Citation
Life history, ecology, and conservation genetics of the striped newt (Notophthalmus perstriatus

Material Information

Title:
Life history, ecology, and conservation genetics of the striped newt (Notophthalmus perstriatus
Creator:
Johnson, Steve A., 1966-
Publication Date:
Language:
English
Physical Description:
ix, 156 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Amphibians ( jstor )
Breeding ( jstor )
Ecological life histories ( jstor )
Emigration ( jstor )
Fences ( jstor )
Highlands ( jstor )
Larvae ( jstor )
Metamorphosis ( jstor )
Newts ( jstor )
Ponds ( jstor )
Dissertations, Academic -- Wildlife Ecology and Conservation -- UF ( lcsh )
Notophthalmus -- Ecology -- Florida ( lcsh )
Notophthalmus -- Genetics ( lcsh )
Notophthalmus -- Life cycles ( lcsh )
Wildlife Ecology and Conservation thesis, Ph.D ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2001.
Bibliography:
Includes bibliographical references (leaves 140-154).
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Steve A. Johnson.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute 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.
Resource Identifier:
027703508 ( ALEPH )
48222078 ( OCLC )

Downloads

This item has the following downloads:


Full Text






LIFE HISTORY, ECOLOGY, AND CONSERVATION GENETICS OF THE
STRIPED NEWT (Notophthalmus perstriatus)
















By

STEVE A. JOHNSON


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


2001













ACKNOWLEDGMENTS

First and foremost, I would like to express my deepest appreciation to my wife

Dale for enduring the stresses of a husband enrolled in graduate school for many years.

Dale has been a constant source of encouragement and intellectual stimulation throughout

my graduate research. She has opened my mind to new ideas, allowed me to see and

appreciate beauty that previously I was blind to, and taught me what is important in life.

Without her support I would not have been able to complete my graduate education.

Dale drafted many of the maps and figures in this dissertation. I also thank my parents,

my mother Bobbie W. Johnson and my late father Gordon E. Johnson, for encouraging

the biological interests of their son through putting up with all the messes I made as a

child.

There are numerous friends and colleagues who came to my aid throughout my

research projects and I deeply appreciate their assistance. These folks helped me collect

samples, install and remove drift-fence arrays, check traps, and tend experimental

animals. I would like to thank the following individuals for their help in this regard: Brad

Austin, Mark Bailey, Jamie Barichivich, Bobby Bass, Laura Becht, Boyd Blihovde,

Cheryl Cheshire, Ken Dodd, Brian Emanuel, Dick Franz, Dan Hipes, John Jensen, Dale

Johnson, Kenney Krysko, Ryan Means, Paul Moler, Bubba Owen, David Printiss, Rob

Robbins, Matt Seguin, Joe Sexton, Parks Small, Lora Smith, Jennifer Staiger, Dirk

Stevenson, and Chad Truxall.







For help with statistical analyses and data interpretation I would like to thank:

Anna Bass, Brian Bowen, Brian Cade, Ginger Clark, and Julie Heath. For providing

constructive criticism on earlier drafts of my dissertation I thank Alicia Francisco, Holly

Freifeld, Dale Johnson, David Leonard, and the members of my graduate committee. For

assistance in the lab working on my genetics project I thank: Anna Bass, Ginger Clark,

and Alicia Francisco. I would also like to express my gratitude to staff in the Department

of Wildlife Ecology and Conservation for all of their help during my graduate studies at

the University of Florida. In particular I would like to thank Laura Hayes, Monica

Lindberg, Caprice MacRae, Sam Jones, and Cynthia Sain for their kind assistance

throughout my graduate career. I thank Kent Vliet and John Reiskind for providing me

with teaching assistantships through the Biological Sciences Program. I also thank the

clerical staff, Kenetha Johnson and Tangelyn Mitchell, for their help while I served as a

TA.

Funding for my dissertation projects was provided by the U.S. Fish and Wildlife

Service, the Gopher Tortoise Council, and the Florida Fish and Wildlife Conservation

Commission. I am especially grateful to Linda LaClaire of the U.S. Fish and Wildlife

Service for administering my grants from this agency. The Lerio Corporation, BEECS

Genetics Analysis Core at the University of Florida, and U.S. Geological Survey donated

materials and provided lab space. In particular, I acknowledge Russ Hall of the U.S.

Geological Survey for providing me with space to rear experimental animals. I would

also like to express my gratitude to the governing board of the Katharine Ordway

Preserve/Swisher Memorial Sanctuary for allowing me to conduct field research on the







Preserve. I am grateful to John Eisenberg, Dick Franz, and Mel Sunquist for facilitating

my work on the Preserve.

For permission to collect newt tissue samples at Ichauway, I thank Lindsey

Boring of the Joseph Jones Ecological Research Center, Newton, GA. In Georgia,

samples were collected under Georgia DNR scientific collecting Permit #00335. I thank

Parks Small for facilitating sample collection at Rock Springs Run State Preserve and the

Florida Division of Forestry for permission to collect at Jennings State Forest.

I would like to acknowledge the support of the members of my graduate

committee: C. Kenneth Dodd, Jr. (committee chair), Dick Franz, Brian Bowen, Mark

Brenner, George Tanner, and Mike Moulton for their guidance and assistance. I extend

special thanks to Dick Franz for everything he has done for me. Dick always looked out

for my best interests and I am most appreciative. Dick was also the person who gave me

my first experience with striped newts.

Finally, I extend my most sincere thanks to all the great friends that Dale and I

have had the pleasure of spending time with during our stay in Gainesville. It was

certainly the highlight of our years here.














TABLE OF CONTENTS

page

ACKN OW LEDGM ENTS ............................................................ .............................. ii

ABSTRACT ............................................................................................................... viii

CHAPTER 1
INTRODUCTION TO THE STRIPED NEWT (Notophthalmus perstriatus).................... 1
The Striped N ewt...................................................................................................... 1
Status of the Striped N ewt ..................................................... ................................. 1
Current Knowledge and Research Justification.................................... .................. 2
Striped Newt Life History and Life-history Stage Terminology ................................... 3
Larvae ................................................................................................................... 4
Efts .............................................................................................................................. 4
Paedom orphs............................................................................................................... 4
Adults.......................................................................................................................... 5
Overview of Dissertation ....................................................................................... 5

CHAPTER 2
LIFE HISTORY OF THE STRIPED NEWT AT A NORTH-CENTRAL FLORIDA
BREEDIN G POND ....................................................................................................... 9
Introduction............ .......................................................................................... .............. 9
M materials and M ethods............................................................................................ 10
Study Site .................................................................................................................. 10
Drift Fence at One Shot Pond............................................... ........................... 11
N ewts Caught at Drift Fences.................................................. ......................... 12
W weather Data ...................................................................................................... 13
Statistical Analyses ............................................................................................. 13
Results............................................. .............................................................................. 13
Seasonal Activity ................................................................................................ 13
Imm igration......................................................................................................... 14
Em igration........................................................................................................... 14
Reproduction............ .................................. ........................................................... ... 15
Population Size Structure.................................................. ................................ 17
Sex Ratios ........................................................................................................... 18
Rainfall and Hydroperiod ......................................................................................... 19
Discussion ..................................................................................................................... 20
Seasonal Activity ................................................................................................ 20
Im m igration......................................................................................................... 21








Emigration................................................................................................................. 21
Reproduction....................................................................................................... 22
Population Size Structure.................................................. ................................ 24
Sex Ratios ........................................................................................................... 24
Hydroperiod and Rainfall ...................................................................................... 25
Comparisons with Notophthalmus viridescens................................................ 26
Implications for Striped N ewt Status Surveys .............................................. .. 29

CHAPTER 3
ORIENTATION AND DISPERSAL DISTANCES OF STRIPED NEWTS AT A
NORTH-CENTRAL FLORIDA BREEDING POND.......................................................46
Introduction.............................................................. .................................................. 46
M ethods......................................................................................................................... 49
Study Site .................................................................................................................. 49
Orientation at One Shot Pond .................................................. ......................... 50
Upland Dispersal................................................................................................. 51
Results...................................................................................................................... ..... 53
Orientation at One Shot Pond .................................................. ......................... 53
Dispersal Into Uplands........................................................................................ 55
Discussion ..................................................................................................................... 58
Orientation .......................................................................................................... 58
Upland D ispersal......... .................................................. .................. ............. .. 60
Conservation Implications ............................................................ .................... 61

CHAPTER 4
INFLUENCE OF GROWTH RATE ON LIFE-HISTORY EXPRESSION OF STRIPED
NEW TS ........................................................................................................................75
Introduction................................................................................................................... 75
M ethods................................................................................................................... 80
Experim ental Design.............................................................................. ........... 80
Procedures................................................................................................................. 81
Food Treatm ents .................................................................................................. 83
Dissections .......................................................................................................... 84
Data Analysis...................................................................................................... 85
Results.......................................................................................... ..................... 87
Larval Growth................................................................................... ..................... 87
Size at M etam orphosis and Larval Period ........................................................... 88
Life-history Pathway Expression............................................................................ 88
Discussion ..................................................................................................................... 90
Expression of Alternative Life-history Pathways .................................. ............ 90
M etam orphosis and M odel Applicability .............................................................. 94

CHAPTER 5
CONSERVATION GENETICS AND PHYLOGEOGRAPHY OF THE STRIPED
NEW T ...................................................................................................................... ........ 106
Introduction.................................................................................... ..................... ....... 106








M ethods................................................................................................................. 108
Sam ple Collection........................................... ..................................................... 108
DNA Isolation and Sequencing ............................................................................. 109
Data Analysis .................................................................................................... 110
Results.............................................................................................................................. 111
Discussion................................................................................................................... 113
Population Structure........................................................................................... 113
Testing a Biogeographic Hypothesis .................................................................... 115
Striped Newt Biogeography and Phylogeography................................................... 116
Conservation and M anagem ent Im plications........ .......................................... 119

CHAPTER 6
SUM M ARY AND CON CLU SION S .................................. ........................................131
Life-history Sum m ary ................................................................................................. 131
Conservation, M anagem ent, and Research Prospectus............................................. 132


LIST OF REFEREN CES ............................................................................................... .. 140

BIOGRAPHICAL SKETCH ....................................... ................ ............................155













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

LIFE HISTORY, ECOLOGY, AND CONSERVATION GENETICS OF THE STRIPED
NEWT (Notophthalmus perstriatus)



By

Steve A. Johnson

August 2001


Chairman: C. Kenneth Dodd, Jr.
Major Department: Wildlife Ecology and Conservation

The striped newt (Notophthalmus perstriatus) is a salamander endemic to south

Georgia and north-central Florida. The species has declined throughout its range because

of habitat destruction and modification. Before my research, little was known about

striped newt life history.

To learn more about striped newt ecology in order to make management and

conservation recommendations, I studied several aspects of striped newt life history. I

used a multidisciplinary approach that incorporated fieldwork, a laboratory experiment,

and DNA sequencing. For 2 years I monitored drift fences at a north-Florida breeding

pond and in the sandhill uplands around the pond. This method was used to determine

basic parameters of the species' life history. In the laboratory experiment I reared

individual larvae on different food regimes to test the influence of growth rate on life-


viii







history expression. A portion of the Cytochrome b gene was sequenced to determine

genetic population structure.

Field-collected data showed that striped newts have a complex life-cycle,

involving terrestrial and aquatic stages. An individual may move between a breeding

pond and upland retreats multiple times during its life. Larval development occurs in the

pond, but once metamorphosis is complete, individuals leave the pond and may disperse

in excess of 500 m from the pond. Striped newts may express one of two life-history

pathways. An individual may initiate metamorphosis and disperse from the pond before

it matures (metamorph), or it may remain in the pond and mature while retaining larval

characteristics (paedomorph). The metamorph vs. paedomorph "decision" is not

controlled by growth rate per se, but is likely influenced by a suite of genes. Based on

DNA sequence data, significant population genetic structure was found among ten

locations sampled throughout most of the species' range, showing that gene flow is

severely restricted among populations.

It appears that striped newts form metapopulations and that long-term survival of

the species depends on preserving those metapopulations that persist. Conservation

efforts should focus on protecting and managing upland and aquatic habitats. A

landscape approach is most effective, and prescribed fire in the landscape is essential.












CHAPTER 1
INTRODUCTION TO THE STRIPED NEWT (Notophthalmus perstriatus)


The Striped Newt

The striped newt (Notophthalmus perstriatus) is a salamander endemic to

southeastern Georgia and north-central Florida (Christman and Means, 1992; Conant and

Collins, 1991; Dodd and LaClaire, 1995; Franz and Smith, 1999; Mecham, 1967).

Individuals are restricted to xeric upland habitats (primarily sandhill and scrub

communities) and breed exclusively in temporary wetlands that lack predaceous fishes

(Campbell and Christman, 1982; Carr, 1940; Christman and Means, 1992; Dodd and

LaClaire, 1995; Dodd et al., in press; Franz and Smith, 1999; Stout et al., 1988). These

upland ecosystems are pyrogenic (Myers, 1990), and fire appears to be crucial for the

persistence of striped newts. Besides having a complex life history involving aquatic and

terrestrial stages (Christman and Means, 1992; Dodd, 1993), individuals commonly

exhibit paedomorphosis, the retention of larval morphology in mature individuals

(Bishop, 1941a, 1943; S. A. Johnson, pers. obs.).


Status of the Striped Newt

Decline of the longleaf pine/wiregrass ecosystem, fire suppression, and the natural

patchy distribution of upland habitats required by striped newts have resulted in the

fragmentation of striped newt populations. Striped newts have declined throughout their

range (Dodd and LaClaire, 1995; Franz and Smith, 1999). A complex life history makes

striped newts vulnerable to threats at breeding ponds (e.g., ditching, draining, and filling







of temporary ponds) and within the surrounding uplands (e.g., fire suppression, various

silviculture practices, and urban and agricultural development). Relative abundance of

striped newts is extremely low at most areas where the species persists (S. A. Johnson, B.

Means, K. Greenberg, and D. Stevenson, unpubl. data). Because of historical declines

and low relative abundance at most locations, the striped newt is recognized as a rare

species throughout its range (Christman and Means, 1992; Cox and Kautz, 2000; Jensen,

1999). Its biological status is under review by the U.S. Fish and Wildlife Service L.

LaClaire, pers. comm.).


Current Knowledge and Research Justification

Striped newts have been characterized as "uncommon and enigmatic" (Christman

and Means, 1992) and "poorly known" (Dodd, 1993; Dodd and LaClaire, 1995) and until

the last decade or so, little was known about their ecology. Most of the literature on

striped newts is limited to the results of surveys (Dodd and LaClaire, 1995; Franz and

Smith, 1999; Hipes and Jackson, 1996) and to species accounts (Ashton and Ashton,

1988; Bishop, 1941a, 1943; Carmichael and Williams, 1991; Carr, 1940; Christman and

Means, 1992; Dodd et al., in press; Mecham, 1967; Petranka, 1998). Johnson and Franz

(1999) documented the occurrence of albinism in the species. Dodd and Charest (1988)

and Dodd (1992) mentioned striped newts as part of the herpetofaunal community of a

north Florida sandhills pond. Dodd (1996) included striped newts in his survey of

terrestrial habitat use by amphibians. Studies of striped newt feeding habits (Christman

and Franz, 1973), natural history at a breeding pond (Dodd, 1993), and orientation into

and away from a breeding pond (Dodd and Cade, 1998) represent the only published

works focusing specifically on striped newt life history. Petranka (1998) provided a







summary of the biology of striped newts based mainly on the work of Dodd and

coauthors.

Taking into account the decline ofN. perstriatus throughout its range, and that its

biological status is under review by the U.S. Fish and Wildlife Service, it is essential that

natural resource managers acquire knowledge of striped newt life history. Such

information will be required in order to draft a recovery plan, which would be required by

law if the species was federally listed. Knowledge of striped newt life history also will

be of immediate use to natural resources managers and may help circumvent the need to

federally protect the species.


Striped Newt Life History and Life-history Stage Terminology

To better understand the following chapters of this dissertation, it would help to

have a basic understanding of striped newt life history and terminology describing the

various life-history stages. The life history stages are complex, and no published source

adequately defines these stages as they relate to N. perstriatus. Throughout the life of an

individual, both aquatic and terrestrial stages occur ontogenetically, and an individual

may move several times between aquatic and terrestrial habitats (Fig. 1-1).

Reproduction occurs primarily in isolated, temporary ponds that lack predaceous

fishes because the ponds dry relatively often. Courtship and oviposition by adult newts

occur in the ponds and females lay eggs one at a time over a protracted period of several

months. Eggs are often attached to aquatic vegetation. Eggs hatch into immature larvae

that feed and grow in the pond.







Larvae

A larva is an immature aquatic stage newt. Larvae have bushy external gills, a

membranous tail fin, and a conspicuous lateral line, which is visible as a series of dorso-

lateral dashes on each side of the animal. Larvae do not posses the namesake lateral

stripe and they do not have swollen vents. After a period of growth, but before sexual

maturation, a larva may metamorphose and leave the pond as an immature eft (i.e.,

metamorphic pathway; Fig. 1-1). On the other hand, a larva may remain in the pond,

continue to grow, and mature while retaining the larval morphology (i.e., paedomorphic

pathway; Fig. 1-1).

Efts

An eft is an immature terrestrial stage newt. Efts lack gills and do not have a tail

fin or lateral line. However, at metamorphosis the namesake dorso-lateral stripe, which is

reddish to orange, appears on each side of an eft. Because efts are immature they do not

have swollen vents. After larvae metamorphose into efts, they disperse into the uplands

surrounding the breeding pond (Fig. 1-1). Efts mature in the uplands, at which point they

are referred to as terrestrial adults.

Paedomorphs

A paedomorph is a mature aquatic stage newt. Paedomorphs are larger than

immature larvae, have bushy external gills, a membranous tail fin, and a conspicuous

lateral line. Paedomorphs do not usually possess the namesake lateral stripe but they do

have swollen vents. Paedomorphs reproduce in the breeding pond, then metamorphose

and disperse into the surrounding uplands (Fig. 1-1). Once a paedomorph transforms and

leaves the pond it is referred to as a terrestrial adult. On rare occasions, a paedomorph

may transform directly into an aquatic adult (e.g., dashed line in Fig. 1-1).







Adults

An adult is a mature terrestrial or aquatic stage newt. Adults lack gills and do not

have a visible lateral line. Adults possess the namesake dorso-lateral stripes and have

swollen vents. They are sexually dimorphic, and males have a light-colored gland that is

visible at the posterior end of the vent. Adults occur in the uplands around breeding

ponds (i.e., terrestrial adults) as well as in the ponds (i.e., aquatic adults), and there is

movement between these habitats during the life of an adult. Aquatic adults develop a

membranous tail fin similar to the tail fin larvae and paedomorphs. They do not regrow

external gills, however. Terrestrial adults lack a tail fin. They differ from efts in that the

vent of a terrestrial adult is swollen, whereas the vent of an eft is not swollen.

In the chapters that follow, I refer to the various life-history stages of striped

newts as outlined above. Readers may need to refer back to this section, as well as Fig.

1-1, until they are familiar with the striped newt life cycle and the different stages that

comprise it.


Overview of Dissertation

Chapters 2 through 5 present the results of a multidisciplinary research project on

striped newt life history; they include data based on a fieldwork component, a laboratory

experiment, and a molecular genetics study. Each of these chapters plays an integral role

in a unifying theme of striped newt life history and conservation. Chapters are written in

manuscript format, each with its own introduction, materials and methods, results, and

discussion, to facilitate publication in peer-reviewed journals.

Prior to my research, knowledge of striped newt life history was limited to results

of studies at a single pond during a drought period. Because life history information is







crucial for conservation and management planning, I conducted a 2-year field study of

striped newt life history at a different pond during a relatively wet period. In addition to

monitoring newts at the breeding pond, I trapped newts at various distances from the

pond in the surrounding uplands. Results from captures at the pond, which are presented

in Chapter 2, demonstrated that striped newts have a complex life history and are adapted

to taking advantage of temporary breeding habitats that fluctuate within and among

seasons. Directionality of newt movements into and away from the pond was nonrandom

(Chapter 3), and results of upland captures, also presented in Chapter 3, showed that

striped newts dispersed hundreds of meters from the pond.

One component of the striped newt's complex life history is the expression of

alternative life-history pathways (i.e., metamorphic pathway vs. paedomorphic pathway).

Field-collected data showed that within a single cohort some larvae metamorphosed and

left the breeding pond before attaining sexual maturity (metamorphic pathway), whereas

others remained in the pond, continued to grow, and matured in the larval morphology

(paedomorphic pathway). In a laboratory experiment, I tested the hypothesis that

expression of these alternative life-history pathways is influenced by larval growth rate.

The experiment also allowed me to test the applicability of two popular models of

amphibian metamorphosis as they pertain to N. perstriatus. As presented in Chapter 4,

growth rate did not significantly affect life-history pathway expression, and neither of the

two models of metamorphosis was totally consistent with the results of the experiment.

Inasmuch as the genetic structuring of populations has important evolutionary,

biogeographical, management, and conservation implications, I conducted analyses of

mitochondrial DNA of striped newts from throughout their range. Results showed that







there has been considerable evolutionary differentiation among the locations sampled,

probably because of genetic drift caused by natural habitat fragmentation. Genetic data

supported data from mark-recapture studies, which suggest that striped newts form

metapopulations. The results of this molecular genetic study are presented in Chapter 5,

and the consequences of metapopulation structure and population fragmentation for the

conservation of this imperiled species are discussed in the last chapter.

Chapter 6, the final chapter, contains a brief summary of striped newt life history

as well as conservation and management recommendations based on the results of the

preceding chapters. In this chapter, I also provide suggestions for additional research on

N. perstriatus.





terrestrial
adult J


eft


4


eggs


adult


Fig. 1-1. Life-history schematic of the striped newt. Life-history stages include aquatic and terrestrial phases. During the life of an
individual, it will move between aquatic and terrestrial habitats. Lines and arrows indicate direction of movement. The dashed line
infers that this developmental path is possible, but uncommon.


larva .












CHAPTER 2
LIFE HISTORY OF THE STRIPED NEWT AT A NORTH-CENTRAL FLORIDA
BREEDING POND


Introduction

Salamanders of the genus Notophthalmus occur exclusively in North America

with three extant species: N. viridescens (eastern newt), N. meridionalis (black-spotted

newt), and N. perstriatus (striped newt). Notophthalmus viridescens ranges throughout

the eastern United States and into southeastern Canada, whereas N. meridionalis is

confined to extreme southeast Texas and northeastern Mexico (Conant and Collins, 1991;

Petranka, 1998). Notophthalmus perstriatus is limited to northern Florida and southern

Georgia (Conant and Collins, 1991; Petranka, 1998). Each of the species exhibits a

complex life cycle, involving aquatic and terrestrial phases. The ecology of N.

viridescens has been well studied (Gill, 1978a, b; Harris, 1987; Harris et al., 1988; Healy,

1970, 1973, 1974a, b, 1975; Hurlbert, 1969; Pope, 1924). On the other hand, far less

research has focused on N. meridionalis and N. perstriatus (Petranka, 1998).

Because of historical declines and current relative abundance, which is low

throughout most of its range, N. perstriatus is recognized as a rare species in Florida and

Georgia (Christman and Means, 1992; Cox and Kautz, 2000; Jensen, 1999). Its

biological status is under review by the U.S. Fish and Wildlife Service (L. LaClaire, pers.

comm.). Most of what has been reported about striped newts has been limited to the

results of surveys (Dodd and LaClaire, 1995; Franz and Smith, 1999; Hipes and Jackson,

1996) and species accounts (Ashton and Ashton, 1988; Bishop, 1941a, 1943; Carmichael







and Williams, 1991; Christman and Means, 1992; Carr, 1940; Dodd et al., in press;

Mecham, 1967; Petranka, 1998). In the only study of N. perstriatus life history, Dodd

(1993) monitored striped newt movements at a single breeding pond from 1985 through

1990. However, a severe drought impacted the pond throughout Dodd's study. Despite

dry conditions, Dodd (1993) determined seasonal activity, population size structure, and

sex ratio. Of necessity these data were limited primarily to the adult life stage. Very

little information was available on metamorphic individuals because the breeding pond

only held water for short periods. It was not clear if the patterns observed by Dodd were

typical of striped newt life history.

To gain a better understanding of striped newt life history, I conducted a 2-year

study at a breeding pond in north-central Florida. I used a drift fence (Gibbons and

Semlitsch, 1981) to monitor striped newt immigration and emigration at the pond. My

objectives were: 1) to determine the timing of immigration and emigration of newts, 2) to

measure breeding success by monitoring emigration of metamorphic animals, 3) to

estimate population size-structure and sex ratio, and 4) to evaluate the influence of

hydroperiod and rainfall on striped newt movements and reproduction.


Materials and Methods

Study Site

The study was conducted at One Shot Pond (OSP), an isolated water body within

a high pine community in north-central Florida, approximately 4 km west of Breezeway

Pond, the site of Dodd's (1993) striped newt study. One Shot Pond is located in Putnam

Co., FL, on the Katharine Ordway Preserve-Swisher Memorial Sanctuary (Fig. 2-1).

Descriptions of the Preserve and its habitats are provided elsewhere (Dodd, 1996;







Eisenberg and Franz, 1995; LaClaire, 1995). One Shot Pond is a sinkhole-depression

pond with a basin area of ca. 0.8 ha (LaClaire, 1995). The hydroperiod of the pond is

variable (hydroperiod refers to the number of days a pond holds water between periods

when it is dry) and it dries periodically. Because of this, OSP does not support fish, and

many species of amphibians breed there (Johnson, 1999).

Drift Fence at One Shot Pond

Newt movements into (i.e., immigration) and away from (i.e., emigration) OSP

were monitored with a continuous drift fence (Gibbons and Semlitsch, 1981) that

encircled the pond. The galvanized metal fence was buried in the ground ca. 15 cm with

ca. 35 cm extending above ground. The circumference of the fence was 190 m, with 38

pitfalls (19 pairs of 19 1 plastic buckets) buried flush with the ground at intervals of ca. 10

m. For each pair of pitfalls, one was buried on the side of the fence toward the pond and

one on the side away from the pond. To reduce mortality of trapped animals, foam

sponges were placed in each trap and cover boards were leaned against the fence over

each pitfall to provide shade. Each time traps were checked I removed invertebrates

(e.g., spiders, predaceous beetles, and centipedes) and added water to keep the sponges

moist. Traps were checked at least 3 times per week and daily during periods of warm

weather and/or high rainfall. Because of relatively high rainfall during the winter of

1997/98, the water in OSP rose to a level at which three pairs of pitfalls became

inundated. In March of 1998, the entire fence was moved ca. 12 m further up slope,

requiring an additional 40 m of flashing. The fence was moved in one day, so there was

no break in trapping. The number of pitfalls remained the same and each trap was

reinstalled in the same relative position at the new fence location. The drift fence was







monitored from 7 Oct-96 to 11 Sep-98. Pitfall traps were open for 705 consecutive days,

for a total of 26,790 trap-nights (i.e., one trap-night = one pitfall trap open for 24 hours).

Newts Caught at Drift Fences

Most newts were marked with a unique toe clip so individuals could be identified

(Donnelly et al., 1994), but recently metamorphosed efts captured after 11 Jul-98 were

only marked with a "daily cohort" toe clip. Newts were weighed to the nearest 0.1 g

using a Pesola scale, and snout-vent length (SVL) was measured to the nearest 1 mm.

The sex of each newt was recorded as male, female, or unknown. Sex was determined by

examining the vent region; adult males have a light-colored gland visible at the posterior

edge of the cloaca (Dodd, 1993). The condition of the cloaca was recorded as swollen,

slightly swollen, or not swollen, indicative of mature, maturing, and immature

individuals, respectively (Chapter 4). Animals were released on the opposite side of the

fence from where they were captured.

Three distinct life-history stages of striped newts were examined: adults, efts

(immature larvae that recently metamorphosed), and paedomorphs (mature larvae that

recently metamorphosed, Table 2-1, Chapter 1). Recently metamorphosed newts retained

vestiges of their gills (i.e., gill buds) for several days after they left the pond. Therefore,

the presence of gill buds indicated that a newt had recently transformed and left the pond.

Data for adults immigrating to the pond include recaptures. Many of the adults captured

in pitfalls on the outside of the drift fence had been previously captured and marked in

the uplands surrounding OSP (Chapter 3). Others were initially marked as they

emigrated from the pond as immature efts. Data for emigrating adults also include

recaptured individuals. These individuals had been initially marked as described above

or when they were captured in pitfalls on the outside of the drift fence as they immigrated







to breed. Data for recently transformed paedomorphs and efts only include initial

captures. By excluding eft and paedomorph recaptures, I obtained a clearer pattern of

striped newt movements.

Weather Data

To evaluate the influence ofhydroperiod and rainfall on striped newt movements

and reproduction, rainfall and pond depth were monitored at OSP. Rainfall (to the

nearest mm) was measured with a rain gauge mounted in the open within the pond basin.

Pond depth was measured with a permanent depth gauge placed in the center of the pond.

I used binoculars to read the depth gauge.

Statistical Analyses

When assumptions of parametric tests were violated, nonparametric methods

(Hollander and Wolfe, 1999) were used to test for differences between data sets. All

statistical analyses, with the exception of 2 tests, were performed using SPSS ver. 10.0.

I used X2 tests (Sokal and Rolf, 1995) to test for departure of 1:1 sex ratios for adults and

paedomorphs. To calculate expected values, I divided the total number of males and

females used in each analysis by two.


Results

Seasonal Activity

During the 2-year study, 10,290 striped newt captures were recorded at the drift-

fence encircling OSP. At least one newt was captured during every week of the study,

although four periods of activity accounted for the vast majority of captures (Fig. 2-2).

Activity during these four peaks included newts moving into (immigration) and away

from (emigration) the pond.







Immigration

Immigration was almost exclusively comprised of breeding migrations of adult

striped newts, and there were four distinct adult immigrations during the study (Table 2-

2, Fig. 2-3A). The largest and most prolonged immigration event was immigration Event

E-3, which lasted for 6 months. Peak movement during this event occurred in Dec-97

(Fig. 2-3A). During this month, 1,567 adults were captured in outside pitfall traps. The

other three immigration events were much smaller and occurred during 2-month or 3-

month periods. Females and males immigrated during the same times of the year (Fig. 2-

3A).

Although many juvenile newts were captured in pitfalls on the outside of the drift

fence, these individuals had recently metamorphosed and were initially captured and

marked on the inside of the fence after they left the pond. Rather than immediately

dispersing into the uplands, some of these individuals headed back toward the pond and

were caught in outside pitfalls. Nonetheless, captures of these animals were the result of

very localized movements and not an indication of immigration by efts. When recaptures

of recently transformed individuals are excluded, almost all (96%) newts captured in

outside pitfalls had swollen vents, indicating that they were sexually mature. The

remainder had slightly swollen vents, indicating that they were close to maturity.

Emigration

Emigration events included individuals representing all three life-history stages:

adults, efts, and recently metamorphosed paedomorphs (Table 2-2). Similar to

immigration, there were four distinct periods of adult emigration (Table 2-2, Fig. 2-4).

The largest of these (E-4) occurred toward the end of the study (Fig. 2-4). This

emigration event accounted for 88% of all emigrating adults with more adults captured







leaving the pond in Jun-98 than any other month of the study. Female and male adults

immigrated during the same times of the year (Fig. 2-3B).

Efts metamorphosed and emigrated during all months of the year except February,

but there were four distinct periods of emigration (Table 2-2, Fig. 2-4). Most efts (81%)

emigrated during the last 5 months of the study period (eft emigration Event E-4).

Recently metamorphosed paedomorphs were captured during three emigration events

(Table 2-2, Fig. 2-4). Most (94%) were captured from Mar-97 through Aug-97

(paedomorph emigration Event E-2).

Reproduction

Totals of 5,296 recently transformed larvae (i.e., efts) and 435 recently

transformed paedomorphs were captured during the 2 years. These individuals likely

represented successful reproduction of four distinct breeding bouts, as indicated by

emigration of recently transformed newts throughout the study (Fig. 2-4). The first

evidence of successful reproduction was provided by captures of emigrating efts (E-1,

Table 2-2) and paedomorphs (E-1, Table 2-2) during the first few months of the study

(Fig. 2-4). During this period, 776 efts and 25 recently transformed paedomorphs were

captured dispersing from OSP (Fig. 2-4). The second period of eft and paedomorph

production occurred in spring and early summer of 1997. During this time, 214 efts and

407 recently transformed paedomorphs were captured (E-2, Table 2-2, Fig. 2-4). Only 16

efts were produced during the third eft emigration event (E-3, Table 2-2), but no recently

transformed paedomorphs were captured during this period (Fig. 2-4). By far, the most

successful reproductive bout during the 2 years was indicated by eft emigration (E-4) that

occurred during the last 5 months of the study. Efts were captured starting in late May-

98, and immature larvae continued to transform and leave the pond until the end of the







study (Fig. 2-4). I likely did not document the full extent of this emigration event since

98 efts were captured in inside pitfalls on the last day of the study. Only nine recently

transformed paedomorphs were caught during this period.

After they transformed and emigrated from the breeding pond, efts migrated into

the surrounding sandhill uplands (Chapter 3). While in the uplands, efts matured before

they returned to the pond to breed as adults. The largest immigration of adults (Event I-

3), which occurred from Oct-97 through Mar-98, consisted of many newts that were

captured initially as they emigrated as efts during the first few months of the study (eft

Event E-1). Although these newts were easily recognized as recaptures when they

returned to the pond, I could not be sure of individual toe clips in some instances. Of 40

newts that I was confident of their toe clip, all of which had been marked as efts when

they immigrated (i.e., vents not swollen), 39 had matured (i.e., vents swollen) by the time

they were recaptured in outside pitfalls. Based on dissections and examination of gonads

(Chapter 4), newts with swollen vents are always sexually mature. Therefore, at least 39

of the 40 recaptured efts had matured in the uplands, then migrated back to the pond to

breed a year or more after they left the pond. These 40 newts had remained at large in the

uplands around OSP for an average of 416 days (SD = 19.7; range = 359 to 456 days).

The average number of days at large since metamorphosis was similar between the sexes.

Males (n = 16) averaged 412 days at large (SD = 22.8; range = 359 to 440 days), whereas

females (n = 24) averaged 419 days (SD = 17.4; range = 394 to 456 days). Net growth,

measured as the difference in SVL between initial capture during emigration and

recapture during immigration, was similar between females and males (Fig. 2-5).







Differences in growth rates (net growth (mm)/days at large) between the sexes (Table 2-

3) were not significantly different (Wilcoxon rank sum test; t, = 0.815, P > 0.4).

Population Size Structure

The size-structure of striped newts differed between immigrating and emigrating

adults. Snout-vent length and mass differed significantly among immigrating and

emigrating males and females (SVL: F3,3136 = 776, P < 0.0001; mass: F3,3135 = 628, P <

0.0001). Post hoc comparisons showed that immigrating adults of both sexes were

significantly smaller than emigrating adults for SVL (Fig. 2-6) and mass (Fig. 2-7). On

average, immigrating females were slightly larger (SVL and mass) than immigrating

males, and this pattern was evident during immigration Events I-1, 1-3, and 1-4 (Table 2-

4). During immigration Event 1-2, adult males and females were almost the same size

(Table 2-4). However, overall differences were not significant for SVL or mass (Figs. 2-

6A, 2-7A). Adult females slightly exceeded males in SVL and mass for emigration

Events E-3 and E-4, but males were slightly larger than females during the first two

emigration events (Table 2-4). Overall, emigrating females were significantly larger in

SVL (Fig. 2-6B) and mass (Fig. 2-7B) than emigrating males (Scheff6's tests, all P <

0.0001).

Recently metamorphosed efts ranged in SVL from 20 32 mm (n = 2605) and

from 0.2 1.0 g in mass (n = 1886). Body size (SVL and mass) of efts differed among

the four emigration events (Table 2-5). They were smallest during Event E-1 and largest

during Event E-4. Snout-vent length and mass of efts differed significantly among three

(E-l, E-2, E-4) of the four emigration events (SVL: F2,2581 = 2114, P < 0.0001; mass:

F2,1862 = 128, P < 0.0001; Table 2-5). Post hoc comparisons showed that both measures

of body size were significantly different for efts during Events E-1, E-2, and E-4







(Scheff6's tests, all P < 0.0001; Table 2-5). Event E-3 was excluded from the analyses

because of small sample sizes.

Male and female paedomorphs were essentially the same size (SVL and mass)

during all three emigration events. Overall, males were slightly longer than females but

the average mass of males and females was the same (Table 2-5). There were no

statistical differences in either measure of body size between the sexes (t-tests; SVL: t =

1.42, P = 0.156; mass: t = -0.159, P = 0.874; Table 2-5). Of the three paedomorph

emigrations, Event E-2 had the largest number of individuals (Table 2-5). Because of

small sample sizes for Events E-1 and E-3, I did not make statistical comparisons of body

size among the three events.

Body sizes of the three different life-history stages differed (Table 2-5).

Emigrating efts that had recently transformed had the smallest body size (SVL and mass),

followed by recently transformed paedomorphs, then emigrating adult males. Emigrating

adult females, on average, were the largest of all stages. These differences were

statistically significant (ANOVA; SVL: F3,4017 = 3789, P < 0.0001; mass: F3,3266 = 1578,

P < 0.0001) and post hoc tests showed that means differed among all four groups eftss,

paedomorphs, males, and females; Scheff6's tests, all P < 0.0001).

Sex Ratios

Sex ratios were male biased during immigration Events I-1 and I-4 but female

biased during Events I-2 and 1-3 (Table 2-6). Because of the relatively large number of

captures during immigration event 1-3, when the sex ratio was 1:1.26 (m:f), the overall

sex ratio of immigrating adults was 1:1.26. During emigration, the sex ratio of adults was

female biased during all events except E-1 (Table 2-6). The relative contribution of sex

ratio data provided by emigration Event E-4 (88% of all emigrating adults) had a large







influence on the overall sex ratio of emigrating adults, which was 1:1.22 (Table 2-6).

Overall adult sex ratio (emigrating and immigrating individuals) was female biased

(Table 2-6; X2 = 43.9, df= 1, P < 0.001).

The sex ratio of recently metamorphosed paedomorphs was highly female-biased

during each of the three emigration events. Event E-2 was by far the largest of the three

events, representing 92% of paedomorph captures. Therefore E2 had a large impact on

the overall sex ratio of paedomorphs, which was significantly skewed toward females

(m:f = 1:4.64; 2 = 161.8, df = 1, P< 0.001).

Rainfall and Hydroperiod

Monthly rainfall at OSP varied ranged from 12 mm to 283 mm (Fig. 2-8). The

driest periods were Nov-96 through Mar-97 and Mar-98 through Jul-98 (Fig. 2-8). The

wettest period was from Jun-97 through Feb-98 because of an El Nifio Southern

Oscillation event. Rainfall exceeded 100 mm during 13 months of the study period, and

beginning in Jun-97, there were 7 months consecutively in which rainfall exceeded 100

mm. Summer rainfall resulted from localized thunderstorms, whereas winter rain was

associated with cold fronts.

Newts tended to move during wetter periods and newt captures were significantly

correlated with rainfall (P < 0.001). Nonetheless, rainfall was a weak predictor of the

magnitude of newt movements and only explained a small portion of variation in

movements of newts at OSP (r2 = 0.06).

One Shot Pond held water throughout the study period (Fig. 2-9). Pond depth was

lowest (68 cm) in Oct-97, but the El Nifio rains filled the pond to its greatest depth (275







cm) the following Apr. Analyses of the influence of pond drying and filling on striped

newt reproduction are precluded because OSP always held water during the study.

Although pond depth exceeded 65 cm for the duration of the study, water depth

may have influenced the survivorship of larvae, and therefore the number of emigrating

efts. Of the four eft emigration events, Event E-3 was the smallest (Table 2-5),

coinciding with the shallowest pond depth during the study (Fig. 2-9). The largest eft

emigration (Event E-4, Table 2-5) began in May-98, when pond depth exceed 260 cm.

Pond depth increased steeply during the months before May-98 (Fig. 2-9), when larvae

that transformed during Event E-4 were growing in the pond. Although pond depth may

have influenced the survivorship of striped newt larvae, conclusions are confounded by

the fact that a variable number of females potentially contributed eggs that resulted in efts

for each emigration event. Fewer than 70 females appear to have contributed to the

production of larvae during eft emigration E-3, whereas ca. 1300 females potentially

produced the larvae that transformed and emigrated during Event E-4.


Discussion

Seasonal Activity

Striped newts were active at OSP during all months of the year, but there were

four periods of activity that accounted for most captures. In the only published study of

striped newt life history, Dodd (1993) also found several periods of activity that

accounted for the majority of his striped newt captures over a 5-year period. At OSP, two

activity periods occurred during the fall/winter, whereas the other two took place during

the spring/summer. At Breezeway Pond, striped newts were mainly active during the

fall/winter portion of the year (Dodd, 1993).







Immigration

Adults moved into OSP to breed from Oct. through Mar. and from Apr. through

Jul. At Breezeway Pond, 75% of adults immigrated from Jan. through Mar. (Dodd,

1993). Although data collected at OSP support the fact that striped newts tend to breed

during the winter (Dodd, 1993; Petranka 1998), there were also two distinct migrations

(presumably breeding migrations) during the spring/summer. Clearly, striped newts are

plastic in the timing of breeding migrations. The only months adults were not

documented moving into OSP were Aug. and Sep. Dodd (1993) suggested that the

extended breeding period of striped newts allows them to take advantage of temporary

breeding habitats that fluctuate within and among seasons. Such a plastic life history is

likely an adaptation to living in an unpredictable environment.

Emigration

As with immigration, there were four distinct periods of emigration. These

periods overlapped with the four immigration events. Adults migrated to the wetland,

then courted and bred I assume, and then moved back into the surrounding uplands. This

pattern persisted throughout the study even though OSP always held water. Therefore,

emigration of adults was not simply because of pond drying, which appeared to be the

case at Breezeway Pond (Dodd, 1993). Based on the interval adults spent in the pond, as

well as laboratory observations of reproductive activity (Johnson, unpubl.), striped newts

have protracted courtship and oviposition. Females, including paedomorphic individuals,

lay eggs one at a time and do so over the course of several months. As a result, adults

that immigrated into OSP during the winter of 1997/98 (Event 1-3) for example, stayed in

the pond until they had finished breeding and then emigrated during the summer of 1998

(Event E-4). This staggered pattern of immigration, later followed by emigration, applies







to the two other immigration events as well. Therefore, adults immigrating during Events

I-1 and I-2 left the pond several months later, during Events E-2 and E-3, respectively

(Tables 2-4, 2-5, Fig. 2-3). A similar pattern is apparent in Dodd's (1993) data early in

his study (Fig. 1 in Dodd, 1993), although the variable hydroperiod of Breezeway Pond

and small number of captures confounds interpretation during the later years.

Reproduction

More than 5,500 recently transformed striped newts were captured as they

emigrated from OSP. Production of very large numbers of metamorphic individuals is

not uncommon for pond-breeding amphibians (Semlitsch et al., 1996). However, no

previous studies have found so many metamorphic N. perstriatus (Dodd, 1993; B. Means,

pers comm.; K. Greenberg, pers. comm.; D. Stevenson, pers. comm.). As a result of a

drought, Dodd (1993) only captured 42 recently metamorphosed newts during the entire

5-year study at Breezeway Pond.

Recently transformed efts emigrated in all months except January, but there were

periods of concentrated migration, three of which accounted for 99.6% of the captures

(Fig. 2-4, Table 2-5). At Breezeway Pond, recently transformed striped newts were only

captured from Jun. through Aug-97. The four eft emigration events at OSP presumably

represent four bouts of reproduction. Eft emigration E-1 during the first two months of

the study resulted from a reproductive event that likely occurred before the study began.

The other three emigrations of efts probably represent reproduction of adults that

immigrated during the study period. For example,' adult immigration Event I-1 produced

larvae that metamorphosed and left the pond during eft emigration Event E-2. Adults

that were captured immigrating during Events 1-2 and 1-3, probably produced most of the

larvae that transformed and left OSP during eft emigration Events E-3 and E-4,







respectively (Figs. 2-3A, 2-4). This staggered pattern of adult immigration, later

followed by eft emigration, was because larvae apparently required approximately 6

months to reach metamorphosis. In the single successful reproductive bout recorded by

Dodd at Breezeway Pond, larvae required a 139-day hydroperiod. Paedomorphs also

likely contributed to production of larvae in OSP but the relative contributions of non-

gilled adults and paedomorphs are unknown.

A successful reproductive event often appeared to produce a bimodal distribution

of emigrating newts. This is because within a single cohort of larvae, both immature and

mature larvae may result. Some immature larvae transformed and exited the pond as efts,

whereas others remained in the pond, attained sexual maturation (i.e., paedomorphs),

reproduced, then transformed and exited the pond. This resulted in a bimodal pattern of

emigration of a cohort with immature efts showing up first, followed by transformed

paedomorphs. For example, the recently transformed paedomorphs captured in the

spring of 1997 (paedomorph Event E-2) were probably members of the same cohort that

produced the efts that emigrated the previous fall (eft Event E-1; Fig. 2-4). The few

paedomorphs that were caught in Jun. and Jul. of 1998 were likely members of the same

cohort that produced the few efts that emigrated during eft Event E-3. Monthly samples

taken in OSP with dip nets showed that the larvae that eventually matured as

paedomorphs remained in the pond after their counterparts had transformed and

emigrated as efts (S. A. Johnson, unpubl. data). Comparisons of size at metamorphosis

and sex ratio for paedomorphs are precluded because no comparable data are available

from other studies.







Population Size Structure

Sizes of recently transformed efts at OSP encompassed the range of sizes of this

life-history stage at Breezeway Pond with the exception of the smallest individuals

captured at Breezeway. Recently transformed juveniles that Dodd (1993) captured at

Breezeway Pond ranged from 18 to 25 mm (n = 47) and 0.1 to 0.4 g (n = 44). Based on

Dodd's (1993) data, a snout-vent length of 18 mm and a mass of 0.1 g. appear to be

absolute minimum body sizes required for a striped newt to initiate metamorphosis.

Although the average SVL of recently transformed efts at OSP was 25.8 mm, the average

SVL during the four emigration events varied significantly across all events. Variation in

zooplankton availability during each of the periods preceding the emigration events may

have caused the differences.

The sizes of adult striped newts at OSP were similar to sizes of adults from

Breezeway Pond. Snout-vent length of females at OSP ranged from 25 to 43 mm,

whereas SVL of females at Breezeway Pond ranged from 26 to 43 mm. One Shot Pond

females ranged from 0.3 to 1.6 g and Breezeway Pond females ranged from 0.3 to 2.0 g.

Adult males at OSP were, on average, slightly smaller than females and ranged in SVL

from 26 to 38 mm with mass ranging from 0.4 to 1.2 g. Dodd (1993) also found that

male striped newts were smaller than females. Snout-vent length of males at Breezeway

Pond was the same as that for OSP, whereas mass ranged from 0.2 1.6 g (Dodd, 1993).

Sex Ratios

Overall sex ratio of striped newts at OSP and Breezeway Pond were significantly

female-biased. At OSP, there was one male for every 1.25 females, and at Breezeway

Pond Dodd (1993) captured one male for every 1.46 females. The significance and







causes) of the female bias in striped newts at the Katharine Ordway Preserve are

unknown.

Hydroperiod and Rainfall

The hydroperiod of amphibian breeding ponds has a strong influence on

reproduction (Pechmann et. al., 1989; Semlitsch, 2000; Semlitsch et al., 1996). If

hydroperiod is too short, larvae do not have adequate time to initiate metamorphosis and

will therefore perish as the pond dries. On the other end of the spectrum, permanent

ponds usually support predacious fishes that can extirpate some aquatic-breeding

amphibians (Semlitsch, 2000). Although OSP held water during the entire study, over

the past 2 decades it has dried often enough to preclude predatory fishes (R. Franz, pers.

comm.). The large number of efts emigrating from OSP during the 2-year study was

probably because the pond held water continuously. In contrast, Dodd (1993) observed

standing water in Breezeway Pond during only 14 months of the 5-year study period.

Breezeway Pond held water in five distinct episodes (Fig. 1 in Dodd, 1993), and the

longest of these episodes was a 139-day hydroperiod. This was the only time during the

study when Dodd (1993) captured recently transformed juveniles. The shorter

hydroperiods precluded larval maturation, and consequently, no recently transformed

paedomorphs were captured at Breezeway Pond.

Long-term rainfall patterns likely have a significant impact on the striped newt

population at the Katharine Ordway Preserve. Variability in hydroperiods of striped newt

breeding ponds over relatively long time periods probably result in "boom or bust"

scenarios for striped newt reproduction. Alternating relatively dry and wet intervals

appear to result in highly variable striped newt reproductive success within and among

ponds. Dodd (1993) captured very few metamorphic newts during his 5-year study and







attributed an observed decline in striped newts at Breezeway Pond to persistent drought

conditions. At OSP on the other hand, I observed an increase in the number of striped

newts, mainly because of the large number of larvae that metamorphosed during the last

several months of the study. The heavy rainfall during the winter of 1997/98 filled the

pond to its greatest depth (275 cm) while these larvae were developing. Because of the

relatively great depth of the pond, I suggest that there was more habitat and food

(zooplankton) available to the larvae. This could have reduced intraspecific competition

and contributed to the reproductive success and corresponding survivorship.

Comparisons with Notophthalmus viridescens

The life history of the red-spotted newt, Notophthalmus viridescens, has been

studied in detail (Gill, 1978a, b; Harris et al., 1988; Healy, 1970, 1973, 1974a, b, 1975;

Hurlbert, 1969; Pope, 1924). Dodd (1993) compared and contrasted the life history of N.

viridescens with N. perstriatus at Breezeway Pond. Data for striped newts at OSP allow

some additional comparisons that Dodd was unable to make because of the poor

reproductive success at Breezeway Pond.

Striped newt larvae appear to be more variable than red-spotted newts with regard

to the time of year when metamorphosis occurs. Recently transformed efts of N.

viridescens have been found from Jun. through Nov. (Bishop, 1941b; Gill, 1978a;

Hurlbert, 1970; Worthington, 1968). Recently transformed striped newt efts at OSP were

found during these months, but they were also captured during Dec., Mar., Apr., and

May. The variation in timing of adult immigration and breeding of striped newts at OSP,

along with variability of the aquatic habitat quality (e.g., food and pond depth), is likely

responsible for the extreme variation in timing of larval metamorphosis. Healy (1973)

estimated that red-spotted newt larvae in Massachusetts had a larval period of







approximately six months. The same was true for striped newt larvae (excluding

paedomorphs) in OSP during my 2-year study. However, red-spotted newt larvae from

other locations have shorter larval periods (Bishop, 1941b; Harris et al., 1988;

Worthington, 1968). Red-spotted newt larvae transform into efts when they reach a SVL

of 19 to 21 mm (Petranka, 1998). The minimum size for metamorphosis of immature

striped newt larvae is similar (Dodd, 1993; this study)

The duration of the eft stage can be shorter for striped newts than red-spotted

newts. At OSP, many efts matured and immigrated to breed after an eft stage of about 14

months. Notophthalmus viridescens efts may remain on land from 2-8 years before

returning to breeding ponds (Bishop, 1941b; Healy, 1974a). Striped newts certainly have

much greater variability in duration of the eft stage than I documented at OSP. Although

many efts matured and returned to OSP after a period of about 14 months, this is likely a

minimum time-frame. Some efts probably remained in the uplands around OSP and did

not migrate to breed. Therefore, these individuals, once they matured and migrated to the

pond to breed, would have an eft stage longer than 14 months. Moreover, if drought

conditions had prevailed in the vicinity of OSP during the winter of 1997/98, the efts that

did migrate to breed would not have had the opportunity to do so, thus increasing the

estimate of duration of the eft stage. Annual variation in rainfall certainly has an impact

on the duration of the terrestrial stage of striped newts.

Growth rates of N. perstriatus efts at OSP appear similar to growth rates of

juvenile red-spotted newts. Healy (1973) calculated growth rates of marked efts in a

Massachusetts's population of red-spotted newts for three consecutive years. Healy

(1973) presented "mean growth increment" values (Table 2 in Healy, 1973), and I







estimated daily growth rates by dividing his mean values by 365. The range of growth

rates (mm SVL / day) for red-spotted newts varied from 0.0048 to 0.0159. Growth rates

of striped newt efts (females: 0.0167 mm SVL / day; males: 0.0183 mm SVL / day) were

similar to growth rates of red-spotted newts in Massachusetts (0.0159 mm SVL / day;

Healy, 1973). Because of the short duration of my study, annual variation in growth rates

of striped newt efts is unknown.

For striped newts and red-spotted newts, life-history pathway has a profound

influence on age at first reproduction. In both species, individuals that omit the eft stage

reach sexual maturity earlier than individuals that metamorphose when immature. For N.

perstriatus, an individual that omits the eft stage matures as a paedomorph and

reproduces at about 1 year old. For N. viridescens, an individual that omits the eft stage

may remain in the pond and later mature as an aquatic adult, or mature as a paedomorph

(Brandon and Bremer, 1966; Healy, 1970, 1974a; Petranka, 1998). According to Healy

(1974a), immature red-spotted newts that remain in the pond and omit the eft stage

reproduce earlier (at 2 years old) than newts that migrate into the uplands as efts. This

life-history pathway has not been detected in stripe newts. Red-spotted newts that

become paedomorphic may reach sexual maturity in as little as seven months (Petranka,

1998). Expression of the paedomorphic life-history pathway is common throughout the

range ofN. perstriatus (S. A. Johnson, unpubl. data; D. Stevenson, pers. comm.; D. B.

Means, pers. comm.; J. Jensen pers. comm.). Paedomorphosis is most common in coastal

populations ofN. viridescens and in places where the terrestrial environment is perceived

as exceptionally harsh (Bishop, 1941b; Brandon and Bremer, 1966; Healy, 1974a;







Petranka, 1998). In both species, genetic and environmental factors are believed to

control the expression of life history pathway (Chapter 4; Harris, 1987).

Implications for Striped Newt Status Surveys

Life history data for striped newts at OSP have implications for management of

the species. Considering the imperiled status throughout its range (Cox and Kautz, 2000;

Jensen, 1999), identifying undocumented breeding ponds and monitoring striped newts at

known breeding ponds will help ensure the long-term persistence of the species.

Probably the most efficient method to survey multiple sites is by sampling breeding

ponds for striped newts. Obviously, such surveys must be conducted during non-drought

periods when breeding ponds hold water. Dodd's (1993) work proved that during

drought conditions, a suitable striped newt pond may only hold water for short periods.

Furthermore, as drought conditions persist, newt abundance declines. Therefore, even

when potential breeding ponds hold water, newts might be present in such low numbers

as to elude detection. Use of drift fences around ponds will increase the likelihood of

detection, but this method is very labor-intensive and is not practical for range-wide

surveys when time and personnel are limited. Failure to detect newts during drought

periods may result from low relative abundance caused by the drought, rather than local

extirpation. On the other hand, during wet conditions, such as was the case at OSP

following the winter of 1997/98, newt abundance can be relatively high. Wet periods

increase the likelihood of detecting the species during aquatic sampling. Surveys

conducted during relatively wet periods will prove fruitful for monitoring persistence and

locating new breeding ponds. Additionally, as suggested by Dodd (1993), surveys for

striped newts should include assessment of biotic and abiotic characteristics of known

and potential breeding sites.







Striped newts of various life-history stages may be found in breeding ponds

during all months of the year. However, I suggest that spring (Apr. through Jun.) is the

best time of the year for conducting aquatic sampling for the species, assuming ponds

hold water. At this time of year, OSP contained all three life-history stages. Adults and

paedomorphs that had recently bred were still in the pond, as were developing larvae.

Sampling for newts in breeding ponds during this time of the year should maximize the

probability of capturing newts. However, considering the temporary nature of striped

newt breeding ponds, individuals conducting striped newt sampling should conduct

surveys whenever breeding ponds hold water.










Table 2-1. Descriptions of the three life-history stages of striped
newts referred to in the Chapter 2.
Stage Description
Adult Mature newt as indicated by a swollen vent;
no evidence of gills; males with a distinct,
light-colored gland at posterior edge of cloaca

Eft Immature newt as indicated by a vent that
is not swollen; gill vestiges present--indicating
recent metamorphosis of an immature branchiate;
sex recorded as unknown


Paedomorph


Mature newt as indicated by a swollen vent, gill
vestiges present--indicating recent metamorphosis
of a mature branchiate; males with a distinct,
light-colored gland at posterior edge of cloaca






Table 2-2. Timing of immigration (I) and emigration (E) events of the three life-history stages of
striped newts captured at One Shot Pond, Putnam Co., FL.
Life-history Stage
Event Adults Efts Paedomorphs
I-1 Oct.-96 through Dec.-96 not applicable not applicable
I-2 Apr.-97 and May-97 not applicable not applicable
I-3 Oct.-97 through Mar.-98 not applicable not applicable
I-4 Jun.-98 and Jul.-98 not applicable not applicable
E-l Oct.-96 through Dec.-96 Oct.-96 through Dec.-96 Oct.-96 through Dec.-96
E-2 Apr.-97 through Aug.-97 Mar.-97 through Jun.-97 Mar.-97 through Aug.-97
E-3 Nov.-97 through Jan.-98 Aug.-97 through Dec.-97 Jun.-98 and Jul.-98
E-4 May.-98 through Aug.-98 May.-98 through Sep.-98 not applicable










Table 2-3. Growth of striped newt efts at One Shot Pond,
Putnam Co., FL. Growth is expressed as mm/day from initial
capture during emigration (shortly after metamorphosis)
until recapture during immigration.
Females Males
n 24 16
Mean 0.0167 0.0183
Range 0.0068 0.0295 0.0129 0.0306
SD 0.0054 0.0043











Table 2-4. Snout-vent length (SVL) and live mass of adult
striped newts caught at One Shot Pond, Putnam Co., FL
during four immigration events.


Event
I-1


n
Mean
Range
SD


I-2 n
Mean
Range
SD


I-3





1-4





Total


n
Mean
Range
SD

n
Mean
Range
SD

n
Mean
Range
SD


Females
SVL (mm) Mass (g)
11 7
36.27 0.80
26-40 0.6-1.0
4.34 0.13


58
33.57
27 40
4.19

1148
30.90
26-41
1.87

10
35.6
33 39
1.90

1227
31.11
26-41
2.24


56
0.72
0.4- 1.1
0.20

1140
0.61
0.4-1.2
0.11

9
1.0
0.8- 1.2
0.14

1212
0.62
0.4-1.2
0.13


Males
SVL (mm) Mass (g)
18 11
33.80 0.62
27-39 0.3-1.0
3.80 0.20


22
33.59
28-40
3.28

924
30.77
26-39
1.66

8
34.25
33 36
0.89

972
30.91
26 40
1.88


22
0.72
0.4- 1.0
0.14

914
0.61
0.3-1.0
0.11

6
0.90
0.8- 1.0
0.11

953
0.61
0.3- 1.0
0.11







Table 2-5. Snout-vent length (SVL) and live mass of adult, recently transformed eft, and recently transformed paedomorph
striped newts captured at One Shot Pond, Putnam Co., FL during four emigration events.


Adults
Females Males


Event
E-I


n
Mean
Range
SD


E-2 n
Mean
Range
SD

E-3 n
Mean
Range
SD


E-4





Total


n
Mean
Range
SD

n
Mean
Range
SD


SVL (mm)
4
29.00
25 38
6.06


70
32.40
26 41
4.07

13
30.54
28-33
1.45

469
36.31
30 43
1.73

556
35.63
25 43
2.73


Mass (g)
2
0.65
0.4 0.9
0.35

66
0.70
0.3 -1.1
0.19

13
0.58
0.4 0.8
0.14

460
0.93
0.5- 1.6
0.18

541
0.88
0.3- 1.6
0.21


SVL (mm)
7
30.57
26 38
4.24


18
33.06
28-37
2.71

11
29.82
26-33
1.99

414
33.72
31 -38
1.29

450
33.55
26 38
1.63


Mass (g)
4
0.73
0.6 0.9
0.15

18
0.68
0.4-1.0
0.14

10
0.54
0.5 0.6
0.05

406
0.82
0.4- 1.2
0.13

438
0.80
0.4- 1.2
0.14


Paedomorphs


Efts
SVL (mm) Mass (g)


772
23.02
20 29
1.34

209
28.29
23 32
1.51

21
24.43
21 -32
2.84

1603
26.74
22-31
1.46

2605
25.75
20-32
2.33


0.31
0.2 0.5
0.07

213
0.55
0.3- 1.0
0.12

21
0.33
0.2 0.8
0.15

1598
0.46
0.2-1.0
0.11

1886
0.47
0.2- 1.0
0.12


SVL (mm)
17
26.47
24-30
1.55

334
31.838323
27 43
2.54

5
37.40
35-41
2.51

not
applicable




356
31.66
24-43
2.83


Females
Mass (g)
6
0.42
0.4 0.5
0.04


320
0.73
0.4- 1.8
0.18

5
1.04
0.5- 1.6
0.40

not
applicable




331
0.73
0.4- 1.8
0.19


SVL (mm)
7
27.00
24 29
1.63


72
32.57
28 40
2.06

2
35.00
33-37
2.83

not
applicable




81
32.15
24 40
2.60


Males
Mass (g)
3
0.48
0.4-0.5
0.06


70
0.73
0.4- 1.0
0.13

2
1.05
0.9-1.2
0.21

not
applicable




75
0.73
0.4-1.2
0.15










Table 2-6. Sex ratios of adult and paedomorph striped newts
captured at One Shot Pond, Putnam Co., FL. Overall and grand
total values include several individuals not accounted for
in the immigration (I) and emigration (E) events listed. Sex ratios
are listed as the ratios of males:females, followed by the number
males and females captured in parentheses.
Event Adults Paedomorph
I-1 1:0.57 (23:13) not applicable
1-2 1:3.00 (22:66) not applicable
1-3 1:1.26 (1038:1307) not applicable
1-4 1:0.69 (32:22) not applicable
Overall-I 1:1.26 (1119:1412) not applicable
E-1 1:0.57 (7:4) 1:2.57 (7:18)
E-2 1:3.74 (19:71) 1:4.64 (72:334)
E-3 1:1.08 (12:13) 1:3.50 (2:7)
E-4 1:1.23 (431:484) not applicable
Overall-E 1:1.22 (469:572) 1:4.43 (81:359)
Grand total 1:1.25 (1588:1984) 1:4.43 (81:359)





















ATLANTIC
OCEAN


Katharine




GULF
OF MEXICO










area
enlarged


r(


N


Fig. 2-1. Location of study area, Katharine Ordway Preserve, in Putnam Co., north-
central Florida.


10 ,"t ..
K> a90D










1800

1600

1400

S1200

4o 1000

800

600

400

200

0


2000


Month
Fig. 2-2. Monthly activity patterns of all striped newt captures at One Shot Pond, Putnam Co., FL.
Adult, recently transformed efts, and recently transformed paedomorphs are included.










1800

1600 -

1400 -

S1200 -

S1000 -
0

S800 -
m-

600 -

400 -

200 -
S-


1000


800


600



400


200


11 I I I I I I I I I I I I I I I I I



Month
Fig. 2-3. Captures of adult female and male striped newts at One Shot Pond, Putnam
Co., FL. A) Immigrating adults. B) Emigrating adults. Note differences in scales along
the Y-axes. Four distinct periods of immigration (I) and emigration (E) are indicated.


A
1 Males
Females







I-3





I-1 1-2 I-4


B











E-4


E-2
E-1 E-3
_ ~ ~ ~ = ___ Ea________11 020 B lM W1111 ----- -B W_ -,-- --- W I W -, --


\J


4 1 #/ #/ / l / -. f 1. < \ I ; t 11 J. .1! 0''
o9' ,i ,'b,4 V Y )







1800


I _


1600 m AUUIts
i- ;1 Paedomorphs
1400 Efts
/ //


C 800 E-4


0
600-


SE-1
400-
E-2


200

E-3
0
S' ","" '1 %sJ\ ,q"


Month
Fig. 2-4. Monthly captures of adults, recently transformed efts, and recently transformed paedomorphs
emigrating from One Shot Pond, Putnam Co., FL. Four distinct periods of newt captures are indicated.


-A i1~~1











* Males
o Females


so


*0


0 0 00


0*


400 420
Days at large


440


Fig. 2-5. Net increase in SVL of female (n = 24) and male (n = 16) efts since initial capture during emigration,
shortly following metamorphosis, and recapture during immigration when they returned to breed. Note X-axis
starts at 340 days and some symbols overlap.


12 -


10


00


00


0 *



0


n


-j


340
340


360


380
380


460


480







300
A
250 Males
[Fi Females

- 200


., 150
0
o
100
z

50 -


0 o -__n _i I i n In Jn. _1 n _
0
20 25 30 35 40 4:


140

B
120


100
-o
1 80
.<-
0
$ 60


Z 40 -


20 -

0 ..n .n II L n
20 25 30 35 40 4
Snout-vent length (mm)
Fig. 2-6. Snout-vent lengths of adult female and male striped newts captured
One Shot Pond, Putnam Co., FL. A) Immigrating adults. B) Emigrating adults.


5


5





43


600

SMales A
500 Females


g 400


4. 300
0
o


S200
Z


100


0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

160

B
140 -

120

100

4 80 -
0
a 60

z
40

20 -
20 mi in


0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

Live Mass (g)
Fig. 2-7. Live mass of adult female and male striped newts captured at One
Shot Pond, Putnam Co., FL. A) Immigrating adults. B) Emigrating adults.









I


I I I I I I I I I I I I I I I I I I I I I I I I





Month


Fig. 2-8. Monthly rainfall recorded at One Shot Pond, Putnam Co., FL.


300




250 -




200 -




150-
a-
1M.



100 -




50 -




S-


v







300


280
260 -
240 -
220 -
200 -
S180-
- 160 -
140 -
-3
o 120
100
80
60
40
20

15 20 25 30 35 40 45 50

^ ~*P ^ ^<^^
^9 ^ ^^ ^^ -^0


55 60 65 70 75 80 85 90 95 100

4 ^ ^> ^ ^


Week (number along X-axis) and Month


Fig. 2-9. Depth of One Shot Pond, Putnam Co., FL at the end of each week. Pond depth was not record until week 15.












CHAPTER 3
ORIENTATION AND DISPERSAL DISTANCES OF STRIPED NEWTS AT A
NORTH-CENTRAL FLORIDA BREEDING POND


Introduction

During the past two decades, amphibian declines have received considerable

attention (Alford and Richards, 1999; Barinaga, 1990; Wake, 1991; Wake et al., 1991).

Although pathogens have been implicated in several die-off events (Berger et al., 1998;

Lips, 1998, 1999), there is a consensus among herpetologists that the global decline is a

result of multiple factors (Alford and Richards, 1999). Habitat modification and

destruction have been identified as significant factors contributing to the global decline

(Alford and Richards, 1999; Dodd, 1997; Duellman, 1999; Semlitsch, 2000). Although

they do not attract the media attention that mass mortality or deformed amphibians

receive, habitat modification and loss are insidious processes that must be addressed if

amphibians are to persist. The effects of habitat changes on amphibian populations are of

particular concern in areas that are characterized by a high density of small, isolated

wetlands (Alford and Richards, 1999; Babbitt and Tanner, 2000; Delis et al., 1996; Dodd,

1997; Greenberg, 2001; Hecnar and M'Closkey, 1996; Knutson et al., 1999; Knutson et

al., 2000; Semlitsch, 2000; Snodgrass et al., 2000). In these areas (e.g., the Southeastern

Coastal Plain of North America), amphibian diversity is high (Duellman and Sweet,

1999) and many species rely solely on small, isolated wetlands as breeding sites (Babbitt

and Tanner, 2000; Dodd, 1997; Semlitsch and Bodie, 1998).







Despite their size (i.e., less than a few hectares), small, isolated wetlands are of

tremendous biological importance, particularly for amphibians. In the Southeastern

Coastal Plain, for example, these wetlands support a rich diversity of amphibian species,

and several ponds that have been studied in detail were found to produce thousands of

metamorphic individuals (Dodd, 1992; Gibbons and Semlitsch, 1981; K. Greenberg, pers.

comm.; Hart and Newman, 1995; Johnson, 1999; R. Means, pers. comm.; Moler and

Franz, 1988; Semlitsch et al., 1996; Semlitsch and Bodie, 1998). Small, isolated

wetlands likely play a vital role in amphibian metapopulation dynamics, and therefore are

essential in maintaining viable populations of amphibians at a landscape level (Semlitsch

and Bodie, 1998; Semlitsch, 2000; Snodgrass et al., 2000). In addition to amphibians,

numerous other vertebrates and a suite of invertebrate species depend on small, isolated

wetlands (Brown et al., 1990; Burke and Gibbons, 1995; Hart and Newman, 1995; Moler

and Franz, 1988; Semlitsch and Bodie, 1998).

Preserving a wetland alone may not result in protection of many of the organisms

that depend upon the wetland. Many amphibians have complex life cycles in which they

require ponds to breed but spend the majority of their lives in surrounding upland habitats

(Dodd, 1997; Dodd and Cade, 1998; Semlitsch, 1998). If sufficient upland habitat

surrounding isolated breeding-ponds is not preserved, amphibians with complex life

cycles are not likely to persist at a local scale. Therefore, at some point the loss of

uplands may lead to extirpation of some amphibian populations because of disruption of

metapopulation dynamics (Semlitsch and Bodie, 1998; Semlitsch, 2000), even when the

ponds themselves are preserved.







One strategy to curtail the loss of amphibians associated with habitat alteration

around small, isolated wetlands is to preserve terrestrial "core zones" of upland habitat

surrounding the ponds. These zones provide habitat for retreats and foraging for those

species with complex life cycles, many of which are now considered common. Without

preservation of appropriate upland habitat, even common species will decline.

Little is known, however, about the extent of upland "core areas" required by

pond-breeding amphibians. Dodd (1996) summarized the literature on upland

movements of amphibians in North America and found that this life stage is poorly

known. From this summary and a review by Semlitsch (1998) on dispersal distances of

ambystomatid salamanders, it is apparent that many amphibians disperse considerable

distances from breeding ponds. Unfortunately, dispersal distances are only available for

a few species and usually are based on a single or a few individuals. Clearly there is need

for data on dispersal distances from breeding sites for most North American amphibians.

These data are essential to justify establishing adequate "core areas" of upland habitat

around amphibian breeding ponds.

I collected data on orientation and dispersal distances for striped newts

(Notopthalmus perstriatus) at a breeding pond and in the surrounding uplands in north-

central Florida. The striped newt breeds exclusively in small, isolated wetlands that lack

fish. It has a complex life cycle and individuals spend much of their lives in uplands

surrounding breeding ponds (Fig. 1-1; Carr, 1940; Christman and Means, 1992; Dodd

and LaClaire, 1995; Dodd et al., in press; Franz and Smith, 1999; Chapter 2). Striped

newts are restricted to xeric uplands (i.e., sandhill and scrub communities) in southern

Georgia and northern Florida, U.S.A. (Fig. 3-1). The species has declined throughout its







range (Dodd and LaClaire, 1995; Franz and Smith, 1999) and its biological status is

under review by the U.S. Fish and Wildlife Service (L. LaClaire, pers. comm.). The

objectives of my study were to 1) determine orientation patterns of striped newts into and

away from a breeding pond and 2) determine dispersal distances of individuals into the

surrounding upland habitat.


Methods

Study Site

The study was conducted on the Katharine Ordway Preserve-Swisher Memorial

Sanctuary, Putnam Co., FL (2941'N, 8200'W; Fig. 2-1). Eisenberg and Franz (1995),

LaClaire (1995), and Dodd (1996) provide descriptions of the Preserve and its habitats.

Data were collected from 7 Oct-96 to 11 Sep-98 at One Shot Pond (OSP). One Shot

Pond is a small, isolated pond with a variable hydroperiod (hydroperiod refers to the

number of days a pond holds water between periods when it is dry) and is located in xeric

sandhill uplands dominated by longleaf pine (Pinus palustris), turkey oak (Quercus

laevis), and wiregrass (Aristida beyrichiana). A small pine plantation (Pinus elliottii) is

located west of the pond basin (Fig. 3-2). Several water bodies are located near OSP

(Fig. 3-2). These water bodies are isolated from one another and only receive water from

rainfall and ground water seepage; their hydroperiods are dictated by fluctuations in the

water table. Fox Pond held water from 26 Nov-97 until the end of the study, whereas

OSP, Berry Pond, and the Anderson Cue Lakes held water throughout the entire study

period. During the study, striped newts were only present in OSP and Fox Pond.

However, only 32 newts (16 adults and 16 juveniles) were captured at Fox Pond

(Johnson, 1999). The Anderson Cue lakes support predatory fishes, and striped newts do







not breed there. No striped newts were captured during periodic sampling throughout the

study period in Berry Pond. Because there were no other breeding ponds within several

kilometers of OSP, I assumed that striped newts caught in upland fences around OSP

originated from within OSP.

Orientation at One Shot Pond

I encircled OSP with a 190-m drift fence made of galvanized metal flashing that

was buried ca. 15 cm below the ground, with ca. 35 cm extending above the ground.

Thirty-eight pitfall traps (19 1 plastic buckets) were buried flush with the ground. Pitfall

traps were placed in pairs, one on each side of the fence, at intervals of about 10 m. I

usually checked traps three to five days per week, depending on weather and movements

of animals. I weighed and measured newts caught in pitfall traps at the pond and in the

surrounding uplands (Chapter 2). Each newt was individually marked by toe clipping

(Donnelly et al., 1994) and released on the opposite side of the fence. Sex of adults was

determined by the presence of a conspicuous whitish gland visible at the posterior edge of

the vent in mature males. Recently transformed newts were recognized by the presence

of gill vestiges visible for several days after metamorphosis. Recently transformed newts

with swollen vents were presumed to be mature (Chapter 4), and aquatic sampling in the

pond showed that such individuals represent paedomorphic animals that recently bred.

I obtained a compass orientation for each pair of pitfall traps surrounding OSP.

To do this, I stood in the center of the pond and took a bearing on each pair of traps at the

drift fence. Following the methods of Dodd and Cade (1998), I used Rao's spacing test

(Batchelet, 1981; Rao 1976) to determine if captures were distributed uniformly around

the drift fence (i.e., random orientation). I analyzed orientation of newts into and away

from the pond by sex and life history stage (paedomorph vs. metamorph; Table 2-1). I







made comparisons between distinct migration events (Chapter 2) within the adult and eft

life history stages. For comparisons between sexes, life history stages, and migration

events, I ran the same multirepsonse permutation procedure (MRPP, Mielke et al., in

press) used by Dodd and Cade (1998). Orientation analyses were performed with the

statistical software package BLOSSOM, which was developed by the U.S. Geological

Survey (Cade and Richards, 2000). BLOSSOM is available free at

www.mesc.usgs.gov/blossom/blossom.html.

Upland Dispersal

Dispersal distances of newts in the sandhill uplands around OSP were determined

through captures in pitfall traps associated with drift fences. Drift fences were oriented in

such a manner as to capture newts during movements to and from the pond (Fig. 3-3). In

year one, five fence sections were established at each of four distances from OSP (20 m,

40 m, 80 m, and 160 m). Fence sections at each distance totaled 20% of the

circumference at that distance from the pond. Fence sections were distributed evenly at

each distance, and they did not overlap with fence sections at the other distances (Fig. 3-

3A). Fence sections at 20 m were 10.0 m long with 4 pitfalls (2 on each side of the

fence); at 40 m, fence sections were 15.1 m with 6 pitfalls; at 80 m, sections were 25.1 m

with 8 pitfalls; at 160 m, sections were 45.2 m with 10 pitfalls. Pitfall traps were

installed on both sides of the upland fences (i.e., pond side and upland side; Fig. 3-3A).

This upland fence array was monitored from 7 Oct-96 to 5 Dec-97, and fences were

constructed similarly to the fence at the pond.

Results from year one demonstrated that striped newts regularly dispersed more

than 160 m. Therefore, a new upland fence array was installed in year two, with upland

drift fences erected much farther away from OSP. On 5 Dec-97, the upland drift fences







described above were replaced with a different array of fence sections (Fig. 3-3B) and the

new fences were in place by 7 Dec-97. These fences were constructed of heavy-gauge

silt-fence material buried ca. 15 cm into the ground; ca. 40 cm extended above ground.

Two fence sections were installed at each of five distances (100 m, 200 m, 300 m, 400 m,

and 500 m) from the pond. Fence sections at each distance totaled 13.4% of the

circumference at that distance from the pond, and fence sections overlapped (Fig. 3-3B).

The two fence sections at 100 m were each 42 m long with 6 pitfalls (3 on each side of

the fence) installed evenly throughout each section; at 200 m, sections were 84 m with 10

pitfalls; at 300 m, sections were 126 m long with 14 pitfalls; at 400 m, sections were 168

m long with 18 pitfalls; at 500 m, sections were 210 m long with 22 pitfalls. Pitfall traps

were oriented in the same manner as year one; pond-side traps were on the side of the

fences toward OSP and upland-side traps were away from OSP (Fig. 3-3B). The upland

fence array in year two was monitored until the study ended on 11 Sep-98.

In total, 280 pitfall traps were installed at upland fence sections and were

monitored during the 2-year study, for a total of 98,140 trap-nights (i.e., one trap-night =

one pitfall trap open for 24 hours). Upland traps were checked on the same schedule as

those at the pond and newts were processed as described above.

I estimated the proportion of the newt population that dispersed different

distances from the pond based on captures at upland fence sections and at the outside of

the drift fence encircling OSP. Data used in the estimates were confined to 7 Dec-97

through 31 Mar-98. During this period, there was a mass migration of newts toward the

pond and very little movement away from the pond (Chapter 2). Ninety-one percent of

upland fence captures during year two occurred during this period. These captures,







however, only represented newts that migrated through a subset of surrounding uplands.

Because upland drift fences sampled only 13.4% of the uplands at each distance, I

multiplied the number of captures in the outside pitfalls by 7.5. The product of this

calculation is an estimate of the number of captures expected at each distance had the

upland fence sections sampled 100% of the uplands at each distance. For each upland

fence section, the estimate was divided by the number of total newt captures on the

outside of the fence at OSP to approximate the proportion of individuals that had

dispersed various distances (i.e., 100 m to 500 m, at 100 m intervals). I assumed that

there was no strong nonrandom orientation of newts moving through the uplands.

Nonetheless, movement of newts into and away from the pond was nonrandom, but there

was no overwhelmingly strong directionality that would violate this assumption.

However, estimates of the proportion of newts that had dispersed various distances from

the pond are probably conservative.

I use the term "migration" to indicate seasonal movements of newts toward or

away from the breeding pond. "Immigration" indicates a general pattern of migration

toward the breeding pond, whereas "emigration" indicates migration away from the pond

(Semlitsch and Ryan, 1999).


Results

Orientation at One Shot Pond

All patterns of adult immigration and emigration were significantly nonrandom

(Fig. 3-4; Rao's Spacing Tests, all P < 0.001). Adult striped newts entered and exited the

pond in all directions. They tended to enter the pond basin primarily from the east and

west (Fig. 3-4). Adults emigrated in all directions but there was a single, distinct angle of







emigration, as indicated by the relatively high number of captures in a pitfall trap located

at a south-southeast direction (Fig. 3-4). Emigration ofpaedomorphs and efts also was

nonrandom (Fig. 3-5; Rao's Spacing Tests, both P < 0.001). There was no obvious

pattern to paedomorph emigration, but emigrating efts exited the pond basin most often in

the southwest quadrant (Fig. 3-5).

Overall patterns of immigration differed significantly from emigration for females

and males (Table 3-1). Although the directionality of immigrating adults appeared

similar between the sexes (Fig. 3-4), patterns were significantly different (MRPP test, P =

0.002). There were three distinct immigration events of adults, but orientation patterns

were significantly different between the sexes only during the third, and largest of these

events (Table 3-2). Differences in emigration between males and females (Fig. 3-4) were

not significant overall or when distinct emigration events were compared (Tables 3-1, 3-

2).

There were two distinct emigration events of recently transformed striped newts

comprising the 1996/97 cohort. The first emigration event took place from Oct. through

Nov. 1996, and the second event from Apr. through Jun. 1997 (Chapter 2). Immature

newts (i.e., efts) comprised the first event, whereas emigration later consisted mostly of

recently transformed paedomorphs (Chapters 2 and 4). Patterns of emigration were

significantly different between the eft and paedomorph life-history stages of the same

cohort (Table 3-2). In addition to the eft emigration of 1996, a second emigration event

of efts took place from Jun. through early Sep. 1998 (Chapter 2). Patterns of eft captures

at OSP differed significantly between these two emigration events and, considering all







efts and all adults, efts exited the pond basin in a different pattern from adults (Tables 3-

1, 3-2).

Data for 44 uniquely-marked efts initially caught leaving the pond in the winter of

1996 and recaptured when they returned to breed in the winter of 1997 indicated that

individuals tended to enter and exit the pond within the same quadrant. Sixty-four

percent of these efts left and returned to OSP in the same quadrant (intervals 0 to 3 in Fig.

3-6) and four individuals (9%) were caught leaving and returning to the pond at the same

pair of pitfall traps (interval 0 in Fig. 3-6). The vast majority of individuals (84%)

entered the pond basin within the same half they had left from the previous year

(intervals 0 to 6).

Dispersal Into Uplands

I captured 831 newts in the upland drift fences during year one (Fig. 3-3A, Table

3-3). Pond-side captures accounted for 73% of total captures, and migration in year one

consisted primarily of recently transformed efts. I captured newts at all of the upland

fence sections (Fig. 3-3A; Table 3-3), and in most (91.4%) of the pond-side pitfall traps.

During each period of migration, the vast majority of newts were captured on the same

sides of upland drift fences. However, for most movement events, a small percentage of

newts were captured in pitfalls on the opposite side of fences from the majority of

captures. I believe this is because there was a small degree of wandering by some newts

in the uplands as they moved to or from OSP. Pond-side captures at upland fences in

year one represented three distinct periods of newt migration, two emigration events and

one immigration event (Table 3-4). Most newts captured on the pond-side of upland

fences in year one (76% of pond-side captures) were caught during the first emigration

event (i.e., E-1), which occurred from Oct-96 through Feb-97 (Table 3-4). Emigration







during this period consisted almost exclusively of immature efts that had recently

transformed. I captured far fewer newts (15% of pond-side captures) during emigration

event two (E-2), which occurred from Apr. through Jul. of 1997 (Table 3-4). This

emigration event was comprised of recently transformed paedomorphic newts (54% of

the migrating newts), as well as recently transformed efts and several adults that likely

had finished breeding and were moving back into the uplands. The third period of

migration, indicated by pond-side fence captures in year one, was the result of an

immigration event (i.e., 1-3) that began in Oct-97 (Table 3-4). There was a major

breeding migration of adults to the pond that began in Oct-97 and pond-side captures at

this time probably resulted from adults that were moving toward the pond but happened

to be captured on the pond-side of the upland drift fences (Table 3-4).

Upland-side captures of striped newts accounted for 27% of captures in year one.

I captured newts at each of the five fence sections (Fig. 3-3A) at each distance from OSP

(Table 3-3), and in most (81.4%) of the pitfall traps on the upland-side of the fences in

year one. Upland-side captures occurred during three distinct periods of newt migration,

all of which were immigration events. These migration events (I-1, 1-2, and 1-3; Table 3-

4) occurred during the same time periods as describe above for pond-side captures (Table

3-4). Immigration event 1-3 accounted for the largest proportion (54%) of upland-side

captures in year one, followed by event I-1 (29%) and 1-2 (17%). All of these migration

events consisted of adult newts moving toward OSP to breed (Table 3-4).

I captured 495 newts in the upland drift fences during year two (Fig. 3-3B, Table

3-3). In contrast to year one, migration consisted primarily of immigrating adults. Pond-

side captures accounted for only 9% of total captures. I captured newts at each of the two







fence sections (Fig. 3-3B) at each distance from OSP (Table 3-3), but captures were

recorded in less than half of the pitfall traps (42.8%) on the pond-side of the upland

fences in year two. Pond-side captures at upland fences in year two represented two

distinct periods of newt migration, one immigration event (i.e., 1-3) and one emigration

event (i.e., E-3). I captured few newts during both of these events; 16 during 1-3, and 25

newts during E-3 (Table 3-4). Captures during migration event 1-3 were adults that were

moving to the pond to breed but were captured in pond-side traps as they wandered

toward the pond. Captures during E-3 were recently transformed newts that were

dispersing from OSP.

In year two, I captured far more newts (91% of total upland captures) on the

upland-side of drift fences than on the pond-side (Table 3-3). I captured newts at all

sections of drift fence and in almost all of the upland-side pitfalls (88.6%). Captures

occurred only during a single immigration event (1-3; Table 3-4) and were exclusively of

adults. The number of captures declined as the distance from the pond increased (Table

3-3). Based on estimated values, at least 360 newts (16% of the breeding migration)

dispersed more than 500 m from OSP (Fig. 3-7). I estimated that 645 newts (29% of the

breeding migration) dispersed at least 400 m. The estimate was the same for 300 m (645

newts). I estimated that 810 (36% of the breeding migration) and 908 (41% of the

breeding migration) of newts dispersed from the pond at least 200 and 100 m,

respectively (Fig. 3-7). Based on these estimates, it appears that roughly 60% of the

striped newts emigrated less than 100 m. However, as indicated by captures at the 500 m

fences, a substantial percentage of individuals comprising the 1997/98 breeding

migration immigrated to OSP from farther than 500 m. In fact one newt that was marked







leaving OSP as an eft on 18 Nov-96, was recaptured on 4 Feb-98 as it colonized Fox

Pond, a dispersal distance of ca. 685 m.


Discussion

Orientation

Striped newts of all life history stages (i.e., adults, recently transformed

paedomorphs and efts) emigrated and immigrated in a significantly nonrandom fashion,

although individuals were captured at all pairs of pitfall traps encircling the pond.

Directionality of emigration and immigration differed significantly between and within

sexes, with the exception of emigration patterns between females and males.

Directionality of emigration also differed significantly among life history stages.

Although the percentage of newts dispersing into the uplands decreased as distance from

the pond increased, I estimated that at least 16% of the breeding population dispersed

more than 500 m.

The distribution of habitats surrounding a breeding pond should have a

fundamental influence on patterns of immigration revealed by captures of salamanders at

the pond. Habitat preferences among species and/or differential survivorship in various

habitat types might be apparent as individuals arrive at the breeding pond. For example,

imagine an amphibian breeding pond in which one half of the uplands surrounding the

pond were pine plantation while the other half remained native uplands. The pattern of

captures at the pond would be expected to reflect the distribution of upland habitats. One

would predict significantly fewer captures along the half of the pond adjacent to the pine

plantation as compared to the native upland half. This is because pond-breeding

salamanders have the ability to select appropriate upland habitats and accurately navigate







through uplands during migration, often using specific habitats (deMaynadier and Hunter,

1999; Hurlbert, 1969; Madison, 1997; Madison and Farrand, 1998; Semlitsch, 1981;

Stenhouse, 1985; Shoop, 1968).

In this study, although newts entered and exited the pond basin from all

directions, migration was nonrandom. Some directions were preferred over others, but

there were no obvious upland habitat features that could explain the newts' orientation

behavior. However, I did not measure habitat variables in the uplands and individuals

could have used micro-topographic features as cues to navigate toward the pond. In a

similar study, Dodd and Cade (1998) concluded that movements of striped newts and

narrowmouth toads were a reflection of the distribution of favorable upland habitats

around the pond. Although the uplands at OSP were primarily sandhill habitat, a small

plantation of slash pine (with intact groundcover) was well within the dispersal

capabilities of migrating newts (Fig. 3-2). In year one I often caught newts at a section of

drift fence in the pine plantation. Newts could have resided within the plantation or have

traveled through it en route to native sandhill. Nevertheless, this plantation represented

only a small portion of the uplands and had no detectable effect on striped newt

movements.

Although upland-habitat preferences and microenvironmental features I did not

measure could have influenced the nonrandom pattern of immigration observed at OSP,

if measured over several seasons, orientation may in fact be random. It is possible that

striped newts are roughly evenly distributed in the uplands around OSP but that only a

portion of the population migrates to the pond during any particular breeding event. If

the portion of individuals moving was not indicative of the whole population, then what







truly should be random orientation would appear as nonrandom because data were

collected for a relatively short time.

Patterns of newt emigration were also nonrandom, and newts exited the pond

basin in all directions. Efts emigrated predominantly in the southwest quadrant of the

pond. The slope of the pond basin was shallowest in this quadrant, and water depth

during metamorphic events could have influenced the behavior of recently transformed

efts as they left the pond. Adults on the other hand, emigrated most often in the south-

southeast portion of the basin. Differences in aquatic habitat preference (e.g., depth)

between adult and immature newts might explain the varying emigration patterns,

although habitat preferences of both life history stages are unknown.

Upland Dispersal

Using upland drift fence arrays in year two, I was able to estimate the percentage

of the striped newt breeding population that migrated different distances (in increments of

100 m) from the pond. Captures at drift fences in the sandhill uplands surrounding OSP

indicated that many striped newts (16%) dispersed more than 500 m from the pond. This

is a conservative estimate because many individual captured in traps closer to the pond

may have dispersed further than indicated by the traps. Captures at the drift fence

surrounding the pond and at upland drift fences at the end of year one showed that a

breeding migration of newts into OSP had begun before the installation of fences for year

two (Chapter 2). Although the proportion of individuals caught at the pond before the

new upland fence array was established was small (7% of the total), some newts already

had moved toward the pond before the upland array was in place. Moreover,

immigrating adults did not arrive at the pond in a random fashion during this breeding

migration. The upland fence arrays in year two were located north and southeast of OSP







and newts were caught at the pond with lowest frequency toward the north. Therefore,

the proportions of the breeding population caught at each distance from the pond in year

two is likely an underestimate of the actual proportion moving to that particular distance.

Many pond-breeding amphibians have complex life-cycles and spend much of

their adult lives in terrestrial habitats away from breeding sites. Distances that

individuals disperse from breeding ponds have been reported for some species (Dodd,

1996; Semlitsch, 1998 and references therein). It is clear that individuals disperse

hundreds of meters from breeding sites into upland habitats, some even thousands of

meters. With few exceptions, however, distance values usually have been presented for

less than 10 individuals per species. The results from my study are the first estimates of

dispersal distances for a breeding population of North American amphibians based on a

substantial sample size.

Conservation Implications

Central to a successful amphibian conservation strategy is the protection of

sufficient breeding and nonbreeding habitat. Studies of amphibian dispersal can provide

the scientific basis for determining directional and distance components that can be used

to establish protected areas around breeding ponds. Brown et al. (1990) used spatial

requirements (i.e., distance dispersed from a wetland), among other data, to recommend

width of "buffer zones" for wildlife protection at wetlands in Florida. Nevertheless, lack

of data for amphibians forced them to use rough estimates for most of the species

considered. Further utility of dispersal distance data can be found in regulations to

protect the flatwoods salamander (Ambystoma cingulatum) which, as a result of severe

population decline (Means et al., 1996), was federally listed as threatened (U.S. Fish and

Wildlife Service, 1999). The U.S. Fish and Wildlife Service restricts specific







silvicultural practices within 450 m of flatwoods salamander ponds. Additionally, only

selective timber harvest at specific times is allowed within a primary radius of 164 m

around breeding ponds (U.S. Fish and Wildlife Service, 1999). The width of the primary

zone was derived from a review of dispersal distances for pond-breeding salamanders of

the genus Ambystoma (Semlitsch, 1998), despite the fact that no data for A. cingulatum

were available. This example underscores the need to determine dispersal distances for

all pond-breeding amphibians. Semlitsch (1998) acknowledged that the core zone

recommended for Ambystoma species may apply to some species of pond-breeding

amphibians, but certainly not all. My data show that recommendations for protecting

terrestrial habitat for ambystomatid salamanders are inadequate for Notophthalmus

perstriatus. Therefore, it is not defensible to extrapolate data across taxa. Clearly, a 164

m protected zone would not protect all of the striped newts breeding at OSP. Based on

extrapolation of dispersal distances revealed by upland drift fences, a protected core zone

extending ca. 1000 m from OSP would likely be needed to encompass all of the newts

that breed there.

Although they have great value as wildlife habitat, small, isolated wetlands in the

United States are afforded little protection from development. Overall, more than 50% of

wetlands have been destroyed by development in the United States (Dahl, 1990), and

much of this loss has been small wetlands. In Florida, a state with an extremely large

number and diversity of wetlands, isolated wetlands less than 0.2 ha receive no protection

from development. This size threshold was adopted by the state's water management

districts "based on a consensus of scientific and regulatory opinion rather than on







biological and hydrological evidence" (Hart and Newman, 1995). Small wetlands are

just as vulnerable at the national level as they are in Florida.

There is strong evidence that protection of core areas of terrestrial habitat

surrounding breeding sites is crucial for persistence of amphibian populations and

species. Data from OSP demonstrate that small, isolated wetlands can support breeding

populations of salamanders that extend hundreds of meters into the surrounding uplands.

Similar studies at other ponds and in different upland types are necessary because data on

upland habitat requirements (quality and quantity) of most amphibian species are lacking.

Without this information, designating terrestrial core habitat to conserve aquatic-breeding

amphibians will largely remain guesswork, with generalizations made from data on

relatively few individuals of a few species. However, unless more protection is afforded

to small, isolated wetlands, arguments to preserve uplands surrounding the wetlands are

irrelevant.










Table 3-1. Overall comparisons of directional orientation patterns for striped newts
entering (immigrating) and leaving (emigrating) One Shot Pond, Putnam Co., FL.
Standardized
Comparison n test statistic P
Immigrating vs. emigrating males 1159,486 -13.317 < 0.001
Immigrating vs. emigrating females 1489, 645 -3.798 0.008
Immigrating males vs. females 1159, 1489 -5.524 0.002
Emmigrating males vs. females 486, 645 -0.437 0.2
Emigrating efts vs. emigrating adults 5008, 1131 -67.639 < 0.001
Emigrating efts vs. emigrating
paedomorphs of the same cohort 745, 407 -9.506 < 0.001






Table 3-2. Comparisons of directional orientation patterns for striped newts entering (immigrating)
and leaving (emigrating) One Shot Pond, Putnam Co., FL.
Standardized
Comparison n test statistic P
Immigrating males vs. immigrating females
Immigration Event 1 23, 13 0.697 0.7
Immigration Event 2 22, 66 -0.130 0.3
Immigration Event 3 1049, 1290 -4.008 0.006
Emigrating males vs. emigrating females
Emigration Event 2 15, 68 0.686 0.7
Emigration Event 3 430, 484 -0.005 0.3
Emigrating efts during metamorphic Event 1 vs.
emigrating efts during metamorphic Event 3 745, 4237 -3.599 0.01






Table 3-3. Numbers of striped newts captured in pitfall traps at drift fence arrays in the sandhill
uplands surrounding One Shot Pond, Putnam, Co., FL. Drift fences were located at
various distances from the pond. See Fig. 3-3 for a depiction of the arrays.
Year 1 Year 2
20m 40m 80m 160m 100m 200m 300m 400m 500 m
Pond-side 140 126 169 172 11 6 10 12 7
Upland-side 79 39 64 42 121 108 86 86 48
Total 219 165 233 214 132 114 96 98 55







Table 3-4. Captures of striped newts in upland fences around One Shot Pond, Putnam Co., FL during distinct
periods of movement.
Side of fences Migration Predominant direction Time period No. inds
of captures event of newt movement of event captured Description
Year 1
Pond-side E-1 away from pond Oct.-96 through Feb.-97 461 Emigrating efts
Pond-side E-2 away from pond Apr.-97 through Jul.-97 91 Primarily emigrating paedomorphs and efts
Pond-side I-3 toward pond Oct.-97 through Dec.-97* 55 Immigrating adults
Upland-side I-1 toward pond Oct.-96 through Jan.-97 65 Immigrating adults, some emigrating efts
Upland-side I-2 toward pond Apr.-97 through Jul.-97 36 Immigrating adults
Upland-side I-3 toward pond Nov.-97 through Dec.-97* 123 Immigrating adults
Year 2
Pond-side I-3 toward pond Dec.-97* through Mar.-98 16 Immigrating adults
Pond-side E-3 away from pond Jun.-98 through Sep.-98 25 Emigrating efts
Upland-side I-3 toward pond Dec.-97* through Mar.-98 449 Immigrating adults
*Fence arrays modified in early Dec.-97



















































Fig. 3-1. Present range of the striped newt. Note the hiatus (?) between the western and
eastern portions of the range. This area may represent a true gap in the species
distribution, rather than an artifact of inadequate survey effort.







Katharine Ordway Preserve


I I

0 1km

Fig. 3-2. The location of One Shot Pond within the Katharine Ordway Preserve, Putnam
Co., FL. The pond is surrounded primarily by sandhill uplands but a small pine
plantation (medium gray) is located to the west of pond. Ponds and lakes are indicated in
black and north is at the top of the figure.








u A'


upland-side A
pitfall traps


pond-side /
pitfall traps




\


.000
iwi


-20m
40m
80m


160m


B


500m

400m


300m --


200m ---

100m -


pond-side upland-side
pitfall traps pitfall traps



. .. .. *

*. .
: :1:
:j: ) J:0
~


One Shot Pond


Fig. 3-3. Upland drift fence arrays around One Shot Pond, Putnam Co., FL. A) Array design in year 1. B) Array design in year 2.
One Shot Pond is depicted as a solid gray circle, and the black circle around it represents the drift fence at the pond.


1 -- 11_11_1_1__








Immigrating males


Immigrating females


Emigrating females

N
70
60
50
40


W E






S


Fig. 3-4. Orientation patterns of immigrating and emigrating striped newt adults
captured in pitfall traps at a drift fence encircling One Shot Pond, Putnam, Co., FL.
Orientation was significantly different from random for all four patterns. The length
of the lines indicate the number of newts entering and exiting the pond basin at
each pitfall trap.


Emigrating males







Emigrating paedomorphs


Emigrating efts


Fig. 3-5. Orientation patterns of emigrating striped newt paedomorphs
and efts captured in pitfall traps at a drift fence encircling One Shot Pond,
Putnam, Co., FL. Orientation was significantly different from random for
both patterns. The length of the lines indicate the number of newts exiting
the pond basin at each pitfall trap.





73

















14
13
12
11 ..
10 -
c 9
~i 8

7
6
5
z
4
3 -

2 :,.
1 : ....... ,*".*,





0 1 2 3 4 5 6 7 8 9

Interval



Fig. 3-6. Distance intervals between locations of initial capture during immigration and
recapture during emigration for individual marked striped newts at One Shot Pond,
Putnam Co., FL. Locations were determined by established pairs of pitfall traps
at a drift fence encircling the pond. Distances between trap pairs (i.e., distance interval)
was approximately 10 m.










2200


2000

S1800

1600 -

4 1400

S1200

1000

^ 800 -

S600

400

200


0 100 200 300 400 500 600 700 800 900 1000

Distance from pond (m)

Fig. 3-7. Estimated numbers of striped newt captures in pitfall traps at drift fences in the sandhill uplands around
One Shot Pond, Putnam Co., FL. Drift fences were located at 100 m intervals up to 500 m from the pond.
The zero point represents captures at a drift fence encircling the pond. See "Methods" for an explanation of how the
"estimated" numbers were calculated.












CHAPTER 4
INFLUENCE OF GROWTH RATE ON LIFE-HISTORY EXPRESSION OF STRIPED
NEWTS


Introduction

The expression of alternative phenotypes as a function of the environment has

been documented for many species of plants and animals (Moran, 1992; Scheiner, 1993;

Steams, 1989; Whiteman 1994; Whiteman et al., 1996). Although the environment can

have a profound influence, an individual's genotype may exert a strong effect on

phenotypic plasticity as well (Scheiner, 1993; Steams, 1989).

Species that exhibit complex life cycles (Wilbur, 1980), such as many

amphibians, are excellent models for studying expression of phenotypic plasticity

(Newman, 1992). In these species there is a distinct ontogenetic change in an

individual's morphology, physiology, and often its habitat, that occurs at metamorphosis.

The larval stage is usually dedicated to feeding and growth, whereas the adult stage

disperses and reproduces (Wilbur, 1980). In anurans, the size of an individual at

metamorphosis and duration of larval period have proven to be extremely variable. A

diversity of environmental factors affect these characters (Alford, 1989; Denver, 1997;

Denver et al., 1998; Kupferberg, 1997; Morin, 1986; Skelly and Werner, 1990; Werner,

1986).

Some salamanders also exhibit complex life cycles, although morphological

changes associated with metamorphosis are not as extreme as in anurans (Wassersug,

1974; Werner, 1986). In addition to those species in which individuals always transform







from an aquatic larval stage into a terrestrial or semiterrestrial stage (i.e., obligate

metamorphosis), there are species in which individuals reproduce while the larval or

branchiate morphology persists. This paedomorphic habit (Gould, 1977) is fixed in those

species that have evolved a simple life cycle. However, in other species, namely some

mole salamanders and newts, paedomorphosis is facultative (Duellman and Trueb, 1986).

In facultatively paedomorphic species (referred to as "paedotypic" by Reilly et al., 1997),

individuals may follow one of two developmental trajectories. Once an individual

reaches a threshold size it may transform into a terrestrial form (i.e., metamorphic

phenotype). Alternatively, it may remain in the aquatic environment and mature while

retaining larval characteristics (i.e., paedomorphic phenotype; Whiteman, 1994). In

facultative species, paedomorphic and metamorphic individuals may be found in the

same pond; furthermore, both phenotypes may result from the same cohort (Chapter 2;

Semlitsch et al., 1990; Winne and Ryan, 2001). The expression of alternative phenotypes

is believed to be environmentally induced and to depend on an interaction between an

individual's genotype and the aquatic environment in which it develops (Whiteman,

1994, 1997).

Experiments, almost exclusively with species ofAmbystoma, have identified a

variety of biotic and abiotic factors that influence expression of the two alternative

phenotypes. These include density of individuals in experimental tanks (Licht, 1992;

Ryan and Semlitsch, 1998; Semlitsch, 1987), presence of fish predators (Jackson and

Semlitsch, 1993), drying regime (i.e., hydroperiod) of tanks (Semlitsch and Gibbons,

1985; Semlitsch, 1987; Semlitsch et al., 1990), and potentially, nongenetic maternal

effects (Licht, 1992; Semlitsch et al., 1990). Because growth rate is implicated as an







important determinant of life history expression by several ecological models (Werner,

1986; Whiteman, 1994; Wilbur and Collins, 1973), the effect of food availability has

been the focus of numerous studies. Results of these studies have varied, depending on

the species of Ambystoma studied and type of food regime (Licht, 1992; Ryan and

Semlitsch, in review; Semlitsch, 1987; Whiteman et al., 1996). Furthermore, studies with

Ambystoma have suggested that the expression of facultative paedomorphosis has a

genetic basis (Harris et al., 1990; Licht, 1992; Semlitsch and Gibbons, 1985; Semlitsch

and Wilbur, 1989; Semlitsch et al., 1990).

Less attention has been given to factors affecting the expression of alternative

life-history pathways in other groups of facultatively paedomorphic salamanders. A

better understanding of the proximate and ultimate causes of this phenomenon can only

be achieved through studies of other species and across populations within species

(Whiteman, 1994). This is especially important considering that paedomorphosis

(facultative and obligate) has evolved numerous times (Duellman and Trueb, 1986;

Griffiths, 1996; Ryan and Bruce, 2000; Shaffer et al., 1991; Shaffer, 1993; Shaffer and

Voss, 1996).

Species in the family Salamandridae (newts) provide an excellent opportunity to

study the evolution and maintenance of facultative paedomorphosis. Newts exhibit a

diversity of reproductive strategies, and facultative paedomorphosis occurs in several

genera (Griffiths, 1996; Halliday, 1990; Petranka, 1998). Within North American newts

(Taricha and Notophthalmus), facultative paedomorphosis is known to occur in Taricha

granulosa (Marangio, 1978), Notophthalmus viridescens (Brandon and Bremer, 1966;

Harris, 1987; Healy, 1970, 1974a), Notophthalmus perstriatus (Dodd et al., in press;







Chapter 2), and possibly in Notophthalmus meridionalis (Mecham, 1968). Nevertheless,

the expression of facultative paedomorphosis has only been experimentally explored in a

single study of Notophthalmus viridescens dorsalis (Harris 1987). Harris (1987) found

that density, which influenced growth rate, had a significant affect on expression of

paedomorphosis. Based on the literature and my research with Notophthalmus

perstriatus, it appears that hypotheses invoked to explain the expression and maintenance

of paedomorphosis may differ between Ambystoma and Notophthalmus species. In fact,

several different hypotheses may explain the maintenance of facultative paedomorphosis

(Whiteman, 1994, 1997). Therefore, results of experiments with Ambystoma may not be

entirely applicable to newts or other salamander species.

Field research I conducted at several breeding ponds in north Florida revealed that

the striped newt (N. perstriatus) has a complex life history (Fig. 1-1). Within a single

breeding season, I found that some larvae transformed before maturing and left the pond

(metamorphs), whereas others remained in the pond and matured while retaining their

larval morphology (paedomorphs). This same pattern appeared to have occurred during

three consecutive breeding seasons. The proximate causes) of the expression of the

paedomorphic phenotype in some individuals but not others is unknown. A growth

advantage for some individuals over others could explain the dichotomy in expression of

the paedomorphic versus metamorphic life-history. Faster growing larvae of other newt

species (N. v. dorsalis, Harris, 1987; Triturus carnifex, Kalezic et al. 1994) were more

likely to become paedomorphic than slower growing larvae.

Some models of amphibian metamorphosis take into account the potential

influence of growth rate on the expression of paedomorphosis in salamanders (Wemer,







1986; Wilbur and Collins, 1973; Whiteman, 1994). In the Wilbur and Collins (1973)

model. "physiological processes that initiate metamorphosis are related to recent growth

history of the individual." Larvae with a growth advantage are predicted to remain in the

aquatic environment and continue to grow, eventually expressing the paedomorphic life-

history pathway. Slow growing larvae are predicted to metamorphose once they have

reached some population-specific or species-specific threshold. Slow growing larvae

thereby escape the potential density-dependent influence of competition with

paedomorphs and larger larvae (Whiteman, 1994; Wilbur and Collins, 1973). Werner's

(1986) model of amphibian metamorphosis also has been extended to paedomorphosis.

In this model, the "decision" of an individual to remain in the aquatic environment or

metamorphose and move into the terrestrial habitat is a result of growth potential in each

habitat weighed against the habitat-specific risk of mortality. All else being equal, larvae

with a growth advantage (i.e., relatively fast growing larvae) should be more likely to

become paedomorphic as compared to slower growing larvae.

In addition to life-history expression, the Wilbur-Collins (1973) model predicts

size and age at metamorphosis. This model can be viewed as a flexible-rate or optimal-

growth rate model because mass-specific growth rates throughout much of the larval

period dictate when metamorphosis should occur and at what body size. An alternative

model, initially proposed by Smith-Gill and Berven (1979), and later extended by Travis

(1984) and Leips and Travis (1994), differs from the Wilbur-Collins model in that rate of

differentiation (i.e., development) of larvae, as opposed to growth per se, dictates larval

period. Therefore, this alternative is referred to as a fixed differentiation-rate model. In

this model, larval period is determined or fixed at some specific point in development.







Differences in growth rate after this stage of development are only predicted to influence

body size at metamorphosis, not larval period. Travis (1984) suggested that this fixation

period occurs early in development. Size at metamorphosis is still predicted to be a result

of larval growth rate however, and because of this, both models account for the extreme

plasticity in metamorphic body size of amphibians. Unlike the optimal growth-rate

model of Wilbur and Collins, the fixed-rate model does not apply to paedomorphosis,

however.

I tested the hypothesis that growth rate influences the expression of

paedomorphosis in striped newts. I used varying food levels to generate different growth

trajectories of larvae raised individually in the lab. The treatments included high and low

levels of food, as well as switches in food levels. The objectives of the experiment were:

1) to determine the influence of growth rate on the expression of alternative life histories

(i.e., metamorph vs. paedomorph), 2) to test the ability of the Wilbur and Collins model

to predict the life-history expression, 3) to determine the influence of growth rate on

larval period and size at metamorphosis, and 4) to simultaneously test the applicability of

two types of models (optimal rate vs. fixed rate) for predicting metamorphosis in

Notophthalmus perstriatus.


Methods

Experimental Design

I used different food treatments (constant High = HH, constant Low = LL, switch

from Low to High = LH, and switch from High to Low = HL) to generate growth

trajectories among four experimental groups of striped newt larvae. I used a randomized

block design, with the two shelves treated as blocks. Ten individuals were randomly







assigned to each of the four treatments. Newts were reared individually in plastic

containers which were randomly assigned a position on one of two shelves. The

experiment was designed to standardize all variables with the exception of food

availability.

Procedures

Newts were housed in Sterilite ClearView storage boxes (43 cm X 28 cm X

16 cm). These were filled with 9 liters of aged water, which was changed every 5-7 days.

During water changes, each container was cleaned with a scrub brush and antibacterial

soap, then rinsed with tap water. Light was supplied by florescent bulbs mounted above

the containers. Automatic timers controlled the lights and were adjusted several times

during the experiment so that the light/dark regime approximated the natural light cycle.

Newts were housed indoors, although the temperature fluctuated daily and seasonally,

tracking the temperature variation for Gainesville, FL during the study period. I weighed

and measured each newt twice a month (every 17 days on average). I weighed and

measured newts singly by first removing an animal from its container and placing it in a

plastic sandwich bag. I measured body length to nearest millimeter with a clear plastic

ruler while the newt was in the bag. Because it was difficult to see the posterior edge of

the vent on small larvae, I measured from the tip of the snout to the anterior edge of the

rear legs where they meet the body (i.e., body length), rather than measure standard

snout-vent length (SVL). Body length is highly correlated with SVL (Spearman

Correlation test: r = 0.988, P < 0.0001, derived from measurements of 30 preserved

striped newt larvae). Each newt was weighed to the nearest 0.1 g by transferring it to a

small plastic cup filled with aged tap water that had already been tared on a Ohaus brand

digital balance. I also recorded the degree of swelling of the vent (not swollen, slightly







swollen, and swollen), condition of the gills and tail fin (full vs. regressing), and other

external characteristics (e.g., presence of secondary sexual characters). These external

characters can be used as indicators of maturation (i.e., vent swelling) and metamorphosis

(i.e., gills and tail fin regress). In addition to using secondary sexual characters, I

dissected each newt and examined its gonads under a dissecting scope to determine if it

was sexually mature.

Newts were derived from eggs laid by paedomorphic adults that I collected on 3

Jan-99 from a breeding pond (Blue Pond) located on the Katharine Ordway Preserve,

Putnam Co., FL. Striped newts are winter-breeders and these individuals were already in

reproductive condition. Forty-eight paedomorphs were housed in 12 glass aquaria (2

females and 2 males per aquarium) and fed black worms every 2-3 days. Each aquarium

contained an air stone, a constant water source, a standpipe, and vegetation on which to

lay eggs. I collected eggs every 7-10 days and transferred them to shallow tubs

containing dechlorinated water. Striped newt females lay eggs one at a time and have a

prolonged breeding season of several months (Chapters 1 and 2). Hatchlings were

transferred to another tub and raised on a diet of zooplankton until they were large

enough to consume whole black worms. Water in the hatchling tubs was changed every

5-7 days. I waited until the larvae were large enough to eat black worms because this

food source was easy to quantify and prior experience had shown that newts thrive on

black worms. Although small larvae would readily eat Artemia, they always died within

2-3 days; they thrived however on native zooplankton. I grew zooplankton in large (1.2

m diameter) plastic (high-density polyethylene) tubs and harvested them with a fine-mesh

net. The tubs were manufactured by the Lerio Co., Kissimmee, FL. The tubs contained







ca. 700 liters of well water, soil collected from a dried pond basin at the Ordway

Preserve, leaf litter, and aquatic vegetation. I also inoculated the tubs several times with

zooplankton collected from local ephemeral ponds. Once enough larvae were reared to a

size capable of consuming black worms, I began the experiment.

The experiment was initiated on 16 Jun-99 and continued until 27 Dec-99, for a

total of 206 days. An individual was removed from its container when it showed obvious

evidence of metamorphosis. This included the appearance of a bold stripe along the

dorso-lateral portions of the body, and regression of the gills and membranous tail fin. A

newt nearing metamorphosis also reduced its food intake and often floated at the surface

of its container. At this point, the newt was transferred to a new container and placed on

a damp paper towel to allow gill regression to continue. Once the gills had been

completely resorbed, usually 2-3 days after removal from the plastic container, the newt

was anesthetized in a chlorotone solution, weighed and measured. Each newt was tagged

with a unique label and preserved in 10% formalin for dissection later. Newts that did

not metamorphose remained in the experiment until its termination on Dec. 27. At this

point, they were anesthetized, weighed, measured, and preserved.

Food Treatments

High food-treatment animals were fed black worms (Lumbriculus) ad libitum; the

Low food-treatment was calculated as of the average number of worms consumed by

the High food-treatment animals during the preceding feeding interval. Low food-

treatment animals were fed every 3-5 days. Therefore, although the number of worms

consumed by individuals in each food treatment increased as the newts grew (i.e.,

approximating mass-specific food levels), the animals on Low food always received of

that of the High food animals. The LH and the HL treatment animals experienced a shift







in available food during the experiment. On 17 Jul-99, I initiated the switch in food

levels. The LH animals, which previously had received of the amount of food as the

High food level animals, now received food ad libitum. The HL animals were now only

fed 4 of the average number of Lumbriculus consumed by High-treatment animals.

High-food newts always had Lumbriculus available in their tanks; Low-food newts

consumed all of their Lumbriculus within a few hours of feeding. The switch was

initiated when mean mass of the HH and HL animals was at least 0.7 g and their body

length was at least 22 mm. These body sizes are greater than the minima required for an

individual striped newt to initiate metamorphosis but below the minima required for

maturation in paedomorphic individuals (Chapter 2). Therefore, the switch in food levels

occurred near the body size where metamorphosis can be initiated in nature but before

maturation occurs. I chose this point to make the food-level switch because it should be a

critical time in development. Once the minimum size for metamorphosis is reached, an

individual may express the metamorphic life-history pathway or postpone metamorphosis

and continue to grow, possibly expressing the paedomorphic pathway later.

Dissections

Internal and external characteristics were used to determine sex and reproductive

condition. External secondary sexual characteristics that indicate maturity in

salamandrids include: swelling of the vent, development of genial glands on the side of

the head of males, and presence of nuptial excrescences on the rear limbs and toe tips of

males (Duellman and Trueb, 1986). Mature striped newts are sexually dimorphic and in

addition to the characters above, the shape of the cloacal lips differ between males and

females. A conspicuous cloacal gland, which is whitish in color, is visible at the

posterior edge of the cloaca in mature and some immature males (Dodd, 1993). I







dissected all newts and examined their urogenital systems under a dissecting scope to

determine reproductive condition. Dissection also allowed me to confirm the sex of

newts that exhibited secondary sexual characteristics, and determine the sex of immature

animals. Maturity of females was indicated by the presence of enlarged ovarian follicles

and enlarged, convoluted oviducts. Maturity in males was indicated by the presence of

enlarged, convoluted, and pigmented Wolffian ducts (Ryan and Semlitsch, in review;

Semlitsch, 1985)

Data Analysis

Response variables used in statistical analyses were body length, body mass,

larval period (number of days from initiation of the experiment to metamorphosis), and

morph type (metamorph or paedomorph). Prior to analysis, each data set was tested for

normality and heteroscedasticity. When assumptions of parametric tests were violated,

nonparametric methods (Hollander and Wolfe, 1999) were used to test for treatment

effects. All statistical analyses, with the exception of goodness-of-fit tests, were

performed using SAS version 6.12 (SAS Institute Inc., 1990).

I used a two-part analysis to test for differences in growth rates among the four

treatment groups. First, a MANOVA (on body length and mass) was used to test for

differences among treatments the day before the switch in food level was initiated.

Second, I tested for differences among growth rates during the remainder of the

experiment. For each individual that remained in the experiment (i.e., had not yet

metamorphosed), I subtracted the size (for BL and mass independently) of the individual

on the most recent day it had been measured (before metamorphosis or at the end of the

experiment for paedomorphs) from its size on the day before the food switch. This value

was then divided by the number of days from the switch until the date a newt had been







most recently measured to give a growth rate (unit of growth/day). Growth rates among

treatments (excluding HL) were then tested with a Kruskal-Wallace test (Hollander and

Wolfe, 1999). Growth rates of HL animals were excluded because all individuals had

metamorphosed before the first scheduled date for measuring individuals following the

food switch.

I used a MANOVA to test for differences among treatments on body length,

mass, and larval period for those individuals that metamorphosed before the end of the

experiment. Following significant univariate tests in MANOVA tests, a Student-

Newman-Keuls (SNK) test was used to compare treatment means. For each MANOVA,

if Wilks' X was significant, alpha was adjusted for univariate tests with the Bonferroni

method (Sokal and Rolf, 1995). I used a goodness-of-fit test (Sokal and Rolf, 1995) to

analyze the different frequencies of metamorphs and paedomorphs across treatments and

between sexes. For influence of treatment on morph type, I calculated the number

expected for each morph in each treatment (expected metamorph:paedomorph =

7.75:2.25) by dividing the total number of metamorphs and paedomorphs by four. For

influence of sex, I partitioned the observed numbers of each morph between the sexes to

calculate expected numbers (expected metamorph:paedomorph = 14:4.5). Goodness-of-

fit was used also to determine if sexes of the newts were distributed equally among the

treatments. Expected values were derived by dividing the total number of males and

females by four (expected male:female = 5.75:3.5)







Results

Larval Growth

On day one of the experiment there were no significant differences in mass

(Kruskal-Wallace test: H= 2.864, df= 3, P = 0.4131) or body length (ANOVA: F3,36 =

0.54, P = 0.6605) of newts across all four food treatments (i.e., HH, HL, LL, LH). On the

day before the switch in food levels (day 31), MANOVA indicated significant differences

in size of newts among treatments even when alpha was adjusted with the Bonferroni

method (Wilks' X = 0.219, F 6,66= 12.53, P < 0.0001). Univariate tests were significant

for both mass (F3,35 = 39.41, P < 0.0001) and body length (F3,35 = 24.61, P < 0.0001);

block was not significant. SNK tests showed that for both body length and mass, the HH

and HL treatment animals did not differ nor did the LL and LH animals. However, HH

and HL newts had significantly greater mass and body length than the LL and LH newts.

Therefore, before the switch in food levels, individuals receiving the H food treatment

had grown faster than individuals fed at the L level (Fig. 4-1).

The switch in food levels was initiated on day 32 (17 Jul-99) of the experiment.

By this point, newts in the HH and HL treatment groups had slightly surpassed the

minimum size required for metamorphosis ofN. perstriatus in nature (Chapter 2),

whereas the LL and LH newts remained smaller than this minimum size. Analysis of

growth rates following the switch day does not include any HL individuals because all of

them metamorphosed before day 48 of the experiment, the first day after the switch when

mass and body length were next measured. After the food switch, the growth rates of HL

newts promptly increased and soon their rate paralleled those of the HH newts, whereas

the LL newts continued to grow slower than individuals receiving the H food level (Fig.

4-1). After the switch in food levels, growth rates, measured both as differences in mass







and body length, differed among remaining branchiates in the HH, LL, and LH

treatments (Kruskal-Wallace test for mass: H= 15.694, df= 2, P = 0.0005; for body

length: H= 15.094, df= 2, P = 0.0004; Fig. 4-1). Overall growth rates in mass and body

length were greatest for LH newts, followed by HH then LL treatment animals (Fig. 4-2).

Size at Metamorphosis and Larval Period

Univariate tests from a MANOVA on mass, BL, and larval period (Wilks' X =

0.119, F 9,66 = 7.57, P < 0.0001) were not significant for either mass at metamorphosis

(F3,22 = 2.71, P = 0.0696) or body length (F3,22 = 0.27, P = 0.8496; Table 4-1); block was

not significant in any of the univariate tests. Larval period on the other hand, was

significantly different among the food treatments (F3,22 = 19.63, P < 0.0001; Table 4-1).

A SNK test showed that the larval period did not differ between the HH and HL groups

but both of these differed from the LL and LH groups, which differed from each other.

The relationships between mass and body length at metamorphosis and larval period is

presented in Fig. 4-3.

Life-history Pathway Expression

At the termination of the experiment, 22.5% of the individuals were mature

branchiates and the rest had metamorphosed earlier in the experiment before attaining

maturity. All newts considered paedomorphs were mature based on external and internal

characters. Paedomorphic females each had many enlarged, pigmented follicles in their

ovaries, as well as enlarged and convoluted oviducts. All paedomorphic females also had

swollen vents. The Wolffian ducts (i.e., vas deferens) of all paedomorphic males were

enlarged, pigmented, and convoluted. Externally, paedomorphic males had very swollen

vents, relatively enlarged rear limbs with comifications on their toe tips, and well-

developed hedonic pits on each side of their head. A light colored gland was obvious at







the posterior end of their vents. None of the metamorphic newts had near the degree of

cloacal swelling exhibited by the paedomorphs. The rear limbs of metamorphic males

were not enlarged and their toe tips were not cornified; hedonic pits were barely visible in

only a few immature males. Internally, neither the Wolffian ducts nor the oviducts were

enlarged or convoluted in metamorphic newts. Follicles in the ovaries of metamorphic

females were small and undeveloped. External gills were present in all paedomorphic

newts at the end of the experiment, and the namesake stripe was well developed.

External gills of experimental paedomorphs were not as large and filamentous as gills of

paedomorphic striped newts observed in natural ponds (S. A. Johnson, pers. obs.).

Distribution of paedomorphs and metamorphs was not even across treatments,

and no paedomorphs were produced in the HL treatment (Fig. 4-4). Nevertheless, a

goodness-of-fit test did not detect significant differences in life history expression across

the food treatments (G = 6.968, df = 3, P = 0.062). However, as compared to the overall

sex ratio of metamorphs and paedomorphs in the experiment (both morphs, M:F = 1:0.6),

significantly more females became paedomorphic (females, paedomorph:metamorph =

1:1.34) than males (males, p:m = 1:6.7; G = 6.333, df = 2, P = 0.042 ). Males and

females were distributed equally among treatments (G = 1.664, df = 3, P = 0.680).

By the end of the experiment, all paedomorphs were longer than the metamorphs

and all but one had greater mass than the metamorphs (Fig. 4-5). The body size of

paedomorphs at the end of the experiment differed among the three treatments in which

paedomorphs were produced. Mean mass and body length were greatest in paedomorphs

produced by the HH food treatment, followed by paedomorphs in the LH then the LL

treatments (Table 4-2).







Discussion

Differences in food levels resulted in differences in growth rates among

treatments before and after the food-level switch (Figs. 4-1 and 4-2). Despite these

differences, mass and body length were not significantly different at metamorphosis for

those newts expressing this life-history pathway (Fig. 4-3, Table 4-1). However,

treatment had a significant effect on duration of larval period. Mean larval period of HH

newts did not differ significantly from that of HL newts, but both of these were

significantly different from the LL and LH treatment individuals (Fig. 4-3, Table 4-1),

which were significantly different from each other. The paedomorphic life-history

pathway was expressed by newts in all treatment groups except the HL group (Fig. 4-4).

Nonetheless, the proportion of paedomorphs and metamorphs resulting from each

treatment did not differ significantly. Therefore, growth trajectories generated by the

four different food regimes did not affect life-history pathway expression. There were

however, significant differences in the expression of life-history pathway among the

sexes. Proportionately more females became paedomorphic than expected considering

the overall sex ratio of animals in the experiment.

Expression of Alternative Life-history Pathways

Optimal growth-rate models, primarily the Wilbur-Collins (1973) and the Werner

(1986) models, make predictions about the expression of life-history pathways in

salamanders. Growth rate is also central to Whiteman's (1994) three alternative

hypotheses to explain the maintenance of facultative paedomorphosis. Each of these

models essentially makes the same predictions with regard to life-history pathway

expression that should have resulted from treatments in the experiment. High-High and

LH newts should have expressed the paedomorphic phenotype, whereas LL and HL







newts should have transformed before maturing. However, the results for life-history

pathway expression clearly did not fit these predictions.

I found no significant influence of treatment, and thus growth trajectory, on life-

history pathway expression. As predicted by the Wilbur-Collins model, HL newts

transformed shortly after they were switched from a High to Low food ration and none

became paedomorphic. However, all but two of the HH newts also transformed shortly

after the switch, and there was not a significant difference in larval period between newts

in the two treatments. No paedomorphs were predicted to result from the LL treatment.

Nevertheless there were more paedomorphs in this treatment than any of the other three

(Fig. 4-4). The LH treatment was predicted to produce only paedomorphs but seven

larvae in this treatment transformed. To reiterate, food level did not influence an

individual's expression of life-history pathway.

The role of food level on life-history expression has been tested in mole

salamanders. Constant food levels did not significantly influence the expression of

paedomorphosis in A. gracile or A. talpoideum (Licht, 1992; Semlitsch, 1987).

Additional experiments with A. talpoideum revealed a significant effect of food level for

individuals that received relatively low food rations late in development (Ryan and

Semlitsch, in review), counter to predictions of the Wilbur-Collins model.

Life-history expression of striped newts depended on sex in my experiment.

Females had a significantly greater propensity to become paedomorphic than males. The

sex ratio of paedomorphs was female biased (m:f = 1:2), whereas the sex ratio of

metamorphs was male biased (m:f = 5:2). Female-biased sex ratios in paedomorphs have




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EECXT5RNM_8SR3DS INGEST_TIME 2013-04-08T20:14:47Z PACKAGE AA00013531_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EKUJN23G3_A3F8MK INGEST_TIME 2013-03-12T13:32:03Z PACKAGE AA00013531_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

LIFE HISTORY ECOLOGY, AND CONSERVATION GENETICS O F THE STRIPED NEWT (Notophthalmus perstriatu s ) By STEVE A. JOHNSON 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 2001

PAGE 2

ACKNOWLEDGMENTS First and foremost I would like to express my deepest appreciation to my wife Dale for enduring the stresses of a husband enrolled in graduate school for many years. Dale has been a constant source of encouragement and intellectual stimulation throughout my graduate research. She has opened my mind to new ideas allowed me to see and appreciate beauty that previously I was blind to and taught me what is important in life. Without her support I would not have been able to complete my graduate education. Dale drafted many of the maps and figures in this dissertation. I also thank my parents my mother Bobbie W. Johnson and my late father Gordon E. Johnson for encouraging the biological interests of their son through putting up with all the messes I made as a child. There are numerous friends and colleagues who came to my aid throughout my research projects and I deeply appreciate their assistance. These folks helped me collect samples install and remove drift-fence arrays check traps and tend experimental animals. I would like to thank the following individuals for their help in this regard: Brad Austin Mark Bailey Jamie Barichivich Bobby Bass Laura Becht Boyd Blihovde Cheryl Cheshire Ken Dodd Brian Emanuel Dick Franz Dan Ripes John Jensen Dale John s on Kenney Krysko Ryan Means Paul Moler Bubba Owen David Printiss Rob Robbins Matt eguin Joe exton Parks mall Lora Smith Jennifer Staiger Dirk teven on and Chad Truxall. 11

PAGE 3

For help with statistical analyses and data interpretation I would lik e to thank : Anna Bass Brian Bowen Brian Cade Ginger Clark and Julie Heath. For providin g constructive criticism on earlier drafts of my dissertation I thank Alicia Francisco, Holly Freifeld, Dale Johnson David Leonard, and the members of my graduate committee. For assistance in the lab working on my genetics project I thank: Anna Bass Ginger Clark and Alicia Francisco. I would also like to express my gratitude to staff in the Department of Wildlife Ecology and Conservation for all of their help during my graduate studies at the University of Florida. In particular I would like to thank Laura Hayes Monica Lindberg Caprice MacRae Sam Jones and Cynthia Sain for their kind assistance throughout my graduate career. I thank Kent Vliet and John Reiskind for providing me with teaching assistantships through the Biological Sciences Program. I also thank the clerical staff, Kenetha Johnson and Tangelyn Mitchell, for their help while I served as a TA. Funding for my dissertation projects was provided by the U.S. Fish and Wildlife Service the Gopher Tortoise Council, and the Florida Fish and Wildlife Conservation Commission. I am especially grateful to Linda LaClaire of the U.S. Fish and Wildlife Service for administering my grants from this agency. The Lerio Corporation BEECS Genetics Analysis Core at the University of Florida, and U.S. Geological Survey donated materials and provided lab space. In particular, I acknowledge Russ Hall of the U.S. Geological Survey for providing me with space to rear experimental animals I would also like to express my gratitude to the governing board of the Katharine Ordway Preserve / Swisher Memorial Sanctuary for allowing me to conduct field research on the lll

PAGE 4

Preserve. I am grateful to John Eisenberg Dick Franz and Mel Sunquist for facilitating my work on the Preserve. For permission to collect newt tissue samples at Ichauway I thank Lindsey Boring of the Joseph Jones Ecological Research Center Newton GA. In Georgia samples were collected under Georgia DNR scientific collecting Permit # 00335. I thank Parks Small for facilitating sample collection at Rock Springs Run State Preserve and the Florida Division of Forestry for permission to collect at Jennings State Forest. I would like to acknowledge the support of the members of my graduate committee: C. Kenneth Dodd Jr. (committee chair) Dick Franz Brian Bowen Mark Brenner George Tanner and Mike Moulton for their guidance and assistance. I extend special thanks to Dick Franz for everything he has done for me. Dick always looked out for my best interests and I am most appreciative. Dick was also the person who gave me my first experience with striped newts. Finally, I extend my most sincere thanks to all the great friends that Dale and I have had the pleasure of spending time with during our stay in Gainesville. It was certainly the highlight of our years here. lV

PAGE 5

TABLE OF CONTENTS ACKNOWLEDGMENTS .................................................................................................. ii ABSTRACT ..................................................................................................................... viii CHAPTER 1 INTRODUCTION TO THE STRIPED NEWT (Notophthalmus perstriatus) .................... 1 The Striped Newt ..................................................................................... ...................... 1 Status of the Striped Newt .............................................................................................. 1 Current Knowledge and Research Justification .............................................................. 2 Striped Newt Life History and Life-history Stage Terminology .................................... 3 Larvae ......................................................................................................................... 4 Efts .............................................................................................................................. 4 Paedomorphs .............................................................................................................. 4 Adults .......................................................................................................................... 5 Overview of Dissertation ................................................................................................ 5 CHAPTER2 LIFE HISTORY OF THE STRIPED NEWT AT A NORTH-CENTRAL FLORIDA BREEDING POND ............................................................................................................. 9 Introduction ................................................................................................................. ... 9 Materials and Methods.................................................................................................. 10 Study Site ............................................................................................... .................. 10 Drift Fence at One Shot Pond ............................................................................ .. .. .. 11 Newts Caught at Drift Fences ................................................................................... 12 Weather Data ........................................................................................................... 13 Statistical Analyses ....................................................................................... ........... 13 Results ..................................................................................................................... .... 13 Seasonal Activity .................................................................................... .. ... .. .. ... .... 13 Immigration ...................................................................................... ............... .. .. .. 14 Emigration ..................................................................................................... ... . .. . . 14 Reproduction ........................................................................................................... .. 15 Population Size Structure .......................................... ........................................... .. 1 7 Sex Ratios ......... ................................................................................................... .. 18 Rainfall and Hydroperiod ......................................................................................... 19 Discussion .............................................................. ...................................................... 20 Seasonal Activity ................................................................... .... .... .... .... .... .. .. .. . 2 0 Immigration ................................................ ............................................... ............. 2 1 V

PAGE 6

Emigration ...... ..................................................... ................................ .. ......... .. ..... 21 Reproduction .................................................................................. ....................... .. 22 Population Size Structure ................................................................................ ......... 24 Sex Ratios ........ ....................................................................................................... 24 Hydroperiod and Rainfall ..... ................. .......... ..................................................... 25 Comparisons with Notophthalmus viridescens ........................................ ..... ......... 26 Implications for Striped Newt Status Surveys . ......................... .. ................ .. . .... 29 CHAPTER3 ORJENTATION AND DISPERSAL DISTANCES OF STRIPED NEWTS AT A NORTH-CENTRAL FLORIDA BREEDING POND ......................................... ............ .46 Introduction ............ ................................................. . ...................................... .. ... .... 46 Methods .. ...................................................................................................................... 49 tudy Site ........................................................................................... ................. .... 49 Orientation at One Shot Pond ......... ......................................................................... 50 Upland Dispersal ............................................................................. ... ................. . .. 51 Results ............................ .................................................................................. ....... ... 53 Orientation at One Shot Pond ........ ...................................................................... .. .. 53 Dispersal Into Uplands ................................................................. ...................... ..... 55 Discussion ............................................................................ ......................... ........... . 5 8 Orientation ....... ................................. ......................................... ...................... ..... 5 8 Upland Dispersal ...................................... ........................................ .. ... .. .. .. ....... 60 Conservation Implications ............................... .. ..................... ... .. .. ................. .. . 61 CHAPTER4 INFLUENCE OF GROWTH RA TE ON LIFE-HISTORY EXPRESSION OF STRIPED NEWTS ..................... .. ... .. ..................... ................ ........................ ............................... 75 Introduction .................................................................................... ........... ...... .. . .. . . 75 Methods ...................... ................................................................... ....................... .... . 80 Experimental Design ... .......................................... .... ............................. .. .. .. . .. .. .. 80 Procedures ....... .... ................................ .......... ............... .................................... ... 81 Food Treatments ............. ............... .. ... .............. .................. .. .......... ......... .. . .. 83 Dissections ... ... ... .... ... ....... .... ....... ....... .. .. .. .. ................................ .. ...... ..... .. . 84 Data Analysis . ...... ....... ...................................... .. ............................. ... ..... .. .. ..... 85 Results ..................... .. ........................... .. ............ .. .. . ................................................... 87 Larval Growth .. .................... .. ...... .. ................................... .. .... ........... ... ..... .. . .. . 87 ize at Metamorphosis and Larval Period .. ................ ...... .. ... .. ... .. .. .... ................... 88 Life-hi s tory Pathway Ex pression . ............. .. .. ... .. .. ............... .. .. ....... ... ... .............. 88 Di s cu ss ion . ... . ..... .... . .... .. ........ .. .. . ....... .. .... . .. .. ...... .. .. . .... ........ .. .. .. .. .. ...... .. . .. . .. 90 x pr ess ion of Alternati v e Life-history Pathway s ... .. ........... ... ... . .. .......... .. ............ 90 M e tamorpho s is and Model Applicability .......... .. .. ...... ....... ... .. ........ . ... ......... .. .. 94 HAP TE R 5 ON RVATION GENETICS AND PHYLOGEOGRAPHY OF THE STRIPED N W T ... .. ........ .. ... ... ... ... ... .. ....... . ... ..... . ..... .. .. .. .... . ..... . .. . .............................. ......... 106 Introdu c tion .. .. ... .. .. . . . ...... .... ........ ... ... .. ... . .. . . ..... ......... . . .... ... .. .. .. ...... .............. 106 Vl

PAGE 7

Methods ................... ...... .. ... ..... ... .. .. .. .. .. .. ..... .. .. .. .. .. .. .. .. .. .. .. .. ...... .... .......... .. .... 108 Sample Collection .................................. ..... .................... .. .. ... ....... ... .. .. .. ......... . 108 DNA Isolation and Sequencing ......... .. . ..... ..... .. ............... . .. ... ...... ... .. .. .. .. .......... 109 Data Analysis ........................................ ................................... ....... ....... ............ 110 Results .............. .............................................. ..... ..................................................... 111 Discussion ......................................... ...................................................... ... .. .. .. ....... 113 Population Structure ... ........................ ................ .. ....... .. .... ............. ......... .. ..... 113 Testing a Bio geographic Hypothesis ......................................... ................... ... .. .. 11 5 Striped Newt Biogeography and Phylogeography ..................... ............. .. .. .. .. .. ... 116 Conservation and Management Implications ...................................................... .... 119 CHAPTER6 SU"MMARY AND CONCLUSIONS .............................................................................. 131 Life-history Summary ........................................................................... ..................... 131 Conservation Management, and Research Prospectus .................... ........ .. ......... .. ..... 132 LIST OF REFERENCES ................................................................................................. 140 BIOGRAPIBCAL SKETCH ... ....................................................................................... 155 Vll

PAGE 8

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 LIFE HISTORY ECOLOGY AND CONSERVATION GENETICS OF THE STRIPED NEWT (Notophthalmus perstriatus) Chairman: C. Kenneth Dodd, Jr. By Steve A. Johnson August 2001 Major Department: Wildlife Ecology and Conservation The striped newt (Notophthalmus perstriatus) is a salamander endemic to south Georgia and north-central Florida The species has declined throughout its range because of habitat destruction and modification. Before my research little was known about striped newt life history. To learn more about striped newt ecology in order to make management and conservation recommendations I studied several aspects of striped newt life history I u s ed a multidisciplinary approach that incorporated fieldwork a laboratory experiment and DNA sequencing. For 2 years I monitored drift fences at a no1th-Florida breeding pond and in the sandhill uplands around the pond. This method was used to determine ba s i c param t e rs o f the species life history In the laboratory exp riment I reared individual larvae on different food regimes to test the influence of growth rate on lifeVlll

PAGE 9

history expression. A portion of the Cytochrome b gene was sequenced to det e rmine genetic population structure Field-collected data showed that striped newts have a complex life-cycle involving terrestrial and aquatic stages. An individual may move between a breeding pond and upland retreats multiple times during its life. Larval development occurs in the pond but once metamorphosis is complete individuals leave the pond and may disperse in excess of 500 m from the pond. Striped newts may express one of two life-history pathways. An individual may initiate metamorphosis and disperse from the pond before it matures (metamorph), or it may remain in the pond and mature while retaining larval characteristics (paedomorph). The metamorph vs. paedomorph decision is not controlled by growth rate per se, but is likely influenced by a suite of genes. Based on DNA sequence data, significant population genetic structure was found among ten locations sampled throughout most of the species' range, showing that gene flow is severely restricted among populations. It appears that striped newts form metapopulations and that long-term survival of the species depends on preserving those metapopulations that persist. Conservation efforts should focus on protecting and managing upland and aquatic habitats. A landscape approach is most effective and prescribed fire in the landscape is essential. IX

PAGE 10

CHAPTER 1 INTRODUCTION TO THE STRIPED NEWT (Notophthalmus perstriatus) The Striped Newt The striped newt (Notophthalmus perstriatus) is a salamander endemic to southeastern Georgia and north-central Florida (Christman and Means 1992; Conant and Collins 1991; Dodd and LaClaire, 1995; Franz and Smith 1999; Mecham 1967). Individuals are restricted to xeric upland habitats (primarily sandhill and scrub communities) and breed exclusively in temporary wetlands that lack predaceous fishes (Campbell and Christman 1982; Carr 1940; Christman and Means, 1992; Dodd and LaClaire 1995; Dodd et al., in press; Franz and Smith 1999 Stout et al. 1988). These upland ecosystems are pyrogenic (Myers, 1990) and fire appears to be crucial for the persistence of striped newts. Besides having a complex life history involving aquatic and terrestrial stages (Christman and Means, 1992; Dodd, 1993), individuals commonly exhibit paedomorphosis, the retention of larval morphology in mature individuals (Bishop 1941a, 1943; S. A. Johnson, pers. obs.). Status of the Striped Newt Decline of the longleaf pine/wire grass ecosystem fire suppression and the naturnl patchy distribution of upland habitats required by striped newts have re ulted in the fragmentation of striped newt populations. triped newts have declined throughout th ir range (Dodd and La laire 1995 ; Franz and mith 1999). A complex life history mak s s trip d newt vulnerable to threats at breeding pond ( e.g. ditching draining and filling

PAGE 11

2 of temporary ponds) and within the surrounding uplands (e.g. fire suppression vanous silviculture practices, and urban and agricultural development). Relative abundance of striped newts is extremely low at most areas where the species persists (S. A. Johnson B. Means, K. Greenberg and D. Stevenson, unpubl. data). Because of historical declines and low relative abundance at most locations the striped newt is recognized as a rare species throughout its range (Christman and Means, 1992; Cox and Kautz 2000; Jensen 1999). Its biological status is under review by the U.S. Fish and Wildlife Service L. LaClaire pers. comm.). Current Knowledge and Research Justification Striped newts have been characterized as uncommon and enigmatic (Christman and Means 1992) and "poorly known" (Dodd 1993; Dodd and LaClaire 1995) and until the last decade or so, little was known about their ecology. Most of the literature on striped newts is limited to the results of surveys (Dodd and LaClaire, 1995; Franz and Smith, 1999; Ripes and Jackson, 1996) and to species accounts (Ashton and Ashton 1988; Bishop, 1941a 1943; Carmichael and Williams, 1991; Carr, 1940; Christman and Means, 1992; Dodd et al., in press; Mecham 1967 Petranka, 1998). Johnson and Franz (1999) documented the occurrence of albinism in the species. Dodd and Charest (1988) and Dodd (1992) mentioned striped newts as part of the herpetofaunal community of a north Florida sandhills pond. Dodd (1996) included striped newts in his survey of terrestrial habitat use by amphibians. Studies of striped newt feeding habits (Christman and Franz 1973), natural history at a breeding pond (Dodd, 1993), and orientation into and away from a breeding pond (Dodd and Cade 1998) represent the only published works focusing specifically on striped newt life history. Petranka (1998) provided a

PAGE 12

3 summary of the biology of striped newts based mainly on the work of Dodd and coauthors. Taking into account the decline of N. perstriatus throughout its range, and that its biological status is under review by the U.S. Fish and Wildlife Service it is essential that natural resource managers acquire knowledge of striped newt life history. Such information will be required in order to draft a recovery plan, which would be required by law if the species was federally listed. Knowledge of striped newt life history also will be of immediate use to natural resources managers and may help circumvent the need to federally protect the species. Striped Newt Life History and Life-history Stage Terminology To better understand the following chapters of this dissertation, it would help to have a basic understanding of striped newt life history and terminology describing the various life-history stages. The life history stages are complex, and no published source adequately defines these stages as they relate to N. perstriatus. Throughout the life of an individual both aquatic and terrestrial stages occur ontogenetically, and an individual may move several times between aquatic and terrestrial habitats (Fig. 1-1 ). Reproduction occurs primarily in isolated, temporary ponds that lack predaceous fishes because the ponds dry relatively often. Courtship and oviposition by adult newts occur in the ponds and females lay eggs one at a time over a protracted period of several months. Eggs are often attached to aquatic vegetation. Eggs hatch into immature larvae that feed and grow in the pond.

PAGE 13

4 Larvae A larva is an immature aquatic stage newt. Larvae have bushy external gills a membranous tail fin and a conspicuous lateral line which is visible as a series of dorso lateral dashes on each side of the animal. Larvae do not posses the namesake lateral stripe and they do not have swollen vents. After a period of growth but before sexual maturation, a larva may metamorphose and leave the pond as an immature eft (i.e metamorphic pathway; Fig. 1-1 ) On the other hand, a larva may remain in the pond continue to grow and mature while retaining the larval morphology (i.e., paedomorphic pathway; Fig. 1-1 ). Efts An eft is an immature terrestrial stage newt. Efts lack gills and do not have a tail fin or lateral line. However at metamorphosis the namesake dorso-lateral stripe which is reddish to orange, appears on each side of an eft. Because efts are immature they do not have swollen vents. After larvae metamorphose into efts, they disperse into the uplands surrounding the breeding pond (Fig. 1-1 ). Efts mature in the uplands at which point they are referred to as terrestrial adults. Paedomorphs A paedomorph is a mature aquatic stage newt. Paedomorphs are larger than immature larvae, have bushy external gills a membranous tail fin, and a conspicuous lateral line. Paedomorphs do not usually possess the namesake lateral stripe but they do have swollen vents. Paedomorphs reproduce in the breeding pond, then metamorphose and disperse into the surrounding uplands (Fig. 1-1 ). Once a paedomorph transforms and leaves the pond it is referred to as a terrestrial adult. On rare occasions, a paedomorph may transform directly into an aquatic adult ( e.g. dashed line in Fig. 1-1 ).

PAGE 14

5 Adults An adult is a mature terrestrial or aquatic stage newt. Adults lack gills and do not have a visible lateral line. Adults possess the namesake dorso-lateral stripes and have swollen vents. They are sexually dimorphic and males have a light-colored gland that is visible at the posterior end of the vent. Adults occur in the uplands around breeding ponds (i.e. terrestrial adults) as well as in the ponds (i.e., aquatic adults) and there is movement between these habitats during the life of an adult. Aquatic adults develop a membranous tail fin similar to the tail fin larvae and paedomorphs. They do not regrow external gills however. Terrestrial adults lack a tail fin. They differ from efts in that the vent of a terrestrial adult is swollen whereas the vent of an eft is not swollen. In the chapters that follow, I refer to the various life history stages of striped newts as outlined above. Readers may need to refer back to this section as well as Fig. 1-1 until they are familiar with the striped newt life cycle and the different stages that comprise it. Overview of Dissertation Chapters 2 through 5 present the results of a multidisciplinary research project on striped newt life history; they include data based on a fieldwork component a laboratory experiment and a molecular genetics study. Each of these chapters plays an integral role in a unifying theme of striped newt life history and conservation. Chapters are written in manuscript format, each with its own introduction materials and methods results and di sc u ssio n to facilitate publication in peer-reviewed journals. Prior to my research knowledge of striped newt life history was limited to results of tudies at a ingle pond during a drought period. Because life history information is

PAGE 15

6 crucial for conservation and management planning I conducted a 2-year field s tudy o f striped newt life history at a different pond during a relatively wet period. In addition to monitoring newts at the breeding pond I trapped newts at various distances from the pond in the surrounding uplands. Results from captures at the pond which are presented in Chapter 2 demonstrated that striped newts have a complex life history and are adapted to taking advantage of temporary breeding habitats that fluctuate within and among seasons. Directionality of newt movements into and away from the pond was nonrandom (Chapter 3), and results of upland captures, also presented in Chapter 3 showed that striped newts dispersed hundreds of meters from the pond. One component of the striped newt s complex life history is the expression of alternative life history pathways (i.e., metamorphic pathway vs. paedomorphic pathway). Field-collected data showed that within a single cohort some larvae metamorphosed and left the breeding pond before attaining sexual maturity (metamorphic pathway) whereas others remained in the pond, continued to grow and matured in the larval morphology (paedomorphic pathway). In a laboratory experiment I tested the hypothesis that expression of these alternative life-history pathways is influenced by larval growth rate. The experiment also allowed me to test the applicability of two popular models of amphibian metamorphosis as they pertain to N. perstriatus. As presented in Chapter 4 growth rate did not significantly affect life-history pathway expression and neither of the two models of metamorphosis was totally consistent with the results of the experiment. Inasmuch as the genetic structuring of populations has important evolutionary biogeographical management, and conservation implications I conducted analyses of mitochondrial DNA of striped newts from throughout their range Results showed that

PAGE 16

7 there has been considerable evolutionary differentiation among the locations sampled probably because of genetic drift caused by natural habitat fragmentation. Genetic data supported data from mark-recapture studies, which suggest that striped newts form meta populations. The results of this molecular genetic study are presented in Chapter 5 and the consequences of metapopulation structure and population fragmentation for the conservation of this imperiled species are discussed in the last chapter. Chapter 6 the final chapter contains a brief summary of striped newt life history as well as conservation and management recommendations based on the results of the preceding chapters. In this chapter, I also provide suggestions for additional research on N. perstriatus.

PAGE 17

eggs terrestrial adult aquatic aault Fig. 1-1. Life-history schematic of the striped newt. Life-history stages include aquatic and terrestrial phases. During the life of an individual it will move between aquatic and terrestrial habitats. Lines and arrows indicate direction of movement. The dashed lin infers that this developmental path is possible but uncommon. 00

PAGE 18

CHAPTER2 LIFE HISTORY OF THE STRIPED NEWT AT A NORTH-CENTRAL FLORIDA BREEDING POND Introduction Salamanders of the genus Notophthalmus occur exclusively in North America with three extant species: N. viridescens (eastern newt) N. meridionalis (black-spotted newt) and N. perstriatus (striped newt). Notophthalmus viridescens ranges throughout the eastern United States and into southeastern Canada, whereas N. meridionalis is confined to extreme southeast Texas and northeastern Mexico (Conant and Collins 1991 Petranka 1998). Notophthalmus perstriatus is limited to northern Florida and southern Georgia (Conant and Collins 1991; Petranka, 1998). Each of the species exhibits a complex life cycle involving aquatic and terrestrial phases. The ecology of N. viridescens has been well studied (Gill, 1978a b; Harris, 1987; Harris et al. 1988 Healy 1970, 1973 1974a b, 1975 ; Hurlbert, 1969; Pope 1924). On the other hand far less research has focused on N. meridionalis and N. perstriatus (Petranka 1998) Because of historical declines and current relative abundance which is low throughout most of its range N. perstriatus is recognized as a rare species in Florida and Georgia ( hristman and Means 1992; Cox and Kautz 2000 Jensen 1999). Its biological status is under review by the U.S. Fish and Wildlife Service (L. La laire pers comm.). Mo t of what has been reported about striped newts has been limited to the r uJt of s urvey s (Dodd and La laire 1995 Franz and Smith 1999 Ripes and Jackson 1996) and pecies accounts (Ashton and Ashton 1988 ; Bishop 1941a 1943 Carmichael 9

PAGE 19

10 and Williams 1991 Christman and Means 1992 Carr 1940 ; Dodd et al. in press Mecham, 1967 Petranka 1998). In the only study of N. perstriatus life history Dodd (1993) monitored striped newt movements at a single breeding pond from 1985 through 1990. However a severe drought impacted the pond throughout Dodd's study. Despite dry conditions, Dodd (1993) determined seasonal activity population size structure and sex ratio. Of necessity these data were limited primarily to the adult life stage. Very little information was available on metamorphic individuals because the breeding pond only held water for short periods. It was not clear if the patterns observed by Dodd were typical of striped newt life history. To gain a better understanding of striped newt life history, I conducted a 2-year study at a breeding pond in north-central Florida. I used a drift fence (Gibbons and Semlitsch, 1981) to monitor striped newt immigration and emigration at the pond. My objectives were: 1) to determine the timing of immigration and emigration of newts, 2) to measure breeding success by monitoring emigration of metamorphic animals, 3) to estimate population size-structure and sex ratio, and 4) to evaluate the influence of hydroperiod and rainfall on striped newt movements and reproduction. Materials and Methods Study Site The study was conducted at One Shot Pond (OSP), an isolated water body within a high pine community in north-central Florida, approximately 4 km west of Breezeway Pond, the site ofDodd's (1993) striped newt study. One Shot Pond is located in Putnam Co. FL on the Katharine Ordway Preserve-Swisher Memorial Sanctuary (Fig. 2-1). Descriptions of the Preserve and its habitats are provided elsewhere (Dodd 1996;

PAGE 20

11 E isenberg and Franz 1995 ; LaClaire 1995). One Shot Pond is a sinkhole-depression pond with a basin area of ca. 0.8 ha (La Claire 1995). The hydroperiod of the pond is v ariable ( hydroperiod refers to the number of days a pond holds water between periods w hen it is dry) and it dries periodically. Because of this OSP does not support fish and many species of amphibians breed there (Johnson 1999). Drift Fence at One Shot Pond Newt movements into (i.e. immigration) and away from (i.e. emigration) OSP were monitored with a continuous drift fence (Gibbons and Semlitsch 1981) that encircled the pond. The galvanized metal fence was buried in the ground ca. 15 cm with ca. 35 cm extending above ground. The circumference of the fence was 190 m with 38 pitfalls (19 pairs of 19 1 plastic buckets) buried flush with the ground at intervals of ca. 10 m. For each pair of pitfalls one was buried on the side of the fence toward the pond and one on the side away from the pond. To reduce mortality of trapped animals foam sponges were placed in each trap and cover boards were leaned against the fence over each pitfall to provide shade. Each time traps were checked I removed invertebrates ( e.g. spiders predaceous beetles and centipedes) and added water to keep the sponges moist. Traps were checked at least 3 times per week and daily during periods of warm w e ather and/or high rainfall. Because of relatively high rainfall during the winter of 1997 / 98 the water in OSP rose to a level at which three pairs of pitfalls became inundat e d. In March of 1998 the entire fence was moved ca. 12 m further up slope r e quirin g an additional 40 m of flashing. The fence was moved in one day so there wa s no br e ak in trappin g The number of pitfalls remained the same and each trap wa r e in tall e d in the s ame relati ve po s ition at the new fence location. The drift fence was

PAGE 21

12 monitored from 7 Oct-96 to 11 Sep-98. Pitfall traps were open for 705 con s ecuti v e da ys, for a total of 26 790 trap-nights (i.e. one trap-night = one pitfall trap op e n for 24 h o ur ). Newts Caught at Drift Fences Most newts were marked with a unique toe clip so individuals could be identified (Donnelly et al. 1994) but recently metamorphosed efts captured after 11 Jul-98 were only marked with a daily cohort toe clip. Newts were weighed to the nearest 0 1 g using a Pesola scale and snout vent length (SVL) was measured to the nearest 1 mm The sex of each newt was recorded as male female or unknown. Sex was determined b y examining the vent region ; adult males have a light-colored gland visible at the posterior edge of the cloaca (Dodd 1993). The condition of the cloaca was recorded as swollen slightly swollen or not swollen indicative of mature maturing and immature individuals respectively (Chapter 4). Animals were released on the opposite side of the fence from where they were captured. Three distinct life-history stages of striped newts were examined: adults efts (immature larvae that recently metamorphosed) and paedomorphs (mature larvae that recently metamorphosed Table 2-1 Chapter 1 ). Recently metamorphosed newts retained vestiges of their gills (i.e., gill buds) for several days after they left the pond. Therefore the presence of gill buds indicated that a newt had recently transformed and left the pond Data for adults immigrating to the pond include recaptmes. Many of the adults captured in pitfalls on the outside of the drift fence had been previously captured and marked in the uplands surrounding OSP (Chapter 3). Others were initially marked as the y emigrated from the pond as immature efts. Data for emigrating adults also include recaptured individuals. These individuals had been initially marked as desc r ibed abo e or when they were captured in pitfalls on the outside of the drift fence as they immi gr at e d

PAGE 22

13 to breed Data for recently transformed paedomorphs and efts only include initial captures. By excluding eft and paedomorph recaptures I obtained a clearer pattern of striped newt movements. Weather Data To evaluate the influence of hydroperiod and rainfall on striped newt movements and reproduction rainfall and pond depth were monitored at OSP. Rainfall (to the nearest mm) was measured with a rain gauge mounted in the open within the pond basin. Pond depth was measured with a permanent depth gauge placed in the center of the pond I used binoculars to read the depth gauge. Statistical Analyses When assumptions of parametric tests were violated, nonparametric methods (Hollander and Wolfe 1999) were used to test for differences between data sets. All statistical analyses, with the exception of x 2 tests were performed using SPSS ver. 10.0. I used x2 tests (Sokal and Rolf 1995) to test for departure of 1: 1 sex ratios for adults and paedomorphs. To calculate expected values I divided the total number of males and females used in each analysis by two. Results Seasonal Activity During the 2-year study, 10 ,2 90 striped newt captures were recorded at the driftfence encircling OSP. At least one newt was captured during every week of the study although four periods of activity accounted for the vast majority of captures (Fig. 2-2). Activity during the se four peak s included newts moving into (immigration) and away from ( emigration) the pond.

PAGE 23

14 Immigration Immigration was almost exclusivel y comprised of breeding migration s of adult striped newts and there were four distinct adult immigrations during the stu dy (Ta bl e 22, Fig. 2-3A). The largest and most prolonged immigration event was immigration vent E-3 which lasted for 6 months. Peak movement during this event occurred in Dec-97 (Fig. 2-3A). During this month 1,567 adults were captured in outside pitfall traps. The other three immigration events were much smaller and occurred during 2-month or 3month periods. Females and males immigrated during the same times of the year (F ig. 23A). Although many juvenile newts were captured in pitfalls on the outside of the drift fence these individuals had recently metamorphosed and were initially captured and marked on the inside of the fence after they left the pond. Rather than immediately dispersing into the uplands some of these individuals headed back toward the pond and were caught in outside pitfalls. Nonetheless captures of these animals were the result of very localized movements and not an indication of immigration by efts. When recaptures of recently transformed individuals are excluded almost all (96%) newts captured in outside pitfalls had swollen vents, indicating that they were sexually mature. The remainder had slightly swollen vents indicating that they were close to maturity Emigration Emigration events included individuals representing all three life-history stages: adults efts, and recently metamorphosed paedomorphs (Table 2-2). Similar to immigration there were four distinct periods of adult emigration (Table 2-2, Fig. 2 -4 ). The largest of these (E-4) occurred toward the end of the study (Fig. 2-4). This emigration event accounted for 88% of all emigrating adults with more adults captured

PAGE 24

15 lea v ing the pond in Jun-98 than any other month of the study. Female and male adults immigrated during the same times of the year (Fig. 2-3B). Efts metamorphosed and emigrated during all months of the year except Februar y but there were four distinct periods of emigration (Table 2-2 Fig. 2-4). Most efts (81 %) emigrated during the last 5 months of the study period (eft emigration Event E-4). Recently metamorphosed paedomorphs were captured during three emigration events ( Table 2-2 Fig. 2-4). Most (94%) were captured from Mar-97 through Aug-97 (paedomorph emigration Event E-2). Reproduction Totals of 5 296 recently transformed larvae (i.e., efts) and 435 recently transformed paedomorphs were captured during the 2 years These individuals likely represented successful reproduction of four distinct breeding bouts as indicated by emigration of recently transformed newts throughout the study (Fig. 2-4). The first evidence of successful reproduction was provided by captures of emigrating efts (E-1 Table 2-2) and paedomorphs (E-1 Table 2-2) during the first few months of the study ( Fig. 2-4). During this period 776 efts and 25 recently transformed paedomorphs were captured dispersing from OSP (Fig. 2-4 ). The second period of eft and paedomorph production occurred in spring and early summer of 1997. During this time 214 efts and 407 recently transformed paedomorphs were captured (E-2 Table 2-2 Fig. 2-4). Only 16 e ft s w e re produced during the third eft emigration event (E-3 Table 2-2) but no recently tran s formed paedomorphs were captured dming this period (Fig 2 -4) By far the most u c c ess ful r e productiv e bout dming the 2 years was indicated by eft emi g ration (E-4) that o cc urr e d dmin g th e la s t 5 month s of the study E fts were captured starting in late May9 8 a nd immatur e l a rv ae co ntinu e d to tran s form and leave the pond until the end of the

PAGE 25

16 study (Fig. 2-4). I likely did not document the full extent of this emigration event s inc e 98 efts were captured in inside pitfalls on the last day of the study. Only nine recently transformed paedomorphs were caught during this period After they transformed and emigrated from the breeding pond efts migrated into the surrounding sandhill uplands (Chapter 3). While in the uplands efts matured before they returned to the pond to breed as adults. The largest immigration of adults (Event I3) which occurred from Oct-97 through Mar-98 consisted of many newts that were captured initially as they emigrated as efts during the first few months of the study (eft Event E-1 ). Although these newts were easily recognized as recaptures when they returned to the pond I could not be sure of individual toe clips in some instances. Of 40 newts that I was confident of their toe clip all of which had been marked as efts when they immigrated (i.e. vents not swollen) 39 had matured (i.e. vents swollen) by the time they were recaptured in outside pitfalls. Based on dissections and examination of gonads (Chapter 4) newts with swollen vents are always sexually mature. Therefore at least 39 of the 40 recaptured efts had matured in the uplands then migrated back to the pond to breed a year or more after they left the pond. These 40 newts had remained at large in the uplands around OSP for an average of 416 days (SD = 19.7; range = 359 to 456 days). The average number of days at large since metamorphosis was similar between the sexes. Males (n = 16) averaged 412 days at large (SD= 22.8 ; range = 359 to 440 days) whereas females (n = 24) averaged 419 days (SD= 17.4; range = 394 to 456 days). Net growth measured as the difference in SVL between initial capture during emigration and recapture during immigration was similar between females and males (Fig. 2-5).

PAGE 26

17 Differences in growth rates (net growth (mm) / days at large) between the sexes (Table 23) were not significantly different (Wilcoxon rank sum test ; t s= 0.815 P > 0.4). Population Size Structure The size-structure of striped newts differed between immigrating and emigrating adults. Snout-vent length and mass differed significantly among immigrating and emigrating males and females (SVL: F 3 3 1 36 = 776 P < 0.0001 ; mass: F 3 3 1 35 = 628 P < 0.0001). Post hoc comparisons showed that immigrating adults of both sexes were significantly smaller than emigrating adults for SVL (Fig. 2-6) and mass (Fig. 27). On average immigrating females were slightly larger (SVL and mass) than immigrating males and this pattern was evident during immigration Events I-1 I-3 and I-4 (Table 24). During immigration Event I-2 adult males and females were almost the same size (Table 2-4). However overall differences were not significant for SVL or mass (Figs. 26A 27 A). Adult females slightly exceeded males in SVL and mass for emigration Events E-3 and E-4 but males were slightly larger than females during the first two emigration events (Table 2-4). Overall emigrating females were significantly larger in SVL (Fig. 2-6B) and mass (Fig. 27B) than emigrating males (Scheffe s tests all P < 0.0001). Recently metamorphosed efts ranged in SVL from 20 32 mm (n = 2605) and from 0 .2 1.0 gin mass (n = 1886). Body size (SVL and mass) of efts differed among the four emigration events (Table 2-5). They were smallest during Event E-1 and largest durin g vent E -4 Snout-vent length and mass of efts differed significantly among three (E1 -2 -4) of the four emigration events (SVL: F 2 25 g 1 = 2114 P < 0.0001 mass: F 2 1 862 = 1 28 P < 0.000 I ; Table 2-5). Po t hoc compari ons showed that both measw-es of body iz were i gn ificantl y diff e rent for efts during vents -1 -2 and E-4

PAGE 27

18 (Scheffe s tests all P < 0.0001 Table 2-5). Event E-3 was excluded from the analyses because of small sample sizes. Male and female paedomorphs were essentially the same size (SVL and mass) during all three emigration events. Overall males were slightly longer than females but the average mass of males and females was the same (Table 2-5). There were no statistical differences in either measure of body size between the sexes (t tests ; SVL: t = 1.42 P = 0.156; mass: t = -0.159, P = 0.874 ; Table 2-5). Of the three paedomorph emigrations Event E-2 had the largest number of individuals (Table 2-5). Because of small sample sizes for Events E-1 and E-3 I did not make statistical comparisons of body size among the three events. Body sizes of the three different life-history stages differed (Table 2-5). Emigrating efts that had recently transformed had the smallest body size (SVL and mass ), followed by recently transformed paedomorphs then emigrating adult males. Emigrating adult females on average were the largest of all stages. These differences were statistically significant (ANOVA; SVL: F 3 40l7 = 3789 P < 0.0001; mass: F 3 32 66 = 1578 P < 0. 000 l) and post hoc tests showed that means differed among all four groups ( efts paedomorphs males and females ; Scheffe s tests all P < 0.0001). Sex Ratios Sex ratios were male biased during immigration Events I-1 and I-4 but female biased during Events I-2 and I-3 (Table 2-6). Because of the relatively large number of captures during immigration event I-3 when the sex ratio was 1 :1.26 (m:f) the overall sex ratio of immigrating adults was 1: 1.26. During emigration the sex ratio of adults wa female biased during all events except E-1 (Table 2-6). The relative contribution of sex ratio data provided by emigration Event E-4 (88% of all emigrating adults) had a large

PAGE 28

19 influence on the overall sex ratio of emigrating adults which was 1: 1.22 (Table 2-6). 0 erall adult sex ratio (emigrating and immigrating individuals) was female biased (Table 2-6x2 = 43.9 df = 1 P < 0.001). The sex ratio of recently metamorphosed paedomorphs was highly female-biased during each of the three emigration events Event E-2 was by far the largest of the three events representing 92% of paedomorph captures. Therefore E2 had a large impact on the overall sex ratio of paedomorphs which was significantly skewed toward females (m:f = I :4.64; x2 = 161.8, df= 1, P < 0.001). Rainfall and Hydroperiod Monthly rainfall at OSP varied ranged from 12 mm to 283 mm (Fig. 2-8). The driest periods were Nov-96 through Mar-97 and Mar-98 through Jul-98 (Fig. 2-8). The wettest period was from Jun-97 through Feb-98 because of an El Nifio Southern Oscillation event. Rainfall exceeded 100 mm during 13 months of the study period and beginning in Jun-97, there were 7 months consecutively in which rainfall exceeded 100 mm. Summer rainfall resulted from localized thunderstorms, whereas winter rain was associated with cold fronts. Newts tended to move during wetter periods and newt captures were significantly correlated with rainfall (P < 0.001 ). Nonetheless rainfall was a weak predictor of the magnitude of newt movements and only explained a small portion of variation in movements of newts at OSP (r 2 = 0.06). One Shot Pond held water throughout the study period (Fig. 2-9). Pond depth was lowe t (68 cm) in Oct-97 but the El Nifio rains filled the pond to its greatest depth (275

PAGE 29

20 cm) the following Apr. Analyses of the influence of pond drying and filling on s triped newt reproduction are precluded because OSP always held water during the s tud y Although pond depth exceeded 65 cm for the duration of the s tud y w ater d e pth may have influenced the survivorship of larvae and therefore the number of emi gr atin g efts. Of the four eft emigration events Event E-3 was the smallest (Table 2-5 ) coinciding with the shallowest pond depth during the study (Fig. 2-9 ) The large s t eft emigration (Event E-4 Table 2-5) began in May-98 when pond depth exceed 2 60 cm. Pond depth increased steeply during the months before May-98 (Fig. 2-9 ), when larvae that transformed during Event E-4 were growing in the pond. Although pond depth may have influenced the survivorship of striped newt larvae conclusions are confounded b y the fact that a variable number of females potentially contributed eggs that resulted in eft s for each emigration event. Fewer than 70 females appear to have contributed to the production oflarvae during eft emigration E-3 whereas ca. 1300 females potentially produced the larvae that transformed and emigrated during Event E-4. Discussion Seasonal Activity Striped newts were active at OSP during all months of the year but there were four periods of activity that accounted for most captures. In the only published stud y of striped newt life history Dodd (1993) also found several periods of activity that accounted for the majority of his striped newt captures over a 5-year period At OSP two activity periods occurred during the fall / winter whereas the other two took place durin g the spring/summer. At Breezeway Pond striped newts were mainly active durin g the fall/winter portion of the year (Dodd 1993)

PAGE 30

21 Immigration Adults moved into OSP to breed from Oct. through Mar. and from Apr. through Jul. At Breezeway Pond 75 % of adults immigrated from Jan. through Mar. (Dodd 1993). Although data collected at OSP support the fact that striped newts tend to breed during the winter (Dodd 1993 ; Petranka 1998) there were also two distinct migrations (presumably breeding migrations) during the spring / summer. Clearly striped newts are plastic in the timing of breeding migrations The only months adults were not documented moving into OSP were Aug. and Sep. Dodd (1993) suggested that the extended breeding period of striped newts allows them to take advantage of temporary breeding habitats that fluctuate within and among seasons. Such a plastic life history is likely an adaptation to living in an unpredictable environment. Emigration As with immigration there were four distinct periods of emigration. These periods overlapped with the four immigration events. Adults migrated to the wetland then courted and bred I assume and then moved back into the surrounding uplands. Thi s pattern persisted throughout the study even though OSP always held water. Therefore emigration of adults was not simply because of pond drying which appeared to be the case at Breezeway Pond (Dodd 1993). Based on the interval adults spent in the pond as well a s laboratory observations of reproductive activity (Johnson unpubl.) striped newt s ha v e protract e d courtship and oviposition Females including paedomorphic individual s l ay eggs on e at a tim e and do s o over the course of several months As a result adults that immi g rat e d into OSP during the winter of 1997 / 98 (Event I-3 ) fore ample stayed in the pond until they had fini s h e d breedin g and then emigrated during the summer of 1998 ( ve nt -4 T hi s ta gge r e d patt e rn o f immi g ration lat e r follow e d by emigration applies

PAGE 31

22 to the two other immigration events as well. Therefore, adults immigrating during Events I-1 and I-2 left the pond several months later, during Events E-2 and E-3 respectively (Tables 2-4 2-5, Fig. 2-3). A similar pattern is apparent in Dodd 's (1993) data early in his study (Fig. 1 in Dodd 1993) although the variable hydroperiod of Breezeway Pond and small number of captures confounds interpretation during the later years. Reproduction More than 5,500 recently transformed striped newts were captured as they emigrated from OSP. Production of very large numbers of metamorphic individuals is not uncommon for pond-breeding amphibians (Semlitsch et al., 1996). However, no previous studies have found so many metamorphic N. perstriatus (Dodd, 1993; B. Means pers comm.; K. Greenberg, pers. comm.; D. Stevenson pers. comm.). As a result of a drought, Dodd (1993) only captured 42 recently metamorphosed newts during the entire 5-year study at Breezeway Pond. Recently transformed efts emigrated in all months except January, but there were periods of concentrated migration, three of which accounted for 99 .6% of the captures (Fig. 2-4 Table 2-5). At Breezeway Pond, recently transformed striped newts were only captured from Jun. through Aug-97. The four eft emigration events at OSP presumably represent four bouts of reproduction. Eft emigration E-1 during the first two months of the study resulted from a reproductive event that likely occurred before the study began. The other three emigrations of efts probably represent reproduction of adults that immigrated during the study period. For example,' adult immigration Event I-1 produced larvae that metamorphosed and left the pond during eft emigration Event E-2. Adults that were captured immigrating during Events I-2 and I-3 probably produced most of th larvae that transformed and left OSP during eft emigration Events E-3 and E-4

PAGE 32

23 respectively (Figs. 2-3A 2-4). This staggered pattern of adult immigration later followed by eft emigration was because larvae apparently required approximately 6 months to reach metamorphosis. In the single successful reproductive bout recorded by Dodd at Breezeway Pond larvae required a 139-day hydroperiod. Paedomorphs also likely contributed to production of larvae in OSP but the relative contributions of non gilled adults and paedomorphs are unknown. A successful reproductive event often appeared to produce a bimodal distribution of emigrating newts. This is because within a single cohort of larvae both immature and mature larvae may result. Some immature larvae transformed and exited the pond as efts whereas others remained in the pond, attained sexual maturation (i.e., paedomorphs) reproduced, then transformed and exited the pond. This resulted in a bimodal pattern of emigration of a cohort with immature efts showing up first, followed by transformed paedomorphs. For example the recently transformed paedomorphs captured in the spring of 1997 (paedomorph Event E-2) were probably members of the same coh01t that produced the efts that emigrated the previous fall ( eft Event E-1; Fig. 2-4 ). The few paedomorphs that were caught in Jun. and Jul. of 1998 were likely members of the same cohort that produced the few efts that emigrated during eft Event E-3. Monthly samples taken in OSP with dip nets showed that the larvae that eventually matured as paedomorphs remained in the pond after their counterparts had transformed and emigrated as efts ( A. Johnson unpubl. data). Comparisons of size at metamorphosis and sex ratio for paedomorphs are precluded because no comparable data are available from other studies.

PAGE 33

24 Population Size Structure Sizes of recently transformed efts at OSP encompassed the range o f s ize s o f th is life-history stage at Breezeway Pond with the exception of the smalle s t indi v idual s captured at Breezeway. Recently transformed juveniles that Dodd (1 993 ) captur e d at Breezeway Pond ranged from 18 to 25 mm (n = 47) and 0.1 to 0.4 g (n = 44 ) Ba se d on Dodd s (1993) data a snout-vent length of 18 mm and a mass of 0 1 g. appear to b e absolute minimum body sizes required for a striped newt to initiate metamorphosi s. Although the average SVL of recently transformed efts at OSP was 25.8 mm the a v era ge SVL during the four emigration events varied significantly across all events Variation in zooplankton availability during each of the periods preceding the emigration events ma y have caused the differences. The sizes of adult striped newts at OSP were similar to sizes of adults from Breezeway Pond. Snout-vent length of females at OSP ranged from 25 to 43 mm whereas SVL of females at Breezeway Pond ranged from 26 to 43 mm. One Shot Pond females ranged from 0.3 to 1.6 g and Breezeway Pond females ranged from 0.3 to 2.0 g. Adult males at OSP were on average slightly smaller than females and ranged in SVL from 26 to 38 mm with mass ranging from 0.4 to 1.2 g. Dodd (1993) also found that male striped newts were smaller than females. Snout-vent length of males at Bree z ewa y Pond was the same as that for OSP whereas mass ranged from 0.2 1.6 g ( Dodd 1993 ) Sex Ratios Overall sex ratio of striped newts at OSP and Breezeway Pond were significantl y female-biased. At OSP there was one male for every 1.25 females and at Bree z ewa y Pond Dodd (1993) captured one male for every 1.46 females. The significance and

PAGE 34

25 cause(s) of the female bias in striped newts at the Katharine Ordway Preserve are unknown. Hydroperiod and Rainfall The hydroperiod of amphibian breeding ponds has a strong influence on reproduction (Pechmann et. al. 1989; Semlitsch 2000; Semlitsch et al., 1996). If hydroperiod is too short, larvae do not have adequate time to initiate metamorphosis and will therefore perish as the pond dries. On the other end of the spectrum, permanent ponds usually support predacious fishes that can extirpate some aquatic-breeding amphibians (Semlitsch, 2000). Although OSP held water during the entire study, over the past 2 decades it has dried often enough to preclude predatory fishes (R. Franz pers. comm.). The large number of efts emigrating from OSP during the 2 year study was probably because the pond held water continuously. In contrast, Dodd (1993) observed standing water in Breezeway Pond during only 14 months of the 5 year study period. Breezeway Pond held water in five distinct episodes (Fig. 1 in Dodd 1993), and the longest of these episodes was a 139-day hydroperiod. This was the only time during the study when Dodd (1993) captured recently transformed juveniles. The shorter hydroperiods precluded larval maturation, and consequently, no recently transformed paedomorphs were captured at Breezeway Pond. Long-term rainfall patterns likely have a significant impact on the striped newt population at the Katharine Ordway Preserve. Variability in hydroperiods of striped newt breeding ponds over relatively long time periods probably result in boom or bust scenarios for striped newt reproduction. Alternating relatively dry and wet intervals appear to result in highly variable striped newt reproductive success within and among pond Dodd (1993) captured very few metamorphic newts during his 5-year study and

PAGE 35

26 attributed an observed decline in striped newts at Breezeway Pond to persistent drought conditions. At OSP on the other hand I observed an increase in the number of striped newts mainly because of the large number of larvae that metamorphosed during the la st several months of the study. The heavy rainfall during the winter of 1997 / 98 filled the pond to its greatest depth (275 cm) while these larvae were developing. Because of the relatively great depth of the pond I suggest that there was more habitat and food (zooplankton) available to the larvae. This could have reduced intraspecific competition and contributed to the reproductive success and corresponding survivorship. Comparisons with Notophthalmus viridescens The life history of the red-spotted newt, Notophthalmus viridescens, has been studied in detail (Gill 1978a b; Harris et al., 1988; Healy 1970 1973 1974a b 1975 ; Hurlbert, 1969 ; Pope, 1924). Dodd (1993) compared and contrasted the life history of N. viridescens with N. perstriatus at Breezeway Pond. Data for striped newts at OSP allow some additional comparisons that Dodd was unable to make because of the poor reproductive success at Breezeway Pond. Striped newt larvae appear to be more variable than red-spotted newts with regard to the time of year when metamorphosis occurs. Recently transformed efts of N. viridescens have been found from Jun. through Nov. (Bishop 1941 b; Gill, 1978a; Hurlbert 1970 ; Worthington, 1968). Recently transformed striped newt efts at OSP were found during these months, but they were also captured during Dec. Mar. Apr. and May. The variation in timing of adult immigration and breeding of striped newts at OSP along with variability of the aquatic habitat quality ( e.g. food and pond depth) is likely responsible for the extreme variation in timing of larval metamorphosis. Healy (1973) estimated that red-spotted newt larvae in Massachusetts had a larval period of

PAGE 36

27 approximately six months. The same was true for striped newt larvae ( excluding paedomorphs) in OSP during my 2-year study. However, red-spotted newt larvae from other locations have shorter larval periods (Bishop, 1941 b; Harris et al. 1988 ; Worthington 1968). Red-spotted newt larvae transform into efts when they reach a SVL of 19 to 21 mm (Petranka 1998). The minimum size for metamorphosis of immature striped newt larvae is similar (Dodd, 1993; this study) The duration of the eft stage can be shorter for striped newts than red-spotted newts. At OSP many efts matured and immigrated to breed after an eft stage of about 14 months. Notophthalmus viridescens efts may remain on land from 2-8 years before returning to breeding ponds (Bishop 1941b; Healy 1974a). Striped newts certainly have much greater variability in duration of the eft stage than I documented at OSP. Although many efts matured and returned to OSP after a period of about 14 months this is likely a minimum time-frame. Some efts probably remained in the uplands around OSP and did not migrate to breed. Therefore these individuals, once they matured and migrated to the pond to breed would have an eft stage longer than 14 months. Moreover if drought conditions had prevailed in the vicinity of OSP during the winter of 1997 / 98 the efts that did migrate to breed would not have had the opportunity to do so thus increasing the estimate of duration of the eft stage. Annual variation in rainfall certainly has an impact on the duration of the terrestrial stage of striped newts. Growth rates of N. perstriatus efts at OSP appear similar to growth rates of ju ven ile red-spotted newts. Healy (1973) calculated growth rates of marked efts in a Massachusetts s population of red-spotted newts for three consecutive years. Healy (1973) presented mean growt h increment' values (Table 2 in Healy 1973) and I

PAGE 37

28 estimated daily growth rates by dividing his mean values by 365 The range o f g rowth rates (mm SVL / day) for red-spotted newts varied from 0.0048 to 0.0159 Growth rate s of striped newt efts (females: 0 0167 mm SVL / day ; males: 0 0183 mm SVL / day ) w e r e similar to growth rates of red-spotted newts in Massachusetts (0.0159 mm SVL / da y; Healy 1973). Because of the short duration of my study annual variation in growth rate s of striped newt efts is unknown. For striped newts and red-spotted newts life-history pathway has a profound influence on age at first reproduction. In both species individuals that omit the eft stage reach sexual maturity earlier than individuals that metamorphose when immature. For N. perstriatus an individual that omits the eft stage matures as a paedomorph and reproduces at about 1 year old. For N. viridescens an individual that omits the eft stage may remain in the pond and later mature as an aquatic adult or mature as a paedomorph (Brandon and Bremer 1966 ; Healy 1970 1974a ; Petranka 1998). According to Healy (1974a) immature red-spotted newts that remain in the pond and omit the eft stage reproduce earlier (at 2 years old) than newts that migrate into the uplands as efts. This life-history pathway has not been detected in stripe newts. Red-spotted newts that become paedomorphic may reach sexual maturity in as little as seven months (Petranka 1998). Expression of the paedomorphic life-history pathway is common throughout the range of N. perstriatus (S. A. Johnson unpubl. data ; D. Stevenson pers. comm. D. B Means pers. comm. ; J Jensen pers. comm.). Paedomorphosis is most common in coastal populations of N. v irid e sc e n s and in places wher e the terrestrial environment is percei v ed as exceptionally harsh (Bishop 1941b ; Brandon and Bremer 1966 ; Healy 1974a ;

PAGE 38

29 Petranka 1998). In both species genetic and environmental factors are believed to control the expression of life history pathway (Chapter 4; Harris 1987). Implications for Striped Newt Status Surveys Life history data for striped newts at OSP have implications for management of the species. Considering the imperiled status throughout its range (Cox and Kautz, 2000; Jensen 1999), identifying undocumented breeding ponds and monitoring striped newts at known breeding ponds will help ensure the long-term persistence of the species. Probably the most efficient method to survey multiple sites is by sampling breeding ponds for striped newts. Obviously, such surveys must be conducted during non-drought periods when breeding ponds hold water. Dodd's (1993) work proved that during drought conditions, a suitable striped newt pond may only hold water for short periods. Furthermore, as drought conditions persist, newt abundance declines. Therefore, even when potential breeding ponds hold water newts might be present in such low numbers as to elude detection. Use of drift fences around ponds will increase the likelihood of detection but this method is very labor-intensive and is not practical for range-wide surveys when time and personnel are limited. Failure to detect newts during drought periods may result from low relative abundance caused by the drought, rather than local extirpation. On the other hand, during wet conditions, such as was the case at OSP following the winter of 1997 /98, newt abundance can be relatively high. Wet periods increase the likelihood of detecting the species during aquatic sampling. Smveys conducted during relatively wet periods will prove fruitful for monitoring persistence and locating new breeding ponds Additionally as suggested by Dodd (1993), surveys for triped newts should include assessment of biotic and abiotic characteristics of known and potential breeding sites.

PAGE 39

30 Striped newts of various life-history stages may be found in breeding ponds during all months of the year. However, I suggest that spring (Apr. through Jun.) is the best time of the year for conducting aquatic sampling for the species, assuming ponds hold water. At this time of year, OSP contained all three life-history stages. Adults and paedomorphs that had recently bred were still in the pond, as were developing larvae. Sampling for newts in breeding ponds during this time of the year should maximize the probability of capturing newts. However, considering the temporary nature of striped newt breeding ponds, individuals conducting striped newt sampling should conduct surveys whenever breeding ponds hold water.

PAGE 40

31 Table 2-1. Descriptions of the three life-history stages of striped newts referred to in the Chapter 2. tage Description Adult Mature newt as indicated by a swollen vent; no evidence of gills; males with a distinct, light-colored gland at posterior edge of cloaca Eft Immature newt as indicated by a vent that is not swollen; gill vestiges present--indicating recent metamorphosis of an immature branchiate; sex recorded as unknown Paedomorph Mature newt as indicated by a swollen vent, gill vestiges present--indicating recent metamorphosis of a mature branchiate; males with a distinct, light-colored gland at posterior edge of cloaca

PAGE 41

Table 2-2. Timing of immigration (I) and emigration (E) events of the three life-history stages of striped newts captured at One Shot Pond Putnam Co. FL. Event I-1 I-2 I-3 I-4 E-1 E-2 E-3 E-4 Adults Oct.-96 through Dec.-96 Apr.-97 and May-97 Oct.-97 through Mar.-98 Jun.-98 and Jul.-98 Oct.-96 through Dec.-96 Apr.-97 through Aug.-97 Nov.-97 through Jan.-98 May.-98 through Aug.-98 Life-history Stage Efts not applicable not applicable not applicable not applicable Oct.-96 through Dec.-96 Mar.-97 through Jun.-97 Aug.-97 through Dec.-97 May.-98 through Sep.-98 Paedomorphs not applicable not applicable not applicable not applicable Oct.-96 through Dec.-96 Mar.-97 through Aug.-97 Jun.-98 and Jul.-98 not applicable w N

PAGE 42

33 Table 2-3. Growth of striped newt efts at One Shot Pond Putnam Co. FL. Growth is expressed as mm/day from initial capture during emigration (shortly after metamorphosis) until recapture during immigration. n Mean Range SD Females 24 0.0167 0.0068 0.0295 0.0054 Males 16 0.0183 0.0129 0.0306 0.0043

PAGE 43

34 Table 2-4. Snout-vent length (SVL) and live mass of adult striped newts caught at One Shot Pond Putnam Co. FL during four immigration events. Females Males Event SVL(mm) Mass (g) SVL (mm) Mass (g) I-1 n 11 7 18 11 Mean 36.27 0.80 33.80 0.62 Range 26 40 0.6 1.0 27 39 0.3 1.0 SD 4.34 0.13 3.80 0.20 I-2 n 58 56 22 22 Mean 33.57 0.72 33.59 0.72 Range 27 40 0.4-1.1 28 40 0.4 1.0 SD 4.19 0.20 3.28 0.14 I-3 n 1148 1140 924 914 Mean 30.90 0.61 30.77 0.61 Range 26 41 0.4 1.2 26 39 0.3 1.0 SD 1.87 0.11 1.66 0.11 I-4 n 10 9 8 6 Mean 35.6 1.0 34.25 0.90 Range 33 39 0.8 1.2 33 36 0.8 1.0 SD 1.90 0.14 0.89 0.11 Total n 1227 1212 972 953 Mean 31.11 0.62 30.91 0.61 Range 26 41 0.4 1.2 26 40 0.3 1.0 SD 2.24 0.13 1.88 0.11

PAGE 44

Table 2-5. nout-vent length (SVL) and live mass of adult recently transformed eft and recently transformed paedomorph striped newts captured at One Shot Pond Putnam Co., FL during four emigration events. Adults Paedomo!:Ehs Females Male s Efts Females Male s E ent VL (mm) Mass (g) SVL (mm) Mass (g) SVL (mm) Mass (g) SVL (mm) Mass (g) SVL (mm} Ma ss (g} E-1 n 4 2 7 4 772 54 17 6 7 3 Mean 29. 00 0.65 30.57 0 .7 3 23.02 0.31 26.47 0.42 27.00 0.48 Ran ge 25 38 0.4 0 9 26 38 0.6 0.9 20 29 0 2 0.5 24 30 0.4 0.5 24 29 0.4 0 .5 D 6.06 0.35 4.24 0.15 1.34 0 07 1.55 0.04 1.63 0.06 E -2 n 70 66 18 18 209 213 334 320 72 70 ean 32.40 0.70 33.06 0 68 28.29 0.55 31.838323 0.73 32.57 0 73 Ran ge 26 41 0 3 -1.1 28 37 0.4 1.0 23 32 0 3 1.0 27 43 0.4 1.8 28 40 0.4 1.0 D 4.07 0.19 2.71 0 14 1.51 0.12 2.54 0.18 2.06 0.13 E-3 n 13 13 11 10 21 21 5 5 2 2 Mean 30.54 0 58 29.82 0 54 24.43 0.33 37.40 1.04 35.00 1.05 vJ Range 28 33 0.4 0.8 26 33 0.5 0 6 21 32 0.2 0.8 35 41 0.5 1.6 33 37 0 9 1.2 Vl SD 1.45 0.14 1.99 0 05 2.84 0 15 2.51 0.40 2.83 0 .2 1 E-4 n 469 460 414 406 1603 1598 not not not not Mean 36.31 0 93 33.72 0 82 26.74 0.46 applicable applicable applicable applicable Range 30 43 0.5 1.6 31 38 0.4 1.2 22 31 0.2 1.0 SD 1.73 0.18 1.29 0 13 1.46 0.11 Total n 556 541 450 438 2605 1886 356 331 81 75 Mean 35.63 0 88 33.55 0 80 25.75 0.47 31.66 0 73 32 15 0 73 Ran ge 25 43 0 .3 1.6 26 38 0.4 1.2 20 32 0 .2 1.0 24 43 0.4 1.8 24 40 0.4 1.2 SD 2.73 0.21 1.63 0.14 2.33 0 12 2.83 0 19 2.60 0 15

PAGE 45

36 Table 2-6. Sex ratios of adult and paedomorph striped newts captured at One Shot Pond Putnam Co. FL. Overall and grand total values include several individuals not accounted for in the immigration (I) and emigration (E) events listed. Sex ratios are listed as the ratios of males:females followed by the number males and females captured in parentheses. Event Adults I-1 1 :0.57 (23: 13) I-2 1 :3.00 (22:66) I-3 1: 1.26 (1038: 1307) I-4 1 :0.69 (32:22) Overall-I 1: 1.26 (1119: 1412) E-1 1 :0.57 (7:4) E-2 1 :3.74 (19:71) E-3 1: 1.08 (12: 13) E-4 1:1.23(431:484) Overall-E 1 :1.22 (469:572) Grand total 1:1.25 (1588:1984) Paedomorph not applicable not applicable not applicable not applicable not applicable 1:2.57(7:18) 1 :4.64 (72:334) 1 :3.50 (2:7) not applicable 1:4.43 (81 :359) 1 :4.43 (81 :359)

PAGE 46

GULF OF MEX I CO .,.E---~-area enlarged 37 ATLANTIC OCEAN ig 2 -1. Location of study area Katharine Ordway Preserve in Putnam Co. north c entral F lorida

PAGE 47

2000 1800 1600 1400 en d) E 1200 g. <.) <.;..; 1000 0 l--, d) ..0 800 z 600 400 200 0 r7 r7 r7 .--, n n 17 r7 I I I I I I I I I I I I I I I I I I I I I I I I b b b \.'O, <4 ?, ?, ?, ?, -s:' -s:' ~?, ?, ~O, n~ ?, ~O, <4?, ?, ?, ?, -S:O, -s:' ~ ?, ?, ~O, ?, / ov ~ o <:;fP "(,,~ ~('t,,, \V ~v"-0 ov ~o <)~CJ "(,, ~ ~ 'b,> ~,;;:,,% Month Fig. 2-2. Monthly activity patterns of all striped newt captures at One Shot Pond Putnam Co. FL. Adult recently tranformed efts and recently transformed paedomorphs are included VJ 00

PAGE 48

39 1800 A 1600 Males C=::J Females 1400 C/'J 1200 d.) I-< .a 1000 C) 4-1 0 800 I-< d.) ..0 1-3 s 600 :::s z 400 200 1-2 1-4 1-1 1000 .---------------------------~ C/'J d.) B 0.. 800 600 4-1 0 I-< d.) ..0 400 z 200 B E-4 E-2 E-1 E-3 0 ....L__ ~---~ ~ ~~-,.__.,.~~-----~~---....iiiiiiL--~~~----ov ~~e;-<::i. Month F i g. 2 3. aptures of adult f e male and male striped newts at One Shot Pond Putnam o. FL. A ) Immi gr atin g adult s. B) E migrating adults. Note differences in scales along th e Yaxes. F our di s tinct p e riod s o f immigration (I) and emigration (E) are indicated

PAGE 49

en Q) 3 ..... 0.. ro (.) 4-,( 0 1-, Q) ..0 z 1800 1600 Adults c=J Paedomorphs I 1400 Efts 800 E-4 600 E-1 400 E-2 200 E-3 0 -'------=,::.:L......1 ....... _...----.-,-.__,_c::;;=_,___ b b b ~' '4 ?-i cf' -s:' -s:' ~' ~' ~' ~' '4~ -s:' -s:' ~' ~' ~ Ov ~ O <:) l?J \~<:;i '\(,, ') \v ~v C:,~'\J,, Ov ~ O <)~v \~<:;i '\(,, ') ~v C:, ~ '\J,, Month Fig. 2-4. Monthly captures of adults recently transformed efts and recently transformed paedomorph s emigrating from One Shot Pond Putnam Co. FL. Four distinct periods of newt captures are indicated 0

PAGE 50

---> r:/'J .5 Q.) en (13 Q.) 1-, (.) s ...... Q.) z 14 .------------------------------------------~ 1 2 0 Males 0 Females 10 0 0 o o 8 0 00 0 0 0 00 o 6 0 o 00 0 4 0 0 0 0 2 0 -+----------,------.---------,---------r-------r----------r-------t 340 360 380 400 420 440 460 480 Days at large Fi g 2-5. Net increase in SVL of female (n = 24) and male (n = 16) efts since initial capture during emigration shortly following metamorphosis and recapture during immigration when they returned to breed. Note X-axis s tarts at 340 day s and s ome symbols overlap. .......

PAGE 51

en ;::s "O ;; .s 4-; 0 I-; d) .J:;) z en ;::s "O ;; ..., "O .s 4-; 0 I-; d) .J:;) s ;::s z 42 300 A 250 Males C=:J Females 2 0 0 150 100 50 0 n J I ~ I n n J n n I I I 20 25 3 0 35 40 45 140 B 120 100 80 60 40 20 0 n n J I ~ I I n n I I I 20 25 30 35 40 45 Snout-vent l ength (mm) Fig. 2-6. Snout-vent lengths of adult female and ma l e striped newts captured One Shot Pond Putnam Co. FL. A) Immigrating adults. B) Emigrating adults.

PAGE 52

C/1 "'O ;;: .s 0 I-< (l.) s ;:::s z C/1 "'O .s 0 I-< (l.) z 600 500 400 300 200 100 0 0.0 160 140 120 100 80 60 40 20 0 0 0 Males c::==J Females I I 0.2 0.4 .n I I I 0 2 0.4 43 A n n I I I I 0.6 0.8 1.0 1.2 1.4 1.6 B I n I I I 0.6 0.8 1.0 1.2 1.4 1.6 Live Mass (g) ig 27. Liv ma s of adult female and male striped newts captrned at One hot Pond Putnam o ., F L. A) Immigrating adults. B) Emigrating adults

PAGE 53

200 ,,-._ '-.-/ 150 .s 100 Month Fig. 2-8. Monthly rainfall recorded at One Shot Pond Putnam Co. FL.

PAGE 54

3 00 80 60 40 22 0 2 00 E 1 80 u ..c 160 ..... a, C1.) 140 "'O "'O c:: 1 2 0 0 0-c 100 80 VI 60 40 20 0 Week (number along X-axis) and Month Fig. 2-9 Depth of One Shot Pond Putnam Co. FL at the end of each week. Pond depth was not recored until week 15

PAGE 55

CHAPTER3 ORIENTATION AND DISPERSAL DISTANCES OF STRIPED NEWTS AT A NORTH-CENTRAL FLORIDA BREEDING POND Introduction During the past two decades amphibian declines have received considerable attention (Alford and Richards 1999 ; Barinaga 1990; Wake 1991 ; Wake et al. 1991 ). Although pathogens have been implicated in several die-off events (Berger et al. 1998 ; Lips 1998 1999), there is a consensus among herpetologists that the global decline is a result of multiple factors (Alford and Richards 1999). Habitat modification and destruction have been identified as significant factors contributing to the global decline (Alford and Richards, 1999; Dodd 1997; Duellman, 1999; Semlitsch 2000). Although they do not attract the media attention that mass mortality or deformed amphibians receive habitat modification and loss are insidious processes that must be addressed if amphibians are to persist. The effects of habitat changes on amphibian populations are of particular concern in areas that are characterized by a high density of small isolated wetlands (Alford and Richards 1999 ; Babbitt and Tanner 2000; Delis et al. 1996 ; Dodd 1997 ; Greenberg, 2001; Hecnar and M Closkey 1996 ; Knutson et al. 1999 ; Knutson et al. 2000; Semlitsch, 2000; Snodgrass et al. 2000). In these areas ( e.g. the Southeastern Coastal Plain of North America) amphibian diversity is high (Duellman and Sweet 1999) and many species rely solely on small isolated wetlands as breeding sites (Babbitt and Tanner 2000 ; Dodd 1997 ; Semlitsch and Bodie 1998) 46

PAGE 56

47 Despite their size (i.e. less than a few hectares) small, isolated wetlands are of tremendous biological importance particularly for amphibians. In the Southeastern Coastal Plain for example these wetlands support a rich diversity of amphibian species and several ponds that have been studied in detail were found to produce thousands of metamorphic individuals (Dodd, 1992 ; Gibbons and Semlitsch, 1981; K. Greenberg pers. comm. ; Hart and Newman, 1995 ; Johnson, 1999; R. Means pers. comm. ; Moler and Franz 1988 ; Semlitsch et al. 1996; Semlitsch and Bodie 1998). Small isolated wetlands likely play a vital role in amphibian metapopulation dynamics, and therefore are essential in maintaining viable populations of amphibians at a landscape level (Semlitsch and Bodie 1998; Semlitsch 2000 ; Snodgrass et al. 2000). In addition to amphibians numerous other vertebrates and a suite of invertebrate species depend on small isolated wetlands (Brown et al., 1990; Burke and Gibbons 1995; Hart and Newman 1995 ; Moler and Franz, 1988; Semlitsch and Bodie, 1998). Preserving a wetland alone may not result in protection of many of the organisms that depend upon the wetland. Many amphibians have complex life cycles in which they require ponds to breed but spend the majority of their lives in surrounding upland habitats (Dodd 1997; Dodd and Cade 1998 ; Semlitsch 1998). If sufficient upland habitat surrounding isolated breeding-ponds is not preserved amphibians with complex life cycles are not likely to persist at a local scale. Therefore at some point the loss of upland s may lead to extirpation of some amphibian populations because of disruption of metapopulation dynamics (Semlitsch and Bodie 1998; Semlitsch 2000) even when the ponds themselves are preserved

PAGE 57

48 One strategy to curtail the loss of amphibians associated with habitat alteration around small isolated wetlands is to preserve terrestrial core zones of upland habitat surrounding the ponds. These zones provide habitat for retreats and foraging for tho se species with complex life cycles many of which are now considered common Without preservation of appropriate upland habitat even common species will decline. Little is known however about the extent of upland core areas required by pond-breeding amphibians. Dodd (1996) summarized the literature on upland movements of amphibians in North America and found that this life stage is poorly known. From this summary and a review by Semlitsch (1998) on dispersal distances of ambystomatid salamanders it is apparent that many amphibians disperse considerable distances from breeding ponds. Unfortunately dispersal distances are only available for a few species and usually are based on a single or a few individuals. Clearly there is need for data on dispersal distances from breeding sites for most North American amphibians. These data are essential to justify establishing adequate "core areas of upland habitat around amphibian breeding ponds. I collected data on orientation and dispersal distances for striped newts (Notopthalmus perstriatus) at a breeding pond and in the surrounding uplands in north central Florida. The striped newt breeds exclusively in small, isolated wetlands that lack fish. It has a complex life cycle and individuals spend much of their lives in uplands surrounding breeding ponds (Fig. 1-1; Carr, 1940; Christman and Means 1992 Dodd and LaClaire 1995 ; Dodd et al. in press; Franz and Smith 1999; Chapter 2). Striped newts are restricted to xeric uplands (i.e. sandhill and scrub communities) in southern Georgia and northern Florida U.S.A. (Fig. 3-1). The species has declined throughout its

PAGE 58

49 range (Dodd and LaClaire, 1995; Franz and Smith 1999) and its biological status is under review by the U.S. Fish and Wildlife Service (L. Laclaire pers. comm.). The objectives of my study were to 1) determine orientation patterns of striped newts into and away from a breeding pond and 2) determine dispersal distances of individuals into the surrounding upland habitat. Methods Study Site The study was conducted on the Katharine Ordway Preserve-Swisher Memorial Sanctuary Putnam Co. FL (29 'N, 82'W; Fig. 2-1). Eisenberg and Franz (1995), LaClaire (1995) and Dodd (1996) provide descriptions of the Preserve and its habitats. Data were collected from 7 Oct-96 to 11 Sep-98 at One Shot Pond (OSP). One Shot Pond is a small isolated pond with a variable hydroperiod (hydroperiod refers to the number of days a pond holds water between periods when it is dry) and is located in xeric sandhill uplands dominated by longleaf pine (Pinus palustris) turkey oak (Quercus laevis), and wiregrass (Aristida beyrichiana). A small pine plantation (Pinus elliottii) is located west of the pond basin (Fig. 3-2). Several water bodies are located near OSP (Fig. 3-2). These water bodies are isolated from one another and only receive water from rainfall and ground water seepage; their hydroperiods are dictated by fluctuations in the water table. Fox Pond held water from 26 Nov-97 until the end of the study whereas 0 P, Berry Pond and the Anderson Cue Lakes held water throughout the entire study period During the s tudy striped newts were only present in OSP and Fox Pond. However only 32 newts (16 adults and 16 juveniles) were captured at Fox Pond (John on 1999) The Anderson ue lakes suppo1t predatory fishes and striped newts do

PAGE 59

50 not breed there. No striped newts were captured during periodic sampling throughout the study period in Berry Pond. Because there were no other breeding ponds within several kilometers of OSP I assumed that striped newts caught in upland fences around OSP originated from within OSP. Orientation at One Shot Pond I encircled OSP with a 190-m drift fence made of galvanized metal flashing that was buried ca. 15 cm below the ground with ca. 3 5 cm extending above the ground Thirty-eight pitfall traps (19 I plastic buckets) were buried flush with the ground. Pitfall traps were placed in pairs one on each side of the fence at intervals of about 10 m. I usually checked traps three to five days per week, depending on weather and movements of animals. I weighed and measured newts caught in pitfall traps at the pond and in the surrounding uplands (Chapter 2). Each newt was individually marked by toe clipping (Donnelly et al., 1994) and released on the opposite side of the fence. Sex of adults was determined by the presence of a conspicuous whitish gland visible at the posterior edge of the vent in mature males. Recently transformed newts were recognized by the presence of gill vestiges visible for several days after metamorphosis. Recently transformed newts with swollen vents were presumed to be mature (Chapter 4), and aquatic sampling in the pond showed that such individuals represent paedomorphic animals that recently bred. I obtained a compass orientation for each pair of pitfall traps surrounding OSP. To do this I stood in the center of the pond and took a bearing on each pair of traps at the drift fence. Following the methods of Dodd and Cade (1998), I used Rao s spacing test (Batchelet, 1981; Rao 1976) to determine if captures were distributed uniformly around the drift fence (i.e., random orientation). I analyzed orientation of newts into and away from the pond by sex and life history stage (paedomorph vs. metamorph Table 2-1). I

PAGE 60

51 made comparisons between distinct migration events (Chapter 2) within the adult and eft life history stages. For comparisons between sexes, life history stages, and migration events I ran the same multirepsonse permutation procedure (MRPP Mielke et al. in press) used by Dodd and Cade (1998). Orientation analyses were performed with the statistical software package BLOSSOM which was developed by the U.S. Geological Survey (Cade and Richards, 2000). BLOSSOM is available free at www.mesc.usgs.gov / blossom/blossom.html. Upland Dispersal Dispersal distances of newts in the sandhill uplands around OSP were determined through captures in pitfall traps associated with drift fences. Drift fences were oriented in such a manner as to capture newts during movements to and from the pond (Fig. 3-3). In year one five fence sections were established at each of four distances from OSP (20 m, 40 m 80 m and 160 m). Fence sections at each distance totaled 20% of the circumference at that distance from the pond. Fence sections were distributed evenly at each distance, and they did not overlap with fence sections at the other distances (Fig. 33A). Fence sections at 20 m were 10.0 m long with 4 pitfalls (2 on each side of the fence); at 40 m, fence sections were 15 .1 m with 6 pitfalls; at 80 m, sections were 25 .1 m with 8 pitfalls ; at 160 m, sections were 4 5 .2 m with 10 pitfalls. Pitfall traps were installed on both sides of the upland fences (i.e., pond side and upland side; Fig. 3-3A). This upland fence array was monitored from 7 Oct-96 to 5 Dec-97, and fences were constructed similarly to the fence at the pond. Results from year one demonstrated that striped newts regularly dispersed more than 160 m. Therefore a new upland fence array was installed in year two with upland drift fences erected much farther away from O P On 5 Dec-97 the upland drift fences

PAGE 61

52 described above were replaced with a different array of fence s ection s (F i g 3 -3B ) and th new fences were in place by 7 Dec-97. These fences were constructed of heavy-gau g e silt-fence material buried ca. 15 cm into the ground ; ca. 40 cm extended above ground Two fence sections were installed at each of five distances (100 m 200 m 300 m 400 m and 500 m) from the pond. Fence sections at each distance totaled 13.4 % of the circumference at that distance from the pond and fence sections overlapped (Fig. 3-3B ) The two fence sections at 100 m were each 42 m long with 6 pitfalls (3 on each side of the fence) installed evenly throughout each section; at 200 m sections were 84 m with 10 pitfalls ; at 300 m sections were 126 m long with 14 pitfalls ; at 400 m sections were 168 m long with 18 pitfalls; at 500 m sections were 210 m long with 22 pitfalls. Pitfall traps were oriented in the same manner as year one ; pond-side traps were on the side of the fences toward OSP and upland-side traps were away from OSP (Fig. 3-3B). The upland fence array in year two was monitored until the study ended on 11 Sep-98 In total 280 pitfall traps were installed at upland fence sections and were monitored during the 2-year study for a total of 98 140 trap-nights (i.e. one trap-night = one pitfall trap open for 24 hours). Upland traps were checked on the same schedule as those at the pond and newts were processed as described above. I estimated the proportion of the newt population that dispersed different distances from the pond based on captures at upland fence sections and at the outside of the drift fence encircling OSP. Data used in the estimates were confined to 7 Dec-97 through 31 Mar-98. During this period there was a mass migration of newts toward the pond and very little movement away from the pond (Chapter 2) Ninety-one percent of upland fence captures during year two occurred during this period. These captures

PAGE 62

53 however only represented newts tI:iat migrated through a subset of surrounding uplands. Because upland drift fences sampled only 13.4% of the uplands at each distance I multiplied the number of captures in the outside pitfalls by 7.5. The product of this calculation is an estimate of the number of captures expected at each distance had the upland fence sections sampled 100% of the uplands at each distance. For each upland fence section, the estimate was divided by the number of total newt captures on the outside of the fence at OSP to approximate the proportion of individuals that had dispersed various distances (i.e., 100 m to 500 m at 100 m intervals). I assumed that there was no strong nonrandom orientation of newts moving through the uplands. Nonetheless, movement of newts into and away from the pond was nonrandom but there was no overwhelmingly strong directionality that would violate this assumption. However estimates of the proportion of newts that had dispersed various distances from the pond are probably conservative. I use the term "migration' to indicate seasonal movements of newts toward or away from the breeding pond. "Immigration" indicates a general pattern of migration toward the breeding pond whereas "emigration" indicates migration away from the pond (Semlitsch and Ryan 1999). Results Orientation at One Shot Pond All patterns of adult immigration and emigration were significantly nonrandom ( ig. 3-4 ; Rao s Spacing Tests all P < 0.001). Adult striped newts entered and exited the pond in all directions. They tended to enter the pond basin primarily from the east and west (Fig. 3-4 ). Adults emigrated in all directions but there was a single distinct angle of

PAGE 63

54 emigration, as indicated by the relatively high number of captures in a pitfall trap located at a south-southeast direction (Fig. 3-4 ) Emigration of paedomorphs and efts also was nonrandom (Fig. 3-5; Rao's Spacing Tests both P < 0.001). There was no obvious pattern to paedomorph emigration but emigrating efts exited the pond basin most often in the southwest quadrant (Fig. 3-5). Overall patterns of immigration differed significantly from emigration for females and males (Table 3-1 ). Although the directionality of immigrating adults appeared similar between the sexes (Fig. 3-4) patterns were significantly different (MRPP test P = 0.002). There were three distinct immigration events of adults but orientation patterns were significantly different between the sexes only during the third and largest of these events (Table 3-2). Differences in emigration between males and females (Fig. 3-4) were not significant overall or when distinct emigration events were compared (Tables 3-1, 3 2). There were two distinct emigration events of recently transformed striped newts comprising the 1996/97 cohort. The first emigration event took place from Oct. through Nov. 1996 and the second event from Apr. through Jun 1997 (Chapter 2). Immature newts (i.e., efts) comprised the first event whereas emigration later consisted mostly of recently transformed paedomorphs (Chapters 2 and 4). Patterns of emigration were significantly different between the eft and paedomorph life-history stages of the same cohort (Table 3-2). In addition to the eft emigration of 1996 a second emigration event of efts took place from Jun. through early Sep. 1998 (Chapter 2). Patterns of eft captures at OSP differed significantly between these two emigration events and considering all

PAGE 64

55 efts and all adults efts e x ited the pond basin in a different pattern from adults (Tables 31 32) Data for 44 uniquely-marked efts initially caught leaving the pond in the winter of 1996 and recaptured when they returned to breed in the winter of 1997 indicated that individuals tended to enter and exit the pond within the same quadrant. Sixty-four percent of these efts left and returned to OSP in the same quadrant (intervals O to 3 in Fig 3-6 ) and four individuals (9%) were caught leaving and returning to the pond at the same pair of pitfall traps (interval O in Fig. 3 6). The vast majority of individuals (84%) entered the pond basin within the same half they had left from the previous year (intervals O to 6). Dispersal Into Uplands I captured 831 newts in the upland drift fences during year one (Fig. 3-3A Table 3-3). Pond-side captures accounted for 73% of total captures and migration in year one consisted primarily of recently transformed efts. I captured newts at all of the upland fence sections (Fig. 3-3A ; Table 3-3) and in most (91.4%) of the pond-side pitfall traps During each period of migration the vast majority of newts were captured on the same s ides of upland drift fences. However for most movement events a small percentage of newts were captured in pitfalls on the opposite side of fences from the majority of captur es I believe this is because there was a small degree of wandering by some newts in the upland s a s they moved to or from OSP. Pond-side captures at upland fences in ye ar on e repre s ented three distinct periods of newt migration two emigration events and on e immi gr ation e v e nt ( Tabl e 3-4 ) Most newts captured on the pond-side of upland fe n ces in ye ar on e ( 76 % o f pond-side captures) were caught during the first emigration eve nt (i.e 1 ) whi c h o cc urr e d from Oct-96 through Feb-97 ( Table 3-4). Emigration

PAGE 65

56 during this period consisted almost exclusively of immature efts that had recently transformed. I captured far fewer newts (15% of pond-side captures) durin g emigration event two (E-2) which occurred from Apr. through Jul. of 1997 (Table 3-4) This emigration event was comprised of recently transformed paedomorphic newts (54% o f the migrating newts), as well as recently transformed efts and several adults that likel y had finished breeding and were moving back into the uplands. The third period of migration indicated by pond-side fence captures in year one was the result of an immigration event (i.e., I-3) that began in Oct-97 (Table 3-4). There was a major breeding migration of adults to the pond that began in Oct-97 and pond-side captures at this time probably resulted from adults that were moving toward the pond but happened to be captured on the pond-side of the upland drift fences (Table 3-4). Upland side captures of striped newts accounted for 27% of captures in year one. I captured newts at each of the five fence sections (Fig. 3-3A) at each distance from OSP (Table 3-3) and in most (81.4%) of the pitfall traps on the upland-side of the fences in year one. Upland-side captures occurred during three distinct periods of newt migration all of which were immigration events. These migration events (I-1 I-2 and I-3; Table 34) occurred during the same time periods as describe above for pond side captures (Table 3-4). Immigration event I-3 accounted for the largest proportion (54%) of upland-side captures in year one followed by event I-1 (29%) and I-2 (17%). All of these migration events consisted of adult newts moving toward OSP to breed (Table 3-4). I captured 495 newts in the upland drift fences during year two (Fig. 3-3B Table 3-3). In contrast to year one migration consisted primarily of immigrating adults. Pond side captures accounted for only 9% of total captures I captured newts at each of the two

PAGE 66

57 fence sections (Fig. 3-3B) at each distance from OSP (Table 3-3) but captures were recorded in less than half of the pitfall traps (42.8%) on the pond-side of the upland fences in year two. Pond-side captures at upland fences in year two represented two distinct periods of newt migration, one immigration event (i.e., I-3) and one emigration event (i.e. E-3). I captured few newts during both of these events; 16 during I-3 and 25 newts during E-3 (Table 3-4). Captures during migration event I-3 were adults that were moving to the pond to breed but were captured in pond-side traps as they wandered toward the pond. Captures during E-3 were recently transformed newts that were dispersing from OSP. In year two, I captured far more newts (91 % of total upland captures) on the upland-side of drift fences than on the pond-side (Table 3-3). I captured newts at all sections of drift fence and in almost all of the upland-side pitfalls (88.6%). Captures occurred only during a single immigration event (I-3; Table 3-4) and were exclusively of adults. The number of captures declined as the distance from the pond increased (Table 3-3). Based on estimated values, at least 360 newts (16% of the breeding migration) dispersed more than 500 m from OSP (Fig. 37). I estimated that 645 newts (29% of the breeding migration) dispersed at least 400 m. The estimate was the same for 300 m (645 newts). I estimated that 810 (36% of the breeding migration) and 908 (41% of the breeding migration) of newts dispersed from the pond at least 200 and 100 m respectively (Fig. 37). Based on these estimates, it appears that roughly 60% of the striped newts emigrated less than 100 m. However, as indicated by captures at the 500 m fences a substantial percentage of individuals comprising the 1997 / 98 breeding migration immigrated to O P from farther than 500 m. In fact one newt that was marked

PAGE 67

58 leaving OSP as an eft on 18 Nov-96 was recaptured on 4 Feb-98 a s it colonized F o x Pond a dispersal distance of ca. 685 m Discussion Orientation Striped newts of all life history stages (i.e. adults recently transformed paedomorphs and efts) emigrated and immigrated in a significantly nonrandom fashion although individuals were captured at all pairs of pitfall traps encircling the pond. Directionality of emigration and immigration differed significantly between and within sexes, with the exception of emigration patterns between females and males. Directionality of emigration also differed significantly among life history stages. Although the percentage of newts dispersing into the uplands decreased as distance from the pond increased I estimated that at least 16% of the breeding population dispersed more than 500 m. The distribution of habitats surrounding a breeding pond should have a fundamental influence on patterns of immigration revealed by captures of salamanders at the pond. Habitat preferences among species and/or differential survivorship in various habitat types might be apparent as individuals arrive at the breeding pond For example imagine an amphibian breeding pond in which one half of the uplands surrounding the pond were pine plantation while the other half remained native uplands. The pattern of captures at the pond would be expected to reflect the distribution of upland habitats One would predict significantly fewer captures along the half of the pond adjacent to the pine plantation as compared to the native upland half. This is because pond-breeding salamanders have the ability to select appropriate upland habitats and accurately navigate

PAGE 68

59 through uplands during migration, often using specific habitats ( deMaynadier and Hunter 1999 Hurlbert 1969 ; Madison, 1997; Madison and Farrand, 1998 Semlitsch 1981 ; tenhouse 1985 Shoop 1968). In this study although newts entered and exited the pond basin from all directions migration was nonrandom. Some directions were preferred over others but there were no obvious upland habitat features that could explain the newts orientation behavior. However I did not measure habitat variables in the uplands and individuals could have used micro-topographic features as cues to navigate toward the pond. In a similar study, Dodd and Cade (1998) concluded that movements of striped newts and narrowmouth toads were a reflection of the distribution of favorable upland habitats around the pond. Although the uplands at OSP were primarily sandhill habitat a small plantation of slash pine (with intact groundcover) was well within the dispersal capabilities of migrating newts (Fig. 3-2). In year one I often caught newts at a section of drift fence in the pine plantation. Newts could have resided within the plantation or have traveled through it en route to native sandhill. Nevertheless, this plantation represented only a small portion of the uplands and had no detectable effect on striped newt movements. Although upland-habitat preferences and microenvironrnental features I did not measure could have influenced the nonrandom pattern of immigration observed at OSP if measured over several seasons, orientation may in fact be random. It is possible that st riped newts are roughly evenly distributed in the uplands around OSP but that only a portion of the population migrates to the pond during any particular breeding event. If the portion of individuals moving was not indicative of the whole population then what

PAGE 69

60 truly should be random orientation would appear as nonrandom becau s e data were collected for a relatively short time. Patterns of newt emigration were also nonrandom and newts exited the pond basin in all directions. Efts emigrated predominantly in the southwest quadrant of the pond. The slope of the pond basin was shallowest in this quadrant and water depth during metamorphic events could have influenced the behavior of recently transformed efts as they left the pond. Adults on the other hand emigrated most often in the south southeast portion of the basin. Differences in aquatic habitat preference ( e.g. depth) between adult and immature newts might explain the varying emigration patterns although habitat preferences of both life history stages are unknown. Upland Dispersal Using upland drift fence arrays in year two I was able to estimate the percentage of the striped newt breeding population that migrated different distances (in increments of 100 m) from the pond. Captures at drift fences in the sandhill uplands surrounding OSP indicated that many striped newts (16%) dispersed more than 500 m from the pond. This is a conservative estimate because many individual captured in traps closer to the pond may have dispersed further than indicated by the traps. Captures at the drift fence surrounding the pond and at upland drift fences at the end of year one showed that a breeding migration of newts into OSP had begun before the installation of fences for year two (Chapter 2). Although the proportion of individuals caught at the pond before the new upland fence array was established was small (7% of the total) some newts alread y had moved toward the pond before the upland array was in place. Moreover immigrating adults did not arrive at the pond in a random fashion during this breeding migration. The upland fence arrays in year two were located north and southeast of O P

PAGE 70

61 and newts were caught at the pond with lowest frequency toward the north. Therefore the proportions of the breeding population caught at each distance from the pond in year two is likely an underestimate of the actual proportion moving to that particular distance Many pond-breeding amphibians have complex life-cycles and spend much of their adult lives in terrestrial habitats away from breeding sites. Distances that individuals disperse from breeding ponds have been reported for some species (Dodd, 1996 ; Semlitsch 1998 and references therein). It is clear that individuals disperse hundreds of meters from breeding sites into upland habitats some even thousands of meters. With few exceptions however, distance values usually have been presented for less than 10 individuals per species. The results from my study are the first estimates of dispersal distances for a breeding population of North American amphibians based on a substantial sample size. Conservation Implications Central to a successful amphibian conservation strategy is the protection of sufficient breeding and nonbreeding habitat. Studies of amphibian dispersal can provide the scientific basis for determining directional and distance components that can be used to establish protected areas around breeding ponds. Brown et al. (1990) used spatial requirements (i.e., distance dispersed from a wetland), among other data to recommend width of buffer zones" for wildlife protection at wetlands in Florida. Nevertheless lack of data for amphibians forced them to use rough estimates for most of the species considered. Further utility of dispersal distance data can be found in regulations to protect the flatwoods salamander (Ambystoma cingulatum) which as a result of severe population decline (Means et al. 1996) was federally listed as threatened (U.S. Fish and Wildlife ervice l 999). The U.S. Fish and Wildlife Service restricts specific

PAGE 71

62 silvicultural practices within 450 m of flatwoods salamander ponds. Additionally only selective timber harvest at specific times is allowed within a primary radius of 164 m around breeding ponds (U.S. Fish and Wildlife Service 1999). The width of the primar y zone was derived from a review of dispersal distances for pond-breeding salamanders of the genus Ambystoma (Semlitsch 1998) despite the fact that no data for A. cingulatum were available. This example underscores the need to determine dispersal distances for all pond-breeding amphibians. Semlitsch ( 1998) acknowledged that the core zone recommended for Ambystoma species may apply to some species of pond-breeding amphibians but certainly not all. My data show that recommendations for protecting terrestrial habitat for ambystomatid salamanders are inadequate for Notophthalmus perstriatus. Therefore it is not defensible to extrapolate data across taxa. Clearly a 164 m protected zone would not protect all of the striped newts breeding at OSP Based on extrapolation of dispersal distances revealed by upland drift fences a protected core zone extending ca. 1000 m from OSP would likely be needed to encompass all of the newts that breed there. Although they have great value as wildlife habitat, small, isolated wetlands in the United States are afforded little protection from development. Overall more than 50% of wetlands have been destroyed by development in the United States (Dahl 1990) and much of this loss has been small wetlands. In Florida a state with an extremely large number and diversity of wetlands isolated wetlands less than 0.2 ha receive no protection from development. This size threshold was adopted by the state s water management districts based on a consensus of scientific and regulatory opinion rather than on

PAGE 72

63 biological and hydrological evidence' (Hart and Newman, 1995). Small wetlands are just as vulnerable at the national level as they are in Florida. There is strong evidence that protection of core areas of terrestrial habitat surrounding breeding sites is crucial for persistence of amphibian populations and species. Data from OSP demonstrate that small, isolated wetlands can support breeding populations of salamanders that extend hundreds of meters into the surrounding uplands. Similar studies at other ponds and in different upland types are necessary because data on upland habitat requirements ( quality and quantity) of most amphibian species are lacking. Without this information designating terrestrial core habitat to conserve aquatic-breeding amphibians will largely remain guesswork, with generalizations made from data on relatively few individuals of a few species. However unless more protection is afforded to small isolated wetlands arguments to preserve uplands surrounding the wetlands are irrelevant.

PAGE 73

64 Table 3-1. Overall comparisons of directional orientation pattern s for s trip e d newt entering (immigrating) and leaving ( emigrating) One Shot Pond Putnam Co. FL. Standardized Comparison n test statistic p Immigrating vs. emigrating males 1159 486 -13.317 < 0.001 Immigrating vs. emigrating females 1489 645 -3 798 0.008 Immigrating males vs. females 1159 1489 -5.524 0.00 2 Emmigrating males vs. females 486 645 -0.437 0.2 Emigrating efts vs. emigrating adults 5008 1131 -67.639 < 0.001 Emigrating efts vs. emigrating paedomorphs of the same cohort 745 407 -9.506 < 0.001

PAGE 74

Table 3-2. Comparisons of directional orientation patterns for striped newts entering (immigrating) and lea ing ( emigrating) One Shot Pond Putnam Co. FL. Comparison Immigrating males vs. immigrating females Immigration Event 1 Immigration Event 2 Immigration Event 3 Emigrating males vs. emigrating females Emigration Event 2 Emigration Event 3 Emigrating efts during metamorphic Event 1 vs. emigrating efts during metamorphic Event 3 n 23, 13 22,66 1049 1290 15,68 430,484 745,4237 Standardized test statistic p 0.697 0.7 -0.130 0 .3 -4.008 0.006 0.686 0.7 -0.005 0.3 -3.599 0.01 0\ Vl

PAGE 75

Table 3-3. Numbers of striped newts captured in pitfall traps at drift fence arrays in the sandhill uplands surrounding One Shot Pond Putnam Co. FL. Drift fences were located at various distances from the pond See Fig. 3-3 for a depiction of the arrays. Pond-side Upland-side Total Year 1 Year 2 20 m 40 m 80 m 160 m 100 m 200 m 300 m 140 126 169 172 11 6 10 79 39 64 42 121 108 86 219 165 233 214 132 114 96 400 m 500 m 12 7 86 48 98 55

PAGE 76

Table 3-4. Captures of striped newts in upland fences around One Shot Pond, Putnam Co. FL during distinct periods of mo ement. ide of fences Migration Predominant direction Time period No inds of caetures event of newt movement of event caetured Descrietion Year 1 Pond-side E-1 away from pond Oct.-96 through Feb.-97 461 Emigrating efts Pond-side E-2 away from pond Apr.-97 through Jul.-97 91 Primarily emigrating paedomorphs and efts Pond-side I-3 toward pond Oct.-97 through Dec -97* 55 Immigrating adults Uplan d-side I-1 toward pond Oct.-96 through Jan. 97 65 Immigrating adults some emigrating efts Upland -side I-2 toward pond Apr.-97 through Jul.-97 36 Immigrating adults Uplan d-side I-3 toward pond Nov -97 through Dec.-97* 123 Immigrating adults Year2 Pond-side I-3 toward pond Dec. 97* through Mar.-98 16 Immigrating adults Pond-side E-3 away from pond Jun.-98 through Sep.-98 25 Emigrating efts Upland-side I-3 toward pond Dec -97* through Mar.-98 449 Immigrating adults *Fence arrays modified in early Dec.-97 0\ -.....)

PAGE 77

. t:Georgia Gul[f Mexico 68 South Carolina Atlantic Ocean Fig. 3-1. Present range of the striped newt. Note the hiatus (?) between the western and eastern portions of the range. This area may represent a true gap in the species distribution rather than an artifact of inadequate survey effort.

PAGE 78

69 Katha ri ne Ordway Preserve One Shot pond e) () Fox pond i 0 1 km ig 3-2. The location of One hot Pond within the Katharine Ordway Preserve Putnam o. L. The pond is surrounded primarily by sandhill up l ands but a small pine plantation (medium gray) is located to the west of pond. Ponds and lakes are indicated in black and north is at the top of the figure

PAGE 79

uplandide pitfall trap ponds ide / pitfall trap s ( I I \ A B 50 0m 400m 300m 2oom --~ lO O m. /1 60m On e Sho t Pond N pondide up l andide pitfa ll t ~ 7 trap t 1 . . F i g. 3-3 Upland drift fence arrays around One Shot P o nd Putnam Co. FL. A) Array design in year 1 B) Array design in y ear 2 One Shot Pond is depicted as a solid gray circle and the black circle around it represents the drift fence at the pond

PAGE 80

Immigrating males N 1 00 s Immigrating females N 140 s 71 Emigrating males 70 60 50 40 30 N s Emigrating females 50 s Fig. 3-4. Orientation patterns of immigrating and emigrating striped newt adults captured in pitfall traps at a drift fence encircling One Shot Pond Putnam Co. FL. Orientation was significantly different from random for all four patterns The length of the line s indicate the number of newts entering and exiting the pond basin at eac h pitfall trap

PAGE 81

72 Emigrating paedomorphs Emigrating efts s Fig. 3-5. Orientation patterns of emigrating striped newt paedomorphs and efts captured in pitfall traps at a drift fence encircling One Shot Pond Putnam Co. FL. Orientation was significantly different from random for both patterns. The length of the lines indicate the number of newts e iting the pond basin at each pitfall trap

PAGE 82

73 14 13 12 11 10 Cl'.) 9 r: Q.) 8 ..... s 4-1 7 0 1-4 Q.) 6 ..0 5 z 4 3 2 1 0 I I I I I 0 1 2 3 4 5 6 7 8 9 Interval Fig. 3-6. Distance intervals between locations of initial capture during immigration and recapture during emigration for individualy marked striped newts at One hot Pond Putnam Co. FL. Locations were determined by established pairs of pitfall traps at a drift fence encircling the pond. Distances between trap pairs (i.e. distance interval) wa approximately 10 m.

PAGE 83

2200 2000 en 1800 (1) ... 1600 (.) 4--( 1400 0 en 1-, (1) 1200 .D 1000 t:::: ""O (1) 800 ... ro 8 600 ... en .,l 400 200 0 0 100 200 300 400 500 600 700 800 900 1000 Distance from pond (m) Fig. 37. Estimated numbers of striped newt captures in pitfall traps at drift fences in the sandhill uplands around One Shot Pond Putnam Co. FL. Drift fences were located at 100 m intervals up to 500 m from the pond The zero point represents captures at a drift fence encircling the pond. See "Methods" for an e x planation o f ho th e "e s timated" number s were calculated.

PAGE 84

CHAPTER4 INFLUENCE OF GROWTH RATE ON LIFE-HISTORY EXPRESSION OF STRIPED NEWTS Introduction The expression of alternative phenotypes as a function of the environment has been documented for many species of plants and animals (Moran 1992; Scheiner 1993 Steams 1989; Whiteman 1994; Whiteman et al., 1996). Although the environment can have a profound influence, an individual's genotype may exert a strong effect on phenotypic plasticity as well (Scheiner, 1993; Steams 1989). Species that exhibit complex life cycles (Wilbur, 1980), such as many amphibians, are excellent models for studying expression of phenotypic plasticity (Newman, 1992). In these species there is a distinct ontogenetic change in an individual s morphology, physiology and often its habitat that occurs at metamorphosis. The larval stage is usually dedicated to feeding and growth whereas the adult stage disperses and reproduces (Wilbur, 1980). In anurans the size of an individual at metamorphosis and duration of larval period have proven to be extremely variable. A diversity of environmental factors affect these characters (Alford, 1989 Denver 1997 Denver et al., 1998; Kupferberg, 1997 ; Morin 1986 ; Skelly and Werner 1990 Werner 1986). ome sa lamander s also exhibit complex life cycles, although morphological changes associated with metamorphosis are not as extreme as in anurans (Wassersug, 1974 Werner 1986). In addition to those sp cies in which individual alway transform 75

PAGE 85

76 from an aquatic larval stage into a terre st rial or se miterrestrial s tage ( i .e., obligate metamorphosis) there are species in which individual s reproduce while the larval or branchiate morphology persists. This paedomorphic habit (Gould 1977) is fixed in those species that have evolved a simple life cycle. However in other species namely some mole salamanders and newts paedomorphosis is facultative (Duellman and Trueb, 1986 ). In facultatively paedomorphic species (referred to as paedotypic by Reilly et al. 1997 ) individuals may follow one of two developmental trajectories. Once an individual reaches a threshold size it may transform into a terrestrial form (i.e., metamorphic phenotype). Alternatively it may remain in the aquatic environment and mature while retaining larval characteristics (i.e., paedomorphic phenotype; Whiteman 1994). In facultative species paedomorphic and metamorphic individuals may be found in the same pond; furthermore both phenotypes may result from the same cohort (Chapter 2; Semlitsch et al. 1990 ; Winne and Ryan 2001 ). The expression of alternative phenotypes is believed to be environmentally induced and to depend on an interaction between an individual s genotype and the aquatic environment in which it develops (Whiteman, 1994, 1997). Experiments almost exclusively with species of Ambystoma, have identified a variety of biotic and abiotic factors that influence expression of the two alternative phenotypes. These include density of individuals in experimental tanks (Licht, 1992 ; Ryan and Semlitsch 1998; Semlitsch 1987) presence of fish predators (Jackson and Semlitsch 1993) drying regime (i.e., hydroperiod) of tanks (Semlitsch and Gibbons 1985 ; Semlitsch 1987 ; Semlitsch et al. 1990) and potentially nongenetic maternal effects (Licht, 1992 ; Semlitsch et al. 1990). Because growth rate is implicated as an

PAGE 86

77 important determinant oflife history expression by several ecological models (Werner 1986 Whiteman 1994 ; Wilbur and Collins 1973) the effect of food availability has been the focus of numerous studies. Results of these studies have varied, depending on the species of Ambystoma studied and type of food regime (Licht, 1992; Ryan and Semlitsch in review ; Semlitsch 1987; Whiteman et al. 1996). Furthermore studies with Ambystoma have suggested that the expression of facultative paedomorphosis has a genetic basis (Harris et al. 1990; Licht 1992; Semlitsch and Gibbons 1985 ; Semlitsch and Wilbur 1989 ; Semlitsch et al. 1990). Less attention has been given to factors affecting the expression of alternative life-history pathways in other groups of facultatively paedomorphic salamanders. A better understanding of the proximate and ultimate causes of this phenomenon can only be achieved through studies of other species and across populations within species (Whiteman, 1994). This is especially important considering that paedomorphosis (facultative and obligate) has evolved numerous times (Duellman and Trueb 1986 Griffiths 1996 ; Ryan and Bruce, 2000 ; Shaffer et al. 1991; Shaffer 1993; Shaffer and Voss 1996). Species in the family Salamandridae (newts) provide an excellent opportunity to study the evolution and maintenance of facultative paedomorphosis Newts exhibit a diver s ity of reproductive strategies, and facultative paedomorphosis occurs in several genera (Griffiths, 1996; Halliday 1990; Petranka 1998). Within North American newts (Taricha and Notophthalmus) facultative paedomorphosis is known to occur in Taricha granulo a (Marangio 1978) Notophthalmu viridesce ns (Brandon and Bremer 1966 ; Harris 1987 Healy, 1970 1974a), Notophthalmus per triatus (Dodd et al. in press ;

PAGE 87

78 Chapter 2) and possibly in Notophthalmu s meridionalis (Mecham, 1968). Nevertheless the expression of facultative paedomorphosis has only been experimentally explored in a single study of Notophthalmus viridescens dorsalis (Harris 1987). Harris (1987) found that density which influenced growth rate had a significant affect on expression of paedomorphosis. Based on the literature and my research with Notophthalmus perstriatus, it appears that hypotheses invoked to explain the expression and maintenance of paedomorphosis may differ between Ambystoma and Notophthalmus species. In fact several different hypotheses may explain the maintenance of facultative paedomorphosis (Whiteman, 1994, 1997). Therefore, results of experiments with Ambystoma may not be entirely applicable to newts or other salamander species. Field research I conducted at several breeding ponds in north Florida revealed that the striped newt (N. perstriatus) has a complex life history (Fig. 1-1 ). Within a single breeding season, I found that some larvae transformed before maturing and left the pond (metamorphs), whereas others remained in the pond and matured while retaining their larval morphology (paedomorphs ). This same pattern appeared to have occurred during three consecutive breeding seasons. The proximate cause(s) of the expression of the paedomorphic phenotype in some individuals but not others is unknown. A growth advantage for some individuals over others could explain the dichotomy in expression of the paedomorphic versus metamorphic life-history. Faster growing larvae of other newt species (N. v. dorsalis, Harris 1987; Triturus carnifex, Kalezic et al. 1994) were more likely to become paedomorphic than slower growing larvae. Some models of amphibian metamorphosis take into account the potential influence of growth rate on the expression of paedomorphosis in salamanders (W emer

PAGE 88

79 1986 Wilbur and Collins 1973 Whiteman 1994). In the Wilbur and Collins (1973) model physiological processes that initiate metamorphosis are related to recent growth history of the individual. Larvae with a growth advantage are predicted to remain in the aquatic environment and continue to grow eventually expressing the paedomorphic life history pathway. Slow growing larvae are predicted to metamorphose once they have reached some population-specific or species-specific threshold. Slow growing larvae thereby escape the potential density-dependent influence of competition with paedomorphs and larger larvae (Whiteman, 1994; Wilbur and Collins, 1973). Werner 's ( 1986) model of amphibian metamorphosis also has been extended to paedomorphosis. In this model the "decision" of an individual to remain in the aquatic environment or metamorphose and move into the terrestrial habitat is a result of growth potential in each habitat weighed against the habitat-specific risk of mortality. All else being equal larvae with a growth advantage (i.e., relatively fast growing larvae) should be more likely to become paedomorphic as compared to slower growing larvae. In addition to life-history expression, the Wilbur-Collins (1973) model predicts size and age at metamorphosis. This model can be viewed as a flexible-rate or optimal growth rate model because mass-specific growth rates throughout much of the larval period dictate when metamorphosis should occur and at what body size. An alternative model initially proposed by Smith-Gill and Berven (1979), and later extended by Travis (1984) and Leips and Travis (1994), differs from the Wilbur-Collins model in that rate of differentiation (i.e. development) of larvae as opposed to growth per se dictates larval period. Therefore this alternative is referred to as a fixed differentiation-rate model. In thi model larval period i s determined or fixed at some specific point in development.

PAGE 89

80 Differences in growth rate after this stage o f development are only predict ed to influence body size at metamorphosis not larval period Travis (1984) s ug geste d that this fixation period occurs early in development. Size at metamorphosis is still predicted to be a result of larval growth rate however and because of this both models account for the ex tr eme plasticity in metamorphic body size of amphibians. Unlike the optimal growth-rate model of Wilbur and Collins the fixed-rate model does not apply to paedomorphosi s however. I tested the hypothesis that growth rate influences the expression of paedomorphosis in striped newts. I used varying food levels to generate different growth trajectories of larvae raised individually in the lab. The treatments included high and low levels of food as well as switches in food levels. The objectives of the experiment were: 1) to determine the influence of growth rate on the expression of alternative life histories (i.e., metamorph vs. paedomorph), 2) to test the ability of the Wilbur and Collins model to predict the life-history expression 3) to determine the influence of growth rate on larval period and size at metamorphosis, and 4) to simultaneously test the applicability of two types of models ( optimal rate vs. fixed rate) for predicting metamorphosis in Notophthalmus perstriatus. Methods Experimental Design I used different food treatments ( constant High = HH constant Low = LL switch from Low to High = LH and switch from High to Low = HL) to generate growth trajectories among four experimental groups of striped newt larvae. I used a randomized block design with the two shelves treated as blocks. Ten individuals were randomly

PAGE 90

81 assigned to each of the four treatments. Newts were reared individually in plastic containers which were randomly assigned a position on one of two shelves. The experiment was designed to standardize all variables with the exception of food availability. Procedures Newts were housed in Sterilite ClearView storage boxes (43 cm X 28 cm X 16 cm). These were filled with 9 liters of aged water, which was changed every 57 days. During water changes, each container was cleaned with a scrub brush and antibacterial soap then rinsed with tap water. Light was supplied by florescent bulbs mounted above the containers. Automatic timers controlled the lights and were adjusted several times during the experiment so that the light/dark regime approximated the natural light cycle. Newts were housed indoors, although the temperature fluctuated daily and seasonally, tracking the temperature variation for Gainesville FL during the study period. I weighed and measured each newt twice a month ( every 17 days on average). I weighed and measured newts singly by first removing an animal from its container and placing it in a plastic sandwich bag. I measured body length to nearest millimeter with a clear plastic ruler while the newt was in the bag. Because it was difficult to see the posterior edge of the vent on small larvae, I measured from the tip of the snout to the anterior edge of the rear legs where they meet the body (i.e. body length) rather than measure standard snout-vent length (SVL). Body length is highly correlated with SVL (Spearman Correlation test: r = 0.988 P < 0.0001 derived from measurements of 30 preserved triped newt larvae) Each newt was weighed to the nearest 0.1 g by transferring it to a mall pla tic cup filled with aged tap water that had already been tared on a Ohaus brand di g ital balance I also recorded the degree of swelling of the vent (not swollen slightly

PAGE 91

82 swollen, and swollen), condition of the gills and tail fin (full vs. regressing) and other external characteristics ( e.g. presence of secondary sexual characters). These external characters can be used as indicators of maturation (i.e. vent swelling) and metamorphosi s (i.e. gills and tail fin regress). In addition to using secondary sexual characters I dissected each newt and examined its gonads under a dissecting scope to determine if it was sexually mature. Newts were derived from eggs laid by paedomorphic adults that I collected on 3 Jan-99 from a breeding pond (Blue Pond) located on the Katharine Ordway Preserve Putnam Co., FL. Striped newts are winter-breeders and these individuals were already in reproductive condition. Forty-eight paedomorphs were housed in 12 glass aquaria (2 females and 2 males per aquarium) and fed black worms every 2-3 days. Each aquarium contained an air stone a constant water source, a standpipe and vegetation on which to lay eggs. I collected eggs every 7-10 days and transferred them to shallow tubs containing dechlorinated water. Striped newt females lay eggs one at a time and have a prolonged breeding season of several months (Chapters 1 and 2). Hatchlings were transferred to another tub and raised on a diet of zooplankton until they were large enough to consume whole black worms. Water in the hatchling tubs was changed every 57 days. I waited until the larvae were large enough to eat black worms because this food source was easy to quantify and prior experience had shown that newts thrive on black worms. Although small larvae would readily eat Artemia they always died within 2 -3 days ; they thrived however on native zooplankton. I grew zooplankton in large (1.2 m diameter) plastic (high-density polyethylene) tubs and harvested them with a fine-mesh net. The tubs were manufactured by the Lerio Co. Kissimmee FL. The tubs contained

PAGE 92

83 c a 7 00 liters of well water soil collected from a dried pond basin at the Ordway Preserve leaf litter and aquatic vegetation. I also inoculated the tubs several times with zooplankton collected from local ephemeral ponds. Once enough larvae were reared to a size capable of consuming black worms I began the experiment. The experiment was initiated on 16 Jun-99 and continued until 27 Dec-99 for a total of 206 days. An individual was removed from its container when it showed obvious evidence of metamorphosis. This included the appearance of a bold stripe along the dorso-lateral portion s of the body and regression of the gills and membranous tail fin. A newt nearing metamorphosis also reduced its food intake and often floated at the surface of its container. At this point the newt was transferred to a new container and placed on a damp paper towel to allow gill regression to continue. Once the gills had been completely resorbed usually 2-3 days after removal from the plastic container the newt was anesthetized in a chlorotone solution weighed and measured. Each newt was tagged with a unique label and preserved in 10% formalin for dissection later. Newts that did not metamorphose remained in the experiment until its termination on Dec. 27. At this point they were anesthetized weighed measured and preserved. Food Treatments High food-treatment animals were fed black worms (Lumbri cu lu s ) ad libitum the L ow food-treatment was calculated as of the average number of worms consumed by the High food-treatment animals during the preceding feeding interval. Low food treatment animals were fed every 3-5 days. Therefore although the nwnber of worms con s umed by individual s in e ach food treatment increased as the newts grew (i.e appro x imatin g ma ss-s p ec ifi c food levels) the animals on Low food always received of that o f th e Hi g h food animal s The LH and the HL treatment animals experienced a shift

PAGE 93

84 in available food during the experiment. On 17 Jul-99 I initiated the switch in food levels. The LH animals which previously had received of the amount of food as the High food level animals now received food ad libitum. The HL animals were now only fed of the average number of Lumbriculus consumed by High-treatment animals. High-food newts always had Lumbriculus available in their tanks ; Low-food newts consumed all of their Lumbriculus within a few hours of feeding. The switch was initiated when mean mass of the HH and HL animals was at least 0.7 g and their body length was at least 22 mm. These body sizes are greater than the minima required for an individual striped newt to initiate metamorphosis but below the minima required for maturation in paedomorphic individuals (Chapter 2). Therefore, the switch in food levels occurred near the body size where metamorphosis can be initiated in nature but before maturation occurs. I chose this point to make the food-level switch because it should be a critical time in development. Once the minimum size for metamorphosis is reached an individual may express the metamorphic life-history pathway or postpone metamorphosis and continue to grow possibly expressing the paedomorphic pathway later. Dissections Internal and external characteristics were used to determine sex and reproductive condition. External secondary sexual characteristics that indicate maturity in salamandrids include: swelling of the vent development of genial glands on the side of the head of males and presence of nuptial excrescences on the rear limbs and toe tips of males (Duellman and Trueb 1986). Mature striped newts are sexually dimorphic and in addition to the characters above, the shape of the cloaca! lips differ between males and females. A conspicuous cloaca} gland which is whitish in color is visible at the posterior edge of the cloaca in mature and some immature males (Dodd 1993 ). I

PAGE 94

85 dissected all newts and examined their urogenital systems under a dissecting scope to determine reproductive condition. Dissection also allowed me to confirm the sex of newts that exhibited secondary sexual characteristics and determine the sex of immature animals. Maturity of females was indicated by the presence of enlarged ovarian follicles and enlarged convoluted oviducts. Maturity in males was indicated by the presence of enlarged convoluted and pigmented Wolffian ducts (Ryan and Semlitsch, in review Semlitsch 1985) Data Analysis Response variables used in statistical analyses were body length, body mass, larval period (number of days from initiation of the experiment to metamorphosis) and morph type (metamorph or paedomorph). Prior to analysis each data set was tested for normality and heteroscedasticity. When assumptions of parametric tests were violated nonparametric methods (Hollander and Wolfe 1999) were used to test for treatment effects. All statistical analyses, with the exception of goodness of-fit tests were performed using SAS version 6.12 (SAS Institute Inc., 1990). I used a two-part analysis to test for differences in growth rates among the four treatment groups. First a MANOVA (on body length and mass) was used to test for differences among treatments the day before the switch in food level was initiated Second I tested for differences among growth rates during the remainder of the experiment. For each individual that remained in the experiment (i.e. had not yet metamorphosed) I subtracted the size (for BL and mass independently) of the individual on th e most recent day it had been measured (before metamorphosis or at the end of the ex periment for paedomorphs) from its size on the day before the food switch. This value wa then divided by the number of days from the switch until the date a newt had been

PAGE 95

86 most recently measured to give a growth rate (unit of growth/day). Growth rates among treatments (excluding HL) were then tested with a Kruskal-Wallace test (Hollander and Wolfe 1999). Growth rates of HL animals were excluded because all individuals had metamorphosed before the first scheduled date for measuring individuals following the food switch. I used a MANOV A to test for differences among treatments on body length mass and larval period for those individuals that metamorphosed before the end of the experiment. Following significant univariate tests in MANOV A tests a Student Newrnan-Keuls (SNK) test was used to compare treatment means. For each MANOVA if Wilks' A was significant, alpha was adjusted for univariate tests with the Bonferroni method (Sokal and Rolf, 1995). I used a goodness-of-fit test (Sokal and Rolf 1995) to analyze the different frequencies of metamorphs and paedomorphs across treatments and between sexes. For influence of treatment on morph type, I calculated the number expected for each morph in each treatment ( expected metamorph:paedomorph = 7.75:2.25) by dividing the total number of metamorphs and paedomorphs by four. For influence of sex I partitioned the observed numbers of each morph between the sexes to calculate expected numbers (expected metamorph:paedomorph = 14:4.5). Goodness-of fit was used also to determine if sexes of the newts were distributed equally among the treatments. Expected values were derived by dividing the total number of males and females by four (expected male:female = 5.75:3.5)

PAGE 96

87 Results Larval Growth On day one of the experiment there were no significant differences in mass ( Kruskal-Wallace test: H = 2.864 df = 3 P = 0.4131) or body length ( ANOVA: F 3 3 6 = 0.54 P = 0.6605) of newts across all four food treatments (i.e. HR HL LL LH). On the day before the switch in food levels ( day 31 ) MANOV A indicated significant differences in size of newts among treatments even when alpha was adjusted with the Bonferroni method (Wilks "A = 0.219 F 6 66 = 12.53 P < 0.0001). Univariate tests were significant for both mass (F 3 3s = 39.41 P < 0.0001) and body length (F 3 3 5 = 24.61 P < 0.0001) ; block was not significant. SNK tests showed that for both body length and mass, the HR and HL treatment animals did not differ nor did the LL and LH animals. However HR and HL newts had significantly greater mass and body length than the LL and LH newts. Therefore before the switch in food levels, individuals receiving the H food treatment had grown faster than individuals fed at the L level (Fig. 4-1 ). The switch in food levels was initiated on day 32 (17 Jul-99) of the experiment. By this point newts in the HR and HL treatment groups had slightly surpassed the minimum size required for metamorphosis of N. perstriatus in nature (Chapter 2) whereas the LL and LH newts remained smaller than this minimum size. Analysis of growth rates following the switch day does not include any HL individuals because all of them metamorphosed before day 48 of the experiment the first day after the switch when ma ss and body length wer e ne x t measur e d. After the food switch the growth rates of HL n e wt s promptly increased and s oon their rate paralleled those of the HR newts whereas th e LL newt s continued to grow slower than individuals receiving the H food level (Fig 4 -1 ). A fte r the witch in food level s g rowth rates measured both as differences in mass

PAGE 97

88 and body length differed among remaining branchiates in the HH LL and LH treatments (Kruskal-Wallace test for mass: H = 15.694 df = 2 P = 0.0005 ; for body length: H = 15.094 df = 2 P = 0.0004; Fig. 4-1). Overall growth rates in mass and body length were greatest for LH newts followed by HH then LL treatment animals (Fig. 4-2 ). Size at Metamorphosis and Larval Period Univariate tests from a MANOV A on mass, BL and larval period (Wilks A = 0.119 F 9 66 = 7.57, P < 0.0001) were not significant for either mass at metamorphosis (F 3 22 = 2.71, P = 0 0696) or body length (F 3 22 = 0.27 P = 0.8496 ; Table 4-1) ; block was not significant in any of the univariate tests. Larval period on the other hand was significantly different among the food treatments (F 3 22 = 19.63, P < 0.0001 ; Table 4-1). A SNK test showed that the larval period did not differ between the HH and HL groups but both of these differed from the LL and LH groups, which differed from each other. The relationships between mass and body length at metamorphosis and larval period is presented in Fig. 4-3. Life-history Pathway Expression At the termination of the experiment, 22.5% of the individuals were mature branchiates and the rest had metamorphosed earlier in the experiment before attaining maturity. All newts considered paedomorphs were mature based on external and internal characters. Paedomorphic females each had many enlarged pigmented follicles in their ovaries, as well as enlarged and convoluted oviducts. All paedomorphic females also had swollen vents. The Wolffian ducts (i.e. vas deferens) of all paedomorphic males were enlarged pigmented and convoluted. Externally paedomorphic males had very swollen vents, relatively enlarged rear limbs with cornifications on their toe tips and well developed hedonic pits on each side of their head. A light colored gland was obvious at

PAGE 98

89 the posterior end of their v ents. None of the metamorphic newts had near the degree of cloaca! swelling exhibited by the paedomorphs. The rear limbs of metamorphic males were not enlarged and their toe tips were not cornified ; hedonic pits were barely visible in only a few immature males. Internally neither the Wolffian ducts nor the oviducts were enlarged or convoluted in metamorphic newts. Follicles in the ovaries of metamorphic females were small and undeveloped. External gills were present in all paedomorphic newts at the end of the experiment, and the namesake stripe was well developed. External gills of experimental paedomorphs were not as large and filamentous as gills of paedomorphic striped newts observed in natural ponds (S. A. Johnson pers. obs.). Distribution of paedomorphs and metamorphs was not even across treatments and no paedomorphs were produced in the HL treatment (Fig. 4-4). Nevertheless a goodness-of-fit test did not detect significant differences in life history expression across the food treatments (G = 6.968 df = 3 P = 0.062). However, as compared to the overall sex ratio of metamorphs and paedomorphs in the experiment (both morphs M:F = 1 :0.6) significantly more females became paedomorphic (females, paedomorph:metamorph = 1: 1.34) than males (males p:m = 1 :6.7; G = 6.333, df = 2 P = 0.042 ). Males and females were distributed equally among treatments (G = 1.664, df = 3, P = 0.680). By the end of the experiment all paedomorphs were longer than the metamorphs and all but one had greater mass than the metamorphs (Fig. 4-5). The body size of paedomorphs at the end of the experiment differed among the three treatments in which paedomorph s were produced. Mean mass and body length were greatest in paedomorphs produced by the HH food treatment followed by paedomorphs in the LH then the LL treatment (Table 4-2).

PAGE 99

90 Discu sion Differences in food levels resulted in differences in growth rates among treatments before and after the food-level switch (Figs. 4-1 and 4-2). Despite these differences, mass and body length were not significantly different at metamorphosis for those newts expressing this life-history pathway (Fig 4-3 Table 4-1). However treatment had a significant effect on duration of larval period. Mean larval period of HH newts did not differ significantly from that of HL newts but both of these were significantly different from the LL and LH treatment individuals (Fig. 4-3 Table 4-1 ) which were significantly different from each other. The paedomorphic life-history pathway was expressed by newts in all treatment groups except the HL group (Fig. 4-4). Nonetheless the proportion of paedomorphs and metamorphs resulting from each treatment did not differ significantly. Therefore, growth trajectories generated by the four different food regimes did not affect life-history pathway expression. There were however significant differences in the expression of life-history pathway among the sexes. Proportionately more females became paedomorphic than expected considering the overall sex ratio of animals in the experiment. Expression of Alternative Life-history Pathways Optimal growth-rate models primarily the Wilbur-Collins (1973) and the Werner ( 1986) models make predictions about the expression of life-history pathways in salamanders. Growth rate is also central to Whiteman s ( 1994) three alternative hypotheses to explain the maintenance of facultative paedomorphosis. Each of these models essentially makes the same predictions with regard to life-history pathway expression that should have resulted from treatments in the experiment. High-High and LH newts should have expressed the paedomorphic phenotype whereas LL and HL

PAGE 100

91 n ewts s hould have transformed before maturing. However the results for life-history pathwa y expression clearl y did not fit these predictions. I found no significant influence of treatment and thus growth trajectory on life history pathway expression. As predicted by the Wilbur-Collins model HL newts transformed shortly after they were switched from a High to Low food ration and none became paedomorphic. However all but two of the HH newts also transformed shortly after the switch and there was not a significant difference in larval period between newts in the two treatments. No paedomorphs were predicted to result from the LL treatment. Nevertheless there were more paedomorphs in this treatment than any of the other three (Fig. 4-4). The LH treatment was predicted to produce only paedomorphs but seven larvae in this treatment transformed. To reiterate food level did not influence an individual s expression of life history pathway. The role of food level on life-history expression has been tested in mole salamanders. Constant food levels did not significantly influence the expression of paedomorphosis in A. gracile or A. talpoideum (Licht 1992; Sernlitsch 1987). Additional experiments with A. talpoideum revealed a significant effect of food level for individuals that received relatively low food rations late in development (Ryan and emlitsch in review) counter to predictions of the Wilbur-Collins model. Life-history expression of striped newts depended on sex in my experiment. Fe mal es had a significantly greater propensity to become paedomorphic than males. The sex r a tio o f paedomorphs wa s female biased ( m : f = l : 2) whereas the sex ratio of m e tamorph s wa s mal e bia s ed (m:f = 5 :2 ). Female-biased sex ratios in paedomorphs ha e

PAGE 101

92 been reported for several species of mole salamanders and newts and may play a role in maintenance of paedomorphosis in facultative species (Whiteman 1997). Even though food level and therefore growth rate did significantly impact the expression of life-history pathways in striped newts, other ecological conditions certainly have at least some effect on this process. My fieldwork indicates that hydroperiod is an important determinant of life-history expression. If pond hydroperiod is too short paedomorphosis is not possible. Larval newts in a pond with a short hydroperiod will not have enough time to mature and will metamorphose before the pond dries assuming they have reached the minimum threshold size. Ifhydroperiod has been long enough for some larvae to mature but the pond dries before the reproduction is complete, paedomorphic newts will initiate metamorphosis to escape the drying pond. Pond hydroperiod influences life history in mole salamanders as well (Semlitsch and Gibbons 1985; Semlitsch 1987; Semlitsch et al. 1990). Along with ecological factors, the genetic component of life-history expression in salamanders has been an arena of active research contributing significantly toward an understanding of paedomorphosis (Harris et al. 1990; Ryan and Semlitsch in review Shaffer and Voss 1996 ; Semlitsch and Gibbons, 1985; Semlitsch and Wilbur 1989 ; Semlitsch et al. 1990; Tompkins 1978; Voss 1995; Voss and Shaffer 1997 2000). Tompkins (1978) initially reported that paedomorphosis in the axolotl (Ambystoma mexicanum, an obligate paedomorphic species) was caused by homozygosity of a single recessive gene. The work of Voss ( 1995) later disproved this idea and additional experiments supported the idea that paedomorphosis is controlled in the axolotl primarily

PAGE 102

93 ia a major gene effect with smaller affect loci presumably influencing trait expression under certain environmental conditions' (Voss and Shaffer 1997). Several studies have revealed genetic variation in life-history expression among local populations of the mole salamander a facultatively paedomorphic species (Harris et al. 1990; Semlitsch and Gibbons, 1985; Semlitsch and Wilbur, 1989 Semlitsch et al. 1990). This variation suggests that facultative paedomorphosis also has a genetic basis. The variation is assumed to have arisen through differential selective pressure on each morph resulting from varying hydroperiods among ponds. Salamanders from ponds with long hydroperiods had a greater propensity to become paedomorphic than those from sites that were more ephemeral (Semlitsch and Gibbons, 1985; Semlitsch and Wilbur 1989). Although the specific genetic architecture controlling expression of alternative phenotypes in facultatively paedomorphic salamanders is still unknown it is clear that the genetic system is complex. Variability exists in the system among species and among populations within species (Shaffer and Voss, 1996). Experiments with Ambystoma sp. and Notophthalmus viridescens have provided evidence that facultative paedomorphosis is likely controlled by a polygenic system (Licht, 1992; Harris, 1987 Semlitsch et al. 1990 ; Voss 1995; Voss and Shaffer 1997). Such a system could involve a major gene effect with loci of smaller effect being influenced by environmental factors, or the system could be comprised of multiple loci, each with a relatively small effect (Ryan and emlitsch in review Voss 1995 ; Voss and haffer 1997). Although my experiment was not designed to reveal the genetic component of paedomorphosis in N. perstriatus the results rule out the possibility of a simple single-locus recessive-allele explanation.

PAGE 103

94 If a single recessive allele controlled paedomorphosis in striped newts all larvae in th e experiment would have become paedomorphic. This is because each experimental animal would have been homozygous for paedomorphosis since the parent generation consisted entirely of paedomorphs. Based on the evidence currently available life history expression in striped newts may be controlled by a polygenic system. Metamorphosis and Model Applicability Optimal growth-rate and fixed differentiation-rate models types make the same predictions regarding body size at metamorphosis for striped newts in the experiment but they differ in their predictions of larval period duration. Food treatments in the experiment were predicted to generate growth trajectories similar to those shown in Fig. 4-6. Both models predicted that body size at metamorphosis (i.e. mass and body length) should follow the pattern of (HH = LH) > HL > LL (Fig. 4-6). For the fixed differentiation-rate model I assumed that the fixation period should have occurred early in larval development (i.e. before the food switch) as Travis (1984) proposed. The difference between the two model predictions should be manifest in duration of larval period. The optimal rate model (Fig. 4-6A) predicts a pattern ofHL < HH < LH < LL for time to metamorphosis. If I assume that the fixed-rate model applies to N. perstriatus and that fixation occurred before the switch in food levels, then duration of larval period is predicted to be (HH = HL) < (LH = LL) (Fig. 4-6B). Size at metamorphosis was not significantly different among the four treatments for either mass or body length for those newts expressing the metamorphic life-history pathway Furthermore mean body length across treatments did not fit the pattern predicted by both models. Mean body length of the newts in the HH and HL treatments was predicted to be the same and the mean body length of animals in these two

PAGE 104

95 treatments should have exceeded those of the HL treatment followed by the LL animals. Mean body length of HH metamorphs was closest to that of the LL metamorphs followed by newts in the LH treatment; HL newts exhibited the smallest mean body length at transformation (Table 4-1). Although not significantly different mean mass at metamorphosis followed the pattern predicted by both models. Both the HH and LH newts had the same mean mass at metamorphosis followed by the HL newts and then the LL animals (Table 4-1). Duration of larval period differed among the four food treatments. Mean larval periods for the HL newts and the HH newts were not significantly different. However larval period for individuals in both of these treatments was significantly shorter than the LH animals which in tum were shorter than the LL newts (Table 4-1 ). Despite the lack of statistical significance between HL and HH newts, larval duration across the four treatments matched the prediction of the Wilbur-Collins model for metamorphic animals. The data were not consistent with the pattern predicted by the fixed differentiation-rate model. The majority of experiments to test the applicability of both types of models have been conducted with anuran larvae. Results have been variable with some experiments providing support for the optimal growth-rate model (Alford and Harris 1988; Wilbur and Collins 1973) and the fixed differentiation-rate model (Beck, 1997; Smith-Gill and Berven 1979). Nonetheless other experiments have shown that neither model fit the data well (Hensley 1993 ; Leips and Travis 1994; Pandian and Marian 1985 Travis 1984 ) Crump (1981) even suggested that in addition to rates of growth and

PAGE 105

96 differentiation accumulation of energy may al s o be ke y to predictin g amphibian metamorphosis Few studies have tested the applicability of the two model s in salamander s and discussions have usually focused on the predictions of the optimal growth-rate model. The results of Beachy s (1995) experiment with Desmognathus ochropha eus strongly suggest that the Wilbur-Collins model is inappropriate for this stream-dwelling salamander. However because his food treatments only consisted of constant levels of food Beachy ( 1995) admitted that his findings do not allow for a complete assessment of the applicability of the model. In a similar experiment with Amb y stoma gracil e (Licht 1992) there were no differences in body mass at metamorphosis for salamanders receiving constant low or high food rations. However the high food salamanders metamorphosed significantly sooner than the low food animals (Licht 1992) consistent with both models as portrayed in Fig. 4-6. Nevertheless since the food treatments did not vary during the larval period the ability of this experiment to serve as a fair test of the Wilbur-Collins model is arguable. Although the experimental design of Ryan and Semlitsch (in review) is appropriate for testing predictions of metamorphosis for Amb y stoma talpoideum made by the Wilbur-Collins model the emphasis of their experiment was on expression of phenotypes (i.e. metamorph vs. paedomorph ). Unfortunately they did not present data for treatment effects on size at metamorphosis and larval duration of metamorphs in their experiment. Therefore it appears as if this experiment with Notophthalm u s perstriat u s is the first to evaluate predictions of the optimal growth-rate and fixed differentiation-rate models for a salamander.

PAGE 106

97 As noted above differences in mass and body length did not vary significantly across the four treatments although the distribution of means for body mass were consistent with both models. The distribution of mean body length across the treatments did not fit the predictions of either model. Had I measured energy accumulation, as suggested by Crump (1981), rather than body length and mass, I may have found stronger support for one of the models. Response variables measured in experimental tests of both models have usually been body mass and/or body length. Use of these parameters rather than some metric of general body condition may in part be responsible for the variation in model applicability found by ecologists. In fact, Wilbur and Collins (1973) stated that a more complete understanding of metamorphosis may suggest that some aspect of body quality rather than quantity is the critical factor in the initiation of metamorphosis.' Differentiation rate, at least in some species, appears to be one measure of body quality that is important for predicting larval period. Larval periods of striped newts were not consistent with predictions of the fixed differentiation-rate model, regardless of when in development one assumes that differentiation is fixed. Duration of larval periods were consistent with the optimal growth-rate model, although larval period did not differ significantly between the HL and HH newts. Neither of the two classic models of amphibian metamorphosis (i.e., optimal growth-rate model or fixed-differentiation rate model) accurately predicted body length or mass at metamorphosis for striped newt larvae. The discrepancies between model predictions and results of this study as well as experiments cited earlier show that neither model is applicable for all situations Such discrepancies may be caused in pa.it by genetic differences among taxa ai1d / or populations. Additionally as suggested by

PAGE 107

98 Crump (1981), standard measures of body size (e.g., body mass and body length) might not be the appropriate response variables to consider when testing predictions of ecological models. Experimental and field-collected data for N. perstriatus provide qualified support for Whiteman's (1994) "paedomorph advantage" hypothesis. This hypothesis largely based on the Wilbur-Collins model predicts that larvae with relatively high growth rates should become paedomorphic, whereas slow growing larvae should metamorphose. Growth rate did not have a significant influence on life-history pathway expression for striped newts in my experiment, counter to predictions of the the hypothesis. Nonetheless other life history parameters provide support for Whiteman's (1994) paedomorph advantage" hypothesis. For example, paedomorphs reproduce at a younger age than metamorphs of the same cohort (Chapter 2). Additionally, as predicted by the paedomorph advantage" hypothesis the minimum body size for metamorphosis in striped newts is smaller than the minimum size for paedomorphosis (Chapter 2). Although ecological factors affect the expression of paedomorphosis in the striped newt, results of my experiment and field data support an important role for a genetic component in life-history pathway expression in this salamander. My experiment precludes a single recessive allele as the genetic basis for life-history expression in N. perstriatus. A polygenic system seems most parsimonious with current data. Additionally striped newts have a distinct reproductive strategy in which most if not all paedomorphic individuals apparently transform after breeding, which occurs at ca. 1 year of age. Therefore, strict ecologically-based models of facultative paedomorphosis need to be extended to account for these phenomena.

PAGE 108

Table 4-1. Bod length (BL) mass and larval period for metamorphic striped newts across the four food treatments. Body length and mass did not differ significantly among the four treatments. Larval period ofHH and HL newts did not differ significantly. Larval period ofHH and HL newts differed significantly from the LL and LH animals, which were significantly different from each other. BL (mm) Mass (g) Larval Period (days) Treatment n mean SE mean SE n mean SE HH HL LL LH 7 9 5 6 28.0 27.1 27.6 27.3 0.23 0.7 0.76 0.78 0.6 0.5 0.4 0.6 Table 4-2. Body length (BL) and mass for paedomorphs in each treatment. There were no paedomorphs produced in the HL treatment group. BL(mm) Mass (g) Treatment n mean SE mean SE HH 2 36.5 0.50 2.5 0.14 LL 4 27.8 0.50 0.9 0.05 LH 3 33.3 0.87 2.0 0.23 0.04 0 03 0.04 0.08 8 10 6 7 42.4 38.4 101.7 63.1 1.73 1.68 15.39 4.38 '-0 '-0

PAGE 109

100 40 A 35 ~ v c;/) ....r;;J--.r;;J " <1.) 30 -= :a / --~ (.) -~--....---~ g 2 5 / 'v 1--< ~--T"-" ..D v-V _..-~ --0 ~ --,-.._ 20 /~ '---" ~if t ...:l 15 --IIlI g O HL <1.) 10 -T-LL -v LH 5 0 0 20 40 60 80 100 120 140 160 180 2 00 3.0 B 2.5 c;/) <1.) ..., ...c:: 2 0 .. ...svv (.) g -1--< ~ --V ..D .-V 0 1.5 ,ff ,-.._ 01) '---" c;/) c;/) ro 8 1.0 g / -1r-----T <1.) -~---.-;rv ..... --T"-" 0.5 _._ .... _..-T 0.0 0 20 40 60 80 100 120 140 160 180 2 00 Day of experiment Fig. 4-1. Growth trajectories of larvae in the four different food treatments. A ) Mean of body length. B) Means of mass. All of the HL treatment larvae transformed shortl y after the food level switch. Arrows indicate day of food level switch. Only mean s for untransformed individuals are plotted and error bars have been omitted for clarity

PAGE 110

0 .2 0.020 0. 2 0 A 0.018 ,-.., 0.18 >-. 0.016 ro -0 ro ] 0 16 -0 bl) 0.014 .._,, 00 -._,, 0.14 00 ro 8 0.012 s 0.12 s V V 0.010 ..... ro 0.10 1-, 1-, ..c ..c: 0.008 1$ 0 08 1$ 0 0 .... 6h bl) 0.006 @ 0.06 V V 0.04 0.004 0.02 0.002 0.00 I I 0.000 I I HR LL LH HR LL LH Treatment Treatment Fig. 4-2. Means comparisons of growth rates for remaining branchiates among three treatments after the switch in food levels. A) Growth rates in body length (BL). B) Growth rates in mass The HL treatment is excluded because all newts in this treatment transformed shortly after the food-level switch. Differences in BL and mass growth rates were significant. Refer to methods for an explanation of how growth rates were calcualted Error bars are +/ 1 SE. B ...... 0 ......

PAGE 111

10 2 30 A 0 'v 25 CD O 'v ,, (/l CD 'v ... ..c:
PAGE 112

1.0 0.9 0.8 0.7 C: 0 0.6 0 0.. 0 0.5 I-< 0.. ..c: e0.4 0 0.3 0 2 0.1 0 0 103 HH HL LL LH Food Treatment Fig. 4-4. Distribution of life-history pathway expression in striped newts among the four food treatments. Paedomorphs are indicated by the black portions of the bars and metamorphs by the diagonal lines. The e ffect of food treatment on pathway e x pression was not significant.

PAGE 113

en 8 7 6 5 (1.) 4 (1.) ..0 3 z 2 1 0 Metamorphs Paedomorphs I I I 0 5 10 15 104 I 20 BL(mm) I 25 A I I 30 35 40 11 ~--------------------~ 10 9 8 7 6 5 4 3 2 1 B 0 -+--........-.-------__,_-+-'----~---.----'--'-r---.---~--+-'--'-..........,_----,--..._~----1 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 Mass (g) Figure 4-5. Size of metamorphic and paedomorphic newts at the end of the experiment. A) Body length (BL) of newts. B) Mass of newts.

PAGE 114

105 A Optimal growth-rate model B Fixed differentiation-rate model b + c -----b t Time t Time Fig. 4 6. Predictions of size at metamorphosis and larval period for two different models of amphibian metamorphosis. A) The optimal growth-rate model (Wilbur and Collins 1973) B) The fixed differentiation-rate model (Smith-Gill and Berven 1979 ; Travis 1984). The figures were modified from Alford and Harris (1988) and Hensley (1993). Predictions are based on the expected outcome of the four treatments used in my experiment. Slopes indicate relative growth rates and the lines end at metamorphosis. The dashed lines indi cated by b and b + c in model A represent the minimum and maximum size thresholds for metamorphosis as presented in the Wilbur-Collins model. The bold arrows along the Time axes indicate when the food level switch occurred.

PAGE 115

CHAPTERS CONSERVATION GENETICS AND PHYLOGEOGRAPHY OF THE S T RIP E D NEWT Introduction The striped newt (Notophthalmus perstriatus) is a salamander endemic to northcentral Florida and southeastern Georgia (Conant and Collins 1991 ) Individuals inhabit xeric upland habitats (e.g. sandhill and scrub communities) and breed exclusively in temporary ponds that lack predatory fishes (Carr 1940 ; Christman and Means 1992 ; Dodd and LaClaire 1995; Franz and Smith 1999). The uplands inhabited by striped newts are fire climax communities (Myers 1990) and fire appears to be crucial for the persistence of striped newts. Little has been published about the striped newt s ecology (Christman and Means 1992; Dodd 1993; Dodd and LaClaire 1995). Most literature on striped newts is limited to survey results and species accounts (Bishop 1941a 1943; Carr 1940 ; Christman and Means 1992 ; Dodd and LaClaire 1995 ; Franz and Smith 1999 ; Ripes and Jackson 1996 ; Meecham 1967). Studies of striped newt feeding habits (Christman and Franz 1973) natural history at a breeding pond (Dodd 1993) and orientation to and from a breeding pond (Dodd and Cade 1998) represent the only published works focusing on this species. Nothing has been reported about population genetics of the striped newt Striped newts have declined throughout their range (Dodd and LaClaire 1995 ; Franz and Smith 1999). Reduction of the longleaf pine / wire grass ecosystem fire suppression and the natural patchy distribution of upland habitats ( i e. sandhill and s crub 106

PAGE 116

107 communities) required by striped newts have resulted in a fragmented and patchy distribution of the species. A complex life history (Fig. 1-1) makes striped newts vulnerable to threats at breeding ponds (e.g. ditching and draining of temporary ponds) as well as in the surrounding uplands ( e.g. silviculture practices fue suppression). Densities of striped newts are very low at most sites where they persist (Dodd and LaClaire 1995 S. A. Johnson B. Means K. Greenberg and D. Stevenson unpubl. data) Because of historical declines and current low population densities, the striped newt is recognized as a rare species in both Florida and Georgia (Christman and Means 1992 Cox and Kautz 2000 ; Jensen 1999) and its biological status is under review by the U S. Fish and Wildlife Service (L. LaClaire, pers. comm.). Determining population genetic structure is an essential foundation for conservation and management of striped newt populations. Striped newt populations persist at only a few stronghold locations (areas where according to recent surveys newts are known to persist) throughout the range of the species. Most stronghold locations have multiple breeding ponds with appropriate upland habitat that allows dispersal to occur among the ponds. In Florida these include Ocala National Forest Katharine Ordway Preserve Camp Blanding Training Site and Apalachicola National F orest. In Georgia strongholds include the Joseph Jones Ecological Research Center property Fort Stewart Military Instillation and possibly a suite of ponds in Jenkins Co .. ignificant genetic divergence among thes e locations would indicate restricted cont e mporary ge ne flow meaning that ex change of individuals among populations is minimal or none x i s tent. C on s equently if local extinctions were to occur it is unlikely that ex tirpated locations would be recolonization by individuals from other populations.

PAGE 117

108 If this is the case each population should be viewed and managed as an independent demographic unit. Identifying such management units (Moritz 1994a) is a powerful and practical application of molecular techniques to define the appropriate geographic scale for monitoring and management (Moritz 1994b ). Moritz (1994a b) defined management units as populations that differ significantly in allele frequencies at nuclear or mitochondrial loci. Population genetic structure must be resolved to identify appropriate management units for striped newt conservation. The objectives of this study were to 1) determine population genetic structure of striped newts to identify potential management units as targets for conservation 2) infer historical biogeographic patterns based on a mitochondrial DNA (mtDNA) gene genealogy 3) test Dodd and LaClaire's (1995) biogeographic hypothesis which predicts that genetic differences exist between western and eastern striped newt populations and 4) provide management agencies with conservation recommendations for the striped newt based on genetic data. Gene flow among locations sampled was estimated with ~ST and Nm values and divergence among haplotypes was estimated with the Kimura two parameter model. Methods Sample Collection Samples were collected from 10 locations throughout the range of N. perstriatus with the exception the Tifton and Vidalia Uplands of Georgia (Table 5-1 Fig. 5-1). Newts were captured mainly with dip nets although in a few cases seine nets or wire screen funnel traps were used A relatively small section (based on the size of the animal) of the distal end of the tail of each individual was removed and placed in a

PAGE 118

109 uniquely labeled vial containing saturated salt buffer (NaCl 25mM EDTA pH 7.5; 20% DMSO; protocol modified from Amos and Hoelzel, 1991 ). Tail sections, which usually amounted to less than 15% of an individual's total length, were removed with dissection scissors scissors were cleaned and sterilized with alcohol between individuals. This tail clip method is nonlethal and individuals were released shortly after the sample was taken. Samples were stored at room temperature until used for DNA extraction. DNA Isolation and Sequencing Whole genomic DNA was isolated from each sample using standard phenol/chloroform extraction followed by ethanol precipitation and storage in Tris/EDT A buffer (Hillis et al. 1996). A 790 base-pair fragment located in the cytochrome b ( cytb) gene of the mitochondrial DNA (mtDNA) genome was amplified by polymerase chain reaction (PCR) methodology using primers H14447 (Edwards et al. 1991) and MVZ15 (Moritz et al., 1992). Amplifications were carried out in 25 l reactions containing: 1 x buffer 3 M MgC12, 200 Meach dNTP, 0.25 Meach primer, 1 U Taq polymerase (Sigma) and 2-5 ng of template DNA. All amplifications were performed in a Biometra UNO thermocycler. Following an initial denaturation at 94C for 3 min, 35 cycles of polymerase chain reaction (PCR) were ran under the following conditions: denaturing for 1 min at 94C annealing for 1 min at 54C and extension for 1 min at 72C. PCR products were purified with 30,000 MW filters (Millipore Inc. Bedford MA). At the University of Florida's DNA Sequencing Core sequencing reactions were conducted with Big Dye technology (Applied Biosystems Inc. Foster City CA) using a robotic workstation (ABI model 800) and the fragments were gel-separated using an automated equencer (ABI model 377) Ambiguous sequences were re-amplified and sequenced to

PAGE 119

110 confirm the accuracy of the nucleotide sequence designations. equences of haplotype from the western portion of the species s range (haplotype F and the eastern portion of the range (haplotype C) were submitted to GenBank (accession numbers AF380362 and AF380363, respectively). Data Analysis Chromatograms were checked against computer base designations, then aligned using Sequencher version 3.1 (Genes Codes Corp., Ann Arbor, MI). Divergence among haplotypes (d) was estimated with the two-parameter model of Kimura (1980). Nucleotide diversity (1t Nei 1987) and haplotype diversity (h; Nei, 1987) were calculated with the program Arlequin, version 1.1 (Schneider, et al., 1997). I estimated the proportion of gene diversity within and between collection sites with ~ST values based on the AMOV A (Excoffier et al., 1992) function in Arlequin. Pair-wise estimates of gene flow among sites (Nm: number of migrants per generation) were determined using the formula Nm =(1/~sT-l)/4 (Slatkin, 1993). The average migration rate (Nm) between the sampled sites was calculated with the private-allele method (Slatkin, 1985) using equation 14 of Slatkin and Barton (1989). Sites with sample sizes less than five (i.e. JSF, n = 1; and NSJC, n = 2) were excluded in pair-wise calculations of ~ST and Nm values. These samples were included, however, in species-wide estimates of genetic diversity. To reveal relationships among haplotypes, phylogenetic trees were generated using the parsimony option in PAUP* version 4.016 (Swofford 1998). An unrooted parsimony network of haplotypes was constructed and imposed on a map to visualize phylogeographic patterns.

PAGE 120

111 Results A 593 bp fragment of the cytb gene was aligned for 86 samples collected throughout most of the range of the striped newt (Table 5-1, Fig. 5-1). Comparisons among the sequences revealed 53 variable nucleotide sites including 37 transitions and 17 transversions (Table 5-3 ; nucleotide site 95 contained both a transition and a transversion). Twenty-seven haplotypes were identified among the 86 samples (Table 54). The percent sequence divergence (d) among the haplotypes ranged from d = 0.00169 to d = 0.03879. The number ofhaplotypes identified at each collection location ranged from one (ONF JSF) to seven (CBTS) and overall haplotype diversity (h) was 0.8996 (Table 5-4). Overall nucleotide diversity (n) among the 27 haplotypes was 7t = 0.00091 (Table 5-4 ). Most sample locations contained endemic haplotypes, and only haplotypes A B and C were found at more than one site. Haplotype C was the most widespread but was shared among only three of the ten locations (KOP NSJC and ODNA; Table 5-4) occurring with the greatest frequency (62.5%) at the ODNA site in Georgia. Haplotype A was the most common and was found at sites KOP and CBTS, both in north central Florida. Eighty-three percent of individuals at KOP had haplotype A whereas only 18.2 % of individuals from CBTS had haplotype A (Table 5-4). Haplotype B was shared between RSPSR and ONF at the southern end of the range of N. perstriatus. All individuals analyzed from the three ponds representing the ONF site had haplotype B whereas 60% of the individuals from the single pond sampled at RSRSP showed haplotype B The remaining 24 haplotypes were endemic to single locations although

PAGE 121

112 multiple individuals within a location had the same haplotype in some in s tanc es (Ta bl e 54). Haplotype diversity varied considerably among sites. CBTS had the greatest h (0.867), with seven unique haplotypes among 11 individuals. At two sites ONF and J F only a single haplotype was observed. In many instances haplotypes differed by only a single nucleotide position and corresponding nucleotide diversity n was relatively low (Table 5-4 ).. Haplotypes S M, and N were the most divergent from the consensus sequence differing at 14, 11 and nine nucleotide positions respectively. The distribution of haplotypes indicates very strong genetic differentiation among collection sites. Overall ~ST = 0.67 ; a value that indicates highly significant population structure across the range of this species and very restricted gene flow among sites Pair wise comparisons of ~ST values between sites are presented in Table 5-5. With the exception of sites excluded because of small sample size (JSF and NSJC) all pair-wise ~ ST values were significantly different (P < 0.05), based on 100 permutations for each companson. The estimated number of migrants per generation (Nm) among sites is presented in Table 5-5. Because ~ST values were used to calculate the Nm values these two measures of population divergence are directly related. Nevertheless Nm values provide perspective on the estimated exchange of individuals among sites in a management context. The overall migration value (excluding the JSF and NSJC sites) was Nm= 0.265. A geographic distribution of the 27 haplotypes illustrates the limited number of haplotypes shared among collection sites (Fig. 5-2). The topology produced by a

PAGE 122

113 neighbor-joining analysis (Fig. 5-3) was consistent with a manually constructed parsimony network (Fig. 5-4). There are distinct western (i.e. ANF and JJEC sites) and eastern phylogroups which are separated by a hiatus in the species's range of about 120 km (Figs. 5-3 and 5-4). The eastern phylogroup is further partitioned into at least two and possibly three, sub-groups. Numerous branch lengths in Fig. 5-3 are relatively short because of the low sequence divergence among some of the haplotypes, but nonetheless indicate a phylogeographic signal. Discussion Population Structure Significant population genetic structure was revealed throughout the range of the striped newt. An overall high AMOV A estimate of ~sr = 0.67 and corresponding low estimate of migration rate among locations sampled (Nm= 0.265) indicates a high degree of genetic divergence because of restricted contemporary gene flow. The low Nm value demonstrates the high degree of isolation among the locations sampled since migration between populations in excess of four migrants per generation (Nm~ 4) is required to homogenize populations at mitochondrial loci (Birky et al. 1983). Gene flow is severely restricted and, consequently, population differentiation has likely occurred through genetic drift. Both genetic data and mark-recapture studies (Chapter 3) indicate that relatively low vagility of striped newts and the patchy distribution of sandhill and scrub habitats limit long distance dispersal. However within an area of contiguous suitable habitat individuals disperse hundreds of meters and they occasionally colonize or recolonize i olated ponds ( hapter 3). For example of the areas sampled the KOP site is best

PAGE 123

114 represented. I analyzed sequences of 24 individuals from seven pond s ( numb ers 4 5 6, 7 9 10 11) at the KOP site (Table 5-1 Fig. 5-5.). Four haplotype s were revealed among the 24 samples with haplotype A found at all ponds (Tables 5-2 5-4). Sharing of haplotype A among the ponds indicates that they are tightly connected by recent gene flow. Moreover a mark-recapture study demonstrated movement of a newt between two of the ponds that are separated by almost 700 m and newts commonly dispersed hundreds of meters from their pond (Chapter 3). Genetic data show strong separation between habitat fragments across the range of N. perstriatus and a mark-recapture study has revealed considerable movement within at least one of these fragments ( e.g. KOP site). I therefore hypothesize that the KOP site supports a metapopulation (Hanski and Simberloff 1997) of striped newts and that striped newts may persist at most locations as metapopulations. I suggest that within a metapopulation ponds act as focal points for demes (i.e., subpopulations) and that there are periods of extirpation and recolonization that characterize each deme over time. The same focal pond may serve as a source at one point and a sink at another. A pond at which newts arrive to breed but where they are unsuccessful because of a short hydroperiod or high predation level could act as a sink. Successful reproduction in the pond may occur during wet periods or when predators are sparse. Over time a specific deme of newts may decline (Dodd, 1993) and possibly become extirpated Recolonization is possible by dispersal through contiguous upland habitat. Although dispersal occurs at a local scale (i.e., metapopulation scale) as shown by mark-recapture study, the pair-wise and overall ~ ST values indicate that gene flow between metapopulations is severely restricted.

PAGE 124

115 Testing a Biogeographic Hypothesis Dodd and LaClaire (1995) hypothesized that striped newts colonized Georgia via two separate corridors along sandy river terraces in the west and along marine terraces (primarily Trail Ridge) in the east. If this supposition is correct they predicted that genetic differences may exist between western striped newt populations and the eastern striped newt populations. There is an apparent large (ca. 125 km) hiatus in the distribution of striped newts in Florida, separating the Florida animals into eastern and western clusters (Dodd et al. in press; Franz and Smith 1999). The genetic data based on cytb sequences support this biogeographic hypothesis A neighbor-joining tree (Fig. 5-3) depicts a distinct western phylogroup, composed of samples from the ANF site in Florida and the JJEC site in Georgia. Some of the haplotypes from these sites were not very divergent (Fig. 5-4) and the pair-wise ~ST value between these two sites was the smallest (0.173) of all pair-wise values (Table 5-5). Therefore gene flow between the sites may have occurred relatively recently. However the two sites are genetically distinct and they did not share haplotypes (Table 5-4, Fig. 52). These two sites comprise a western phylogroup (Fig. 5-3) that is genetically divergent from the eastern phylogroup. The eastern phylogroup appears to be divided into two or thee additional groups (Fig 5-3) all of which are closely associated with relict coastal ridge systems in peninsular Florida and eastern Georgia. Sequence divergence among most of the eastern haplotypes is relatively shallow and there is no overt phylogeographic signal that is absolutely parsimonious with the distribution of the s everal ridge s ystem s that characterize peninsular Florida and eastern Georgia (Hall 1966 White 1970). Nonetheless if newts persist long enough into the future, it is

PAGE 125

116 possible that the isolated newt metapopulations associated with the different ridge systems will continue to diverge genetically because of restricted gene flow and linea ge sorting (Avise, 1994). Eventually a stronger phylogeographic signal may develop concordant with the distribution of the different ridge systems such as has been revealed for the Florida endemic scrub lizard (Clark et al. 1999). Although samples were analyzed from throughout the range of the striped newt one area of interest in Georgia was not sampled. Consequently the genetic affinities of striped newts that persist in apparent isolation within the Tifton Uplands in Irwin Co. (Dodd and LaClaire 1995) are unknown. However based on Dodd and LaClaire s interpretation of newt biogeography I speculate that newts from Irwin Co. may be closely allied with the western phylogroup. Dodd and LaClaire (1995) suggested that newts in the western portion of the specie's range (i.e., western phylogroup) may have dispersed into Georgia along sandy terraces running parallel to rivers that drain into the Gulf of Mexico. On the other hand newts of the eastern phylogroup in Georgia are thought to have dispersed along the "Trail Ridge and other marine terraces parallel to the coast and perpendicular to rivers draining into the Atlantic (Dodd and LaClaire 1995). Newts in Irwin Co. are located close to the Alapaha River (Dodd and LaClaire 1995) which is part of the Suwannee River basin a Gulf Coast drainage. If Irwin Co. newts prove to be allied with the western phylogroup this would further support the Dodd and LaClaire (1995) biogeographic scenario. Striped Newt Biogeography and Phylogeography Genetic data have been analyzed for many vertebrate species in the southeastern United States thus providing a region-wide phylogeographic interpretation across numerous tax a (A vise, 1996 2000; Walker and A vise 1998). Most species surveyed

PAGE 126

117 show significant phylogeographic structure. The striped newt showed significant mtD A structure across its range as well. The phylogeographic pattern revealed for striped newts best approximates the category ill pattern most or all haplotypes are related closely yet localized geographically", described by Avise (2000). Sequence divergence (p) among the 27 haplotypes was ranged fromp = 0.00169 top = 0.03879. Nevertheless, most haplotypes were endemic, with only three haplotypes shared among sample collection locales (Table 5-4; Fig. 5-2). Another pattern observed in many of the species previously assayed in the Southeast, which included freshwater and terrestrial forms is the presence of a fundamental phylogeographic split separating species into eastern and western phylogroups. Similar to these species, striped newts showed a pattern in which a western phylogroup is distinct from an eastern phylogroup. The location of the split between the eastern and western phylogroups for the striped newt is remarkably concordant with the boundary between eastern and western clades of the gopher tortoise (Osentoski and Lamb, 1995) and the white-tailed deer (Ellsworth et al. 1994). In each of the three species mtDNA analysis revealed a phylogeographic boundary in panhandle Florida and southeastern Georgia east of the Apalachicola River drainage. In Florida, this area is recognized as a significant zoogeographic boundary, influencing the distribution of freshwater fishes (Gilbert 1987) Although phylogeographic patterns of other species also show distinct eastern and western genetic assemblages the precise geographic location and evolutionary depth of this east/west split varies among taxa (A vise 1996). hanges in sea level and climate influenced Florida's natural communities throughout the Pleistocene (Webb 1990) and this ce1tainly impacted the present-day

PAGE 127

118 genetic population structure of striped newts. The difference in mtDNA se qu e n ces between the most divergent striped newt haplotypes was 3.9%. Using an estimate of 0.8% sequence divergence per million years suggested by Tan and Wake ( 1995 ) for North American salamandrids it is clear that events during the late Pliocene and especially during the middle Pleistocene had a major influence on the phylogeograph y of striped newts. Historical events have led to a patchy distribution of xeric upland habitat s required by striped newts presently. Likely having evolved from an ancestor adapted to the xeric conditions of the Madro-Tertiary Geoflora the range of the striped newt has probably always been linked to the distribution of xeric habitats mainly scrub and sandhill. Current distribution of scrub communities is just a remnant of a historically much more extensive ecosystem (Myers 1990). Dry savanna-like habitat (resembling modem sandhill) was also more widespread in Florida during the early Pleistocene (Meylan 1981; Webb and Wilkins 1984; Webb 1990) and certainly xeric habitats predominated across much of the peninsula during the Wisconsinan glaciation (Myers 1990 ; Watts and Stuiver 1980 ; Webb 1990). Presently scrub and sandhill communities occur as isolated habitat islands in panhandle and peninsular Florida (Myers, 1990). The natural patchy distribution of striped newt habitat probably contributed to the apparent development of metapopulations throughout the range of the species. Much of the naturally fragmented striped newt habitat has been severely impacted by humans reducing the extent of suitable habitat to a mere fraction of what is was just a century ago (Means 1996 ; Myers 1990). Human disturbance has further isolated metapopulations and probably caused the extirpation of some local populations. Considering the fragmented and isolated nature of present striped newt habitat and the relatively limited

PAGE 128

119 dispersal abilities of newts it should not be surprising that significant genetic structure was found across the range of N. perstriatus. Conservation and Management Implications Genetic and drift-fence data show that gene flow occurs among ponds at a local scale (i.e., within metapopulations). However, as indicated by pair-wise estimates of migration rates between sites (Table 5-5), gene flow is severely restricted between the sites sampled. The fact that only a few haplotypes were shared among populations and the $sr values were significant in all pair-wise comparisons supports the designation of site-specific management units. In fact, each management unit probably represents a different metapopulation. The current patchy distribution of high pine uplands (Myers, 1990) certainly contributes to restricted gene flow observed and probably influence metapopulation structuring across the species range. Based on these data, the sites sampled are demographically isolated from one another and should be considered as management units. Management units are populations recognized to have significant divergence of allele frequencies at nuclear or mitochondrial loci (Moritz, 1994a). Practical considerations also indicate that each location sampled (i.e. metapopulation) should be managed as an independent demographic unit. As a result of habitat loss and degradation, as well as the patchy distribution of upland xeric habitats required by striped newts the species is now almost entirely restricted to public lands (Dodd and LaClaire 1995 ; Franz and Smith 1999) and many different organizations are ultimately responsible for management of these public lands (e.g U.S. Forest Service U .. Department of Defense Florida Division of Forestry Florida Department of nvironmental Protection Georgia Department of Natural Re ource ). If triped newts

PAGE 129

120 become extirpated at any of these sites the genetic data indicate that no recolonization will occur within a timeframe meaningful to resource managers. Clearly most land management practices at one site will have no effect on newt populations at other sites. Although each organization is responsible for making management decisions to secure the long-term persistence of an isolated striped newt metapopulation(s) similar conservation and management procedures should be followed at all sites. Managing the habitat at a landscape scale (i.e., ecosystem management) will benefit numerous other taxa in addition to the striped newt. Conserving and protecting small temporary wetlands where striped newts breed as well as the adjacent uplands will also enhance populations of the numerous other species that rely on these habitats (Dodd and Charest, 1988; Dodd 1992; Guyer and Bailey 1993; Johnson 1999; Moler and Franz 1988). Managing upland habitat and restoring historical conditions seem the most effective approach to striped newt conservation. Such an approach will rely heavily on prescribed burn programs. Regular burning of appropriate upland habitat will maintain the dynamics of metapopulations facilitating dispersal of individuals among demes. In addition to prescribed burning and protecting isolated wetlands mechanical disturbance of herbaceous ground cover should be avoided or eliminated. Such disturbance, associated with silvicultural practices is one form of habitat degradation that has contributed to the decline of N. perstriatus. If the species becomes federally listed those agencies and organizations responsible for land management will have to become more attentive to these recommendations to ensure that their land management practices enhance the quality of striped newt habitat.

PAGE 130

121 Table 5-1 Locations of striped newt breeding ponds in Georgia and Florida wh re amples were collected. Location No ponds no tate County sampled Location of pond(s) Florida Orange 1 Rock Springs Run State Preserve (RSRSP) Spears Scrub Pond ; Sanford SW Quad. ; sec6 T20 R29E 2 Florida Marion 2 Ocala National Forest (ONF) Neofiber Pond ; Lake Delancy Quad sec22-23 Tl 2S R25E Mud Bog Pond ; Lake Kerr Quad. ; sec36 Tl3S R25E Putnam Greenberg Pond 6 ; Lake Delancy Quad .sec36 Tl 2 R25E 3 Florida Putnam 7 Katharine Ordway Preserve-Swisher Memorial Sanctuary (KOP ) Blue Pond ; Putnam Hall Quad.sec15-16 T9S R23E Smith Lake; Putnam Hall Quad.; sec23-24 T9S R23E Harry Prairie Sinkhole Pond; Putnam Hall Quad .; secl5-16 T9S R23E Fox Pond ; Putnam Hall Quad .; sec28 T9S R23E Clear Pond ; Putnam Hall Quad sec24 T9S R23E Breezeway Pond ; Putnam Hall Quad.sec24 T9 R23E One Shot Pond ; Melrose Quad .; sec28 T9S R23E 4 Florida Clay Camp Blanding Training Site (CBTS) unnamed pond ; Gold Head Branch Quad. ; sec35 T7S R23E 5 Florida Clay Jennings State Forest (JSF) Franz pond 7 ; Middleburg Quad.sec32 T4S R24E 6 Florida St. Johns Northern St. Johns County (NSJC) unnamed pond; no location available 7 Georgia Bryan 3 Ft. Stewart Military Installation (FSMI) DS pond 1 ; no location available DS pond 2 no location available DS pond 3 no location available 8 Georgia manuel Ohoopee Dunes Natural Area (OONA) unnamed pond ; Norristown Quad .; 23 36.23' N 82 24 97' W 9 Florida Leon Apalachicola National Forest (ANF) borrow pit pond ; Lake Munson Quad .; sec3 T2 RIW 10 Geor g ia Baker 3 Joseph Jones cological Research Center (JJ ) Pond 40 ; Bethany Quad : 31 15 66' N 84 31.75' W Pond 41 ; Bethany Quad .; 31 15 65' N 84 31.74' W Rhexia Pond !model Quad .31 16 22' N 84 29 73' W

PAGE 131

122 Table 5-2. Number of samples analyzed and haplotype discovered at each location Individual sample ID numbers are pres nted so samples can be referenced to stored DNA and tissue samples at the University of Florida Location Location No samples Haplotypes Sample Pond no acronym analyzed observed JD nos name( s ) RSRSP 5 B Np47 48 50 Spears Scrub Pond H Np49 54 pears Scrub Pond 2 ONF 8 B Np66 67 68 69 Neofiber Pond Np80 81 82 Mud Bog Pond Np26 Greenberg Pond 6 3 KOP 24 A Np2 171 Blue Pond Np142 Smith Lak e Npl49 151 153 154 Harry Prairie Sinkhole Pond Np158 160 161 162 Fox Pond Np167 Clear Pond Npl75 176 178 179 Breezeway Pond Np127 128 129 130 One Shot Pond C Npl 1 Blue Pond Npl50 Harry Prairie Sinkhole Pond R Np3 Blue Pond T Npl59 Fox Pond 4 CBTS 11 A Np95 99 unnamed pond E Np93 100 102 104 unnamed pond M Np98 unnamed pond N Npl0l unnamed pond 0 Np103 unnamed pond p Np94 unnamed pond Q Np97 unnamed pond 5 JSF z Np16 Franz pond 7 6 NSJC 2 C Np59 unnamed pond u Np58 unnamed pond 7 FSMl 13 D Npl05-109 Ill 112 114 DS pond 3 Np4 DS pond l Npl 16 DS pond 2 J Np5 DS Pond 1 K Npl 15 DS Pond 2 L Npl 18 DS Pond I 8 ODNA 9 C Np72 74 77 unnamed pond G Np70 71 78 unnamed pond AA Np79 unnamed pond 9 ANF 8 F Np7 8 42 44 45 borrow pit pond V Np43 borrow pit pond w Np41 borrow pit pond s Np40 borrow pit pond 10 JJEC 5 X Np30 37 Pond 40 y Np31 38 Pond 41 Np35 Rhe ia Pond

PAGE 132

Tab! 5-3. foll Haplotypes Concen us C B H G L 0 D J K s E A p Q z R T M V F y w X ariable nucleotide sites found in striped newt cytochrome b sequences. The 27 haplotypes are identified by letters nwnbers read top to bottom) to indicate nucleotide locations of polymorphic sites within the 593-base sequence. 2 7 I 2 2 3 3 4 5 6 6 7 8 9 9 9 l I I I l l l I 2 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 I 3 4 6 0 7 5 6 8 3 5 7 9 I l 2 2 2 2 4 7 0 0 2 3 4 5 5 3 3 4 6 7 9 0 0 0 4 4 4 5 5 9 0 I I I 2 6 7 6 7 2 3 8 9 4 2 4 7 3 2 6 I 2 3 5 8 8 5 9 3 7 8 0 7 8 1 6 I 7 0 5 6 9 7 5 ATTG ATTTTTTTAACCTCATAGGTTAAATGTGATTGAAACATCCCCTTATTT C T G C G ---G-----G----CC C G C --------T -G C C G G C -G-C--G---TC C G C ------C-A ----GC ------C--G----GC C G G CCG -C-CC-TCCC C --G-T-C ---G----A-CC C G-G-A-CC C C G C ---G---CA------G---G--CA------------G----G----G-CA----G T ---CA-G --CA---T-G ---T---CA-G -G C -G-G-GG C A T T -GA C T T G -C-G-TC G --C---C-G---T---C C G --C---C--G---TT----C G C C G C

PAGE 133

Table 5-4. Striped newt haplotypes haplotype diversity (h ) and nucleotide diversity (n) at each location. Haplotypes Location A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA Total h n 1) Orange Co 3 2 5 0 600 0 001 RSRSP 2) Marion Co. 8 8 1.000 0 000 ONF 3) Putnam Co. 20 2 1 1 24 0 .2 97 0.00 2 KOP 4) Clay Co. 2 4 1 1 1 1 1 11 0.867 0.01 2 CBTS 5) Clay Co. ----------------------1JSF 6) St. Johns Co. 1 1 NSJC 7 ) Bryan Co. 8 2 1 1 1 FSMI 8) Emanuel Co. 5 3 1 ODNA 9) Leon Co. ANF 10 ) Baker Co. ----5 ---------1 -1 1 ---3 2 1 2 13 9 8 5 JJ EC Tota l s 22 1 1 8 8 4 5 3 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3 2 1 86 0 628 0 .00 1 0.639 0.001 0.64 3 0.008 0.600 0.001 ....... N

PAGE 134

Table 5-5. Pairwise m (above diagonal) and ~ST va l ues among the 10 striped newt samp l e locations. Location 1 2 3 4 5 6 7 8 9 10 1 Orange o. 0.51 0.06 0.59 0 08 0 14 0.15 0 02 RRP 2 arion Co 0.331 0.06 0.47 0 06 0.10 0.12 0.01 0 3 ) Putnam Co. 0.797 0.818 0.31 0.04 0.05 0.09 0 03 KOP 4) Clay Co 0.299 0 347 0.443 0.20 0 31 0.38 0 33 CBTS 5 Cla Co ,-.... JSF N Vt 6) t. Johns Co SJC 7) Bryan Co 0.748 0.800 0.858 0.557 0.17 0.07 0.02 F MI 8) Emanuel Co 0.633 0 722 0.831 0.445 0 599 0.10 0 02 OD A 9) Leon Co. 0.620 0 675 0 725 0.396 0.771 0.711 1.20 F 10) Baker Co 0.912 0.962 0.817 0.429 0 920 0.913 0.173 JJEC

PAGE 135

Georgia I ,,._ 9 ,: , I , Gulf of Mexico , I I I I I I \ \ , \ \ \ , , ' ' _, I I I I , I I , , , , 126 I I I \ I 7\ ', , I , I ------... I \ e2\ I \ l: ., I ........ ., ------Florida S01,1th Carolma (? Atlantic Ocean Fig. 5-1. Sampling locations of striped newts in Florida and Georgia. Dots on the map are approximate locations. Refer to Table 5-1 for specific locations. The dashed line shows the present range of the species.

PAGE 136

\ Georgia , , 127 , -----...... ' ' I C,G,AAI ' ' ' ' I I I ' I I ' :[ X I ' ' ' ' .. .. .. .. ' ' ' ' \ , ' ' .. is,F,v,w I I I I I ',, I A,E,M,N,O,P,Q I j.--D-, 1-, J-, K-, ~------, \ ) I I I I I I I I \ I B I \ ' B, H [ F lo ri d a South Carolina A t lan t ic Ocean ig. 5-2. Geographic distribution of the 27 striped newt haplotypes among the 10 ites ampled in lorida and Georgia. Dots on th map are approximate location R fer to Table 51 for specific locations The da h d l ine hows the pre ent range of th pecie

PAGE 137

AA D G u C H B A Ip I Q z R T V -{F -{y w --X 0.1 % sequence divergence 12 8 L I K J 0 E I M N a. :::J 0 L... C) 0 >, .c a. C: L... Q) ..... en ro UJ ==== a. :::J s e C) 0 >, .c a. C: L... Q) ..... en Q) .. s Fig. 5-3. Ne i gh bo rjo ining tree (mi d point rooted) showing the p hylogenetic relationship among the 27 mt D NA ha pl otypes for the str i pe d n ewt. Note the two major phylogroup indicate d at t h e right.

PAGE 138

0 Georgia Gulf of Mexico 129 Florida Fig. 5-4. Hand-generated parsimony network hawing th numb r of nucleotide diffi renc between eac h of the hap l otype The network is roughly o rlaid n the lo ation her ampl were c llect d B cau e of the map' ca l l ocati n of hap! typ ar only approximat Tab l e 51 and 5-2 g iv e pecific location wh re haplotyp w re foW1d. p l ain lin conn cting hap! type indicate o n nucleotide differenc betw nth haplotyp Thin cro da h indicate addit i nal nucleotide differ e nces ( l per da h) and thick dash r pre ent 5 nucl otid differ n e Hypothetical haplotypes or po ibly contemporary on not di co ered during amp lin g are indicated in th va l a hypo

PAGE 139

Katharine Ordway Preserve N 1.5 km I p Fig. 5-5. Map of Katharine Ordway Preserve Putnam Co. Florida (i.e., KOP site). Known striped newt breeding ponds are identified with numbers and are shown in black. These ponds include: 1) Recess 2) South Fence 3) Barry 4) One Shot 5) Fox 6) Blue 7) Harry Prairie Sinkhole, 8) Harry Prairie 9) Clear 10) Smith Lake 11) Breezeway 12) Breezeway Sandhill. Other ponds and lake s are s hown in dark gray with their associated prairies and swamps in light gray. w 0

PAGE 140

CHAPTER6 SUMMARY AND CONCLUSIONS Life-history Summary Field-collected data at OSP (Chapter 2) and Breezeway Pond (Dodd, 1993) demonstrate that striped newts have a complex life-cycle involving terrestrial and aquatic stages (Fig. 1-1 ). Breeding occurs exclusively in aquatic habitats of fishless ponds that are small and have variable hydroperiods (Carr, 1940; Campbell and Christman 1982 Christman and Means, 1992; Dodd and LaClaire 1995; Dodd et al., in press; Franz and Smith, 1999; Stout et al., 1988). Larvae feed and grow in ponds until they reach a SVL of 18-20 mm, which appears to be the minimum size necessary for striped newts to initiate metamorphosis (Chapter 2; Dodd, 1993). If metamorphosis is not initiated at this size, an individual will continue to grow and will follow one of two life-history pathways (Chapter 4). After more growth, but before sexual maturation is reached it may metamorphose (metamorphic pathway) and leave the pond as an immature eft. The other life-history pathway results in maturation while the larval morphology is retained (paedomorphic pathway). Paedomorphs reproduce when they are about 1 year old then metamorphose and leave the pond even if the pond holds substantial water. The life history pathway decision appears to be controlled in part by expression of genes that may be influenced by environmental factors (Chapter 4). Striped newts spend much of their lives in uplands surrounding breeding ponds and individuals disperse hundreds of meter s after metamorphosis or reproduction (Chapter 3 ; Dodd 1996). The quality of 131

PAGE 141

132 upland habitat has an important influence on the local persistence of s triped newt s. F i r e exclusion in uplands and conversion of native longleaf pine (Pinu s palu stris) upland s to monoculture pine plantations (e.g. slash pine P. elliottii loblolly pine P. ta e d a, and sand pine P clausa) especially if done in conjunction with mechanical s ite preparation have contributed to the decline of N. perstriatus throughout the species range (Dodd and LaClaire 1995; Franz and Smith 1999 ; Jensen 1999 ; R. C. Means pers. comm. ) Because of the current patchy distribution of localities where striped newts persist as a result of anthropogenic and natural causes remaining striped newt populations are effectively isolated. Cytochrome b sequence data from samples collected throughout the range of striped newts show that contemporary gene flow between habitat fragments is severely restricted (Chapter 5). On a local scale gene flow occurs among some breeding ponds. Taken together these data suggest that striped newts form metapopulations and that the long-term survival of the species may depend on preserving existing metapopulations. Conservation, Management, and Research Prospectus Current knowledge of striped newt life history status and distribution and population genetic structure can be used to make recommendations for the conservation and management of the species. In general presence of newts at a location (Dodd and La Claire 1995 ; Franz and Smith 1999) suggests that land managers are managin g the habitat effectively. Nevertheless present management practices may not ensure the lon g term persistence of newts where they are currently found. Land mana g ers s hould consider the following recommendations carefully. Some of th e sug g e s tions are not novel ; for example Christman and Means ( 199 2) Dodd ( 1993 ) Dodd and L a C l a i r

PAGE 142

133 (1995 ) and Franz and Smith ( 1999). Most of the recommendations which follow are based on the results of my studies or they reinforce the ideas previously presented. Managing upland habitats to reestablish historical ecosystem function is the most effective approach to achieve the conservation of striped newts. Such an approach will rely on a prescribed bum program. Managing upland and pond habitats as a unified landscape should prove an effective conservation strategy. Protecting and managing small seasonally-ponded wetlands where striped newts breed as well as the adjacent uplands also will enhance populations of the numerous other species that rely on these habitats (Dodd, 1992; Dodd and Charest 1988 ; Guyer and Bailey 1993 ; Johnson 1999 Moler and Franz 1988). Semlitsch (2000) presented an excellent template for land managers concerned with aquatic breeding amphibians. My recommendations lend support for his protocols. The order in which the following recommendations are listed does not imply any hierarchy of importance. Suggestions for additional research on striped newts follow most recommendations. Recommendations for management and additional research are often not mutually exclusive. I. Manage striped newt metapopulations as independent demographic units. Remaining populations in both Florida and Georgia are confined to isolated areas because of the natural patchy distribution of appropriate upland habitats as well as habitat loss and habitat degradation caused by anthropogenic factors. Genetic data (Chapter 5) show s tron g partitioning of striped newt metapopulations and there is essentially no c ontemporar y g ene flow a mong metapopulations as indicated by pairwise values of ~ ST and an o v erall ~ ST o f 0 67 These data imply that if a local extinction of a metapopulation

PAGE 143

134 occurs it will not be recolonized naturally from other metapopulation s on a contemporary time scale. Based on the genetic data presented in Chapter 5, the location s sampled meet the requirements of Management Units (Moritz 1994a b). Although samples were analyzed from throughout the range of the striped newt one region of interest in Georgia was not sampled. Consequently the genetic affinities of striped newts that persist in apparent isolation within the Tifton Uplands in Irwin Co. (Dodd and LaClaire, 1995) are unknown. Based on the biogeographic scenario proposed by Dodd and LaClaire (1995), these newts are probably allied with the western phylogroup (Chapter 5). Tissue samples of newts from this location should be analyzed to determine the genetic relationship of newts in this area. 2. Protect small, isolated wetlands where striped newts are known to occur or may potentially occur. Striped newts breed exclusively in ponds that are devoid of predatory fishes. Such ponds are isolated and usually dry frequently; they are often less than one hectare in extent (LaClaire, 1995; Means et al., 1994). Small, isolated wetlands receive little protection at local or federal levels (Chapter 3; Semlitsch, 2000). If striped newts are to be conserved, small, isolated wetlands must be afforded protection. Furthermore as initially suggested by Christman and Means (1992), stocking of fish in known or suspected breeding ponds should never be allowed. Research on striped newt breeding ponds is needed. Dodd and LaClaire (1995) reported biotic and abiotic characteristics of striped newt ponds in Georgia and LaClaire ( 199 5) presented vegetative and soil characteristics of dry pond basins for several known newt ponds in Florida. Hydroperiods of striped newt breeding ponds vary considerably (R. Means pers. comm.; K. Greenberg pers. comm.). A very short a hydroperiod will

PAGE 144

135 preclude or negate reproduction but a long hydroperiod may result in colonization by predatory fishes ( Chapter 2 Semlitsch 2000). The window of hydroperiod lengths that allows local persistence of striped newts is unknown and needs to be established. Such data would be useful to identify potential breeding sites. In addition wildlife managers need to know which microhabitat features of a breeding pond affect metamorp hosis of striped newt larvae ( e.g. types of predators density of larvae food availability) In captivity larval striped newts eat zooplankton but the diet of larvae in natural ponds is unknown. Knowledge of food requirements of larvae could help identify potential breeding ponds 3. Initiate or continue prescribed bum programs in upland sites near striped newt breeding ponds. Although there has never been an empirical test of the effect of fire suppression on striped newts surveys conducted by several researchers have suggested that local extinction has occurred at sites where fire has been suppressed (Franz and Smith 1999 ; R Means pers. comm. ; S. A. Johnson unpubl. data). However fire suppression at some of these sites has been concurrent with conversion of uplands to pine plantations so interpreting the direct imp~ct of fire suppression is confounded by silviculture Fire probably plays a crucial role in maintaining productive breeding ponds for striped newts and other pond-breeding amphibians in the southern Coastal Plain. Periodic burning of dry pond basins may be necessary to maintain the quality of breeding pond s. Fire s a s evidenced by charring on the stems of shrubs and trees frequently occur within th e ba s in s o f many striped newt breeding ponds Land management practices that di s coura ge fir e in s mall i s olat e d w e tland s should b e discouraged. Studies of the

PAGE 145

136 influence of prescribed fire in breeding ponds are crucial to effective conservation planing for striped newts and other pond-breeding amphibians. Although regular burning of upland habitat may be essential for the persistence o f striped newts the optimal frequency of prescribed fire is unknown as is the most appropriate season(s) to burn and the optimal fire intensity. It would be valuable to analyze prescribed burn records from locations that support the highest densities of striped newts such as the Katharine Ordway Preserve and Ocala National Forest to determine if there is any correlation between burning regime and striped newt relative abundance at breeding ponds. Experiments that manipulate fire frequency, intensity and burn season would prove informative for managing striped newts, but will be difficult to undertake because of sample size considerations. Research needs to address the fundamental issue of why striped !}ewts appear to persist only at sites that burn regularly. Research should also determine the direct impact of fire suppression in the absence of confounding factors, especially silviculture. 4. Preserve core areas and buffer zones of protected upland habitat around breeding ponds. Striped newts spend most of their lives in upland habitats. A large percentage of striped newts at OSP, located in a sandhill longleaf pine habitat, dispersed hundreds of meters and many newts (an estimated 16% of the breeding population) exceeded 500 m (Chapter 3). Striped newts have been found as far as 709 m from the closest breeding pond (Dodd, 1996). Protecting adequate upland habitat is crucial for persistence of striped newts. Core areas and their associated buffer zones should be as large as possible, but managers should strive to preserve upland sandhill habitat that extends at least 1000 m from the pond edge. Data from OSP show that protected areas of

PAGE 146

137 this amount of habitat will encompass most of the newts that use the pond. Dispersal distances of striped newts in other upland habitat types ( e.g., scrub) and at other breeding ponds needs to be studied. In any case, the extent of upland used by striped newts appears to be much greater than that used by pond-breeding salamanders of the genus Ambystoma (Semlitsch, 1998). Therefore, extrapolating data across amphibian genera or species may not be justified. 5. A void mechanical disturbance of native vegetation in upland habitats, especially near breeding ponds. Silvicultural practices that disturb the herbaceous ground layer and disrupt the soil ( e.g., extensive mechanical site preparation) should be avoided because they appear to lead to local extirpation of striped newts (Dodd and La Clair~ 1995). Protecting the ground cover and subsurface soil structure in established core areas and buffer zones is crucial for the persistence of striped newt populations. 6. Maintain corridors of managed habitat at sites where there are striped newt breeding ponds. Genetic and field data demonstrated that striped newts disperse among breeding ponds (Chapter 5). One individual captured as an eft dispersing from OSP was later captured at a neighboring pond (Fox Pond). The genetic data from the Katharine Ordway Preserve indicate that newts disperse between ponds often enough to homogenize the metapopulation at mitochondrial loci. Throughout their range striped newts appear to persist mainly at locations where multiple breeding ponds have been identified. In concert with dispersal records from OSP and genetic data this supports the hypothesis that striped newts occur in metapopulations (Chapter 5) and likely require a metapopulation structure for long-term persistence. Maintaining connectivity among

PAGE 147

138 ponds via upland habitat will facilitate metapopulation function and provide a buffer against local extirpation. Where connectivity among ponds is maintained with habitat corridors research needs to determine if newts will use the corridors and what corridor characteristics influence newt dispersal among breeding ponds. 7. Surveys should be conducted to locate aditional striped newt populations. Striped newts have declined throughout their range during the past decades. Therefore all remaining populations are vital to the long-term persistence of the species. Striped newts may persist in some areas and have not been detected because of low population density or lack of survey effort. Given the imperiled status of the species it is important to identify all remaining sites that support striped newts so that they can be managed properly. Sites found on private land should be purchased or protected through a conservation easement. Permission to manage the property for striped newts should be guaranteed. Based on a habitat model the Florida Fish and Wildlife Conservation Commission has identified areas that should contain suitable habitat and could serve as target areas for future surveys (Cox and Kautz 2000). Dodd and LaClaire (1995) identified areas in Georgia that should be surveyed. 8. Striped newts should be regularly monitored at sites where they occur. Regularly monitoring striped newts will enable biologists to determine natural levels of population fluctuation and possibly identify the cause(s) if a decline or local extinction occurs. With this information declines might be mitigated or prevented at other sites. It is crucial to understand population dynamics so that natural fluctuation can be recogni z ed as distinct from declines caused by human impacts (Pechmann et al. 1991 ).

PAGE 148

139 In conclusion an effective conservation strategy for striped newts requires a landscape approach Large tracts of upland habitats containing multiple breeding ponds of varying hydroperiods must be preserved. These areas should be managed to facilitate natural ecosystem processes fire in the landscape is crucial. Clearly more information on the upland habitat requirements of striped newts is urgently needed.

PAGE 149

LIST OF REFERENCES Alford R. A. 1989. Variation in predator phenology affects predator performance and prey community composition. Ecology 70:206-219. Alford R. A. and R. N. Harris. 1988. Effects of larval growth history on anuran metamorphosis. The American Naturalist 131:91-106. Alford R. A. and S. J. Richards. 1999. Global amphibian declines: a problem in applied ecology. Annual Review of Ecology and Systematics 30:133-165 Ashton R. E. Jr. and P. S. Ashton. 1988 Handbook of Reptiles and Amphibians of Florida: Part Three the Amphibians. Winward Publishing, Inc. Miami Florida. Avise J C. 1994. Molecular Markers Natural History and Evolution Chapman and Hall New York New York. Avise J.C. 1996. Toward a regional conservation genetics perspective: phylogeography of faunas in the southeastern United States. Pages 431-470 in: J.C. A vise and J. L. Hamrick (eds.) Conservation Genetics Case Histories from Nature. Chapman and Hall New York New York. A vise J. C. 2000. Phylogeography: The History and Formation of Species. Harvard University Press Cambridge Massachusetts. Babbitt K. J. and G. W. Tanner. 2000. Use of temporary wetlands by anurans in a hydrologically modified landscape. Wetlands 20:313-322. Barinaga M. 1990. Where have all the froggies gone? Science 247:1033-1034. Batschelet E. 1981. Circular statistics in biology. Academic Press London. Beachy C K. 1995. Effects of larval growth history on metamorphosis in a streamdwelling salamander ( D esm ognath us ochrophaeus). Journal of Herpetology 29:375-382. Beck C. W. 1997 Effect of changes in resource level on age and size at metamorphosis in H y /a squire/la. Oecologia 112:187-192. 140

PAGE 150

141 Berger L., R. Speare P. Daszak D. E. Green, A. A. Cunningham C. L. Goggin R. Slocombe M.A. Ragan, A. D. Hyatt K. R. McDonald, H.B. Hines K. R. Lips G. Marantelli and H. Parkes. 1998. Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America. Proceedings of the National Academy of Sciences (USA) 95 :9031-9036. Birky C. W. Jr., T. Matuyama and P. Fuerst. 1983. An approach to population and evolutionary genetic theory for genes in mitochondria and chloroplasts and some results. Genetics 103:513-527. Bishop S. C. 1941a. Notes on salamanders with descriptions of several new forms. Occasional Papers of the Museum of Zoology, University of Michigan 451:5-21. Bishop S. C. 1941b. Salamanders ofNew York. New York State Museum Bulletin 324:1-365. Bishop, S. C. 1943. Handbook of salamanders. Comstock Publishing Company, Inc. Ithica, New York. Brandon R. A., and D. J. Bremer. 1966. Neotenic newts, Notophthalmus viridescens louisianensis, in southern Illinois. Herpetologica 22:213-217. Brown, M. T. J.M. Schaefer and K. H. Brandt. 1990. Buffer zones for water, wetlands and wildlife in east central Florida. Publication 89-07. Center for Wetlands, University of Florida, Gainesville, Florida. Burke V. J., and J. W. Gibbons. 1995. Terrestrial buffer zones and wetland conservation: a case study of freshwater turtles in a Carolina bay. Conservation Biology 9:1365-1369. Cade, B. S., and J. D. Richards. 2000. User manual for BLOSSOM statistical software. Midcontinent Ecological Science Center, Biological Resources Division, U. S. Geological Survey Fort Collins, Colorado. Campbell, H. W. and S. P. Christman. 1982. The herpetological components of Florida sandhill and sand pine scrub associations. Pages 163-171 in: Herpetological Communities. N. J. Scott (ed.). U.S. Department of the Interior Wildlife Research Report 13, Washington D. C. armichael P., and W. Williams. 1991. Florida s Fabulous Reptiles and Amphibians. World Publications, Tampa, Florida. arr A. 1940. A contribution to the herpetology of Florida. University of Florida Biological ciences Series 3: 1-118.

PAGE 151

142 Christman S. P. and L. R. Franz. 1973 Feeding habits of the striped newt Notophthalmus perstriatus. Journal of Herpetology 7: 133-135. Christman S. P. and D. B Means. 1992. Striped newt. Pages 62-65 in: Rare and Endangered Biota of Florida Volume 3 Amphibians and Reptiles P. E. Moler (ed.). University Presses of Florida Gainesville Florida. Clark A. M., B. W. Bowen, and L. C. Branch. 1999. Effects of natural habitat fragmentation on an endemic scrub lizard (Sceloporus woodi): an historical perspective based on mitochondrial DNA gene genealogy. Molecular Ecology 8:1094-1104. Conant R. and J. T. Collins. 1991. A field guide to reptiles and amphibians : Eastern and central North America. Houghton Mifflin Company, Boston Massachusetts. Cox J. A. and R. S. Kautz. 2000. Habitat Conservation Needs of Rare and Imperiled Wildlife in Florida. Office of Environmental Services Florida Fish and Wildlife Conservation Commission Tallahassee Florida. Crump, M. L. 1981. Energy accumulation and amphibian metamorphosis. Oecologia 49:167-169. Dahl, T. E. 1990. Wetlands: losses in the United States 1780's to 1980's. U.S. Fish and Wildlife Service, Washington D.C. Delis, P.R., H. R. Mushinsky, and E. D. McCoy. 1996. Decline of some west-central Florida anuran populations in response to habitat degradation. Biodiversity and Conservation 5: 1579-1595. deMaynadier, P. G., and M. L. Hunter Jr. 1999. Forest canopy closure and juvenile emigration by pool-breeding amphibians in Maine. Journal of Wildlife Management 63 :441-450. Denver, R. J. 1997. Proximate mechanisms of phenotypic plasticity in amphibian metamorphosis. American Zoologist 3 7: 1 72-184. Denver, R. J., N. Mirhadi, and M. Phillips. 1998. Adaptive plasticity in amphibian metamorphosis: response of Scaphiopus hammondii tadpoles to habitat desiccation. Ecology: 1859-1872. Dodd C. K. Jr. 1992. Biological diversity of a temporary pond herpetofauna in north Florida sandhills. Biodiversity and Conservation 1: 125-152. Dodd C. K., Jr. 1993. Cost ofliving in an unpredictable environment: the ecology of striped newts Notophthalmus perstriatus during a prolonged drought. Copeia 1993:605-614.

PAGE 152

143 Dodd C. K ., Jr. 1996. Use of terrestrial habitats by amphibians in the sandhill uplands of north-central Florida. Alytes 14:42-52. Dodd C. K. Jr. 1997. Imperiled amphibians : a historical perspective. Pages 165-200 in: Aquatic Fauna in Peril: The Southeastern Perspective G. W. Benz and D E: Collins (eds.). Special Publication 1 Southeast Aquatic Research Institute Lenz Design and Communications Decatur Georgia. Dodd C. K. Jr., and B. S. Cade. 1998. Implications of movement patterns for the conservation of amphibian communities in small temporary wetlands. Conservation Biology 12:331-339. Dodd C. K. Jr. and B G Charest. 1988. The herpetofaunal community of temporary ponds in north Florida sandhills: species composition, temporal use, and management implications. Pages 87-97 in: Management of Amphibians, Reptiles and Small Mammals in North America. R. C. Szaro, K.E. Severnson and D.R. Patton (eds.). USDA Forest Service General Technical Report RM-166 Ft. Collins Colorado. Dodd C. K. Jr., and L. V. LaClaire. 1995. Biogeography and status of the striped newt (Notophthalmus perstriatus) in Georgia USA. Herpetological Natural History 3:37-46. Dodd C. K ., Jr D. B. Means and S. A. Johnson .. In Press. Notophthalmus perstriatus Pages 2402-2411 in: Status and Conservation of U.S. Amphibians Volume 2: Species Accounts. M.J. Lanoo (ed.). University of California Press Berkley, California. Donnelly, M. A. C. Guyer J. E. Jutterbock and R. A. Alford. 1994. Techniques for marking amphibians. Pages 277-284 in: Measuring and Monitoring Biological Diversity: Standard Methods for Amphibians. W. R. Heyer M. A. Donnely R. W McDiarmid L. C. Hayek, and M. S. Foster (eds.). Smithsonian Institution Press Washington D.C. Duellman W. E. 1999. Global distribution of amphibians: patterns conservation and future challenges. Pages 1-30 in: Patterns of Distribution of Amphibians. W. E. Duellman (ed.). The Johns Hopkins University Press Baltimore Maryland Duellman W E. and S.S. Sweet. 1999 Distribution patterns of amphibians in the Nearctic region of North America. Pages 31-109 in: Patterns of Distribution of Amphibians. W. E. Duellman (ed.). The Johns Hopkins University Press Baltimore, Maryland. Duellman, W E and L. True b. 1986. Biology of Amphibians. The Johns Hopkins University Pre ss Baltimore, Maryland

PAGE 153

144 Edwards S. V. P. Arctander and A. C. Wilson. 1991. Genealogical tree for perchin g birds. Proceedings of the Royal Society of London 243:99-107. Eisenberg J. F., and R. Franz. 1995. Foreword. Pages i-v in: The Katharine Ordway Carl Swisher Memorial Sanctuary Volume: Herpetology and Mammalogy. J F. Eisenberg and R. Franz (eds.). Bulletin of the Florida Museum of Natural History 3 8 Gainesville Florida. Ellsworth D. L. R. L. Honeycutt, N. J. Silvy J. W. Bickham and W. D. Klimstra 1994. Historical biogeography and contemporary patterns of mitochondrial DNA variation in white-tailed deer from the southeastern United States. Evolution 48:122-136. Excoffier L. P. E. Smouse and J.M. Quattro. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial restriction data. Genetics 131:479-491. Franz R. and L. L. Smith 1999. Distribution and status of the striped newt and Florida gopher frog in peninsula Florida. Florida Fish and Wildlife Conservation Commission. Final Report Tallahassee Florida. Gibbons J. W. and R. D. Semlitsch. 1981. Terrestrial drift fences with pitfall traps: an effective technique for quantitative sampling of animal populations. Brimleyana 7:1-16. Gilbert C. R. 1987. Zoo geography of the freshwater fish fauna of southern Georgia and peninsular Florida. Brimleyana 13:25-54. Gill D. E. 1978a. The metapopulation ecology of the red-spotted newt Notophthalmu s viridescens (Rafinesque ). Ecological Monographs 32: 145-166. Gill D. E. 1978b. Effective population size and interdemic migration rates in a metapopulation of the red-spotted newt Notophthalmus viridescens (Rafinesque). Evolution 32:839-849. Gould S. J. 1977 Ontogeny and Phylogeny Harvard University Press Cambridge Massachusetts. Greenberg C. H. 2001. Spatio-temporal dynamics of pond use and recruitment in Florida gopher frogs (Rana capita aesopus). Journal of Herpetology 35:74-85. Griffiths R. A. 1996. The Newts and Salamanders of Europe. Academic Press San Diego California.

PAGE 154

145 Guyer C. and M.A. Bailey. 1993. Amphibians and reptiles oflongleafpine communities. Pages 139-158 in: The Longleaf Pine Ecosystem: Ecology Restoration and Management Proceedings of the 18th Tall Timbers Fire Ecology Conference. S. M. Herman (ed.). Tall Timbers Research Inc. Tallahassee Florida. Hall R. L. 1966. Pleistocene terraces in Georgia. Unpublished Master of Science Thesis University of Florida, Gainesville, Florida. Halliday T. R. 1990. The evolution of courtship behavior in newts and salamanders. Advances in the Study of Behavior. 19:137-169. Hanski I. A. and D. Simbeloff. 1997. The metapopulation approach its history conceptual domain and application to conservation. Pages 5-26 in: Metapopulation Biology: Ecology, Genetics and Evloution. I. A. Hanski and M. E. Gilpin (eds.). Academic Press, San Diego California. Harris R. N. 1987. Density-dependent paedomorphosis in the salamander Notophthalmus viridescens dorsalis. Ecology 68:705-712. Harris R. N., R. A. Alford, and H. M. Wilbur. 1988. Density and phenology of Notophthalmus viridescens dorsalis in a natural pond. Herpetologica 44:234-242. Harris, R. N., R. D. Semlitsch, H. M. Wilbur, and J.E. Fauth. 1990. Local variation in the genetic basis of paedomorphosis in the salamander Ambystoma talpoideum. Evolution 44:1588-1603. Hart R. and J. R. Newman. 1995. The importance of isolated wetlands to fish and wildlife in Florida. Florida Game and Fresh Water Fish Commission. Final Report Tallahassee, Florida. Healy W.R. 1970. Reduction ofneoteny in Massachusetts populations of Notophthalmus viridescens. Copeia 1970:578-581. Healy W.R. 1973. Life history variation and the growth of juvenile Notophthalmus viridescens from Massachusetts. Copeia 1973 :641-64 7. Healy W. R. 197 4a. Population consequences of alternative life histories in Notophthalmus v. viridescens. Copeia 1974:221-29. Healy W.R. 1974b. Sex ratio variation in samples of adultNotophthalmus viridesc e ns. The American Midland Naturalist 92:492-495. Healy W R. 1975. Terrestrial activity and home range in efts of Notophthalmus virid esce n s The American Midland Naturalist 93: 131-138.

PAGE 155

146 Hecnar S. J and R. T. M'Closkey 1996. Regional dynamic s and the sta tu s of amphibians. Ecology 77:2091-2097 Hensley F. R. 1993. Onto genetic loss of phenotypic plasticity of age at metamorpho sis in tadpoles. Ecology 74:2405-2412. Hillis D. M. B. K. Mable A A Larson S. K. Davis and E. A. Zimmer. 1996. Nucleic acids IV: sequencing and cloning. Pages 321-381 in: Molecular Systematics. D M. Hillis C. Moritz and B. K. Mable ( eds.). Sinauer Associates Inc. Sunderland Massachusetts. Ripes D. L. and D. R. Jackson. 1996. Rare vertebrate fauna of Camp Blanding Training Site, a potential landscape linkage in northeastern Florida. Florida Scientist 59:96-114. Hollander M. and D. A. Wolfe. 1999. Nonparametric Statistical Methods. Second Edition. John Wiley and Sons Inc. New York, New York. Hurlbert S. H. 1969. The breeding migrations and interhabitat wandering of the vermilion-spotted newt Notophthalmus viridescens (Rafinesque). Ecological Monographs 39:465-488. Hurlbert, S. H. 1970. The post-larval migration of the red-spotted newt Notophthalmus viridescens (Rafinesque) Copeia 1970:515-528 Jackson, M. E. and R. D. Semlitsch. 1993. Paedomorphosis in the salamander Ambystoma talpoideum: effects of a fish predator. Ecology 74:342-350. Jensen J.B. 1999. Striped newt (Notophthalmus perstriatus). Pages 102-103 in: Protected Animals of Georgia. P. W. Johnson J.C. Ozier J. L. Bohannon J.B. Jensen and C. Skelton (eds.). Georgia Department ofNatural Resources Social Circle Georgia. Johnson S. A. 1999. Natural history and reproductive ecology of the striped newt (Notophthalmus perstriatus) in north Florida. U.S. Fish and Wildlife Service. Final Report Jackson Mississippi. Johnson S. A., and R. Franz. 1999. Notophthalmus perstriatus Coloration. Herpetological Review 30:89. Kalezic M. L. D. Cvetkovic A. Djorovic and G. Dzukic. 1994. Paedomorphosis and differences in life-history traits of two neighboring crested newt (Triturus carnifex) populations. Herpetological Journal 4:151-158.

PAGE 156

147 Kimura M. 1980. A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution. 16 : 111-120. Knutson M. G. J. R. Sauer D. A. Olsen M. J. Mossman L. M. Hemesath and M. J. Lannoo. 1999. Effects of landscape composition and wetland fragmentation on frog and toad abundance and species richness in Iowa and Wisconsin U.S.A Conservation Biology 13:1437-1446. Knutson M. G. J. R. Sauer D. A Olsen M. J. Mossman L. M. Hemesath and M J. Lannoo. 2000. Landscape associations of frog and toad species in Iowa and Wisconsin, U.S.A. Journal of the Iowa Academy of Sciences 107:134-145. Kupferburg S. J. 1997. The role of larval diet in anuran metamorphosis. American Zoologist 37:146-149. La Claire L. 1995. Vegetation of selected upland ponds in north and north-central Florida. Bulletin of the Florida Museum of Natural History 38 Part I:69-96. Leips J. and J. Travis. 1994. Metamorphic response to changing food levels in two species ofhylid frogs. Ecology 75:1345-1356. Licht L. E. 1992. The effect of food level on growth rate and frequency of metamorphosis and paedomorphosis in Ambystoma gracile. Canadian Journal of Zoology 70:87-93. Lips K. R. 1998. Decline of a tropical montane amphibian fauna. Conservation Biology 12:1-13. Lips K. R. 1999. Mass mortality and population declines of anurans at an upland site in western Panama. Conservation Biology 13: 11 7-125. Madison D. M. 1997. The emigration of radio-implanted spotted salamanders Amb ys toma maculatum. Journal of Herpetology 31 :542-551. Madison D M ., and L. Farrand III. 1998 Habitat use during breeding and emigration in radio-implanted tiger salamanders Amb y stom a tigrin u m. Copeia 1998:40 2 410. Maran g io M. S 1978. The occurrence of neotenic rough-skinned newts (Tari c ha g ranulo s a) in montane lakes of southern Oregon Northwest cience 52:343-350. M e an s D. B 1996 Lon g l e a f pine fore s t going going ... Page s 210-229 in: Eastern Oldg rowth ore s t s . M B Davi s (ed ). I s land Press Washington D. C.

PAGE 157

148 Means D B. T. E. Ostertag and D. Prinitss 1994. Di s tribution habitat eco lo gy and management of the striped newt (Notophthalmus perstriatus) in the Apa la chico l a National Forest Florida. U.S. Forest Service. F inal Report Tallhassee F lo rida Means D. B. J G. Palis and M. Baggett 1996. Effe cts of slash pine si lvicultur e on a Florida population of flatwoods salamander. Conservation Biology 10:426-437. Mecham J. S. 1967. Notophthalmus p erstriatus Catalog of American Amphibians and Reptiles pp 38.1-38.2. Mecham J. S. 1968. On the relationships between Notophthalmus meridionalis and Notophthalmus kallerti Journal of Herpetology 2:121-127. Meylan P. A. 1981. The squamate reptiles of the Inglis IA fauna (lrvingtonian: Citrus County Florida). Bulletin of the Florida State Museum Biological Sciences 27:1-85. Mielke P. W. Jr. K. J. Berry and H.K. Iyer. In Press. Permutation methods in statistics: a distance function approach. Springer-Verlag. Moler P. E., and R. Franz. 1988. Wildlife values of small, isolated wetlands in the southeastern coastal plain. Pages 234-241 in: Proceedings of the Southeast Nongame and Endangered Wildlife Symposium. R. R. Odum K. A. Riddleberger, and J. C. Ozier ( eds.). Georgia Department of Natural Resources Social Circle Georgia. Moran N. 1992. The evolutionary maintenance of alternative phenotypes. The American Naturalist 139:971-989. Morin, P. J. 1986. Interactions between intraspecific competition and predation in an amphibian predator-prey system. Ecology:713720. Moritz C. 1994a. Defining evolutionary significant units" for conservation Trends In Ecology and Evolution 9:373-375. Moritz C. 1994b. Application of mitochondrial DNA analysis in conservation: a critical review. Molecular Ecology 3 : 401-411. Moritz C. C. J. Schneider and D. B. Wake. 1992. Evolutionary relationships within the Ensatina eschscholtzii complex confirm the ring species interpretation Systematic Biology 41 :273-291. Myers R. L. 1990. Scrub and high pine Pages 150-193 in: Ecosystems of Florida. R. L. Myers and J. J. Ewel (eds.). University of Central Florida Press Orlando Florida.

PAGE 158

149 Nei M. 1987. Molecular Evolutionary Genetics. Columbia University Press New York New York. Newman R. A. 1992. Adaptive plasticity in amphibian metamorphosis. BioScience 42:671-678. Osentoski M. F. and T. Lamb. 1995. Intraspecific phylogeography of the gopher tortoise Gopherus pol yp hemus: RFLP analysis of amplified mtDNA segments. Molecular Ecology 4:709718. Pandian T. J. and M. P. Marian. 1985. Predicting anuran metamorphosis and energetics. Physiological Zoology 58:538-552 Pechmann J H.K. D. E. Scott J W. Gibbons and R. D. Semlitsch. 1989. Influence of wetland hydroperiod on diversity and abundance of metamorphosing juvenile amphibians. Wetlands Ecology and Management 1 :3-11. Pechmann J. H.K. D. E. Scott R. D. Semlitsch J.P. Caldwell L. J. Vitt and J. W. Gibbons. 1991. Declining amphibian populations: the problem of separating human impacts from natural fluctuations. Science 253:892-895. Petranka J. W. 1998. Salamanders of the United States and Canada. Smithsonian Institution Press Washington, D.C. Pope P.H. 1924. The life-history of the common water-newt (Notophthalmus viridescens), together with observations on the sense of smell. Annals of the Carnegie Museum 15:305-368. Rao J S. 1976. Some tests based on arc-lengths for the circle. Sankhya Series B 38:329-338. Reilly S M. E. 0. Wiley and D. J. Meinhardt. 1997. An itegrative approach to heterochrony: the distinction between interspecific and intraspecific phenomena. Biological Journal of the Linnean Society 60:119-143 Ryan T. J. and R. C. Bruce 2000. Life history evolution and adaptive radiation of hemidactyliine salamanders. Pages 303-326 in: The Biology of Plethodontid Salamanders. R. C. Bruce R.G. Jaeger and L. D Houck ( eds.). Plenum Press Inc. New York, New York. R yan T. J. and R. D. Semlitsch. 1998. Intraspecific heterochrony and life history evolution: decouplin g somatic and sexual development in a facultatively paedomorphic sa l aman der. Proceedings of the National Academy of Sciences (U A) 95:5643-5648.

PAGE 159

150 Ryan, T. J. and R. D. Semlitsch. In Review. Growth and life cycle polymorphi s m in Ambystoma talpoideum I: expression of phenotypes. Evolutionary Ecology Research. SAS Institute Inc. 1990. SAS / STAT User s Guide version 6. 4th Edition, Volume 2. SAS Institute Inc. Cary North Carolina Scheiner S. M. 1993. Genetics and evolution of phenotypic plasticity. Annual Review of Ecology and Systematics 24:35-68. Schneider S. D. Roessli, and L. Excoffier. 1997. Arlequin. A software for population genetics data analysis version 1.1. Genetics Biometry Lab Dept. of Anthropology University of Geneva, Switzerland. Semlitsch, R. D. 1981. Terrestrial activity and summer home range of the mole salamander (Ambystoma talpoideum). Canadian Journal of Zoology 59:315-322. Semlitsch R. D. 1985. Reproductive strategy of a facultatively paedomorphic salamander Ambystoma talpoideum. Oecologia 65:305-313. Semlitsch R. D. 1987. Paedomorphosis in Ambystoma talpoideum: effects of density food and pond drying. Ecology 68:994-1002. Semlitsch R. D. 1998. Biological delineation of terrestrial buffer zones for pond breeding salamanders. Conservation Biology 12:113-119. Semlitsch R. D. 2000. Principles for management of aquatic-breeding amphibians. Journal of Wildlife Management 64:615-631. Semlitsch, R. D. and J. R. Bodie. 1998. Are small, isolated wetlands expendable? Conservation Biology 12:1129-1133. Semlitsch, R. D., and J. W. Gibbons. 1985. Phenotypic variation in metamorphosis and paedomorphosis in the salamander Ambystoma talpoideum. Ecology 66:11231130. Semlitsch R. D. R. N. Harris and H. M. Wilbur. 1990. Paedomorphosis in Ambystoma talpoideum: maintenance of population variation and alternative life-history pathways. Evolution 44: 1604-1613. Semlitsch R. D. and T. J. Ryan. 1999. Migration Amphibians. Pages 221-227 in: Encyclopedia of Reproduction. E. Knobil and J. D. Neill (eds.). Academic Press Inc. San Diego California.

PAGE 160

151 Semlitsch R. D. D E. Scott J H.K. Pechmann and J. W Gibbons. 1996. Structure and dynamics of an amphibian community: Evidence from a 16-year study of a natural pond. Pages 217-248 in: Long-term Studies of Vertebrate Communities. M L. Cody and J. A. Smallwood (eds.). Academic Press Inc. San Diego California. Semlitsch R. D. and H. W. Wilbur 1989 Artificial selection for paedomorphosis in the salamander Amb y stoma talpoideum. Evolution 43:105-112 Shaffer H.B. 1993. Systematics of model organisms: the laboratory axolotl Amb y stoma mexicanum. Systematic Biology 42:508 522. Shaffer H.B. J.M Clark and F. Kraus. 1991. When molecules and morphology clash: a phylogenetic analysis of the North American ambystomatid salamanders (Caudata: Ambystomatidae). Systematic Zoology. 40:284-303. Shaffer H. B. and S. R. Voss. 1996. Phylogenetic and mechanistic analysis of a developmentally integrated character complex: alternate life history modes in ambystomatid salamanders. American Zoologist 36:24-35. Shoop C. R. 1968. Migratory orientation of Amb y stoma maculatum: movements near breeding ponds and displacements of migrating individuals. The Biological Bulletin 135 :230-238. Skelly D. K. and E. E. Werner. 1990. Behavioral and life-historical responses of larval American toads to an odonate predator. Ecology 71:2313-2322. Slatkin M. 1985. Rare alleles as indicators of gene flow Evolution 39:53-65. Slatkin M. 1993. Isolation by distance in equilibrium and non-equilibrium populations. Evolution 47:264-279. Slatkin M ., and N. H Barton. 1989. A comparison of three indirect estimates for estimating average levels of gene flow. Evolution 43: 1349-1368. Smith-Gill S. J. and K A. Berven. 1979 Predicting amphibian metamorphosis. The American Naturalist 113:563-585. Snodgrass J. W. M J Komoroski A. L. Bryan Jr. and J. Burger. 2000. Relationships among isolated wetland size hydroperiod and amphibian species richness: implication s for wetland regulations. Conservation Biology 14 :414-419. okal R R ., and F J Rolf. 1995. Biometry Third edition. W. H. Freeman and Co ., N e w York N e w Yo r k.

PAGE 161

152 Stearns, S. C. 1989. The evolutionary significance of phenotypic plasticity. Bio c1ence 39:436-445. Stenhouse S. L. 1985 Migratory orientation and homing in Ambystoma maculatum and Ambystoma opacum. Copeia 1985:631-637. Stout, I. J. D.R. Richardson and R. E. Roberts. 1988. Management of amphibians reptiles and small mammals in xeric pinelands of peninsular Florida. Pages 98108 in: Management of Amphibians Reptiles and Small Mammals. R. C. Szaro K.E. Severnson and D.R. Patton (eds.). USDA Forest Service General Technical Report RM-166 Ft. Collins Colorado. Swofford, D.L. 1998. PAUP* 4.0bl: Phylogeny analysis using parsimony. Sinauer Sunderland Massachusetts. Tan A. M., and D. B. Wake. 1995. MtDNA phylogeography of the California newt Taricha torosa (Caudata Salamandridae). Molecular Phylogenetics and Evolution 4:383-394. Tompkins R. 1978. Genetic control of axolotl metamorphosis. American Zoologist 18:313-319. Travis J. 1984. Anuran size at metamorphosis: experimental test of a model based on intraspecific competition. Ecology 65: 115 5-1160. U.S. Fish and Wildlife Service. 1999. Final rule to list the flatwoods salamander as a threatened species. Federal Register 64: 15691-15704. Voss S. R 1995. Genetic basis of paedomorphosis in the axolotl, Ambystoma mexicanum: a test of the single-gene hypothesis. Journal of Heredity 86:441-447 Voss S. R., and H.B. Shaffer. 1997. Adaptive evolution via a major gene effect: paedomorphosis in the Mexican axolotl. Proceedings of the National Academy of Sciences 94:14185-14189. Voss S. R., and H. B. Shaffer. 2000. Evolutionary genetics of metamorphic failure using wild-caught vs. laboratory axolotls (Ambystoma mexicanum ). Molecular Ecology 9: 1401-1407. Wake D. B. 1991. Declining amphibian populations. Science 253:860. Wake D B. Morowitz H.J. Blaustein A. Bradford D. Bury R. B. Caldwel J. Corn P. S. Dubois A. Hayes M. Inger R ., Nettmann H. K. Rand A. S. Smith D. Tyler M ., and Vitt L. 1991. Declining amphibian populations a global phenomenon? Findings and recommendations Alytes 9:33-42

PAGE 162

153 Walker D. and J.C. A vise. 1998. Principles of phylogeography as illustrated by freshwater and terrestrial turtles in the southeastern United States. Annual Review of Ecology and Systematics 29:23-58. Wassersug R. J. 1974. Evolution of anuran life cycles. Science 185 :377-378. Watts W. A. and M. Stuiver. 1980. Late Wisconsin climate of northern Florida and the origin of species-rich deciduous forest. Science 210:325 :327. Webb S. D. 1990. Historical biogeography. Pages 70-100 in: Ecosystems of Florida. R. L. Myers and J. J. Ewel (eds.). University of Central Florida Press Orlando Florida. Webb S. D. and K. T. Wilkins. 1984. Historical biogeography of Florida Pleistocene Mammals. Pages 370-383 in: Contributions in Quaternary Vertebrate Paleontology: A Volume in Memorial to John E. Guilday. H. H. Genoways and M. R. Dawson (eds.). Special Publication of the Carnegie Museum ofNatural History No. 8. Werner E. E. 1986. Amphibian metamorphosis: growth rate predation risk and the optimal size at transformation. The American Naturalist 128 :319-341. White W. B. 1970. The geomorphology of the Florida peninsula. Florida Geological Survey Bulletin no. 51. Whiteman, H. H. 1994. Evolution of facultative paedomorphosis in salamanders. The Quarterly Review of Biology 69:205-221. Whiteman H. H. 1997. Maintenance of polymorphism promoted by sex-specific payoffs. Evolution 51 :2039-2044. Whiteman H. H. S. A. Wissinger, and W S. Brown. 1996. Growth and foraging consequences of facultative paedomorphosis in the tiger salamander Ambystoma tigrinum nebulosum. Evolutionary Ecology 10:433-446. Wilbur H. M. 1980. Complex life cycles Annual Review of Ecology and Systematics 11 :67-93. Wilbur H. M. and J.P. Collins. 1973. Ecological aspects of amphibian metamorphosis. Science 182:1305-1314. Winne C. T. and T. J. Ryan. 2001. Aspects of sex-specific differences in the expression of an alternative life cycle in the salamander Ambystoma talpoideum. Copeia: 2 001:143-149

PAGE 163

154 Worthington, R. D. 1968. Observations on the relative sizes of three species of salamander larvae in a Maryland pond. Herpetologica 24:242-246.

PAGE 164

BIOGRAPHICAL SKETCH Steve A. Johnson was born 19 April 1966 in St. Petersburg Florida. He grew up in Orlando, Florida where he attended Boone High School graduating in 1984. He attended Valencia Community College and transferred to the University of Central Florida, where he graduated with a Bachelor of Science degree in biology in 1990. While he was an undergraduate, Steve became involved with the U.C.F. Marine Turtle Research Group, under the direction of Dr. L. M. Ehrhart. Steve entered the graduate program at U.C.F. in 1991 and studied sea turtle ecology for his Master of Science research. His thesis was entitled "Reproductive Ecology of the Florida Green Turtle" and he graduated with his Master of Science degree in Biological Sciences in the summer of 1994. Steve began his Ph.D. program at the University of Florida in the spring of 1994 and thus was dual enrolled at U.F. and U.C.F. for two semesters. His initial research at U.F. was on the effects of organized sea turtles watches on loggerhead nesting behavior and hatchling production. However, to broaden his graduate education and research experience he initiated studies of the striped newt in 1996. While at U.F. Steve worked as a lab technician, graduate teaching assistant and graduate research assistant; and was funded for 2 years by a grant from the U .. Fish and Wildlife Service. He is a member of severa l professional soc ieties. 155

PAGE 165

156 Steve is married to his soul mate of ten years and counting Dale Ann John s on an artist and biological illustrator. They live with their four cats (Jojo Ginger Jasmine and Cedar), dog (Dude) parrot (Zipper) and several turtles. Steve plans to fish more after graduation.

PAGE 166

I certify that I have read this study and that in my opinion it conforms to acceptable tandards of scholarly presentation and is fully adequate in scope and quality, as a dissertation for the degree of Doctor of Philosophy {!.~fL~ C. Kenneth Dodd, Jr. C Associate Professor, Wildlife Ecology and Conservation I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy da~LQ~. L. Richard Franz Associate, Wildlife Ecology and Conservation I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is full adequate, in scope and quality as a dissertation for the degree of Doctor of Philosoph I certify that I have read this study and that in my opinion it conform to acceptable standards of scholarly presentation and is fully adequate in cope and quality as a diss rtation for the degree of Doctor of Phil~: hael P Moulton !~ Associate Professor, Wildlife Ecology and Con ervation I c rtify that I haver ad this study and that in my opinion it c nfi rm to acceptable tandard of cholarly pre entation and i fully adequat in c p and quality a a di ertat i n fi r the d egr e of Doctor of Philo ophy Brian n A i ant Profe r Fi heri and Aquatic

PAGE 167

I certify that I have read this study and that in my opinion it conform s to acceptable standards of scholarly presentation and is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy Mark Brenner Assistant Professor Fisheries and Aquatic Sciences This dissertation was submitted to the Graduate Faculty of the College of Agricultural and Life Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy August 2001 Dean College of Agricult Sciences Dean, Graduate School

PAGE 170

UNIVERSITY OF FLORIDA II I II IIIII I Ill I l l ll l ll l l lll I I l ll l ll l l ll ll l l ll l l 1 1 111111 11 111111 1 3 1262 08554 4525