n .......................................................................................6 1
M methods and M materials .................................. ... ....... ............ 62
Tools and Trap Preparation ........................................ ........................... 62
F ie ld S ite s ............................................................................6 3
F ield P lacem ent ............................................................................................. 63
Field Collection and Laboratory Rearing ..........................................................64
Specim en Identifications .............................................................. ...............65
Statistics and Calculations ..................................................... 65
R e su lts ...........................................................................................6 8
F ie ld site s ............................ .......................................................................... 6 8
Subsites and sampling month ................................................68
Effect of burning on species richness .......................................................... 68
Effect of burning on diversity ...........................................69
Sim ilarity of burned and unburned .............................................................69
Functional groups ........... ...... ............... .. ... ................. 69
Most abundant species/ species groups.....................................................70





v









D isc u ssio n .............................................................................................................. 7 0
C o n c lu sio n ...................................................................................................... 7 4
A cknow ledgem ents ......................... ........................ .. .............. .............75

5 BIOLOGY, PREY AND NESTS OF THE POTTER-WASP Monobia quadridens
L. (HYM ENOPTERA: VESPIDAE).................................... ............................ 79

Abstract ................ .............. ............................... 79
Introduction .............................. .........................80
M methods and M materials ........................................................................ ..................8 1
Tools and Trap Preparation ............................................ ........................... 81
Field Sites .................................................................................. ..................... 81
F ield P lacem en t .............................................................................................. 8 2
Field Collection and Laboratory Rearing ............ ................ ...............82
Identifications ............. ... .. .............. ........................ ....... ..... 83
Statistical A analysis ...................... .................. ........................... 84
Results .........................................................84
Nest Architecture ............. ........... ...... ............... ....... 84
S ex R atio ...............................................................8 5
P rey ................................................................................................ 8 5
D iscu ssion ............... .... ...... .......................................................... .....86
Nest Architecture ............. ........... ...... ............... ....... 86
S ex R atio ....................................................... 8 7
P re y .........................................................................................................8 7
C o n c lu sio n ...................................................................................................... 9 0
A cknow ledgem ents ......................... ........................ .. .............. .............90

6 BIODIVERSITY OF TRAP-NESTING HYMENOPTERA OF FIVE NORTH
FLOR ID A STA TE PARK S................................................ .............................. 92

A b stra c t ......................................................................... ................ 9 2
In tro d u ctio n .......................... .. ......... ... .. ................................................ 9 2
M methods and M materials .............................. ...................... .. .............. .............93
Tools and Trap Preparation ........................................... ............................ 93
F ie ld S ite s .................................................................9 3
F ield P lacem en t .............................................................................................. 9 4
Field Collection and Laboratory Rearing ............ ................ ...............94
Specim en Identifications ............................................................................. 95
Statistical A nalysis.................................................. 96
R e su lts .................98.............................................
D isc u ssio n ................................................... .........................................9 9
A cknow ledgem ents .................................................. .................................... 104

APPENDIX

A ADDITIONAL FIGURES AND SPECIMEN PHOTO GUIDE..........................111









B SELECTED SPECIMEN DIAGNOSTICS AND IDENTIFICATION................... 143

L IST O F R E F E R E N C E S ............. .............................................................................. 145

B IO G R A PH IC A L SK E T C H ........... ..........................................................................156
















LIST OF TABLES


Table page

2-1 Comparison of Isodontia auripes and I. mexicana ...............................................32

2-2 Frequency of cavities nested in by I. auripes and I. mexicana .............................33

2-3 Summary of emerged I. auripes and I. mexicana ......................................... 33

2-4 Prey records for I. mexicana .............................................................................. 34

2-5 Prey records for Isodontia auripes ...... ..................................35

3-1 Spiders found as prey in nests of Trypoxylon lactitarse in north central Florida ....56

3-2 Similarity indexes and comparisons for spider prey ............................................. 58

3-3 Summary of diversity values for prey items tabulated by site ..............................59

4-1 Species trapped in burned and unburned sandhill pine habitat .............................77

5-1 Number nest diameters occupied by Monobia quadridens......................... .... 91

6-1 Summary of trap-nesting arthropods captured in five state parks ........................108

6-2 Similarity indexes and comparisons for trap-nesting Hymenoptera .......... ......110
















LIST OF FIGURES


Figure page

2-1 Cross section of Isodontia mexicana nest in a 12.7mm cavity ...........................29

2-2 Isodontia auripes larvae on provisioned Scudderiafurcata............................29

2-3 Isodontia cocoon ........ ........................... ........... .. .. .. ........ .... 30

2-4 Frequency of cavities nested in by I. auripes and I. mexicana........................30

2-5 Summary of emerged Isodontia mexicana and Isodontia auripes from
captured nests ..................... ...................... ... ........ ..... ............. 31

3-1 Ten most abundant spider prey species for all sites pooled.............................50

3-2 Five most abundant spider prey species at Suwannee River State Park.............50

3-3 Five most abundant spider prey species at San Felasco State Park ..................51

3-4 Five most abundant spider prey species at Silver River State Park ..................51

3-5 Five most abundant spider prey species at Gold Head Branch State Park .........52

3-6 Five most abundant spider prey species at Devils' Millhopper State Park.........52

3-7 Abundance and percentage of spider prey families captured at all sites ............53

3-8 Pooled rank proportional abundance of spider species collected from five
Florida state parks. ........................... .................... ... .. ...... .... ...... ...... 53

3-9 Site rank proportional abundance of spider species collected at each state
p a rk ........................................................................... 5 4

3-10 Species richness estimation for spider prey tabulated by site.............................54

3-11 Species richness estimator performance for spider prey tabulated by site .........55

4-1 Rank proportional abundance of species in burned and unburned sandhill
pin e h ab itats .......................................................................76

6-1 Actual observed species richness, tabulated by site.......................................105









6-2 Shannon index of diversity and Shannon evenness values for trap nesting
Hymenoptera and associated arthropods at five state parks ...........................105

6-3 Simpson index of diversity for trap nesting Hymenoptera and associated
arthropods at five state parks. ........................................ ....................... 106

6-4 Species richness estimators tabulated by site............................... ...............106

6-5 Species richness estimator performance per site ...........................................107

A T rap s ........................................................ ....... ............................. 1 1 1

A -2 R hearing room ............................... ......................... ................... 111

A-3 Male Monobia quadridens.................... ......................... 112

A-4 Female Monobia quadridens .................. ..................... ...............113

A-5 Antenna of Monobia quadridens ................ ............................... 113

A -6 M ale Isodontia auripes ......................................................... .............. 114

A -7 Fem ale Isodontia auripes ........................................................ ............... 114

A-9 Fem ale Isodontia m exicana.......................................................... .. ........ 115

A -10 Anthrax analis ....... ........................... ........ ... .............. ........ .. .. 116

A- 11 Anthrax aterrimus ............. ... ...... ............................. ............... 116

A-12 Lepidophora lepidocera ........................................................ ............. 117

A -13 Toxophora amphitea .................................... ......... ................. ............... 117

A-14 A wasp in the family Chrysididae........ ........................... ............... 118

A-15 A series of Chrysidid wasps demonstrating variation in size and color .........118

A -16 Lecontella brunnea ............................................. .................. ......................119

A-17 M acrosigon cruentum ............... ....... .. .............. .............................. 119

A -18 N em ognatha punctulata......... ................. ............................... ............... 20

A-19 Ancistorcerus ............................... .. ..... ......... .. ............120

A -20 Euodynerus m egaera ............................................... ............................. 121

A -21 Pacnodynerus erynnis ......................................................... .............. 121









A -22 Stenodynerus sp A .................................................. ............................... 122

A -23 Stenodynerus sp b ........................................... ....................................... 122

A -24 Camponotus Red ......... ....................... ............. ....................... ............... 123

A-29 A wasp of the family Leucospididae ............................... ..................126

A -30 D olicostelis louisa................................................. ................................ 127

A -31 Coelioxys sayi .......................................... .. .. ............... ....... 127

A -32 Coelioxys dolichos ...................................... .............................................. 128

A -33 Coelioxys texana .......................................................................... ............... 128

A -34 M egachile campanulae ............................................. ............................ 129

A-35 M egachile mendica ......................................... ................................ 129

A -36 M egachile c. w ilm ingtoni....................................................... ............... 130

A -37 M egachile g eorg ica ........................................... ......................................... 130

A-38 Megachile xylocopoides female................................. 131

A-39 M egachile xylocopoides male ................................... .............. .............. 131

A -40 O sm ia sandhouseae ................................................ .............................. 132

A-41 Sphaeropthalma pensylvanicafloridensis .............. .. ................. 132

A -42 O rocharis luteolira ................................................. ............... ............... 133

A -43 Ampulex canaliculata.......................... ......... .......................... ............... 134

A -45 P odium rufipes ....................................................... ......... ............135

A-46 Trypoxylon clavatum clavatum ...... ........... ............... 136

A-47 Face of Trypoxylon c. clavatum. Note golden vessiture............... ...............136

A -48 Trypoxylon carinatum ............................. ........................ ............... 137

A-50 Trypoxylon collinum collinum ................................................................... 138

A -5 1 Trypoxylon johnsoni ........................................................... ............... 138

A -52 Trypoxylon lactitarse ............................................... ............................. 139









A -53 Vesp ula m aculifrons........................................................................... ... ..... 39

A-54 Xylocopa virginica, male ............................................................................140

A-55 Xylocopa virginica, female .................................... ..................................... 140

A -59 Centruiodes hentzi ................................................. ............................... 142















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

BIOLOGY, ECOLOGY, BEHAVIOR, PARASITOIDS AND RESPONSE TO
PRESCRIBED FIRE OF CAVITY NESTING HYMENOPTERA IN NORTH
CENTRAL FLORIDA


By

David Serrano

August 2006

Chair: John L. Foltz
Major Department: Entomology and Nematology

This study examined the biology, ecology, behavior, parasitoids and response to

fire of an understudied group of insects, the cavity nesting Hymenoptera. Five state parks

in north central Florida were surveyed for two years with trap nests yielding over 3,000

captured nests. Trap-nesting Hymenoptera represent important guilds, such as predators

and pollinators, within these surveyed habitats and are an integral part of maintaining

desired biodiversity of both flora and fauna. Over the two year period, biology, ecology,

and prey of a potter wasp, Monobia quadridens, a mud-dauber wasp, Trypoxylon

lactitarse, and two grass carrying wasps, Isodontia auripes and Isodontia mexicana, were

examined in depth. In addition, more than 100 species of trap-nesting Hymenoptera and

associated arthropods were examined yielding data on distribution, host ranges, biology

and ecology. Also, a detailed inventory of identified trap-nesting hymenoptera and

associated arthropods is provided to expand park faunal records.









In addition to examining biology and ecology of this group, this study examines the

effect prescribed fire has on these insects. Prescribed fire is a commonly used practice in

managed parks and natural areas to restore and maintain native and protected habitat and

these insects, as pollinators and predators of plant feeders, may play an important role in

the succession of desirable, native habitat after the fire event. Prescribed fire was used by

the park managers in such a manner that allowed for comparison of equally sized areas of

identical habitat. Overall, the community of trap-nesting Hymenoptera was affected by

the scale of prescribed fire used by the park service in terms of overall diversity and

abundance of key species. The diversity and richness of cavity nesting Hymenoptera may

be used as an indicator of when to use prescribed fire to maintain native ecosystems and

foster a healthy biodiversity.














CHAPTER 1
INTRODUCTION AND LITERATURE REVIEW

Introduction

During the late 1960s and 1970s, the United States began to confront the costs of

unrestricted patterns of settlement and land use on the environment (Porter and Marsh

2005). Wetlands, water and air quality, and threatened species were all protected by new

laws, with many states passing laws and regulations to further address these concerns.

Conservation and restoration of natural areas are now common requirements of land and

community development mandated by local, state, and federal agencies.

As human population and development increase and the relative amount of these

natural areas decreases, the quality and health of these natural areas are becoming of

more importance. Agricultural practices, resource gathering, waterway diversions,

fragmentation of natural habitat and alteration of patterns of natural vegetation are

increasing with human population expansion and affecting even previously protected and

isolated natural areas (Collinge 1996, Dale et al. 1998, Kramer 2005). Degrading habitat

health, measured by biodiversity, is a major concern for natural resource managers and

governmental agencies. Currently, the primary driver for the loss in biodiversity is

habitat modification and destruction due to changes in land-use practices (Kramer 2005).

As human populations expand and the wildland-urban interface increases, more natural

and protected areas are increasingly affected by these changes in land use. In addition,

resource managers use many land management techniques, such as prescribed fire and

mechanical removal of vegetation, directly in natural and protected areas in order to









protect human interests. These direct impacts to the natural and protected areas should be

monitored to assure minimal negative impact to biodiversity.

As more natural and protected areas are subjected to these changing forces, land

managers must be able to quantify changes in order to identify areas at risk. There are

many tools in the environmental and ecological sciences for quantifying changes and

differences in biodiversity, yet an initial inventory or measurement is needed for future

assessments.

Species Richness And Diversity: Estimating Their Values

Diversity has been a persistent theme in ecology and is frequently seen as an

indicator of ecological health (Magurran 1988). Although often incorrectly used

interchangeably, species richness and diversity are distinct entities that relay sometimes

quite different information.

Species richness, the number of species in an area, is a simple yet informative

measurement of a community. Intuitively, this simple measurement is ideal for

comparing communities in conservation and management of biodiversity, assessing

anthropogenic effects on protected lands and influencing public policy. Yet, this measure

is not simple to accurately attain. Complete species inventories usually require huge

amounts of resources and expertise and are impractical and quite often impossible to

compile. Almost every taxonomic survey will undoubtedly have undiscovered species.

This trend is especially true with hyper-diverse taxonomic groups such as arthropods,

nematodes, bacteria, and fungi. These groups are impossible to completely survey.

Many groups of interest that are especially sensitive to anthropogenic disturbance are

these hyper-diverse taxonomic groups. In these cases, the best option for measuring









richness is through sampling of target environments or particular regions of interest, such

as a habitat slated for conservation or development.

Three broad categories of estimators exist to estimate the number of species in a

community: estimators that fit a lognormal abundance distribution and estimate the

hidden or unsampled portion of the curve, estimators that fit asymptotic equations to

species accumulation curves, and non-parametric estimators that use relative abundance

of rare species to estimate the number of unseen species. In recent years, there has been a

heightened interest in biodiversity that has resulted in new measurement techniques

including niche apportionment models, new techniques for measuring taxonomic

diversity, and improved methods of species richness estimation (Heltshe and Forrester

1983, Chao 1984, Hughes 1986, Chao 1987, Magurran 1988, Chao and Lee 1992,

Colwell and Coddington 1994, Longino et al. 2002, Colwell et al 2004, Magurran 2004,

Chao et al. 2005). Fortunately, long difficult mathematical calculations are usually no

longer needed to estimate species richness through the many computer software packages

readily available such as EstimateS (Colwell, 2005), Distance (Thomas et al. 2005),

WS2m (Turner et al. 2003) and COMDYN (Hines et al. 1999), among others. Magurran

(2004) provides more examples of such programs and discusses their use, theory and

sources of acquisition.

These estimators of species richness are useful for letting researchers know when

they have sampled sufficiently to have confidently surveyed the majority of species

present. Such information is crucial since funds, time, and the taxonomic experts needed

for reliable identification are usually in short supply (Hopkins and Freckleton 2002).









Species richness is an informative measure but can, on its own, be misleading.

Species richness is simply the number of species within a geographic area, but how

desirable these species are is another matter. Quite often such information is used in

conservation, preservation, restoration efforts and public policy. However, degraded

areas not desirable for conservation may have a relatively high value of species richness

(when compared to its desirable counterpart) due to exotic, feral and transient species.

Disturbed habitats may serve as sinks drawing in species from surrounding habitats.

During my undergraduate research (unpublished data) in Everglades National Park we

saw more species of flower-visiting insects on the mowed roadsides (a disturbed habitat)

than in the neighboring marl prairie. Utilizing solely species richness, we could suggest

that building more roads in the marl prairie would increase pollinator species richness of

the park. Of course such a conclusion and suggestion is ludicrous. Consequently, a

detailed understanding of ecological relationships of the species sampled is essential

when applying such studies to land-use practices and policies (McCraken and Bignal

1998).

Much more ecological insight can be attained though measurement of diversity

than solely through species richness estimation. Diversity is a measure of species

richness and the abundance distribution of these species, and therefore can detect effects

unseen by species richness alone. Disturbance effects may cause changes in diversity

through shifts in the abundance of species or increases in the dominance of some species.

Magurran (1988) details the theory and calculation of various diversity measures

including the Shannon (1949) and Simpson (1949) indices of diversity. Such indices are

easily calculated by traditional means, but many software packages can quickly calculate









these values, including updated versions of the indices that take abundance data into

account. Computer software such as EstimateS (Colwell 2005) is available and can

easily calculate richness and diversity values with a variety of estimators.

Overview of Trap-nesting Hymenoptera

Aculeate Hymenoptera are an integral part of most terrestrial ecosystems (Jenkins

and Matthews 2004), including natural and disturbed habitats of north central Florida.

These Hymenopterans fill many important roles, most commonly as pollinators,

predators, and parasitoids. Changes in their populations would have a cascading effect,

altering the habitat's flora and fauna (Raw 1988, LaSalle and Gauld 1993, Neff and

Simpson 1993, Jenkins and Matthews 2004). The majority of these Hymenopteran

species are solitary in behavior.

The nests of solitary bees and wasps are usually difficult to find and examine

(Krombein 1967, Jayasingh and Freeman 1980, Alves-dos-Santos 2003). Many nest in

pre-existing cavities in various substrates such as wood, clay, rock and man-made

structures (Bequaert 1940, Krombein and Evans 1954, Krombein 1967, 1970, Bohart and

Menke 1976, Coville and Coville 1980, Coville 1982). This practice makes their nests

not only difficult to find, but also extremely difficult to successfully extract and examine.

The majority of these insects readily accept trap-nests (drilled wooden blocks) since they

normally nest in preexisting cavities created by other creatures. In addition, many of

these insects frequently reuse cavities for nesting, allowing for long term observations on

biology and ecology. Many successful studies of solitary bees and wasps have used trap-

nests to examine various aspects of their biology and ecology. Trap-nests have been used

to examine species composition and diversity at particular sites (Parker and Bohart 1966,

1968, Krombein 1967, 1970 and Camillo et al. 1995), population dynamics of occupants









(Jayasingh and Freeman 1980), evaluation of habitat health and effects of habitat

fragmentation (Frankie et al. 1998 and Tscharntke et al. 1998), genetic study (Packer et

al. 1995) and survey of exotic species (Mangum and Sumner 2003). Trap-nests are

extremely useful when the study focuses on gathering biological data of occupant

species. Various studies have examined nesting behavior and architecture (Medler 1967,

Krombein 1967, 1970, Camillo et al. 1993, Pereira et al. 1999 and Alves-dos-Santos

2003), prey captured (Krombein 1967, 1970, and Camillo and Brescovit 1999, 2000), and

associated parasitoids (Krombein 1967, 1970, Wcislo et al. 1996 and Scott et al. 2000).

Trap-nests are a powerful survey tool that allows collection of data on abundance, prey,

habitat, phenology, and nest architecture that are not detectable through other survey

methods that target Hymenoptera (Gathmann et al. 1994, Steffan-Dewenter 2002,

Miyano and Yamaguchi 2001, Jenkins and Matthews 2004).

Effect Of Fire on Trap-Nesting Hymenoptera

Fire is an integral part of forest and grassland ecosystems throughout the United

States. Native Americans used fire for many purposes, such as a tool to clear areas for

agriculture. Fire in natural areas, however, poses many hazards especially when in close

proximity to urban areas. In response to this threatening hazard policies of complete fire

suppression became popular in the 1920s and 1930s (Long et al. 2005). One result of the

absence of periodic fires was a buildup of woody understory and excessive fuels. This

caused subsequent fires to become more intense, damaging and unmanageable. Forest

management with prescribed burning is the current popular tool. Prescribed fire has been

shown to be effective in reducing hazardous fuels, disposing of logging debris, preparing

sites for seeding or planting, improving wildlife habitat, managing competing vegetation,

managing invasive weeds, controlling insects and diseases, improving forage for grazing,









enhancing appearance, improving access and in perpetuating fire-dependent species

(Cumming 1964, Helms 1979, Wade et al. 1988, Biswell 1999, DellaSala and Frost 2001,

Fuller 1991, Mutch 1994, Paynter and Flanagan 2004 and Long et al. 2005). Many

studies have been conducted to examine the effect of fire on plant (Main 2002, Vazquez

et al. 2002, Laterra 2003, Lloret 2003, Reinhart 2004, Schoennagel 2004, Barton 2005,

Overbeck et al 2005, Ansley et al. 2006) and animal communities (Chew et al. 1959,

Kahn 1960, Lawrence 1966, Simons 1989, Mushinsky 1992, Saab and Vierling 2001,

Cunningham et al. 2002, Meehan and George 2003) with some examining arthropods.

Fire not only causes direct mortality in arthropods (Fay and Samenson 1993, Bolton and

Peck 1946, Miller 1978, Evans 1984) but also indirectly affects arthropod communities

via changes in plant community composition and habitat alteration (Lawton 1983, Evans

1984). Unfortunately, most invertebrate studies have focused on terrestrial arthropods

monitored via sweeping or pitfall traps (Bess 2002, Brand 2002, Niwa 2002, Clayton

2002, Fay 2003, and Koponen 2005) and have overlooked trap-nesting Hymenoptera and

other aerial insects.

Hymenopterans Sampled and Summary

This study provides a record and survey of trap-nesting Hymenoptera in five

Florida state parks to further enhance the understanding of these habitats. It also provides

a bench mark for future assessments of the insect fauna in these habitats.

The following chapters provide a detailed inventory and biological notes on many

trap-nesting Hymenopterans and associated arthropods. These findings may be used in

assessing habitat quality and perhaps aid in identifying any changes in biodiversity over

time for the five state parks studied.









Observations of biology and natural history are important to document, especially

at the extremes of geographical range and in unique areas for a species that is

cosmopolitan in range, such as Isodontia. In particular I. mexicana has become

established in Hawaii and France (Bohart and Menke 1976) providing unique habitats in

comparison to its native North America. This case provides observations of mexicana

in the southeast extreme of the geographical range. O'Neil and O'Neil (2003) recently

studied a population of mexicana in Montana and should provide a nice comparison to

this Florida population. Debates of species and subspecies versus dines and ecotypes

commonly arise and it is important that we identify possible subjects to further examine

the mechanisms of speciation and hybrid zones. For example, the splitting ofAnisota

senatoria into A. senatoria and A. peigleri by Riotte (1975) has been questioned by

Tuskes et al. (1996) and throughout many taxa, taxonomists that are lumperss" or

splitterss" are constantly at odds.

In addition, such information is valuable to help identify possible projects for

evolutionary biologists and studies in biogeography. For example, Mark Deyrup and

Thomas Eisner (2003) examined the differences of coloration between Florida

Hymenoptera and their northern relatives (subspecies, clinal types, etc.) utilizing museum

specimens and natural history data from past studies. Their preliminary observations

called for further exploration of Florida biogeography to help recognize distinctive

species and examine mimetic complexes. Isodontia mexicana is a cosmopolitan species

that is easily captured, studied and occurs sympatrically with I auripes at this site.

Therefore chapter 2 examines ecology and natural history of a cosmopolitan species of a

trap-nesting Hymenopteran, Isodontia mexicana (Hymenoptera: Sphecidae), and a sister









species, I. auripes, in these Florida study sites. These observations give a basis for

geographical comparison for this wide-ranging species. Observations and findings for .

mexicana populations have been recently published by O'Neill (2001), O'Neill and

O'Neill (2003), and historically by Bohart and Menke (1963), Lin (1966), Krombein

(1967), Bohart and Menke (1976). These studies took place in distant parts of the

geographical range, as compared to the Florida population observed, and may offer

insight and inspiration for additional study. Such studies, for example, Sears et al.

(2001), may examine behavioral, biological and natural history differences of a species in

two extremes of its range. Such information may also be useful for examining aspects of

biogeography and evolutionary history (Deyrup and Eisner 2003). For chapter 2 there are

two main objectives: 1) Examine similar features and aspects of natural history of Florida

population that was examined in Montana populations and infer whether these features

warrant further biogeographical examination, 2) and since the two sister species (I.

mexicana and I. auripes) occur sympatrically in these Florida sites, determine if they

differ substantially in the examined features to inspire a closer look at possible

speciation/separating mechanisms.

In Chapter 5 a second species, Monobia quadridens (Hymenoptera: Vespidae), is

examined in the same manner. Monobia quadridens is also a cosmopolitan species in

terms of geographical range and has been previously studied in different areas by

Bequaert (1940), Krombein (1967), and Krombein et al. (1979). Monobia quadridens

has not been studied in the recent literature in terms of biology and natural history. The

objectives for chapter 5 are to 1) determine if this wasp has preference in cavity size for

nesting, 2) determine the nest architecture, 3) determine the range of prey provisioned by









this wasp in this portion of its geographic range, comparing findings to other published

records, 4) and determine the emerging sex ratio of trap-nested individuals.

In Chapter 3 the spider prey captured by the most abundant trap-nesting

Hymenopteran in this study, Trypoxylon lactitarse (Hymenoptera: Sphecidae) is

examined. Previous studies (Rau 1928, Krombein and Evans 1954, Krombein 1956,

1967, Medler 1965, Lin 1969, Coville 1979, 1981, 1982, Coville and Coville 1980,

Genaro et. al. 1989, Camillo et. al. 1993, Genaro and Alayon 1994, Jimenez and Tejas

1994, Camillo and Brescovit 1999, 2000) have highlighted differing prey preferences in

various species of Trypoxylon and Rehnberg (1987) and Camillo and Breviscovit (1998)

examined prey preferences for Trypoxylon lactitarse. This chapter, therefore, has five

main objectives: 1) Determine what prey T. lactitarse is provisioning at these Florida

sites, 2) Determine if T lactitarse is a generalist or specialist in terms of prey

provisioned, 3) Determine what, if any, is T. lactitarse's prey preference, 4) Does T.

lactitarse's prey preference seem to differ between sites? 5) Determine the benefits and

potential problems with using this wasp (and potentially other spider-provisioning trap-

nesters) as a sampling tool for estimating spider abundance and species richness.

In Chapter 4 the impact prescribed fire has on trap-nesting Hymenoptera and

associated arthropods is examined and poses the following questions: 1) Does overall

diversity and species richness of trap-nesting Hymenoptera differ between burned and

unburned sites? 2) In terms of species sampled, how similar are the burned and unburned

sites? 3) Is the diversity of sampled functional groups (predator, parasitoid and pollen

specialists) affected by fire? 4) What species, if any, seem to be negatively or positively

affected by fire in terms of abundance?









There was a large amount of data collected in the process of addressing the

objectives of chapters 2-5. This final chapter summarizes and reports this mass data,

which are highly desirable, and required by the Florida State Department of

Environmental Protection. The objectives of chapter six are 1) report the abundances and

species richness of all trap-nesting Hymenoptera and associated arthropods sampled at

each of the five surveyed Florida State Parks and 2) determine, by using estimators, if the

inventory offered can be considered adequate and, if adequate, estimate total species

richness and diversity of trap-nesting hymenopterans and associated arthropods for each

state park surveyed.














CHAPTER 2
NEST ARCHITECTURE, PREY, AND SEXUAL DIMORPHISM IN THE GRASS-
CARRYING WASPS ISODONTIA (MURRAYELLA) MEXICANA (SAUSSURE) AND
ISODONTIA A URIPES (FERNALD) (HYMENOPTERA: SPHECIDAE: SPHECINAE)

Abstract

Isodontia (Murrayella) mexicana (Saussure) and Isodontia (Murrayella) auripes

(Femald) nested in trap nests at four different state parks in north central Florida. Nests

consisted of fragments of native nutrush grasses (Cyperaceae: Scleria sp.). Females

provisioned either a communal cell, or separated cells with 1 to 15 tree crickets

(Gryllidae: Oecanthinae: Oecanthus), bush crickets (Gryllidae: Eneopterinae, Orocharis),

meadow katydids (Tettigoniidae: Conocephalinae, Odontoxiphidium), coneheaded

katydids (Tettigoniidae: Copiphorinae, Belocephalus, Conocephalus and

Neoconocephalis), and false katydids (Tettigoniidae: Phaneroperinae, Scudderia). Both

Isodontia species displayed the sexual size difference typically found in the Sphecidae

with females significantly larger than males. Female-biased provisioning has been shown

to occur in other populations of mexicana and seems to occur in the Florida populations

as well. Although I. auripes exhibits this sexual size trend, the communal brood chamber

of the nest architecture rules out any provisioning difference as the cause for the size

difference.

Introduction

Isodontia (Hymenoptera) is one of the cavity-nesting genera of the Sphecid

subfamily Sphecinae and this genus is unique in its nesting biology. While other solitary

aculeate wasps that nest in pre-existing cavities use mud, agglutinated sand, plant resin or









masticated plant materials as nest partitions and plugs (O'Neill and O'Neill 2003, O'Neill

2001), Isodontia use dry grass leaves that they cut, pack, and twist into position within

the nest (Lin 1966, Krombein 1967, Bohart and Menke 1976, O'Neill and O'Neill 2003).

They are commonly known as the grass-carrier wasps since they can be observed flying

with bits of grass as long as 80 mm in their mandibles (Bohart and Menke 1963). Several

species of Isodontia construct nests with a common brood chamber that contains as many

as 12 larvae feeding on a common prey mass (Bohart and Menke 1976, O'Neill and

O'Neill 2003). Here I report on a two-year study where female Isodontia mexicana and

Isodontia auripes nested in trap-nests set up in four state parks in north central Florida. I

examined 90 nests of mexicana and 89 nests of I auripes out of 235 total Isodontia

nests and recorded information on nest structure, prey, sex ratio, sexual size dimorphism,

emergence schedules and parasitoids. I then determined if they exhibited sexual

dimorphism typically seen in other Sphecid wasps, identified nest architecture

highlighting the difference between the two species, identified prey used to provision

nests and examined survival of brood including parasitoids and predators of these wasps.

There are two main objectives for this study: 1) Examine similar features and aspects of

natural history of Florida population that were examined in Montana populations and

infer whether these features warrant further biogeographical examination, and 2) since .

mexicana and I. auripes occur sympatrically in these Florida sites, determine if they

differ substantially in the examined features to inspire a closer look at possible

speciation/separating mechanisms.









Methods and Materials

Tools and Trap Preparation

The traps used in this study were fabricated from seasoned 37-mm x 86-mm x 2.4m

pine/spruce timbers obtained from a local home improvement store. The pine/spruce

timbers were cut into 10-cm-long blocks. Two cavities of one of five diameters (3.2, 4.8,

6.4, 7.9 or 12.7-mm) were drilled into each block. Cavities were drilled to a depth of 80

mm on each short side (the 37-mm side), offset approximately 10-mm from the center

point. Traps were assembled using one block of each diameter with the smallest cavity

on top and the largest on the bottom. Blocks were stacked so that no cavity was situated

directly above or below a cavity in the adjacent block. The five blocks were bound

together with strapping tape (3M St Paul, MN), and 16-gauge wire was used to further

bind the stack and suspend the trap from trees and shrubs at the field sites. Each bundle

of five blocks was considered to be a single trap.

Field Sites

I set traps at five locations: 1) Suwannee River State Park in Suwannee County (30

23.149' N, 083 10.108' W), 2) Mike Roess Gold Head Branch State Park in Clay County

(29 50.845'N, 081 57.688' W), 3) Devil's Millhopper Geological State Park in Alachua

County (29 42.314'N, 082 23.6924' W), 4) San Felasco Hammock Preserve State Park

(29 42.860' N, 082 27.656' W) in Alachua County and 5) Silver River State Park in

Marion County (29 12.317'N, 082 01.128' W). The habitats surveyed at Suwannee

River State Park were burned and unburned sand hill habitat, while the habitat at Mike

Roess Gold Head Branch State Park was burned sand hill pineland and ravine. Sites at

San Felasco Hammock Preserve State Park consisted of upland and mesic hardwood

hammock. Surveyed areas of Devil's Millhopper Geological State Park consisted of pine









flatwood habitat and sites at Silver River State Park consisted of river habitat and upland

mesic forest. Descriptions of these habitats can be found in Franz and Hall (1991).

Field Placement

Transects were set up with ten traps placed approximately 10 m apart and hung

approximately 1.5 m off the ground on trees or limbs with placement on dead standing

wood preferred. Transects were initially established (direction and distance from center

of plot) randomly. Four transects were established in Suwannee River State Park while

three transects were established Mike Roess Gold Head Branch State Park. Three

transects were established in San Felasco State Park but size constraints only allowed a

single transect in Devil's Millhopper State Park. Finally, two transects were set up in

Silver River State Park. Transects were in the field from April 2003 until January 2005.

Field Collection and Laboratory Rearing

Traps remained in the field two years and were checked monthly. Preliminary field

tests revealed that one-month intervals were sufficient to avoid trap saturation (no

available cavities). Traps were considered occupied when insects were observed actively

nesting, harboring or had sealed a cavity with mud or plant material. Occupied traps

were removed and replaced with a new trap. These occupied traps were brought into the

forest entomology lab at the University of Florida in Gainesville, FL, for processing.

Occupied blocks were removed for observation while unoccupied blocks were

reincorporated into replacement traps. Each occupied cavity was given a unique

reference number.

Location, date of collection, diameter of cavity, and various notes describing the

nature of the occupants and/or plug were recorded for each reference number. Occupied

cavities were then covered with a 2, 4, 6, or 8-dram glass shell vial. The shell vials were









attached to the wood section with masking tape (Duck, Henkel Consumer Adhesive

Inc., Akron Ohio) appropriate for wood application. These sections were then placed in a

rearing room and observed daily for emergence. The rearing room was maintained as

nearly as possible at outside mean temperatures for Gainesville, Florida.

When emergence occurred, the specimens were removed, preserved and given the

same reference number as the cavity from which they had emerged. Dates of emergence,

identification of occupants, measurements and notes were taken for each cavity at

emergence. When an insect was harboring or actively tending a nest, it was captured,

identified, and given a reference number corresponding to the cavity. The contents of the

nest/cavity were then extracted and recorded. After the contents were extracted, the

wood block was reused in replacement traps. These processed blocks were re-drilled to

the next larger diameter cavity to eliminate any alterations or markings (either physical or

chemical) by the previous occupant prior to reuse.

Specimen Diagnostics and Identification

Isodontia auripes and I. mexicana occur sympatrically and it is important to

distinguish between the two species. Appendix B (adapted from Bohart and Menke

(1963)) provides characters to distinguish between the species and sexes of each species.

All cavity nesters and their prey were identified by the author with some specimens

identified and/or verified by entomologists Jim Wiley1, Lionel Stange1, Thomas Walker2,

and John M. Leavengood Jr.1'2 (Florida State Collection of Arthropods1 Gainesville, FL

and University of Florida2, Gainesville, FL). Voucher specimens have been deposited at

the Florida State Collection of Arthropods in Gainesville, Florida.









Statistical Analysis

Descriptive statistics (means, ranges, SD, etc.) were calculated the Microsoft Excel

statistical package (Microsoft, Inc, CA). Since all sites where both Isodontia species

occurred were similar (sand hill habitat) data were pooled for analysis. One main concern

may be the difference of burned and unburned sandhill habitats. Collections for both

species in unburned habitat was quite low with only 11% of total abundance for I.

mexicana and 10% of total abundance for I. auripes. Chi-squared goodness of fit test

was used to examine nest diameter preference for pooled habitats and burned and

unburned habitats separately. The assumption was that wasps would nest equally in all

diameters. Since neither species nested in 3.2-mm diameter cavities, that cavity size was

omitted from analysis. Head width of adults was measured to the nearest 0.01mm using

an ocular micrometer. Head capsule comparison was analyzed using, the Mann-Whitney

test.

Results

Habitat

Both Isodontia species were captured at Suwannee River, Gold Head, and Devil's

Millhopper State Parks. Only Isodontia mexicana was captured at San Felasco S. P., and

neither Isodontia species was captured at Silver River S. P.

Nest Architecture

I examined 90 nests of Isodontia mexicana and 89 nests of auripes. An additional

56 cavities that had Isodontia nests were trapped but species identification of these nests

was not possible due to predation or disturbance. Females of mexicana preferred to

nest in 7.9-mm cavities (20 of 90 nests) and 12.7-mm cavities (68 of 90 nests; Table 2)

and none nested in 3.2-mm and 4.8-mm cavities. Only 2 of 90 females nested in a 6.4-









mm cavity. Females did not equally nest in all diameters (chi-squared contingency table,

X2= 133.5, df= 3, P< .001). Results were similar when the population was separated

into burned and unburned habitats (P< 0.001, df=3, Xburned= 108.4, Nburned= 10,

Xunburned= 30, Nunbumed=80).

Likewise, females of auripes nested in mostly 12.7-mm cavities (85 of 89

nests). Only one nest of auripes was placed in a 4.8-mm cavity and 3 nests were

placed in 7.9-mm cavities (figure 2-4)(chi-squared contingency table, X2 = 236.1, df= 3,

P<0.001). Results were similar when the population was separated into burned and

unburned habitats (P <0.001, df=3, X2bued=215.2, Nbured = 8, 2unbumed= 17.2, Nunbumed=

81)

Both species used grass as back wall and opening plugs, and L. mexicana also made

brood cell partitions out of the grass material. The grass did not have any binding agents

(such as resin or secretions), but was twisted and compacted into position. Many nests

had a slight amount of grass within the brood cell(s) when extracted and the pupae had a

fair amount of grass pieces adhering to them (Figure 2-3). The occurrence of grass within

the brood cell suggests that females may line the cell, but female and larval activity

cannot be distinguished without further observation. All I. mexicana observed utilized

separated brood cells (figure 2-1), with individual cells each 20-30mm. Each egg was laid

on an orthopteran prey item and then separated from the next egg and provisioned prey

mass by a tightly packed partition of grass. These nests had a mean of 2.50 brood cells

(SD = 0.88, range = 1-4, N= 90). The majority of mexicana nests had 3 cells, however

the "over-wintering" type of nest with only one brood cell may have resulted in a lower

mean. In fact, when these outliers are removed the mean becomes 2.86 (SD= 0.50, range









2-4 N= 72). O'Neill and O'Neill (2003) observed I. mexicana in Montana had a range of

1-6 separated cells in cavities 15 cm deep. In contrast, Isodontia auripes consistently

used a common brood cell in its nest architecture. Cells tended to be 50-60 mm in length.

All provisions and eggs were laid in a single cell without any internal partitions. Both

species used neatly coiled, tightly packed plugs of grass for the back end of the nest and

to close the opening. These tightly packed plugs ranged from 6-10 mm thick for all nests

in 12.7-mm and 7.9-mm diameter cavities. The tightly packed plugs in the few nests in

smaller diameter (4.8-mm and 6.4-mm) cavities tended to be slightly thicker at 10-15

mm. Both species also used a loose plug between the opening and the outermost tightly

packed plug. These loose plugs tended to include longer lengths of the grass and

occasionally contained seed heads. Loose plugs occupied the outermost 5-20 mm of the

cavity and always extended beyond the cavity opening. These plugs of grass resemble

broom-like tufts and regularly extended 30-60 mm beyond the cavity opening and

occasionally reached up to 100 mm beyond the cavity opening. The few seed heads

included in the loose plug material allowed for identification of the grasses used by these

wasps. Mark Garland (Botanist at the Florida Department Agriculture, Division of Plant

Industry, Gainesville, Florida) identified the materials as the native nutrushes Scleria sp

(ciliate/pauciflora) (Cyperaceae).

Sex Ratio and Sexual Dimorphism

Isodontia mexicana that emerged from trap nests had a sex ratio of 1.2 males per

female (N= 119). Isodontia auripes that emerged from trap nests had a sex ratio of 5.3

males per female (N= 144).

Isodontia mexicana display sexual size dimorphism typical for the Sphecidae

(O'Neill 2001). Females that emerged from nests (mean head width = 3.10 mm, SD =









0.20 mm, range 2.7-3.6 mm, N= 53) were larger than males (mean head width = 2.84

mm, SD = 0.18 mm, range 2.4-3.2 mm, N= 65; Mann-Whitney U= 2930, P < 0.0001).

Twenty-three percent of the females were larger than the largest male and 15% of the

males were smaller than the smallest female. These differences are far less than in the

Montana populations of mexicana examined by O'Neill and O'Neil (2003).

Isodontia auripes females that emerged from nests (mean head width = 3.30 mm,

SD = 0.397 mm, range 2.4-3.9 mm, N= 23) were larger than males (mean head width =

3.01, SD = 0.259 mm, range 2.2-3.5, N = 121; Mann-Whitney U= 2276.5, P < 0.0001),

with 30% of females larger than the largest male and 3% of the males being smaller than

the smallest female.

Overall, auripes tended to be larger than I mexicana (females: Mann-Whitney

U= 867.5, P < 0.01, males: Mann-Whitney U= 6270, P< 0.001).

Prey

Extracting the contents of 20 Isodontia nests yielded samples with provisions in an

identifiable condition. Thomas J. Walker (Professor Emeritus, University of Florida,

Gainesville, Fl.) positively identified prey provisions from these nests. Nests ofL.

mexicana contained Odontoxiphidium apterum Morse 1891 (Tettigoniidae:

Conocephalinae), Oecanthus quadripunctatus Beutenmuller 1894 (Gryllidae:

Oecanthinae), Belocephalus sp. (Tettigoniidae: Copiphorinae), Orocharis luteolira T

Walkerl969 (Gryllidae: Eneopterinae) and Scudderia sp. (juv) (Tettigoniidae:

Phaneroperinae) (Table 2.4).

Nests of auripes contained Odontoxiphidium apterum, Oecanthus celerinictus T

Walker 1963 (Gryllidae: Oecanthinae), Oecanthus niveus (De Geer 1773) (Gryllidae:

Oecanthinae), Orocharis luteolira, Neoconocephalis spp. (juv) (Tettigoniidae:









Copiphorinae), Conocephalus brevipennis (Tettigoniidae: Conocephalinae), Scudderia

furcata sp. (juv) (Tettigoniidae: Phaneroperinae), and other Oecanthus spp. (juv) (Table

2.5).

Amount of prey provisioned varied greatly. Provisioned prey ranged from 1-19

prey items per nest. Nests and/or brood cells that had one or few prey items tended to

contain large adult tettigoniids and those nests and/or brood cells with many prey tended

to contain juveniles and/or small species of gryllids.

Survival

Ants (Crematogaster spp.) pillaged many Isodontia nests and it was impossible to

identify the species of Isodontia. Therefore, overall Isodontia survival was calculated. A

total of 235 nests were examined yielding 320 individuals of 529 resulting in a

survival/emergence percentage of 60.49%. Mean brood per nest was 2.27 (SD= 1.311),

yet this number is not useful since the two species differ in nesting strategies. Fifteen

nests were lost to Crematogaster ant raids which accounted for 8.7% mortality of brood.

Seven nests were lost to bombyliid fly parasitoids in the genera Anthrax and Lepidophora

accounting for 3.4% of brood mortality. In addition, 2 nests were lost to a phorid fly

parasitoid and 1 nest was lost to a male mutillid in the genus Spheropthalma

(Sphaeropthalma), most likely the species pensylvanica. One nest was lost to supersedure

(the act of taking over by a second individual of the same or different species of a cavity

partially stored by the first individual) when a vespid, Stenodynerus sp. placed her mud

nest in front of the I. auripes nest in progress.

Discussion

Documented observations of biology and natural history are important especially at

the extremes of geographical range and in unique areas, for a species that is cosmopolitan









in range, such as the genus Isodontia. In particular I. mexicana has become established in

Hawaii and France (Bohart and Menke 1976) providing unique habitats in comparison to

its native North America. Florida populations, as in this case, provide observations in the

southeast extreme of the geographical range. O'Neil and O'Neil (2003) examined a

population in Montana. With the ever-ongoing debate of species and subspecies versus

dines and ecotypes it is important that we identify possible subjects to further examine

the mechanisms of speciation and hazy hybrid zones. For example, the splitting of

Anisota senatoria into A. senatoria and A. peigleri by Riotte (1975) has been called into

question by Tuskes et al. (1996) and through out many groups taxonomists that are

lumperss" or splitterss" are constantly at odds.

In addition, such information is valuable to help identify possible projects for

evolutionary biologists and studies in biogeography. For example, Mark Deyrup and

Thomas Eisner (2003) examined the differences of coloration between Florida

Hymenoptera and their northern relatives (subspecies, clinal types, etc.) utilizing museum

specimens and natural history data from past studies. Their preliminary observations

called for further exploration of Florida biogeography to help recognize distinctive

species and examine mimetic complexes. Mimetic complexes are adaptive syndromes

that reflect the evolutionary history of species, and historical events that cannot be

repeated by an investigator and leave no fossil record (Deyrup and Eisner 2003), yet data

such as these can help examine such events. Isodontia mexicana is a cosmopolitan

species that is easily captured, studied and occurs sympatrically with I. auripes at these

Florida sites. Such data can be extremely useful when examining speciation by









highlighting divergence of these two sympatric sister species and by highlighting the

divergence of mexicana in its geographical extremes.

My study is by no means comprehensive, but does offer substantial data in a unique

part of mexicana range that may offer insight into divergence and speciation processes.

Habitat

Isodontia auripes was not captured in San Felasco Hammock Preserve State Park,

but was captured a few kilometers away at Devil's Millhopper Geological State Park.

Isodontia mexicana was not particularly abundant at San Felasco and I. auripes was

probably present at San Felasco just not captured. All transects with Isodontia were in or

adjacent to sand hill habitat that tended to be xeric. Silver River State Park does not have

sandhill habitat that was particularly of substantial size or xeric in nature and lacks both I.

auripes and I. mexicana. Both species were present in Suwannee River S.P. in both

recently burned and unburned sandhill habitat, although the majority of nests (82% for .

mexicana and 87% for I auripes) were captured in burned habitats (see chapter 4).

Nest Architecture

Several species of Isodontia have nests that contain a common brood cell where up

to 12 larvae will feed on a single prey mass. Isodontia auripes exhibited this behavior

and all nests (apart from single "over-wintering" emergence) of this species had common

brood cells. Although I. mexicana has been reported to have a common brood cell in

some populations(Krombein 1967), the populations I studied had separated brood cells

within each nest (Figure 2-1). Bohart and Menke (1963) reported that some Isodontia

use grass to line the nest. O'Neill and O'Neill (2003) reported that the population of .

mexicana they observed in Montana did not line nest cells. Pupae of both species had

some amount of grass incorporated into the cocoon suggesting there was some manner of









grass lining each cell (Figure 2-3). Krombein (1970) observed larvae oflsodontia

auripes pulling grass fragments from the plugs and incorporating into the spinning of the

cocoon. This behavior may be the explanation for the grass fragments incorporated into

cocoons. The small amount of grass incorporated into the cocoons in addition to the lack

of remaining grass in the chambers suggests that these species in fact do not actively line

the brood cells with grass. The nature of my traps did not allow for direct observation of

pupating activity.

Isodontia females of both species plug the opening of nest cavities with clumps of

loosely packed grass. These plugs of grass resemble broom-like tufts and regularly

extended 30-60 mm beyond the cavity opening and occasionally reached up to 100 mm.

Bohart and Menke (1976) reported these plugs extending only up to 50 mm beyond the

cavity opening. In Montana populations ofl. mexicana, many plugs were flush with the

opening, that tufts apparently being clipped short by the female (O'Neil and O'Neil

2003). I did not observe any of this clipping in Florida. The only nests that did not have

the tufts of the closure plug extending beyond the opening were those that completely

lacked the closure plug. These nests had only the final tightly packed partition

suggesting that the plug had fallen out, the females had not completed her nest at time of

collection or she had died before nest completion.

I observed an interesting deviation of mexicana nest architecture. About 18

nests were found to have one separated brood cell was provisioned with prey and a single

egg, then the remaining nest was packed with both tightly packed partitions and loose

plugs. Only 18 such nests were extracted and recorded, yet in the spring many other

nests yielded only one adult without evidence of other pupae. It was impossible to









determine if these nests were a fall-winter behavior (rather than the other chambers eggs

not successfully hatching or being pillaged by ants) since the emerging adults are quite

destructive of the nest commonly pushing the entire contents of the nest out of the cavity.

These types of nest may be the result of an end of life span behavior of the nesting

females. These fall-winter types were usually found in the fall with the earliest collected

in August. However, normal nests with multiple brood cells were found throughout the

year including in December.

Sex Ratio and Sexual Dimorphism

The two species of Isodontia had dramatically different sex ratios. Isodontia

mexicana displayed a sex ratio of 2.1: 1 (M: F). However, I. auripes displayed a sex ratio

drastically different at 5:1 (M: F). O'Neill and O'Neill (2003) found that males of I.

mexicana tended to emerge from smaller diameter nesting cavities. Yet, the majority of

L. auripes occupied the largest diameter nest (12.7 mm) and one of the few nests in

smaller diameters yielded a female (7.9 mm). Therefore, this conclusion does not seem

applicable. An alternative explanation is the Trivers-Willard hypothesis which states that

sex allocation is condition dependant (Trivers and Willard 1973). This assumes that

females can control the sex of the offspring, and it has been shown that nest-provisioning

hymenopterans precisely determine the sex of each offspring (Green et al., 1982, O'Neill

2001 O'Neill and O'Neill 2003). In fact, the majority oflsodontia nests were located in

recently burned (within 1 and 2 years) sandhill habitat. Females can be expected to

produce more of the sex for which quality makes the greatest difference in reproductive

success (Clutten-Block et al 1984, Miller and Aviles 2000). Therefore, in poor

conditions, males should be produced and in good conditions females should be

produced, assuming the burned condition is a detrimental condition. Unfortunately, the









relatively low amount of Isodontia nests captured in unburned areas did not allow a

meaningful comparison. It can be speculated that the fire event eliminated nesting

materials and reduced prey populations. The sites were surveyed for two years beyond

the fire event. Grasses tend to respond positively and quickly after a fire event. The

reduction of plant biomass may have provided less harborage for prey items and therefore

easier hunting for the wasps. Yet, the sex ratios were equal in both years following the

fire event. The relative amount of prey and nesting material needs to be known as well

as further study beyond the fire event to detect any response lag. However, I. mexicana

did not exhibit such a skewed sex ratio that I auripes displayed. Since both species have

similar biology (same prey, nesting habitat, nesting materials) and occur sympatrically,

could interspecific competition be the driving force? Unfortunately the current data set

cannot suggest any answers with any kind of confidence. Additional research focusing

on interspecific competition is needed.

Prey

The prey provisioned by both Isodontia species has substantial overlap between my

records and those reported in the literature (Table 2.4 & 2.5). Orocharis luteolira

(Gryllidae: Eneopterinae) and Belocephalus sp (Tettigoniidae: Copiphorinae) were the

only prey for I. mexicana that were not reported in the literature. Oecanthus celerinictus

(Gryllidae: Oecanthinae), Odontoxiphidium apterum (Tettigoniidae: Conocephalinae) and

Neoconocephalus sp (Tettigoniidae: Copiphorinae) were the only prey for I. auripes not

reported in the literature. Both Isodontia species examined provisioned 3 species in

common, but I. mexicana provisioned 6 unique species while I. auripes provisioned 2

unique species (Table 2.1). Under closer observation with more nests dissected, the

range of prey items for both Isodontia may become more similar. Amount of prey









provisioned varied greatly. Provisioned prey ranged from 1-19 prey items per nest.

However, since these wasps prey on both adult and juvenile prey, biomass of the

provisions is probably more important than quantity. Nests and/or brood cells that had

one or few prey items tended to contain large adult tettigoniids and those nests and/or

brood cells with many prey tended to contain juveniles and/or small species of gryllids.

Therefore, prey per nest and prey items per brood cell are statistics of questionable value.

Nesting females seem to be filling brood cells rather than provisioning a particular

number of prey items per egg.


These Isodontia species displayed the typical sexual size difference found in the

Sphecidae, suggesting a possible provisioning strategy by nesting females. Female-

biased provisioning has been shown to occur in Montana populations (O'Neill and

O'Neill 2003) of mexicana, and it also seems to occur in Florida populations of .

mexicana. However, auripes exhibited this sexual size trend, yet the communal brood

chamber nest architecture rules out any provisioning difference as the cause for the size

difference since both sexes regularly emerged from the same nest. A possible

explanation could be that females deposit female eggs on prey earlier than male eggs.

Since eggs that are deposited earlier tend to hatch earlier, female larvae would have more

time with the prey mass and would most likely consume more of the prey mass.

Krombein (1970) reported that the last egg to hatch in a nest of auripes where 6 eggs

were laid actually died from lack of food.

One nest experienced supersedure. A vespid, Stenodynerus sp., usurped a nesting .

auripes female and placed her mud nest in front of the I. auripes nest in progress of

provisioning. Two auripes successfully developed behind the Stenodynerus nest, but









were unable to break through the mud partitions of the vespid nest and subsequently died.

These I. auripes adults were of normal size suggesting that the female oviposits on prey

after there are a number of sufficient prey items in the communal chamber to support that

egg. This was an isolated observation, however and more data are needed to substantiate

this hypothesis.

Conclusion

Nest structure and prey in the observed Florida populations were similar to those

reported by Medler (1965), Krombein (1967), Bohart and Menke (1976) and O'Neil and

O'Neil (2003). Both Isodontia mexicana and I. auripes displayed sexual dimorphism

typically found in the Sphecidae. More importantly, these two sister species occur

sympatrically and have a broad range of overlap in biology and prey species, yet they had

extremely different sex ratios. Sex allocation is typical in the Sphecidae and a skewed

ratio suggests a harsh environment. There is substantial overlap in prey items, but there

are unique prey items to each species. Whether those prey items remain unique as

sampling for provisioned prey is increased is unknown. In addition, prey populations

were not sampled or estimated leaving possible disparity of unique prey unknown.

Whether the skewed sex ratio is a result of direct interspecific competition or differences

in mutually exclusive prey populations, the relationship of these two sympatric sister

species may offer evaluation of competitive exclusion, divergence and possible

speciation events.

Acknowledgements

All research and collections were completed with permission of the Florida

Department of Environmental Protection Division of Parks and Recreation under permit

numbers 11250310 and 08170410.






























Figure 2-1. Cross section of Isodontia mexicana nest in a 12.7mm cavity


Figure 2-2. Isodontia auripes larvae on provisioned Scudderiafurcata








































Figure 2-3. Isodontia cocoon


Frequency of cavities nested in by I. auripes and I. mexicana


90

80

70

u 60

S50 1. auripes
o 40 mexicana
40

2 30
U-
20

10


3.2 4.8 6.4 7.9 12.7
Cavity diameters (mm)


Figure 2-4. Frequency of cavities nested in by I. auripes and I mexicana












Summary of emerged Isodontia sp.


nests


male


1 auripes
SI. mexicana


female


Figure 2-5. Summary of emerged Isodontia mexicana and Isodontia auripes from
captured nests









Table 2-1. Comparison of Isodontia auripes and I. mexicana
Isodontia auripes Isodontia mexicana
Identification Violaceous wings Clear wings, with black veins and
Red-brown legs smoky brown along anterior margin
Male: 18 mm, Female: 19 Black legs
mm Male: 16 mm, Female: 17 mm
Architecture Single chamber 50-60mm 3 (1-4) chambers 20-30mm each
Prey (Gryllidae: Oecanthinae); (Gryllidae: Oecanthinae);
(Bold = Prey Oecanthus celerinictus T Oecanthus quadripunctatus
provisioned by Walker 1963 Beutenmuller 1894
both species) Oecanthus niveus (De Geer (Gryllidae: Eneopterinae);
1773) Orocharis luteolira, T Walkerl969
Oecanthus spp. (juv). (Tettigoniidae: Conocephalinae);
(Gryllidae: Eneopterinae); Odontoxiphidium apterum Morse
Orocharis luteolira, T 1891
Walkerl969 (Tettigoniidae: Copiphorinae);
(Tettigoniidae: Belocephalus sp.
Conocephalinae); (Tettigoniidae: Phaneroperinae);
Odontoxiphidium apterum, Scudderia sp. (juv)
Morse 1891
Conocephalus brevipennis
Scudder 1862
(Tettigoniidae:
Copiphorinae);
Neoconocephalis spp. (juv)
(Tettigoniidae:
Phaneroperinae);
Scudderia furcata Bruner
1878
Scudderia sp. (juv)
Dimorphism Females: 3.30 + 0.397 Females: 3.10 + 0.20
(Mean head width Males: 3.01 + 0.259 Males: 2.84 + 0.18
(mm))
Sex ratio (M:F) 5:1 1.2:1
Cavity diameters 6.4, 7.9, 12.7 4.8, 7.9, 12.7
(mm) nested
Habitat Sandhill, Ravine (adjacent to Sandhill, Ravine (adjacent to
sandhill), Pine Flatwoods sandhill), Pine Flatwoods, Mesic
hardwood hammock
State Parks Suwannee River State Park, Suwannee River State Park,
Mike Roess Gold Head Mike Roess Gold Head Branch State
Branch State Devil's Devil's Millhopper Geological State
Millhopper Geological State ParkSan Felasco Hammock Preserve
Park State Park









Table 2-2. Frequency of cavities nested in by I. auripes and I. mexicana
Cavity diameter (mm) I. auripes I. mexicana
3.2 0 0
4.8 1 0
6.4 0 2
7.9 3 20
12.7 85 68
Total nests 89 90


Table 2-3 Summary of emerged I auripes and I. mexicana
Number of nests Male Female Total adults
(Emerged)
Isodontia auripes 90 66 53 119
Isodontia mexicana 89 131 23 154
Isodontia (unknown 56

Total 235 273









Table 2-4 Prey records for mexicana
Prey Present O'Neill & Bohart & Krombein Lin Medler
study O'Neill 2003 Menke 1976 1967 1966 1965
Gryllidae: X X
Gryllinae
Gryllus sp.
Gryllidae: X X
Eneopterinae
Orocharis sp X X
Orocharis X
luteolira T Walker
Gyllidae: X X X X X X
Oecanthinae
Oecanthussp. X X X X X X
0. exclamationis
Davis*
0. nigricornis F X X
Walker
0. quadripunctatus X X X
Beutenmuller
0. fultoni T X X
Walker
0. niveus X X X
(DeGeer)
0. saltator Uhler X
0. celerinictus T
Walker*
Neoxabea sp. X X
N. bipunctata X
(DeGeer)
Tettigoniidae: X X X
Conocephalinae
Conocephalus sp X X X
C. brevipennis
(Scudder)*
C. fasciatus X
(DeGeer)
C. saltans X
(Scudder)
Odontoxiphidium X X X
sp
0. apterum Morse X X
Orchelimum sp. X X
Tettigoniidae:
Copiphorinae










Table 2-4 Continued
Prey Present O'Neill & Bohart & Krombein Lin Medler
study O'Neill Menke 1967 1966 1965
2003 1976
Belocephalus sp. X
Neoconocephalus X
sp.
Tettigoniidae:
Phaneropterinae
Scudderia sp. X X
Scudderiafurcata
Brunner*
Tettigoniidae:
Tettigoniinae*
Atlanticus sp.*
A. gibbosus
Scudder*
Neobarretta sp. X
Not found but present inlsodontia auripes nests

Table 2-5 Prey records for Isodontia auripes
Prey Present Bohart & Krombein Krombein
study Menke 1976 1970 1967
Gryllidae: Gryllinae
Gryllus sp.
Gryllidae: Eneopterinae X X X
Orocharis sp X X X
Orocharis luteolira T X X
Walker
Gryllidae: Oecanthinae X X X
Oecanthus sp. X X X
0. exclamationis Davis X
O. nigricornis F Walker*
0. quadripunctatus
Beutenmuller*
0. fultoni T Walker*
0. niveus (DeGeer) X X
0. saltator Uhler X
0. celerinictus T Walker X
Neoxabea sp. X X
N. bipunctata (DeGeer) X
Tettigoniidae: X X
Conocephalinae
Conocephalus sp X X
C. brevipennis (Scudder)*










Table 2-5. Continued
Prey Present Bohart & Menke Krombein Krombein
study 1976 1970 1967
C. fasciatus (DeGeer)*
C. saltans (Scudder)*
Odontoxiphidium sp*
O. apterum Morse X
Orchelimum sp. X X
Tettigoniidae: X
Copiphorinae
Belocephalus sp.*
Neoconocephalus sp. X
Tettigoniidae: X X
Phaneropterinae
Scudderia sp. X X
Scudderiafurcata
Brunner*
Tettigoniidae: X X
Tettigoniinae
Atlanticus sp. X X
A. gibbosus Scudder X
Neobarrettia sp.*
Not found but present in Isodontia mexicana nests














CHAPTER 3
SPIDER PREY IN NESTS OF THE MUD DAUBER WASP Trypoxylon lactitarse
(HYMENOPTERA: SPHECIDAE)

Abstract

Prey from 88 nests of Trypoxylon lactitarse in five state parks in north central

Florida were examined, yielding 1173 individual spiders from 15 families, 40 genera and

64 species. Overall, Neoscona sp. was the most commonly collected prey species

(23.78%), followed by Mimetus sp. (12.27%) and Pisaurida mira (8.86%). Araneidae

(56.26%) was the most commonly collected family, followed by Mimetidae (12.27%)

and Pisauridae (10.57%). Trypoxylon lactitarse tended to be a generalist in its prey

preference with a fairly even diversity of prey captured. Although the majority of prey

items were common web-spinners, many rarely surveyed spiders, such as Aniphedids and

some Salticids, were collected. Since T lactitarse hunts for spiders in wide-ranging

microhabitats and with more intensity than human collectors, surveying nest contents is

an extremely useful tool to expand spider species richness estimates, species inventories,

and natural history data.

Introduction

Trypoxylon lactitarse is a solitary wasp found in the western hemisphere from

Canada to Argentina. Females of T lactitarse nest in cavities constructing linear cells

subdivided by partitions of mud and provision these cells with numerous paralyzed

spiders. Data concerning prey are normally difficult to obtain, but this wasp deposits

prey in nests that are easily collected (Camillo and Breviscovit 1998, Rehnberg 1987).









The wasp nests in preexisting cavities and readily accepts trap-nests, allowing for prey

data to be easily collected. Previous studies (Rau 1928, Krombein and Evans 1954,

Krombein 1956, 1967, Medler 1965, Lin 1969, Coville 1979, 1981, 1982, Coville and

Coville 1980, Genaro et. al. 1989, Camillo et. al. 1993, Genaro and Alayon 1994,

Jimenez and Tejas 1994, Camillo and Brescovit 1999, 2000) have shown that different

species of Trypoxylon have different prey preferences. These differences in prey

preference can be in proportion of each family, genus, or species taken; amount of

families taken; and relative proportion of spider groups (orb-weaving, hunting or

wandering) in the prey. Coville (1987) suggested that preferences for different species or

species groups may arise because of different hunting behaviors of the wasps, different

microhabitats hunted, or the wasps are conditioned to a certain type of spider. Some

species capture spiders predominately from one family and occasionally spiders from

other families (Camillo and Brescovit 2000), while some species, including T. lactitarse,

prey on spiders of many different families (Camillo and Brescovit 1999).

Nests of Trypoxylon lactitarse provide large amounts of spiders from various

families, including spiders rarely caught by humans. Because these wasps hunt

extensively in different microhabitats and areas rarely sampled by humans, sampling their

nests and prey may provide additional information on spiders in the area.

I set out to investigate the following questions about Trypoxylon lactitarse and its

nest contents to determine: 1) What prey is T. lactitarse provisioning at these Florida

sites? 2) Is T lactitarse a generalist or specialist in terms of prey provisioned in Florida?

3) What, if any, is T. lactitarse's prey preference? 4) Does T. lactitarse's prey preference

seem to differ between sites? 5) What are the benefits and problems with using this wasp









(and potentially other spider-provisioning trap-nesters) as a sampling tool for estimating

spider abundance and species richness?

Methods and Materials

Tools and Trap Preparation

The traps used in this study were fabricated from seasoned 37-mm x 86-mm x 2.4-

m pine/spruce timbers obtained from a local home improvement store. The pine/spruce

timbers were cut into 100, 10-cm-long blocks. Two cavities of one of five diameters

(3.2, 4.8, 6.4, 7.9 or 12.7-mm) were drilled into each block. Cavities were drilled to a

depth of 80 mm on each short side (the 37-mm side), offset approximately 10-mm from

the center point. Traps were assembled using one block of each diameter with the

smallest cavity on top and the largest on the bottom. Blocks were stacked so that no

cavity was situated directly above or below a cavity in the adjacent block. The five

blocks were bound together with strapping tape (3M St Paul, Minnesota), and 16-gauge

wire was used to further bind the stack and suspend the trap from trees and shrubs at the

field sites. Each bundle of five blocks was considered to be a single trap.

Field Sites

I set trap nests at five locations: 1) Suwannee River State Park in Suwannee County

(30 23.149' N, 083 10.108' W), 2) Mike Roess Gold Head Branch State Park in Clay

County (29 50.845'N, 081 57.688' W), 3) Devil's Millhopper Geological State Park in

Alachua County (29 42.314'N, 082 23.692' W), 4) San Felasco Hammock Preserve State

Park (29 42.860' N, 082 27.656' W) in Alachua County and 5) Silver River State Park in

Marion County (29 12.317'N, 082 01.128' W). The habitats surveyed at Suwannee

River State Park were burned and unburned sand hill habitat, while the habitat at Mike

Roess Gold Head Branch State Park was burned sand hill pineland and ravine. Sites at









San Felasco Hammock Preserve State Park consisted of upland and mesic hardwood

hammock. Surveyed areas of Devil's Millhopper Geological State Park consisted of pine

flatwood habitat and sites at Silver River State Park consisted of river habitat and upland

mesic forest. Descriptions of these habitats can be found in Franz and Hall (1991).

Field Placement

Transects were set up with 10 traps placed approximately 10 m apart and hung

approximately 1.5 m off the ground on trees or limbs with placement on dead standing

wood preferred. Transects were initially established (direction and distance from center

of plot) randomly. Four transects were established in Suwannee River State Park while

three transects were established in Mike Roess Gold Head Branch State Park. Three

transects were established in San Felasco State Park but size constraints only allowed a

single transect in Devil's Millhopper State Park. Finally, two transects were set up in

Silver River State Park. Transects were in the field from April 2003 until January 2005.

Field Collection and Laboratory Rearing

Traps remained in the field two years and were checked monthly. Preliminary field

tests revealed that one-month intervals were sufficient to avoid trap saturation (no

available cavities). Traps were considered occupied when insects were observed actively

nesting, harboring or had sealed a cavity with mud or plant material. Occupied traps

were removed and replaced with a new trap. These occupied traps were brought into the

forest entomology lab at the University of Florida in Gainesville, FL, for processing.

Occupied blocks were removed for observation while unoccupied blocks were

reincorporated into replacement traps. Each occupied cavity was given a unique

reference number.









In order to examine prey items of Trypoxylon lactitarse, traps were dissected for

nest contents when a wasp was encountered provisioning, guarding, or sealing a nest

during collection runs. Prey items were removed, preserved, and given the same

reference number as the cavity from which they had been removed. After the contents

were extracted, the wood block was reused in replacement traps. These processed blocks

were re-drilled to the next cavity diameter to eliminate any alterations or markings (either

physical or chemical) by the previous occupant prior to reuse.

Specimen Identifications

All specimens were identified by the author with most of the spider prey specimens

identified and verified by G. B. Edwards at the Florida State Collection of Arthropods in

Gainesville, Fl. Voucher specimens have been deposited at the Florida State Collection

of Arthropods.

Statistical Analysis

Similarity was calculated with Jaccard's similarity index (ISj) (Southwood 1978).

This index is the proportion of the combined set of species present at either site that are

present in both sites. This value ranges from 0 to 1, with 0 meaning no similarity (no

species in common) in both sites and 1 meaning all species are present at both sites. The

value is calculated using the following equation:

ISj= c /(a + b + c)

Where c is the number of species common to both sites and a and b respectively are

the species exclusive to those sites

Similarity was also calculated with Sorensen's similarity index (ISs) (Sorensen

1948). This index is the proportion of the combined set of species present at both sites

that are present in both sites. This value ranges from 0 to 1, with 0 meaning no similarity









(no species in common) in both sites and 1 meaning all species are present at both sites.

The value is calculated using the following equation:

ISs = 2c / (a + b)

Where c is the number of species common to both sites and a and b are respectively the

total number of species at each site '

Chao-Jaccard raw (uncorrected for unseen species) abundance-based similarity

index, Chao-Jaccard estimate (corrected for unseen species) abundance-based similarity

index, Chao-Sorensen raw (uncorrected for unseen species) abundance-based similarity,

and Chao-Sorensen estimate (corrected for unseen species) abundance-based similarity

(Chao et al. 2005) was calculated with EstimateS 7.5 (Colwell 2005).

Diversity was calculated using Simpson's index of diversity and Simpson's index

of dominance (Simpson 1949). Simpson's index of diversity values range from 1 to S,

where S is the total number of species. Simpson's index of dominance ranges from 0-1.

Simpson's index of dominance, h is given by:


X= (n/N)2
1=1

where n is the total number of organisms of the ith species and N is the total number of

organisms of all species.

Simpson's Index of Diversity is given by: 1/k

Diversity was also calculated using the Shannon-index (Shannon and Weaver 1949) H',

given by:


H'= pi In pi
1=1









where pi = n/ N and n is the total number of organisms of a particular species and N is the

total number of organisms of all species

Diversity is a combination of species richness (number of species) and evenness of

species abundance. Therefore, Shannon's index of evenness, J (Pielou 1966), is given

by:

J = H' / In s where s is the total number of species

Species richness was estimated using rarefaction curves (Colwell et al. 2004). This

estimate of species richness is based on a sub-sample of pooled species actually

discovered. In addition, three non-parametric species richness estimators, ACE

(Abundance-based Coverage Estimator: Chao et al. 2000, Chazdon et al. 1998), first

order jackknife (Burnham and Overton 1978, 1979, Smith and van Belle 1984, Heltshe

and Forrester 1983) and Chao 1(Chao 1984) were used. These estimators produce

estimates of total species richness including species not present in any sample. Most of

the indices and all of the richness estimators were computed using EstimateS 7.5

(Colwell, 2005).

Results

I examined 88 nests of Trypoxylon lactitarse and found 1173 individual spiders

from 15 families, 40 genera, and 64 species from all five state parks (Table 3-1). Overall,

Neoscona sp. was the most commonly collected species (23.78%), followed by Mimetus

sp. (12.27%) and Pisaurida mira (8.86%). See Table 3-1 for a summary of captured

species tabulated by site of capture. When subdivided by site, (Figures 3-2 through 3-6)

the top five species of spider prey for each site was similar to overall pooled results. The

most abundant species, Neoscona sp., was the most abundant species for three of the sites

and second and third most abundant for the remaining two sites. For all sites, the five









most abundant species were included in the ten most abundant species for pooled data,

except for 2 species at Gold Head Branch. At the Gold Head Branch site, the most

abundant species, Trachelas sp., is third most abundant species for pooled data, Theridion

sp., and fifth most abundant species, Neoscona crucifera, were not included in the pooled

top ten most abundant species captured.

The most abundant families of spiders collected were Araneidae (56.26%),

Mimetidae (12.27 %), Pisauridae (10.57%), Salticidae (6.82%) and Tetragnathidae

(6.82%) (Figure 3-7).

Prey diversity as reported by Shannon's and Simpson's indices is similar among

sites except for the Simpson's index value for Devil's Millhopper which is remarkably

different at 6.81 (Table 3-3). The Shannon evenness index for all but one site is above

0.7, (Table 3-3) suggesting a fair degree of evenness. The evenness of the prey diversity

is illustrated on the rank proportional abundance graphs (Figures 3-8, 3-9).

Similarity of prey between sites, reported by Jaccard's similarity, Sorensen's

similarity, Chao-Jaccard raw (uncorrected for unseen species) abundance-based

similarity, Chao-Sorensen raw (uncorrected for unseen species) abundance-based

similarity, Chao-Jaccard estimate (corrected for unseen species) abundance-based

similarity, Chao-Sorensen estimate (corrected for unseen species) abundance-based

similarity (Chao et al. 2005) indexes are summarized in Table 3-3. Overall, the classic

formulas for Jaccard's and Sorensen's indices gave the lowest values with Jaccard's

index being the lower of the pair. This is intuitive since these indices are calculated with

actual observed species. All the Jaccard's indices, including the Chao versions, were

more conservative by yielding lower values than their Sorensen counterparts. The only









exception to this trend was a single site comparison, Silver River vs. San Felasco were

both estimate versions of Chao-Jaccard's and Chao-Sorensen's gave a value of 1.0 for

complete similarity.

Species richness estimates given by rarefaction, first order jackknife, ACE

(abundance-based coverage estimator) and Chao 1 estimator are given in Figures 3-10.

The only site at which all these estimators stabilized, however, was at San Felasco State

Park and the Chao 1 and ACE estimators stabilized for Gold Head Branch. The other

sites and estimators did not completely reach an asymptote and can be viewed with

skepticism (figure 3-11).

Discussion

Prey diversity varied among sites due to a large difference in species composition;

however, the Shannon evenness index for all but one site is above 0.7, suggesting a fair

amount of evenness. This trend can also be seen on the rank proportional abundance

graphs (figures 3-1, 3-2). This level of evenness suggests that Trypoxylon lactitarse is

not specializing on a few prey species with occasional secondary species, but rather

behaving as a generalist and hunting a wide variety of available spider prey including

both web spinning and hunting spiders.

The somewhat low levels of similarity for Jaccard's and Sorenson's (Table 3-6)

indices between all sites suggest a distinct variation in spider prey composition. The

Chao-Jaccard estimate abundance-based and the Chao-Sorensen estimate abundance-

based similarity indexes show a higher degree of similarity between sites than their raw

estimate counterparts. These estimate-based indexes are corrected for under-sampling

bias and suggest that sites are more similar that the current observations reveal. Since

under-sampling or limited sampling effort is the generally the case, the Chao-Jaccard and









Chao-Sorensen estimates would be the best choice. Of these two, the Chao-Jaccard

estimate is generally the more conservative yielding (slightly) lower estimates of

similarity. The highest estimated similarity values were between Silver River state park

and San Felasco state park and the lowest similarity was between Devil's Millhopper

state park and Gold Head Branch state park, yet the estimators for these sites did not

stabilize suggesting these sites were under-sampled. These findings are expected since

the similarity of the respective habitats coincides with prey item similarity. This further

suggests that Trypoxylon lactitarse is a generalist predator capturing prey that is abundant

in the habitat and not searching for a particular species within any habitat. Yet, since the

sample sizes for each of the sites were different, due to opportunistic nature of the

sampling, an additional study with a more systematic, even sampling focusing on

obtaining nest contents is needed to provide more confident results. Furthermore,

estimators for Devil's Millhopper and Mike Roess Gold Head Branch did not stabilize

due to small sample sizes, so these results should be viewed with skepticism. The

estimators for the remaining sites did stabilize (except for the Chao 1 estimator in Silver

River) and we can be confident in these species richness estimations.

Finally, do Trypoxylon lactitarse nest provisions provide useful data on spider

populations? The characteristics of the nest provisions confirm that T lactitarse is a

generalist predator of spiders, which is ideal for surveying a population. Intuitively, these

wasps collect spiders much more intensively and efficiently than human collectors.

These nests also yielded a fair amount of rare species uniquee, singletons, and

doubletons) further suggesting a complete survey of the target group. These three factors

suggest that nests of T lactitarse are an ideal survey tool for spiders. In addition, species









richness estimators (such as in the freeware EstimateS (Colwell, 2005)) extrapolate total

species richness for a site and, therefore, suggest a sufficient level of species inventory

for a particular site. Sufficient sampling, however, is crucial for successful estimation.

In most sites, the estimators did not stabilize due to undersampling (figure 3-11). The

estimators for San Felasco site did stabilize and two estimators stabilized at Gold Head

Branch suggesting that sufficient samples were taken to estimate species richness for

Trypoxylon lactitarse provisioned prey with confidence at those sites (figure 3-11). The

fact that estimators did not stabilize for Silver River and half of the estimators for Gold

Head Branch did not stabilize is not surprising because of the smaller sample sizes. What

is surprising, however, is that none of the estimators stabilized for Suwannee River.

Although similar numbers of samples (contents of a single nest) were taken in the two

parks, all estimators for San Felasco stabilized between 10-20 samples while no

estimators stabilized for Suwannee River after 28 samples. I suggest that each survey

effort monitor estimators for stabilization for each site individually in order to determine

sufficient sampling.

These findings in no way suggest that this sampling represents the total spider

fauna of the particular sites, but simply that we have sufficiently examined the prey of

Trypoxylon lactitarse. These estimators are indeed practical to determine the richness of

spiders preyed upon by the wasps and when this sampling has been sufficient. As

discussed earlier, T. lactitarse is a generalist in Florida and in tropical regions, and

provides a wide range of spider prey. Trap-nests also have the advantages of other spider

provisioning wasps being trapped. Trypoxylonjohnsonii, T. carinatum, T. collinum

collinum, T. clavatumjohanis, and T clavatum clavatum are other spider provisioning









wasps that were also captured at these Florida field sites, but their nest contents were not

extracted. Even though some wasps may be specialists (Camillo and Brescovit 2000) in

addition to generalists such as T lactitarse (Camillo and Brescovit 1999), these wasps

intuitively search longer and in different microhabitats and, therefore, provide more

abundance and possibly variety of spiders than hand collecting alone. Yet, some wasps

do have preferences for one family or another. Even the generalist hunters may

periodically favor one group of spiders that are locally abundant or more easily captured

at that time over groups they would normally prey upon. It may be prudent, therefore, to

take samples at various times of the year to avoid temporal population cycles of spiders.

This technique for sampling spider fauna would be ineffective alone, however, as a

part of a structured inventory protocol including other techniques, such as hand collecting

and pitfall traps, may provide more complete and accurate cataloguing of spider faunas.

This is especially true since underestimates have been shown to most commonly be

derived from shortcomings of sampling techniques rather than sampling effort (Longino

and Colwell 1997, King and Porter 2005). When various techniques are integrated

together to create a structured inventory procedure, such as the Ants of the Leaf Litter

(ALL) protocol for sampling ant communities (Agosti et al. 2000) and the methodology

proposed by Coddington et al (1991) for spiders, they can be extremely powerful and

reliable tools (Toti et al. 2000). Various techniques such as Malaise traps (Jennings and

Hilburn 1988) and trap nests can be used in addition to the standard hand collecting,

sweeping, and pitfall trapping, to provide an efficient and complete method of

determining spider fauna of an area when long term sampling is an option.






49


Acknowledgements

I thank G. B. Edwards of the Florida State Collection of Arthropods in Gainesville,

Florida, for the voluminous amount of identification, verification, and help with all of the

spider specimens. All research and collection were completed with permission of the

Florida Department of Environmental Protection Division of Parks and Recreation under

permit numbers 11250310 and 08170410







50



Ten most abundant spider prey species


300
250
200
150
100
50
0


1 2 3 4 5 6 7 8 9 10


Abundance -- Percent

Figure 3-1. Ten most abundant spider prey species for all sites pooled


Suwannee River State Park

70 20
18
60
16
50- 14


0= 10 2
30- 8 4
20- 6
4
10
2
i: B I :s


Figure 3-2. Five most abundant spider prey species at Suwannee River State Park


SAbundance -- Percent








51




San Felasco State Park


120 25

100

80 1

60 2
t-o
10"
< 40

20 5

0 0
Neoscona sp. Pisaurina mira Mimetus sp. Wagneriana Eustala anastera
tauricomis

1 2 3 4 5

Abundance + Percent

Figure 3-3. Five most abundant spider prey species at San Felasco State Park



Silver River State Park


100 60
90
80- 50
70 40
40
o 60
CC
'a 50 30
C C
4 40-
20
30
20 10
1010
10


I i Abundance Percent

Figure 3-4. Five most abundant spider prey species at Silver River State Park








52




Goldhead Branch State Park

30 35

25 30

25
S20
20 "
15
15 I
10 1
10

5 5

0 0
Trachelas sp. Neoscona sp. Theridion sp. Thiodina sylvana Neoscona
crucifera

1 2 3 4 5

Abundance --- Percent

Figure 3-5. Five most abundant spider prey species at Gold Head Branch State Park



Devil's Millhopper State Park


40 45

35 40

30 35

25-
030
5 5
20 25 c--
-a
C-
020 D
< 15
15
10 10

5 -

0 0
Neoscone sp. Pisaurina mire Wagnetiana Eustale anastera Mecynogee
tauricomis lemniscata

1 2 3 4 5


Figure 3-6. Five most abundant spider prey species at Devils' Milihopper State Park












Abundance and percentage of spider prey families captured at all sites

700 60

600 50

500
40

S400
30
300

20
200

100 10

0- 0






g Number of individuals -- Percent

Figure 3-7. Abundance and percentage of spider prey families captured at all sites


Pooled Rank Proportional Abundance


0.25



0.2

0

So0.15



0.


S0.05



0


1 10

Species rank


Figure 3-8 Pooled rank proportional abundance of spider species collected from five
Florida state parks.











Site Rank Proportional Abundance


-- DM
-u-SR
-A- SW
-x-SF
-- GH


Species rank

Figure 3-9. Site rank proportional abundance of spider species collected at each state
park. DM = Devil's Millhopper State Park, SF = San Felasco State Park, GH
= Mike Roess Gold Head Branch State Park, SR = Silver River State Park,
SW = Suwannee River State Park

1 st order Jackknife species richness estimator ACE (Abundance-based Coverage Estimator)

80


i *l 1 l
70
1' 60
50
o 40










M SF DM SR SF GH SW
Chao 3 Species richness estimator
lho eSobs (Sample based rarefaction)
50
45
40
35
30
25
S20
E 15-
z10
5
DM SR SF GH SW
Site

Figure 3-10. Species richness estimation for spider prey tabulated by site: DM = Devil's
Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold
Head Branch State Park, SR = Silver River State Park, SW = Suwannee River
State Park


Neoscona sp.


EIS-II














Chao 1









'---fl-n--

0 5 10 15 20 25 30 35
Cumulative number of samples

-*--sf --sw gh sr


Sample based rarefaction











0 5 10 15 20 25 30 35
cumulative samples

---SF---SW GH SR


ACE











0 5 10 15 20 25 30 35
Cumulative samples

---SF---sw gh sr

Jack 1











0 5 10 15 20 25 30 35
cumulative number of samples

--sf--- SW gh sr


Figure 3-11. Species richness estimator performance for spider prey tabulated by site:

DM = Devil's Millhopper State Park, SF = San Felasco State Park, GH =

Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW

= Suwannee River State Park










Table 3-1 Spiders found as prey in nests of Trypoxylon lactitarse in north central Florida
Family Genus Species DM SR SF GH SW Sum
Agelenidae Agelenopsis sp. 1 0 0 0 2 3
Aniphedidae sp. 0 0 1 1 3 5
Anyphaenidae Hibana sp. 0 1 1 0 1 3
Anyphaenidae Hibana velox 0 0 10 0 3 13
Anyphaenidae Lupettiana mordax 2 0 1 0 1 4
Araneidae Acacesia hamata 1 0 2 0 5 8
Araneidae Araneus bicentenareus 0 1 11 0 1 13
Araneidae Araneus juniperii 0 0 3 0 15 18
Araneidae Araneus miniatus 1 0 1 1 9 12
Araneidae Araneus pegnia 1 1 2 1 8 13
Araneidae Araneus sp. 1 0 18 1 14 34
Araneidae Argiope aurantia 0 0 0 0 1 1
Araneidae Argiope sp. 0 0 0 0 1 1
Araneidae Eriophora ravilla 0 0 4 0 0 4
Araneidae Eustala anastera 0 5 25 0 5 35
Araneidae Eustala sp. 5 48 29 0 9 83
Araneidae Kaira alba 0 0 1 0 0 1
Araneidae Larina direct 1 0 0 0 1 2
Araneidae Mecynogea lemniscata 6 0 15 0 52 71
Araneidae Metapeira sp. 0 0 0 0 6 6
Araneidae Metazygia zilloides 0 1 0 0 0 1
Araneidae Metepeira labyrinthea 0 0 7 0 0 7
Araneidae Neoscona arabesca 0 0 6 0 3 9
Araneidae Neoscona crucifera 3 0 0 2 0 5
Araneidae Neoscona sp. 37 87 100 15 41 279
Araneidae Ocrepeira sp. 0 1 9 0 1 11
Araneidae Scoloderus sp. 0 1 0 0 0 1
Araneidae Wagneriana tauricornis 7 2 28 0 8 45
Araneida Parauixia sp. 0 0 1 0 1 2
Clubionidae Elaver except 0 0 1 1 0 2
Corinnidae Trachelas similes 0 0 1 1 0 2
Corinnidae Trachelas sp. 0 0 0 24 1 25
Mimetidae Mimetus sp. 7 17 58 0 62 144
Philodromidae Philodromus sp. 1 0 1 0 1 1 1
Philodromidae Philodromus sp. 2 0 2 0 0 1 3
Philodromidae Philodromus sp. 3 0 0 0 0 1 3
Pisauridae Dolomedes albineus 0 3 1 2 3 9
Pisauridae Dolomedes sp. 0 1 4 0 2 7
Pisauridae Pisaurina mira 17 0 76 0 11 104
Pisauridae Pisaurina sp. 0 0 0 0 2 2













Table 3-1 Continued.
Family Genus Species DM SR SF GH SW Sum
Salticidae Hentzia mitrata 0 0 0 0 2 0
Salticidae Lyssomanes viridis 0 0 0 7 13 15
Salticidae Metacyrba floridana 0 0 0 0 1 1
Salticidae Phidippus pulcherrimus 0 0 2 0 0 2
Salticidae Phidippus regius 0 0 0 1 0 1
Salticidae Platycryptus undatus 0 0 0 1 2 3
Salticidae Thiodina sp. 0 0 1 0 0 1
Salticidae Thiodina sylvana 2 1 24 7 20 54
Salticidae Zygoballus sexpunctatus 0 0 0 0 3 3
Segestriidae Ariadna bicolor 0 0 1 1 0 2
Tetragnathidae Leucauge venusta 0 4 11 0 2 2
Tetragnathidae Leucauge sp. 0 1 1 0 0 17
Tetragnathidae Nephila clavipes 2 0 24 0 37 61
Theridiidae Argyrodes sp. 0 0 0 0 1 1
Theridiidae Theridion sp. 0 0 7 11 0 11
Thomisidae Misumenops oblongus 0 0 0 0 2 11
Thomisidae Misumenops sp. 0 0 0 0 1 2
Thomisidae Synema parvula 0 0 0 0 2 2
Thomisidae Tmarus sp. 0 0 0 24 1 1
Argla giaparatia 0 0 0 0 1 1
Total 94 178 467 73 361 1173
DM = Devil's Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess
Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State
Park










Table 3-2. Similarity indexes and comparisons for spider prey
Site Observed Jaccard Chao- Chao- Sorensen Chao- Chao-
shared (Classic) Jaccard Jaccard Classic Sorensen Sorensen-
species raw estimate raw estimate
abundance- abundance- abundance- abundance-
based based based based
DM vs. 6 0.214 0.577 0.605 0.353 0.731 0.754
SR
DM vs. 13 0.351 0.749 0.789 0.52 0.856 0.882
SF
DM vs. 6 0.222 0.264 0.31 0.364 0.417 0.474
GH
DM vs. 15 0.326 0.756 0.882 0.492 0.861 0.938
SW
SR vs. 14 0.368 0.62 1.0 0.538 0.766 1.0
SF
SR vs. 5 0.167 0.269 0.367 0.286 0.424 0.537
GH
SR vs. 15 0.313 0.454 0.621 0.476 0.624 0.766
SW
SF vs. 10 0.244 0.223 0.418 0.392 0.364 0.59
GH
SF vs. 25 0.463 0.847 0.953 0.633 0.917 0.976
SW
GH vs. 11 0.216 0.47 0.47 0.355 0.45 0.64
SW


DM = Devil's Millhopper State Park, SF


San Felasco State Park, GH = Mike Roess


Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State
Park






59


Table 3-3. Summary of diversity values for prey items tabulated by site
Statistic DM SR SF GH SW
Individuals 94 178 467 73 361
Simpson index of diversity 6.81 9.4 9.69 10.13 10.25
Shannon index of diversity 2.21 2.65 2.78 2.87 2.91
Shannon evenness 0.735 0.594 0.755 0.735 0.772
DM = Devil's Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess
Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State
Park














CHAPTER 4
EFFECTS OF PRESCRIBED FIRE ON BIODIVERSITY AND SPECIES RICHNESS
OF CAVITY NESTING HYMENOPTERA IN SUWANNEE RIVER STATE PARK,
FLORIDA

Abstract

I examined the effect prescribed fire management had on the biodiversity and

species richness of populations of trap-nesting Hymenoptera and associated arthropods in

Florida. Four sandhill pine habitat sites (two burned sites and two unburned sites) at

Suwannee River State Park were examined over a two-year period. For trap-nesting

Hymenoptera, overall, species richness was different between treatment sites, and

diversity was significantly different (p < 0.05) between burned and unburned sites.

Overall diversity was not significantly different over time. Both unburned and burned

sites showed similarity in species composition, which was especially high when an

abundance-based estimate of similarity was used. When functional groups of trap-

nesting Hymenoptera were analyzed (predators, parasitoids and pollen collectors),

pollinators and parasitoids were not significantly different between burned and burned

sites. Predators were more abundant (p = 0.10) in unburned habitat. Of the six most

abundant species examined, Trypoxylon lactitarse (Hymenoptera: Sphecidae) was

significantly more abundant on unburned sites (p < 0.05), while Isodontia spp. (I. auripes

andI. mexicana, Hymenoptera: Sphecidae) were significantly more abundant on burned

sites (p < 0.05). Xylocopa virginica (Hymenoptera: Anthophoridae) had significantly

higher abundance on burned habitat than unburned habitat (p = 0.10). Chrysididae spp.,

Megachilidae spp. and Monobia quadridens (Hymenoptera: Vespidae) were not









significantly different, in terms of abundance, between burned and unburned sites.

Overall, prescribed fire employed by the park service to maintain natural sandhill pine

habitat has some impact on trap-nesting Hymenoptera and associated arthropods in terms

species richness and diversity. Although diversity and species richness changes were

determined, the use of trap-nesting Hymenoptera to detect community changes from

small-scale fires such as prescribed fire on their own may not be an appropriate choice to

detect community changes owing to the substantial flight ranges of these insects.

Introduction

Fire is an integral part of forest and grassland ecosystems throughout the United

States. Urban sprawl has increased the wildland-urban interface, causing increased

concerns about wildfire. Prescribed fire is a useful tool to reduce the intensity of

wildfires and has been shown to be effective in reducing hazardous fuels, disposing of

logging debris, preparing sites for seeding or planting, improving wildlife habitat,

managing competing vegetation, controlling insects and disease, improving forage for

grazing, enhancing appearance, improving access and in perpetuating fire-dependent

species (Biswell 1999, Cumming 1964, DellaSala and Frost 2001, Fuller 1991, Helms

1979, Long et. al. 2005, Mutch 1994, Wade et. al. 1988). Prescribed fire is a proven,

frequently used tool that works well in many aspects of wildland management. Many

natural areas, including state parks, utilize prescribed fire to restore and preserve native

habitat and plant communities in addition to reducing the risk of uncontrolled wildfire

(Siemann et al. 1997, Daubenmire 1968, Hurlbert 1965). Many studies have been

conducted to examine the effect of fire on animal and plant communities, with some

examining arthropods. Fire not only causes direct mortality in arthropods (Fay and

Samenson 1993, Bolton and Peck 1946, Miller 1978, Evans 1984), but also indirectly









affects arthropod communities via changes in plant community composition and habitat

alteration (Lawton 1983, Evans 1984). However, most invertebrate studies focus on

terrestrial arthropods monitored via sweeping or pitfall traps (Bess 2002, Brand 2002,

Niwa 2002, Clayton 2002, Fay 2003, and Koponen 2005) and overlook aerial insects. I

examined the effect of prescribed fire on the community of trap-nesting Hymenoptera

and associated arthropods. Trap-nesting hymenopterans are a diverse group of insects

that include various functional groups and interspecies interactions. This diverse group

includes predators, pollen specialists, and parasitoids. Fire may affect the various

subgroups differently depending on the alteration of resources. I investigated the effect

of fire on the biodiversity of these insects by using traps in Florida state parks that have

regularly and recently use prescribed fire. I set out to investigate the following questions

about these trap-nesting insects and associated arthropods: 1) Do overall diversity and

species richness differ between burned and unburned sites? 2) Do overall diversity and

species richness differ between the sampled years within burned sites? 3) In terms of

species sampled, how similar are the burned and unburned sites? 4) Are sampled

functional groups, in terms of abundance (predator, parasitoid and pollen specialists)

affected by fire? 5) Which of the most abundant species, if any, are negatively or

positively affected by fire in terms of abundance?

Methods and Materials

Tools and Trap Preparation.

The traps used in this study were fabricated from seasoned 37-mm x 86-mm x 2.4m

pine/spruce timbers obtained from a local home improvement store. The pine/spruce

timbers were cut into 100 10-cm-long blocks. Two cavities of one of five diameters (3.2,

4.8, 6.4, 7.9 or 12.7-mm) were drilled into each block. Cavities were drilled to a depth of









80 mm on each short side (the 37-mm side), offset approximately 10-mm from the center

point. Traps were assembled using one block of each diameter with the smallest cavity

on top and the largest on the bottom. Blocks were stacked so that no cavity was situated

directly above or below a cavity in the adjacent block. The five blocks were bound

together with strapping tape (3M St Paul, MN), and 16-gauge wire was used to further

bind the stack and suspend the trap from trees and shrubs at the field sites. Each bundle

of five blocks was considered to be a single trap

Field Sites.

I set four trap lines at two locations in Suwannee River State Park in Suwannee

County (30 23.149' N, 083 10.108' W). The habitats surveyed were burned and

unburned sand hill habitat. Descriptions of this habitat can be found in Franz and Hall

(1991). Sites that were burned within the current year were considered burned sites and

sites that had significant understory growth resulting from at least three years free of fire

were considered unburned. These unburned sites tended to have a thick understory and

were slated for prescribed bum if possible in the next couple of seasons. Four subsites,

two for each burn treatment, were established in the park. These subsites were

designated burned 1, burned 2, unburned 1, and unburned 2. Neither of the unburned

subsites was burned allowing for a two-year period of observation.

Field Placement

Transects were set up with ten traps placed approximately 10 m apart and were

hung approximately 1.5 m off the ground on trees or limbs with placement on dead

standing wood preferred. Transects were initially established (direction and distance

from center of plot) randomly. Four transects were established in Suwannee River state

park. Two transects were established in recently burned habitat (with smoldering still









ongoing) and two additional transects were established in habitat that had not been

burned in at least 3 years and had a significant understory of mixed broad leaf with some

palmetto. Transects were in the field from April 2003 until January 2005.

Field Collection and Laboratory Rearing

Traps remained in the field for two years and were checked monthly. Preliminary

field tests revealed that one-month intervals were sufficient to avoid trap saturation (no

available cavities). Traps were considered occupied when insects were observed actively

nesting, harboring or had sealed a cavity with mud or plant material. Occupied traps

were removed and replaced with a new trap. These occupied traps were brought into the

forest entomology lab at the University of Florida (Gainesville, FL) for processing.

Occupied blocks were removed for observation while unoccupied blocks were

reincorporated into replacement traps. Each occupied cavity was given a unique

reference number.

Location, date of collection, diameter of cavity, and various notes describing the

nature of the occupants and/or plug were recorded for each reference number. Occupied

cavities were then covered with a 2, 4, 6, or 8-dram glass shell vial. The shell vials were

attached to the wood section with masking tape (Duck, Henkel Consumer Adhesive

Inc., Akron OH) appropriate for wood application. These sections were then placed in a

rearing room and observed daily for emergence. The rearing room was maintained as

nearly as possible at outside mean temperatures for Gainesville, Florida.

When emergence occurred, the specimens were removed, preserved, and given the

same reference number as the cavity from which they had emerged. Dates of emergence,

identification of occupants, measurements, and notes were taken for each cavity at

emergence. Remaining nest fragments and debris were kept for further analysis when









possible. When an insect was harboring or actively tending a nest, it was captured,

identified, and given a reference number corresponding to the cavity. The contents of the

nest/cavity were then extracted and recorded. After the contents were extracted, the wood

block was reused in replacement traps. These processed blocks were re-drilled to

eliminate any alterations or markings (either physical or chemical) by the previous

occupant prior to reuse.

Specimen Identifications

All specimens were identified by the author with some specimens identified and or

verified by entomologists Jim Wiley1, Lionel Stange1, G. B. Edwards1, John B. Heppner1

and John M. Leavengood Jr. 1,2 (Florida State Collection of Arthropods1, Gainesville, FL

and University of Florida2, Gainesville, FL) Voucher specimens have been deposited at

the Florida State Collection of Arthropods in Gainesville, Florida.

Statistics and Calculations

I examined the difference between sites that were recently burned (treatment) and

those that had not been burned (non-treatment). Sites were examined for differences in

abundance of functional groups and abundance of the six most abundant species with

Analysis of Variance (ANOVA) using a Generalized Linear Mixed Model (GLMM) and

ANOVA using a Linear Mixed Model (LMM) was used to examine for differences in

species diversity index values. All models were computed by George Papageorgiou of

the IFAS statistical help lab at the University of Florida, Gainesville, Fl, using the

GLIMMIX procedure and SAS system for mixed models (SAS Institute, Inc., Cary NC) .

In addition, similarity of species presence in different sites and diversity statistics

were determined to further compare burned and unburned sites. Similarity was

calculated with Jaccard's similarity index (ISj) (Southwood 1978). This index is the









proportion of the combined set of species present at either site that are present in both

sites. This value ranges from 0 to 1, with 0 meaning no similarity (no species in common)

in both sites and 1 meaning all species are present at both sites. The value is calculated

using the following equation:

ISj= c /(a + b + c)

where c is the number of species common to both sites and a and b respectively are

the species exclusive to those sites.

Similarity was also calculated with Sorensen's similarity index (ISs) (Sorensen

1948). This index is the proportion of the combined set of species present at both sites

that are present in both sites. This value ranges from 0 to 1, with 0 meaning no similarity

(no species in common) in both sites and 1 meaning all species are present at both sites.

The value is calculated using the following equation:

ISs = 2c / (a + b)

where c is the number of species common to both sites and a and b are respectively the

total number of species at each site.

Similarity was also calculated for estimated population (in order to correct for

under sampling bias) using Chao-Jaccard abundance based estimate similarity index

(Chao et al 2005).

Diversity was calculated using Simpson's index of Diversity and Simpson's index

of dominance (Simpson 1949). Simpson's index of diversity value ranges from 1 to S,

where S is the total number of species. Simpson's index of dominance ranges from 0-1.

Simpson's index of dominance, K is given by:


X= (n/N)2
1=1









where n is the total number of organisms of the ith species and N is the total number of

organisms of all species.

Simpson's Index of Diversity is given by: 1/k

Diversity was also calculated using the Shannon index (Shannon and Weaver 1949) H',

given by:


H' = V (n / N) In (n / N); where n is the total number of organisms of a


particular species and N is the total number of organisms of all species.

Diversity is a combination of species richness (number of species) and evenness of

species abundance. Therefore, Shannon's index of evenness, J (Pielou 1966), is given

by:

J = H' / In s, where s is the total number of species

Species richness was estimated using rarefaction curves (Colwell et al. 2004). This

estimate of species richness is based on a sub-sample of pooled species actually

discovered.

In addition, two non-parametric species richness estimators, ACE (Abundance

based Coverage Estimator: Chao et al. 2000, Chazdon et al. 1998) and Chao 1(Chao

1984) were used. These estimators produce estimates of total species richness including

species not present in any sample. Most of the indices and all of the richness estimators

were computed using EstimateS 7.5 (Colwell, 2005)









Results

Field sites

During the two years of trapping, I collected 471 nests (34 pillaged by ants) in

burned areas and 700 nests (62 pillaged by ants) in unburned areas. These nests yielded

53 species from 25 families and 8 orders. These results are compiled in Table 4.1.

Subsites and sampling month

Rarefaction curves of observed species richness were produced for treatment sub-

site (burned 1 and burned; unbumedl and unburned2) in order to detect a possible site

effect. The resulting curves reached an asymptote and had over-lapping 95% confidence

intervals showing no significant difference, in terms of specie richness, between

treatment sub-sites. The identification of no significant sub-site bias allowed for sub-

sites to be pooled for further analysis. In addition, the statistical model used, mixed

linear model, factors out possible subsite, temporal and positional effects.

In terms of sampling months, all but two analyses did not have a significant

sampling month effect (Mixed linear model: P > 0.10). Two functional groups, predators

and parasitoids, had a significant sampling month effect (Mixed linear model: Ppredator

0.004, Pparasitoid = 0.02).

Effect of burning on species richness

Actual observed species richness was 38 species in burned habitat and 44 species in

unburned habitat. Rarefaction curves of observed species richness were produced for

burned and unburned sites. The rarefaction curve estimated 35 species in burned habitat

and 46 species in unburned habitat. The rarefaction curve, however, did not completely

attain an asymptote and should be viewed with skepticism, especially since the estimate

for burned habitat richness is lower than the observed richness. The Abundance based









Coverage Estimator (ACE) did not completely stabilize and its estimate of 46.84 species

in burned habitat and 62.04 species in unburned habitat should also be viewed with

skepticism. The Chao 1 estimate of species richness did stabilize and yielded estimates

of 47.5 species in burned habitat and 68.5 species in unburned habitat.

Effect of burning on diversity

Overall, the values for the Simpson index of diversity were 13.0 in burned habitat

and 3.38 in unburned habitat. Simpson's index of diversity values were significantly

different between burned and unburned habitats (LMM: F = 5.13, df = 13, P = 0.041)

with burned sites having a higher index of diversity. There was no significant month

effect (LMM: F = 1.21, df = 13, P = 0.3655)

Shannon's index of diversity showed burned sites were somewhat more diverse

than unburned sites (unburned = 2.17, burned = 2.87). Evenness in unburned sites was

less even than burned sites (Shannon evenness: unburned = 0.57, burned = 0.79)

The rank proportional abundance curve (Figure 4-1) also shows that the species

abundance in unburned plots was less even.

Similarity of burned and unburned

Similarity, measured by Sorensen's index yielded a value of 0.682. Jaccard's index

yielded a value of 0.5181. In addition, Chao-Jaccard estimate similarity index, which

provides similarity values based on estimated populations to correct for under-sampling,

gave a value of 0.928.

Functional groups

In terms of abundance, the predator group was significantly different (at p = 0.10)

between burned and unburned sites, with higher abundance in unburned sites (GLMM: F

= 3.76, P = 0.0745) and has a significant month effect (GLMM: F = 4.78, P < 0.001).









Both pollinators and parasitoid groups did not differ between treatments (GLMM:

pollinator: F = 0.95, df = 13, P = 0.3472; parasitoid F = 0.13, df = 13, P= 0.7277).

Pollinators did not have a significant month effect (GLMM: F = 0.45, df = 13, P = 0.91),

but parasitoids did have a significant month effect (GLMM: F = 3.13, df = 13, P = 0.02).

Most abundant species/ species groups

Of the six most abundant species, Trypoxylon lactitarse was significantly more

abundant in unburned habitat and Isodontia spp. were significantly more abundant in

burned habitat (Table 4-1). Xylocopa virginica was significantly more abundant in

burned habitat at p = 0.10 (GLMM: T. lactitarse F = 12.85, df = 14, P < 0.01; Isodontia

sp.: F = 11.18, df = 14, P < 0.01; X virginica: F = 3.84, df = 28, P = 0.07). Monobia

quadridens, Megachilidae species and Chrysididae species were not significantly

different between burned and unburned sites (GLMM: M. quadridens: F = 1.50, df = 14,

P > 0.10; Megachilidae sp: F = 0.40, df = 14, P > 0.10, Chrysididae: F = 0.89, df = 14, P

> 0.10). None of the top six most abundant species showed a month effect (GLMM: P >

0.10)

Discussion

The major justifications for using prescribed fire in state parks and natural areas are

prevention of uncontrollable wildfire and maintenance/ restoration of native/ natural

habitat. The Florida state park system uses informational displays and signs to emphasize

the importance of fire to maintain various native habitats such as sandhill pine and

rockland pine habitats. Many consider biodiversity to be an important indicator of

environmental health (Magurran 1988).

In terms of species richness (both observed and estimated), unburned sites had

higher values than burn sites. The only estimator that stabilized, Chao 1, estimated









higher species richness in unburned sites than burned sites. For this group, especially

when lower sampling effort cannot be avoided the Chaol estimator would seem to be the

best choice as it stabilized earlier and we can therefore have confidence in the estimate.

The Chao 1 estimator, however, assumes homogeneity and should not be used to

compare site with large compositional differences.

The similarity indices were high, especially the abundance based Chao-Jaccard

estimate index which had many sub-sites as completely similar. Of the similarity indices

used, Jaccard's index consistently gave the lowest, most conservative, estimate. Choice

of similarity index used should depend on several things. First sampling effort is a main

concern, especially in areas that have a high level of dominance and rare species are

frequently overlooked. In cases of small sampling effort or undersampling it would be

prudent to use the Chao-Jaccard estimate similarity index in order to correct for this bias.

Secondly, the level of identification is important and can skew similarity values. As in

this case, there were some groups that are notoriously difficult to identify, even by

authorities and this may influence the index value if they are pooled into a morpho-

species or species group. In such cases the more conservative index, Jaccard's index and

Chao-Jaccard estimate similarity index should be used so that similarity is not

overestimated.

Depending only on species richness values, however, can be misleading,

especially when investigating an event (such as fire in Florida) that the native fauna have

evolved with and to which have possibly adapted. Such events would unlikely cause

localized extinction (which would change species richness) but rather alter relative

abundances and dominance of species that have adapted to fire in varying degrees (which









is detected by diversity). Changes in species richness may have applications in situations

where the event is one that has not evolved with the fauna such as exotic species

introduction or anthropogenic disturbance. As seen here, the diversity of these insects

was significantly different between burn treatment sites. In this situation, the relative

abundance of Trypoxylon lactitarse, the overall most abundant species, was significantly

less abundant in burned sites. In unburned sites T lactitarse is much more dominant than

in burned sites (Figure 4-1). Trypoxylon lactitarse is so dominant in unburned areas that

it lowers the evenness values and therefore overall diversity, even though unburned sites

had higher values of species richness. Although the burned sites had lower species

richness than unburned sites, diversity was higher because of greater evenness values.

Low or lowered evenness values can be seen as a sign of disturbance. This suggests that

the unburned condition is actually the disturbed state for this habitat. This makes

intuitive sense since the fauna have evolved with the yearly and regular intervals of fires

that are suppressed by the park service. In essence, the disturbance is the removal of fire

from the ecosystem by man.

In addition, any difference in sampling years may be the result of natural habitat

succession after a fire event. Most habitats have succession periods that span years and

continually change over decades (Siemann et al. 1997, Swengel 2001). Therefore,

change in diversity may be a result of succession and not a difference in sample years

(e.g. especially cold winter, drought, etc.). This idea is further supported since there was

no temporal effect detected for the majority of the analyzed groups over the two-year

sampling period and longer intervals are needed to detect the faunal response.









Focusing on functional groups, it was surprising to not see a difference in

abundance for the pollinator group. Bee communities tend to respond positively, after

the initial catastrophic mortality, post-burn in response to increased floral resources (Potts

et al. 2003). Pollen group species should see an increase, minimally a difference in

sampled years of the burned sites, but did not differ in abundance. Most likely, the scale

of these small prescribed fires and the mobility of these insects eliminated any impact fire

may have had.

The difference in predator and parasitoid diversity follow previous observations of

reductions in abundance and resulting diversity in burned areas. These groups depend on

abundance of prey items and these prey items tend to have varying response to fire. Bock

and Bock (1991) showed that although all grasshoppers were affected by burning, certain

groups suffered higher mortality and populations took longer to regenerate. Some prey

groups may be better adapted to fire and these populations rebound quicker than other

prey groups (Dunwiddiel991). Parasitoids and predators are dependant on prey

population and their varying ability to respond to fire and this unequal return to previous

abundance pattern will inherently affect the abundance patterns of predators. This should

be especially true for spider-hunting wasps and parasitoids of predators.

Hymenoptera, especially bees and wasps, are generally strong fliers with flight

ranges that can span kilometers and prescribed fire tend to be restricted to variously sized

sections that are commonly 10 hectares or less. Newly burned areas are easily accessible

by these insects from the surrounding unburned habitat. These prescribed fires do an

excellent job of removing dead standing and felled wood, but rarely eliminate all such

material (pers. obs.), especially in controlled burns that tend to be less intense than









uncontrolled wildfire. Uncontrolled, intense wildfire may remove most dead timber but

tend to kill younger trees and, in the cases of the fire crowning, may kill mature trees

resulting in possible increased nesting sites. In the end, these insects use dead wood as

nesting sites and are not completely deprived of this resource within burned sites.

Furthermore, the preferred nesting sites and required materials for some species, such as

resin and grasses, increase in abundance and availability in response to burning.

The nature of prescribed fire does not allow for controlled experimentation, and

burn site establishment is dependent on weather and park management. In order to

minimize a possible site bias, the sites chosen were initially (pre-bum) identical in terms

of flora and habitat. In addition, repeated measures were used for statistical analysis to

further minimize the possibility of a site bias confounding the results. Even though I am

confident that the measures taken to reduce influence from a site bias were adequate, the

possibility of sit effects cannot be completely disregarded.

Conclusion

Overall species richness and diversity did differ between burned and unburned

sites, and sites were not different between sampling dates, indicating that burning affects

trap-nesting hymenopterans and associated arthropods from burning treatment.

None of the three functional groups (pollinators, predators, and parasitoids) were

affected by the bum treatment. Of the six most abundant species captured, only two were

significantly different (p < 0.05) between burned and unbumed habitat, with one more

abundant in burned habitat and the other in unburned habitat.

Even though diversity and species richness changes were determined, the use of

trap-nesting Hymenoptera on their own may not be an appropriate choice to detect

community changes from small-scale fires such as prescribed fire. These insects are









volant with substantial flight range and this flight range may allow these insects to

respond to a source-sink of nesting materials (resin, grasses, and cavities), yet still forage

beyond the scope of the relatively small scale of the prescribed fire and perhaps eliminate

any effect the treatment may have had on these groups. In the case of wildfire, where the

scale is usually exponentially larger, the species richness and diversity of these insects

may be more indicative of the community as a whole.

Monitoring diversity and abundance of trap-nesting hymenopterans in unburned

sandhill pine habitat, however, may be an appropriate application to monitor the

community. When the community becomes less even with few species, such as T

lactitarse dominating the proportional abundance, this may be an indicator of disturbance

and that a burn is needed to maintain the desired sandhill pine habitat.

Acknowledgements

All research and collection were completed with permission of the Florida

Department of Environmental Protection Division of Parks and Recreation under permit

numbers 11250310 and 08170410.







76



Rank proportional Abundance


1 10 1
Species Rank

-- Bued site o Unburned site


Figure 4-1. Rank proportional abundance of species in burned and unburned sandhill
pine habitats










Table 4-1. Species trapped in burned and unburned sandhill pine habitat
Number of nests in habitat
Order Family Genus Species Functional Burned Unburned
group
Araneida Salticidae Platycryptus undatus Predator 3 2
Segestriidae Ariadna bicolor Predator 2 4
Clubionidae Elaver except Predator 2 5
Blatteria Blattaria sp B 2 4
Coleoptera Carabidae Cymindus platycollis Parasitoid 0 2
Tenebrionidae 0 1
Cleridae Cymatodera Parasitoid 0 1
Cleridae Lecontella brunnea Parasitoid 1 0
Cleridae ii. ,,i, Parasitoid 0 1
Elateridae 0 1
Diptera Bombyliidae Anthrax analis Parasitoid 9 22
Bombyliidae Anthrax aterrimus Parasitoid 6 25
Bombyliidae Lepidophora lepidocera Parasitoid 4 2
Bombyliidae Toxophora amphitea Parasitoid 1 0
Conopidae Parasitoid 0 1
Orthoptera Gryllidae Orocharis luteolira 14 7
Hymenoptera Anthophoridae Xylocopa virginica Pollen 7 1
Formicidae Crematogaster Blk Predator 6 26
Formicidae Crematogaster Red Predator 11 2
Chrysididae Parasitoid 40 33
Ichneumonidae Parasitoid 1 1
Leucospidae Parasitoid 1 0
Megachilidae Dolicostelis louisa Parasitoid 1 0
Megachilidae Coelioxys sayi Parasitoid 2 1
Megachilidae Megachile campanulae Pollen 1 0
Megachilidae Megachile georgica Pollen 7 5
Megachilidae Megachile mendica Pollen 9 1
Megachilidae Megachile xylocopoides Pollen 0 3
Megachilidae Osmia sandhouseae Pollen 2 0
Mutillidae Sphaeropthalma pensylvanica Pollen 1 0
Pompilidae Ampulex canaliculata Predator 1 0
Pompilidae Dipogon graenicheri Predator 1 3
Sphecidae Isodontia auripes Predator 60 9
Sphecidae Isodontia mexicana Predator 28 6
Sphecidae Liris beata Predator 0 1
Sphecidae Podium rufipes Predator 5 15
Sphecidae Trypoxylon clavatum Predator 0 12
Sphecidae Trypoxylon collinum Predator 21 17
Sphecidae Trypoxylon clavatumjohanis Predator 4 15
Sphecidae Trypoxylon carinatum Predator 0 1
Sphecidae Trypoxylon lactitarse Predator 56 339
Sphecidae Trypoxylon Red Predator 1 3
Sphecidae Trypoxylon Sm Predator 0 3
Vespidae Vespula maculifrons Predator 2 1
Vespidae Euodynerus megaera Predator 13 7
Vespidae Monobia quadridens Predator 47 33
Vespidae Stenodynerus Sp A Predator 53 7
Vespidae Sp C Predator 1 0
Lepidoptera Pyralidae Uresiphita reversalis 0 2







78





Table 4-1 Continued. Species trapped in burned and unburned sandhill pine habitat
Number of nests in habitat
Order Family Genus Species Functional Burned Unburned
group _
Noctuidae Cerma cerintha 0 1
Noctuidae sp B 0 4
Scorpionida Buthidae Centruroides hentzi Predator 12 0
Chilopoda Scolopendridae Hemiscolopendra punctiventris Predator 0 5














CHAPTER 5
BIOLOGY, PREY AND NESTS OF THE POTTER-WASP Monobia quadridens L.
(HYMENOPTERA: VESPIDAE)

Abstract

I observed the potter wasp, Monobia quadridens L, nesting in predrilled wooden

trap-nests at five state parks in north central Florida. Wasps nested mostly in 12.7-mm

diameter cavities (97 of 129 nests) and occasionally nested in 7.9-mm diameter cavities

(26 of 129 nests). Females rarely nested in 6.4-mm (5 of 129) and 4.8-mm diameter

cavities (1 of 129). All cavities were 80-mm deep. Females used mud to make

provisioned cells, partitions, intercalary cells, vestibular cells and a closure plug, yet did

not line the inside of any cells with material. They constructed nests with an average of

1.69 provisioned cells (range = 1-3, SD = 0.47), a vestibular cell, and from 0-3

intercalary cells. All nests were solely provisioned with paralyzed caterpillars ofMacalla

sp. (thrysisalis orphaeobasalis) (Lepidoptera: Pyralidae). Cells with female brood had a

mean length of 24.01 mm (range = 20-30, SD = 3.59 N= 16) while cells that resulted in

males had a mean length of 18.22 mm (range = 14-25, SD =2.68, N = 23). Intercalary

cells were highly variable with a mean of 13.45 mm, (range = 5-35, SD = 5.69, N = 28)

as were vestibular cells with a mean of 12.78 mm (range = 1-35, SD= 9.055, N = 23).

Resulting sex ratio of emerging adults was 1.2 males per female. In conclusion, nest

architecture of Monobia quadridens is variable, females tend to nest in cavities with

diameters greater than 7.9 mm, and females preyed on a single species of the Pyralid

caterpillar, Macalla sp. (thrysisalis phaeobasalis). This apparent prey specialization is









unique when compared to M quadridens in other parts of its range and even to historical

Florida data. Such a difference and possible shift of behavior within Florida warrants

further study.

Introduction

Monobia quadridens is a common wasp in the eastern United States where it is

reported from Massachusetts, Rhode Island, Connecticut, New York, New Jersey,

Pennsylvania, Delaware, Maryland, West Virginia, District of Columbia, Virginia, North

Carolina, South Carolina, Alabama, Georgia, Florida, Mississippi, Louisiana, Arkansas,

Kentucky, Tennessee, Texas, Oklahoma, Kansas, New Mexico, Missouri, Indiana,

Illinois and Ohio (Bequaert 1940). Specimens captured by Krombein (1967) at the

Archbold Biological Station at Lake Placid, Florida (Highlands County) and specimens at

the Florida State Collection of Arthropods in Gainesville, Florida indicate that it ranges

throughout peninsular Florida, including the Florida Keys and Everglades National Park.

This wasp normally nests in abandoned nests and burrows of other insects, such as

carpenter bees that nest in clay banks and wood. In Florida, Monobia quadridens is the

largest wasp that nested in traps.

When female M. quadridens find an acceptable cavity, they line the back end of the

cavity with mud and suspend a single egg from the top of the cavity. Females hunt and

paralyze caterpillars to provision the nest. When prey populations are sufficient, they

tend to provision the same species of caterpillars (Krombein 1967). Females are also

known to have a leisurely nesting rate taking up to a week to complete a nest with three

provisioned cells and a single intercalary cell (Krombein 1967). Adults emerge from the

nest two days to two weeks after completion of the nest, but actual development time is

about 10-14 days.









Practically all recent research on Monobia quadridens has been on the biochemistry

and physiology of organisms and proteins extracted from its hemolymph and little has

been done to examine the ecology of the wasp. The objectives for this study are to

determine the preferred nesting cavity diameter, determine what prey the females are

provisioning, describe nest architecture, and determine emerging sex ratio of this wasp,

Monobia quadridens, in north central Florida.

Methods and Materials

Tools and Trap Preparation

The traps used in this study were fabricated from seasoned 37 mm x 86 mm x

2.4m pine/spruce timbers obtained from a local home improvement store. The

pine/spruce timbers were cut into 100 10-cm-long blocks. Two cavities of one of five

diameters (3.2, 4.8, 6.4, 7.9, or 12.7-mm) were drilled into each block. Cavities were

drilled to a depth of 80 mm on each short side (the 37-mm side), offset approximately 10-

mm from the center point. Traps were assembled using one block of each diameter with

the smallest cavity on top and the largest on the bottom. Blocks were stacked so that no

cavity was situated directly above or below a cavity in the adjacent block. The five

blocks were bound together with strapping tape (3M St Paul, MN), and 16-gauge wire

was used to further bind the stack and suspend the trap from trees and shrubs at the field

sites. Each bundle of five blocks was considered to be a single trap.

Field Sites

I set traps at five locations: 1) Suwannee River State Park in Suwannee County (30

23.149' N, 083 10.108' W), 2) Mike Roess Gold Head Branch State Park in Clay County

(29 50.845'N, 081 57.688' W), 3) Devil's Millhopper Geological State Park in Alachua

County (29 42.314'N, 082 23.6924' W), 4) San Felasco Hammock Preserve State Park









(29 42.860' N, 082 27.656' W) in Alachua County and 5) Silver River State Park in

Marion County (29 12.317'N, 082 01.128' W). The habitats surveyed at Suwannee

River State Park were burned and unburned sand hill habitat, while the habitat at Mike

Roess Gold Head Branch State Park was burned sand hill pineland and ravine. Sites at

San Felasco Hammock Preserve State Park consisted of upland and mesic hardwood

hammock. Surveyed areas of Devil's Millhopper Geological State Park consisted of pine

flatwood habitat and sites at Silver River State Park consisted of river habitat and upland

mesic forest. Descriptions of these habitats can be found in Franz and Hall (1991).

Field Placement

Transects were set up with ten traps placed approximately 10 m apart and hung

approximately 1.5 m off the ground on trees or limbs with placement on dead standing

wood preferred. Transects were initially established (direction and distance from center

of plot) randomly. Four transects were established in Suwannee River State Park while

three transects were established Mike Roess Gold Head Branch State Park. Three

transects were established in San Felasco State Park but size constraints only allowed a

single transect in Devil's Millhopper State Park. Finally, two transects were set up in

Silver River State Park. Transects were in the field from April 2003 until January 2005.

Field Collection and Laboratory Rearing

Traps remained in the field two years and were checked monthly. Preliminary field

tests revealed that one-month intervals were sufficient to avoid trap saturation (no

available cavities). Traps were considered occupied when insects were observed actively

nesting, harboring or had sealed a cavity with mud or plant material. Occupied traps

were removed and replaced with a new trap. These occupied traps were brought into the

forest entomology lab at the University of Florida in Gainesville, Fl for processing.









Occupied blocks were removed for observation while unoccupied blocks were

reincorporated into replacement traps. Each occupied cavity was given a unique

reference number.

Location, date of collection, diameter of cavity, and various notes describing the

nature of the occupants and/or plug were recorded for each reference number. Occupied

cavities were then covered with a 2, 4, 6, or 8-dram glass shell vial. The shell vials were

attached to the wood section with Duck (Henkel Consumer Adhesive Inc., Akron OH)

masking tape appropriate for wood application. These sections were then placed in a

rearing room and observed daily for emergence. The rearing room was maintained as

nearly as possible at outside mean temperatures for Gainesville, FL.

When emergence occurred, the specimens were removed, preserved and given the

same reference number as the cavity from which they had emerged. Dates of emergence,

identification of occupants, measurements and notes were taken for each cavity at

emergence. When a female M. quadridens was harboring or actively tending a nest, it

was captured, identified, and given a reference number corresponding to the cavity. The

contents of the nest/cavity were then extracted and recorded. After the contents were

extracted, the wood block was reused in replacement traps. These processed blocks were

re-drilled to the next size diameter cavity to eliminate any alterations or markings (either

physical or chemical) by the previous occupant prior to reuse. All adult Monobia

quadridens that successfully emerged from cavities were curated and sexed.

Identifications

Monobia quadridens is readily distinguished from other vespid wasps. Specimen

diagnostics are given in Appendix B and figures of both sexes of Monobia quadridens are

given in Appendix A.









All cavity nesters and their prey were identified by the author with some specimens

identified or verified by entomologists Jim Wiley1, Lionel Stange1, John B Heppner1,

John M. Leavengood Jr.1,2 (Florida State Collection of Arthropods, Gainesville, FL and

University of Florida2, Gainesville, FL). Voucher specimens have been deposited at the

Florida State Collection of Arthropods.

Statistical Analysis

Descriptive statistics (means, ranges, standard deviation) were calculated using

Microsoft Excel statistical package. Chi-squared goodness of fit was used to examine

nest diameter preference. The assumption was that wasps would nest equally in all

acceptable diameters.

Results

Nest Architecture

Monobia quadridens nested in 129 cavities over the two-year period of

observation, mostly nesting in 12.7-mm cavities (97 of 129 nests) and occasionally

nesting in 7.9-mm cavities (26 of 129 nests). Females rarely nested in 6.4 mm-cavities (5

of 129) and 4.8-mm cavities (1 of 129). Females did not nest equally in all diameters

(chi-squared contingency table, X 2= 184.51, df = 3, P< 0.001) when the data were

pooled. Females did not nest equally in all diameters at individual sites (P< 0.001:

X 2Devil's Millhopper= 23.25, 2Goldhead = 134.56, 2San Felasco = 65.33, 2Suwannee River = 32.62,

and P < .05: and 2Silver River = 8.2) and during particular years (P< 0.001, x 2003 =101.0,

X 22004= 144.69). See table 5.1 for a summary of nesting results.

Whenever females were observed plugging a cavity, the nests (n = 32) were later

dissected to examine nest architecture. Females used mud to make the cell partitions and









closing plug yet did not line the sides of cells with material. Nests averaged 1.69

provisioned cells (SD = 0.47, range 1-3, N= 32), with vestibular cells (an empty cell

between the cavity opening and provisioned cells) and intercalary spaces (empty,

unprovisioned spaces between provisioned cells and behind the vestibular cell). Cells

with female brood had a mean length of 24.01 mm (SD= 3.59, range = 20-30, N= 16)

while cells containing males had a mean length of 18.22 mm (SD = 2.68, range = 14-25,

N = 23). Intercalary spaces were highly variable with a mean of 13.45 mm (SD = 5.69,

range = 5-35, N = 28), as were vestibular cells with a mean of 12.78 mm (SD = 9.055,

range = 1-35, N = 23). On average, nests had two provisioned cells usually separated by

at least one intercalary space. Use of the intercalary spaces varied; some nests had an

intercalary space in front of each provisioned cell while other nests had a single

provisioned cell with multiple intercalary spaces leading to the vestibular cell. Nests had

a mean of 1.28 (SD = 0.631,range = 0-3, N= 32) intercalary spaces per cavity.

Sex Ratio

I collected 129 nests ofMonobia quadridens from traps. These nests yielded a sex

ratio of 1.2 males per 1 female (N= 245).

Prey

Females suspended a single egg from the top of each cell with a filament and would

then search for prey. All prey items were the caterpillar, Macalla sp. (thrysisalis/

phaeobasalis) (Lepidoptera: Pyralidae). These two species cannot be distinguished

without rearing out to adult stage, but this is impossible since caterpillars have been

paralyzed by the wasp. Females would position these paralyzed caterpillars

longitudinally with the head toward the rear of the cavity. Developing larvae would









leave caterpillar head capsules, allowing for prey confirmation after emergence. On

average, there were 3.0, (SD = 0.67, range 2-5, N= 98) caterpillars per cell.

Discussion

Nest Architecture

Krombein (1967) reported that in some populations, including a central Florida

population, females used agglutinated sand to construct partitions and plugs. None of the

populations I observed used agglutinated sand although all trap lines were within 2

kilometers of a mud resource such as roads or bodies of water. Nests had a mean of 1.69

cells but in actuality nests usually had 2 provisioned cells in 80-mm long cavities, with

occasional nests with a single provisioned cell lowering the mean. Bequaert (1940)

reported that the typical pattern for solitary eumenid wasps, including M. quadridens, is a

nest containing up to 12 cells in preexisting cavities. These cavities are usually burrows

made by some other insect in clay banks or wood and do not usually have the confines of

the traps used that had cavities only 80-mm long. Krombein (1967) reported that M

quadridens primarily nested in 12.7-mm diameter cavities with rare (2 of 78) occasions

of nesting in 6.4-mm diameter cavities. He stated that most females are too large to enter

a 6.4-mm cavity which restricts nesting activity to diameters larger than 6.4-mm. The

Florida populations I observed were similar in this respect. Monobia quadridens mostly

nested in 12.7-mm cavities with some in 7.9-mm cavities and rarely (5 of 129) in 6.4-mm

cavities. I did trap an uncompleted nest in a 4.8-mm cavity, however the nest was

abandoned after a single egg was laid and two caterpillars were provisioned. The single

cell was not sealed and did not develop, yet since an egg was deposited and prey

provisioned, the female most likely may have died and not necessarily absconded




Full Text

PAGE 1

BIOLOGY, ECOLOGY, BEHAVIOR, PA RASITOIDS AND RESPONSE TO PRESCRIBED FIRE OF CAVITY NESTING HYMENOPTERA IN NORTH CENTRAL FLORIDA By DAVID SERRANO 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 2006

PAGE 2

Copyright 2006 by David Serrano

PAGE 3

iii ACKNOWLEDGMENTS I thank my graduate advisor, John L. Foltz, for being understanding and patient with all of lifes complicati ons that I have had to endur e throughout my program. His insight, suggestions, advice and patience made hi m the best advisor I could have chosen. I also thank my graduate committee (Drs. Lionel Stange, Robert McSorley, and Emilio M. Bruna) for their guidance, insight and maki ng this work substant ially better. I also thank Jim Wiley, one of the most helpful a nd humble men I know, for his helpful efforts and mastery of Hymenoptera. I thank my pa rents for teaching me the values of hard work and perseverance that were integral in obtaining my PhD. I also thank my wife, Esther S. Serrano (DPM 2005), for her help by pushing me acro ss the finish line.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT.....................................................................................................................xiii CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW....................................................1 Introduction...................................................................................................................1 Species Richness And Diversity: Estimating Their Values..........................................2 Overview of Trap-nesting Hymenoptera......................................................................5 Effect Of Fire on Trap-Nesting Hymenoptera..............................................................6 Hymenopterans Sampled and Summary.......................................................................7 2 NEST ARCHITECTURE, PREY, AND SEXUAL DIMORPHISM IN THE GRASS-CARRYING WASPS Isodontia ( MURRAYELLA ) mexicana (SAUSSURE) AND Isodontia auripes (FERNALD) (HYMENOPTERA: SPHECIDAE: SPHECINAE).....................................................................................12 Abstract.......................................................................................................................12 Introduction.................................................................................................................12 Methods and Materials...............................................................................................14 Tools and Trap Preparation.................................................................................14 Field Sites............................................................................................................14 Field Placement...................................................................................................15 Field Collection and Laboratory Rearing............................................................15 Specimen Diagnostics and Identification............................................................16 Statistical Analysis..............................................................................................17 Results........................................................................................................................ .17 Habitat.................................................................................................................17 Nest Architecture.................................................................................................17 Sex Ratio and Sexual Dimorphism......................................................................19 Prey......................................................................................................................20 Survival................................................................................................................21

PAGE 5

v Discussion...................................................................................................................21 Habitat.................................................................................................................23 Nest Architecture.................................................................................................23 Sex Ratio and Sexual Dimorphism......................................................................25 Prey......................................................................................................................26 Conclusion...........................................................................................................28 Acknowledgements.............................................................................................28 3 SPIDER PREY IN NESTS OF THE MUD DAUBER WASP Trypoxylon lactitarse (HYMENOPTERA: SPHECIDAE)...........................................................37 Abstract.......................................................................................................................37 Introduction.................................................................................................................37 Methods and Materials...............................................................................................39 Tools and Trap Preparation.................................................................................39 Field Sites............................................................................................................39 Field Placement...................................................................................................40 Field Collection and Laboratory Rearing............................................................40 Specimen Identifications.....................................................................................41 Statistical Analysis......................................................................................................41 Results........................................................................................................................ .43 Discussion...................................................................................................................45 Acknowledgements.............................................................................................49 4 EFFECTS OF PRESCRIBED FIRE ON BIODIVERSITY AND SPECIES RICHNESS OF CAVITY NESTIN G HYMENOPTERA IN SUWANNEE RIVER STATE PARK, FLORIDA............................................................................60 Abstract.......................................................................................................................60 Introduction.................................................................................................................61 Methods and Materials...............................................................................................62 Tools and Trap Preparation.................................................................................62 Field Sites...................................................................................................................63 Field Placement...................................................................................................63 Field Collection and Laboratory Rearing............................................................64 Specimen Identifications.....................................................................................65 Statistics and Calculations...................................................................................65 Results........................................................................................................................ .68 Field sites.............................................................................................................68 Subsites and sampling month.......................................................................68 Effect of burning on species richness...........................................................68 Effect of burning on diversity......................................................................69 Similarity of burned and unburned...............................................................69 Functional groups.........................................................................................69 Most abundant species/ species groups........................................................70

PAGE 6

vi Discussion...................................................................................................................70 Conclusion...........................................................................................................74 Acknowledgements.............................................................................................75 5 BIOLOGY, PREY AND NEST S OF THE POTTER-WASP Monobia quadridens L. (HYMENOPTERA: VESPIDAE)..........................................................................79 Abstract.......................................................................................................................79 Introduction.................................................................................................................80 Methods and Materials...............................................................................................81 Tools and Trap Preparation.................................................................................81 Field Sites............................................................................................................81 Field Placement...................................................................................................82 Field Collection and Laboratory Rearing............................................................82 Identifications......................................................................................................83 Statistical Analysis..............................................................................................84 Results........................................................................................................................ .84 Nest Architecture.................................................................................................84 Sex Ratio......................................................................................................85 Prey...............................................................................................................85 Discussion...................................................................................................................86 Nest Architecture.................................................................................................86 Sex Ratio.............................................................................................................87 Prey......................................................................................................................87 Conclusion...........................................................................................................90 Acknowledgements.............................................................................................90 6 BIODIVERSITY OF TRAP-NESTING HYMENOPTERA OF FIVE NORTH FLORIDA STATE PARKS........................................................................................92 Abstract.......................................................................................................................92 Introduction.................................................................................................................92 Methods and Materials...............................................................................................93 Tools and Trap Preparation.................................................................................93 Field Sites............................................................................................................93 Field Placement...................................................................................................94 Field Collection and Laboratory Rearing............................................................94 Specimen Identifications.....................................................................................95 Statistical Analysis......................................................................................................96 Results........................................................................................................................ .98 Discussion...................................................................................................................99 Acknowledgements...........................................................................................104 APPENDIX A ADDITIONAL FIGURES AND SP ECIMEN PHOTO GUIDE..............................111

PAGE 7

vii B SELECTED SPECIMEN DIAGNOST ICS AND IDENTIFICATION...................143 LIST OF REFERENCES.................................................................................................145 BIOGRAPHICAL SKETCH...........................................................................................156

PAGE 8

viii LIST OF TABLES Table page 2-1 Comparison of Isodontia auripes and I. mexicana ..................................................32 2-2 Frequency of cavities nested in by I. auripes and I. mexicana ................................33 2-3 Summary of emerged I. auripes and I. mexicana ....................................................33 2-4 Prey records for I. mexicana ....................................................................................34 2-5 Prey records for Isodontia auripes ...........................................................................35 3-1 Spiders found as prey in nests of Trypoxylon lactitarse in north central Florida....56 3-2 Similarity indexes and co mparisons for spider prey................................................58 3-3 Summary of diversity values for prey items tabulated by site.................................59 4-1 Species trapped in burned a nd unburned sandhill pine habitat................................77 5-1 Number nest diameters occupied by Monobia quadridens ......................................91 6-1 Summary of trap-nesting arthr opods captured in five state parks..........................108 6-2 Similarity indexes and compar isons for trap-nesting Hymenoptera......................110

PAGE 9

ix LIST OF FIGURES Figure page 2-1 Cross section of Isodontia mexicana nest in a 12.7mm cavity...........................29 2-2 Isodontia auripes larvae on provisioned Scudderia furcata ...............................29 2-3 Isodontia cocoon.................................................................................................30 2-4 Frequency of cavities nested in by I. auripes and I. mexicana ...........................30 2-5 Summary of emerged Isodontia mexicana and Isodontia auripes from captured nests......................................................................................................31 3-1 Ten most abundant spider pr ey species for all sites pooled................................50 3-2 Five most abundant spider prey species at Suwannee River State Park.............50 3-3 Five most abundant spider prey species at San Felasco State Park....................51 3-4 Five most abundant spider prey species at Silver River State Park....................51 3-5 Five most abundant spider prey spec ies at Gold Head Branch State Park.........52 3-6 Five most abundant spider prey sp ecies at Devils Millhopper State Park.........52 3-7 Abundance and percentage of spider prey families captured at all sites............ 53 3-8 Pooled rank proportional abundance of spider species collected from five Florida state parks...............................................................................................53 3-9 Site rank proportional abundance of sp ider species collected at each state park.....................................................................................................................54 3-10 Species richness estimation for spider prey tabulated by site.............................54 3-11 Species richness estimator performan ce for spider prey tabulated by site.........55 4-1 Rank proportional abundance of spec ies in burned and unburned sandhill pine habitats........................................................................................................76 6-1 Actual observed species ri chness, tabulated by site..........................................105

PAGE 10

x 6-2 Shannon index of diversity and Shan non evenness values for trap nesting Hymenoptera and associated arthro pods at five state parks.............................105 6-3 Simpson index of diversity for tr ap nesting Hymenoptera and associated arthropods at five state parks............................................................................106 6-4 Species richness estimators tabulated by site....................................................106 6-5 Species richness estimator performance per site..............................................107 A-1 Traps................................................................................................................111 A-2 Rearing room...................................................................................................111 A-3 Male Monobia quadridens ................................................................................112 A-4 Female Monobia quadridens ............................................................................113 A-5 Antenna of Monobia quadridens ......................................................................113 A-6 Male Isodontia auripes......................................................................................114 A-7 Female Isodontia auripes..................................................................................114 A-9 Female Isodontia mexicana...............................................................................115 A 10 Anthrax analis ...................................................................................................116 A-11 Anthrax aterrimus .............................................................................................116 A-12 Lepidophora lepidocera ....................................................................................117 A-13 Toxophora amphitea .........................................................................................117 A-14 A wasp in the family Chrysididae.....................................................................118 A-15 A series of Chrysidid wasps demons trating variation in size and color...........118 A-16 Lecontella brunnea ...........................................................................................119 A-17 Macrosigon cruentum .......................................................................................119 A-18 Nemognatha punctulata ....................................................................................120 A-19 Ancistorcerus ....................................................................................................120 A-20 Euodynerus megaera ........................................................................................121 A-21 Pacnodynerus erynnis .......................................................................................121

PAGE 11

xi A-22 Stenodynerus sp A.............................................................................................122 A-23 Stenodynerus sp b.............................................................................................122 A-24 Camponotus Red...............................................................................................123 A-29 A wasp of the family Leucospididae................................................................126 A-30 Dolicostelis louisa .............................................................................................127 A-31 Coelioxys sayi ...................................................................................................127 A-32 Coelioxys dolichos ............................................................................................128 A-33 Coelioxys texana ...............................................................................................128 A-34 Megachile campanulae .....................................................................................129 A-35 Megachile mendica ...........................................................................................129 A-36 Megachile c. wilmingtoni ..................................................................................130 A-37 Megachile georgica ..........................................................................................130 A-38 Megachile xylocopoides female........................................................................131 A-39 Megachile xylocopoides male...........................................................................131 A-40 Osmia sandhouseae ..........................................................................................132 A-41 Sphaeropthalma pensylvanica floridensis ........................................................132 A-42 Orocharis luteolira ...........................................................................................133 A-43 Ampulex canaliculata ........................................................................................134 A-45 Podium rufipes ..................................................................................................135 A-46 Trypoxylon clavatum clavatum .........................................................................136 A-47 Face of Trypoxylon c. clavatum. Note golden vessiture...................................136 A-48 Trypoxylon carinatum .......................................................................................137 A-50 Trypoxylon collinum collinum.........................................................................138 A-51 Trypoxylon johnsoni ..........................................................................................138 A-52 Trypoxylon lactitarse ........................................................................................139

PAGE 12

xii A-53 Vespula maculifrons ..........................................................................................139 A-54 Xylocopa virginica male..................................................................................140 A-55 Xylocopa virginica female...............................................................................140 A-59 Centruiodes hentzi ............................................................................................142

PAGE 13

xiii 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 BIOLOGY, ECOLOGY, BEHAVIOR, PA RASITOIDS AND RESPONSE TO PRESCRIBED FIRE OF CAVITY NESTING HYMENOPTERA IN NORTH CENTRAL FLORIDA By David Serrano August 2006 Chair: John L. Foltz Major Department: Entomology and Nematology This study examined the biology, ecology, behavior, parasitoids and response to fire of an understudied group of insects, the cavity nesting Hymenoptera. Five state parks in north central Florida were surveyed for two years with trap nests yielding over 3,000 captured nests. Trap-nesting Hymenoptera repr esent important guilds, such as predators and pollinators, within these surveyed habitats and are an integral part of maintaining desired biodiversity of both flora and fauna Over the two year period, biology, ecology, and prey of a potter wasp, Monobia quadridens a mud-dauber wasp, Trypoxylon lactitarse, and two grass carrying wasps, Isodontia auripes and Isodontia mexicana were examined in depth. In addition, more than 100 species of trap-n esting Hymenoptera and associated arthropods were examined yiel ding data on distribution, host ranges, biology and ecology. Also, a detailed inventory of identified trap-nesting hymenoptera and associated arthropods is provided to expand park faunal records.

PAGE 14

xiv In addition to examining biology and ecology of this group, this study examines the effect prescribed fire has on these insects. Prescribed fire is a co mmonly used practice in managed parks and natural areas to restore a nd maintain native and protected habitat and these insects, as pollinators and predators of plant feeders, may play an important role in the succession of desirable, nati ve habitat after the fire event. Prescribed fire was used by the park managers in such a manner that allo wed for comparison of equally sized areas of identical habitat. Overall, the community of trap-nesting Hymenoptera was affected by the scale of prescribed fire used by the pa rk service in terms of overall diversity and abundance of key species. The diversity and richness of cavity nesting Hymenoptera may be used as an indicator of when to use prescr ibed fire to maintain native ecosystems and foster a healthy biodiversity.

PAGE 15

1 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW Introduction During the late 1960s and 1970s, the United St ates began to confront the costs of unrestricted patterns of settlement and land use on the environmen t (Porter and Marsh 2005). Wetlands, water and air quality, and thre atened species were all protected by new laws, with many states passing laws and regulat ions to further address these concerns. Conservation and restoration of natural ar eas are now common re quirements of land and community development mandated by local state, and federal agencies. As human population and development incr ease and the relative amount of these natural areas decreases, the quality and health of these natural areas are becoming of more importance. Agricultural practices, resource gathering, waterway diversions, fragmentation of natural hab itat and alteration of pattern s of natural vegetation are increasing with human population expansion and affecting even previ ously protected and isolated natural areas (Collinge 1996, Dale et al. 1998, Kramer 2005). Degrading habitat health, measured by biodiversity, is a majo r concern for natural resource managers and governmental agencies. Currently, the primar y driver for the loss in biodiversity is habitat modification and destruction due to ch anges in land-use practices (Kramer 2005). As human populations expand and the wildland -urban interface increases, more natural and protected areas are increasi ngly affected by these changes in land use. In addition, resource managers use many land management techniques, such as prescribed fire and mechanical removal of vegetation, directly in natural and protected areas in order to

PAGE 16

2 protect human interests. These direct impacts to the natural and prot ected areas should be monitored to assure minimal nega tive impact to biodiversity. As more natural and protected areas are subjected to these changing forces, land managers must be able to quantify changes in order to identify area s at risk. There are many tools in the environmental and ecological sciences for quantifying changes and differences in biodiversity, yet an initial inventory or measurement is needed for future assessments. Species Richness And Diversity: Estimating Their Values Diversity has been a persistent theme in ecology and is frequently seen as an indicator of ecological h ealth (Magurran 1988). Alt hough often incorrectly used interchangeably, species richness and diversity are distinct entities that relay sometimes quite different information. Species richness, the number of species in an area, is a simple yet informative measurement of a community. Intuitively, this simple measurement is ideal for comparing communities in conservation and management of biodiversity, assessing anthropogenic effects on protected lands and in fluencing public policy. Yet, this measure is not simple to accurately attain. Comple te species inventories usually require huge amounts of resources and expertise and are impractical and quite often impossible to compile. Almost every taxonomic survey will undoubtedly have undiscovered species. This trend is especially true with hyper-diverse taxonomic groups such as arthropods, nematodes, bacteria, and fungi. These groups are impossible to completely survey. Many groups of interest that are especially sensitive to anthro pogenic disturbance are these hyper-diverse taxonomic gr oups. In these cases, the best option for measuring

PAGE 17

3 richness is through sampling of ta rget environments or particul ar regions of interest, such as a habitat slated for conservation or development. Three broad categories of estimators exist to estimate the number of species in a community: estimators that fit a lognorma l abundance distribution and estimate the hidden or unsampled portion of the curve, esti mators that fit asymptotic equations to species accumulation curves, and non-parametri c estimators that use relative abundance of rare species to estimate the number of unseen species. In recent years, there has been a heightened interest in biodi versity that has resulted in new measurement techniques including niche apportionment models, ne w techniques for measuring taxonomic diversity, and improved methods of species richness estimation (Heltshe and Forrester 1983, Chao 1984, Hughes 1986, Chao 1987, Magurran 1988, Chao and Lee 1992, Colwell and Coddington 1994, Longino et al. 2002, Colwell et al 2004, Magurran 2004, Chao et al. 2005). Fortunately, long difficult math ematical calculations are usually no longer needed to estimate speci es richness through the many co mputer software packages readily available such as EstimateS (Colwell, 2005), Distance (Thomas et al. 2005), WS2m (Turner et al. 2003) and COMDYN (Hines et al. 1999), among others. Magurran (2004) provides more examples of such pr ograms and discusses their use, theory and sources of acquisition. These estimators of species richness are useful for letting re searchers know when they have sampled sufficiently to have conf idently surveyed the majority of species present. Such information is crucial since funds, time, and the taxonomic experts needed for reliable identification are usually in short supply (Hopkins and Freckleton 2002).

PAGE 18

4 Species richness is an informative meas ure but can, on its own, be misleading. Species richness is simply the number of species within a geographic area, but how desirable these species are is another matter. Quite often such information is used in conservation, preservation, re storation efforts and public policy. However, degraded areas not desirable for conservation may have a relatively high valu e of species richness (when compared to its desirable counterpart) due to exotic, feral a nd transient species. Disturbed habitats may serve as sinks draw ing in species from surrounding habitats. During my undergraduate rese arch (unpublished data) in Ev erglades National Park we saw more species of flower-visiting insects on the mowed roadsides (a disturbed habitat) than in the neighboring marl pr airie. Utilizing solely species richness, we could suggest that building more roads in the marl prairie would increase pollinato r species richness of the park. Of course such a conclusion a nd suggestion is ludicrous. Consequently, a detailed understanding of ecol ogical relationships of the sp ecies sampled is essential when applying such studies to land-use pr actices and policies (McCraken and Bignal 1998). Much more ecological insight can be at tained though measurement of diversity than solely through species richness estima tion. Diversity is a measure of species richness and the abundance distribution of thes e species, and therefore can detect effects unseen by species richness alone. Disturban ce effects may cause changes in diversity through shifts in the abundance of species or increases in the domina nce of some species. Magurran (1988) details the th eory and calculation of va rious diversity measures including the Shannon (1949) and Simpson (1949) indices of diversity. Such indices are easily calculated by traditional means, but ma ny software packages can quickly calculate

PAGE 19

5 these values, including update d versions of the indices th at take abundance data into account. Computer software such as Estim ateS (Colwell 2005) is available and can easily calculate richness and diversity values with a variety of estimators. Overview of Trap-nesting Hymenoptera Aculeate Hymenoptera are an integral part of most terrestrial ecosystems (Jenkins and Matthews 2004), including natural and distur bed habitats of north central Florida. These Hymenopterans fill many important roles, most commonly as pollinators, predators, and parasitoids. Changes in thei r populations would have a cascading effect, altering the habitats flora and fauna (Raw 1988, LaSalle and Gauld 1993, Neff and Simpson 1993, Jenkins and Ma tthews 2004). The majority of these Hymenopteran species are solitary in behavior. The nests of solitary bees and wasps are usually difficult to find and examine (Krombein 1967, Jayasingh and Freeman 1980, Al ves-dos-Santos 2003). Many nest in pre-existing cavities in various substrates such as wood, clay, rock and man-made structures (Bequaert 1940, Krombein a nd Evans 1954, Krombein 1967, 1970, Bohart and Menke 1976, Coville and Coville 1980, Coville 1982). This practice makes their nests not only difficult to find, but also extremely difficult to successfully extract and examine. The majority of these insects readily accept tr ap-nests (drilled wooden blocks) since they normally nest in preexisting cav ities created by other creatu res. In addition, many of these insects frequently reuse cavities for ne sting, allowing for long term observations on biology and ecology. Many successful studies of solitary bees and wasps have used trapnests to examine various aspects of their biol ogy and ecology. Trap-nests have been used to examine species composition and diversity at particular sites (P arker and Bohart 1966, 1968, Krombein 1967, 1970 and Camillo et al. 1995), population dynamics of occupants

PAGE 20

6 (Jayasingh and Freeman 1980), evaluation of habitat health and effects of habitat fragmentation (Frankie et al. 1998 and Tscharntke et al. 1998), genetic study (Packer et al. 1995) and survey of exotic species (Ma ngum and Sumner 2003). Trap-nests are extremely useful when the study focuses on gathering biological data of occupant species. Various studies have examined nest ing behavior and arch itecture (Medler 1967, Krombein 1967, 1970, Camillo et al. 1993, Pereira et al. 1999 and Alves-dos-Santos 2003), prey captured (Krombein 1967, 1970, a nd Camillo and Brescovit 1999, 2000), and associated parasitoids (Krombein 1967, 1970, Wcislo et al. 1996 and Scott et al. 2000). Trap-nests are a powerful survey tool that a llows collection of data on abundance, prey, habitat, phenology, and nest arch itecture that are not dete ctable through other survey methods that target Hymenoptera (Gathmann et al. 1994, Steffan-Dewenter 2002, Miyano and Yamaguchi 2001, Jenki ns and Matthews 2004). Effect Of Fire on Trap-Nesting Hymenoptera Fire is an integral part of forest and grassland ecosystems throughout the United States. Native Americans used fire for many purposes, such as a tool to clear areas for agriculture. Fire in natural areas, however, poses many hazards especially when in close proximity to urban areas. In response to this threatening hazard policies of complete fire suppression became popular in the 1920s and 1930s (Long et al. 2005). One result of the absence of periodic fires was a buildup of woody understory and excessive fuels. This caused subsequent fires to become more in tense, damaging and unmanageable. Forest management with prescribed burning is the curr ent popular tool. Prescribed fire has been shown to be effective in reducing hazardous fuels, disposing of logging debris, preparing sites for seeding or planting, improving wild life habitat, managing competing vegetation, managing invasive weeds, controlling insect s and diseases, improving forage for grazing,

PAGE 21

7 enhancing appearance, improving access and in perpetuating fire-dependent species (Cumming 1964, Helms 1979, Wade et al. 1988, Biswell 1999, Della Sala and Frost 2001, Fuller 1991, Mutch 1994, Paynter and Flanagan 2004 and Long et al. 2005). Many studies have been conducted to examine the effect of fire on plant (Main 2002, Vazquez et al. 2002, Laterra 2003, Lloret 2003, Reinha rt 2004, Schoennagel 2004, Barton 2005, Overbeck e t al 2005, Ansley et al. 2006) and animal communities (Chew et al. 1959, Kahn 1960, Lawrence 1966, Simons 1989, Mushinsky 1992, Saab and Vierling 2001, Cunningham et al. 2002, Meehan and George 2003) with some examining arthropods. Fire not only causes direct mortality in arthropods (Fay and Samenson 1993, Bolton and Peck 1946, Miller 1978, Evans 1984) but also indirectly affects arthropod communities via changes in plant commun ity composition and habitat alteration (Lawton 1983, Evans 1984). Unfortunately, most inve rtebrate studies have focu sed on terrestrial arthropods monitored via sweeping or pitfall trap s (Bess 2002, Brand 2002, Niwa 2002, Clayton 2002, Fay 2003, and Koponen 2005) and have ov erlooked trap-nesting Hymenoptera and other aerial insects. Hymenopterans Sampled and Summary This study provides a record and survey of trap-nesting Hymenoptera in five Florida state parks to further enhance the unders tanding of these habitats. It also provides a bench mark for future assessments of the insect fauna in these habitats. The following chapters provide a detailed inventory and biological notes on many trap-nesting Hymenopterans and associated ar thropods. These findings may be used in assessing habitat quality and pe rhaps aid in identifying any changes in biodiversity over time for the five state parks studied.

PAGE 22

8 Observations of biology and natural histor y are important to document, especially at the extremes of geographical range and in unique areas for a species that is cosmopolitan in range, such as Isodontia In particular I. mexicana has become established in Hawaii and France (Bohart a nd Menke 1976) providing un ique habitats in comparison to its native North America. This case provi des observations of I. mexicana in the southeast extreme of the geographica l range. ONeil and ONeil (2003) recently studied a population of I. mexicana in Montana and should provide a nice comparison to this Florida population. Debate s of species and subspecies versus clines and ecotypes commonly arise and it is important that we identify possible subject s to further examine the mechanisms of speciation and hybrid zones. For example, the splitting of Anisota senatoria into A. senatoria and A. peigleri by Riotte (1975) has been questioned by Tuskes et al. (1996) and throughout many taxa, taxonom ists that are lumpers or splitters are constantly at odds. In addition, such informa tion is valuable to help id entify possible projects for evolutionary biologists and studies in biogeography. For example, Mark Deyrup and Thomas Eisner (2003) examined the diffe rences of coloration between Florida Hymenoptera and their northern re latives (subspecies, clinal types, etc.) utilizing museum specimens and natural history data from past studies. Their pre liminary observations called for further exploration of Florida biogeography to help recognize distinctive species and examine mimetic complexes. Isodontia mexicana is a cosmopolitan species that is easily captured, studied and occurs sympatrically with I. auripes at this site Therefore chapter 2 examines ecology and natu ral history of a cosmopolitan species of a trap-nesting Hymenopteran, Isodontia mexicana (Hymenoptera: Sphecidae) and a sister

PAGE 23

9 species, I. auripes, in these Florida study sites. Th ese observations give a basis for geographical comparison for this wide-rangi ng species. Observations and findings for I. mexicana populations have been recently published by ONeill (2001), ONeill and ONeill (2003), and historically by Bohart and Menke (1963), Lin (1966), Krombein (1967), Bohart and Menke (1976). These stud ies took place in di stant parts of the geographical range, as compared to the Fl orida population obse rved, and may offer insight and inspiration fo r additional study. Such st udies, for example, Sears et al. (2001), may examine behavioral, biological and na tural history differences of a species in two extremes of its range. Such information may also be useful for examining aspects of biogeography and evolutionary history (Deyrup and Eisner 2003). For chapter 2 there are two main objectives: 1) Examine similar features and aspects of natural history of Florida population that was examined in Montana popu lations and infer whether these features warrant further biogeographical examination, 2) and since the tw o sister species ( I. mexicana and I. auripes ) occur sympatrically in these Florida sites, determine if they differ substantially in the examined featur es to inspire a clos er look at possible speciation/separating mechanisms. In Chapter 5 a second species, Monobia quadridens (Hymenoptera: Vespidae), is examined in the same manner. Monobia quadridens is also a cosmopolitan species in terms of geographical range and has been previously studied in different areas by Bequaert (1940), Krombein (1967), and Krombein et al (1979). Monobia quadridens has not been studied in the recent literature in terms of biology and natural history. The objectives for chapter 5 are to 1) determine if this wasp has preference in cavity size for nesting, 2) determine the nest architecture, 3) determine the range of prey provisioned by

PAGE 24

10 this wasp in this portion of its geographic range, compareing findings to other published records, 4) and determine the emerging se x ratio of trap-neste d individuals. In Chapter 3 the spider prey cap tured by the most abundant trap-nesting Hymenopteran in this study, Trypoxylon lactitarse (Hymenoptera: Sphecidae) is examined. Previous studies (Rau 1928, Krombein and Evans 1954, Krombein 1956, 1967, Medler 1965, Lin 1969, Coville 1979, 1981, 1982, Coville and Coville 1980, Genaro et. al. 1989, Camillo et. al. 1993, Genaro and Alayon 1994, Jimenez and Tejas 1994, Camillo and Brescovit 1999, 2000) have hi ghlighted differing prey preferences in various species of Trypoxylon and Rehnberg (1987) and Camillo and Breviscovit (1998) examined prey preferences for Trypoxylon lactitarse This chapter, therefore, has five main objectives: 1) Determine what prey T. lactitarse is provisioning at these Florida sites, 2) Determine if T. lactitarse is a generalist or special ist in terms of prey provisioned, 3) Determine what, if any, is T. lactitarse s prey preference, 4) Does T. lactitarse s prey preference seem to differ between sites? 5) Determine the benefits and potential problems with using this wasp (and potentially other sp ider-provisioning trapnesters) as a sampling tool for estimati ng spider abundance and species richness. In Chapter 4 the impact prescribed fire has on trap-nesting Hymenoptera and associated arthropods is examined and pos es the following questions: 1) Does overall diversity and species richne ss of trap-nesting Hymenoptera differ between burned and unburned sites? 2) In terms of species samp led, how similar are the burned and unburned sites? 3) Is the diversity of sampled func tional groups (predator, parasitoid and pollen specialists) affected by fire? 4) What species, if any, seem to be negatively or positively affected by fire in terms of abundance?

PAGE 25

11 There was a large amount of data collect ed in the process of addressing the objectives of chapters 2-5. This final chap ter summarizes and reports this mass data, which are highly desirable, and require d by the Florida State Department of Environmental Protection. The objectives of chapter six are 1) re port the abundances and species richness of a ll trap-nesting Hymenoptera and asso ciated arthropods sampled at each of the five surveyed Florida State Parks and 2) determine, by using estimators, if the inventory offered can be considered adequate and, if adequate, estimate total species richness and diversity of tr ap-nesting hymenopterans and a ssociated arthropods for each state park surveyed.

PAGE 26

12 CHAPTER 2 NEST ARCHITECTURE, PREY, AND SE XUAL DIMORPHISM IN THE GRASSCARRYING WASPS ISODONTIA ( MURRAYELLA ) MEXICANA (SAUSSURE) AND ISODONTIA AURIPES (FERNALD) (HYMENOPTERA: SPHECIDAE: SPHECINAE) Abstract Isodontia ( Murrayella ) mexicana (Saussure) and Isodontia ( Murrayella ) auripes (Fernald) nested in trap nests at four different state parks in north central Florida. Nests consisted of fragments of native nutrush grasses (Cyperaceae: Scleria sp.). Females provisioned either a communal cell, or sepa rated cells with 1 to 15 tree crickets (Gryllidae: Oecanthinae: Oecanthus ), bush crickets (Gryllidae: Eneopterinae, Orocharis ), meadow katydids (Tettigoniidae: Conocephalinae, Odontoxiphidium ), coneheaded katydids (Tettigoniidae: Copiphorinae, Belocephalus, Conocephalus and Neoconocephalis ), and false katydids (Tet tigoniidae: Phaneroperinae, Scudderia ). Both Isodontia species displayed the sexual size diffe rence typically found in the Sphecidae with females significantly larger than male s. Female-biased provisioning has been shown to occur in other populations of I. mexicana and seems to occur in the Florida populations as well. Although I. auripes exhibits this sexual size tr end, the communal brood chamber of the nest architecture rule s out any provisionin g difference as the cause for the size difference. Introduction Isodontia (Hymenoptera) is one of the cavit y-nesting genera of the Sphecid subfamily Sphecinae and this genus is unique in its nesting biology. While other solitary aculeate wasps that nest in pr e-existing cavities use mud, aggl utinated sand, plant resin or

PAGE 27

13 masticated plant materials as nest partiti ons and plugs (ONeill and ONeill 2003, ONeill 2001), Isodontia use dry grass leaves that they cut, pack, and twist into position within the nest (Lin 1966, Krombein 1967, Bohart and Menke 1976, ONeill and ONeill 2003). They are commonly known as the grass-carrie r wasps since they can be observed flying with bits of grass as long as 80 mm in th eir mandibles (Bohart and Menke 1963). Several species of Isodontia construct nests with a common br ood chamber that contains as many as 12 larvae feeding on a common prey mass (Bohart and Menke 1976, ONeill and ONeill 2003). Here I report on a two-year study where female Isodontia mexicana and Isodontia auripes nested in trap-nests se t up in four state parks in north central Florida. I examined 90 nests of I. mexicana and 89 nests of I. auripes out of 235 total Isodontia nests and recorded information on nest structure, prey, sex ratio, sexual size dimorphism, emergence schedules and parasitoids. I th en determined if they exhibited sexual dimorphism typically seen in other Spheci d wasps, identified nest architecture highlighting the difference between the two species, identified prey used to provision nests and examined survival of brood including parasi toids and predators of these wasps. There are two main objectives for this study: 1) Examine similar features and aspects of natural history of Florida population that were examined in Montana populations and infer whether these features warrant furthe r biogeographical examination, and 2) since I. mexicana and I. auripes occur sympatrically in these Fl orida sites, determine if they differ substantially in the examined featur es to inspire a clos er look at possible speciation/separating mechanisms.

PAGE 28

14 Methods and Materials Tools and Trap Preparation The traps used in this study were fabr icated from seasoned 37-mm x 86-mm x 2.4m pine/spruce timbers obtained from a local home improvement store. The pine/spruce timbers were cut into 10-cm-long blocks. Tw o cavities of one of five diameters (3.2, 4.8, 6.4, 7.9 or 12.7-mm) were drilled into each block. Cavities were drilled to a depth of 80 mm on each short side (the 37-mm side), o ffset approximately 10-mm from the center point. Traps were assembled using one block of each diameter with the smallest cavity on top and the largest on the bottom. Blocks were stacked so that no cavity was situated directly above or below a cavity in the ad jacent block. The fi ve blocks were bound together with strapping tape (3M St Paul, MN), and 16-gauge wire was used to further bind the stack and suspend the trap from trees and shrubs at the field sites. Each bundle of five blocks was considered to be a single trap. Field Sites I set traps at five locations: 1) Suwann ee River State Park in Suwannee County (30 23.149 N, 083 10.108 W), 2) Mike Roess Gold Head Br anch State Park in Clay County (29 50.845 N, 081 57.688 W), 3) Devils Millhopper Geolog ical State Park in Alachua County (29 42.314 N, 08223.6924 W), 4) San Felasco Hammock Preserve State Park (29 42.860 N, 08227.656 W) in Alachua County and 5) Silver River State Park in Marion County (29 12.317 N, 082 01.128 W). The habitats surveyed at Suwannee River State Park were burned and unburned sa nd hill habitat, while the habitat at Mike Roess Gold Head Branch State Park was burne d sand hill pineland a nd ravine. Sites at San Felasco Hammock Preserve State Park consisted of upland and mesic hardwood hammock. Surveyed areas of Devils Millhopper Geological State Park consisted of pine

PAGE 29

15 flatwood habitat and sites at S ilver River State Park consiste d of river habitat and upland mesic forest. Descriptions of these habitats can be found in Franz and Hall (1991). Field Placement Transects were set up with ten traps placed approximately 10 m apart and hung approximately 1.5 m off the ground on trees or limbs with placement on dead standing wood preferred. Transects were initially es tablished (direction a nd distance from center of plot) randomly. Four tran sects were established in Suwannee River State Park while three transects were established Mike Roe ss Gold Head Branch State Park. Three transects were established in San Felasco St ate Park but size constraints only allowed a single transect in Devils M illhopper State Park. Finally, two transects were set up in Silver River State Park. Transects were in the field from Apr il 2003 until January 2005. Field Collection and Laboratory Rearing Traps remained in the field two years and were checked monthly. Preliminary field tests revealed that one-month intervals were sufficient to avoid trap saturation (no available cavities). Traps were considered o ccupied when insects we re observed actively nesting, harboring or had sealed a cavity w ith mud or plant material. Occupied traps were removed and replaced with a new trap. These occupied traps were brought into the forest entomology lab at the University of Florida in Gainesville, FL, for processing. Occupied blocks were removed for obse rvation while unoccupied blocks were reincorporated into replacement traps. Each occupied cavity was given a unique reference number. Location, date of collection, diameter of cavity, and various notes describing the nature of the occupants and/or plug were r ecorded for each reference number. Occupied cavities were then covered with a 2, 4, 6, or 8dram glass shell vial. The shell vials were

PAGE 30

16 attached to the wood section with maski ng tape (Duck, Henkel Consumer Adhesive Inc., Akron Ohio) appropriate for wood applicati on. These sections were then placed in a rearing room and observed daily for emergen ce. The rearing room was maintained as nearly as possible at outside mean temp eratures for Gainesville, Florida. When emergence occurred, the specimens were removed, preserved and given the same reference number as the cavity from which they had emerged. Dates of emergence, identification of occupants, measurements and notes were taken for each cavity at emergence. When an insect was harbori ng or actively tending a nest, it was captured, identified, and given a reference number corre sponding to the cavity. The contents of the nest/cavity were then extracted and recorde d. After the contents were extracted, the wood block was reused in replacement traps. These processed blocks were re-drilled to the next larger diameter cavity to eliminate any alterations or markings (either physical or chemical) by the previous occupant prior to reuse. Specimen Diagnostics and Identification Isodontia auripes and I. mexicana occur sympatrically a nd it is important to distinguish between the two species. Appe ndix B (adapted from Bohart and Menke (1963)) provides characters to distinguish between the species and sexes of each species. All cavity nesters and their prey were iden tified by the author with some specimens identified and/or verified by entomologists Jim Wiley1, Lionel Stange1, Thomas Walker2, and John M. Leavengood Jr.1,2 (Florida State Collection of Arthropods1 Gainesville, FL and University of Florida2, Gainesville, FL). Voucher sp ecimens have been deposited at the Florida State Collection of Arthr opods in Gainesville, Florida.

PAGE 31

17 Statistical Analysis Descriptive statistics (means, ranges, SD, etc.) were calculated the Microsoft Excel statistical package (Microsoft, Inc, CA). Since all sites where both Isodontia species occurred were similar (sand hill habitat) da ta were pooled for analysis. One main concern may be the difference of burned and unburned sandhill habitats. Collections for both species in unburned habitat was quite lo w with only 11% of total abundance for I. mexicana and 10% of total abundance for I. auripes. Chi-squared goodness of fit test was used to examine nest diameter pref erence for pooled habitats and burned and unburned habitats separately. The assumption was that wasps would nest equally in all diameters. Since neither species nested in 3.2-mm diameter cavities, that cavity size was omitted from analysis. Head width of adults was measured to the nearest 0.01mm using an ocular micrometer. Head capsule comp arison was analyzed using, the Mann-Whitney test. Results Habitat Both Isodontia species were captured at Suwannee River, Gold Head, and Devils Millhopper State Parks. Only Isodontia mexicana was captured at San Felasco S. P., and neither Isodontia species was captured at Silver River S. P. Nest Architecture I examined 90 nests of Isodontia mexicana and 89 nests of I. auripes An additional 56 cavities that had Isodontia nests were trapped but species identification of these nests was not possible due to predation or disturbance. Females of I. mexicana preferred to nest in 7.9-mm cavities (20 of 90 nests) a nd 12.7-mm cavities (68 of 90 nests; Table 2) and none nested in 3.2-mm and 4.8-mm cavities. Only 2 of 90 females nested in a 6.4-

PAGE 32

18 mm cavity. Females did not equa lly nest in all diameters (c hi-squared contingency table, X2 = 133.5, df = 3, P< .001). Results were si milar when the population was separated into burned and unburned habitats (P< 0.001, df=3, X2 burned= 108.4, Nburned = 10, X2 unburned= 30, Nunburned=80). Likewise, females of I. auripes nested in mostly 12.7-mm cavities (85 of 89 nests). Only one nest of I. auripes was placed in a 4.8-mm cavity and 3 nests were placed in 7.9-mm cavities (figure 24)(chi-squared contingency table, X2 = 236.1, df = 3, P<0.001). Results were similar when the population was separated into burned and unburned habitats (P <0.001, df=3, X2 burned=215.2, Nburned = 8, X2 unburned= 17.2, Nunburned= 81) Both species used grass as b ack wall and opening plugs, and I. mexicana also made brood cell partitions out of the grass material The grass did not have any binding agents (such as resin or secretions), but was twisted and compacted into position. Many nests had a slight amount of grass within the br ood cell(s) when extrac ted and the pupae had a fair amount of grass pieces adhering to them (Figure 2-3). The occurrence of grass within the brood cell suggests that females may line the cell, but female and larval activity cannot be distinguished wit hout further observation. All I. mexicana observed utilized separated brood cells (figure 2-1), with indivi dual cells each 20-30mm. Each egg was laid on an orthopteran prey item and then separa ted from the next egg and provisioned prey mass by a tightly packed partition of grass. These nests had a mean of 2.50 brood cells (SD = 0.88, range = 1-4, N= 90). The majority of I. mexicana nests had 3 cells, however the over-wintering type of nest with only one brood cell may have resulted in a lower mean. In fact, when these outliers are removed the mean becomes 2.86 (SD= 0.50, range

PAGE 33

19 2-4 N= 72). ONeill and ONeill (2003) observed I. mexicana in Montana had a range of 1-6 separated cells in cavities 15 cm deep. In contrast, Isodontia auripes consistently used a common brood cell in its ne st architecture. Cells tended to be 50-60 mm in length. All provisions and eggs were laid in a singl e cell without any intern al partitions. Both species used neatly coiled, tight ly packed plugs of grass for the back end of the nest and to close the opening. These ti ghtly packed plugs ranged from 6-10 mm thick for all nests in 12.7-mm and 7.9-mm diameter cavities. The tightly packed plugs in the few nests in smaller diameter (4.8-mm and 6.4-mm) cavities te nded to be slightly thicker at 10-15 mm. Both species also used a loose plug between the opening and the outermost tightly packed plug. These loose plugs tended to include longer lengths of the grass and occasionally contained seed heads. Loose plugs occupied the out ermost 5-20 mm of the cavity and always extended beyond the cavity opening. These plugs of grass resemble broom-like tufts and regularly extended 30-60 mm beyond the cavity opening and occasionally reached up to 100 mm beyond the cavity opening. The few seed heads included in the loose plug material allowed fo r identification of the grasses used by these wasps. Mark Garland (Botanist at the Flor ida Department Agriculture, Division of Plant Industry, Gainesville, Florida) identified the materials as the native nutrushes Scleria sp ( ciliate/ pauciflora ) (Cyperaceae). Sex Ratio and Sexual Dimorphism Isodontia mexicana that emerged from trap nests had a sex ratio of 1.2 males per female (N= 119). Isodontia auripes that emerged from trap nests had a sex ratio of 5.3 males per female (N= 144). Isodontia mexicana display sexual size dimorphism typical for the Sphecidae (ONeill 2001). Females that emerged from nests (mean head width = 3.10 mm, SD =

PAGE 34

20 0.20 mm, range 2.7-3.6 mm, N= 53) were larger than males (mean head width = 2.84 mm, SD = 0.18 mm, range 2.4-3.2 mm, N= 65; Mann-Whitney U= 2930, P < 0.0001). Twenty-three percent of the females were la rger than the largest male and 15% of the males were smaller than the smallest female. These differences are far less than in the Montana populations of I. mexicana examined by ONeill and ONeil (2003). Isodontia auripes females that emerged from nests (mean head width = 3.30 mm, SD = 0.397 mm, range 2.4-3.9 mm, N= 23) were larger than males (mean head width = 3.01, SD = 0.259 mm, range 2.2-3.5, N = 121; Mann-Whitney U= 2276.5, P < 0.0001), with 30% of females larger than the largest male and 3% of the males being smaller than the smallest female. Overall, I. auripes tended to be larger than I. mexicana (females: Mann-Whitney U= 867.5, P < 0.01, males: Mann-Whitney U= 6270, P< 0.001). Prey Extracting the contents of 20 Isodontia nests yielded samples with provisions in an identifiable condition. Thomas J. Walker (P rofessor Emeritus, University of Florida, Gainesville, Fl.) positively identified prey provisions from these nests. Nests of I. mexicana contained Odontoxiphidium apterum Morse 1891 (Tettigoniidae: Conocephalinae), Oecanthus quadripunctatus Beutenmuller 1 894 (Gryllidae: Oecanthinae), Belocephalus sp. (Tettigoniidae: Copiphorinae), Orocharis luteolira T Walker1969 (Gryllid ae: Eneopterinae) and Scudderia sp. (juv) (Tettigoniidae: Phaneroperinae) (Table 2.4). Nests of I. auripes contained Odontoxiphidium apterum Oecanthus celerinictus T Walker 1963 (Gryllidae: Oecanthinae), Oecanthus niveus (De Geer 1773) (Gryllidae: Oecanthinae), Orocharis luteolira Neoconocephalis spp. (juv) (Tettigoniidae:

PAGE 35

21 Copiphorinae), Conocephalus brevipennis (Tettigoniidae: Conocephalinae), Scudderia furcata sp. (juv) (Tettigoniidae: Phaneroperinae), and other Oecanthus spp. (juv) (Table 2.5). Amount of prey provisioned varied great ly. Provisioned prey ranged from 1-19 prey items per nest. Nests and/or brood cells that had one or few prey items tended to contain large adult tet tigoniids and those nests and/or br ood cells with many prey tended to contain juveniles and/or sma ll species of gryllids. Survival Ants (C rematogaster spp.) pillaged many Isodontia nests and it was impossible to identify the species of Isodontia Therefore, overall Isodontia survival was calculated. A total of 235 nests were examined yieldi ng 320 individuals of 529 resulting in a survival/emergence percentage of 60.49%. Mean brood per nest was 2.27 (SD= 1.311), yet this number is not useful since the two sp ecies differ in nesting strategies. Fifteen nests were lost to Crematogaster ant raids which accounted fo r 8.7% mortality of brood. Seven nests were lost to bombyliid fly parasitoids in the genera Anthrax and Lepidophora accounting for 3.4% of brood mortality. In ad dition, 2 nests were lost to a phorid fly parasitoid and 1 nest was lost to a male mutillid in the genus Spheropthalma ( Sphaeropthalma ), most likely the species pensylvanica One nest was lost to supersedure (the act of taking over by a s econd individual of the same or different species of a cavity partially stored by the first individual) when a vespid, Stenodynerus sp. placed her mud nest in front of the I. auripes nest in progress. Discussion Documented observations of biology and natu ral history are importa nt especially at the extremes of geographical range and in uni que areas, for a species that is cosmopolitan

PAGE 36

22 in range, such as the genus Isodontia. In particular I. mexicana has become established in Hawaii and France (Bohart and Menke 1976) prov iding unique habitats in comparison to its native North America. Florida populations, as in this case, provi de observations in the southeast extreme of the ge ographical range. ONeil and ONeil (2003) examined a population in Montana. With the ever-ongoing debate of species and subspecies versus clines and ecotypes it is important that we identify possible subjects to further examine the mechanisms of speciation and hazy hybr id zones. For example, the splitting of Anisota senatoria into A. senatoria and A. peigleri by Riotte (1975) has been called into question by Tuskes et al. (1996) and through out many gr oups taxonomists that are lumpers or splitters are constantly at odds. In addition, such informa tion is valuable to help id entify possible projects for evolutionary biologists and studies in biogeography. For example, Mark Deyrup and Thomas Eisner (2003) examined the diffe rences of coloration between Florida Hymenoptera and their northern re latives (subspecies, clinal types, etc.) utilizing museum specimens and natural history data from past studies. Their pre liminary observations called for further exploration of Florida biogeography to help recognize distinctive species and examine mimetic complexes. Mimetic complexes are adaptive syndromes that reflect the evolutionary history of species, and histor ical events that cannot be repeated by an investigator and leave no foss il record (Deyrup and Eisner 2003), yet data such as these can help examine such events. Isodontia mexicana is a cosmopolitan species that is easily captured, studied and occurs sympatrically with I. auripes at these Florida sites Such data can be extremely useful when examining speciation by

PAGE 37

23 highlighting divergence of th ese two sympatric sister sp ecies and by highlighting the divergence of I. mexicana in its geographical extremes. My study is by no means comprehensive, but does offer substantial data in a unique part of I. mexicana range that may offer insight into divergence and speciation processes. Habitat Isodontia auripes was not captured in San Felasco Hammock Preserve State Park, but was captured a few kilometers away at Devils Millhopper Geological State Park. Isodontia mexicana was not particularly abundant at San Felasco and I. auripes was probably present at San Felasco just not captured. All transects with Isodontia were in or adjacent to sand hill habitat that tended to be xeric. Silver River State Park does not have sandhill habitat that was particul arly of substantial size or xe ric in nature and lacks both I. auripes and I. mexicana. Both species were present in Suwannee River S.P. in both recently burned and unburned sandhill habitat, although the majority of nests (82% for I. mexicana and 87% for I. auripes ) were captured in burned habita ts (see chapter 4). Nest Architecture Several species of Isodontia have nests that contain a common brood cell where up to 12 larvae will feed on a single prey mass. Isodontia auripes exhibited this behavior and all nests (apart from singl e over-wintering emergence) of this species had common brood cells. Although I. mexicana has been reported to have a common brood cell in some populations(Krombein 1967), the populatio ns I studied had separated brood cells within each nest (Figure 2-1). Bohart and Menke (1963) reported that some Isodontia use grass to line the nest. ONeill and O Neill (2003) reported that the population of I. mexicana they observed in Montana did not line nest cells. Pupae of both species had some amount of grass incorporated into th e cocoon suggesting ther e was some manner of

PAGE 38

24 grass lining each cell (Figure 2-3). Krombein (1970) observed larvae of Isodontia auripes pulling grass fragments from the plugs and incorporating into the spinning of the cocoon. This behavior may be the explanati on for the grass fragments incorporated into cocoons. The small amount of grass incorporated into the cocoons in addition to the lack of remaining grass in the cham bers suggests that these specie s in fact do not actively line the brood cells with grass. The nature of my traps did not allow for direct observation of pupating activity. Isodontia females of both species plug the opening of nest cavities with clumps of loosely packed grass. Thes e plugs of grass resemble br oom-like tufts and regularly extended 30-60 mm beyond the cavity opening and occasionally reached up to 100 mm. Bohart and Menke (1976) reported these plugs extending only up to 50 mm beyond the cavity opening. In Montana populations of I. mexicana, many plugs were flush with the opening, that tufts apparently being clippe d short by the female (ONeil and ONeil 2003). I did not observe any of this clipping in Florida. The only ne sts that did not have the tufts of the closure plug extending be yond the opening were those that completely lacked the closure plug. These nests ha d only the final tightly packed partition suggesting that the plug had fallen out, the fema les had not completed her nest at time of collection or she had died before nest completion. I observed an intere sting deviation of I. mexicana nest architecture. About 18 nests were found to have one separated brood ce ll was provisioned with prey and a single egg, then the remaining nest was packed with both tightly packed partitions and loose plugs. Only 18 such nests were extracted and recorded, yet in the spring many other nests yielded only one adult without evidence of other pupa e. It was impossible to

PAGE 39

25 determine if these nests were a fall-winter be havior (rather than the other chambers eggs not successfully hatching or being pillaged by ants) since the emerging adults are quite destructive of the nest commonly pushing the entire contents of the nest out of the cavity. These types of nest may be the result of an end of life span behavior of the nesting females. These fall-winter types were usuall y found in the fall with the earliest collected in August. However, normal nests with multiple brood cells were found throughout the year including in December. Sex Ratio and Sexual Dimorphism The two species of Isodontia had dramatically different sex ratios. Isodontia mexicana displayed a sex ratio of 2.1: 1 (M: F). However, I. auripes displayed a sex ratio drastically different at 5:1 (M: F). O Neill and ONeill (2003) found that males of I. mexicana tended to emerge from smaller diameter ne sting cavities. Yet, the majority of I. auripes occupied the largest diameter nest ( 12.7 mm) and one of the few nests in smaller diameters yielded a female (7.9 mm). Therefore, this conclusion does not seem applicable. An alternative explanation is th e Trivers-Willard hypothesis which states that sex allocation is condition depe ndant (Trivers and Willard 1973). This assumes that females can control the sex of the offspring, a nd it has been shown th at nest-provisioning hymenopterans precisely determine th e sex of each offspring (Green et al., 1982, ONeill 2001 ONeill and ONeill 2003). In fact, the majority of Isodontia nests were located in recently burned (within 1 and 2 years) sandhill habitat. Females can be expected to produce more of the sex for which quality ma kes the greatest difference in reproductive success (Clutten-Block et al 1984, Miller and Aviles 2000). Therefore, in poor conditions, males should be produced and in good conditions females should be produced, assuming the burned condition is a detrimental condition. Unfortunately, the

PAGE 40

26 relatively low amount of Isodontia nests captured in unburned areas did not allow a meaningful comparison. It can be speculate d that the fire event eliminated nesting materials and reduced prey populations. The sites were surveyed for two years beyond the fire event. Grasses tend to respond pos itively and quickly after a fire event. The reduction of plant biomass may have provided less harborage for prey items and therefore easier hunting for the wasps. Yet, the sex ratios were equal in both years following the fire event. The relative amount of prey a nd nesting material needs to be known as well as further study beyond the fire event to detect any response lag. However, I. mexicana did not exhibit such a skewed sex ratio that I. auripes displayed. Since both species have similar biology (same prey, nesting habitat, nesting materials) a nd occur sympatrically, could interspecific competition be the driving force? Unfort unately the current data set cannot suggest any answers with any kind of confidence. Additional research focusing on interspecific competition is needed. Prey The prey provisioned by both Isodontia species has substantial overlap between my records and those reported in th e literature (Table 2.4 & 2.5). Orocharis luteolira (Gryllidae: Eneopterinae) and Belocephalus sp (Tettigoniidae: Copiphorinae) were the only prey for I. mexicana that were not reporte d in the literature. Oecanthus celerinictus (Gryllidae: Oecanthinae) Odontoxiphidium apterum (Tettigoniidae: Conocephalinae) and Neoconocephalus sp (Tettigoniidae: Copiphorin ae) were the only prey for I. auripes not reported in the literature. Both Isodontia species examined provisioned 3 species in common, but I. mexicana provisioned 6 unique species while I. auripes provisioned 2 unique species (Table 2.1). Under closer observation with more nests dissected, the range of prey items for both Isodontia may become more similar. Amount of prey

PAGE 41

27 provisioned varied greatly. Provisioned prey ranged from 1-19 prey items per nest. However, since these wasps prey on both a dult and juvenile prey, biomass of the provisions is probably more important than quantity. Nests and/or brood cells that had one or few prey items tended to contain larg e adult tettigoniids and those nests and/or brood cells with many prey tended to contain juveniles and/or small species of gryllids. Therefore, prey per nest and pr ey items per brood cell are statis tics of questionable value. Nesting females seem to be filling brood cells rather than provisioning a particular number of prey items per egg. These Isodontia species displayed the typical se xual size difference found in the Sphecidae, suggesting a possible provisioni ng strategy by nesting females. Femalebiased provisioning has been shown to o ccur in Montana populations (ONeill and ONeill 2003) of I. mexicana and it also seems to occu r in Florida populations of I. mexicana. However, I. auripes exhibited this sexual size trend, yet the communal brood chamber nest architecture rule s out any provisioning differen ce as the cause for the size difference since both sexes regularly emer ged from the same nest. A possible explanation could be that females deposit fema le eggs on prey earlier than male eggs. Since eggs that are deposited earlier tend to hatch earlier, fe male larvae would have more time with the prey mass and would most likely consume more of the prey mass. Krombein (1970) reported that the la st egg to hatch in a nest of I. auripes where 6 eggs were laid actually died from lack of food. One nest experienced supersedure. A vespid, Stenodynerus sp., usurped a nesting I. auripes female and placed her mud nest in front of the I. auripes nest in progress of provisioning. Two I. auripes successfully developed behind the Stenodynerus nest, but

PAGE 42

28 were unable to break through the mud partitions of the vespid nest and subsequently died. These I. auripes adults were of normal size suggesti ng that the female oviposits on prey after there are a number of sufficient prey items in the communal chamber to support that egg. This was an isolated observation, however and more data are needed to substantiate this hypothesis. Conclusion Nest structure and prey in the observed Fl orida populations were similar to those reported by Medler (1965), Kr ombein (1967), Bohart and Menke (1976) and ONeil and ONeil (2003). Both Isodontia mexicana and I. auripes displayed sexual dimorphism typically found in the Sphecidae. More importantly, these two sister species occur sympatrically and have a broad range of overlap in biology and prey species, yet they had extremely different sex ratios. Sex allocati on is typical in the S phecidae and a skewed ratio suggests a harsh environment. There is substantial overlap in prey items, but there are unique prey items to each species. Wh ether those prey items remain unique as sampling for provisioned prey is increased is unknown. In addition, prey populations were not sampled or estimated leaving possible disparity of unique prey unknown. Whether the skewed sex ratio is a result of di rect interspecific competition or differences in mutually exclusive prey populations, the relationship of these two sympatric sister species may offer evaluation of compet itive exclusion, dive rgence and possible speciation events. Acknowledgements All research and collections were comp leted with permission of the Florida Department of Environmental Protection Divi sion of Parks and Recreation under permit numbers 11250310 and 08170410.

PAGE 43

29 Figure 2-1. Cross section of Isodontia mexicana nest in a 12.7mm cavity Figure 2-2. Isodontia auripes larvae on provisioned Scudderia furcata

PAGE 44

30 Figure 2-3. Isodontia cocoon Frequency of cavities nested in by I. auripes and I. mexicana0 10 20 30 40 50 60 70 80 90 3.24.86.47.912.7 Cavity diameters (mm)Frequency of nests I. auripes I. mexicana Figure 2-4. Frequency of cavities nested in by I. auripes and I. mexicana

PAGE 45

31 Summary of emerged Isodontia sp.0 20 40 60 80 100 120 140 nestsmalefemaleFrequency I. auripes I. mexicana Figure 2-5 Summary of emerged Isodontia mexicana and Isodontia auripes from captured nests

PAGE 46

32 Table 2-1. Comparison of Isodontia auripes and I. mexicana Isodontia auripes Isodontia mexicana Identification Violaceous wings Red-brown legs Male: 18 mm, Female: 19 mm Clear wings, with black veins and smoky brown along anterior margin Black legs Male: 16 mm, Female: 17 mm Architecture Single chamber 50-60mm 3 (1-4) chambers 20-30mm each Prey (Bold = Prey provisioned by both species) (Gryllidae: Oecanthinae); Oecanthus celerinictus T Walker 1963 Oecanthus niveus (De Geer 1773) Oecanthus spp. (juv). (Gryllidae: Eneopterinae); Orocharis luteolira, T Walker1969 (Tettigoniidae: Conocephalinae); Odontoxiphidium apterum, Morse 1891 Conocephalus brevipennis Scudder 1862 (Tettigoniidae: Copiphorinae); Neoconocephalis spp. (juv) (Tettigoniidae: Phaneroperinae); Scudderia furcata Bruner 1878 Scudderia sp. (juv) (Gryllidae: Oecanthinae); Oecanthus quadripunctatus Beutenmuller 1894 (Gryllidae: Eneopterinae); Orocharis luteolira, T Walker1969 (Tettigoniidae: Conocephalinae); Odontoxiphidium apterum Morse 1891 (Tettigoniidae: Copiphorinae); Belocephalus sp. (Tettigoniidae: Phaneroperinae); Scudderia sp. (juv) Dimorphism (Mean head width (mm)) Females: 3.30 + 0.397 Males: 3.01 + 0.259 Females: 3.10 + 0.20 Males: 2.84 + 0.18 Sex ratio (M:F) 5:1 1.2:1 Cavity diameters (mm) nested 6.4, 7.9, 12.7 4.8, 7.9, 12.7 Habitat Sandhill, Ravine (adjacent to sandhill), Pine Flatwoods Sandhill, Ravine (adjacent to sandhill), Pine Flatwoods, Mesic hardwood hammock State Parks Suwannee River State Park, Mike Roess Gold Head Branch State Devils Millhopper Geological State Park Suwannee River State Park, Mike Roess Gold Head Branch State Devils Millhopper Geological State ParkSan Felasco Hammock Preserve State Park

PAGE 47

33 Table 2-2. Frequency of cavities nested in by I. auripes and I. mexicana Cavity diameter (mm) I. auripes I. mexicana 3.2 0 0 4.8 1 0 6.4 0 2 7.9 3 20 12.7 85 68 Total nests 89 90 Table 2-3 Summary of emerged I. auripes and I. mexicana Number of nestsMale Female Total adults (Emerged) Isodontia auripes 90 66 53 119 Isodontia mexicana 89 131 23 154 Isodontia (unknown sp) 56 _ Total 235 273

PAGE 48

34 Table 2-4 Prey records for I. mexicana Prey Present study ONeill & ONeill 2003 Bohart & Menke 1976 Krombein 1967 Lin 1966 Medler 1965 Gryllidae: Gryllinae Gryllus sp. X X Gryllidae: Eneopterinae X X Orocharis sp X X Orocharis luteolira T Walker X Gyllidae: Oecanthinae X X X X X X Oecanthus sp. X X X X X X O. exclamationis Davis* O. nigricornis F Walker X X O. quadripunctatus Beutenmuller X X X O. fultoni T Walker X X O. niveus (DeGeer) X X X O. saltator Uhler X O. celerinictus T Walker* Neoxabea sp. X X N. bipunctata (DeGeer) X Tettigoniidae: Conocephalinae X X X Conocephalus sp X X X C. brevipennis (Scudder)* C. fasciatus (DeGeer) X C. saltans (Scudder) X Odontoxiphidium sp X X X O. apterum Morse X X Orchelimum sp. X X Tettigoniidae: Copiphorinae

PAGE 49

35 Table 2-4 Continued Prey Present study ONeill & ONeill 2003 Bohart & Menke 1976 Krombein 1967 Lin 1966 Medler 1965 Belocephalus sp. X Neoconocephalus sp. X Tettigoniidae: Phaneropterinae Scudderia sp. X X Scudderia furcata Brunner* Tettigoniidae: Tettigoniinae* Atlanticus sp.* A. gibbosus Scudder* Neobarretta sp. X Not found but present in Isodontia auripes nests Table 2-5 Prey records for Isodontia auripes Prey Present study Bohart & Menke 1976 Krombein 1970 K rombein 1967 Gryllidae: Gryllinae Gryllus sp. Gryllidae: Eneopterinae X X X Orocharis sp X X X Orocharis luteolira T Walker X X Gryllidae: Oecanthinae X X X Oecanthus sp. X X X O. exclamationis Davis X O. nigricornis F Walker* O. quadripunctatus Beutenmuller* O. fultoni T Walker* O. niveus (DeGeer) X X O. saltator Uhler X O. celerinictus T Walker X Neoxabea sp. X X N. bipunctata (DeGeer) X Tettigoniidae: Conocephalinae X X Conocephalus sp X X C. brevipennis (Scudder)*

PAGE 50

36 Table 2-5. Continued Prey Present study Bohart & Menke 1976 Krombein 1970 Krombein 1967 C. fasciatus (DeGeer)* C. saltans (Scudder)* Odontoxiphidium sp* O. apterum Morse X Orchelimum sp. X X Tettigoniidae: Copiphorinae X Belocephalus sp.* Neoconocephalus sp. X Tettigoniidae: Phaneropterinae X X Scudderia sp. X X Scudderia furcata Brunner* Tettigoniidae: Tettigoniinae X X Atlanticus sp. X X A. gibbosus Scudder X Neobarrettia sp.* Not found but present in Isodontia mexicana nests

PAGE 51

37 CHAPTER 3 SPIDER PREY IN NESTS OF THE MUD DAUBER WASP Trypoxylon lactitarse (HYMENOPTERA: SPHECIDAE) Abstract Prey from 88 nests of Trypoxylon lactitarse in five state parks in north central Florida were examined, yielding 1173 individu al spiders from 15 families, 40 genera and 64 species. Overall, Neoscona sp. was the most commonly collected prey species (23.78%), followed by Mimetus sp. (12.27%) and Pisaurida mira (8.86%). Araneidae (56.26%) was the most commonly collected family, followed by Mimetidae (12.27%) and Pisauridae (10.57%). Trypoxylon lactitarse tended to be a generalist in its prey preference with a fairly even diversity of prey captured. Although the majority of prey items were common web-spinners many rarely surveyed spiders, such as Aniphedids and some Salticids, were collected. Since T. lactitarse hunts for spiders in wide-ranging microhabitats and with more intensity than hum an collectors, surveying nest contents is an extremely useful tool to expand spider sp ecies richness estimates, species inventories, and natural history data. Introduction Trypoxylon lactitarse is a solitary wasp found in the western hemisphere from Canada to Argentina. Females of T. lactitarse nest in cavities constructing linear cells subdivided by partitions of mud and provisi on these cells with numerous paralyzed spiders. Data concerning prey are norma lly difficult to obtain, but this wasp deposits prey in nests that are easily collected (Camillo and Breviscovit 1998, Rehnberg 1987).

PAGE 52

38 The wasp nests in preexisting cavities and readily accepts trap-nests, allowing for prey data to be easily collected. Previous studies (Rau 1928, Kr ombein and Evans 1954, Krombein 1956, 1967, Medler 1965, Lin 1969, Coville 1979, 1981, 1982, Coville and Coville 1980, Genaro et. al. 1989, Camillo et. al. 1993, Genaro and Alayon 1994, Jimenez and Tejas 1994, Camillo and Bresc ovit 1999, 2000) have shown that different species of Trypoxylon have different prey preferences These differences in prey preference can be in proportion of each fa mily, genus, or species taken; amount of families taken; and relative proportion of spider groups (orb-weaving, hunting or wandering) in the prey. Coville (1987) suggest ed that preferences for different species or species groups may arise because of different hunting behaviors of the wasps, different microhabitats hunted, or the wasps are conditi oned to a certain type of spider. Some species capture spiders predominately from one family and occasionally spiders from other families (Camillo and Brescovit 2000), while some species, including T. lactitarse prey on spiders of many different fam ilies (Camillo and Brescovit 1999). Nests of Trypoxylon lactitarse provide large amounts of spiders from various families, including spiders rarely caught by humans. Because these wasps hunt extensively in different microhabitats and ar eas rarely sampled by humans, sampling their nests and prey may provide additional information on spiders in the area. I set out to investigate th e following questions about Trypoxylon lactitarse and its nest contents to determine: 1) What prey is T. lactitarse provisioning at these Florida sites? 2) Is T. lactitarse a generalist or specialist in terms of prey provisioned in Florida? 3) What, if any, is T. lactitarse s prey preference? 4) Does T. lactitarse s prey preference seem to differ between sites? 5) What are th e benefits and problems with using this wasp

PAGE 53

39 (and potentially other spider-p rovisioning trap-nesters) as a sampling tool for estimating spider abundance and species richness? Methods and Materials Tools and Trap Preparation The traps used in this study were fa bricated from seasoned 37-mm x 86-mm x 2.4m pine/spruce timbers obtained from a local home improvement store. The pine/spruce timbers were cut into 100, 10-cm-long blocks. Two cavities of one of five diameters (3.2, 4.8, 6.4, 7.9 or 12.7-mm) were drilled into e ach block. Cavities were drilled to a depth of 80 mm on each short side (the 37-mm side), offset approximately 10-mm from the center point. Traps were assembled us ing one block of each diameter with the smallest cavity on top and the largest on the bottom. Blocks were stacked so that no cavity was situated directly above or belo w a cavity in the adjacent block. The five blocks were bound together with strapping tape (3M St Paul, Minnesota), and 16-gauge wire was used to further bind the stack and su spend the trap from trees and shrubs at the field sites. Each bundle of fi ve blocks was considered to be a single trap. Field Sites I set trap nests at five lo cations: 1) Suwannee River State Park in Suwannee County (30 23.149 N, 083 10.108 W), 2) Mike Roess Gold Head Branch State Park in Clay County (29 50.845 N, 081 57.688 W), 3) Devils Millhopper Geological State Park in Alachua County (29 42.314 N, 08223.692 W), 4) San Felasco Hammock Preserve State Park (29 42.860 N, 08227.656 W) in Alachua County and 5) Silver River State Park in Marion County (29 12.317 N, 082 01.128 W). The habitats surveyed at Suwannee River State Park were burned and unburned sa nd hill habitat, while the habitat at Mike Roess Gold Head Branch State Park was burne d sand hill pineland a nd ravine. Sites at

PAGE 54

40 San Felasco Hammock Preserve State Park consisted of upland and mesic hardwood hammock. Surveyed areas of Devils Millhopper Geological State Park consisted of pine flatwood habitat and sites at S ilver River State Park consiste d of river habitat and upland mesic forest. Descriptions of these habitats can be found in Franz and Hall (1991). Field Placement Transects were set up with 10 traps placed approximately 10 m apart and hung approximately 1.5 m off the ground on trees or limbs with placement on dead standing wood preferred. Transects were initially es tablished (direction a nd distance from center of plot) randomly. Four tran sects were established in Suwannee River State Park while three transects were established in Mike Ro ess Gold Head Branch State Park. Three transects were established in San Felasco St ate Park but size constraints only allowed a single transect in Devils M illhopper State Park. Finally, two transects were set up in Silver River State Park. Tr ansects were in the field fr om April 2003 until January 2005. Field Collection and Laboratory Rearing Traps remained in the field two years and were checked monthly. Preliminary field tests revealed that one-month intervals were sufficient to avoid trap saturation (no available cavities). Traps were considered o ccupied when insects we re observed actively nesting, harboring or had sealed a cavity w ith mud or plant material. Occupied traps were removed and replaced with a new trap. These occupied traps were brought into the forest entomology lab at the University of Florida in Gainesville, FL, for processing. Occupied blocks were removed for obse rvation while unoccupied blocks were reincorporated into replacement traps. Each occupied cavity was given a unique reference number.

PAGE 55

41 In order to examine prey items of Trypoxylon lactitarse, traps were dissected for nest contents when a wasp was encountered provisioning, guarding, or sealing a nest during collection runs. Prey items were removed, preserved, and given the same reference number as the cavity from which th ey had been removed. After the contents were extracted, the wood block was reused in replacement traps. These processed blocks were re-drilled to the next cavity diameter to eliminate any alterations or markings (either physical or chemical) by the previ ous occupant prior to reuse. Specimen Identifications All specimens were identified by the author with most of the spider prey specimens identified and verified by G. B. Edwards at the Florida State Coll ection of Arthropods in Gainesville, Fl. Voucher specimens have been deposited at the Florida State Collection of Arthropods. Statistical Analysis Similarity was calculated with Jaccards similarity index (ISj) (Southwood 1978). This index is the proportion of the combined set of species pres ent at either site that are present in both sites. This value ranges from 0 to 1, with 0 meaning no similarity (no species in common) in both sites and 1 meani ng all species are present at both sites. The value is calculated using the following equation: ISJ= c / (a + b + c) Where c is the number of species common to both sites and a and b respectively are the species exclusive to those sites Similarity was also calculated with Sorensens similarity index (ISs) (Sorensen 1948). This index is the propor tion of the combined set of species present at both sites that are present in both sites. This value ra nges from 0 to 1, with 0 meaning no similarity

PAGE 56

42 (no species in common) in both sites and 1 m eaning all species are present at both sites. The value is calculated using the following equation: ISs = 2c / (a + b) Where c is the number of species common to both sites and a and b are respectively the total number of species at each site ` Chao-Jaccard raw (uncorrected for unseen species) abundance-based similarity index, Chao-Jaccard estimate (corrected fo r unseen species) abundance-based similarity index, Chao-Sorensen raw (uncorrected for unseen species) abundance-based similarity, and Chao-Sorensen estimate (corrected fo r unseen species) abundance-based similarity (Chao et al 2005) was calculated with EstimateS 7.5 (Colwell 2005). Diversity was calculated using Simpsons index of diversity and Simpsons index of dominance (Simpson 1949). Simpsons index of diversity values range from 1 to S, where S is the total number of species. Simpsons index of dominance ranges from 0-1. Simpsons index of dominance, is given by: = s i 1(n / N)2 where n is the total number of organisms of the ith species and N is the total number of organisms of all species. Simpsons Index of Diversity is given by: 1/ Diversity was also calculated using th e Shannon-index (Shannon and Weaver 1949) H given by: H = s i 1pi ln pi

PAGE 57

43 where pi = n/ N and n is the total number of orga nisms of a particular species and N is the total number of organisms of all species Diversity is a combination of species ri chness (number of speci es) and evenness of species abundance. Therefore, Shannons i ndex of evenness, J (Pielou 1966), is given by: J = H / ln s where s is the total number of species Species richness was estimated using rarefaction curves (Colwell et al. 2004). This estimate of species richness is based on a sub-sample of pooled species actually discovered. In addition, three non-paramet ric species richness estimators, ACE (Abundance-based Coverage Estimator: Chao et al. 2000, Chazdon et al. 1998), first order jackknife (Burnham and Overton 1978, 1979, Smith and van Belle 1984, Heltshe and Forrester 1983) and Chao 1(Chao 1984) were used. These estimators produce estimates of total species richness including sp ecies not present in any sample. Most of the indices and all of the richness estimators were computed using EstimateS 7.5 (Colwell, 2005). Results I examined 88 nests of Trypoxylon lactitarse and found 1173 individual spiders from 15 families, 40 genera, and 64 species from all five state parks (Table 3-1). Overall, Neoscona sp. was the most commonly collected species (23.78%), followed by Mimetus sp. (12.27%) and Pisaurida mira (8.86%). See Table 3-1 for a summary of captured species tabulated by site of capture. When subdivided by site, (Figures 3-2 through 3-6) the top five species of spider prey for each si te was similar to overall pooled results. The most abundant species, Neoscona sp., was the most abundant sp ecies for three of the sites and second and third most abundant for the rema ining two sites. For all sites, the five

PAGE 58

44 most abundant species were included in the ten most abundant species for pooled data, except for 2 species at Gold Head Branch. At the Gold Head Branch site, the most abundant species, Trachelas sp., is third most abundant species for pooled data, Theridion sp., and fifth most abundant species, Neoscona crucifera, were not included in the pooled top ten most abundant species captured. The most abundant families of spider s collected were Araneidae (56.26%), Mimetidae (12.27 %), Pisauridae (10.57%), Salticidae (6.82%) and Tetragnathidae (6.82%) (Figure 3-7). Prey diversity as reported by Shannons and Simpsons indices is similar among sites except for the Simpsons index value fo r Devils Millhopper which is remarkably different at 6.81 (Table 3-3). The Shannon ev enness index for all but one site is above 0.7, (Table 3-3) suggesting a fair degree of ev enness. The evenness of the prey diversity is illustrated on the rank proportional abundance graphs (Figures 3-8, 3-9). Similarity of prey between sites, repo rted by Jaccards similarity, Sorensens similarity, Chao-Jaccard raw (uncorrected for unseen species) abundance-based similarity, Chao-Sorensen raw (uncorrect ed for unseen species) abundance-based similarity, Chao-Jaccard estimate (correct ed for unseen species) abundance-based similarity, Chao-Sorensen estimate (corre cted for unseen species) abundance-based similarity (Chao et al 2005) indexes are summarized in Table 3-3. Overall, the classic formulas for Jaccards and Sorensens indices gave the lowest values with Jaccards index being the lower of the pair. This is intuitive since these indices are calculated with actual observed species. All th e Jaccards indices, includi ng the Chao versions, were more conservative by yielding lower values than their Sorensen counterparts. The only

PAGE 59

45 exception to this trend was a single site co mparison, Silver River vs. San Felasco were both estimate versions of Chao-Jaccards and Chao-Sorensens gave a value of 1.0 for complete similarity. Species richness estimates given by ra refaction, first order jackknife, ACE (abundance-based coverage estimator) and Chao 1 estimator are given in Figures 3-10. The only site at which all these estimators stabilized, however, was at San Felasco State Park and the Chao 1 and ACE estimators stabilized for Gold Head Branch. The other sites and estimators did not completely r each an asymptote and can be viewed with skepticism (figure 3-11). Discussion Prey diversity varied among sites due to a large difference in species composition; however, the Shannon evenness index for all bu t one site is above 0.7, suggesting a fair amount of evenness. This trend can also be seen on the rank proportional abundance graphs (figures 3-1, 3-2). This level of evenness suggests that Trypoxylon lactitarse is not specializing on a few prey species with occasional secondary species, but rather behaving as a generalist and hunting a wide va riety of available spider prey including both web spinning and hunting spiders. The somewhat low levels of similarity for Jaccards and Sorensons (Table 3-6) indices between all sites suggest a distinct variation in spider prey composition. The Chao-Jaccard estimate abundance-based and the Chao-Sorensen estimate abundancebased similarity indexes show a higher degree of similarity between sites than their raw estimate counterparts. These estimate-based indexes are corrected for under-sampling bias and suggest that sites are more similar that the current observ ations reveal. Since under-sampling or limited sampling effort is th e generally the case, the Chao-Jaccard and

PAGE 60

46 Chao-Sorensen estimates would be the best choice. Of these two, the Chao-Jaccard estimate is generally the more conservative yielding (slightly) lower estimates of similarity. The highest estimated similarity values were between Silver River state park and San Felasco state park and the lowest similarity was between Devils Millhopper state park and Gold Head Branch state park, yet the estimators for these sites did not stabilize suggesting these sites were under-sam pled. These findings are expected since the similarity of the respective habitats coinci des with prey item similarity. This further suggests that Trypoxylon lactitarse is a generalist pred ator capturing prey that is abundant in the habitat and not searching for a particular species within any habi tat. Yet, since the sample sizes for each of the sites were di fferent, due to opportunistic nature of the sampling, an additional study with a more systematic, even sampling focusing on obtaining nest contents is needed to provi de more confident results. Furthermore, estimators for Devils Millhopper and Mike Ro ess Gold Head Branch did not stabilize due to small sample sizes, so these results should be viewed w ith skepticism. The estimators for the remaining sites did stabilize (except for the Chao 1 estimator in Silver River) and we can be confident in these species richness estimations. Finally, do Trypoxylon lactitarse nest provisions provide useful data on spider populations? The characteristics of the nest provisions confirm that T. lactitarse is a generalist predator of spiders, which is idea l for surveying a populat ion. Intuitively, these wasps collect spiders much more intensivel y and efficiently than human collectors. These nests also yielded a fair amount of rare species (uniques, singletons, and doubletons) further suggesting a complete survey of the target group. These three factors suggest that nests of T. lactitarse are an ideal survey tool for spiders. In addition, species

PAGE 61

47 richness estimators (such as in the freewar e EstimateS (Colwell, 2005)) extrapolate total species richness for a site and, therefore, s uggest a sufficient level of species inventory for a particular site. Sufficient sampling, however, is crucial for successful estimation. In most sites, the estimators did not stabi lize due to undersampling (figure 3-11). The estimators for San Felasco site did stabilize and two estimators stabilized at Gold Head Branch suggesting that sufficient samples were taken to estimate species richness for Trypoxylon lactitarse provisioned prey with confidence at those sites (figure 3-11). The fact that estimators did not stabilize for Silv er River and half of the estimators for Gold Head Branch did not stabilize is not surprising because of the smaller sample sizes. What is surprising, however, is that none of the estimators stabilized for Suwannee River. Although similar numbers of samples (contents of a single nest) were taken in the two parks, all estimators for San Felasco st abilized between 10-20 samples while no estimators stabilized for Suwann ee River after 28 samples. I suggest that each survey effort monitor estimators for stabilization for each site individually in order to determine sufficient sampling. These findings in no way suggest that this sampling represents the total spider fauna of the particular sites, but simply that we have sufficiently examined the prey of Trypoxylon lactitarse These estimators are indeed pract ical to determine the richness of spiders preyed upon by the wasps and when this sampling has been sufficient. As discussed earlier, T. lactitarse is a generalist in Florid a and in tropical regions, and provides a wide range of spider prey. Trap-nes ts also have the advant ages of other spider provisioning wasps being trapped. Trypoxylon johnsonii, T. carinatum, T. collinum collinum, T. clavatum johanis, and T. clavatum clavatum are other spider provisioning

PAGE 62

48 wasps that were also captured at these Florida field sites, but their ne st contents were not extracted. Even though some wasps may be sp ecialists (Camillo and Brescovit 2000) in addition to generalists such as T. lactitarse (Camillo and Brescovit 1999), these wasps intuitively search longer and in different microhabitats and, therefore, provide more abundance and possibly variety of spiders than hand collecting alone. Yet, some wasps do have preferences for one family or another. Even the generalist hunters may periodically favor one group of spiders that are locally abundant or more easily captured at that time over groups they would normally pr ey upon. It may be prudent, therefore, to take samples at various times of the year to avoid temporal population cycles of spiders. This technique for sampling spider fauna would be ineffective alone, however, as a part of a structured inventor y protocol including ot her techniques, such as hand collecting and pitfall traps, may provide more complete and accurate cataloguing of spider faunas. This is especially true since underestimat es have been shown to most commonly be derived from shortcomings of sampling tech niques rather than sampling effort (Longino and Colwell 1997, King and Porter 2005). Wh en various techniques are integrated together to create a structured inventory procedure, such as the Ants of the Leaf Litter (ALL) protocol for sampling ant communities (Agosti et al. 2000) and the methodology proposed by Coddington et al (1991) for spiders, they can be extremely powerful and reliable tools (Toti et al. 2000). Various techniques such as Malaise traps (Jennings and Hilburn 1988) and trap nests can be used in addition to the st andard hand collecting, sweeping, and pitfall trappi ng, to provide an efficient and complete method of determining spider fauna of an area when long term sampling is an option.

PAGE 63

49 Acknowledgements I thank G. B. Edwards of the Florida Stat e Collection of Arthropods in Gainesville, Florida, for the voluminous amount of identific ation, verification, and he lp with all of the spider specimens. All resear ch and collection were completed with permission of the Florida Department of Environmental Protec tion Division of Parks and Recreation under permit numbers 11250310 and 08170410

PAGE 64

50 Ten most abundant spider prey species0 50 100 150 200 250 300Neoscona sp. Mimetus sp. Pisaurina mira Eustala sp. Mecynogea lemniscata Nephila clavipes Thiodina sylvana Wagneriana tauricornis Eustala anastera Araneus sp.12345678910 Abundance0 5 10 15 20 25Percent Abundance Percent Figure 3-1. Ten most abundant spider prey species for all sites pooled Suwannee River State Park0 10 20 30 40 50 60 70 Mimetus sp.Mecynogea lemniscata Neoscona sp.Nephila clavipesThiodina sylvana 12345 Abundane0 2 4 6 8 10 12 14 16 18 20Percent Abundance Percent Figure 3-2. Five most abundant spider prey species at Suwannee River State Park

PAGE 65

51 San Felasco State Park0 20 40 60 80 100 120 Neoscona sp.Pisaurina miraMimetus sp.Wagneriana tauricornis Eustala anastera 12345 Abundance0 5 10 15 20 25Percent Abundance Percent Figure 3-3. Five most abundant spider pr ey species at San Felasco State Park Silver River State Park0 10 20 30 40 50 60 70 80 90 100 Neoscona sp.Eustala anasteraMimetus sp.Eustala anasteraLeucauge venuste 12345 Abundance0 10 20 30 40 50 60Percent Abundance Percent Figure 3-4. Five most abundant spider prey species at Silver River State Park

PAGE 66

52 Goldhead Branch State Park0 5 10 15 20 25 30 Trachelas sp.Neoscona sp.Theridion sp.Thiodina sylvanaNeoscona crucifera 12345 Abundance0 5 10 15 20 25 30 35Percent Abundance Percent Figure 3-5. Five most abundant spider prey species at Gold Head Branch State Park Devil's Millhopper State Park0 5 10 15 20 25 30 35 40 Neoscona sp.Pisaurina miraWagneriana tauricornis Eustala anasteraMecynogea lemniscata 12345 Abundance0 5 10 15 20 25 30 35 40 45Percent Figure 3-6. Five most abundant spider prey species at Devils Millhopper State Park

PAGE 67

53 Abundance and percentage of spider prey families captured at all sites0 100 200 300 400 500 600 700A r a n ei d a e Mimetidae P isau r idae S a l tic i da e Tetragnathidae C o r in n id a e A n y phae n idae Theri d idae Philo d rom id a e T h o m i si d ae Aniphedidae A g e leni d aeAbundance0 10 20 30 40 50 60Percent Number of individuals Percent Figure 3-7. Abundance and percentage of spid er prey families captured at all sites Figure 3-8 Pooled rank proporti onal abundance of spider spec ies collected from five Florida state parks. Pooled Rank Proportional Abundance0 0.05 0.1 0.15 0.2 0.25 110100 Species rankProportional abundance Neoscona sp. Mimetus sp. Pisaurina mira Eustala sp.

PAGE 68

54 Site Rank Proportional Abundance 0 0.1 0.2 0.3 0.4 0.5 0.6 110100Species rankProportional abundance DM SR SW SF GH Figure 3-9. Site rank proportional abundance of spider species collected at each state park. DM = Devils Millhopper State Pa rk, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park 1st order Jackknife species richness estimator0 10 20 30 40 50 60 70 80 90 DMSRSFGHSWNumber of Species ACE (Abundance-based Coverage Estimator)0 10 20 30 40 50 60 70 80 DMSRSFGHSWNumber of species Chao 1 Species richness estimator0 10 20 30 40 50 60 70 DMSRSFGHSWNumber of Species Sobs (Sample based rarefaction) 0 5 10 15 20 25 30 35 40 45 50 DMSRSFGHSW SiteNumber of Species Figure 3-10. Species richness estimation for sp ider prey tabulated by site: DM = Devils Millhopper State Park, SF = San Felasc o State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park Neoscona sp.

PAGE 69

55 Chao 1 0 10 20 30 40 50 60 70 80 90 100 05101520253035 Cumulative number of samplesnumber of species sf sw gh sr ACE0 10 20 30 40 50 60 70 80 05101520253035 Cumulative samplesnumber of species SF sw gh sr Sample based rarefaction0 5 10 15 20 25 30 35 40 45 50 05101520253035 cumulative samplesnumber ofspecies SF SW GH SR Jack 10 10 20 30 40 50 60 70 80 05101520253035 cumulative number of samplesnumber ofspecies sf SW gh sr Figure 3-11. Species richness estimator perfor mance for spider prey tabulated by site: DM = Devils Millhopper State Park SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Pa rk, SR = Silver River State Park, SW = Suwannee River State Park

PAGE 70

56 Table 3-1 Spiders found as prey in nests of Trypoxylon lactitarse in north central Florida Family Genus Species DM SR SF GH SW Sum Agelenidae Agelenopsis sp. 1 0 0 0 2 3 Aniphedidae sp. 0 0 1 1 3 5 Anyphaenidae Hibana sp. 0 1 1 0 1 3 Anyphaenidae Hibana velox 0 0 10 0 3 13 Anyphaenidae Lupettiana mordax 2 0 1 0 1 4 Araneidae Acacesia hamata 1 0 2 0 5 8 Araneidae Araneus bicentenareus 0 1 11 0 1 13 Araneidae Araneus juniperii 0 0 3 0 15 18 Araneidae Araneus miniatus 1 0 1 1 9 12 Araneidae Araneus pegnia 1 1 2 1 8 13 Araneidae Araneus sp. 1 0 18 1 14 34 Araneidae Argiope aurantia 0 0 0 0 1 1 Araneidae Argiope sp. 0 0 0 0 1 1 Araneidae Eriophora ravilla 0 0 4 0 0 4 Araneidae Eustala anastera 0 5 25 0 5 35 Araneidae Eustala sp. 5 48 29 0 9 83 Araneidae Kaira alba 0 0 1 0 0 1 Araneidae Larina directa 1 0 0 0 1 2 Araneidae Mecynogea lemniscata 6 0 15 0 52 71 Araneidae Metapeira sp. 0 0 0 0 6 6 Araneidae Metazygia zilloides 0 1 0 0 0 1 Araneidae Metepeira labyrinthea 0 0 7 0 0 7 Araneidae Neoscona arabesca 0 0 6 0 3 9 Araneidae Neoscona crucifera 3 0 0 2 0 5 Araneidae Neoscona sp. 37 87 100 15 41 279 Araneidae Ocrepeira sp. 0 1 9 0 1 11 Araneidae Scoloderus sp. 0 1 0 0 0 1 Araneidae Wagneriana tauricornis 7 2 28 0 8 45 Araneida Parauixia sp. 0 0 1 0 1 2 Clubionidae Elaver excepta 0 0 1 1 0 2 Corinnidae Trachelas similes 0 0 1 1 0 2 Corinnidae Trachelas sp. 0 0 0 24 1 25 Mimetidae Mimetus sp. 7 17 58 0 62 144 Philodromidae Philodromus sp. 1 0 1 0 1 1 1 Philodromidae Philodromus sp. 2 0 2 0 0 1 3 Philodromidae Philodromus sp. 3 0 0 0 0 1 3 Pisauridae Dolomedes albineus 0 3 1 2 3 9 Pisauridae Dolomedes sp. 0 1 4 0 2 7 Pisauridae Pisaurina mira 17 0 76 0 11 104 Pisauridae Pisaurina sp. 0 0 0 0 2 2

PAGE 71

57 Table 3-1 Continued. Family Genus Species DM SR SF GH SW Sum Salticidae Hentzia mitrata 0 0 0 0 2 0 Salticidae Lyssomanes viridis 0 0 0 7 13 15 Salticidae Metacyrba floridana 0 0 0 0 1 1 Salticidae Phidippus pulcherrimus 0 0 2 0 0 2 Salticidae Phidippus regius 0 0 0 1 0 1 Salticidae Platycryptus undatus 0 0 0 1 2 3 Salticidae Thiodina sp. 0 0 1 0 0 1 Salticidae Thiodina sylvana 2 1 24 7 20 54 Salticidae Zygoballus sexpunctatus 0 0 0 0 3 3 Segestriidae Ariadna bicolor 0 0 1 1 0 2 Tetragnathidae Leucauge venusta 0 4 11 0 2 2 Tetragnathidae Leucauge sp. 0 1 1 0 0 17 Tetragnathidae Nephila clavipes 2 0 24 0 37 61 Theridiidae Argyrodes sp. 0 0 0 0 1 1 Theridiidae Theridion sp. 0 0 7 11 0 11 Thomisidae Misumenops oblongus 0 0 0 0 2 11 Thomisidae Misumenops sp. 0 0 0 0 1 2 Thomisidae Synema parvula 0 0 0 0 2 2 Thomisidae Tmarus sp. 0 0 0 24 1 1 Argyia giaparatia 0 0 0 0 1 1 Total 94 178 467 73 361 1173 DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park

PAGE 72

58 Table 3-2. Similarity indexes and comparisons for spider prey Site Observed shared species Jaccard (Classic) ChaoJaccard raw abundancebased ChaoJaccard estimate abundancebased Sorensen Classic ChaoSorensen raw abundancebased ChaoSorensenestimate abundancebased DM vs. SR 6 0.214 0.577 0.605 0.353 0.731 0.754 DM vs. SF 13 0.351 0.749 0.789 0.52 0.856 0.882 DM vs. GH 6 0.222 0.264 0.31 0.364 0.417 0.474 DM vs. SW 15 0.326 0.756 0.882 0.492 0.861 0.938 SR vs. SF 14 0.368 0.62 1.0 0.538 0.766 1.0 SR vs. GH 5 0.167 0.269 0.367 0.286 0.424 0.537 SR vs. SW 15 0.313 0.454 0.621 0.476 0.624 0.766 SF vs. GH 10 0.244 0.223 0.418 0.392 0.364 0.59 SF vs. SW 25 0.463 0.847 0.953 0.633 0.917 0.976 GH vs. SW 11 0.216 0.47 0.47 0.355 0.45 0.64 DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park

PAGE 73

59 Table 3-3. Summary of diversity values for prey items tabulated by site Statistic DM SR SF GH SW Individuals 94 178 467 73 361 Simpson index of diversity 6.81 9.4 9.69 10.13 10.25 Shannon index of diversity 2.21 2.65 2.78 2.87 2.91 Shannon evenness 0.735 0.594 0.755 0.735 0.772 DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park

PAGE 74

60 CHAPTER 4 EFFECTS OF PRESCRIBED FIRE ON BIODIVERSITY AND SPECIES RICHNESS OF CAVITY NESTING HYMENOPTERA IN SUWANNEE RIVER STATE PARK, FLORIDA Abstract I examined the effect prescribed fire management had on the biodiversity and species richness of populations of trap-nesting Hymenoptera a nd associated arthropods in Florida. Four sandhill pine habitat sites (two burned sites and two unburned sites) at Suwannee River State Park were examined over a two-year period. For trap-nesting Hymenoptera, overall, species richness was different between treatment sites, and diversity was significantly different (p < 0.05) between burned and unburned sites. Overall diversity was not significantly diffe rent over time. Both unburned and burned sites showed similarity in species com position, which was especially high when an abundance-based estimate of similarity wa s used. When functional groups of trapnesting Hymenoptera were analyzed (preda tors, parasitoids a nd pollen collectors), pollinators and parasitoids were not significantly different between burned and burned sites. Predators were more abundant (p = 0.10) in unburned habitat. Of the six most abundant species examined, Trypoxylon lactitarse (Hymenoptera: Sphecidae) was significantly more abundant on unburned sites (p < 0.05), while Isodontia spp. ( I. auripes and I. mexicana, Hymenoptera: Sphecida e) were significantly more abundant on burned sites (p < 0.05). Xylocopa virginica (Hymenoptera: Anthophor idae) had significantly higher abundance on burned ha bitat than unburned habitat (p = 0.10). Chrysididae spp., Megachilidae spp. and Monobia quadridens (Hymenoptera: Vespidae) were not

PAGE 75

61 significantly different, in terms of abundan ce, between burned and unburned sites. Overall, prescribed fire employed by the park service to maintain natural sandhill pine habitat has some impact on trap-nesting Hyme noptera and associated arthropods in terms species richness and diversity. Although dive rsity and species richness changes were determined, the use of trap-nesting Hymenopt era to detect community changes from small-scale fires such as prescribed fire on th eir own may not be an appropriate choice to detect community changes owing to the substa ntial flight ranges of these insects. Introduction Fire is an integral part of forest and grassland ecosystems throughout the United States. Urban sprawl has increased the wildland-urban interface, causing increased concerns about wildfire. Prescribed fire is a useful tool to re duce the intensity of wildfires and has been shown to be effectiv e in reducing hazardous fuels, disposing of logging debris, preparing sites for seedi ng or planting, improving wildlife habitat, managing competing vegetation, controlling in sects and disease, improving forage for grazing, enhancing appearance, improving access and in perpetuating fire-dependent species (Biswell 1999, Cumm ing 1964, DellaSala and Frost 2001, Fuller 1991, Helms 1979, Long et. al. 2005, Mutch 1994, Wade et. al 1988). Prescribed fire is a proven, frequently used tool that works well in many aspects of wildland management. Many natural areas, including state parks, utilize pr escribed fire to restore and preserve native habitat and plant communities in addition to reducing the risk of uncontrolled wildfire (Siemann et al. 1997, Daubenmire 1968, Hurlbert 1965). Many studies have been conducted to examine the effect of fire on animal and plant communities, with some examining arthropods. Fire not only causes direct mortality in arthropods (Fay and Samenson 1993, Bolton and Peck 1946, Miller 1978, Evans 1984), but also indirectly

PAGE 76

62 affects arthropod communities via changes in plant community composition and habitat alteration (Lawton 1983, Evans 1984). Howeve r, most invertebrate studies focus on terrestrial arthropods monitored via sweep ing or pitfall traps (Bess 2002, Brand 2002, Niwa 2002, Clayton 2002, Fay 2003, and Koponen 2005) and overlook aeri al insects. I examined the effect of prescribed fire on the community of tr ap-nesting Hymenoptera and associated arthropods. Trap-nesting hym enopterans are a diverse group of insects that include various functional groups and interspecies intera ctions. This diverse group includes predators, pollen speci alists, and parasitoids. Fi re may affect the various subgroups differently depending on the alteration of resources. I investigated the effect of fire on the biodiversity of these insects by using traps in Florida state parks that have regularly and recently use prescribed fire. I set out to investigate the following questions about these trap-nesting insects and associat ed arthropods: 1) Do overall diversity and species richness differ between burned and unbur ned sites? 2) Do overall diversity and species richness differ between the sampled y ears within burned sites? 3) In terms of species sampled, how similar are the bur ned and unburned sites? 4) Are sampled functional groups, in terms of abundance (pre dator, parasitoid and pollen specialists) affected by fire? 5) Which of the most a bundant species, if any, are negatively or positively affected by fire in terms of abundance? Methods and Materials Tools and Trap Preparation. The traps used in this study were fabr icated from seasoned 37-mm x 86-mm x 2.4m pine/spruce timbers obtained from a local home improvement store. The pine/spruce timbers were cut into 100 10-cm-long blocks. Two cavities of one of five diameters (3.2, 4.8, 6.4, 7.9 or 12.7-mm) were drilled into each bloc k. Cavities were drilled to a depth of

PAGE 77

63 80 mm on each short side (the 37-mm side), offset approximately 10-mm from the center point. Traps were assembled using one block of each diameter with the smallest cavity on top and the largest on the bottom. Blocks were stacked so that no cavity was situated directly above or below a cavity in the ad jacent block. The fi ve blocks were bound together with strapping tape (3M St Paul, MN), and 16-gauge wire was used to further bind the stack and suspend the trap from trees and shrubs at the field sites. Each bundle of five blocks was considered to be a single trap Field Sites. I set four trap lines at two locations in Suwannee River State Park in Suwannee County (30 23.149 N, 083 10.108 W). The habitats su rveyed were burned and unburned sand hill habitat. Descriptions of th is habitat can be found in Franz and Hall (1991). Sites that were burned within the cu rrent year were consid ered burned sites and sites that had significant understo ry growth resulting from at least three years free of fire were considered unburned. These unburned site s tended to have a th ick understory and were slated for prescribed burn if possible in the next couple of seasons. Four subsites, two for each burn treatment, were established in the park. These subsites were designated burned 1, burned 2, unburned 1, and unburned 2. Neither of the unburned subsites was burned allowing for a two-year period of observation. Field Placement Transects were set up with ten traps pl aced approximately 10 m apart and were hung approximately 1.5 m off the ground on trees or limbs with placement on dead standing wood preferred. Transects were in itially established (d irection and distance from center of plot) randomly. Four transe cts were established in Suwannee River state park. Two transects were established in re cently burned habitat (with smoldering still

PAGE 78

64 ongoing) and two additional transects were es tablished in habitat that had not been burned in at least 3 years and had a significant understory of mixed br oad leaf with some palmetto. Transects were in the field from April 2003 until January 2005. Field Collection and Laboratory Rearing Traps remained in the field for two year s and were checked monthly. Preliminary field tests revealed that one-month intervals were sufficient to avoi d trap saturation (no available cavities). Traps were considered o ccupied when insects we re observed actively nesting, harboring or had sealed a cavity w ith mud or plant material. Occupied traps were removed and replaced with a new trap. These occupied traps were brought into the forest entomology lab at the University of Florida (Gainesville, FL) for processing. Occupied blocks were removed for obse rvation while unoccupied blocks were reincorporated into replacement traps. Each occupied cavity was given a unique reference number. Location, date of collection, diameter of cavity, and various notes describing the nature of the occupants and/or plug were r ecorded for each reference number. Occupied cavities were then covered with a 2, 4, 6, or 8dram glass shell vial. The shell vials were attached to the wood section with maski ng tape (Duck, Henkel Consumer Adhesive Inc., Akron OH) appropriate for wood applicati on. These sections were then placed in a rearing room and observed daily for emergen ce. The rearing room was maintained as nearly as possible at outside mean temp eratures for Gainesville, Florida. When emergence occurred, the specimens were removed, preserved, and given the same reference number as the cavity from which they had emerged. Dates of emergence, identification of occupants, measurements, and notes were taken for each cavity at emergence. Remaining nest fragments and de bris were kept for further analysis when

PAGE 79

65 possible. When an insect was harboring or actively tending a nest, it was captured, identified, and given a reference number corre sponding to the cavity. The contents of the nest/cavity were then extracte d and recorded. After the cont ents were extracted, the wood block was reused in replacement traps. These processed blocks were re-drilled to eliminate any alterations or markings (eit her physical or chemical) by the previous occupant prior to reuse. Specimen Identifications All specimens were identified by the author with some specimens identified and or verified by entomologists Jim Wiley1, Lionel Stange1, G. B. Edwards1, John B. Heppner1 and John M. Leavengood Jr. 1,2 (Florida State Co llection of Arthropods1, Gainesville, FL and University of Florida2, Gainesville, FL) Voucher spec imens have been deposited at the Florida State Collection of Arth ropods in Gainesville, Florida. Statistics and Calculations I examined the difference between sites that were recently burned (treatment) and those that had not been burned (non-treatment). Sites were examined for differences in abundance of functional groups and abundance of the six most abundant species with Analysis of Variance (ANOVA) using a Gene ralized Linear Mixe d Model (GLMM) and ANOVA using a Linear Mixed Model (LMM) wa s used to examine for differences in species diversity index values. All models were computed by Ge orge Papageorgiou of the IFAS statistical help lab at the Univers ity of Florida, Gainesville, Fl, using the GLIMMIX procedure and SAS system for mixed models (SAS Institute, Inc., Cary NC) In addition, similarity of species presence in different sites and diversity statistics were determined to further compare bur ned and unburned sites. Similarity was calculated with Jaccards similarity index (ISj) (Southwood 1978). This index is the

PAGE 80

66 proportion of the combined set of species presen t at either site that are present in both sites. This value ranges from 0 to 1, with 0 meaning no similarity (no species in common) in both sites and 1 meaning all species are pres ent at both sites. The value is calculated using the following equation: ISJ= c / (a + b + c) where c is the number of species common to both sites and a and b respectively are the species exclusive to those sites. Similarity was also calculated with Sorensens similarity index (ISs) (Sorensen 1948). This index is the propor tion of the combined set of species present at both sites that are present in both sites. This value ra nges from 0 to 1, with 0 meaning no similarity (no species in common) in both sites and 1 m eaning all species are present at both sites. The value is calculated using the following equation: ISs = 2c / (a + b) where c is the number of species common to both sites and a and b are respectively the total number of species at each site. Similarity was also calculated for esti mated population (in order to correct for under sampling bias) using Chao-Jaccard a bundance based estimate similarity index (Chao et al 2005) Diversity was calculated using Simpsons index of Diversity and Simpsons index of dominance (Simpson 1949). Simpsons index of diversity value ranges from 1 to S, where S is the total number of species. Simpsons index of dominance ranges from 0-1. Simpsons index of dominance, is given by: = s i 1(n / N)2

PAGE 81

67 where n is the total number of organisms of the ith species and N is the total number of organisms of all species. Simpsons Index of Diversity is given by: 1/ Diversity was also calculated using th e Shannon index (Shannon and Weaver 1949) H given by: H = s i 1(n / N) ln (n / N); where n is the total number of organisms of a particular species and N is the total number of organisms of all species. Diversity is a combination of species ri chness (number of speci es) and evenness of species abundance. Therefore, Shannons i ndex of evenness, J (Pielou 1966), is given by: J = H / ln s, where s is the total number of species Species richness was estimated using rarefaction curves (Colwell et al. 2004). This estimate of species richness is based on a sub-sample of pooled species actually discovered. In addition, two non-parametric specie s richness estimators, ACE (Abundance based Coverage Estimator: Chao et al. 2000, Chazdon et al. 1998) and Chao 1(Chao 1984) were used. These estimators produce esti mates of total species richness including species not present in any samp le. Most of the indices and all of the richness estimators were computed using EstimateS 7.5 (Colwell, 2005)

PAGE 82

68 Results Field sites During the two years of trapping, I collect ed 471 nests (34 pillaged by ants) in burned areas and 700 nests (62 pillaged by an ts) in unburned areas. These nests yielded 53 species from 25 families and 8 orders. These results are compiled in Table 4.1. Subsites and sampling month Rarefaction curves of observed species richness were produced for treatment subsite (burned 1 and burned2; unburned1 and unburned2) in order to detect a possible site effect. The resulting curves reached an as ymptote and had over-lapping 95% confidence intervals showing no significant difference, in terms of specie richness, between treatment sub-sites. The identification of no significant sub-site bias allowed for subsites to be pooled for furthe r analysis. In addition, the statistical model used, mixed linear model, factors out possible subsite temporal and positional effects. In terms of sampling months, all but tw o analyses did not have a significant sampling month effect (Mixed linear mode l: P > 0.10). Two functional groups, predators and parasitoids, had a significant sampli ng month effect (Mixed linear model: Ppredator = 0.004, Pparasitoid = 0.02). Effect of burning on species richness Actual observed species richness was 38 sp ecies in burned habitat and 44 species in unburned habitat. Rarefaction curves of observed species richness were produced for burned and unburned sites. The rarefaction cu rve estimated 35 species in burned habitat and 46 species in unburned habitat. The rarefa ction curve, however, did not completely attain an asymptote and should be viewed w ith skepticism, especia lly since the estimate for burned habitat richness is lower than the observed richness. The Abundance based

PAGE 83

69 Coverage Estimator (ACE) did not completely stabilize and its estimate of 46.84 species in burned habitat and 62.04 species in unburne d habitat should also be viewed with skepticism. The Chao 1 estimate of species richness did stabilize and yielded estimates of 47.5 species in burned habitat and 68.5 species in unburned habitat. Effect of burning on diversity Overall, the values for the Simpson index of diversity were 1 3.0 in burned habitat and 3.38 in unburned habitat. Simpsons inde x of diversity values were significantly different between burned and unburned ha bitats (LMM: F = 5.13, df = 13, P = 0.041) with burned sites having a higher index of diversity. There was no significant month effect (LMM: F = 1.21, df = 13, P = 0.3655) Shannons index of diversity showed burne d sites were somewhat more diverse than unburned sites (unburned = 2.17, burned = 2.87). Evenness in unburned sites was less even than burned sites (Shannon evenness: unburned = 0.57, burned = 0.79) The rank proportional abundance curve (Figur e 4-1) also shows that the species abundance in unburned plots was less even. Similarity of burned and unburned Similarity, measured by Sorensens index yielded a value of 0.682. Jaccards index yielded a value of 0.5181. In addition, Chao -Jaccard estimate similarity index, which provides similarity values based on estimat ed populations to correct for under-sampling, gave a value of 0.928. Functional groups In terms of abundance, the predator group was significantly different (at p = 0.10) between burned and unburned sites, with hi gher abundance in unburned sites (GLMM: F = 3.76, P = 0.0745) and has a significant mont h effect (GLMM: F = 4.78, P < 0.001).

PAGE 84

70 Both pollinators and parasitoid groups did not differ between treatments (GLMM: pollinator: F = 0.95, df = 13, P = 0.3472; parasitoid F = 0.13, df = 13, P= 0.7277). Pollinators did not have a significant mont h effect (GLMM: F = 0.45, df = 13, P = 0.91), but parasitoids did have a significant m onth effect (GLMM: F = 3.13, df = 13, P = 0.02). Most abundant species/ species groups Of the six most abundant species, Trypoxylon lactitarse was significantly more abundant in unburned habitat and Isodontia spp. were significantly more abundant in burned habitat (Table 4-1). Xylocopa virginica was significantly more abundant in burned habitat at p = 0.10 (GLMM: T. lactitarse F = 12.85, df = 14, P < 0.01; Isodontia sp.: F = 11.18, df = 14, P < 0.01; X. virginica: F = 3.84, df = 28, P = 0.07). Monobia quadridens, Megachilidae species and Chrysidi dae species were not significantly different between burned and unburned sites (GLMM: M. quadridens: F = 1.50, df = 14, P > 0.10; Megachilidae sp: F = 0.40, df = 14, P > 0.10, Chrysididae: F = 0.89, df = 14, P > 0.10). None of the top six most abundant species showed a month effect (GLMM: P > 0.10) Discussion The major justifications for using prescribed fire in state parks and natural areas are prevention of uncontrollable wildfire and ma intenance/ restoration of native/ natural habitat. The Florida state park system uses informational displays and signs to emphasize the importance of fire to maintain various native habitats such as sandhill pine and rockland pine habitats. Many consider biodiversity to be an important indicator of environmental health (Magurran 1988). In terms of species richness (both obse rved and estimated), unburned sites had higher values than burn sites. The only es timator that stabilized, Chao 1, estimated

PAGE 85

71 higher species richness in unburned sites than bu rned sites. For this group, especially when lower sampling effort cannot be avoided the Chao1 estimator would seem to be the best choice as it stabilized earlier and we can therefore have confidence in the estimate. The Chao 1 estimator, however, assumes homogeneity and should not be used to compare site with large co mpositional differences. The similarity indices were high, es pecially the abundan ce based Chao-Jaccard estimate index which had many sub-sites as completely similar. Of the similarity indices used, Jaccards index consisten tly gave the lowest, most conservative, estimate. Choice of similarity index used should depend on severa l things. First sampling effort is a main concern, especially in areas that have a hi gh level of dominance and rare species are frequently overlooked. In cases of small sampling effort or undersampling it would be prudent to use the Chao-Jaccard estimate similarity index in order to correct for this bias. Secondly, the level of identifica tion is important and can skew similarity values. As in this case, there were some groups that ar e notoriously difficult to identify, even by authorities and this may influence the inde x value if they are pooled into a morphospecies or species group. In such cases the more conservative inde x, Jaccards index and Chao-Jaccard estimate similarity index shoul d be used so that similarity is not overestimated. Depending only on species richness values, however, can be misleading, especially when investigating an event (such as fire in Florida) that the native fauna have evolved with and to which have possibly ad apted. Such events would unlikely cause localized extinction (which would change sp ecies richness) but rather alter relative abundances and dominance of species that have adapted to fire in varying degrees (which

PAGE 86

72 is detected by diversity). Ch anges in species richness may have applications in situations where the event is one that has not evolved with the fauna such as exotic species introduction or anthropogenic disturbance. As seen here, the diversity of these insects was significantly different between burn treatm ent sites. In this situation, the relative abundance of Trypoxylon lactitarse, the overall most abundant sp ecies, was significantly less abundant in burned site s. In unburned sites T. lactitarse is much more dominant than in burned sites (Figure 4-1). Trypoxylon lactitarse is so dominant in unburned areas that it lowers the evenness values and therefor e overall diversity, even though unburned sites had higher values of species richness. Although the burned sites had lower species richness than unburned sites, diversity was hi gher because of greater evenness values. Low or lowered evenness values can be seen as a sign of disturbance. This suggests that the unburned condition is actually the disturbed state for th is habitat. This makes intuitive sense since the fauna have evolved w ith the yearly and regular intervals of fires that are suppressed by the park se rvice. In essence, the distur bance is the removal of fire from the ecosystem by man. In addition, any difference in sampling years may be the resu lt of natural habitat succession after a fire event. Most habitats have succession periods that span years and continually change over decades (Siemann et al. 1997, Swengel 2001). Therefore, change in diversity may be a result of su ccession and not a difference in sample years (e.g. especially cold winter, dr ought, etc.). This idea is fu rther supported since there was no temporal effect detected for the majority of the analyzed groups over the two-year sampling period and longer intervals are needed to detect the faunal response.

PAGE 87

73 Focusing on functional groups, it was surprising to not see a difference in abundance for the pollinator group. Bee commu nities tend to respond positively, after the initial catastrophic mortality, post-burn in re sponse to increased floral resources (Potts et al. 2003). Pollen group species should see an increase, minimally a difference in sampled years of the burned sites, but did not differ in abundance. Most likely, the scale of these small prescribed fires and the mobility of these insects eliminated any impact fire may have had. The difference in predator and parasitoid diversity follow previous observations of reductions in abundance and resulting diversity in burned areas. These groups depend on abundance of prey items and these prey items tend to have varying response to fire. Bock and Bock (1991) showed that although all gr asshoppers were affected by burning, certain groups suffered higher mortality and populations took longer to regenerate. Some prey groups may be better adapted to fire and these populations rebound quicker than other prey groups (Dunwiddie1991). Parasitoids and predator s are dependant on prey population and their varying ability to respond to fire and this unequal return to previous abundance pattern will inherently affect the abundance patterns of predators. This should be especially true for spider-hunting wa sps and parasitoids of predators. Hymenoptera, especially bees and wasps, ar e generally strong fliers with flight ranges that can span kilometers and prescribed fire tend to be restrict ed to variously sized sections that are commonly 10 hectares or le ss. Newly burned areas are easily accessible by these insects from the su rrounding unburned habitat. Th ese prescribed fires do an excellent job of removing dead standing and felled wood, but rarely eliminate all such material (pers. obs.), especially in controll ed burns that tend to be less intense than

PAGE 88

74 uncontrolled wildfire. Uncontro lled, intense wildfire may re move most dead timber but tend to kill younger trees and, in the cases of the fire crowning, may kill mature trees resulting in possible increased nesting sites. In the end, th ese insects use dead wood as nesting sites and are not completely deprived of this resource within burned sites. Furthermore, the preferred nesting sites and required materials for some species, such as resin and grasses, increase in abundance and availability in response to burning. The nature of prescribed fire does not a llow for controlled experimentation, and burn site establishment is dependent on weat her and park management. In order to minimize a possible site bias, the sites chosen were initially (pre-burn) identical in terms of flora and habitat. In add ition, repeated measures were used for statistical analysis to further minimize the possibility of a site bi as confounding the results. Even though I am confident that the measures taken to reduce in fluence from a site bias were adequate, the possibility of sit effects cannot be completely disregarded. Conclusion Overall species richness and diversit y did differ between burned and unburned sites, and sites were not diffe rent between sampling dates, indicating that burning affects trap-nesting hymenopterans a nd associated arthropods fr om burning treatment. None of the three functional groups (pollin ators, predators, and parasitoids) were affected by the burn treatment. Of the six mo st abundant species captured, only two were significantly different (p < 0.05) between bur ned and unburned habitat, with one more abundant in burned habitat and the other in unburned habitat. Even though diversity and sp ecies richness changes were determined, the use of trap-nesting Hymenoptera on their own may not be an appropriate choice to detect community changes from small-scale fires such as prescribed fire. These insects are

PAGE 89

75 volant with substantial flight range and this flight range may allow these insects to respond to a source-sink of nesting materials (res in, grasses, and cavitie s), yet still forage beyond the scope of the relatively small scale of the prescribed fire and perhaps eliminate any effect the treatment may have had on thes e groups. In the case of wildfire, where the scale is usually exponentially larger, the spec ies richness and divers ity of these insects may be more indicative of th e community as a whole. Monitoring diversity and abundance of trap-nesting hymenopterans in unburned sandhill pine habitat, however, may be an appropriate application to monitor the community. When the community becomes less even with few species, such as T. lactitarse dominating the proportional abundance, this may be an indicator of disturbance and that a burn is needed to maintain the desired sandhill pine habitat. Acknowledgements All research and collection were comp leted with permission of the Florida Department of Environmental Protection Di vision of Parks and Recreation under permit numbers 11250310 and 08170410.

PAGE 90

76 Rank proportional Abundance0 0.1 0.2 0.3 0.4 0.5 0.6 110100 Species RankProprtional Abundance Burned site Unburned site Figure 4-1. Rank proportional abundance of species in burned and unburned sandhill pine habitats Trypoxylon lactitarse

PAGE 91

77 Table 4-1. Species trapped in bur ned and unburned sandhill pine habitat Number of nests in habitat Order Family Genus Species Functional group Burned Unburned Araneida Salticidae Platycryptus undatus Predator 3 2 Segestriidae Ariadna bicolor Predator 2 4 Clubionidae Elaver excepta Predator 2 5 Blatteria Blattaria sp B 2 4 Coleoptera Carabidae Cymindus platycollis Parasitoid 0 2 Tenebrionidae 0 1 Cleridae Cymatodera Parasitoid 0 1 Cleridae Lecontella brunnea Parasitoid 1 0 Cleridae Nemognatha Parasitoid 0 1 Elateridae 0 1 Diptera Bombyliidae Anthrax analis Parasitoid 9 22 Bombyliidae Anthrax aterrimus Parasitoid 6 25 Bombyliidae Lepidophora lepidocera Parasitoid 4 2 Bombyliidae Toxophora amphitea Parasitoid 1 0 Conopidae Parasitoid 0 1 Orthoptera Gryllidae Orocharis luteolira 14 7 Hymenoptera Anthophoridae Xylocopa virginica Pollen 7 1 Formicidae Crematogaster Blk Predator 6 26 Formicidae Crematogaster Red Predator 11 2 Chrysididae Parasitoid 40 33 Ichneumonidae Parasitoid 1 1 Leucospidae Parasitoid 1 0 Megachilidae Dolicostelis louisa Parasitoid 1 0 Megachilidae Coelioxys sayi Parasitoid 2 1 Megachilidae Megachile campanulae Pollen 1 0 Megachilidae Megachile georgica Pollen 7 5 Megachilidae Megachile mendica Pollen 9 1 Megachilidae Megachile xylocopoides Pollen 0 3 Megachilidae Osmia sandhouseae Pollen 2 0 Mutillidae Sphaeropthalma pensylvanica Pollen 1 0 Pompilidae Ampulex canaliculata Predator 1 0 Pompilidae Dipogon graenicheri Predator 1 3 Sphecidae Isodontia auripes Predator 60 9 Sphecidae Isodontia mexicana Predator 28 6 Sphecidae Liris beata Predator 0 1 Sphecidae Podium rufipes Predator 5 15 Sphecidae Trypoxylon clavatum Predator 0 12 Sphecidae Trypoxylon collinum Predator 21 17 Sphecidae Trypoxylon clavatum johanis Predator 4 15 Sphecidae Trypoxylon carinatum Predator 0 1 Sphecidae Trypoxylon lactitarse Predator 56 339 Sphecidae Trypoxylon Red Predator 1 3 Sphecidae Trypoxylon Sm Predator 0 3 Vespidae Vespula maculifrons Predator 2 1 Vespidae Euodynerus megaera Predator 13 7 Vespidae Monobia quadridens Predator 47 33 Vespidae Stenodynerus Sp A Predator 53 7 Vespidae Sp C Predator 1 0 Lepidoptera Pyralidae Uresiphita reversalis 0 2

PAGE 92

78 Table 4-1 Continued. Species trapped in burned and unburned sandhill pine habitat Number of nests in habitat Order Family Genus Species Functional group Burned Unburned Noctuidae Cerma cerintha 0 1 Noctuidae sp B 0 4 Scorpionida Buthidae Centruroides hentzi Predator 12 0 Chilopoda Scolopendridae Hemiscolopendra punctiventris Predator 0 5

PAGE 93

79 CHAPTER 5 BIOLOGY, PREY AND NEST S OF THE POTTER-WASP Monobia quadridens L. (HYMENOPTERA: VESPIDAE) Abstract I observed the potter wasp, Monobia quadridens L, nesting in predrilled wooden trap-nests at five state park s in north central Florida. Wasps nested mostly in 12.7-mm diameter cavities (97 of 129 nests) and occasionally nested in 7.9-mm diameter cavities (26 of 129 nests). Females rarely nested in 6.4-mm (5 of 129) and 4.8-mm diameter cavities (1 of 129). All cavities were 80-mm deep. Females used mud to make provisioned cells, partitions, in tercalary cells, vestibular ce lls and a closure plug, yet did not line the inside of any cells with material. They construc ted nests with an average of 1.69 provisioned cells (range = 1-3, SD = 0.47) a vestibular ce ll, and from 0-3 intercalary cells. All nests were solely provisioned with para lyzed caterpillars of Macalla sp. ( thrysisalis or phaeobasalis ) (Lepidoptera: Pyralidae). Cells with female brood had a mean length of 24.01 mm (range = 20-30, SD = 3.59 N= 16) while cells that resulted in males had a mean length of 18.22 mm (range = 14-25, SD =2.68, N = 23). Intercalary cells were highly variable with a mean of 13.45 mm, (range = 5-35, SD = 5.69, N = 28) as were vestibular cells with a mean of 12.78 mm (range = 1-35, SD= 9.055, N = 23). Resulting sex ratio of emerging adults was 1.2 males per female. In conclusion, nest architecture of Monobia quadridens is variable, females tend to nest in cavities with diameters greater than 7.9 mm, and females preyed on a single species of the Pyralid caterpillar, Macalla sp. ( thrysisalis / phaeobasalis ). This apparent pr ey specialization is

PAGE 94

80 unique when compared to M. quadridens in other parts of its range and even to historical Florida data. Such a difference and possible shift of behavior within Florida warrants further study. Introduction Monobia quadridens is a common wasp in the eastern United States where it is reported from Massachusetts, Rhode Isla nd, Connecticut, New York, New Jersey, Pennsylvania, Delaware, Maryland, West Virgin ia, District of Columbia, Virginia, North Carolina, South Carolina, Alabama, Georgia, Florida, Mississippi, Louisiana, Arkansas, Kentucky, Tennessee, Texas, Oklahoma, Ka nsas, New Mexico, Missouri, Indiana, Illinois and Ohio (Bequaert 1940). Speci mens captured by Krombein (1967) at the Archbold Biological Station at Lake Placid, Florida (Highlands County) and specimens at the Florida State Collection of Arthropods in Gainesville, Florida i ndicate that it ranges throughout peninsular Florida, including the Fl orida Keys and Everglades National Park. This wasp normally nests in abandoned nests and burrows of other insects, such as carpenter bees that nest in cl ay banks and wood. In Florida, Monobia quadridens is the largest wasp that nested in traps. When female M. quadridens find an acceptable cavity, they line the back end of the cavity with mud and suspend a single egg fr om the top of the cavity. Females hunt and paralyze caterpillars to provi sion the nest. When prey popul ations are sufficient, they tend to provision the same species of cater pillars (Krombein 1967). Females are also known to have a leisurely nesting rate taking up to a week to complete a nest with three provisioned cells and a single intercalary cell (Krombein 1967). Adults emerge from the nest two days to two weeks after completion of the nest, but actual development time is about 10-14 days.

PAGE 95

81 Practically all r ecent research on Monobia quadridens has been on the biochemistry and physiology of organisms and proteins ex tracted from its hemolymph and little has been done to examine the ecology of the was p. The objectives for this study are to determine the preferred nesting cavity diamet er, determine what prey the females are provisioning, describe nest architecture, a nd determine emerging sex ratio of this wasp, Monobia quadridens, in north central Florida. Methods and Materials Tools and Trap Preparation The traps used in this study were fa bricated from seas oned 37 mm x 86 mm x 2.4m pine/spruce timbers obtained from a local home improvement store. The pine/spruce timbers were cut into 100 10-cm -long blocks. Two cavities of one of five diameters (3.2, 4.8, 6.4, 7.9, or 12.7-mm) were dril led into each block. Cavities were drilled to a depth of 80 mm on each short side (the 37-mm side), offset approximately 10mm from the center point. Traps were assemble d using one block of each diameter with the smallest cavity on top and the largest on the bottom. Blocks were stacked so that no cavity was situated directly above or belo w a cavity in the adjacent block. The five blocks were bound together with strapping tape (3M St Paul, MN), and 16-gauge wire was used to further bind the stack and suspend the trap from trees and shrubs at the field sites. Each bundle of five blocks was cons idered to be a single trap. Field Sites I set traps at five locations: 1) Suwann ee River State Park in Suwannee County (30 23.149 N, 083 10.108 W), 2) Mike Roess Gold Head Br anch State Park in Clay County (29 50.845 N, 081 57.688 W), 3) Devils Millhopper Geolog ical State Park in Alachua County (29 42.314 N, 08223.6924 W), 4) San Felasco Hammock Preserve State Park

PAGE 96

82 (29 42.860 N, 08227.656 W) in Alachua County and 5) Silver River State Park in Marion County (29 12.317 N, 082 01.128 W). The habitats surveyed at Suwannee River State Park were burned and unburned sa nd hill habitat, while the habitat at Mike Roess Gold Head Branch State Park was burne d sand hill pineland a nd ravine. Sites at San Felasco Hammock Preserve State Park consisted of upland and mesic hardwood hammock. Surveyed areas of Devils Millhopper Geological State Park consisted of pine flatwood habitat and sites at S ilver River State Park consiste d of river habitat and upland mesic forest. Descriptions of these habita ts can be found in Franz and Hall (1991). Field Placement Transects were set up with ten traps placed approximately 10 m apart and hung approximately 1.5 m off the ground on trees or limbs with placement on dead standing wood preferred. Transects were initially esta blished (direction and distance from center of plot) randomly. Four tran sects were established in Suwannee River State Park while three transects were established Mike Roe ss Gold Head Branch State Park. Three transects were established in San Felasco St ate Park but size constraints only allowed a single transect in Devils M illhopper State Park. Finally, two transects were set up in Silver River State Park. Tr ansects were in the field fr om April 2003 until January 2005. Field Collection and Laboratory Rearing Traps remained in the field two years and were checked monthly. Preliminary field tests revealed that one-month intervals were sufficient to avoid trap saturation (no available cavities). Traps were considered o ccupied when insects we re observed actively nesting, harboring or had sealed a cavity w ith mud or plant material. Occupied traps were removed and replaced with a new trap. These occupied traps were brought into the forest entomology lab at the University of Florida in Gainesville, Fl for processing.

PAGE 97

83 Occupied blocks were removed for obse rvation while unoccupied blocks were reincorporated into replacement traps. Each occupied cavity was given a unique reference number. Location, date of collection, diameter of cavity, and various notes describing the nature of the occupants and/or plug were r ecorded for each reference number. Occupied cavities were then covered with a 2, 4, 6, or 8dram glass shell vial. The shell vials were attached to the wood section with Duck (H enkel Consumer Adhesive Inc., Akron OH) masking tape appropriate for wood applicati on. These sections were then placed in a rearing room and observed daily for emergen ce. The rearing room was maintained as nearly as possible at outside mean temperatures for Gainesville, FL. When emergence occurred, the specimens were removed, preserved and given the same reference number as the cavity from which they had emerged. Dates of emergence, identification of occupants, measurements and notes were taken for each cavity at emergence. When a female M. quadridens was harboring or activ ely tending a nest, it was captured, identified, and given a refere nce number corresponding to the cavity. The contents of the nest/cavity were then extrac ted and recorded. Afte r the contents were extracted, the wood block was reused in replace ment traps. These processed blocks were re-drilled to the next size di ameter cavity to eliminate any alterations or markings (either physical or chemical) by the previous o ccupant prior to reuse. All adult Monobia quadridens that successfully emerged from cavities were curated and sexed. Identifications Monobia quadridens is readily distinguished from other vespid wasps. Specimen diagnostics are given in Appendix B and figures of both sexes of Monobia quadridens are given in Appendix A.

PAGE 98

84 All cavity nesters and their prey were iden tified by the author with some specimens identified or verified by entomologists Jim Wiley1, Lionel Stange1, John B Heppner1, John M. Leavengood Jr.1,2 (Florida State Collection of Ar thropods, Gainesville, FL and University of Florida2, Gainesville, FL). Voucher speci mens have been deposited at the Florida State Collection of Arthropods. Statistical Analysis Descriptive statistics (means, ranges, st andard deviation) we re calculated using Microsoft Excel statistical package. Chisquared goodness of fit was used to examine nest diameter preference. The assumption was that wasps would nest equally in all acceptable diameters. Results Nest Architecture Monobia quadridens nested in 129 cavities ove r the two-year period of observation, mostly nesting in 12.7-mm cav ities (97 of 129 nests) and occasionally nesting in 7.9-mm cavities (26 of 129 nests). Fe males rarely nested in 6.4 mm-cavities (5 of 129) and 4.8-mm cavities (1 of 129). Female s did not nest equally in all diameters (chi-squared contingency table, 2 = 184.51, df = 3, P< 0.001) when the data were pooled. Females did not nest equally in all diameters at indi vidual sites (P< 0.001: 2 Devils Millhopper = 23.25, 2 Goldhead = 134.56, 2 San Felasco = 65.33, 2 Suwannee River = 32.62, and P < .05: and 2 Silver River = 8.2) and during particular years (P< 0.001, 2 2003 =101.0, 2 2004 = 144.69). See table 5.1 for a su mmary of nesting results. Whenever females were observed plugging a cavity, the nests (n = 32) were later dissected to examine nest architecture. Fema les used mud to make the cell partitions and

PAGE 99

85 closing plug yet did not line the sides of cells with material. Nests averaged 1.69 provisioned cells (SD = 0.47, ra nge 1-3, N= 32), with vestib ular cells (an empty cell between the cavity opening and provisioned cells) and intercalary spaces (empty, unprovisioned spaces between provisioned cells and behind the vestibular cell). Cells with female brood had a mean length of 24.01 mm (SD= 3.59, range = 20-30, N= 16) while cells containing males had a mean length of 18.22 mm (SD = 2.68, range = 14-25, N = 23). Intercalary spaces were highly variable with a mean of 13.45 mm (SD = 5.69, range = 5-35, N = 28), as were vestibular cells with a mean of 12.78 mm (SD = 9.055, range = 1-35, N = 23). On average, nests ha d two provisioned cells usually separated by at least one intercalary space. Use of the in tercalary spaces varied; some nests had an intercalary space in front of each provis ioned cell while other nests had a single provisioned cell with multiple intercalary spaces leading to the vestibular cell. Nests had a mean of 1.28 (SD = 0.631,range = 0-3, N= 32) intercalary spaces per cavity. Sex Ratio I collected 129 nests of Monobia quadridens from traps. These nests yielded a sex ratio of 1.2 males per 1 female (N= 245). Prey Females suspended a single egg from the t op of each cell with a filament and would then search for prey. All prey items were the caterpillar, Macalla sp. ( thrysisalis/ phaeobasalis ) (Lepidoptera: Pyralidae). These two species cannot be distinguished without rearing out to adult st age, but this is impossible since caterpillars have been paralyzed by the wasp. Females woul d position these para lyzed caterpillars longitudinally with the head toward the re ar of the cavity. Developing larvae would

PAGE 100

86 leave caterpillar head capsules, allowing fo r prey confirmation after emergence. On average, there were 3.0, (SD = 0.67, range 2-5, N= 98) caterpi llars per cell. Discussion Nest Architecture Krombein (1967) reported that in some populations, including a central Florida population, females used agglutinated sand to cons truct partitions and plugs. None of the populations I observed used agglutinated sa nd although all trap lines were within 2 kilometers of a mud resource such as roads or bodies of water. Nests had a mean of 1.69 cells but in actuality nests usually had 2 pr ovisioned cells in 80-mm long cavities, with occasional nests with a single provisioned cell lowering the mean. Bequaert (1940) reported that the typical pattern fo r solitary eumenid wasps, including M. quadridens is a nest containing up to 12 cells in preexisting cavities. These cavities are usually burrows made by some other insect in clay banks or wood and do not usually have the confines of the traps used that had cavities only 80 -mm long. Krombein (1967) reported that M. quadridens primarily nested in 12.7-mm diameter cav ities with rare (2 of 78) occasions of nesting in 6.4-mm diameter cavities. He stat ed that most females are too large to enter a 6.4-mm cavity which restricts nesting activ ity to diameters larger than 6.4-mm. The Florida populations I observed we re similar in this respect. Monobia quadridens mostly nested in 12.7-mm cavities with some in 7.9mm cavities and rarely (5 of 129) in 6.4-mm cavities. I did trap an uncompleted ne st in a 4.8-mm cavity, however the nest was abandoned after a single egg was laid and tw o caterpillars were provisioned. The single cell was not sealed and did not develop, ye t since an egg was deposited and prey provisioned, the female most likely may have died and not necessarily absconded

PAGE 101

87 because the cavity was too small. The pa rtial provisioning and oviposition, however, indicates that smaller M. quadridens could nest in diameters as small as 4.8 mm. Cavity diameter did not seem to alter the si ze of male or female cells. In fact, one nest in a 7.9-mm cavity containe d a provisioned cell that was at the top end of the size range (30mm) and provisioned cell per nest range (3.0). These nests in 7.9-mm cavities tended to use fewer and smaller non-provisioned elements (vestibule cells and intercalary spaces) perhaps compensating for loss of volum e due to the smaller diameter. Of the nests dissected and examined for cell di mensions, only three nests were in 7.9-mm diameter cavities with the remaining nests being in 12.7-mm cavities. The other, rarely caught diameters (6.4-mm and 4.8-mm) cavities were not examined for cell dimensions. Sex Ratio Krombein (1967) observed a similar sex ratio of 1:1 (M: F, 65 males, 63 females) in a Florida population of M. quadridens He observed female bias ed sex ratios of 1: 1.8 (M: F, N= 50) in Maryland and 1:1.6 (M: F, N= 41) in North Carolina. I found a male biased sex ratio of 1.2:1 (M: F) more simila r to Krombeins (1967) Florida observation. Sex ratios tend to be skewed in relation to resources demands, which intuitively may suggest for the differences in these populati ons. Our Florida populati ons were similar to each other while slightly differing from th e more northern populations. These two different parts of the range intuitively have different habitats and prey, but Krombeins (1967) reported sex ratios are not drastically skewed to suggest re source strain on the population. Prey The observed prey items provisioned by Monobia quadridens were identical for all sites. Over the entire two-year peri od and the 98 nests were prey were verified,

PAGE 102

88 only Macalla sp. ( thrysisalis / phaeobasalis ) (Lepidoptera: Pyralidae) was provisioned. Krombein et al. (1979) reported Neophopteryx uvinella (Rag.), Neophopteryx sp., Phycitinae species, Epipaschia superatilis Clem., Epipaschia sp. Tetralopha asperatella (Clem.), Tetralopha sp., Epipaschiinae species, Desmia funeralis (Hbn.), Pyraustinae species, Stenoma schaegeri Zell., Stenoma sp., Stenomidae species., Psilocorsis sp., Gelechiidae species., Platynota sp., and Tortricidae species. In Krombeins (1967) Florida population (Lake Placid), he found 9 species of prey from 4 families extracted from 11 nests, in North Carolina he found 4 species from 3 families extracted from 3 nests and in Maryland he found 4 species from 3 families extracted from 7 nests. A ll of the species were included in the list from Krombein et al. (1979). Although our observed prey species, Macalla sp. ( thrysisalis / phaeobasalis ), was not previously reported, it is a member of the family (Pyralidae) previously reported as prey for M. quadridens. Macalla thrysisalis and M. phaeobasalis are the only Macalla sp. recorded from Florida (Kimball 1965). Macalla thrysisalis has only been recorded to feed on Mahogany and M. phaeobasalis on but complete host ranges have not been explored for these sp ecies or many other species of Lepidoptera (Pers. Comm. J.B. Heppner Florida Stat e Collection of Arthropods, Gainesville, Florida). The fact that my observed populations, however, only yielded a single species poses interesti ng questions that warrant further investigation. With relatively few samples Krombein (1967), obtaine d a fairly wide host range for M. quadridens spanning across families. He also stated that M. quadridens seem to concentrate on storing caterpillars of a singl e species when abundant, but th e trend he observed was in

PAGE 103

89 short time periods and even then some cells would contain a singl eton of another prey species. This activity would most likely be the result of a source sink, such as a population boom of a caterpillar species on a nearby host tree. Yet, my populations yielded only one species over the course of two years w ith 98 nests sampled during various times of the year. This strongly suggests that M. quadridens has been specializing on this prey speci es in these surveyed areas. Furthermore, has the Lake Placid, Florida population continued to be more of a generalist or has it become more of a specialist as its north Florida count erpart in the more than 30 years since Krombein surveyed them? Either way, su ch an event could provide interesting investigations into bioge ography and evolutionary hist ory for this species. Krombein (1967) also reported Lecontella cancellata (LeC.), a Cleridae species and Dermestidae species as predators, but I did not observe any predator activity. A chrysidid wasp and an ichneumonid wasp were found parasitizing M. quadridens. Many unsuccessful pupae of cycl oraphan flies, and occasiona lly the fly itself were commonly found in association with M. quadridens nests. Yet it is most likely that these flies emerged from the caterpillar s that had been oviposited on prior to provisioning in the nest sin ce the majority of larvae of M. quadridens successfully completed development. Furthermore, when M. quadridens successfully completed development the fly pupae did not complete development and when flies completed development adult M. quadridens were either slightly smaller or dead intact wasp larvae were found in the cell. These findi ng suggest that the maggots were competing for the provisioned caterpillar and not att acking the wasp larvae. There were no observations indicating that wasp larv ae consumed the maggots.

PAGE 104

90 Conclusion The surveyed populations of Monobia quadridens were similar to observations in other parts of its range it terms of nest architecture, cavity preference, and sex ratio. Prey of this wasp, however, was quite different wh en compared to other populations. Not only was the prey provisioned at these sites not previous ly reported, but M. quadridens appears to be specializing on this caterpillar, Macalla sp. ( thrysisalis / phaeobasalis ) when in other populations M. quadridens was a generalist prey ing on various species from various families of caterpillars. I strongly suggest further examination of this observation focusing on biogeography and evolut ionary history of this species. Acknowledgements All research and collections were comp leted with permission of the Florida Department of Environmental Protection Di vision of Parks and Recreation under permit numbers 11250310 and 08170410

PAGE 105

91 Table 5-1 Number nest diameters occupied by Monobia quadridens Diameter (mm) DM SR SF GH SW Total 3.2 0 0 0 0 0 0 4.8 0 0 0 0 1 1 6.4 1 0 1 1 2 5 7.9 4 0 2 6 14 26 12.7 7 4 20 39 27 97 DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silv er River State Park, SW = Suwannee River State Park

PAGE 106

92 CHAPTER 6 BIODIVERSITY OF TRAP-NESTING HYMENOPTERA OF FIVE NORTH FLORIDA STATE PARKS Abstract Maintaining biodiversity is one of multiple objectives of land managers of state parks and other natural areas. I surveyed th e species richness and biodiversity of trapnesting Hymenoptera using pre-drilled wooden tr ap-nests at five st ate parks in north central Florida. I found 85 species or species groups in total from all of the parks and provide estimates of over-all species richne ss based on this sampling for each park. Various values of biodiversity (Simpsons in dices of diversity and dominance, Shannons indices of diversity and eve nness) are reported for each park. Surveying of trap-nesting Hymenoptera and obtaining their biodiversity values by using the described traps are practical methods for land managers that requir e minimal resources from park staff. This allows for practical replication of the survey by detecting changes in biodiversity of these insects. The inventory of species identifie d may also expand park faunal records. Introduction Loss of biodiversity in natural and protect ed areas is a major concern for natural area resource managers (Kramer 2005). The main source of this loss is the impact resulting from human land-use, including agri culture, development, waterway diversion and habitat fragmentation. As human populatio ns grow and urban areas further encroach into rural areas, protected natu ral areas and parks formally isolated from such influences are increasingly becoming affected. With rapi d urbanization and resulting impacts, it is

PAGE 107

93 becoming more important to establish inventories of cu rrent flora and fauna as benchmarks for future comparisons. The objectives of this chap ter are to: 1) Report the a bundances and species richness of all trap-nesting hymenoptera and associated arthropods sampled at each of the five surveyed Florida State Parks 2) Determine, by using estimators, if the inventory offered can be considered adequate and, if adequate estimate total speci es richness of trapnesting hymenopterans and associated arthropods for each state park surveyed. Methods and Materials Tools and Trap Preparation The traps used in this study were fa bricated from seasoned 37-mm x 86-mm x 2.4m pine/spruce timbers obtained from a local home improvement store. The pine/spruce timbers were cut into 100 10-cm-long blocks. Two cavities of one of five diameters (3.2, 4.8, 6.4, 7.9 or 12.7-mm) were drilled into each bloc k. Cavities were drilled to a depth of 80 mm on each short side (the 37-mm side), offset approximately 10-mm from the center point. Traps were assembled using one block of each diameter with the smallest cavity on top and the largest on the bottom. Blocks were stacked so that no cavity was situated directly above or below a cavity in the ad jacent block. The fi ve blocks were bound together with strapping tape (3M St Paul, Minnesota), and 16-gauge wire was used to further bind the stack and suspend the trap from trees and shrubs at th e field sites. Each bundle of five blocks was considered to be a single trap. Field Sites I set trap nests at five lo cations: 1) Suwannee River State Park in Suwannee County (30 23.149 N, 083 10.108 W), 2) Mike Roess Gold Head Branch State Park in Clay County (29 50.845 N, 081 57.688 W), 3) Devils Millhopper Geological State Park in

PAGE 108

94 Alachua County (29 42.314 N, 08223.692 W), 4) San Felasco Hammock Preserve State Park (29 42.860 N, 08227.656 W) in Alachua County and 5) Silver River State Park in Marion County (29 12.317 N, 082 01.128 W). The habitats surveyed at Suwannee River State Park were burned and unburned sa nd hill habitat, while the habitat at Mike Roess Gold Head Branch State Park was burne d sand hill pineland a nd ravine. Sites at San Felasco Hammock Preserve State Park consisted of upland and mesic hardwood hammock. The area surveyed at Devils Mill hopper Geological State Park was of pine flatwood habitat and sites at S ilver River State Park consiste d of river habitat and upland mesic forest. Descriptions of these habitats can be found in Franz and Hall (1991). Field Placement Transects were set up with ten traps placed approximately 10 m apart and hung approximately 1.5 m off the ground on trees or limbs with placement on dead standing wood preferred. Transects were initially esta blished (direction and distance from center of plot) randomly. Four tran sects were established in Suwannee River State Park while three transects were established Mike Roe ss Gold head Branch State Park. Three transects were established in San Felasco St ate Park but size constraints only allowed a single transect in Devils M illhopper State Park. Finally, two transects were set up in Silver River State Park. Tr ansects were in the field fr om April 2003 until January 2005. Field Collection and Laboratory Rearing Traps remained in the field two years and were checked monthly. Preliminary field tests revealed that one-month intervals were sufficient to avoid trap saturation (no available cavities). Traps were considered o ccupied when insects we re observed actively nesting, harboring or had sealed a cavity w ith mud or plant material. Occupied traps were removed and replaced with a new trap. These occupied traps were brought into the

PAGE 109

95 forest entomology lab at the University of Florida in Gainesville, Fl, for processing. Occupied blocks were removed for obse rvation while unoccupied blocks were reincorporated into replacement traps. Each occupied cavity was given a unique reference number. Location, date of collection, diameter of cavity, and various notes describing the nature of the occupants and/or plug were r ecorded for each reference number. Occupied cavities were then covered with a 2, 4, 6, or 8dram glass shell vial. The shell vials were attached to the wood section with Duck (H enkel Consumer Adhesive Inc., Akron OH) masking tape appropriate for wood applicati on. These sections were then placed in a rearing room and observed daily for emergen ce. The rearing room was maintained as nearly as possible at outside mean temperatures for Gainesville, FL. When emergence occurred, the specimens were removed, preserved, and given the same reference number as the cavity from which they had emerged. Dates of emergence, identification of occupants, measurements, and notes were taken for each cavity at emergence. When an insect was harbori ng or actively tending a nest, it was captured, identified, and given a reference number corre sponding to the cavity. The contents of the nest/cavity were then extracted and recorde d. After the contents were extracted, the wood block was reused in replacement traps. These processed blocks were re-drilled to the next size diameter cavity to eliminate a ny alterations or marki ngs (either physical or chemical) by the previous occupant prior to reuse. Specimen Identifications All specimens were identified by the author with some specimens identified and or verified by entomologists Jim Wiley1, Lionel Stange1, G. B. Edwards1, John B. Heppner1 and John M. Leavengood Jr. 1,2 (Florida State Co llection of Arthropods1, Gainesville, FL

PAGE 110

96 and University of Florida2, Gainesville, FL). Voucher sp ecimens have been deposited at the Florida State Collection of Arth ropods in Gainesville, FL. Statistical Analysis Diversity was calculated using Simpsons index of diversity and Simpsons index of dominance (Simpson 1949). Simpsons index of diversity values ranges from 1 to S, where S is the total number of species. Simpsons index of dominance ranges from 0-1. Simpsons index of dominance, is given by: = s i 1(n / N)2 where n is the total number of organisms of the ith species and N is the total number of organisms of all species. Simpsons Index of Diversity is given by: 1/ Diversity was also calculated using th e Shannon-index (Shannon and Weaver 1949) H given by: H = s i 1(n / N) ln (n / N) where n is the total number of organisms of a particular species and N is the total numbe r of organisms of all species Diversity is a combination of species ri chness (number of speci es) and evenness of species abundance. Therefore, Shannons i ndex of evenness, J (Pielou 1966), is given by: J = H / ln s where s is the total number of species. Shannons diversity index and Simpsons diversity index can be produced in programs such as EstimateS 7.5 (Colwell, 2005). Species richness was estimated using rarefaction curves (Colwell et al. 2004). This estimate of species richness is based on a sub-sample of pooled species actually discovered. In addition, three non-paramet ric species richness estimators, ACE

PAGE 111

97 (Abundance based Coverage Estimator: Chao et al. 2000, Chazdon et al. 1998), first order jackknife (Burnham and Overton 1978, 1979, Smith and van Belle 1984, Heltshe and Forrester 1983) and Chao 1(Chao 1984) were used. These estimators produce estimates of total species richness including sp ecies not present in any sample. Most of the indices and all of the richness estimators were computed using EstimateS 7.5 (Colwell, 2005). Similarity was calculated with Jaccards similarity index (ISj) (Southwood 1978). This index is the proportion of the combined set of species pres ent at either site that are present in both sites. This value ranges from 0 to 1, with 0 meaning no similarity (no species in common) in both sites and 1 meani ng all species are present at both sites. The value is calculated using the following equation: ISJ= c / (a + b + c) Where c is the number of species common to both sites and a and b respectively are the species exclusive to those sites Similarity was also calculated with Sorensens similarity index (ISs) (Sorensen 1948). This index is the propor tion of the combined set of species present at both sites that are present in both sites. This value ra nges from 0 to 1, with 0 meaning no similarity (no species in common) in both sites and 1 m eaning all species are present at both sites. The value is calculated using the following equation: ISs = 2c / (a + b) where c is the number of species common to both sites and a and b are respectively the total number of species at each site `

PAGE 112

98 Chao-Jaccard raw (uncorrected for unseen species) abundance based similarity index, Chao-Jaccard estimate (corrected fo r unseen species) abundance-based similarity index, Chao-Sorensen raw (uncorrected for unseen species) abundance based similarity, and Chao-Sorensen estimate (corrected fo r unseen species) abundance-based similarity (Chao et al 2005) were calculated with Es timateS 7.5 (Colwell, 2005). Results Overall, I captured 85 species or specie s groups from 2953 nests captured at all of the sites surveyed. Broken down by park, I found 33 species at Devils Millhopper S.P., 39 species at Silver River S. P., 40 species at San Felasco S. P., 54 species at Gold Head Branch S. P., and 53 species at Suwannee Ri ver S. P. (Figure 6-1). I provide an inventory of these trap-nesting Hymenopter a and associated arthropods, tabulated by park, in Table 6-1. I calculated commonly used diversity st atistics (Species richness, Simpson diversity, Shannon diversity, Shannon evenness) and species richness estimators (first order jackknife, ACE and Chao 1) from the nests captured from each of the five state parks surveyed (Figures 6-1, 6-2, 6-3, 6-4) Species richness estimators stabilized, suggesting sufficient sampling occurred to conf idently estimate species richness for each of the sites. Species richness estimate s ranged from 41.8 to 109 species. The lowest estimator (sample based rarefaction) estimat ed 41.8 species at De vils Millhopper, 59.6 species Silver River, 70.9 species at San Fe lasco, 79 species at Mike Roess Gold Head Branch, and 85 species at Suwannee River. The higher (more conservative) estimates were 52.0 species at Devil s Millopper (ACE), 76.5 speci es at Silver River ( 1st order Jackknife), 92.2 species at San Felasco ( 1st order Jackknife), 102.6 species at Mike Roess

PAGE 113

99 Goldhead Branch ( 1st order Jackknife), and 109 at Suwannee River( 1st order Jackknife). See figure 6-4 for all estimates tabulated by site All estimators stabilized except for the Jack 1 and ACE estimators at San Felasco (Figure 6-5). Similarity of prey between sites, repo rted by Jaccards similarity, Sorensens similarity, Chao-Jaccard raw (uncorrected for unseen species) abundance based similarity, Chao-Sorensen raw (uncorrect ed for unseen species) abundance based similarity, Chao-Jaccard estimate (correct ed for unseen species) abundance-based similarity, Chao-Sorensen estimate (corre cted for unseen species) abundance-based similarity (Chao et al 2005) indexes are summarized in Table 6-2. Overall, the classic formulas for Jaccards and Sorensens indices gave the lowest values with Jaccards index being the lower of the pair. This is intuitive since these indices are calculated with actual observed species. All th e Jaccards indices, includi ng the Chao versions, were more conservative by yielding lower values than their Sorensen counterparts. The only exception to this trend was two site compar isons, Goldhead vs. San Felasco were both estimate versions of Chao-Jaccards and Chao-Sorensens gave a value of 1.0 for complete similarity and Devils Millhopper vs. San Felasco were the Jaccard estimate abundance-based was slightly higher (Table 6-2). Discussion The findings of this survey offer an i nventory of trap-nesting Hymenoptera and the corresponding diversity values and estimates of overall species richness for each state park surveyed. Although the faunal list ma y not be expansive with only 85 observed species overall and a range of 33-54 species for each park, species richness estimators have been used to extrapol ate total species richness fo r trap-nesting Hymenoptera of these sites. These extrapolating estimators provide a lower limit estimate for the amount

PAGE 114

100 of species in the sampled habitat, including uns een rare species. In essence, the estimated species richness value provided in Figure 61 is the minimum number of trap-nesting arthropods estimated to exist at that site. All of the estimators for each of the sites stabilized suggesting sampling was sufficient and we can have confidence in thes e estimates of species richness (Figure 6-5), with the exception of two estimators (Jack 1 and ACE) at San Felasco. These species richness estimates are usef ul when a complete taxonomic inventory is desired, especially when funding is an issu e and an end point is needed. These species richness values are also us eful and commonly used to ga uge environmental well-being and monitor for change in that well-bei ng, but relying only on this value can be misleading. Changes in species richness can indicate major changes in habitat wellbeing, but will not detect changes in comm unity structure. Change in community structure can be detrimental to the overall health and sustainability of a particular habitat. There are numerous examples where two areas can have the same number of species but completely different community structure. One general scenario is where an exotic species may enter and completely dominate an area. This same habitat may have the exact same (or even higher richness if no species was driven out) amount of species richness but the exotic species may account for 80% of all of the individuals. In terms of species richness, exotic species may decrease (inhibition), increase (facilitation) or not affect the overall value (equivalency/ co mpensation). Changes to the ecosystem, measured via species richness, would go unde tected if equivalency or compensation occurred in the monitored species assemblage (Sax et al. 2005). For an effective comparison or monitoring program it is essentia l to calculate diversity statistics (which

PAGE 115

101 factor in species densities) in addition to spec ies richness. In the pa st, diversity statistics were complex and difficult to calculate, but computer software (usually the same program used to estimate species richness) can easily calculate these values. San Felasco and Devils Millhopper are ge ographically separated by less than 5 miles and they share similar habitat, yet spec ies richness and diversity values for Devils Millhopper were all lower than San Felasco. In fact, Devils Millhopper consistently had the lowest species richness for all estimator s (figure 6-4) and lowest diversity value (figures 6-2, 6-3) when compared to a ll other sites. Alt hough protected, Devils Millhopper has become an island of habitat from encroaching urbanization. The rapid growth of the city of Gainesville in the past twenty years has been s ubstantial. Previously the park was far from the city and its effect s, but now the suburbs and their effects fully border the park. For about a third of the park s main hiking trail, houses are 20-50 meters away and one part of the tr ail curves to avoid a support cable for a neighboring radio tower. Intuitively, it would seem that encr oaching urbanization ma y have caused the low diversity and richness of Devils Millhopper, but further study is needed before a definitive statement may be made. Devils Millhopper is a unique system and therefore may be difficult to compare to other stat e parks. Future survey and comparison, therefore, may provide an insight to th e effects of urbanization for this site. Of the species richness estimators the Chao 1 estimator stabilized for all sites and gave the highest estimate for three of the five sites, while rarefaction (Sobs) always gave the lowest estimate. The Chao 1 estimator is a non-parametric estimator, which removes assumptions about the population, and Colw ell and Coddington (1994) argue that these are superior to parametric estimators (like rarefaction). Although Chao 1 stabilized well

PAGE 116

102 at all sites, it provides an estimate of minimum richness and some of the assumptions with the procedure make it inappropriate to co mpare values between areas that have large compositional differences. Without a complete faunal list (complete species richness) for comparison, however, the true performance of these estimators for this group of insects cannot be determined. Further study such as King and Porters (2005) evaluation of estimators for use in ant fauna is needed to fully examine performance of these estimators for cavity nesting Hymenoptera. The Chao-Jaccard estimate abundance ba sed and the Chao-Sorensen estimate abundance based similarity indexes show a highe r degree of similarity between sites than their raw estimate counterparts. These es timate-based indexes are corrected for undersampling bias and suggest that sites are more si milar that the current observations reveal. According to Chao (2005), since under-sam pling or limited sampling effort is the generally the case the Chao-Jaccard and Chao -Sorensen estimates would be the best choice especially were limited sampling is a concern and rare species make up a large proportion of the fauna. Of these two, the Ch ao-Jaccard estimates is generally the more conservative yielding (slightly) lower estimates of similarity with this data. The highest estimated similarity values were found for S ilver River vs. San Felasco and Gold Head vs. San Felasco the lowest similarity was between Silver River and Suwannee River For the both the Jaccard and Sorensens indices, Gold Head vs. Suwannee River and Devils Millhopper vs. San Felasco were the most similar while Silver River vs. Gold Head were the most dissimilar. Based on the overall a pparent habitat similar ity between sites the trend reported by these classic indices seems a ppropriate. The estimate similarity values were high for all site comparisons and this is expected since the fauna sampled via traps

PAGE 117

103 that target cavity nesters, yet some of the comparisons yielded higher values than expected by results form their classic counter parts and apparent habitat comparison. For some non-parametric species richness estimato rs, such as Chao 1, assumptions are made about the populations that make it use inappropriate. Some of the sites are very different in habitat and such a bias would seem a problem, but Chao (2005) makes no mention of such a restraint for this estimator. When sampling is sufficient as determined by species richness estimators, however, I prefer the clas sic indices especially when habitats vary greatly. The system used here may be a practi cal approach for conducting future faunal inventories of trap-nesting Hyme noptera. The simplicity of the trap system allows for practical replication. The tr aps themselves are of a simple design with all material available from a hardware or home improvement store. A small room or an outdoor area can be used to store traps while waiting fo r emergence. In ad dition, the manual skills needed to construct and maintain the traps are commonly required of park staff. The nature of the monthly sampling required adds a minimal amount of manual labor added to existing park workforce. Most species involved may be recognized easily with reference collections and some familiarity with insects. Species difficult to identify, such as species from the family Chrysididae, are entered as a group. Although genus and species level identification of the Chrysidida e are extremely difficult for experts, this family as a group is easily distinguish ed from other trap-nesting Hymenoptera by amateurs. In addition, the method of capturi ng insects emerging from traps allows for minimum knowledge in entomology when outside help is available. Vials that receive emerging specimens confine and protect the specimen until it can be collected. Almost

PAGE 118

104 all the insects reported in this survey could be left in the vials for long periods of time without degrading the specimen quality for id entification. Some nests that have many cells per nest, such as the Megachilid ae, may have many specimens destroyed by siblings, but there usually are a couple of specimens in an acceptable condition for identification. This attribute of the tra pping method allows for park staff to send specimens to universities or museums for iden tification by just adding alcohol to a vial. This would allow a park employee to complete ly carry out the field survey without any knowledge of entomology and never handling live insects. These attributes allow for the survey to be repeated without additional staff and at minimal cost to the institution. Such future comparisons may be excellent measurem ents of biodiversity and detecting change in these measurements. Such observations may be especially powerful when added to other low cost surveys such as bird watche r inventories. In addition, the normally complicated task of using species richness estimators has been simplified through availability of free comp uter software such as EstimateS (Colwell, 2005). Acknowledgements All research and collection were completed wi th permission of the Florida Department of Environmental Protection Division of Pa rks and Recreation under permit numbers 11250310 and 08170410

PAGE 119

105 Actual Observed species richness 0 10 20 30 40 50 60Number of species Actual Observed species richness 3329405453 DMSRSFGHSW Figure 6-1. Actual observed species richness, tabulated by site. DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park Shannon index of diversit y and Shannon evenness tabulated by site0 0.5 1 1.5 2 2.5 3 3.5 DMSRSFGHSWDiveristy values0 0.02 0.04 0.06 0.08 0.1 0.12 0.14Evenness value Shannon index of diversity Evenness value Figure 6-2. Shannon index of diversity and Shannon evenness values for trap nesting Hymenoptera and associated arthro pods at five state parks. DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park

PAGE 120

106 Simpson index of diversity 0 2 4 6 8 10 12 DMSRSFGHSWSimpson value Figure 6-3. Simpson index of diversity fo r trap nesting Hymenoptera and associated arthropods at five state parks. DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park 1st order Jackknife estimate of species richness0 10 20 30 40 50 60 70 80 90 100 110 120 DMSRSFGHSWNumber of species A CE (Abundance Based Coverage Estimate) estimate of species richness0 10 20 30 40 50 60 70 80 90 100 DMSRSFGHSWEstimate number of species Chao 1 species richness estimate0 10 20 30 40 50 60 70 80 90 100 DMSRSFGHSWEstimate number of species Rarefaction species richness estimation 0 10 20 30 40 50 60 70 80 90 DMSRSFGHSWNumber of species Figure 6-4. Species richness es timators tabulated by site. DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park

PAGE 121

107 Suwannee River0 10 20 30 40 50 60 70 020040060080010001200 Cumulative individualsNumber of species Sobs chao 1 Jack 1 ACE Devil's Millhopper State Park0 5 10 15 20 25 30 35 40 45 020406080100120140160180200 Cumulative individualsCumulative species Chao1 ACE jack1 Sobs Goldhead0 10 20 30 40 50 60 70 050100150200250300350 Cumulative individualsNumber of species Sobs Chao 1 Jack 1 ACE San Felasco0 10 20 30 40 50 60 01002003004005006007008009001000 Cumulative individualsNumber of species Sobs Chao 1 jack 1 ACE Silver River 0 5 10 15 20 25 30 35 40 45 050100150200250300350 Cumulative individualsNumber of species Chao 1 Sobs Jack 1 ACE Figure 6-5. Species richness esti mator performance per site

PAGE 122

108 Table 6-1. Summary of trap-nesting arth ropods captured in five state parks Number of nests at site Order Family Genus Species DM SR SF GH SW Araneae Salticidae Phillipus regians 0 1 0 0 0 Salticidae Platycryptus undatus Segestriidae Ariadna bicolor 0 0 6 1 2 Clubionidae Elaver excepta 10 5 20 16 2 Blatttodea Blattaria Eurycotis floridans 0 0 5 0 0 Blattaria Sp A 4 0 10 0 0 Blattaria Sp B 0 0 0 2 2 Blattaria Sp C 0 0 2 0 0 Coleoptera Carabidae Cymindus platycollis 0 0 0 0 2 Carabidae 0 0 0 1 0 Elateridae 0 0 0 1 1 Tenebrionidae Platydema flavipes 0 0 1 1 0 Tenebrionidae 0 0 5 3 1 Trogossitidae 0 0 0 1 0 Trogossitidae Airora cylindrica 0 0 1 0 0 Cleridae Cymatodera 0 0 0 1 1 Cleridae Lecontella brunnea 0 0 0 0 1 Meloidae Nemognatha punctulata 3 0 0 0 1 Elateridae 0 0 0 3 1 Rhipiphoridae Macrosiagon cruentum 1 0 0 0 0 Diptera Bombyliidae Anthrax analis 3 4 11 4 9 Bombyliidae Anthrax aterrimus 1 4 9 7 6 Bombyliidae Lepidophora lepidocera 1 2 5 9 4 Bombyliidae Toxophora amphitea 0 0 0 5 1 Tachinidae 0 0 0 1 0 Hemiptera Egg mass 0 2 0 0 0 Conopidae 0 0 0 1 1 Hymenoptera Anthophoridae Xylocopa virginica 0 0 1 10 7 Chrysididae 4 7 31 33 73 Leucospidae 0 0 0 1 1 Formicidae Crematogaster Blk 3 1 16 2 32 Formicidae Crematogaster Red 0 0 1 0 13 Formicidae Crematogaster Red/blk 0 0 0 3 0 Formicidae Crematogaster minutissima 0 0 5 0 0 Formicidae Camponotus Red 2 1 7 10 0 Formicidae Camponotus black 0 0 2 3 0 Formicidae Pseudomrymex 12 1 31 4 0 Ichneumonidae 1 0 1 2 2 Megachilidae Dolicostelis louisa 0 0 0 1 2 Megachilidae Coelioxys sayi 0 0 0 0 3 Megachilidae Coelioxys dolichos 0 0 0 2 0 Megachilidae Coelioxys texana 0 0 0 1 0 Megachilidae Coelioxys 1 1 0 1 0 Megachilidae Megachile campanulae 0 1 0 0 1 Megachilidae Megachile mendica 9 0 0 4 10 Megachilidae Heriades carinata 0 0 0 2 2 Megachilidae Megachile c. wilmingtoni 0 0 0 5 0 Megachilidae Megachile georgica 2 10 0 10 12 Megachilidae Megachile pruina 0 0 0 1 0 Megachilidae Megachile rubi 0 0 0 4 0

PAGE 123

109 Table 6-1, Continued. Summary of trap-nesting arthropods captured Number of nests at site Order Family Genus Species DM SR SF GH SW Megachilidae Megachile xylocopoides 5 2 2 2 3 Megachilidae Osmia sandhouseae 0 0 0 0 2 Mutillidae Sphaeropthalma pensylvanica 0 0 7 1 1 Pompilidae Ampulex canaliculata 0 0 0 0 1 Pompilidae Dipogon graenicheri 5 7 35 6 4 Sphecidae Isodontia auripes 6 0 1 19 62 Sphecidae Isodontia mexicana 2 0 1 55 35 Sphecidae Liris beata 0 0 0 0 1 Sphecidae Podium rufipes 3 0 8 7 20 Sphecidae Trypoxylon clavatum 1 0 0 0 12 Sphecidae Trypoxylon collinum 6 0 19 17 38 Sphecidae Trypoxylon clavatum. johanis 6 2 12 31 19 Sphecidae Trypoxylon carinatum 3 5 12 0 1 Sphecidae Trypoxylon johnsonii 0 0 13 1 0 Sphecidae Trypoxylon lactitarse 76 249 510 13 392 Sphecidae Trypoxylon Red 0 0 1 0 4 Sphecidae Trypoxylon Small 2 1 2 0 3 Vespidae Ancistocerus sp. 0 1 3 0 0 Vespidae Monobia quadridens 22 4 48 97 80 Vespidae Vespula maculifrons 0 0 0 0 3 Vespidae Euodynerus megaera 13 0 24 5 30 Vespidae Stenodynerus sp A 0 0 1 54 58 Vespidae Stenodynerus sp B 0 18 0 7 0 Vespidae sp C 2 0 0 0 1 Vespidae Pacnodynerus erynnis 2 1 0 28 0 Isoptera 0 0 0 2 0 Lepidoptera Pyralidae Uresiphita reversalis 0 0 0 0 4 Noctuidae Cerma cerintha 0 0 0 0 1 Noctuidae Litoprosus frutilis 0 2 0 0 0 Noctuidae sp B 1 4 Orthoptera Gryllidae Orocharis luteolira 3 1 6 7 21 Orthoptera Egg mass 1 2 Scorpiones Buthidae Centruroides hentzi 0 0 0 2 12 Scolopendromorpa Scolopendridae Hemiscolopendra punctiventris 0 2 0 0 5 Scolopendridae Scolopendra viridis 0 2 0 0 0 DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silv er River State Park, SW = Suwannee River State Park

PAGE 124

110 Table 6-2. Similarity indexes and comp arisons for trap-nesting Hymenoptera Site Observed shared species Jaccard (Classic) ChaoJaccard raw abundance based ChaoJaccard estimate abundance based Sorensen Classic ChaoSorensen raw abundancebased ChaoSorensenestimate abundancebased DM vs. SR 16 0.432 0.714 0.945 0.604 0.833 0.972 DM vs. SF 23 0.535 0.869 0.996 0.697 0.930 0.998 DM vs. GH 20 0.345 0.589 0.945 0.513 0.741 0.972 DM vs. SW 22 0.373 0.695 0.781 0.543 0.820 0.877 SR vs. SF 19 0.396 0.804 0.925 0.567 0.891 0.961 SR vs. GH 17 0.274 0.578 0.867 0.43 0.732 0.929 SR vs. SW 19 0.302 0.621 0.701 0.463 0.766 0.924 SF vs. GH 27 0.415 0.768 1.0 0.587 0.869 1.0 SF vs. SW 29 0.439 0.814 0.900 0.611 0.897 0.947 GH vs. SW 39 0.574 0.833 0.879 0.729 0.909 0.936 DM = Devils Millhopper State Park, SF = San Felasco State Park, GH = Mike Roess Gold Head Branch State Park, SR = Silver River State Park, SW = Suwannee River State Park

PAGE 125

111 APPENDIX A ADDITIONAL FIGURES AND SP ECIMEN PHOTO GUIDE Figure A-1. Traps Figure A-2. Rearing room

PAGE 126

112 The following figures are intended to aid in identification of species frequently captured in traps. Figures A-1 through A7 are reference photographs for diagnostic and identification guides provided in appendix B. The remaining figures are photographs of commonly captured species, provid ed strictly a reference or starting point. Identification of specimens should be executed using expe rts, diagnostic keys, monographs and other sources in the scientific literature. Figure A-3. Male Monobia quadridens

PAGE 127

113 Figure A-4. Female Monobia quadridens Figure A-5. Antenna of Monobia quadridens

PAGE 128

114 Figure A-6. Male Isodontia auripes Figure A-7. Female Isodontia auripes

PAGE 129

115 Figure A-8. Male Isodontia mexicana Figure A-9. Female Isodontia mexicana

PAGE 130

116 Figure A 10. Anthrax analis Figure A-11. Anthrax aterrimus

PAGE 131

117 Figure A-12. Lepidophora lepidocera Figure A-13. Toxophora amphitea

PAGE 132

118 Figure A-14. A wasp in the family Chrysididae Figure A-15. A series of Chrysidid wasps demonstrating variati on in size and color

PAGE 133

119 Figure A-16. Lecontella brunnea Figure A-17. Macrosigon cruentum

PAGE 134

120 Figure A-18. Nemognatha punctulata Figure A-19. Ancistorcerus

PAGE 135

121 Figure A-20. Euodynerus megaera Figure A-21. Pacnodynerus erynnis

PAGE 136

122 Figure A-22. Stenodynerus sp A Figure A-23. Stenodynerus sp b

PAGE 137

123 Figure A-24. Camponotus Red Figure A-25. Camponotus Black

PAGE 138

124 Figure A-26. Crematogaster minutissima Top: worker Bottom: Queen

PAGE 139

125 Figure A-27. Crematogaster species Figure A-28. Pseudomyrmex species

PAGE 140

126 Figure A-29. A wasp of the family Leucospididae

PAGE 141

127 Figure A-30. Dolicostelis louisa Figure A-31. Coelioxys sayi

PAGE 142

128 Figure A-32 Coelioxys dolichos Figure A-33. Coelioxys texana

PAGE 143

129 Figure A-34 Megachile campanulae Figure A-35. Megachile mendica

PAGE 144

130 Figure A-36. Megachile c. wilmingtoni Figure A-37. Megachile georgica

PAGE 145

131 Figure A-38 Megachile xylocopoides female Figure A-39. Megachile xylocopoides male

PAGE 146

132 Figure A-40. Osmia sandhouseae Figure A-41. Sphaeropthalma pensylvanica floridensis

PAGE 147

133 Figure A-42. Orocharis luteolira

PAGE 148

134 Figure A-43. Ampulex canaliculata Figure A-44. Liris beata

PAGE 149

135 Figure A-45. Podium rufipes

PAGE 150

136 Figure A-46. Trypoxylon clavatum clavatum Figure A-47. Face of Trypoxylon c. clavatum. Note golden vessiture

PAGE 151

137 Figure A-48. Trypoxylon carinatum Figure A-49. Trypoxylon clavatum johannis

PAGE 152

138 Figure A-50. Trypoxylon collinum collinum Figure A-51. Trypoxylon johnsoni

PAGE 153

139 Figure A-52. Trypoxylon lactitarse Figure A-53. Vespula maculifrons

PAGE 154

140 Figure A-54. Xylocopa virginica male Figure A-55. Xylocopa virginica female

PAGE 155

141 Figure A-56. Cerma cerintha Figure A-57. Litoprosopus frutilis

PAGE 156

142 Figure A-58. Uresphita reversali Figure A-59. Centruiodes hentzi

PAGE 157

143 APPENDIX B SELECTED SPECIMEN DIAGNOST ICS AND IDENTIFICATION Specimen Diagnostics And Identifications For Monobia Quadridens (Adapted from Bequaert 1940) Pronotum extensively and most of postsc utellum creamy-yellow; propodium with lateral angles pointed; propodium black, with the exception of a small spot on the dorsal areas; wings dark violaceous. Male: (Figure A-3) Antennae 13-segmente d, last segment folded back as a hook (Figure A-5). Clypeus distinctly bidentate. Clypeus creamy-yellow (except for denticles), sometimes with areas or patterns of orange-red. Female: (Figure A-4) Antennae 11-segmented, last segment normal, not folded back as a hook as in male (Figure A-5). Clype us distinctly bidentate and totally black, the anterior margin slightly concave. Specimen Diagnostics and Identification for Isodontia auripes and Isodontia mexicana Isodontia auripes and I. mexicana occur sympatrically a nd it is important to distinguish between the two species. The fo llowing diagnostic descri ption (Adapted from Bohart and Menke (1963)) prov ides characters to distinguish between the species and sexes of each species. Isodontia auripes : black; apex of hind femur, tibia and tarsi reddish brown; wings dark violaceous;

PAGE 158

144 Male: (Figure A-6) Average length 18mm; antenna with eleven flagellomeres; abdomen with seven visible tergites. Female (Figure A-7): Average length 19 mm; antenna with ten flagellomeres; abdomen with six visible tergites. Isodontia mexicana : black; legs black; wings clea r in cellular area, suffused with brown along anterior margin, veins black brown Male (Figure A-8): Average length 16mm; antenna with eleven flagellomeres; abdomen with seven visible tergites. Female: (Figure A-9) Average length 17 mm; antenna with ten flagellomeres; abdomen with six visible tergites.

PAGE 159

145 LIST OF REFERENCES Agosti, D., J.D. Majer, L.E. Alonso and T.R. Schultz. 2000. Ants: Standard Methods for Measuring and Monitoring Biodiversity Smithosonian Institution Press. Washington DC. Alves-dos-Santos, I. 2003. Trap-nesting bees and wasps on the University Campus in Sao Paulo, Southeastern Brazil (Hymenoptera: Aculeata). Journal of the Kansas Entomological Society 76:328-334. Ansley, R.J., H.T. Wiedemann, M.J. Cast ellano and J.E. Slosser. 2006. Herbaceous restoration of juniper dominated gr asslands with chaining and fire. Rangeland Ecological Management. 59:171-178. Armbrust, E.A. 2004. Resource use and nesting behavior of Megachile prosopidis and M. chilopsidis with notes on M. discorhina (Hymenoptera: Megachilidae). Journal of the Kansas Entomological Society 77: 89-98. Barton, A.M. 2005. Response of Arbutus arizonica (Arizona Madrone) to fire in southeastern Arizona. Southwestern Naturalist 50:7-11. Bess, E.C., R.R. Parmenter, S. McCoy and M. C. Molles Jr. 2002. Responses of a riparian forest-floor arthropod comm unity to wildfires in the middle Rio Grande Valley, New Mexico. Environmental Entomology 31: 774-784. Bequaert, J. 1940a. Monobia, Montezuia and Pa chymenes, neotropical elements in the nearctic Fauna (Hymenoptera, Vespidae). Annals of the Entomological Society of America. 33: 95-102. Bequaert, J. 1940b. Synopsis of Monobia de Sa ussure, an American genus of solitary wasp (Hymenoptera, Vespidae). Revista de Entomologia 11: 822-842. Bohart, R.M. and A.S. Menke. 1963. A reclassifi cation of the Sphecinae with a revision of the nearctic species of the tr ibes Sceliphronini and Sphecini. University of California publications in Entomology. 30: 91-182 Bohart, R.M., and A.S. Menke. 1976. Sphecid Wasps of the Worl d: A Generic Revision University of California Pre ss, Berkeley, California. 695 pp Bolton, J.L. and O. Peck. 1946. Alfalfa seed production in northern Saskatchewan as affected by Lygus bugs, with a report on their control by burning. Science in Agriculture 26: 130-137.

PAGE 160

146 Biswell, H.H. 1999. Prescribed Burning in Cal ifornia Wildlands Vegetation Management. Berkeley, CA: University of California Press. Brand, R.H. 2002. The effect of prescribed burning on epigeic springtails (Insecta: Collembola) of woodland litter. American Midland Naturalist 148:383-393. Burnham, K.P. and W.S. Overton. 1978. Estim ation of the size of a closed population when capture probabilities vary among animals. Biometrika 65: 623-633. Burnham, K.P. and W.S. Overton. 1979.R obust estimation of population size when capture probabilities vary amoung animals. Ecology 60: 927-936. Camillo, E. and A. D. Brescovit .1999. Spiders (Araneae) captured by Trypoxylon (Trypargilum) lactitarse (Hymenoptera: Sphecidae) in southwestern Brazil. Revista Biologia Tropical 47: E-article Camillo, E. and A. D. Brescovit. 2000. Spiders (Araneae) captured by Trypoxylon (Trypargilum) rogenhoferi (Hymenoptera: Sphecidae) in southeastern Brazil. Revista Biologia Tropical 48: E-article Camillo, E., C.A. Garofalo, G. Muccillo a nd J.C. Serrano. 1993. Bi ological observations on Trypoxylon (Trypargilum) lactitarse Saussure in southeastern Brazil (Hymenoptera: Sphecidae) Revista de Brazil Entomologia 37: 769-778. Camillo, E., C.A. Garofalo, J.C. Serrano and G. Muccillo. 1995. Diversidade e abundancia sazonal de abelhas e vespas solitarias em ninhos armadilhas (Hymenoptera, Apocrita, Aculeata). Revista Brasileira de Entomologia 39: 459470. Chao, A. 1984. Non-parametric estimation of the number of cl asses in a population. Scandinavian journal of Statistics 11: 265-270. Chao, A. 1987. Estimating the population size for capture-recapture data with unequal catchability. Biometrics 43: 783-791 Chao, A. and S.M. Lee. 1992. Estimating the number of classes via sample coverage. Journal of the American Statistical Association 87: 210-217 Chao, A., W.H. Hwang, Y. C. Chen and C.Y. Kuo. 2000. Estimating the number of shared species in two communities. Statistica Sinica 10: 227-246. Chao, A, R.L. Chazdon, R.K. Colwell and T.-J Shen. 2005.A new statistical approach for assessing compositional si milarity based on incidence and abundance data. Ecology Letters 8:148-159.

PAGE 161

147 Chazdon, R.L., R.K. Colwell, J.S. denslow and M.R. Guariguata. 1998. Statistical methods for estimating species richness of woody regeneration in primary and secondary rain forests of NE Costa Rica. Pp 285-309 in F. Dallmeier and J.A. Comiskey, eds. Forest biodiversity researc h, monitoring amd modeling: Conceptual background and Old World case studies. Parthenon Publishing, Paris Chew, R.M., B.B. Butterworth and R. Gretchman. 1959. The effects of fire on the small mammal populations of chaparral. Journal of Mammalogy 40: 253 Clayton, J.C. 2002. The effect of clearcutting and wildfire on grasshopper and crickets (Orthoptera) in an intermountain ecosystem. Journal of Orthoptera Research 11: 163-167. Clutton-Brock, T.H., K.E. Rose and F.E. Guinness. 1984. Maternal dominance, breeding success and birth ratio in red deer. Nature 308:358-360 Coddington, J.A., C. Griswald, D. Davila, E. Penaranda and S. Larcher. 1991. Designing and testing protocols to estimate biodivers ity in tropical systems. Pp 44-60 in, E. Dudley (Ed.) The Unity of Evolutionary Biology. Proceedings of the fourth International Congress of Syst ematic Evolutionary Biology. Discorides Press, Portland, Oregon. Collinge, S.K. 1996. Ecological consequences of habitat fragmentation: implications for landscape architectur e and planning. Landscape and Urban Planning 36: 59-77. Colwell, R.K. 2005. EstimateS: Statistical es timation of species richness and shared species from samples. Version 7.5. Use rs guide and application published at: http://purl.oclc.org/estimates (Last accessed, August 2006) Colwell, R.K. and J.A. Coddington.1994. Estimating terrestrial biodiversity. Philosophical Transactions of the Royal Society of London B. 345: 101-118. Colwell, R.K., C. Rahbek and N.J. Gotell i. 2004, The Mid-domain effect and species richness patterns: what ha ve we learned so far? The American Naturalist 163: Earticle. Coville, R.E. 1979. Biological observations on Trypoxylon (Tr y pargilum ) orizabense Richards in Arizona (H ymenoptera: Sphecidae) Journal of the Kansas Entomological Society 52: 613-620. Coville, R.E. 1981. Biological observations on three Trypoxylon wasp in the subgenus Trypargilum from Costa Rica: T. nitidum schultessi, T. saussurei and T. lactitarse (Hymenopera: Sphecidae). Pan-Pacific Entomologist 57: 332-340. Coville, R.E. 1982. Wasps of the Genus Trypoxylon Subgenus Trypargilum in North America. University of California, Berkeley, California 147 pp

PAGE 162

148 Coville, R.E. 1987. Spider-hunting sphecid wasps, pp 309-318. In W. Nintwig (ed.). Ecophysiology of spiders Springer, Berlin. 448p Coville, R.E. and P.L. Coville. 1980. Nesting biology and male behavior of Trypoxylon (Trypargilum) tenoctitlan in Costa Rica (Hymenoptera: Sphecidae) Annals of the Entomological Society of America 73: 110-119. Cumming, J.A. 1964. Effectiveness of prescrib ed burning in reduced wildfire damaged during periods of abnormally high fire danger, Journal of Forestry 62: 535-537. Cunningham, C.S., R.D. Babb, T. R. Jones, B.D. Taubert and R. Vega. 2002. Reaction of lizard populations to a catastrophic wildfi re in a central Arizona mountain range. Biological Conservation 107: 193-201. Dale, V.H., A.W. King, L.K. Mann, R.A. Washington-Allen, and R.A.McCord. 1998. Assessing land-use impacts on natural resources, Environmental Management 22: 203-211. Daubenmire, R. 1968. Ecology of fire in grasslands. Advanced Ecological Research 5: 209-266 DellaSala, D.A. and E. Frost 2001. An ecologi cally based strategy for fire and fuels management in National Forest roadless areas. Fire Management Today 61:12-23 Evans, E.W. 1984. Fire as a natural disturba nce to grasshopper a ssemblages of tallgrass prairie. Oikos 43: 9-16. Fay, P.A. 2003 Insect diversity in two burned and grazed grasslands. Environmental Entomology 32: 1099-1104. Fay, P.A and R.J. Samenus Jr. 1993. Gall wasp (Hymenoptera: Cynipidae) mortality in a spring tallgrass prairie fire. Environmental Entomology 22: 1333-1337 Frankie, G.W., R.W. Thorp, L.E. Newstrom-Lloyd, M.A. Rizzard, J.F. Barthell, T.L. Griswold, J.Y. Kim and S. Kappagoda. 1998. M onitoring solitary bees in modified wild land habitats: implications for bee ecology and conservation. Environmental Entomology 27: 1137-1148. Franz, R. and D.W. Hall. 1991. Vegetative communities and annot ated plant lists for the Katherine Ordway Preserve-Swisher Memori al Sanctuary, Putnam county Florida. Ordway Preserve Research Series, Re port No. 3. Florida Museum of Natural History, University of Florid a, Gainesville, Florida.65 pp Fuller, M. 1991. Forest fires: An Introduction to Wildland Fire Behavior, Management, Fire Fighting and Prevention Wiley Nature Editions. New York, NY: Wiley & Sons.

PAGE 163

149 Gathmann, A., H. J. Greiler and T. Tsch arntke. 1994. Trap-nesting bees and wasps colonizing set-aside fields: succession and body size, management by cutting and sowing. Oecologia 98: 8-14 Genaro, J.A. and G. Alayon. 1994. Las presas (Araneae) de Trypoxylon (Trypargilum) subimpressum (Hymeoptera: Sphecidae) en Cuba. Revista Bilogia Tropical 42: 353-356. Genaro, J.A., C.S. Sanchez and G. Alayon. 1989. Notas sobre la conducta de nidificacion de Trypoxylon (Trypargilum) subimpressum Smith (Hymenoptera: Sphecidae). Caribbean Journal of Science 25:228-229. Helms, J.A. 1979. Positive effects of pr escribed burning on w ildfire intensity, Fire Management Notes 40:10-13 Heltshe, J. and N.E. Forrester. 1983. Estima ting species richness using the jackknife procedure. Biometrics 39: 1-11. Hines, J.E., T. Boulinier, J.D. Nichols, J.R. Sauer, K.H. Pollock. 1999. COMDYN: Software to study the dynamics of animal communities using a capture-recapture approach. Bird Study 46 (suppl.): S209-217. Hopkins, G.W. and R.P. Freckleton. 2002. Decl ines in the numbers of amateurs and professional taxonomists: implications for conservation. Animal Conservation 5:245-249. Horn, S. and J.L. Hanula. 2004. A survey of cavity-nesting bees a nd wasps in Loblolly pine stands of the Savannah River site, Aiken County, South Carolina. Entomological Science 39: 464-469. Hughes, R.G. Theories and models of species abundance. 1986. The American Naturalist 128: 879-899. Hurlbert, L.C. 1969. Fire and litter effects in undisturbed bluestem prairie in Kansas. Ecology 50: 874-877. Jaysingh, D.B. and B.E. Freeman. 1980. The comparative population dynamics of eight solitary bees and wasps (Aculeata: Apoidea: Hymenoptera) trap nested in Jamaica Biotropica 12: 214-219. Jenkins, D.A. and R.W. Matthews. 2004. Ca vity-nesting Hymenopt era in disturbed habitats of Georgia and South Carolina: Nest Architecture and seasonal occurrence. Journal of the Kansas Entomological Society 77: 203-214. Jennings, D.T.and D.J. Hilburn. 1988. Spiders (Araneae) captured in malaise traps in spruce-fir forests of west-central Maine. Journal of Arachanology 16: 85-94.

PAGE 164

150 Jimenez, M.L. and A. Tejas. 1994. Las aranas presas de la avispa lodera Trypoxylon (Trypargilum) tridentatum tridentatum en Baja California Sur, Mexico. Southwestern Entomologist 19: 173-180. Kahn, W.C. 1960. Observations on the e ffect of a burn on population of Sceloporus occidentalis. Ecology 41: 358-359. Kimball, C.P. 1965. Lepidoptera of Florida. Arthropods of Flori da and neighboring land areas. Volume 1. Division of Plant Industr y, State Department of Agriculture. Gainesville, Florida. King, J.R., and S.D. Porter. 2005. Evaluation of sampling methods and species richness estimators for ants in upland ecosystems in Florida. Environmental Entomology 34: 1566-1578. Koponen, S. 2005. Early succession of a boreal sp ider community after forest fire. 2005. The Journal of Arachnology. 33:230-235. Kramer, K. 2005. Landscape Assessment In : Forests at the Wildland-Urban Interface S.W. Vince, M.L. Duryea, E.A. Macie and L.A. Hermansen, Eds CRC Press, Boca Raton, Florida. 293 pp. Krombein, K.V. 1956. Biological and taxonomic notes on the wasps of Lost River State Park, West Virginia, with additions to the faunal list (Hymenoptera, Aculeata) ). Proceedings of the Entomological Society of Washington 58: 153-161. Krombein, K.V. 1967. Trap-nesting Wasps and Bees: Life Histories, Nests and Associates Smithsonian Press, Washington DC. Krombein, K.V. 1970. Behavioral and Life-His tory Notes on three Floridian solitary wasps (Hymenoptera: Sphecidae). Smithsonian Contributions to Zoology No 46 Smithsonian Press, Washington DC. Krombein, K.V. and H.E. Evans. 1954. A list of wasps collected in Florida, March 29 to April 5, 1953, with biological a nnotation (Hymenoptera, Aculeata). Proceedings of the Entomological Soc iety of Washington 56: 225-236. LaSalle, J., and I.D. Gauld. 1993. Hymenoptera: their diversity, and their impact on the diversity of other organisms. In LaSalle and Gauld (eds.) Hymenoptera and Diversity. C.A.B. International, UK. 348 pp Laterra, P. O.R. Vigolio, M.P. Linares, A. Giaquinta and N. Maceira. 2003. Cumulative effects of fire on a tussock pampas grassland. Journal of Vegetation Science 14: 4354. Lawrence, G.E. 1966. Ecology of vertebrate anim als in relation to chaparral fire in the Sierra Nevada foothills. Ecology 47: 278-291.

PAGE 165

151 Lawton, J.H. 1983. Plant architecture and the diversity of phytophagous insects. Annual Revue of Entomology 28:23-39. Lin, C.S. 1966. Bionomics of Isodontia mexicana with a review of generic ethology(Hymenoptera: Sphecidae:Sphecinae.) the Wasmann Journal of Biology 24: 239-247. Lin, C.S. 1969. Biology of Trypoxylon spinosum with notes on T. texense and T. politum at Lake Texoma (Hymenoptera: Sphecidae: Trypoxyloninae) Wasmann Journal of Biology 27: 125-133. Lloret, F. and M. Vila. 2003. Diversity pattern of plant functional type s in relation to fire regime and previous land us e in Mediterranean woodlands. Journal of Vegetation Science. 14: 387-398. Long, J.A., D.D. Wade and F.C. Beall. 2005. Managing for fire in the interface: Challenges and Opportunities. In : Forests at the Wildland-Urban Interface S.W. Vince, M.L. Duryea, E.A. Macie and L.A. Hermansen, Eds CRC Press, Boca Raton, Florida. 293 pp. Longino, J.T., J. Coddington and R. Colwell. 2002. The ant fauna of a tropical forest: estimating species richness three different ways. Ecology. 83: 689-702. Main, M.B. and M.J. Barry. 2002. Influence of season of fire on flowering of wet prairie grasses in south Florida, USA. Wetlands 22: 430-434. Mangum, W.A. and S. Sumner. 2003. A Survey of the North American range of Megachile (Callomegachile) scupturalis, an adventive species in North America. Journal of the Kansas Entomological Society 76: 658-662. Magurran, A.E. 1988. Ecological Diversity and its Measurement Princeton University Press. Princeton, New Jersey. Magurran, A.E. 2004. Measuring Biological Diversity Blackwell Science Ltd. Oxford, UK. McCraken, D.I. and E.M. Bignal. 1998. Appl ying the results of eco logical studies to land-use policies and practices. Journal of Applied Ecology 35: 961-967. Medler, J.T. 1967. Biology of Trypoxylon in trap nests in Wisconsin (Hymenoptera: Sphecidae) American Midland Naturalist 78: 344-358. Meehan, T.D. and T.L. George. 2003. Short-term effects of moderate to high-severity wildfire on a disturbancedependent flycatcher in northwest California. The Auk 120: 1102-1113.

PAGE 166

152 Miller, D.G. III and L. Aviles.2000 Sex Ratio and brood size in a monophagous outcrossing gall wasp, Tamalia coweni (Homoptera: Aphididae). Evolutionary Ecology Research 2: 745-759. Miller, W.E. 1978. Use of prescribed burning in seed production areas to control red pine cone beetle. Environmental Entomology 7: 698-496. Miyano, S. and T. Yamaguchi. 2001. Ants re duce nest building activities of tubedwelling wasps and bees (Hymenoptera). Entomological Science 4: 243-246. Mushinsky, H.R. 1992. Natural history and abunda nce of southeastern five lined skinks, Eumeces inexpectatus on a periodically burn ed sandhill in Florida. Herpetologica 48: 307-312. Mutch, R.W. 1994. Fighting fire with prescrib ed fireA return to ecosystem health. Journal of Forestry 92: 31-33. Neff, J.L. and B.B. Simpson. 1993. Bees, pollination and plant diversity. In LaSalle and Gauld (eds.) Hymenoptera and Diversity C.A.B. International, UK. 348 pp Niwa, C.G. and R.W. Peck. 2002. Influence of Prescribed fire on Carabid Beetle (Carabidae) and Spider (Araneae) Assemb lages in Forest Litter in Southwest Oregon. Environmental Entomology 31: 785-796. Overbeck, G.H., S.C. Muller, V.D. Pillar and J. Pfadenhauer. 2005. Fine-scale post-fire dynamics in southern Brazili an subtropical grasslands. Journal of Vegetation Science. 16:655-664. Parker, F.D. and R.M. Bohart. 1966. Host-par asite associations in some twig-nesting Hymenoptera from western North America. Pan-Pacific Entomology 42:91-98. Parker, F.D. and R.M. Bohart. 1968. Host-par asite associations in some twig-nesting Hymenoptera from western North America part II. Pan-Pacific Entomology 44: 1-6 Paynter, Q. and G.J. Flanagan. 2004. Integrat ing herbicide and mechanical treatments with fire and biological control to manage an invasive wetland shrub, Mimosa pigra Journal of Applied Ecology. 41: 615-629. Pereira, M., C.A. Garofalo, E. Camillo and J.C. Serrano. 1999. Nesting biology of Centris (Hemisiella) vittata Lepeletier in southeaste rn Brazil (Hymenoptera, Apidae, Centridini). Apidologie 30: 327-338. Peruquetti, R.C. 2005. Use of trap ne sts with a neotropical mud-dauber, Trypoxylon ( Trypargilium ) albitarse Fabricius, 1804 (Hymenoptera: Sphecidae). Journal of the Kansas Entomological Society 78: 84-87 Pielou, E.C. 1966. An Introduction to Mathematical Ecology John Wiley, New York, 286 pp

PAGE 167

153 Pielou, E.C. 1975. Ecological Diversity John Wiley, New York, 165pp Porter, R.D. and L.L. Marsh. 2005. Developing Land while retaining Environmental Values: A modern search for the grail. In : Forests at the Wildland-Urban Interface S.W. Vince, M.L. Duryea, E.A. Macie and L.A. Hermansen, Eds CRC Press, Boca Raton, Florida. 293 pp. Potts, S.G., B. Vulliamy, A. Dafni, G. Neema n, C. OToole, S. Roberts and P. Willmer. 2003. Response of plant-pollinator commun ities to fire: changes in diversity, abundance and floral reward structure. Oikos 101:103-112 Rau, P. 1928. Field studies in the be havior of the non-social wasps Transactions of the Academy of Science of St. Louis 25:325-362. Raw, A. 1988. Social wasps (H ymenoptera: Vespidae) and ins ect pests of crops of the Surui and Cinta Larga Indi ans in Ronodonia, Brazil. The Entomologist 107: 104109. Reinhart, K.O. and E.S. Menges. 2004. Effect of re-introducing fire to a central Florida sandhill community. Applied Vegetation Science 7: 141-150. Riotte. 1975. Eine neue Art der amerikanischen Gattung Anisota (Lep., Saturniidae). Entomol. Z. 85:105-110. Saab, V.A. and K.T. Vierling. 2001. Reproduc tive success of Lewiss woodpecker in burned pine and cottonw ood riparian forests. The Condor 103: 491-501. Sax, D.F., B.P. Kinlan and K.F. Smith. 2005. A conceptual framework for comparing species assemblages in native and exotic habitats. Oikos 108: 457-464. Schoennagel, T., D.M. Waller, M.G. Turner and W.H. Romme. 2004. The effect of fire interval on post-fire understory commun ities in Yellowstone National Park. Journal of Vegetation Science. 15:797-806. Scott, V.L., S.T. Kelly and K. Stri ckler. 2000. Reproductive biology of two Coelioxys cleptoparasite in relation to their Megachile hosts (Hymenoptera: Megachilidae). Annals of the Entomological Society of America 93(4): 941-948. Sears, A.L., J.T. Smiley, M. Hilker, F. Mu ller and N.E. Rank.2001. Nesting behavior and prey use in two geographically separa ted populations of the specialist wasp Symmorphus cristatus (Vespidae: Eumeninae). American Midland Naturalist 145: 233-246. Shannon, C.E. and W. Weaver. 1949. The mathematical theory of communication. University of Illinois Press. Urbana, Illinois. Siemann, E., J. Haarstad and D. Tilman. 1997. Short-term and long-term effect of burning on Oak Savanna arthropods. American Midland Naturalist 137: 349-361.

PAGE 168

154 Simons, L. H. 1989 Vertebrates killed by desert fire. The Southwestern Naturalist. 34: 144-145Simpson, E.H. (1949). Measurement of Diversity. Nature 163: 688. Smith, E.P. and van Belle, G. 1984. Nonpara metric estimation of species richness. Biometrics 40, 119-129 Sorensen, T. 1948. A method of establishing group of equal amplitude in plant sociobiology based on similarity of species content and it s application to analyses of the vegetation on Danish commons. Biol. Skr 5: 1-34. Southwood, T.R.E. 1978. Ecological Methods 2nd ed. Chapman Hall, New York. 524 pp. Swengel, A.B. A literature review of insect responses to fire, compared to other conservation managements of open habitat. Biodiversity and Conservation 10: 114-1169. Steffan-Dewenter, I. 2002. Landscape context affect trap-nesting bees, wasps and their natural enemies. Ecological Entomology 27: 631-637 Thomas, L., J.L. Laake, S. Strindberg, F.F.C. Marques, S.T. Buckland, D.L.Borchers, D.R. Anderson, K.P. Burnham, S.L. Hedle y, J.H. Pollard, J.R.B. Bishop and T.A. Marques. 2005. Distance 5.0. Research Unit for Wildlife Population. University of St Andrews UK. http://www.ruwpa.st-and.ac.uk/distance (Last accessed August 2006) Toti, D.S., F.A. Coyle and J.A. Miller. 2000. A structured inventory of Appalachian grass bald and heath bald spider assemblages and a test of species richness estimator performance. The journal of Arachnology 28:329-345. Trivers, R.L. and D.E. Willard. 1973. Natural se lection of parental ab ility to vary the sex ratio of offspring. Science. 179: 90-92. Tscharntke, T., A. Gathmann and I.S. De wenter. 1998. Bioindication using trap-nesting bees and wasps and their natural enemies: community structure and interactions. Journal of Applied Ecology 35: 708-719. Turner, W, W Leitner, and M Rosenzweig. 2003. WS2M: Software for the measurement and analysis of species diversity.http://eebweb.arizona.edu/diversity (Last accessed, August 2006) Tuskes, P.M., J.P. Tuttle and M.M. Collins.1996. The Wild Silk Moths of North America Cornell University Press. Ithaca, New York. ONeill, K.M. 2001. Solitary Wasps: Behavior and Natural History Cornell University Press, Ithaca, NY.

PAGE 169

155 ONeill, K.M. and R.P. ONeill. 2003. Sex allocation, Nests and prey in the Grasscarrying Wasp Isodontia mexicana (Saussure)(Hymenoptera: Sphecidae). Journal of the Kansas Entomological Society 76: 447-454. Vasquez, A., B. Perez, F. Fernandez-Gonzalez and J.M. Moreno. 2002. Journal of Vegetation Science. 13:663-676. Wade, D.D. and J.D. Lunsford. 1989. A Guide for prescribed Fire in Southern Forests. Tech. Pub. R8-TP11, U.S. Dept of Agricult ure, forest Service, Southern Region, Atlanta. Wcislo, W.T. 1996. Parasitism rates in rela tion to nest site in bees and wasps (Hymenoptera: Apoidea). Journal of Insect Behavior 9:643-656.

PAGE 170

156 BIOGRAPHICAL SKETCH David Serrano was born in Miami, Florida, in 1977 to Cuban immigrants. He graduated high school in 1995 in Miami. He then attended the University of Miami where he earned a Bachelor of Science (major: biology; minor: chemistry). Through many volunteer projects and superv ised projects at the University of Miami he discovered his love for ecology and biology. His expe riences with Drs. Keith D. Waddington, Theodore H. Fleming, David Janos, Michael S. Gaines and Paul R. Neal helped guide David to his current career. David accepted a teaching assistantship at the University of Florida, Department of Entomology and Ne matology, in the Fall of 1999. After earning his Master of Science he remained at the University of Florida. He accepted a Minority Alumni Fellowship and began his PhD progr am in the fall 2001. David has a daughter, Isabella Carin Serrano (born May 4th 2001). David married Esther Sarah Dunn (UF, Doctor of Plant Medicine 2005) on January 15th, 2006.