`

1

INFLUENCE OF GEOMORPHOLOGY ON THE POPULATION STRUCTURE AND ECOLOGY OF THE HELLBENDER SALAMANDER (CRYPTOBRANCHUS

ALLEGANIENSIS)

By

KIRSTEN A. HECHT-KARDASZ

A THESIS PRESENTED TO THE GRADUATE SCHOOL

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT

OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2011

`

2

© 2011 Kirsten A. Hecht-Kardasz

`

3

To my son, Dmitry, and those who have dedicated their lives to conserving the natural world for future generations

`

4

ACKNOWLEDGMENTS

I am indebted to Dr. Max Nickerson and Dr. Perran Ross for their advice, ideas,

and assistance with my research project and manuscript. I would also like to thank

them in addition to The School of Natural Resources and Environment and the Ross

Family Syndicate for financial assistance that prevented homelessness and starvation

during my time as a student. Thank you to Phil Colclough and the Knoxville Zoo, Dr.

Michael Freake, and Dr. Marcy Souza for providing data, technical support, and input.

Special thanks to Andrea Drayer for help and advice, members of the Williams Lab at

Purdue University and all volunteers for their field assistance, and Melrose Flockhart for

lending me a vehicle during my time in need. Financial support for this research was

provided by the Great Smoky Mountains Conservation Association: Carlos C. Campbell

Fellowship, The Reptile and Amphibian Conservation Corp, and the Cryptobranchid

Interest Group: Jennifer Elwood Conservation Grant. For housing assistance during my

field research, I would like to thank The Great Smoky Mountains Institute at Tremont,

the Lodge at Valley View, and Century 21 Great Smoky Mountains Realty. I would also

like to acknowledge the Great Smoky Mountains National Park and Paul Super for

allowing me to conduct my research within the park and for their assistance. Last, but

certainly not least, I would like to thank my family, particularly my husband and mother,

for their support. This work would not have been possible without them. Research was

conducted under National Park Service scientific research permit GRSM-2008-SCI-

0052 and University of Florida ARC Protocol #017-08WEC.

`

5

TABLE OF CONTENTS page

ACKNOWLEDGMENTS .................................................................................................. 4

LIST OF TABLES ............................................................................................................ 7

LIST OF FIGURES .......................................................................................................... 8

LIST OF ABBREVIATIONS ........................................................................................... 10

ABSTRACT ................................................................................................................... 11

CHAPTER

1 INTRODUCTION .................................................................................................... 13

Factors Influencing Animal Populations .................................................................. 13

The Eastern Hellbender .......................................................................................... 17 General Description .......................................................................................... 18

Life History ....................................................................................................... 20 Population Studies ........................................................................................... 21

Geomorphology of the Smoky Mountains and Geology of Little River .................... 23 Objectives ............................................................................................................... 25

2 MATERIALS AND METHODS ................................................................................ 34

Study Site ............................................................................................................... 34

Field Sampling Methods ......................................................................................... 36 Data Analysis .......................................................................................................... 38

3 RESULTS ............................................................................................................... 44

General Results ...................................................................................................... 44

Population Structure ............................................................................................... 44 Microhabitat ............................................................................................................ 45

Body Condition ....................................................................................................... 46 Larval Diet ............................................................................................................... 47

4 DISCUSSION ......................................................................................................... 68

Population Structure ............................................................................................... 68

Microhabitat ............................................................................................................ 72 Body Condition ....................................................................................................... 78

Larval Diet ............................................................................................................... 80

5 CONCLUSIONS ..................................................................................................... 88

APPENDIX: WENTWORTH PARTICLE SIZE CATEGORIES ...................................... 91

`

6

LIST OF REFERENCES ............................................................................................... 92

BIOGRAPHICAL SKETCH .......................................................................................... 101

`

7

LIST OF TABLES

Table page

3-1 Variable estimates and odds ratios from an ordinal logistic regression model based on streambed particle size classes at sites used by larval (n=25), sub-adult (n=26), and adult (n=38) hellbenders (Cryptobranchus alleganiensis) captured in Little River, TN. ................................................................................ 49

3-2 Variable estimates and odds ratios from a binomial logistic regression model based on streambed particle size classes at sites used by hellbenders (Cryptobranchus alleganiensis) (n=89) and random locations (n=50) within Little River, TN. ................................................................................................... 50

3-3 Variable estimates and odds ratios from a binomial logistic regression model based on streambed particle size classes (with particles <32 mm combined into one category) at sites used by hellbenders (Cryptobranchus alleganiensis) (n=89) and random locations (n=50) within Little River, TN. ........ 51

3-4 Variable estimates and model fit for linear regressions of hellbender (Cryptobranchus alleganiensis) body condition (mass (g) vs. transformed total length (mm)) in three rivers. ........................................................................ 52

3-5 Contents of diet samples taken from larval hellbenders (Cryptobranchus alleganiensis) in Little River, TN ......................................................................... 53

`

8

LIST OF FIGURES

Figure page

1-1 Historic range of the eastern hellbender (Cryptobranchus alleganiensis alleganiensis) and the Ozark hellbender (Cryptobranchus alleganiensis bishopi) in the eastern United States .................................................................. 28

1-2 Adult hellbender (Cryptobranchus alleganiensis) in Little River, TN demonstrating morphological adaptations to the stream environment including dorsally flattened body, cryptic coloration, lateral folds, and toe discs. .................................................................................................................. 29

1-3 Gilled larval hellbender (Cryptobranchus alleganiensis) measuring 69 mm in total length captured in Little River, TN.. ............................................................. 30

1-4 Percentage of larval hellbenders in sampled populations of hellbenders in the North Fork of White River, MO (n=10); Niangua River, MO (n=3); Little River, TN (n=16), and other Missouri rivers (n=1) (Spring River, Eleven Point River, Gasconade River, Big Piney River).. .................................................................. 31

1-5 Histogram of the size distribution of hellbenders captured in 2000 (n=33) during surveys of Little River, TN. ....................................................................... 32

1-6 Physiographic provinces of the Appalachians .................................................... 33

2-1 Human recreational use on Little River, TN in Great Smoky Mountains National Park.. .................................................................................................... 42

2-2 Study site (Little River, TN; USA). ...................................................................... 43

3-1 Histogram of size distribution of captured hellbenders (Cryptobranchus alleganiensis) from 2000-2010 in Little River, TN (n=500). ................................. 54

3-2 Yearly size distribution histograms of captured hellbenders (Cryptobranchus alleganiensis) from 2000-2010 in Little River, TN. .............................................. 55

3-3 Comparison of hellbender (Cryptobranchus alleganiensis) size class distributions sampled from Little River, TN in 2006 (n=113) and 2008 (n=117), with the North Fork of the White River, MO in 1969 (n=478) ................ 56

3-4 Scatter plot with linear regression line of water temperature (C) vs. hellbender total length (mm) in Little River, TN (n=102). .................................... 57

3-5 Scatter plot with linear regression line of water depth (mm) vs. hellbender total length (mm) in Little River, TN (n=104). ...................................................... 58

`

9

3-6 Scatter plot with linear regression line of shelter size (mm) vs. hellbender total length (mm) in Little River, TN (n=217). ...................................................... 59

3-7 Box plots comparing shelter size (mm) among three hellbender stage classes, larvae (n=61), sub-adults (n=56), and adults (n=100), in Little River, TN . ..................................................................................................................... 60

3-8 Bar graph showing mean ±standard error of the mean (SEM) for shelter size (mm) used by three stage classes of hellbenders, larvae (n=61), sub-adults (n=56), and adults (n=100), in Little River, TN.. .................................................. 61

3-9 Results of Wolman pebble count survey in the Little River, TN showing streambed particle size distribution in Little River, TN (D50=small cobble). ....... 62

3-10 Bar graph comparing streambed particle size categories found below shelter rocks among hellbender stage classes, larvae (n=25), sub-adults (n=26), and adults (n=38) in Little River, TN. ......................................................................... 63

3-11 Bar graph comparing streambed particle size categories found at sites used by hellbenders (n=89) and random locations (n=50) in Little River, TN. ............. 64

3-12 Examples of abnormalities of Cryptobranchus alleganiensis captured in Little River, TN.. .......................................................................................................... 65

3-13 Scatter plot with regression lines comparing body condition of hellbenders (Cryptobranchus alleganiensis) from three rivers (Little River, TN (n=527); Hiwassee River, TN (n=507); North Fork of the White River; MO (n=463) with differing crayfish relative frequencies. ................................................................ 66

3-14 Pie chart of total food items identified from larval hellbender (Cryptobranchus alleganiensis) diet samples (n=23) taken from Little River, TN. .......................... 67

4-1 Grouped histogram showing differences in size class distributions of hellbenders (Cryptobranchus alleganiensis) captured in Little River, TN from 2000-2010 (n=500) and the North Fork of the White River, MO in 1969 (n=478). .............................................................................................................. 85

4-2 Map of the eastern United States showing protected areas in the southern Appalachian and Ozark regions.......................................................................... 86

4-3 Annual peak streamflow at Little River, TN USGS station within Great Smoky Mountains National Park .................................................................................... 87

`

10

LIST OF ABBREVIATIONS

ANCOVA Analysis of covariance

ANOVA Analysis of variance

CITES Convention on International Trade in Endangered Species of Wild Flora and Fauna

FISP Federal Interagency Sedimentation Project

GSM Great Smoky Mountains

GSMNP Great Smoky Mountains National Park

HR Hiwassee River, Tennessee

IUCN International Union for Conservation of Nature

LR Little River, Tennessee

MPLR Middle Prong of Little River, Tennessee

NFWR North Fork of the White River, Missouri

PIT Passive Integrated Transponder

SVL Snout-vent length

TDS Total dissolved solids

TL Total length

USGS United States Geological Survey

VIE Visible Implant Elastomer

`

11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the

Requirements for the Degree of Master of Science

INFLUENCE OF GEOMORPHOLOGY ON THE POPULATION STRUCTURE AND ECOLOGY OF THE HELLBENDER SALAMANDER (CRYPTOBRANCHUS

ALLEGANIENSIS)

By

Kirsten A. Hecht-Kardasz

December 2011

Chair: Max A. Nickerson Cochair: James Perran Ross Major: Interdisciplinary Ecology

The hellbender (Cryptobranchus alleganiensis) is an imperiled salamander that

has experienced declines in many parts of its range. Little knowledge of the general

ecology of the larval stage exists because studies including all stage classes,

particularly larvae, are rare. In 2000, a brief study in Great Smoky Mountains National

Park discovered a population of C. alleganiensis where larvae were regularly

encountered and few adults were captured. The authors of the 2000 study

hypothesized that the size structure of hellbenders in Little River, Tennessee was

potentially influenced by differences in larval hellbender habitat and crayfish abundance

from other studied localities due to the geologic structure of the streambed. To further

investigate this hypothesis, this study examined three main components: trends in

hellbender population structure, microhabitat use of hellbender stage classes within

Little River, and body condition of hellbenders in rivers with different crayfish

abundances. Diurnal skin diving surveys were conducted in the summer and early fall

months of 2008-2010 to locate hellbenders and collect habitat, morphometric, and diet

data. Surveys conducted since 2000 in Little River suggest that the hellbender

`

12

population appears stable, with abundant larvae and regular recruitment. The number

of larvae in Little River appears to be much higher than in most studied hellbender

populations. Flooding induced mortality of larvae may affect long-term population

structure in Little River due to the influence of stream geomorphology on larval habitat

use. Of the habitat variables measured during this survey, only shelter size appears to

differ among stage classes, with larvae utilizing smaller shelters on average than adults

and sub-adults. Very coarse gravel was positively associated with hellbender

occupancy in Little River. Abundance of crayfish, based on crayfish relative

frequencies, correlated to overall body condition of hellbenders in the three rivers

examined. Stomach samples collected from larvae suggest that hellbenders experience

an ontogenetic shift in diet, with young individuals primarily consuming aquatic insect

larvae. These results help fill in knowledge gaps regarding the larval stage of the

hellbender, as well as highlight the potential impacts of stream geomorphology on the

ecology of a hellbender population.

`

13

CHAPTER 1 INTRODUCTION

Factors Influencing Animal Populations

Mechanisms influencing dynamics and regulation of animal populations have

been a major research focus and a source of debate throughout the history of ecology

(Murdoch, 1994). Although understanding these mechanisms can shed light on a

number of ecological questions including the dynamics of communities, biotic

interactions, evolutionary history, and conservation status of populations, determining

which factors ultimately affect populations is complex as several factors, both biotic and

abiotic, can work concurrently (Murdoch, 1994; Rodenhouse et al., 1997). Furthermore

individual populations and even subsets within a particular population can experience

different responses to similar mechanisms (Dobson and Oli, 2001). Despite these

inherent difficulties, identifying mechanisms is important because of their influence on

the basic demographics of populations including birth and death rates, immigration,

emigration, growth rates, and fecundity.

Factors influencing populations are often classified into density-dependent and

density-independent categories. The impacts of biotic factors such as disease and

biotic interactions, including predation and competition, are typically linked to population

densities and are therefore considered density-dependent. In contrast, many abiotic

factors are categorized as density independent. Climatic extremes and natural

disasters such as flooding are examples of factors that work independently of

population density. The physical habitat where a population lives may also serve as an

important abiotic influence on population dynamics. Habitat characteristics such as

`

14

shelter type and availability, chemistry, soils, and geology all potentially affect the

demographics of populations.

In aquatic systems the role of the physical environment is particularly important.

Obligate aquatic species cannot typically relocate to new environments due to

physiologic limitations. Physical tolerances also limit the range of species within the

aquatic system itself. Stream environments are unique and harsh environments to

inhabitant due to constant environmental fluctuations and large spatial and temporal

variation (Giller and Malmqvist, 1998; Peterson et al., 1983). The environment,

therefore, is a critical factor in determining the density, structure, and distribution of

populations within streams. The range of stream organisms is essentially a marriage of

the physiochemical environments present in the stream with the physical tolerances of

the animals (Giller and Malmqvist, 1998).

Habitat variables that influence inhabitants vary depending on scale. At smaller

scales stream flow patterns, temperature, and substrate are key components which

influence how and where organisms live in the stream environment (Giller and

Malmqvist, 1998). The geologic setting, riparian vegetation, and land use surrounding

streams become more important factors at larger scales (Giller and Malmqvist, 1998).

Geomorphology, defined as the study of Earth’s landforms and the processes that

shape them, directly and indirectly shapes many stream attributes. The geologic history

of a stream determines its bedrock structure, which in turn affects stream characteristics

and processes. Geomorphology can influence benthic substrate, flow patterns, pH,

water chemistry, and nutrient levels within a stream, thereby influencing stream biota

and communities on both local and regional scales (Swanson, 1980). Differences in

`

15

production and density of stream macro-fauna have been linked to variations in

geomorphology at both local levels and between different geologic provinces (Huryn

and Wallace, 1987; Hwa-Seong and Ward, 2007; Raymond Bouchard, unpubl. data).

Eco-regions based on geological provinces have also been noted to contain different

assemblage groups (Rashleigh, 2004).

Research examining influences on population dynamics has historically focused

on overall population density or abundance, which can give an incomplete picture of

population status. The structure of a population can reveal a great deal more about the

status of a population than population size estimates alone and is particularly valuable

in conservation where understanding the threat of extinction for a population is

important (Alexander, 1958; Downing, 1980; Gillespie, 2010). In species of

conservation concern, age or stage composition can indicate overall population stability

and lead to more accurate predictions regarding future population trends (Crowder et

al., 1994). A population composed primarily of older individuals may be at risk of

decline or extirpation due to low recruitment (Alexander, 1958; Downing, 1980). A

population with few older individuals, but many young individuals could indicate

population growth, high adult mortality, or a failure to recruit young life stage classes

into adults (Alexander, 1958; Downing, 1980). Understanding the structure of a

population is also important because different age or stage classes can react differently

to their environment and dissimilarly affect overall population dynamics (Dobson and

Oli, 2001).

In aquatic environments organisms often adapt life strategies that can cause

differences in demographic rates among age or stage classes. Many stream species,

`

16

including fish, aquatic insects, and amphibians develop complex life cycles or

ontogenetic shifts in habitat use and diet, which are believed to be adaptations for

increasing survival in a stressful environment (Werner and Gilliam, 1984; Foster et al.,

1988; Giller and Malmqvist, 1998). These shifts can serve as a type of refugia, limiting

intra-specific competition and predation (Werner and Gilliam, 1984; Colley et al., 1989;

McGrath et al., 2007). While these adaptations may help reduce individual mortality,

they can also make studying populations more complicated.

Studies examining amphibian population structures are particularly needed.

While amphibian populations are currently declining worldwide (Alford and Richards,

1999; Vié et al., 2009), population dynamics and demographics of many amphibians

remain unstudied (Duellman and Trueb, 1986; Alford and Richards, 1999; Swanack et

al., 2009; Gillespie, 2010; Lips, 2011). As obtaining amphibian population and life

history data that accurately considers all stage classes is difficult due to complex life

cycles and ontogenetic shifts, data is often lacking for specific size or stage classes

(Swanack et al., 2009; Gillespie, 2010). Larval and juvenile classes can be difficult to

study because they are generally cryptic, small, and sometimes use different habitat

than other stages (Gillespie, 2010). Therefore, understanding the class-structured

dynamics of amphibian populations remains problematic. The resulting information

gaps have hindered researchers from fully comprehending the scope of amphibian

declines and prevented the identification of mechanisms affecting individual populations

(Alford and Richards, 1999; Gillespie, 2010; Lips, 2011). Once population declines

occur, information is even more difficult to obtain as individuals become rare (Gillespie,

2010).

`

17

One amphibian species with few studies regarding its basic demographics and

population dynamics is the hellbender salamander, Cryptobranchus alleganiensis

(Daudin 1803). Currently listed as near threatened on the International Union for

Conservation of Nature (IUCN) red list (Hammerson and Phillips, 2004), hellbender

populations appear to be declining in many parts of the range (Trauth et al., 1992;

Wheeler et al., 2003; Briggler et al., 2007; Foster et al., 2009; Nickerson et al., 2009;

Burgmeier, 2011b). The exact cause or causes of the declines remain difficult to

pinpoint, but stream siltation, disease, collection, species introductions, and habitat loss

are just a few of the cited problems facing this species (Trauth et al., 1992; Hiler et al.,

2005; Briggler et al., 2007; Nickerson and Briggler, 2007; Nickerson et al., 2009). Due

to these declines the hellbender is protected at the state-level throughout most of its

range, and was recently listed in Appendix III of CITES (Convention on International

Trade in Endangered Species of Wild Flora and Fauna) and the Federal endangered

species list (Anonymous, 2011).

The Eastern Hellbender

Cryptobranchus alleganiensis is a member of the ancient giant salamander

family Cryptobranchidae, which was present over 160 million years ago in China based

on the fossil record (Gao and Shubin, 2003). Evidence of Cryptobranchidae in North

America begins in the Upper Paleocene (Naylor, 1981). Today, the lone extant North

American species of these long-lived, large, aquatic salamanders resides primarily in

cool, oxygen-rich streams in the mountainous regions of the eastern United States.

There are currently two accepted subspecies: C. a. alleganiensis, the eastern

hellbender, which is found in the majority of the range, and C. a. bishopi, the Ozark

hellbender which is found in a small area of Missouri and Arkansas (Figure 1-1).

`

18

General Description

Hellbenders are one of the largest salamander species in North America. The

largest hellbender ever recorded was captured in the Little Pigeon River just outside

Great Smoky Mountains National Park (GSMNP) in Gatlinburg, TN. This female

measured 740 mm in total length (TL) (Fitch, 1947). The largest known male at 686

mm TL was captured near the same locality in 1936 (King, 1939). Despite females

generally reaching overall longer lengths than males (Bishop, 1941; Humphries and

Pauley, 2005), the length-specific mass of the sexes can vary depending on locality and

time of year (Bishop, 1941; Taber et al., 1975; Peterson et al., 1983; Peterson et al.,

1998). Studies also suggest that the sexes differ in age at sexual maturity with females

generally being older and thus larger before reaching adulthood (Dundee and Dundee,

1965; Taber et al., 1975; Peterson et al., 1983). Due to the lack of distinct

morphological differences, the sexes are extremely difficult to differentiate in the field

throughout most of the year (Bishop, 1941; Hillis and Bellis, 1971; Taber et al., 1975).

During the breeding season in late summer and early fall, however, males develop

swollen cloacae making them easy to discern from females (Bishop, 1941; Nickerson

and Mays, 1973).

Cryptobranchus alleganiensis has adapted to life in stream riffles by developing

many life strategies common to animals in stream environments (Figure 1-2). Cryptic

coloration helps them blend in with the streambed substrate where they reside

(Nickerson and Mays, 1973). Coloration includes dark brown, grey, olive-green,

orange, and yellow with darker mottling commonly present (Nickerson and Mays, 1973;

Petranka, 1998). A large keeled tail and rough toe pads assists the hellbender in

navigating stream bottoms (Nickerson and Mays, 1973). Their heads and bodies are

`

19

dorso-laterally flattened allowing them to squeeze into small spaces beneath rocks and

minimize resistance from stream flow (Nickerson and Mays, 1973).

The hellbender produces skin secretions, which may help them slide beneath

rocks as well as serve other adaptive purposes (Nickerson and Mays, 1973). These

secretions are believed to be somewhat toxic and distasteful, serving as a potential

defensive mechanism against predation (Brodie, 1971; Nickerson and Mays, 1973).

Individuals under stress, such as during capture, are known to “slime” profusely, making

them slippery and difficult to handle, which may represent an additional deterrent

against predators. Mucous skin secretions may also aid in respiration (Nickerson and

Mays, 1973).

The genus Cryptobranchus, which translates as “hidden gills”, was named in

reference to the lack of external gills in adults. Although external gills are lost as larvae

transition into sub-adults (Bishop, 1941), hellbenders retain gill slits throughout life, and

are thus considered paedomorphic (Nickerson and Mays, 1973; Petranka, 1998;

Nickerson, 2003). Mature C. alleganiensis primarily respire cutaneously, with lateral

folds on the body trunk assisting in this process (Guimond, 1970; Nickerson and Mays,

1973). Adults also possess lungs, but these appear to serve mostly as instruments for

buoyancy, as they are essentially non-functioning for respiration (Hughes, 1967;

Guimond, 1970).

Despite small lidless eyes, which are flattened like many aquatic amphibians,

hellbenders appear to rely on visual cues in addition to chemical and tactile stimulus for

feeding (Nickerson and Mays, 1973). They predominantly consume whole live food

items by asymmetrical suction feeding where suction is created when one side of the

`

20

mandible rapidly descends (Nickerson, 2003). Cryptobranchus alleganiensis is largely

nocturnal and forages at night and occasionally during overcast daylight hours

(Nickerson and Mays, 1973). Throughout their range, crayfish comprise the majority of

the adult hellbender diet, but they are known to consume a number of additional prey

items including fish, snails, worms, aquatic insects, amphibians, aquatic reptiles,

carrion, and even lampreys in some localities (Smith, 1907; Netting, 1929; Green, 1935;

Bishop, 1941; Nickerson and Mays, 1973; Nickerson et al., 1983, Peterson et al.,1989).

Cryptobranchus alleganiensis is also cannibalistic and will eat eggs and smaller

conspecifics (King, 1939; Nickerson and Mays, 1973; Humphries et al., 2005). Non-

food items, including rocks and leaves, have been found in stomach samples and are

believed to be incidental due to the feeding strategy of C. alleganiensis (Netting, 1929;

Nickerson and Mays, 1973; Peterson, 1989).

Life History

Cryptobranchus alleganiensis passes through four distinct life stages: egg, larva,

sub-adult, and adult. Reproductive timing varies throughout the species range, but

generally occurs in the late summer and early fall months (Smith, 1907; Smith, 1912;

Green, 1933; King, 1939; Bishop, 1941; Dundee and Dundee, 1965; Nickerson and

Mays, 1973). Hellbenders congregate at large boulder nest-rocks and sometimes

exhibit communal breeding (Smith, 1907; Peterson, 1988). After courtship, females lay

large strings containing approximately 270 to 450 eggs, which are fertilized externally by

males and often cannibalized by both sexes (Smith, 1907; Bishop, 1941; Nickerson and

Mays, 1973; Peterson, 1988). Following fertilization one male guards the nest from

predation and “fans” the eggs providing aeration (Smith, 1907; Bishop, 1941). Eggs

incubate approximately 1-2 months before hatching (Bishop, 1941; Peterson, 1988).

`

21

Hatchlings emerge at approximately 20-30 mm TL with a yellow yolk sac visible

in their abdomens, external gills, and undeveloped limbs (Smith,1907; Bishop 1941).

They are drab in color and lack the lateral skin folds noted on adults. Larvae quickly

adapt the typical coloration of C. alleganiensis and absorb their yolk sac (Figure 1-3).

After 1-2 years individuals attain 130 mm TL (Bishop, 1941) and reabsorb their external

gills, becoming sub-adults. Sub-adults closely resemble adult C. alleganiensis but have

not yet reached maturity. Although location and sex appear to influence the timing of

adulthood, hellbenders generally reach sexual maturity between 5-8 years of age

(Dundee and Dundee, 1965; Taber et al., 1975; Peterson et al., 1988). Life expectancy

is unknown in the wild, but one individual lived 29 years in captivity (Nigrelli, 1954).

While a basic understanding of the life cycle of the hellbender is known, much

less is known about the sub-adult stage of this species, and virtually nothing is known

about larvae in the wild. Only two larval diet samples have been published (Smith,

1907; Pitt and Nickerson, 2006) and very few studies have focused on the ecology,

habitat and behavior of larvae mostly due to the rarity of encountering larvae in most

localities. In contrast, knowledge of the ecology and natural history of the adult

hellbender, despite some gaps, is relatively abundant. It is therefore not surprising that

studies regarding the population ecology of the species have been biased towards

larger individuals.

Population Studies

Despite the conservation interest in C. alleganiensis, data is sparse regarding the

population dynamics of this species. Many localities lack historic data regarding

population size and status, and many demographics of this species remain poorly

understood. Population studies have primarily focused on snapshot estimates of

`

22

population size or adult population structure. Few studies have examined growth rates,

fecundity, and survivability, and these studies have only been completed in a few

localities in the Ozark region, which may not be representative for hellbenders across

their range, particularly for the eastern subspecies (Topping and Ingersol, 1971; Taber

et al., 1975; Peterson et al., 1988).

Limited historical data from a few previously studied drainages in New York and

Missouri have given better insight into long-term population trends (Wheeler et al.,

2003; Foster et al., 2009). Recent research in these areas indicated that some

populations are declining as well as shifting in overall structure. Comparisons of

historical and recent data in Missouri populations suggested that in declining hellbender

populations, size class distributions were shifting towards larger individuals, possibly

indicating inadequate recruitment (Wheeler et al., 2003). Foster et al. (2009) noted not

only overall declines in the population size of hellbenders in New York’s Alleghany River

drainage, but also shifts in sex ratio towards a male-biased population. In both of these

studies, young individuals <20 cm (i.e. larvae and small sub-adults) were missing from

samples. It remains uncertain whether these size classes were absent from the

population or inadequately sampled perhaps due to their association with interstitial

spaces in gravel beds (Nickerson and Krysko, 2003). Regardless, little is known about

larval hellbenders and few studies include data on larvae. However, in 2000, Nickerson

et al. (2002) discovered a unique hellbender population while conducting research in a

Tennessee stream.

During efforts by the U.S. Geological Survey (USGS) to catalog the amphibians

of GSMNP, surveys were conducted in many of the park’s streams to search for C.

`

23

alleganiensis as well as Necturus maculosus, the common mudpuppy. A short survey

of the hellbender population in Little River (LR), Tennessee, yielded 33 individuals, of

which 48% (n=16) were larval sized (<130 mm) (Nickerson et al. 2002). This

percentage was in stark contrast to those recorded for other hellbender populations

(e.g. Peterson et al., 1988; Wheeler et al., 2003; Foster et al., 2009) (Figure 1-4).

Furthermore, the proportion of adult hellbenders to larvae within LR (Figure 1-5) was the

lowest of any studied river system (Nickerson et al., 2003). Additional research has

confirmed that larvae are regularly captured in this stream from year to year, and that

although adult hellbenders are more abundant than originally thought, very large adults

are still rare (Freake, unpubl. data).

Geomorphology of the Smoky Mountains and Geology of Little River

The Great Smoky Mountains (GSM) are part of the Southern Appalachian

Region, which lies within the Appalachian Highlands of the eastern United States. This

area is comprised primarily of four physiographic provinces: Appalachian Plateau,

Valley and Ridge, Blue Ridge, and Piedmont (Figure 1-6). Fenneman (1917) formed

these classifications based primarily on differences in geomorphology and geologic

history. Rocks in Southern Appalachia vary throughout the region and include

limestone, dolomite, shale, sandstone, siltstone, gneiss, schist, phyllite, as well as a

variety of metamorphosed rock. The Appalachian region displays this diversity in rock

type and structure due to the unique geologic history of each province. The GSM are

located within the Blue Ridge Province.

The Blue Ridge Province is a high mountainous area of the Southern

Appalachians and also includes the Unaka and Blue Ridge Mountains. Metamorphic

rock comprises most of the region. Rocks underlying the Southern Appalachians,

`

24

including the Blue Ridge Province, formed over one billion years ago, when the

supercontinent Rodinia formed. As the continental crust expanded 750 million years

ago, Rodinia began to pull apart, forming a deep basin. The Ocoee Basin, which was

located in present day east Tennessee, western North and South Carolina, and

northern Georgia, filled with saltwater. Vast amounts of river transported sediments,

including sand, silt, clay, and gravel, settled within the basin, creating what would

become the bedrock of the GSM. After millions of years of sedimentation within the

basin, continents began to move together again. Around 270 million years ago, the

continents that became modern day North America and Africa collided, essentially

pushing the formed rocks upward and forming the Appalachian Mountains. In the Blue

Ridge Province, the pressure and heat from the continental collision metamorphosed

much of the sedimentary rock. Following millions of years of weathering and erosion,

the landscape we see today developed.

LR lies entirely within the southern portion of the Blue Ridge physiographic

province. The bedrock of LR is comprised primarily of late Precambrian Elkmont and

Thunderhead metamorphosed sandstone (Mast and Turk, 1999). Over time, the flowing

water has eroded away some exposed bedrock leaving large densities of dense

rounded boulders, cobble, and gravel in the streambed. Hellbenders are generally

associated with large flat rocks, which they utilize for shelter and nest rocks (Bishop,

1941, Hillis and Bellis, 1971). Many of the well-studied hellbender streams contain

bedrock comprised largely of limestone and other sedimentary rocks. These types of

rocks are more susceptible to weathering and erosion than metamorphic rock and

typically fragment and erode into flat slabs rather than the rounded rocks characteristic

`

25

of LR. Interstitial habitat is limited within the LR streambed as sand often fills in many

portions of the gravel beds. The lack of interstitial space in the gravel substrate as well

as the availability of rounded shelters makes LR different from most other studied

hellbender localities in its streambed morphology.

Objectives

Nickerson et al. (2003) hypothesized that “the less secure larval habitat within the

LR makes larvae more susceptible to capture, and coupled with reduced crayfish

populations, translates to fewer adult C. alleganiensis.” Hellbender larvae found in

other locations are often located within the interstitial spaces of gravel (Smith, 1912;

Nickerson and Mays, 1973; Nickerson et al., 2003). These spaces are unavailable

within LR due to the characteristics of the streambed. Furthermore, crayfish, the

principle component of the adult hellbender diet, appear to be scarcer and smaller in

size in LR than in other well-studied hellbender localities (Nickerson et al, 2003). This

trend may be influenced by the bedrock structure of LR as previous researchers have

noted smaller densities of crayfish in streams with non-carbonated bedrock in

comparison to sites comprised of carbonate rocks (Raymond Bouchard, pers. comm.).

Trends of higher production rates in hard water streams have also been noted in a

number of other macro-invertebrate species and may be caused by the influence of

water chemistry on algae and microbes, which ultimately impacts other biota through

trophic interactions (Hwa-Seong and Ward, 2007).

Based on the hypothesis of Nickerson et al. (2003) which links the population

structure of hellbenders to geologic characteristics as well as crayfish frequencies, this

study was conducted to further examine the role of the geomorphology on the general

ecology and population dynamics of C. alleganiensis within LR. This study examined

`

26

three main components: population structure, microhabitat, and body condition of LR’s

hellbenders. The first objective of this study was to examine the population structure in

more detail, and investigate the population structure of hellbenders over multiple years.

The findings of Nickerson et al. (2002) were limited by small sample size and reduced

search hours. Additional study was needed to confirm the differences in hellbender

population structure from well-studied streams.

The second objective was to quantify microhabitat use of C. alleganiensis,

particularly among the larval stage, within LR during the summer and early fall months.

Microhabitat available to stream-dwelling animals is often reliant upon the general

geomorphology of the stream. In the case of LR, the geologic structure of the

streambed has mostly eliminated the type of habitat typical of larvae found in other

streams. Although the lack of interstitial spaces within the gravel has been suspected of

limiting the recruitment of young individuals within LR, no specific study of larval habitat

has been undertaken. While Nickerson et al. (2003) suggested that larval habitat

differed in LR when compared to other localities, the actual habitat being utilized in LR

has been only anecdotally noted. I hypothesized that despite the lack of interstitial

spaces, there would still be some type of ontogenetic shift in microhabitat use among

stage classes. These shifts serve as a form of refugia, reducing rates of intraspecific

competition and predation among stage classes, and are common among aquatic

organisms (Werner and Gilliam, 1984; Giller and Malmqvist, 1988).

The low abundance of crayfish was also cited as a potential cause for the lack of

large adult C. alleganiensis within LR (Nickerson et al., 2003). If prey densities do

indeed impact the survival of adult hellbender, I would expect them impact the body

`

27

condition of hellbenders. To investigate the impact of low prey densities on the body

condition of hellbenders, body condition was compared in rivers with differing crayfish

frequencies. I hypothesized that hellbenders from a river with a low crayfish relative

frequency would exhibit poorer body condition than individuals from rivers exhibiting

higher crayfish abundance. As the natural diet of larval hellbenders is relatively

unknown, the diet of larvae was also investigated to determine if low crayfish densities

might also affect the youngest stage class.

`

28

Figure 1-1. Historic range of the eastern hellbender (Cryptobranchus alleganiensis

alleganiensis) and the Ozark hellbender (Cryptobranchus alleganiensis bishopi) in the eastern United States (Modified from U.S. Geological Survey National Amphibian Atlas, 2010)

`

29

Figure 1-2. Adult hellbender (Cryptobranchus alleganiensis) in Little River, TN

demonstrating morphological adaptations to the stream environment including dorsally flattened body, cryptic coloration, lateral folds, and toe discs. Photo courtesy of Kirsten Hecht-Kardasz.

`

30

Figure 1-3. Gilled larval hellbender (Cryptobranchus alleganiensis) measuring 69 mm in

total length captured in Little River, TN. Photo courtesy of Kirsten Hecht-Kardasz.

`

31

n=765 n=1132 n=1209 n=33

Figure 1-4. Percentage of larval hellbenders in sampled populations of hellbenders in

North Fork of White River, MO (n=10); Niangua River, MO (n=3); Little River, TN (n=16), and other Missouri rivers (n=1) (Spring River, Eleven Point River, Gasconade River, Big Piney River). (Modified from Nickerson et al., 2003).

`

32

Figure 1-5. Histogram of the size distribution of hellbenders captured in 2000 (n=33) during surveys of Little River, TN.

(Modified from Nickerson et al., 2003)

`

33

Figure 1-6. Physiographic provinces of the Appalachians (Modified from Fenneman and

Johnson, 1946)

`

34

CHAPTER 2 MATERIALS AND METHODS

Study Site

LR, located in eastern Tennessee’s Blue Ridge Province, originates on the north

slope of the highest point in both the state and GSMNP, Clingmans Dome. After

reaching Elkmont, the river flows alongside Little River Road (Scenic TN 73) to the

Townsend “Wye”, which refers to the spot where the Middle and West Prongs meet with

the main section of LR before exiting the park entrance at Townsend. After flowing 29

km within GSMNP the river continues through the towns of Townsend, Maryville, Alcoa,

and Rockford before joining the Tennessee River. The LR watershed drains an area of

approximately 980 km2.

Historically, human disturbance, including farming and logging related activities,

occurred within the present park boundary (Mast and Turk, 1999). The area became an

extremely profitable logging area during the first half of the 20th century. The region was

heavily clear-cut, primarily by the Little River Lumber Company, and forest fires were

also common. The Little River Railroad passed through many portions of the GSM,

including the length of LR, to enable the transport of lumber. Logging ended in 1939

after ~60% of the area had been logged, after the formation of GSMNP (Madden et al.,

2004). Many of the forests are still in successional stages (Madden et al., 2004).

Since the first half of the 20th century, few large-scale landscape alterations have

occurred in the park area adjacent to LR, but human recreational use is common.

Spanning 2108 km2, GSMNP is the most visited national park in the United States and

receives over 9 million visitors each year. The Townsend “Wye” attracts tourists and

fishermen year-round, and a large number of swimmers, snorkelers, and inner tube

`

35

users during the warmer months (Figure 2-1). Building temporary rock dams, disturbing

rocks, and kayaking are other common activities in the stream. The former logging

railroad along LR was converted into heavily traveled Scenic TN 73 that serves as the

main route between Cades Cove and Gatlinburg, TN. Several concrete/gravel parking

lots and pull-offs providing walking access to LR. The river is still difficult to access near

some pull-offs because of steep boulder covered slopes. Small amounts of treated

wastewater are released into the river within GSMNP from a campground near Elkmont

and an education center located next to Middle Prong Little River (MPLR) (Mast and

Turk, 1999).

Despite the long history of human use on the LR, hellbenders still reside within

the stream. Based on the results of previous studies (Nickerson et al., 2003; Freake,

unpubl. data), hellbender surveys were conducted within an ~3 km section of the river

(Figure 2-2) investigated by Nickerson et al (2002) and in portions of the MPLR to

ensure capture of all three stage classes. Elevation within the study area ranged from

327-407 m. Macroscopic in-stream vegetation was rare. The surrounding upland

habitat was comprised primarily of pine and river cove hardwood forest (Madden et al.,

2004). The area has a temperate climate, with high levels of precipitation. A five-year

precipitation mean at a similar elevation level within the park was measured at 147 cm

(Shanks, 1954), and snowfall at lower elevations within the park is relatively infrequent

in comparison to higher elevations. The average high temperature in nearby

Gatlinburg, TN during the coldest month is 8.89°C, white the low temperature averages

-3.89°C. High temperatures in the summer months average 27-29°C.

`

36

Field Sampling Methods

To locate C. alleganiensis, diurnal skin diving surveys were conducted in LR

during the summer and early fall months of 2008-2010. Skin-diving was chosen as the

survey method due to its success in locating all stage classes of hellbenders (Nickerson

and Krysko, 2003). Surveyors utilized snorkels and wetsuits to promote continuous

surveying. On most occasions, one individual conducted surveys, but occasionally

groups of two to seven people assisted. The amount of time each individual surveyor

spent searching for hellbenders was recorded. Surveyors worked upstream, against the

current, to prevent visibility issues from displaced sand and silt. Rocks were hand

turned towards the surveyor to limit disturbance to the streambed particles and replaced

in their original position and orientation. Hellbenders encountered were captured by

hand and placed in water filled plastic containers for data collection and tagging. Rocks

serving as shelter for hellbenders were marked temporarily by inscribing their surface

with another rock to indicate the exact location of capture.

TL and snout-vent length (SVL) of each hellbender was measured in millimeters

(mm) with the aid of a modified PVC pipe. The mass of each individual was recorded in

grams with an Ohaus CS2000 compact digital scale (Ohaus Corporation, Parsippany,

NJ. USA). Sex was recorded if it could be determined based on the swelling of male

cloacal glands in August and September (Nickerson and Mays, 1973). Hellbenders

were individually marked to ensure future identification. Biomark 9mm and 12.5mm

Passive Integrated Transponder (PIT) tags (Destron-Fearing, South Saint Paul, MN,

USA) were injected in adult and most sub-adult individuals dorsally near the base of the

tail. New Skin liquid bandage (Prestige Brands, Inc., Irvington, NY, USA) was applied

`

37

at injection sites. Unique individual combinations of Visible Implant Elastomer (VIE)

(Northwest Marine Technology, Inc., Shaw Island, WA, USA) were injected in

individuals too small for PIT tag injection. PIT tag needles were disinfected with 70%

ethanol between uses, while VIE needles, if used multiple times, were sanitized with

rubbing alcohol wipes. Diet samples from larval C. alleganiensis were collected using

the Easy Feeder Nipple Tip Syringe (Four Paws Products, Ltd., Hauppauge, NY, USA)

and river water to flush out stomach contents. Contents were preserved in either 70%

ethanol or buffered 10% dilution of concentrated formalin. Individuals were returned to

their exact capture site following data collection, and GPS localities were recorded using

an eTrex Legend or GPSMAP 76CSx (Garmin International, Inc., Olathe, KS, USA).

Microhabitat parameters were measured directly at the point of capture. Water

temperature, pH, conductivity in μS/m, and total dissolved solids (TDS) in ppm were

measured using the Combo pH/EC/TDS/Temperature Tester with Low Range EC and

Watercheck pH and TDS reader (HANNA Instruments, Woonsocket, RI, USA). Flow

was recorded with a Global Water Flow Probe (Global Water Instrumentation, Inc.,

College Station, TX, USA). Water depth and shelter size, defined as the longest length

of the shelter rock, was also recorded. A sample of the streambed particles under each

shelter rock were measured using the Federal Interagency Sedimentation Project

(FISP) US SAH-97 sediment size analyzer, also known as a gravelometer.

Streambed particle composition and the mean particle size (D50), representing

the particle size where 50% of stream particles are equal to or less than the value, of

the riffles within the LR study area was determined following the general protocol of the

Wolman Pebble Count (Wolman, 1954). TDS, water depth, and streambed particles

`

38

were measured within the study area at fifty random localities chosen with the aid of a

random number table. Prey abundance was measured by calculating crayfish relative

frequencies as in Nickerson et al. (2003). During some survey periods in 2009, the

number of rocks turned and the number of crayfish encountered underneath the rocks

was counted to calculate the percentage of rocks harboring crayfish. To determine a

rough index of human recreational use in LR, individuals using inner tubes on the river

were counted during a half hour period on seven occasions during summer 2010.

Data Analysis

All statistical analyses were completed using Microsoft Excel for Mac (2008) and

Program R (version 2.12.2) (R Development Core Team, 2008). Significance levels for

all tests were set at =0.05. Data from Nickerson et al. (2003) as well as unpublished

results from studies by the Knoxville Zoo, University of Tennessee-Knoxville, and Lee

University were combined with results of this study to investigate population structure

and dynamics over a 10-year period (Phil Colclough and Marcy Souza, unpubl. data

used with permission; Michael Freake, unpubl. data used with permission). Search

effort was calculated as the number of person hours required to locate one hellbender.

Mean mass and TL of hellbenders sampled across all years was calculated.

Histograms of annual and combined C. alleganiensis size class distribution in LR were

constructed based on individual TL. All histograms used 25 mm intervals. Recaptured

hellbenders were only represented once in the combined histogram, but only individuals

recaptured within a single year were eliminated from the yearly histograms. To

determine if the size distribution of LR’s hellbenders was statistically different from a

representative sampled population, TL data were compared to data from one of most

`

39

well-studied hellbender streams, North Fork of White River (NFWR), MO (Nickerson

and Mays, 1973). Data from the 1969 NFWR population were used for this comparison

because the population has since experienced substantial declines (Wheeler et al.,

2003; Nickerson and Briggler, 2007), and these data are the best available baseline. To

reduce potential bias from unmarked individuals in LR, data from only the two years with

the largest sample sizes that were not directly impacted by flooding (2006 and 2008)

were used for analysis. Data were tested against the NFWR historical data using two-

sample boot-strap Kolmogorov-Smirnov tests. The ks.boot function, from R Package

“Matching” (Sekhon, 2011), tested whether probability densities for TL data from the two

rivers were the same.

Individual hellbenders were also classified into stage classes using TL. Based

on previous research, individuals <125 mm in TL, both gilled and non-gilled, were

classified as larvae (Bishop, 1941; Nickerson and Mays, 1973). Larvae were also

classified into first (<90 mm TL) and second year (>100 mm TL) age classes for shelter

size analysis based on previous studies and the results of surveys in LR (Smith, 1907;

Bishop, 1941). Individuals between 90-100 mm TL could not be classified to an age

class and were therefore not used in analysis comparing larval age classes. Previous

research suggests that size at sexual maturity differs among sex and locality, but

generally ranges from 300-390 mm TL (Dundee and Dundee, 1965; Taber et al., 1975;

Peterson et al.,1988). While most animals captured during this study period could not

be sexed, one small male of 285 mm TL was determined to be sexually mature because

of a swollen cloaca during late summer. Due to this capture as well as the general lack

`

40

of larger adults in LR, sexual maturity was estimated at 275 mm TL. All individuals

measuring 125-275 mm TL were considered sub-adults.

Microhabitat data, including water quality and shelter parameters were analyzed

using descriptive statistics, linear regression, Analysis of Variance (ANOVA), Kruskal-

Wallis rank sum tests, and t-tests. Unpublished data on shelter size taken by Dr.

Michael Freake was included with permission in all analyses on shelter size. Linear

regressions were used to examine the relationship between shelter size, water depth,

and water temperature at capture site and hellbender TL. These parameters were also

compared among life stages. As water depth and larval shelter size data were not

normally distributed, these parameters were tested using Kruskal-Wallis rank sum tests.

The remaining parameters were evaluated using ANOVA and t-tests. In order to control

family wise error rate at 0.05, Bonferroni’s correction was used for the individual

pairwise test of means. Therefore the results of t-tests comparing means of habitat

variables among stage class groups were only considered significant if p<0.0167.

All streambed particle sizes were classified into categories according the

Wentworth particle scale (Appendix) (Wentworth, 1922). Due to the low presence of

some categories, all particles <4 mm were combined into one category, and large and

very large cobble counts were also pooled before the data was used for statistical

analysis. The presence/absence of streambed particle size at the site of capture was

compared among stage classes using an ordinal logistic regression. Presence/absence

of particle categories at used sites was also compared to presence/absence of particle

categories at random locations using a binary logistic regression model.

`

41

To evaluate the effect of prey abundance on hellbender body condition,

combined data from four studies completed in LR since 2000 were compared to

Nickerson and May’s (1973) historic data from NFWR in 1969, as well as data from

2004-2010 surveys of Hiwassee River (HR), TN (Phil Colclough and Marcy Souza,

unpubl. data; Michael Freake, unpubl. data). These rivers were chosen because of their

differences in prey availability based on crayfish relative frequencies. All TL

measurements were transformed by cubing and then dividing by 10,000 to linearize the

relationship of TL and mass. Linear regressions of transformed TL vs. mass from LR

were compared to data from the two other rivers using analysis of covariance

(ANCOVA).

Diet samples were analyzed using a Bausch and Lomb 0.75-3.0X binocular

microscope, and identified to the lowest possible taxonomic level based on the condition

of the samples. Due to the small sample size, no statistical analysis was conducted.

`

42

Figure 2-1. Human recreational use on Little River, TN in Great Smoky Mountains

National Park. Photo courtesy of Kirsten Hecht-Kardasz.

`

43

Figure 2-2. Study site (Little River, TN; USA). Photo courtesy of Kirsten Hecht-

Kardasz.

`

44

CHAPTER 3 RESULTS

General Results

During my study from 2008-2010, a total of 125 hellbenders, including 55 adults,

29 sub-adults, and 41 larvae, were captured over 394 total survey hours. Three larval-

sized individuals had lost their external gills. Survey effort varied among years, but

averaged 2.88 hours per hellbender capture in the main portion of LR. In the MPLR, a

search effort of 7.73 survey hours per hellbender capture was noted in all surveyed

areas, but hellbenders were only located in one riffle which had a survey effort of 2.91

hours per capture. True effort may be slightly lower than recorded as people using

inner tubes on the river often interrupted surveying efforts. During five total hours of

sampling, 281 tubers were counted in LR. The mean rate was ~1 tuber passing per

minute but varied from 0.43 to ~3 per minute.

Eight hellbenders were recaptured during my study period. Only one recaptured

individual was initially marked during my study. This sub-adult was relocated several

days after its initial capture within 1 m of the original capture site. One individual from

the original study by Nickerson et al. (2002) was relocated in the same riffle and had

grown from a sub-adult (197 mm TL) to an adult with a TL of 360 mm. The remaining

recaptures were all individuals tagged in other studies within LR since 2004.

Population Structure

During all surveys from 2000-2010, there were 533 total hellbender captures

(168 larvae, 159 sub-adults, and 206 adults) including 33 recaptures of 27 individuals.

356 individual hellbenders were tagged. Sex was determined for 38 individuals (23

males; 15 females). Mean TL for hellbenders across all years in LR (n=500) was 218.1

`

45

mm (±130.1). The combined histogram of size class distributions during the decade

revealed an overall stable population structure with a sharp decline from the 50-75 mm

class to the 75-100 mm class (Figure 3-1). Larval sized individuals represented 25% of

the total captured individuals among all years. Size class distribution varied among

years, but larvae were generally abundant (Figure 3-2). Hellbender size class

distributions from LR in 2006 (n=113) and 2008 (n=117) were statistically different from

the 1969 NFWR population (n=478) (Figure 3-3) based on results of Kolmogorov-

Smirnov bootstrap tests (D=0.584, p<0.001; D=0.284, p<0.001).

Microhabitat

All physical parameters were measured at the point of capture. Average pH at

the site of capture was 7.23 (n=95). With a small sample size (n=12), capture site

stream flow ranged from ranged from <0.5 to 1 m/s with a median value of 0.5 m/s.

Although regression analysis suggested a linear relationship between hellbender TL

and water temperature (n=102) was present, water temperature was not a strong

predictor of hellbender TL (R2=0.042; p=0.039) (Figure 3-4). Linear regression analysis

also revealed no relationship between hellbender size (TL) and water depth (n=104)

(R2=0.024; p=0.12) (Figure 3-5), but a weak correlation between hellbender TL and

shelter size (n=217) was noted (R2=0.266; p<0.001) (Figure 3-6). Additional tests

revealed no significant difference in average water depth or temperature among stage

classes.

Although overall shelter size among the stage classes overlapped, average

shelter size differed significantly among stage classes (F(2, 214)=32.82; p<0.001)

(Figure 3-7; Figure 3-8). Using t-tests, mean shelter size of larvae (n=61) was

significantly different from both adults (t = 8.11, df = 159, p-value = <0.001) and sub-

`

46

adults (t=-4.83, df = 115, p-value = <0.001). Sub-adults (n=56) and adults (n=100) also

differed significantly in mean shelter size (t = 2.55, df = 154, p-value = 0.012). There

was no statistical difference between mean shelter size between first and second year

larvae in LR. However, first year larvae utilized some larger shelter sizes, including one

of 1085 mm while the largest shelter size of second year larvae was 610 mm. One

individual of 90 mm TL found beneath a 1286 mm boulder could not conclusively be

categorized as first or second year larvae.

Results of pebble counts conducted within the riffles of the study area showed

the D50 value in the small cobble category (64-90 mm) (Figure 3-9). Streambed particle

classes under shelter rocks of larvae (n=25), sub-adults (n=26), and adults(n=38)

(Figure 3-10) did not differ significantly (Table 3-1). However, when comparing random

samples to locations of capture (Figure 3-11), hellbenders appeared to utilize shelters

underlain at least partially by very coarse gravel more than would be expected by

chance based on the results of the logistic regression model (Table 3-2). This model

also showed a negative association between hellbender use and rock shelters

overlaying fine gravel. Based on correlations between streambed particle size

categories, an additional model was tested combining all particles smaller than 32 mm

into one category, which listed very coarse gravel as the only significant variable (Table

3-3).

Body Condition

Mean mass of all LR hellbenders (n=494) was 115.1g (±142.5), but was

influenced by the large number of larval individuals. Mean mass of adults (n=183) was

266.6 g (±128.3). Although individuals often appeared thin, most hellbenders appeared

to be in good overall health with few injuries. Common abnormalities were minor and

`

47

included missing/extra digits and scars. A few captured individuals presented with toe

malformations. Only a very small number of potentially serious abnormalities such as

ulcers and severe limb injuries were noted (Figure 3-12). Crayfish relative frequencies

in LR were low, ranging from 2-16% (mean=6.8%). At the NFWR site crayfish relative

frequencies were considered high during the study period in 1969 and ranged from 56-

67%, while the average mass of hellbenders (n=463) was 371.3 g (±240.4) (Nickerson

and Mays,1973). Crayfish relative frequencies in HR ranged from 21-28.5% with a

mean of 24.7%, and hellbender mass (n=414) averaged 139.9 g (±123.6) (Phil

Colclough, unpubl. data; Michael Freake, unpubl. data).

Results of linear regression analysis of body condition in the three rivers are

listed in Table 3-4. An ANCOVA comparing linear regression lines of body conditions in

all three rivers (Figure 3-13) was significant (F(2, 1490)=137.8, p<0.001). Individual

pair-wise comparisons of hellbender body condition in the three populations confirmed

that the linear regression slope of LR was significantly different from both NFWR

(F(1,986)=194.6, p<0.001) and HR (F(1, 1029)=16.6, p<0.001). The slope of hellbender

body condition in HR was also significantly different from the slope of hellbender body

condition in NFWR (F(1,965)=92.5, p<0.001). LR had the smallest expected mass per

adjusted total length of the three rivers.

Larval Diet

A total of 23 larval diet samples were collected. Larval diet samples contained

primarily larval staged aquatic insects, but crayfish were also identified (Table 3-5).

Ephemeroptera and Trichoptera were the most common insect orders consumed by

larval C. alleganiensis sampled in LR (Figure 3-14). Crayfish were also regularly

consumed. The only vertebrate diet item noted was a ~40 mm TL Eurycea salamander

`

48

larvae regurgitated by a 50 mm TL hellbender larvae. Plant matter and gravel also

presented in the samples. One sample from a sub-adult (204 mm in TL) was identified

as an Ephemeroptera nymph (family Heptageniidae).

`

49

Table 3-1. Variable estimates and odds ratios from an ordinal logistic regression model based on streambed particle size classes at sites used by larval (n=25), sub-adult (n=26), and adult (n=38) hellbenders (Cryptobranchus alleganiensis) captured in Little River, TN.

Variable Estimate Standard error

Wald statistic (Z) p-value Odds ratio

<4 mm 1.07 1.37 0.78 0.44 2.91 Fine gravel 0.61 1.13 0.54 0.59 1.84 Medium gravel -0.35 0.54 -0.66 0.51 0.70 Coarse gravel -0.31 0.49 -0.64 0.52 0.73 Very coarse gravel 2.14 1.19 1.79 0.07 8.50 Small cobble -0.54 0.43 -1.26 0.21 0.58 Medium cobble -0.40 0.50 -0.80 0.42 0.67 Large/very large

cobble -0.38 0.52 -0.73 0.47 0.68

`

50

Table 3-2. Variable estimates and odds ratios from a binomial logistic regression model based on streambed particle size classes at sites used by hellbenders (Cryptobranchus alleganiensis) (n=89) and random locations (n=50) within Little River, TN.

Variable Estimate Standard error

Wald statistic (Z) p-value Odds ratio

Intercept -0.75 0.75 -1.00 0.32 0.47 <4 mm -1.36 0.81 -1.68 0.09 0.26 Fine gravel -1.85 0.71 -2.62 0.01 0.16 Medium gravel -0.27 0.58 -0.46 0.64 0.76 Coarse gravel 0.93 0.54 1.72 0.09 2.54 Very coarse gravel 1.57 0.64 2.47 0.01 4.83 Small cobble -0.17 0.48 -0.36 0.72 0.84 Medium cobble 0.24 0.57 0.42 0.68 1.27 Large/very large

cobble 0.97 0.68 1.43 0.15 2.65

`

51

Table 3-3. Variable estimates and odds ratios from a binomial logistic regression model based on streambed particle size classes (with particles <32 mm combined into one category) at sites used by hellbenders (Cryptobranchus alleganiensis) (n=89) and random locations (n=50) within Little River, TN.

Variable Estimate Standard error

Wald statistic (Z) p-value Odds ratio

Intercept -1.93 0.70 -2.78 0.006 0.15 <32 mm 0.17 0.49 0.34 0.73 1.18 Very coarse gravel 2.67 0.55 4.83 <0.001 14.43 Small cobble 0.21 0.44 0.49 0.63 1.24 Medium cobble 0.32 0.52 0.62 0.54 1.38 Large/very large

cobble 0.86 0.62 1.39 0.17 2.36

`

52

Table 3-4. Variable estimates and model fit for linear regressions of hellbender (Cryptobranchus alleganiensis) body condition (mass (g) vs. transformed total length (mm)) in three rivers.

Variables Estimate Standard error

t p R2 F(df) p

Little River, TN - - - - 0.91 5392 (1, 525)

<0.001

Intercept 3.63 2.45 1.49 0.14 - - - Transformed total

length (mm) 0.050 0.0007 73.43 <0.001 - - -

Hiwassee River, TN - - - - 0.92 5975 (1, 504)

<0.001

Intercept 12.40 2.44 5.08 <0.001 - - - Transformed total

length (mm) 0.054 0.0007 77.30 <0.001 - - -

North Fork of White River, MO

- - - - 0.91 4419 (1, 461)

<0.001

Intercept 15.55 6.36 2.44 0.02 - - - Transformed total

length (mm) 0.067 0.001 66.47 <0.001 - - -

`

53

Table 3-5. Contents of diet samples taken from larval hellbenders (Cryptobranchus alleganiensis) in Little River, TN

Sample TL(mm) Mass(g) Contents

1 65.0 2 1 Trichoptera; Gravel 2 66.0 2 2 Trichoptera (2 Polycentropodidae; 1 Polycentropus) 3 70.0 3 1 Ephemeroptera (Heptageniidae); 1 Crayfish; Plant

Matter 4 70.0 2 1 Unknown 5 74.0 3 4 Ephemeroptera (1 Heptageniidae; 2 Baetidae);

1 Plecoptera (Perlidae) 6 78.0 3 1 Ephemeroptera (Heptageniidae); 1 Trichoptera

(Polycentropodidae; Polycentropus); 1 Unknown; Plant Matter

7 78.0 3 1 Ephemeroptera; 1 Plecoptera (Leuctridae) 8 77.0 4 2 Ephemeroptera (1 Heptageniidae); 1 Plecoptera 9 66.0 2 1 Unknown Insect; 1 Unknown; Plant Matter

10 75.0 3 1 Ephemeroptera

11 73.0 3 1 Ephemeroptera; 1 Trichoptera(Hydropsychidae); 1

Unknown 12 63.0 2 1 Plecoptera (Perlidae) 13 63.5 2 1 Unknown; Gravel 14 69.0 2 1 Crayfish 15 70.0 2 1 Trichoptera; 1 Diptera 16 60.0 2 1 Ephemeroptera(Heptageniidae); 1 Unknown; Gravel 17 60.0 2 2 Ephemeroptera (Heptageniidae); 1 Crayfish 18 73.0 2 1 Diptera; 1 Crayfish 19 62.0 2 1 Coleoptera (Adult Elmidae), Gravel 20 118.0 9 2 Unknown (1 possible crayfish/1 insect) 21 76.0 3 2 Ephemeroptera 22 40.0 4 1 Trichoptera 23 50.0 3 1 Eurycea larvae

`

54

Figure 3-1. Histogram of size distribution of captured hellbenders (Cryptobranchus alleganiensis) from 2000-2010 in Little

River, TN (n=500).

`

55

Figure 3-2. Yearly size distribution histograms of captured hellbenders (Cryptobranchus

alleganiensis) from 2000-2010 in Little River, TN.

`

56

Figure 3-3. Comparison of hellbender (Cryptobranchus alleganiensis) size class

distributions sampled from Little River, TN in 2006 (n=113) and 2008 (n=117), with the North Fork of the White River, MO in 1969 (n=478)

`

57

Figure 3-4. Scatter plot with linear regression line of water temperature (C) vs. hellbender total length (mm) in Little

River, TN (n=102).

`

58

Figure 3-5. Scatter plot with linear regression line of water depth (mm) vs. hellbender total length (mm) in Little River, TN

(n=104).

`

59

Figure 3-6. Scatter plot with linear regression line of shelter size (mm) vs. hellbender total length (mm) in Little River, TN

(n=217).

`

60

Figure 3-7. Box plots comparing shelter size (mm) among three hellbender stage classes, larvae (n=61), sub-adults

(n=56), and adults (n=100), in Little River, TN .

`

61

Figure 3-8. Bar graph showing mean ± standard error of the mean (SEM) for shelter

size (mm) used by three stage classes of hellbenders, larvae (n=61), sub-adults (n=56), and adults (n=100), in Little River, TN. Bar graphs with different letters above are significantly different (p<0.05).

`

62

Figure 3-9. Results of Wolman pebble count survey in Little River, TN showing streambed particle size distribution in

Little River, TN (D50=small cobble).

`

63

Figure 3-10. Bar graph comparing streambed particle size categories found below shelter rocks among hellbender stage classes, larvae (n=25), sub-adults (n=26), and adults (n=38) in Little River, TN.

`

64

Figure 3-11. Bar graph comparing streambed particle size categories found at sites used by hellbenders (n=89) and random locations (n=50) in Little River, TN.

`

65

Figure 3-12. Examples of abnormalities of Cryptobranchus alleganiensis captured in Little River, TN. Photos courtesy of Kirsten Hecht-Kardasz.

`

66

Figure 3-13. Scatter plot with regression lines comparing body condition of hellbenders (Cryptobranchus alleganiensis) from three rivers (Little River, TN (n=527); Hiwassee River, TN (n=507); North Fork of the White River; MO (n=463) with differing crayfish relative frequencies.

`

67

Figure 3-14. Pie chart of total food items identified from larval hellbender (Cryptobranchus alleganiensis) diet samples (n=23) taken from Little River, TN.

`

68

CHAPTER 4 DISCUSSION

Population Structure

This study sought to examine the influence of geomorphology on the population

structure of Cryptobranchus alleganiensis by examining several factors: microhabitat

use and diet among stage classes, and the comparison of body condition of individuals

in streams with different crayfish relative frequencies. However, an understanding of

the overall population structure, particularly over time, was needed to verify that the LR

population was in fact unique from the majority of studied populations before potential

mechanisms affecting the population could be examined.

Overall, the LR population appears stable over the last decade with regular

recruitment of young individuals and representation of all size classes. Consistent with

the results of Nickerson et al. (2002) larvae represented a significant proportion of the

population both overall, and in individual years. Although more adults were captured

than by Nickerson et al. (2002), the general trend regarding large adults remained with

only a handful of individuals over 450 mm captured. Comparing these results directly to

historical data taken from NFWR in 1969 illustrated the differences in the LR population

from a well-studied C. alleganiensis population structure. Despite a slightly larger

sample size, researchers still captured fewer adults in every size interval in LR than the

NFWR in 1969 (Figure 4-1). The distribution differences between the two rivers are

particularly apparent in hellbenders over 475 mm TL. However, it remains unclear

whether these observations truly represent differences in population structure or are

due to differences in detectability.

`

69

Results from this study suggest that larvae in LR are primarily utilizing cobble

and boulder for shelter. Unlike other rivers where larvae have been located within

gravel beds, larval hellbenders in LR can be sampled easily using standard skin-diving

methods (Nickerson and Krysko, 2003; Nickerson et al., 2003). Hellbender researchers

typically haven’t used additional methods to search habitats, such as gravel beds,

associated with larval hellbenders. A recent study in the Allegheny River drainage of

New York found that despite a decrease in the density of C. alleganiensis at study sites

within the last 20 years, more individuals <20 mm were captured recently than in the

1980s presumably because of methods specifically targeting these size classes (Foster

et al., 2009). It is also unclear how deep larvae may reside within gravel beds in other

localities so many larvae may not be accessible even with methods specifically targeting

their habitat. Therefore larval hellbenders are potentially present in other studied sites,

but may not be adequately represented in the sample due to low detectability rates.

Larger adults may also avoid detection in LR. Due to the density of rocks and the

presence of very large boulders that could not be lifted, many individuals may have

been missed during surveys. In addition, deep pools, which have been known to house

hellbenders, were not surveyed (Green, 1933; Nickerson and Mays, 1973).

Recent studies conducted in the Blue Ridge Province have also produced young

C. alleganiensis (Maxwell, 2009; Burgmeier et al., 2011b; Groves and Williams, 2011;

Freake, unpubl. data). Approximately 21% of hellbenders captured during surveys in

HR within the boundary of the Cherokee National Forest in Tennessee were larval sized

individuals (Freake, unpubl. data). Short surveys in the Pigeon River of North

Carolina’s Blue Ridge region, produced three larvae out of only six individuals captured

`

70

(Maxwell, 2009). Larvae have also been located in northern Georgia and additional

areas in western North Carolina (Burgmeier et al., 2011b; Groves and Williams, 2011).

These Blue Ridge populations also do not appear to be impacted by disease and/or

serious abnormalities (Gonynor et al., 2011; Groves and Williams, 2011; Souza, unpubl.

data) as in other regions (Hiler et al., 2005; Miller and Miller, 2005; Nickerson et al.,

2009).

Due to geology, topography, and history, the Blue Ridge Province, which has the

highest proportion of interior forest habitat in the Southern Appalachian region, remains

80% forested (SAMAB 1996a, 1996b). Relatively large portions of the Blue Ridge,

including the greatest concentration of public lands in the eastern United States, are

now protected due to aesthetics and ecological value (SAMAB 1996a, 1996b) (Figure 4-

2). Therefore, the abundance of larvae seen throughout the Blue Ridge Province may

be partially due to a decrease in factors suspected in hellbender declines such as

siltation, channelization, agriculture, mining, logging, and pollution (Dundee, 1971;

Nickerson and Mays, 1973; Bury et al., 1980). Recent studies by Groves and Williams

(2011) noted a negative correlation between human development and hellbender

densities, but the finding was not statistically significant. Many C. alleganiensis

populations in eastern West Virginia’s Appalachian Plateau and Valley and Ridge

regions appear to be declining, with the exception of some located within the protected

Monongahela National Forest (Keitzer, 2007). This supports the hypothesis that human

disturbance, rather than geology alone, may have a major influence on hellbender

populations.

`

71

Although size classes were relatively well represented in LR across all years,

some were absent or low in abundance during individual years. Water regimes can

influence the population structure of stream-dwelling amphibians by affecting mortality

and recruitment (Metter, 1968; Duellman and Trueb, 1986). Flooding events influenced

the long-term population structure of Ascaphus truei in unprotected streams in Idaho

and Oregon by almost completely eliminating tadpoles and thus reducing recruitment in

certain years (Metter, 1968). Flooding has also been suspected as a source of mortality

and displacement in hellbenders (Wiggs, 1977; Trauth et al., 1992; Humphries, 2005;

Miller and Miller, 2005; Nickerson et al., 2007), but its influence on population dynamics

remains unclear.

Nickerson et al. (2007) noted that following floods in 2003, no individuals were

captured within MPLR the following year despite previously locating four larvae in only

eight hours of searching. Second year larvae were also absent from the main portion of

the LR in 2004 (Freake, unpubl. data). In 2005, Freake captured no individuals from

125-150 mm, and only a very small number of individuals from 150-200 mm TL.

Additional small-scale flooding events in 2009 correlated with a missing stage class the

following year: small sub-adults from 125-150 mm TL.

Nickerson et al. (2007), which examined the potential impacts of flooding on

hellbenders in MPLR, cited USGS (2001) stream flow readings from station 03497300

beginning in 1997. An examination of peak stream flow data taken at the LR station

prior to 1997 revealed an extreme flooding event in 1994, where peak stream flow was

over 26,000 cfs (Figure 4-3). No data on C. alleganiensis populations in LR are

available prior to 2000 to illuminate the effects of the flood on hellbender population

`

72

structure. However, data from GSMNP’s fisheries division found no young-of-year

brown trout (Salmo trutta) and few young-of-year rainbow trout (Oncorhynchus mykiss)

following the 1994 flooding, suggesting that other taxa may have been affected by the

flooding (Kulp, 2011). It is therefore possible that this extreme flooding event had a

substantial impact on the hellbenders in LR, and potentially contributed to the lack of

large individuals seen in the river today.

As C. alleganiensis growth rates decelerate with age (Taber et al., 1975;

Peterson et al., 1988) and hellbender growth data is not available from LR, it is difficult

to follow cohorts in LR across subsequent years to monitor the long-term effects of low

recruitment during a particular year. However, two under-represented size classes

marked LR’s hellbender population in 2010, correlating with flooding events in 2003 and

2009. Current extremes may be an important influence on hellbender recruitment in LR

that could lead to long-term impacts on the population structure of hellbenders.

Potential reductions in recruitment following flooding events could be related to larval C.

alleganiensis habitat use within LR.

Microhabitat

The examination of C. alleganiensis microhabitat associations in this study

required two main assumptions: 1) Hellbenders closely correlated to the microhabitat at

diurnal capture sites for significant time periods and 2) The microhabitat at capture sites

remained relatively constant through time. To satisfy the first assumption, it is pivotal

that hellbenders regularly inhabit, rather than temporarily utilize, shelters.

Aggressiveness towards intruding conspecifics and the rarity of shared shelters outside

of the breeding season suggests that adult hellbenders may be territorial over shelter

rocks (Smith, 1907; Hillis and Bellis, 1971; Nickerson and May, 1973). Wiggs (1977)

`

73

estimated that at least 93% of captured hellbenders at his site were in their home riffle,

indicating that transitory individuals are rare. Adult C. alleganiensis have relatively

small home ranges and exhibit site fidelity (Hillis and Bellis, 1971; Nickerson and Mays,

1973; Wiggs, 1977; Ball, 1992; Blais, 1996). Radio telemetry studies have commonly

found hellbenders utilizing one or two individual shelters for months (Wiggs, 1977; Ball,

1992; Blais, 1996), and often homing to their original riffles and shelters when displaced

(Hillis and Bellis, 1971; Wiggs, 1977; Blais, 1996), even following months of captivity

(Nickerson, 1980). However, seasonal changes in habitat use have been noted in

some localities (Smith, 1907; Green, 1933; Nickerson, 1978; Ball, 2001).

A radio telemetry study by Ball (2001) found that hellbenders in a North Carolina

stream typically used two boulder shelters throughout the year: one in shallow water

during the spring and summer, and a larger one in a deep pool possibly to overwinter.

Green (1933) noted that C. alleganiensis in West Virginia moved to deep pools during

summer months, presumably due to temperature increases, while hellbenders in a

substantially spring-fed Missouri stream moved from pools to riffles during the same

time period (Nickerson, 1978). Some authors have suggested that hellbenders may

move long distances during breeding periods (Smith, 1907). However, movements of

hellbenders in a North Carolina stream were less frequent and shorter in distance

during the summer and fall months. Due to these seasonal variations, this study only

attempted to assess diurnal microhabitat associations of hellbenders during the studied

seasonal period (i.e. summer and early fall).

While a majority of studies support an extended association of adult hellbenders to

specific seasonal habitats, information regarding movement, activity, and site fidelity of

`

74

immature hellbenders is extremely limited. To my knowledge undisturbed larvae and

small sub-adults in situ have never been observed in the open. It is unclear whether C.

alleganiensis larvae are nocturnal or diurnal in the wild, although Smith (1907) noted

that hatchlings avoided light. While it is also not known whether hellbender larvae in the

wild ever leave shelter to forage, amphibian larvae commonly reduce activity levels in

the presence of predators, including cannibalistic conspecifics (Colley et al., 1989)

which are abundant in LR. In addition macro-invertebrates found in larval hellbender

diets during this study are plentiful beneath rocks in LR, thus low larval hellbender

activity would be expected. Larvae presumably overwinter at male-guarded nest sites,

and are believed to disperse sometime in spring or early summer (Bishop, 1941), prior

to the seasonal timeframe of this study. Larvae and sub-adults were almost entirely

solitary during this study, opening the possibility that young hellbenders also become

territorial soon after dispersing from the nest. While it is not unreasonable to assume

that young hellbenders, like adults, are associated with microhabitats for extended

periods, it cannot be confirmed and therefore the results of this analysis should be

interpreted with caution.

No correlations between hellbender TL or stage class and measured water quality

parameters were noted. Microhabitat parameters were assumed to be relatively

constant through time. Water parameters including pH and TDS showed little temporal

or spatial variation during the survey period, but as LR is fed by surface water, water

depth and water temperate varied considerably due to fluctuations in precipitation.

Therefore, this study cannot conclusively rule out the effects of water depth and water

temperature on hellbender habitat use.

`

75

While hellbender stage classes in LR were captured in sites similar in pH, TDS,

water depth, and temperature, the physical habitat used by larval hellbenders in LR

might increase their susceptibility to flooding. One of the most important factors

affecting stream species is current (Giller and Malmqvist, 1998). Organisms must find

ways to adapt to stream flow, particularly during flooding conditions, due to sheer forces

and energy considerations. While stream organisms typically adapt a suite of

characteristics to help them survive in normal currents, these adaptations are typically

inadequate during periods of intense flow. Many organisms survive spates by seeking

refugia (Giller and Malmqvist, 1998), including the interstitial spaces in the benthic

layers, where larval C. alleganiensis have been located in other localities (Smith, 1912;

Nickerson and Mays, 1973; Nickerson et al., 2003). As this habitat is not available to

larval hellbenders in LR due to sandstone bedrock breaking down into sand and filling

these spaces, larvae are utilizing the space under rocks at the surface of the streambed

which may be less secure during flooding periods. While larvae utilized a wide variety

of shelters in LR, their habitat included much smaller shelter sizes than other stage

classes including medium to very large cobble, and the average shelter size used by

larvae was significantly smaller than sub-adults and adults. Smaller shelters may be

moved easily by increased water current, increasing the risk of the hellbender larvae

underneath being crushed, swept away in the current, or exposed to predators.

The use of refugia is also a common strategy to reduce predation risk. In

cannibalistic species, such as C. alleganiensis, shifts in habitat use among size or stage

classes is a common adaption to reduce mortality of young individuals by intraspecific

predation (Foster et al., 1988). In other localities, hellbender larvae have been

`

76

associated with completely separate habitat from their adult counterparts, such as the

interstitial spaces within the gravel beds (Smith, 1912; Nickerson and Mays, 1973;

Nickerson et al., 2003; Foster et al., 2009). As this typical larval habitat in other rivers is

unavailable in LR, larvae are instead utilizing similar habitat to adults: a wide size range

of rocks at the riverbed surface.

This wide range of shelter sizes used by larvae includes a direct overlap in

shelter size with sub-adults and adults, which may be partially due to some young

individuals dispersing from their site of hatching later than others. Young hellbenders

may remain in nesting sites for prolonged periods, as larval hellbenders have been

observed at guarded nest sites in May and June (Jeff Humphries, unpubl. data).

Second year larvae may be more selective in their choice of shelter size due to

experience with predators. Despite the wide range of shelter sizes utilized by larvae in

LR and the overlap of larval shelter type with older stage classes, there is a shift in

average shelter size used among stage classes. Therefore it appears that an

ontogenetic shift in hellbender habitat use still occurs in LR despite the similar habitat

type among stage classes, which may serve as a form of refugia against intraspecific

predation.

The relationship, although weak, of shelter size and hellbender TL found during

this study is notable because previous studies examining habitat use by hellbenders

have found no association between shelter size and hellbender size (Hillis and Bellis,

1971; Humphries and Pauley, 2005). However, these studies have focused primarily on

adult-sized hellbenders. A correlation was found between hellbender total length and

`

77

shelter size when examining shelter use in a hellbender population with many young

individuals.

Like shelter size, other studies have examined the role of streambed particle size

on the occupancy of C. alleganiensis, but have been unable to compare streambed

particle association among stage classes. Most studies have focused on general

categories of particles rather than the more fine scale categories used in this study.

Previous studies have found a general association between gravel and cobble

substrates and hellbender occupancy (Keitzer, 2007; Maxwell, 2009; Burgmeier et al.,

2011a). These types of streambed particles are known to harbor a number of

salamander species including hellbender larvae (Smith, 1912; Nickerson and Mays,

1973; Tumlinson et al., 1990) and also serve as important macro-invertebrate habitat

(Giller and Malmqvist, 1998; Hwa-Seong and Ward, 2007), which represent the most

utilized food source for hellbenders of all sizes.

Due to the lack of gravel bed habitat in LR, the interstitial spaces among the

gravel, cobble, and boulders beneath the larger shelter rocks may be particularly

important to hellbender larvae for additional protection and food access to smaller food

items. However, most larvae were found directly under shelter rocks rather than

underlying cobble or gravel, and no difference in stream particle sizes below shelter

rocks was noted among the stage classes. This suggests that although the stream

particle sizes associated with shelters may still be important for larval refugia, other

factors might be influencing habitat selection by hellbenders in relation to substrate

beneath shelter sites.

`

78

Comparing streambed particle sizes at sites utilized by hellbenders of all stage

classes to randomly sampled localities revealed a negative association of occupancy

with fine gravel, and a positive association of occupancy with very coarse gravel. It is

unclear if these associations are due to habitat preferences and/or prey availability, or

are simply related to space availability beneath shelter rocks. Smaller streambed

particles could fill in the spaces underneath rocks, embedding them and leaving no area

available for hellbenders to occupy. Stream embeddedness has been negatively

associated with the presence of other species of salamanders (Tumlinson and Cline,

2003). Conversely, boulders or large cobble may leave too much space available

beneath shelter rocks, leaving hellbenders with reduced protection from stream flow,

predators, and con-specifics. The association of shelters used by hellbenders and

medium sized particles, like very coarse gravel, may represent a balance of space

availability and protection as well as food availability.

Body Condition

The comparison of linear regression lines of hellbender body condition in three

rivers with different crayfish relative frequencies (NFWR, LR, and HR) suggests that the

relative frequency of crayfish in the rivers correlates with overall body condition of C.

alleganiensis. Crayfish relative frequency values appear to corroborate the

observations of Bouchard correlating crayfish abundance with streambed rock

composition. The streambed of NFWR, the only stream largely influenced by carbonate

rock, is mostly a mixture of chert, dolomite, and sandstone (Nickerson et al., 2003).

Hellbenders in the NFWR, which had high crayfish relative frequencies, had a higher

expected mass at a given total length than LR, which possessed comparatively low

crayfish relative frequencies. HR, which had the middle crayfish relative frequency

`

79

measure of the three rivers, also had the middle expected hellbender mass at a given

total length. The HR study site lies within the Sandsuck Formation of the Walden Creek

Group and is comprised primarily of shale, sandstone, and quartz-pebble conglomerate.

A recent study on the nutritional content of crayfish from MO revealed that

crayfish are high in proteins and fatty acids, but extremely low in fat content (Dierenfeld

et al., 2009). Despite the low metabolism of hellbenders, the low fat content of crayfish

could lead to an overall reduction in body condition at sites with significantly fewer

crayfish. Recent studies in NFWR revealed an overall decrease in crayfish relative

frequencies since 1969 (Nickerson et al., 2009), but no significant reduction in overall

hellbender body condition during the last 20 years (Wheeler et al., 2003). However,

crayfish relative frequencies are still relatively large in comparison to both LR and HR.

While these results in regard to body condition are noteworthy, it is not known whether a

decrease in body condition has any impact on overall survival in hellbenders.

As hellbenders experience indeterminate growth, crayfish frequencies may also

have an impact on the overall growth rates and maximum size of hellbenders. Growth

rates of brown trout (Salmo trutta) in Pennsylvania streams correlated with specific

conductance, which the authors related to the geological characteristics of the studied

streams (McFadden and Cooper, 1962). Fish in rivers associated with limestone

exhibited higher growth rates than those in sandstone and shale areas. During this

study period fewer large adult hellbenders were present in LR than are typically found in

other populations with carbonate rocks like the NFWR. King (1939) reported a large

female, 635 mm TL, taken upstream from the current study sites in LR. Although there

are no additional data to suggest larger adults were at one time more prevalent, this

`

80

specimen suggests that hellbenders in LR were at least capable of attaining larger sizes

historically. No historical data on crayfish in LR is available to examine if prey

populations have changed over time, which could have impacted long-term population

structure trends. It remains unclear if large adults are absent because hellbenders do

not grow as large in LR due to low prey populations, experience increased mortality due

to reduced prey availability, or are missing due to other factors such as flooding events

and recruitment.

Larval Diet

While ontogenetic habitat shifts are common in cannibalistic species, ontogenetic

shifts in diet are also common in species ingesting entire food items or those that

undergo some form of metamorphosis, such as C. alleganiensis (Giller and Malmqvist,

1998). Knowledge of diet can help elucidate aspects of the general ecology of a

species as well as assist in conservation efforts. While a number of diet studies have

been completed on C. alleganiensis, these samples have come almost entirely from

adults (Smith, 1907; Netting, 1929; Green, 1933; Green, 1935; Bishop, 1941; Nickerson

and Mays, 1973; Nickerson et al., 1983, Peterson et al.,1989). Only two previous diet

samples from larvae have been published (Smith, 1907; Pitt and Nickerson, 2006).

Larval hellbenders are quite small compared to mature individuals, and therefore a shift

in diet due to gape limitation would be expected. Consistent with the findings of Pitt and

Nickerson (2006), most diet items collected from larvae in LR were comprised of aquatic

insects. In general, the items found within larval hellbender stomach samples were taxa

associated with the habitat they were found to be utilizing during the study period.

Immature individuals of the order Ephemeroptera, the mayflies, were the most

common food item identified. The family most commonly identified in the diet samples,

`

81

Heptageniidae, is generally associated with fast moving water. Many species in this

family are classified as clingers and are found attached to the underside of submerged

rocks and logs in riffles where many species scrape algae for consumption (Merritt and

Cummins, 1996). Representatives of this family were regularly encountered while

turning rocks during surveys in LR, and appear to be widely available to larval

hellbenders within the safety of their shelter sites. One larva also had eaten two

members of the family Baetidae, also commonly found on or under rocks and stones in

shallow riffle areas (Merritt and Cummins, 1996).

The next most commonly identified order, the caddisflies (Trichoptera), primarily

represented one family: Polycentropodidae. Individuals from this family are generally

net builders and construct net-like structures substrates in moving water to provide

refuge from flow and predators, as well as capture food particles caught in the stream

drift (Merritt and Cummins, 1996). These “nets” were regularly observed on and near

rocks in riffle and run areas within LR during the study period.

Parts from small juvenile crayfish were the third most common item found in diet

samples. Based on observation, these sized crayfish were not typically found in the

open areas of riffles and runs of LR, but rather along the stream margins or areas

protected from stream flow. This trend has been documented in other localities (Creed,

1994), and crayfish are known to exhibit habitat differences dependent on size

(Flinders, 2007). However, hellbender larvae were observed throughout the width of

LR, and were not limited to stream margins. Data regarding the location of hellbender

capture in reference to the stream banks were not taken during this survey, so it is

`

82

unclear if larvae that had consumed crayfish were utilizing habitat near shore or if larvae

encountered this prey item due to some other reason.

Like the mayfly and caddisfly families identified, the remaining identifiable items

in the larval hellbender diet samples are regular inhabitants of benthic substrates in

flowing water. The family Perlidae represented the order Plecoptera in the diet

samples. Members of this family are predators commonly found under rocks in streams

(Merritt and Cummins, 1996). One representative from the order Coleoptera was found.

This adult beetle was identified as a riffle beetle (family Elmidae) that, true to its name,

crawls along the gravel and rocky substrates of stream riffles (Merritt and Cummins,

1996).

Based on observation, larval salamanders of other species, particularly

Desmognathus and Eurycea appeared to occasionally utilize habitat similar to that of

larval hellbenders, as they were encountered with some regularity during survey efforts.

Therefore it was not entirely surprising to report a larval Eurycea in a larval hellbender

diet sample. The size of the vertebrate item in comparison to its predator, however,

was surprising. The consumption of a 40 mm item by a 50 mm larval C. alleganiensis

suggests that hellbender larvae are able to consume a wide size variety of prey items.

Previous reports indicated that a second year hellbender larvae had eaten a smaller

conspecific (Smith, 1907). While most prey items identified during this study were

small, larval hellbenders appear to be efficient predators able to consume relatively

large prey.

An examination of the ontogeny of dentition in C. alleganiensis by Greven and

Clemen (2009) found that the transformation of teeth in this species took place

`

83

particularly early, especially in reference to the loss of gills. Unlike a 29 mm larva, all

teeth present in a 47 mm TL larva were entirely pedicellate and bicuspid, characteristics

associated with maturity in salamanders. This early transformation is unique among

caudates, and teeth transformation in C. alleganiensis appears to occur around the time

of hind limb maturation rather than gill transformation. Like adults, larvae also lacked

vomerine tooth patches, which may be advantageous for swallowing whole find items

quickly. This suggested that even young larvae have adapted similar feeding

mechanisms as adults, and can thus capture and swallow large prey items with relative

ease.

The types of organisms found within the stomach samples of larval hellbenders

captured in LR shed light on the foraging behavior and general ecology of larvae within

LR. Larvae appeared to be mostly consuming items in or near their shelter sites, and

not actively foraging in the open stream or in the interstitial spaces of the streambed or

streambed particles. For a small organism in a stream with a variety of predators such

as larger hellbenders, a wide variety of fish, aquatic insects, snakes, otters, and birds,

actively foraging would presumably be dangerous. Like adults, variation in the diet of

larval hellbender, dependent on locality and food availability probably occurs, but

aquatic insects appeared to be regular food items of larvae in general.

While previous studies found that adult C. alleganiensis do occasionally consume

aquatic insects (Green, 1935; Peterson, 1989), there is no indication that they make up

a large proportion of their diet. Therefore an ontogenetic shift in diet appears likely

among hellbenders, at least in LR. This is a significant finding because conservation

efforts surrounding the hellbender typically focus primarily on crayfish populations as

`

84

prey. However, aquatic insects, and their habitat, may be important for the survival and

recruitment of young hellbenders, and should be considered in management and

conservation actions.

`

85

Figure 4-1. Grouped histogram showing differences in size class distributions of hellbenders (Cryptobranchus

alleganiensis) captured in Little River, TN from 2000-2010 (n=500) and the North Fork of the White River, MO in 1969 (n=478).

`

86

Figure 4-2. Map of the eastern United States showing protected areas in the southern

Appalachian and Ozark regions (Modified from Fenneman and Johnson, 1946; U.S. Geological Survey, 2011)

`

87

Figure 4-3. Annual peak streamflow at Little River, TN USGS station within Great

Smoky Mountains National Park (Courtesy of U.S. Geological Survey, 2001).

`

88

CHAPTER 5 CONCLUSIONS

This research aimed to investigate the impacts of stream geomorphology on the

population structure and ecology of hellbenders within LR. The geologic history of the

GSM created a riverbed composed primarily of metamorphosed sandstone, which

appears to be an important influence in the ecology of C. alleganiensis found within LR.

The geomorphology of this stream potentially affected detectability, prey availability,

habitat use, and the overall population structure of hellbenders.

The size structure of hellbenders in LR was unique compared to most studied

localities in its large proportion of young individuals, particularly larvae, as well as the

relatively small number of large adults. The increase in larvae may have been partially

due to higher detectability of hellbenders within LR caused by differences in bedrock

geology and habitat use from other sites, allowing larvae to be more accurately

sampled. Recent efforts to specifically target potential larval habitat in other rivers have

increased captures of young individuals in some sites, but not near the proportion seen

in LR. Adult C. alleganiensis populations, however, have experienced documented or

suspected declines in many other streams (Trauth et al., 1992; Wheeler et al., 2003;

Briggler et al., 2007; Foster et al., 2009; Nickerson et al., 2009; Burgmeier, 2011b),

which could lead to an overall reduction in larval recruitment simply due to fewer

reproducing individuals. Although historic data is not available for LR, the current adult

hellbender population appeared stable and healthy. LR began in the confines of

GSMNP, and therefore human related issues that could potentially affect reproduction

and recruitment in other localities, such as pollution, endocrine disruptors, and habitat

alterations, were not evident there.

`

89

As originally hypothesized by Nickerson et al. (2003), the metamorphosed

sandstone that comprised the bedrock of LR appeared to be a particularly important

factor affecting the habitat use of larval hellbenders. As the bedrock eroded, it

eventually broke down into small sand particles which filled in the space within gravel

beds, making that habitat, which had been associated with larvae in other localities,

unavailable to larvae. Bedrock erosion also left abundant rounded cobble and boulders

on the stream bottom, which larval hellbenders utilized as shelter in LR. This habitat

was very similar to the overall shelter type (boulder), streambed particle size, and water

depth utilized by both adult and sub-adult hellbenders within LR. However, a wide

availability of different sized shelter rocks allowed larval hellbenders to access smaller

shelters on average than older hellbenders, potentially reducing intra-specific

competition and cannibalism. As the habitat type utilized by larvae in LR appeared less

secure than that used by larvae in other localities, mortality may be higher in young

hellbenders due to a potential increase in overall predation risk and susceptibility to

flooding, leading to possible long-term alterations in population structure.

The bedrock geology of LR may have impacted the hellbender population

structure by influencing crayfish populations, the main prey item of adult C.

alleganiensis. Previous observation by Bouchard (unpubl. data) suggested that rivers

with non-carbonate bedrock typically exhibited lower crayfish densities. Crayfish

relative frequencies in LR were lower in comparison to the HR and NFWR. Hellbenders

in LR also showed a significantly lower linear regression slope when comparing trends

in body condition between the three rivers, suggesting that hellbenders in LR are

expected to have the lowest mass at a given total length of the three sites. NFWR,

`

90

which is the only stream to have a carbonate rock influence, had the highest crayfish

relative frequencies as well as the highest overall hellbender body condition. These

trends were less clear in small individuals, which, based on the results of this study,

primarily fed on aquatic insect larvae rather than crayfish in LR. It is not clear whether

reduction in body condition was detrimental to hellbenders or led to increased mortality

in larger adults. Hellbenders might have also experienced decreased growth at larger

sizes due to low prey availability, explaining the smaller number of large adults.

This study suggested that the geologic history and geomorphology of the

bedrock structures of C. alleganiensis rivers had an important influence on the general

ecology and population structure of this species. Therefore, it is important to consider

the impacts of streambed geomorphology on stream ecology when considering

conservation and management of C. alleganiensis. These findings also contribute

towards our knowledge of the general ecology of larval hellbenders in the wild, a topic

that has been largely overlooked for the past 100 years.

`

91

APPENDIX WENTWORTH PARTICLE SIZE CATEGORIES

Category Particle size (mm)

Sand <2 Very fine gravel 2-4 Fine gravel 4-8 Medium gravel 8-16 Coarse gravel 16-32 Very coarse gravel 32-64 Small cobble 64-90 Medium cobble 90-128 Large cobble 128-180 Very large cobble 180-256 Small boulder 256-512 Medium boulder 512-1024 Large boulder 1024-2048 Very large boulder 2048-4096

`

92

LIST OF REFERENCES

ALEXANDER, M. M. 1958. The place of aging in wildlife management. American

Scientist 46(2):123-137.

ALFORD, R. A., AND S. J. RICHARDS. 1999. Global amphibian declines: A problem in

applied ecology. Annual Review of Ecology and Systematics 30(1999):133-165. ANONYMOUS. 2011. Endangered and threatened wildlife and plants; Endangered

status for the Ozark hellbender salamander. Federal Register 76(194):61956-61978.

BALL, B. S. 2001. Habitat use and movements of eastern hellbenders,Cryptobranchus alleganiensis alleganiensis: A radiotelemetric study. M.S. Thesis, Appalachian State University, Boone, NC.

BISHOP, S. C. 1941. The Salamanders of New York. New York State Museum Bulletin,

The University of the State of New York, Albany, NY.

BLAIS, D. P. 1996. Movement, home range, and other aspects of the biology of the

eastern hellbender (Cryptobranchus alleganiensis alleganiensis): A

radiotelemetric study. M.S. Thesis, State University of New York at Binghamton, Binghamton, NY.

BRIGGLER, J., J. UTRUP, C. DAVIDSON, J. HUMPHRIES, J. GROVES, T. JOHNSON, J. ETTLING, M. WANNER, K. TRAYLOR-HOLZER, D. REED, V. LINDGREN, AND O. BYERS (eds.). 2007. Hellbender Population and Habitat

Viability Assessment: Final Report. IUCN/SSC Conservation Breeding Specialist Group, Apple Valley, MN.

BRODIE, E. D. JR. 1971. Two more toxic salamanders: Ambystoma maculatum and Cryptobranchus alleganiensis. Herpetological Review 3(1):8.

BURGMEIER, N., T. M. SUTTON, AND R. N. WILLIAMS. 2011a. Spatial ecology of the eastern hellbender (Cryptobranchus alleganiensis alleganiensis) in Indiana. Herpetologica 67(2):135-145.

BURGMEIER, N., S. D. UNGER, T. M. SUTTON, AND R. N. WILLIAMS. 2011b.

Population status of the eastern hellbender (Cryptobranchus alleganiensis

alleganiensis) in Indiana. Journal of Herpetology 45(2):195-201. BURY, R. B., C. K. DODD JR., AND G. M. FELLERS. 1980. Conservation of the

Amphibia of the United States: a Review. Resource Publication 134, U.S. Fish and Wildlife Service, Washington, D.C..

`

93

COLLEY, S. W., W. H. KEEN, AND R. W. REED. 1989. Effects of adult presence on behavior and microhabitat use of juveniles of a Desmognathine salamander.

Copeia 1989(1):1-7. CREED, R. C. JR. 1994. Direct and indirect effects of crayfish grazing in a stream

community. Ecology 75(7):2091-2103. CROWDER, L. B., D. T. CROUSE, S. S. HEPPELL, AND T. H. MARTIN. 1994.

Predicting the impact of turtle excluder devices on loggerhead sea turtle populations. Ecological Applications 4(3):437-45.

DIERENFELD, E. S., K. J. MCGRAW, K. FRITSCHE, J. T. BRIGGLER, AND J. ETTLING. 2009. Nutrient composition of whole crayfish (Orconectes and Procambarus species) consumed by hellbenders (Cryptobranchus alleganiensis).

Herpetological Review 40(3):324-330. DOBSON, F. S., AND M. K. OLI. 2001. The demographic basis of population regulation

in Columbian ground squirrels. The American Naturalist 158(3):236-247. DOWNING, R. L. 1980. Vital Statistics of Animal Populations. In S. D. Schemnitz (ed.),

Wildlife Management Techniques Manual. pp. 247-266. The Wildlife Society, Inc., Bethesda, Maryland.

DUELLMAN, W. E., AND L. TRUEB. 1986. Biology of Amphibians. McGraw–Hill Book Company, New York, NY.

DUNDEE, H. A. 1971. Cryptobranchus alleganiensis. Catalogue of American Amphibians and Reptiles, SSAR:101.1-101.4.

DUNDEE, H. A., AND D. S. DUNDEE. 1965. Observations on the systematics and ecology of Cryptobranchus from the Ozark Plateaus of Missouri and Arkansas. Copeia 1965(3):369-370.

FENNEMAN, N. M. 1917. Physiographic subdivision of the United States. Proceedings

of the National Academy of Science 3(1):17-22.

FENNEMAN, N. M., AND D. W. JOHNSON. 1946. Physical divisions of the United

States: U.S. Geological Survey: [cited July 2011]. Available from:

http://water.usgs.gov/GIS/metadata/usgswrd/. FITCH, F. W. 1947. A record Cryptobranchus alleganiensis. Copeia 1947(3):210.

FLINDERS, C. A., AND D. D. MAGOULICK. 2007. Habitat use and selection within

Ozark lotic crayfish assemblages: spatial and temporal variation. Journal of

Crustacean Biology 27(2):242-254.

`

94

FOSTER, R. L., A. M. MCMILLAN, AND K. J. ROBLEE. 2009. Population status of hellbender salamanders (Cryptobranchus alleganiensis) in the Alleghany River

Drainage of New York State. Journal of Herpetology 43(4):579-588. FOSTER, S. A., V. B. GARCIA, AND M. Y. TOWN. 1988. Cannibalism as the cause of

an ontogenetic shift in habitat use by fry of the threespine stickleback. Oecologia 74(4):577-585.

GAO, K., AND N. H. SHUBIN. 2003. Earliest known crown-group salamanders. Nature 422 (6930):424–428.

GILLER, P. S., AND B. MALMQVIST. 1998. The Biology of Streams and Rivers. Oxford University Press, Inc., New York, NY.

GILLESPIE, G. 2010. Population age structure of the spotted tree frog (Litoria spenceri): insights into population declines. Wildlife Research 37:19-26.

GONYNOR, J. L., M. J. YABSLEY, AND J. J. JENSEN. 2011. A preliminary survey of Batrachochytrium dendrobatidis exposure in hellbenders from a stream in Georgia, USA. Herpetological Review 42(1):58-59.

GREEN, N. B. 1933. Cryptobranchus alleganiensis in West Virginia. Proceedings of the

West Virginia Academy of Sciences 7:28-30.

GREEN, N. B. 1935. Further notes on the food habits of the water dog, Cryptobranchus

alleganiensis Daudin. Proceedings of the West Virginia Academy of Sciences

9:36.

GREVEN, H., AND G. CLEMEN. 2009. Early tooth transformation in the paedomorphic

Hellbender Cryptobranchus alleganiensis (Daudin, 1803)(Amphibia: Urodela). Vertebrate Zoology 29(1):71-79.

GROVES, J. D., AND L. A. WILLIAMS. 2011. In search of the hellbender in North Carolina. CONNECT Magazine, March 2011, American Zoological Association; [cited August 2011]. Available from:

http://www.aza.org/Membership/detail.aspx?id=17494. GUIMOND, R. W. 1970. Aerial and aquatic respiration in four species of paedomorphic

salamanders: Amphiuma means, Cryptobranchus alleganiensis alleganiensis, Necturus maculosus maculosus, and Siren lacertina. PhD Dissertation, University of Rhode Island, Kingston, RI.

HAMMERSON, G., AND C. PHILLIPS. 2004. Cryptobranchus alleganiensis. IUCN 2011.

IUCN Red List of Threatened Species. Version 2011.1; [Cited August 2011].

Available from: http://www.iucnredlist.org.

`

95

HILER, W. R., B. A. WHEELER, AND S. E. TRAUTH. 2005. Abnormalities in the Ozark hellbender (Cryptobranchus alleganiensis bishopi) in Arkansas: A comparison

between two rivers with a historical perspective. Journal of Arkansas Academy of Science 59:88-94.

HILLIS, R. E., AND E. D BELLIS. 1971. Some aspects of the ecology of the hellbender, Cryptobranchus alleganiensis alleganiensis, in a Pennsylvania stream. Journal of Herpetology 5(3-4):121-126.

HUGHES, G. M. 1967. Evolution between air and water. In A. V. S. de Reuch and Ruth

Porter (eds.), Ciba Foundation Symposium Development of the Lung. pp. 64-84.

Little, Brown, and Co., Boston, MA. HUMPHRIES, W. J. 2005. Cryptobranchus alleganiensis (Hellbender). Displacement by

a flood. Herpetological Review 36(4):428.

HUMPHRIES, W. J, AND T. K. PAULEY. 2005. Life history of the Hellbender,

Cryptobranchus alleganiensis, in a West Virginia stream. American Midland Naturalist 154:125-142.

HUMPHRIES, W. J., M. SOLIS, C. CARDWELL AND A. SALVETER. 2005. Cryptobranchus alleganiensis (Hellbender). Cannibalism. Herpetological Review 36(4):428.

HURYN, A. D., AND J. B. WALLACE. 1987. Local geomorphology as a determinant of

macrofaunal production in a mountain stream. Ecology 68(6):1932-1942.

HWA-SEONG, J., AND G. M. WARD. 2007. Life history and secondary production of

Glossosoma nigrior Banks (Trichoptera: Glossosomatidae) in two Alabama

streams with different geology. Hydrobiologia 575:245-258. KEITZER, S. C. 2007. Habitat preferences of the eastern hellbender in West Virginia.

M.S. Thesis, Marshall University, Huntington, WV. KING, W. 1939. A survey of the herpetology of the Great Smoky Mountains National

Park. American Midland Naturalist 21(3):531-582. KULP, M. A. 2011. Unpublished public presentation. National Park Service, Great

Smoky Mountains National Park, Gatlinburg, TN. LIPS, K. R. 2011. Museum collections: Mining the past to manage the future.

Proceedings of the National Academy of Sciences 108(9):9323-9324.

`

96

MADDEN, M., R. WELCH, T. R. JORDAN, AND P. JACKSON. 2004. Digital Vegetation Maps for the Great Smoky Mountains National Park. Final Report to U.S.

Department of Interior. National Park Service, Great Smoky Mountains National Park, Gatlinburg, TN, Cooperative Agreement No. 1443-CA-5460-98-019.

MAST, M. A., AND J. T. TURK. 1999. Environmental characteristics and water quality of Hydrologic Benchmark Network stations in the Eastern United States, 1963-95: U.S. Geological Survey Circular 1173-A.

MAXWELL, N. J. 2009. Baseline survey and habitat analysis of aquatic salamanders in

the Pigeon River, North Carolina. M.S. Thesis. University of Tennessee,

Knoxville, TN. MCFADDEN, J. T., AND E. L. COOPER. 1962. An ecological comparison of six

populations of brown trout (Salmo trutta). Transactions of the American Fisheries Society 91(1):53-62.

MCGRATH, K. E., E. T. H. M. PETERS, J. A. J. BEIJER, AND M. SCHEFFER. 2007. Habitat-mediated cannibalism and microhabitat restriction in the stream invertebrate Gammarus pulex. Hydrobiologia 589:155-164.

MERRIT, R. W., AND K. W. CUMMINS (eds.). 1996. An Introduction to the Aquatic

Insects of North America. 3rd Edition. Kendall/Hunt Publishing Company,

Dubuque, Iowa. METTER, E. 1968. The influence of floods on population structure of Ascaphus truei

Stejneger. Journal of Herpetology 1(1):105-106. MILLER, B. T., AND J. L. MILLER. 2005. Prevalence of physical abnormalities in eastern

hellbender (Cryptobranchus alleganiensis alleganiensis) populations of middle Tennessee. Southeastern Naturalist 4(3):513-520.

MURDOCH, W. E. 1994. Population regulation in theory and practice. Ecology 75(2):271-287.

NAYLOR, B. G. 1981. Cryptobranchid salamanders from the Paleocene and Miocene of Saskatchewan. Copeia 1981(1):76-86.

NETTING, M. G. 1929. The food of the hellbender, Cryptobranchus alleganiensis (Daudin). Copeia 1929(170):23-24.

NICKERSON, M. A. 1978. Maintaining hellbender salamanders in captivity: The evolution of our knowledge. Proceedings of the American Association of Zoological Parks and Aquariums (1977-1978):396-398.

`

97

NICKERSON, M. A. 1980. Return of captive Ozark hellbenders, Cryptobranchus alleganiensis alleganiensis, to site of capture. Copeia 1980(3):536-537.

NICKERSON, M. A. 2003. Asiatic giant salamanders and hellbenders. In M.H. Hutchins

and W.E. Duellman (eds.), Gzrimek’s Animal Life Encyclopedia, 2nd Edition,

Amphibia 6. pp. 343–347. Gale Group Inc., Detroit, MI. NICKERSON, M. A., AND J. T. BRIGGLER. 2007. Harvesting as a factor in population

decline of a long-lived salamander; the Ozark hellbender, Cryptobranchus alleganiensis bishopi Grobman. Applied Herpetology 4:207-216.

NICKERSON, M. A., AND K. L. KRYSKO. 2003. Surveying for hellbender salamanders, Cryptobranchus alleganiensis (Daudin): A review and critique. Applied Herpetology 1:37-44.

NICKERSON, M. A., AND C. E. MAYS. 1973. The Hellbenders: North American Giant

Salamanders. Milwaukee Public Museum, Milwaukee, WI.

NICKERSON, M. A., R. E. ASHTON, AND A. L BRASWELL. 1983. Lampreys in the diet

of the hellbender Cryptobranchus alleganiensis (Daudin) and the Neuse River

waterdog Necturus lewisi (Brimley). Herpetological Review 14:10.

NICKERSON, M. A., K. L. KRYKSO, AND R. D. OWEN. 2002. Ecological status of the

Hellbender (Cryptobranchus alleganiensis) and the Mudpuppy (Necturus maculosus) salamanders in the Great Smoky Mountains National Park. Journal of the North Carolina Academy of Science 118:27-34.

NICKERSON, M. A., K. L. KRYKSO, AND R. D. OWEN. 2003. Habitat differences

affecting age class distributions of the Hellbender salamander, Cryptobranchus

alleganiensis. Southeastern Naturalist 2:619-629. NICKERSON, M. A., A. L. PITT, AND M. D. PRYSBY. 2007. The effects of flooding on

Hellbender salamander, Cryptobranchus alleganiensis Daudin, 1803, populations. Salamandra 43(2):111-117.

NICKERSON, M. A., A. L. PITT, AND J. T. TAVANO. 2009. Decline of the Ozark hellbender (Cryptobranchus alleganiensis bishopi) in the North Fork White River, Ozark County, Missouri: A historical habitat perspective. A final report

submitted to the Saint Louis Zoological Park and the Reptile and Amphibian Conservation Corp. Reptile and Amphibian Conservation Corps: Gainesville, FL.

NIGRELLI, G. K. 1954. Some longevity records of vertebrates. Transactions of the New York Academy of Science 16(6):296-299.

PETERSON, C. L. 1988. Breeding activities of the hellbender in Missouri. Herpetological Review 19:28-29.

`

98

PETERSON, C. L. 1989. Seasonal food habits of Cryptobranchus alleganiensis (Caudata: Cryptobranchidae). The Southwestern Naturalist 34(3):438-441.

PETERSON, C. L., R. F. WILKINSON, M. S. TOPPING, AND D. E. METTER. 1983. Age

and growth of the Ozark hellbender (Cryptobranchus alleganiensis bishopi).

Copeia 1983:225-231. PETERSON, C. L., D. E. METTER, B. T. MILLER, R. F. WILKINSON, AND M. S.

TOPPING. 1988. Demography of the hellbender Cryptobranchus alleganiensis in the Ozarks. American Midland Naturalist 119:291-303.

PETRANKA, J. W. 1998. Salamanders of the United States and Canada. Smithsonian Institution Press, Washington D.C..

PITT, A. L, AND M. A. NICKERSON. 2006. Cryptobranchus alleganiensis (Hellbender salamander). Larval diet. Herpetological Review 37(1):69.

POWERS, M. E., R. J. STOUT, C. E. CUSHING, P. P. HARPER, F. R. HAUER, W. J. MATTHEWS, P. B. MOYLE, B. STATZNER, AND I. R. WAIS DE BADGEN. 1998. Biotic and abiotic controls in river and stream communities. Journal of the North

American Benthological Society 7(4):456-479.

R DEVELOPMENT CORE TEAM. 2008. R: A language and environment for statistical

computing. R Foundation for Statistical Computing, Vienna, Austria. Available from: http://www.R-project.org.

RASHLEIGH, B. 2004. Fish assemblage groups in the Upper Tennessee River Basin. Southeastern Naturalist 3(4):621-636.

RODENHOUSE, N. L., T. W. SHERRY, AND R. T. HOLMES 1997. Site-dependent regulation of population size: A new synthesis. Ecology 78(7):2025-2042.

SAMAB (SOUTHERN APPALACHIAN MAN AND THE BIOSPHERE). 1996a. The Southern Appalachian Assessment Social/Cultural/Economic Technical Report. Report 4 of 5. Atlanta: U.S. Department of Agriculture, Forest Service, Southern

Region. SAMAB (SOUTHERN APPALACHIAN MAN AND THE BIOSPHERE). 1996b. The

Southern Appalachian Assessment Terrestrial Technical Report. Report 5 of 5. Atlanta: U.S. Department of Agriculture, Forest Service, Southern Region.

SEKHON, J. S. 2011. Multivariate and Propensity Score Matching Software with Balance Optimization: The Matching Package for R. Journal of Statistical Software 42(7):1-52

SHANKS, R. E. 1954. Climates of the Great Smoky Mountains. Ecology 35(3):354-361.

`

99

SMITH, B. G. 1907. The life history and habits of Cryptobranchus alleganiensis. Biological Bulletin 13(1):5-39.

SMITH, B. G. 1912. Embryology of Cryptobranchus alleganiensis, including

comparisons with some other vertebrates. Journal of Morphology 23(1):61-157.

SWANACK, T. M., W. E. GRANT, AND M. R. J. FORSTNER. 2009. Projecting

population trends of endangered amphibian species in the face of uncertainty: A

pattern-oriented approach. Ecological Modelling 220(2009):148-159. SWANSON, F. J. 1980. Geomorphology and Ecosystems. In R. H. Waring (ed).,

Forests: Fresh Perspectives from Ecosystem Analysis, pp.159-170. Oregon State University Press, Corvallis, OR.

TABER, W. E., R. F. WILKINSON JR., AND M. S. TOPPING. 1975. Age and growth of hellbenders in the Niangua River, Missouri. Copeia 1975(4):633-639.

TOPPING, M. S., AND C. A. INGERSOL. 1981. Fecundity in the hellbender, Cryptobranchus alleganiensis. Copeia 1981(4):873-876.

TRAUTH, S. E., J. D. WILHIDE, AND P. DANIEL. 1992. Status of the Ozark hellbender, Cryptobranchus bishopi (Urodela: Cryptobranchidae) in the Spring River, Fulton County, Arkansas. Proceedings of the Arkansas Academy of Science 46:83-86.

TUMLISON, R. AND G. R. CLINE. 2003. Association between the Oklahoma

Salamander (Eurycea tynerensis) and Ordovician-Silurian Strata. The

Southwestern Naturalist 48(1):93-95. TUMLISON, R., G. R. CLINE, AND P. ZWANK. 1990. Surface habitat associations of the

Oklahoma salamander (Eurycea tynerensis). Herpetologica 46:169-175. U.S. GEOLOGICAL SURVEY. 2001, National Water Information System Data Available

on the World Wide Web (Water Data for the Nation). USGS 03497300 Little River above Townsend, TN; [cited June 2011]. Available from: http://waterdata.usgs.gov/usa/nwis/uv?03497300.

US GEOLOGICAL SURVEY GAP ANALYSIS PROGRAM (GAP). 2011. Protected

Areas Database of the United States (PADUS), version 1.2; [cited August 2011].

Available from: http://gapanalysis.usgs.gov/PADUS. U.S. GEOLOGICAL SURVEY NATIONAL AMPHIBIAN ATLAS. 2010. Hellbender

(Cryptobranchus alleganiensis). Version Number 2.1. USGS Patuxent Wildlife Research Center, Laurel, Maryland; [cited August 2011]. Available from: http://www.pwrc.usgs.gov:8080/mapserver/naa/.

`

100

VIÉ, J.-C., C. HILTON-TAYLOR, AND STUART, S.N. (eds.). 2009. Wildlife in a Changing World – An Analysis of the 2008 IUCN Red List of Threatened Species. IUCN,

Gland, Switzerland. WENTWORTH, C. K. 1922. A scale of grade and class terms for clastic sediments.

Journal of Geology 30:377-392. WERNER E. E. AND J. E. Gilliam. 1984. The ontogenetic niche and species interactions

in size-structured populations. Annual Review of Ecology and Systematics 15(1984):393-425.

WHEELER, B. A., E. PROSEN, A. MATHIS, AND R. F. WILKINSON. 2003. Population declines of a long-lived salamander: A 20+ year study of hellbenders, Cryptobranchus alleganiensis. Biological Conservation 109(2003):151-156.

WIGGS, R. L. 1977. Movement and homing in the hellbender, Cryptobranchus

alleganiensis, in the Niangua River, Missouri. M.A. Thesis, Southwest Missouri

State, Springfield, MO. WOLMAN, M. G., 1954, A method of sampling coarse river-bed material: Transactions

of the American Geophysical Union (EOS) 35:951-956.

`

101

BIOGRAPHICAL SKETCH

Kirsten Hecht-Kardasz was born and raised in northern Ohio, and grew up

chasing toads and garter snakes in her back yard. She attended Perkins High School in

Sandusky, OH while also competitively showing her Appaloosa horses. In 2004, she

graduated cum laude, with honors from The Ohio State University with a B.S. in

evolution and ecology and a minor in natural resource management. After graduation

she had her first experience with hellbenders as an intern at Oglebay’s Good Zoo in

Wheeling, West Virginia. Following her time as a tech for a Virginia Polytechnic Institute

study focused on terrestrial salamanders in eastern West Virginia, Kirsten moved to

Florida where she worked on gopher tortoise and flatwoods salamander projects for the

Florida Fish and Wildlife Conservation Commission. She was accepted as a student in

the Interdisciplinary Ecology program at the University of Florida in 2008 and returned

her focus to her favorite species, the hellbender. Her research interests include

behavioral ecology, population ecology, and conservation biology of herpetofauna,

particularly salamanders. Kirsten and her husband, Paul, currently reside in

Gainesville, FL along with their 2-year-old son, Dmitry. Following graduation, Kirsten

plans to continue focusing on conservation issues surrounding amphibians.