Environmental influences on reproductive potential, clutch viability and embryonic mortality of the American alligator i...

MISSING IMAGE

Material Information

Title:
Environmental influences on reproductive potential, clutch viability and embryonic mortality of the American alligator in Florida
Physical Description:
iv, 124 leaves : ill. ; 29 cm.
Language:
English
Creator:
Masson, Gregory R., 1952-
Publication Date:

Subjects

Subjects / Keywords:
American alligator -- Reproduction -- Florida   ( lcsh )
American alligator -- Effect of environment on -- Florida   ( lcsh )
American alligator -- Embryos -- Physiology   ( lcsh )
Zoology thesis, Ph. D
Dissertations, Academic -- Zoology -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1995.
Bibliography:
Includes bibliographical references (leaves 113-123).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Gregory R. Masson.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 022436091
oclc - 34351275
System ID:
AA00022461:00001

Table of Contents
    Title Page
        Page i
        Page ii
    Table of Contents
        Page iii
    Abstract
        Page iv
        Page v
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Chapter 2. Review of literature
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
    Chapter 3. Variation in clutch viability of wild American alligators in Florida
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
    Chapter 4. Embryonic mortality in Florida alligators
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
    Chapter 5. Zygote mortality in Florida alligators
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
    Chapter 6. Conclusion
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
    References
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
    Biographical sketch
        Page 124
        Page 125
Full Text












ENVIRONMENTAL INFLUENCES ON REPRODUCTIVE POTENTIAL, CLUTCH
VIABILITY AND EMBRYONIC MORTALITY OF THE AMERICAN ALLIGATOR IN
FLORIDA















By

GREGORY R. MASSON


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

UNIVERSITY OF FLORIDA


1995





























































~1 /












TABLE OF CONTENTS

A BSTRA CT ................................................................................................................. iii

CHAPTERS

1 INTRODU CTION ............................................................................... 1

2 REVIEW OF LITERA TURE ........................................................... 13

Reproductive Cycle ................................................................... 17
Egg Biology .............................................................................. 19
Nest Production .......................................................................... 21
D eterm ination of Viability ....................................................... 23
Em bryonic M ortality ................................................................. 29

3 VARIATION IN CLUTCH VIABILITY OF WILD AMERICAN
ALLIGATORS IN FLORIDA

Introduction .............................................................................. 37
M ethods ..................................................................................... 40
A nalyses ..................................................................................... 49
Results ....................................................................................... 52
D iscussion ................................................................................ 57

4. EMBRYONIC MORTALITY IN FLORIDA ALLIGATORS

Introduction .............................................................................. 74
M ethods ..................................................................................... 77
Results ....................................................................................... 78
D iscussion ............................................................................... 80
5. ZYGOTE MORTALITY IN FLORIDA ALLIGATORS

Introduction .............................................................................. 89
M ethods ..................................................................................... 91
Results ....................................................................................... 92
D iscussion ................................................................................ 93

6 CON CLU SION .................................................................................... 100

REFEREN CES ............................................................................................................. 113

BIOGRAPH ICAL SK ETCH ....................................................................................... 124














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

ENVIRONMENTAL INFLUENCES ON REPRODUCTIVE POTENTIAL, CLUTCH
VIABILITY AND EMBRYONIC MORTALITY OF THE AMERICAN ALLIGATOR
IN FLORIDA

By

Gregory R. Masson

December 1995

Chairman Louis J. Guillette, Jr.
Major Department Zoology

The alligator is a member of an ancient order of reptiles, the Crocodilia. Although

it has been in existence in its present state for a considerable time (approximately one to

three million years), little is known of its reproductive biology. The female reproductive

database for alligators has been generated from populations residing in primarily

homogeneous habitats. Although they occur naturally from the coast of Texas to North

Carolina, few studies have been conducted examining geographic and temporal restraints

on clutch viability. It was recognized that Florida had low viability rates of alligator eggs

among wild populations when compared to South Carolina and Louisiana populations.

Moreover, the infertility rate of eggs from Florida was estimated at 30%.

The purpose of this work was to ascertain what effect environmental factors had

on clutch viability and embryonic mortality in populations of alligators from Florida.

Seven study sites were selected from three drainages over differing habitats. Wild eggs

were collected and transported to a research facility for artificial incubation under

controlled conditions. In general the viability rates of alligators throughout Florida

i.















appear depressed in comparison to previous published studies of other regions. Rates in

Florida were variable within a lake, between lakes and among years. However, the low

viability rate is not attributable to infertility but to zygote mortality. Embryonic mortality

occurs in high percentages from eggs less than 10 days through their 65 day incubation

period. Early embryonic mortality is not due to the nesting environment but to some

intrinsic or extrinsic factors not yet identified. These factors may be stress induced from

social interactions, nutrition, or environmental contaminants.














CHAPTER 1

INTRODUCTION



Reproductive potential is the theoretical limit of an organism's ability to

produce viable offspring that can ultimately attain sexual maturity. The basis for this

concept comes from attempting to understand the evolutionary constraints on the

number and size of offspring produced. The production of progeny capable of

reproducing is the ultimate goal for the individual and the species. Within the

vertebrates, there is a continuum of strategies for reproduction, from a few, large

offspring (e.g., ungulates) to many, relatively small offspring (e.g., salmonids). This

continuum includes the production of altricial as well as precocial young, although the

majority of species produce many offspring which are usually precocial.

Due to the broad nature of this topic, this discussion will be limited to

vertebrates. Reproduction is an energetically expensive necessity, which has been

approached in a myriad of ways by animals. The energetic investment for

reproduction must be allocated by the female from her total energy budget. This total

budget can be simplistically broken down into somatic and reproductive portions.

The somatic portion is used for maintenance and growth, while the reproductive

portion is allocated to each phase of reproduction. Reproductive apportionment may

include mate selection, combat, territorial defense and/or selection, as well as care









2
and feeding of offspring by males or females. However, the female generally invests

a significant amount of energy into egg production, such as in the caecilian,

(Dermophis mexicanus) which gives birth to a clutch weighing as much as 65% of the

post-partum female (Steams, 1992). Female investment in reproduction may occur

just once as in semelparous species in which she maximizes her investment without

the need to maintain reserves for future growth. Iteroparous species have evolved

strategies for balancing a maximum female investment against future reproductive

effort. Teleologically, this means the female will maximize her effort at each

reproductive interval while maintaining sufficient energy reserves for her growth,

maintenance and future reproduction. Each species has evolved a life history strategy

in relation to the reproductive effort of individuals within that species.

The majority of work regarding reproductive potential has been compiled

within avian and mammalian taxa. The idea of the Lack clutch, defined as the

number of eggs which produces the most fledglings (Stearns, 1992), has been

promoted as a standard model for oviparous vertebrates yet is based on data for

iteroparous, altricial birds. Assumptions of the Lack clutch are: no trade-offs occur

between clutch size and size of individuals at sexual maturity; and no trade-offs on

survivorship of offspring and survivorship of parents between generations. Additional

assumptions include no change in frequency, size or survival of the next clutch and

clutch size is not constrained to a fixed number.

A correlation of clutch size to body size has been shown in mammals (Steams,

1992) and reptiles (Vitt and Congdon, 1978; Vitt, 1981; Wilkinson, 1984; Congdon










and Gibbons, 1985; Hall, 1990). However, it is important to realize if a correlation

is not found there can still be a relationship between clutch size and female size. For

example, within long lived species constraints on clutch size for animals of

comparable age and size can be variant. This variance could be attributed in part to

reduced physiological performance toward reproduction. That is, with a finite amount

of energy available for maintenance and reproduction, the female may be restricted by

food availability, territory or population density and have to allocate part of her

reproductive energy budget toward maintenance. Reduction in energy destined for

reproduction could result in production of a smaller clutch than comparable size

females. Therefore, environmental pressures on the female can lead to atresia and a

reduction in clutch size (Mendez de la Cruz et al., 1993). Atresia is as a short term

response by the female to conditions which produce an acute response in clutch size.

Additionally, clutch size may be reduced when environmental conditions vary little

and are not conducive to breeding success (Stearns, 1992). If environmental

conditions are not variable and are restricted to breeding success the impact on clutch

size would be over successive breeding seasons. Environmental conditions important

for breeding success are a limiting factor in the females ability to produce eggs (van

Tienhoven, 1983; Gutzke and Packard, 1984). The ultimate factor controlling clutch

size would be follicular recruitment followed by a proximate factor, follicular atresia

(Mendez de la Cruz et al., 1993).

Reproductive strategies may be separated into two broad categories of "more is

better" and "bigger is better" which can become evolutionary endpoints for a species.










The successful approach to the dichotomy of more is better or bigger is better can be

seen readily within the reptiles. In order to look at these strategies, two groups

containing individuals that are exclusively oviparous can be compared, the families of

Chelonidae and Dermochylidae of the order Testudines (turtles) and the families

Crocodylinae and Alligatorinae of the order Crocodilia. Two representative species

which have females of relatively the same body mass (the Loggerhead sea turtle,

Caretta and the American alligator, Alligator mississippiensis) exhibit markedly

different reproductive strategies. The sea turtles are an extremely fecund group which

produce a great number of eggs (> 100), within each clutch and lay multiple clutches

(up to 5) per season (Witherington, 1986; Hays and Speakman, 1991), whereas

alligators produce a single moderate-sized clutch of eggs (40), per breeding season

(Carbonneau, 1987; Jennings and Abercrombie, 1987). High mortality for sea turtles

comes during incubation and after hatching, as the female oviposits the eggs within

the nest and returns to the ocean; therefore, excluding maternal care of the young

(Horikoshi, 1992). However, alligator neonates are protected by the presence of the

female and the secluded location of the clutch within the marsh after hatching

(Woodward et al., 1987; Brandt, 1989).

The life history curves (type III survival curve; Wilson, 1980) for these

species are similar from hatching to maturity (Figure 1.1). That is, there is a high

mortality rate for the neonates which decreases as the animals grow and become less

vulnerable to predation. Additionally, both species (C. caretta, A. mississippiensis)

are long lived and do not reach maturity for many years (20 years in C. caretta,








Witherington, 1986; and 12-15 years in A. mississippiensis, Woodward et al., 1987).

Because of energetic costs of traversing the surf and beach, digging a nest and

ovipositing eggs, female sea turtles maximize their effort in egg production for each

excursion. An advantage to producing a large clutch, within the loggerhead

population examined by Hays and Speakman (1991), is the lack of a relationship

between clutch size and chance of biotic or abiotic mortality. That is, if small and

large clutches have the same probability of mortality, those animals with large

clutches, potentially would have more offspring reach sexual maturity. It becomes an

evolutionary advantage for the sea turtle to produce large clutches. Laying several

clutches per year is also advantageous in that the sea turtle is maximizing reproductive

effort on each trip up the beach for nesting, yet she is not investing all her effort into

one clutch. With multiple nesting by a female, there is a reduction in the chance one

catastrophic event would eliminate her reproductive effort for that nesting season.

Catastrophic events include flooding of the nest, predation or disturbance of the nest.

Physiological and physical constraints on the female would limit the number of eggs

and number of clutches produced.

Alligators present a different approach, as they produce a moderate size clutch

(40 eggs) for a body size comparable to C. caretta and lay a single clutch per year

(Lance, 1989). Energy expenditure for a female alligator consists of nest

construction, oviposition, nest attendance and liberation of the young from the nest as

well as staying with the neonates during the fall months (Joanen and McNease, 1989).

The reproductive strategy differences are highlighted by the female alligator utilizing

her energy in nest guarding and attendance during the incubatory period. Female

alligators also liberate neonates from the nest and protect them throughout the fall and








winter. Each species has shown that their pattern is successful on an evolutionary

scale.

Life history curves historically have described the reproductive potential of

species from birth or hatching to maturity. However, there are two other periods

which influence calculation of survival curves, embryonic development and

vitellogenesis, that are not normally included in these curves. A reduction in the

potential number of offspring occurs during the vitellogenic phase with follicular

atresia (see Mendez de la Cruz et al., 1993). Embryonic mortality occurs and also

reduces the potential offspring (Putney, 1988). Thus, there are five distinct times

when the reproductive potential can be estimated (Figure 1.2). It is not known

whether varying mortality patterns exists for the follicular or embryonic phases. An

assumption can be made that if a pattern exists within these time frames it could be

similar to a type I, II or III survival curve (Figure 1.3).

Until recently, the reproductive biology of all members of the reptilian order

Crocodilia was poorly studied. Due to their rising economic and environmental

importance (Ashley and David, 1987), they are now receiving increased attention

within the research community (see Webb et al., 1987). Their size and potential

threat to human and domestic animal life have garnered them much notoriety, and

their presence within tropical and temperate wetlands has not only attracted tourism

but also has become an expected sight.

Crocodilians have not been studied as intensely as other reptiles, although all

22 extant species are listed as threatened or endangered by IUCN (the World

Conservation Union). The historical approach to the preservation and conservation of

this group has, until the last 10 years, involved the study of basic ecology and








behavior of wild and captive populations. The data derived from these studies were

taken at face value and used to modify existing policies. These surveys, however,

have not investigated the variations in reproductive potential of populations over a

limited or extensive range of individual species. For example, studies examining the

nesting biology of the American alligator have been conducted in areas that are

primarily monotypic in regard to vegetation, nest site location and available foods

(Joanen and McNease, 1989; Wilkinson, 1983; Schulte and Chabreck, 1990). Due to

the pioneering nature of the early research and the limited field sites, many

assumptions regarding nest biology and reproduction were and continue to be made.

These assumptions have fostered many generalizations which require further

investigation. The major studies of alligators within the United States were carried

out in South Carolina (Wilkinson, 1984) and Louisiana (Joanen and McNease, 1980).

Little variation within or between these areas in regard to the reproductive cycle of

the alligator has been noted. However, both areas possess gross ecological

similarities with nesting occurring along man-made dikes. In South Carolina,

alligator populations are found in old impoundments for rice cultivation and the

accompanying canal systems. Personnel in Louisiana have compiled a massive

database on their populations within the confines of the Rockefeller National Wildlife

Refuge which is primarily short grass marshes with an abundance of nutria

(Mycogaster coypu) as a food source. These areas have provided the largest

percentage of published literature on the reproduction of alligators. They have also

generated the promulgation of many hypotheses that need to be tested.

Alligators are reproductively synchronous, and produce eggs of a size

convenient to work with (72 mm x 42 mm) and in sufficient quantities for statistical








analyses. The eggs may be used as bioindicators of contaminants within the females

and/or the environment. Females produce an egg with yolk, albumen, and a calcium

enriched shell, each component of an egg may be used to evaluate conditions within

the population. For example, analysis of the egg components could provide

explanations for actions of contaminants on embryonic development. Additionally,

knowing numbers and sizes of eggs within nests and numbers of nests within a system

can be an indicator of populational responses to varying environmental conditions,

such as a reduction of female animals within a system could lower population

densities, thus eliciting larger clutch sizes (Stearns, 1992). Also, if larger animals

produce larger and a greater number of eggs, the age structure of the population could

be estimated. Analyses can be made of the egg components to identify and quantify

the transfer or accumulation of contaminants by the female. Furthermore, the nesting

environment is reported to have a direct effect on eggs and their survivorship

(Ferguson, 1982). This could allow eggs to be used in evaluating levels of

contaminants and egg survival used to monitor vegetative changes in wetlands.

In Florida, alligators are found in a wide variety of wetland systems. The

variation in habitat offers an opportunity to examine potential reproductive differences

occurring in different regions of the state. Recognition of the alligator as a species of

economic and biological importance to the state of Florida has created an atmosphere

of mutual interest and cooperation among state and federal regulatory agencies and

university scientists and farmers.

This cooperation made possible a large, long term study of the reproductive

biology, egg viability and embryonic mortality of the American alligator from varying

wetlands in Florida. Several hypotheses were developed based on the data obtained






9

in previous studies from other states. Specifically, the following null hypotheses were

tested:

#1. Different habitats within the state of Florida have no affect on the

viability of alligator eggs.

#2. There is no annual effect on alligator egg viability within these

habitats.

#3. Nesting material used during nest construction does not affect the

viability of alligator eggs during incubation.

#4. Infertility is not the cause of poor alligator egg viability in Florida

ecosystems.

#5. Mortality is not different among nests, nor among years or among

embryonic stages.





















Embryonic
Mortality


Survivorship
Curve


Time


Theoretical survivorship curve for vertebrates


Follicular
Atresia


Figure 1.i1:



































































Neonates.
.. .. . . . ..
............*A


Figure 1.2: Diagram of populational reductions over time












Survival Curve


Figure 1.3:


Survivorship curve for vertebrates described by
E.O. Wilson


hatching


maturity














CHAPTER 2

REVIEW OF THE LITERATURE



The ancient reptilian order, Crocodilia, first appeared in the early to mid

Triassic period approximately 130 million years ago (Romer, 1956; Porter, 1972).

The stem reptiles (Order Cotylosauria) gave rise to the captorhinomorphs (suborder)

which are considered the ancestor to all extant reptiles (Romer, 1956; Porter, 1972).

Reptiles, based on temporal fenestrae, are divided into 5 subclasses: 1) Anapsida; 2)

Synapsida; 3) Synaptosauria; 4) Parapsida; 5) Diapsida (Porter, 1972). Crocodilians

are within the subclass Diapsida, all of which possess two temporal openings

separated by the postorbital and squamosal bones. Within the subclass Diapsida, is

the superorder Archosauria which consist of the crocodilians, birds and extinct

dinosaurs. Traditionally, the Crocodilia have been separated into three suborders,

Protosuchia, Mesosuchia, and Eusuchia (Tarsitano et al., 1989). However, there is

much debate as to the true origin and subsequent lineages of several genera (see

Densmore and Owen, 1989). There are three extant Crocodilian families of the

original five to 14, these being the crocodylinae, gavialinae (tomistominae) and the

alligatorinae (Romer, 1956; Alvarez del Toro, 1974; Joanen and McNease, 1989;

Taplin and Grigg, 1989; Tarsitano et al., 1989). At present, there are 22 extant

species of crocodilians (Tarsitano et al., 1989). Alligatorinae members include the

13










American alligator, caimans of the new world and the Chinese alligator (Alligator

sinensis) from Asia (Densmore and Owen, 1989). Based on fossil evidence, the

American alligator (Alligator mississippiensis) has been in existence since the lower

Miocene, approximately 12 million years (see Taplin and Grigg, 1989).

Currently, alligators are widely distributed within the United States from Texas

to North Carolina. Principally, they are found in marshes and swamps but can be

encountered in rivers, lakes and impoundments. They are a relatively temperate

species of crocodilian and can exist in areas experiencing short periods of freezing

conditions (Coulson and Hernandez, 1983). The factor limiting their temperate

distribution may be metabolic processes required for individual survival and not

necessarily a lack of reproductive effort (Coulson and Hernandez, 1983). Courtship

and mating occur when the water temperature of their habitat is below 70 0 F, the

temperature at which they begin feeding (Wilkinson, unpubl. data). No direct

correlation has been determined between their reproductive onset and ambient

temperature. The only necessity appears to be a nest environment that can maintain

sufficient incubation temperatures (28 34 0 C) (Ferguson, 1982,1985).

Human exploitation of alligators has long been a significant factor in their

local distribution and density (Jennings, 1986). The aesthetic value and durability of

crocodilian leather products makes for a very marketable commodity (Ashley and

David, 1987). Harvesting of alligators both legally and illegally, prior to and during

the first two-thirds of this century, decimated alligator populations to a level where

they were placed on the federal list of endangered species in 1966 (Nichols et al.,










1976). The amendment of the Lacey act in 1970, prohibiting interstate shipment of

alligators, gave further credence to the plight of their existence (Nichols et aI., 1976).

Although today it is recognized that they were probably not in eminent danger from

extinction, these measures were fortuitous in publicizing the dire consequences of

human actions on seemingly indestructible beasts, such as the American alligator.

As early as the late 1700's, William Bartram stated "Alligators feeding in a

river in Florida were in such incredible numbers and so close together from shore to

shore that it would have been easy to have walked across on their heads" (Bartram,

1791). Florida has a sizable population of alligators within the confines of its

wetlands, with estimates of 11/kn (Woodward and Moore, 1990). Alligator

population dynamics are difficult to assess, although night light counts and computer

models are currently used to estimate population densities in Florida (Woodward and

Marion, 1978; Wood et al., 1985; Nichols, 1987; Woodward, 1987; Woodward and

Moore, 1990). Alligator distribution in the state of Florida has, in recent years, been

limited and further constrained by the large influx of people and subsequent loss of

suitable habitat (Woodward and Moore, 1990). Estimates of their present status are

not easily obtained, due to the myriad of habitats they occupy in Florida. However,

yearly surveys are conducted on wetlands to facilitate management practices in

monitoring population fluxes. Over a few short years, the population of alligators

within Florida appears to have rebounded significantly, as no ted by the number of

nuisance alligator reports received by the Florida Game and Fresh Water Fish

Commission (D. David pers. comm.). The number of reports steadily increased from








16
4900 complaints and 1871 harvested in 1975 to 10300 complaints and 4464 harvested

in 1990. The potential threat to humans and their domestic animals is often serious

enough for removal of approximately 50% of the reported nuisance alligators (D.

David pers. comm.). However, the increase in nuisance complaints is either

attributed to a significantly enhanced reproductive success or by the tremendous influx

of people into Florida (33 % increase during this time) and subsequent invasion of

alligator habitat thus making alligators a high profile species.



Reproduction



Reproductive strategies of reptiles have been a source of intense interest in

recent years (see Ballinger, 1978; Callard and Ho, 1980; Licht, 1984; van Tienhoven,

1983; Congdon and Gibbons, 1985 ). The species which appear to gamer most of the

attention are those which exhibit conspicuous nest formation, oviposition and nest

guarding (e.g., sea turtles) or those whose populations are easily manipulated (e.g.,

lizards ; Steams, 1992). In order to evaluate the reproductive potential of reptiles,

studies have concentrated on female condition, size, and endocrine control of clutch

size (Ballinger, 1978; Jones, 1978; Sinervo and Licht, 1991). Clutch size, egg size,

and female size have been investigated in the testudines and crocodilia, yet yield

varying conclusions for the species studied (Caiman, Thorbjarnarson, 1990;

Chrysemys, Congdon and Gibbons, 1985; Alligator, Wilkinson, 1983, Hall, 1990).

Thorbjarnarson (1990), found that there was no relationship between clutch size and










the size of female Caiman, while Wilkinson (1983), reported a correlation (approx.

60%) between clutch size and female size in alligators. Hall (1990), reported a

correlation of greater than 75% between clutch mass and female size although based

on small sample sizes (N < 20). Those data available for Chrysemys (Congdon and

Gibbons, 1985) indicate that larger females do not necessarily lay larger clutches.

Interest in alligators and their reproduction has been largely due to their

mystique as a large wetlands predator. Crocodilians are seasonal breeders, laying

hard, calcium-rich eggs in a mound or cavity nest. Much of our understanding of

crocodilian reproductive biology has been obtained in the last 20 years. Below I will

discuss briefly the reproductive cycle, reproductive anatomy and nesting biology of

the crocodilians studied to date.



Reproductive cycle



Our knowledge of crocodilian reproductive cyclicity has increased significantly

over the last few years (see Webb et al., 1987; Lance, 1989; Kofron, 1990). These

studies have, however, not addressed the need for basic information on variation in

reproduction over geographic areas or variation through time. For example, the

reproductive cycle of the female American alligator has been assumed to begin with

vitellogenesis in March, followed by mating in May or June (Lance, 1987; 1989).

Recent data for females in central Florida indicate that during September, October,

November and December, some have ovaries containing 40-50 vitellogenic follicles










with an average diameter of 12 mm (Guillette, 1992). Additionally, RIA analyses

demonstrate elevated plasma estradiol 17-B (E2) concentrations during the fall months

indicative of ovarian activity during this assumed "non-reproductive" time of the year

(Guillette, Woodward, Masson, unpubl. data). The elevated plasma E2 is biologically

active, as the females exhibit electrophoretically-detectable vitellogenin concentrations

in their plasma during the fall months (Guillette, Masson, Bugarin and Woodward,

unpubl. data). Vitellogenin is a yolk precursor protein synthesized in the liver under

estrogen stimulation (Ho et al., 1978).

Female alligators are reported to retain their eggs for approximately three

weeks after ovulation in order to secrete and lay down the albumen, eggshell fiber,

and eggshell calcium prior to oviposition (Lance, 1987, 1989; Palmer and Guillette,

1992). There has been little, if any, research completed on embryonic activity or

viability and status of the egg during the in utero period (Webb et al., 1987; Joanen

and McNease, 1989; Lance, 1989; Webb and Cooper-Preston, 1989). However, the

reproductive cyclicity for several species (Crocodylus porosus, C. johnsoni, C.

niloticus, Caiman crocodilus) has been described (Webb et al., 1987; Kofron, 1990;

Thorbjarnarson, 1990). It is believed female crocodilian s begin vitellogenesis in the

spring followed by mating (Webb et al., 1987; Kofron, 1990; Thorbjarnarson, 1990).

Approximately three weeks after mating, eggs are oviposited in either a hole or

mound nest for incubation. The incubation period varies with each species as C.

porosus and Gavialis gangeticus may be as much as 90 days (Maskey, 1989).










Male American alligators are reported to begin testicular recrudescence and

mating displays in March and April, completing their reproductive cycle after mating

in May and June (Lance, 1989). However, as in females, recent data suggests that

males may begin testicular recrudescence during the fall months as elevated plasma

testosterone concentrations have been observed (Guillette, Vliet and Masson unpubl.

data; Lance, 1989). Territorial behavior and fighting is prominent during the spring

recrudescence period. Bellowing, head slapping, water dance, and posturing in male

alligators has been well documented for wild and captive populations (Lance, 1989;

Vliet, 1989).



Egg biology



The biology of the crocodilian egg has been studied and comparisons made

among different species in regard to their patterns of embryonic development, oxygen

consumption, and lipid utilization (Ferguson, 1987; Grigg, 1987; Joanen et al., 1987;

Manolis et al., 1987; Webb et al., 1987; Webb and Manolis, 1987; Whitehead, 1987;

Deeming and Ferguson, 1989a, 1989b, 1990a, 1990b; Thompson, 1989). Other

studies have focused on incubation conditions (i.e. temperature, hydration and/or gas

exchange) of artificially incubated eggs (Joanen et al., 1989). Basic egg anatomy or

sex determination in alligators and crocodiles has been studied (see Ferguson, 1985;

Webb et al., 1987). Crocodilians produce a hard-shelled amniote egg which is

variable in size depending upon the species (Webb et al., 1987; Whitehead, 1987).










The Saltwater crocodile (Crocodylus porosus) has the largest egg (113 g) (Webb et

al., 1987; Ferguson, 1985; Maskey, 1989), with alligator eggs averaging 72g in

weight (Masson et al., 1991). Crocodilian eggs are primarily ellipsoidal and most

often do not possess a narrow or a blunt end as found in avian eggs. Generally eggs

within a clutch are generally uniform in shape and size, however, between clutch

variation in egg size can be extensive.

The utilization of lipids within the yolk has been investigated by Noble et al.,

(1990) to try and determine which classes of lipids are present and their use during

embryo development. They reported lower concentrations of low-density lipoproteins

than that found in an avian species, the hen (Gallus domesticus). In addition, they

found a possible glycoprotein of molecular weight 20,000 not found in the hen.

However, the other proteins found in Crocodylus porosus were similar to the hen and

differed only in relative amounts and in water content. Palmer (1990) had similar

findings in analyzing the albumen of Alligator mississippiensis where the major

proteins were similar but occurred in different proportions and again there was an

unknown protein (molecular weight 56,000) found. The proteins (e.g. ovotransferrin)

found in the albumen were antibacterial and antifungal and are undoubtedly one

reason eggs may not become adled when left at ambient temperatures for an extended

period (greater than 1 year, Masson unpubl. data). However, the major protein in

alligator albumen is not related to the protein (ovalbumin) the major protein found in

hen egg albumen (Palmer and Guillette, 1991).









21
The eggshell was found to be of calcium carbonate composition and the inner

shell membrane of polymerized fibers (Whitehead, 1987; Palmer, 1990; Packard and

DeMarco, 1992). Shells of viable crocodilian eggs are known to degradate during

incubation (Ferguson, 1982, 1985; Joanen and McNease, 1989). This degradation of

the eggshell during incubation was originally thought to be a function of the embryo

using calcium from the bacterial breakdown of the shell. Data from studies isolating

developing eggs in a sterile environment are conclusive in showing the degradation is

from respiratory by-products (Guillette, Heaton-Jones and Percival, unpubl. data).

These experiments support the hypothesis that the presence of carbon dioxide and

water yield a carbonic acid reaction for the mobilization of the calcium and

subsequent eggshell breakdown.



Nest production



Nesting biology and maternal defense of the nest has been well documented for

alligators (McIlhenny, 1934; Joanen, 1969; Joanen and McNease, 1970, 1975, 1987,

1989; Chabreck, 1973, 1975, 1978; Garrick et al., 1978; Deitz and Hines, 1980;

Ferguson, 1985; Kushlan and Kushlan, 1980; Wilkinson, 1983; Jennings et al., 1987)

as well as for other crocodilians (see Pooley, 1969; Webb et al., 1987). Crocodilians

exhibit two basic modes of nest production. They are either (1) hole nesters (i.e.,

Gavialis gangeticus, Crocodylus niloticus) or (2) mound builders (i.e. Alligator

mississippiensis, Crocodylus porosus, Crocodylus palustris). Nest construction has










been described for many crocodilian species (see Pooley, 1969; Wilkinson, 1984;

Joanen and McNease, 1989; Maskey, 1989; Vliet, 1989).

Female alligators begin nesting behavior in mid-June to mid-July. The onset

of nest construction for the American alligator is generally considered to be two to ten

days prior to oviposition (Mcllhenny, 1934; Bara, 1972; Chabreck, 1973, 1975;

Goodwin and Marion, 1978; Joanen et al., 1987; Joanen and McNease, 1989, 1989;

Deitz and Hines, 1980; Wilkinson, 1983; Licht, 1984; Lance, 1987, 1989; Jennings

et al., 1987; Woodward et al., 1989). Following hatching, the female remains with

the pod through the fall and possibly for as long as a year. Often cohorts from the

previous year are found near the female and/or the present year's hatchlings. These

observations suggest some females can reproduce in consecutive year's. However,

there are no direct data available from wild populations demonstrating breeding in

consecutive years. Estimates have been made for the proportion of nesting females

within a population (Wilkinson, 1983; Joanen and McNease, 1980; Kushlan and

Kushlan, 1980). Joanen and McNease (1980) estimated 63 % of the adult females

nested yearly in Louisiana whereas Kushlan and Kushlan (1980) reported an annual

mean of 29% for animals in south Florida. Using plasma calcium levels and nesting

surveys as criteria, it was reported that 27.5% of the females in South Carolina were

nesters or potential nesters (Wilkinson, 1983). Additional studies indicate that

differences in female size could contribute to the frequency of nesting, with a higher

percentage of larger females nesting in a given year (Chabreck, 1973; Joanen, 1969).

Moreover, in South Carolina the proportion (42.9%) of larger females (>2.34 m









23
total length) found to be nesting was shown to be greater than the proportion (14.8%)

of smaller females (2-2.34 m total length) (Wilkinson, 1983). Using an increase in

plasma calcium concentration as an indicator of nesting, Lance (1989) estimated an

overall population nesting percentage in alligators from Louisiana from < 10 to

> 50%, dependent upon locale. The incubation period for alligators has been

reported as approximately 63 days in Louisiana (Joanen, 1969) and 65 days in Florida

(Jennings et al., 1987). Ferguson (1985) observed a mean incubation period of 75

days for alligator eggs at 30 0 C. However, the developmental rate of embryogenesis

is dependent on incubation temperature (Ferguson and Joanen, 1982; Ferguson, 1985,

1987; Joanen et al., 1987; Thompson, 1989; Deeming and Ferguson, 1989, 1990;

Phelps, 1992) and thus, varying nest temperature will influence the length of

incubation.



Determination of viability



Clutch viability for this study was defined as the proportion of neonates

hatching and living at least one day, divided by the total of all eggs within that nest.

The inverse of clutch viability is the ratio of non-viable eggs within that clutch which

represents embryonic mortality. Many different definitions of viability have been

used in other studies (Carbonneau, 1987; Joanen et al., 1987; Webb, 1987; Brandt,

1989; Joanen and McNease, 1989; Carbonneau and Chabreck, 1990). Often the use

of viability has been interchanged with hatching success to discuss reproduction in









24
alligators (Joanen and McNease, 1989). Hatching success for this study was defined

as the proportion of neonates hatching from only those eggs deemed viable (when

transilluminated), and set in the incubator. One index used to distinguish viable from

non-viable eggs is the presence of a "band" on the egg (Deitz and Hines, 1980;

Ferguson, 1982). The band appears as an opaque chalky spot on the uppermost

surface of the egg; the embryo and its attached membranes are always directly

beneath it (Ferguson, 1982, 1985). It is important to note that at oviposition the

embryo is at the late gastrula stage and extraembryonic membranes have begun to

form (Ferguson, 1987). The band progresses ventrally around the waist of the egg

and toward the poles as embryonic development progresses. It is known that the

initial "spotting" (onset of band formation) occurs within 24 hours of oviposition.

"Banding" progresses around the waist of the egg until a complete band encompasses

the circumference within 3 days (Ferguson, 1982). A pearl white band and the light

rose tint of the non-banded areas (by transillumination) aid in determining viable from

non-viable eggs. As embryonic development progresses, the band traverses the length

and circumference of the egg and the diminishing non-banded area becomes a darker

red (Ferguson, 1982, 1985). "Dead" eggs can be readily visualized and removed

during incubation using the absence or presence of a band as a clear designation of

viability.

As incubation and embryonic development progresses, a visual diminution of

the eggshell can be seen. The eggshell appears more porous and develops

longitudinal cracks. Several hypotheses have been suggested to explain this









25
modification of shell morphology (Ferguson, 1982, 1985, 1987; Packard and Packard,

1988). Ferguson (1982) stated that the degradation of the eggshell was due to

intrinsic bacteria found within the nesting media. Although crocodilians have some

similarities in their incubation biology, a tremendous amount of variation exists in

their nests. The crocodilians which inhabit estuarine environments (e.g., Crocodylus

porosus and C. novaguinea) can build either a nest of vegetation (similar to the

American alligator) or dig a hole nest in the sand of nearby shorelines (Webb, 1987).

The fresh water varieties (e.g., C. niloticus, Gavialis gangeticus, Alligator

mississippiensis) also show both types of nest construction (Joanen and McNease,

1987). It is known from field observations that infertile eggs or those exhibiting very

early zygote mortality exhibit no degradation of the eggshell even though "incubated"

for months or even a year in a nest consisting of rotting vegetation (Woodward,

Percival, Masson, unpubl. data). Recent experimental studies show that eggshell

degradation occurs in eggs in which the eggshell is cleaned with an

antibacterial/antiviral disinfectant and incubated in sterile, humid air (Guillette,

Heaton-Jones, Percival, unpubl. data). These field observations and experimental

data suggest that shell degradation is due to the presence of an embryo, not bacteria

in the nesting environment.

It has been determined that all crocodilians examined to date have

environmental sex determination (Ferguson and Joanen, 1982; Ferguson, 1982, 1985,

1987). However, the influence of the nesting environment on other aspects of

embryonic biology, such as viability of the eggs and subsequent growth and










development of hatchlings has been investigated only marginally (Goodwin and

Marion, 1978; Jennings, 1987; Joanen, and McNease, 1987, 1989; Webb et al.,

1987). These influences have been categorized as predation and flooding, although a

more recent study has looked at location (marsh or levee of nests, (N = 6), and the

potential effects this has on clutch size and hatchability (Schulte and Chabreck, 1990).

Other factors influencing the incubation of American alligator eggs have been

examined, including the effect of nesting environment on gas exchange and

vocalization (Joanen and McNease, 1987; Thompson, 1989). The insulating and

temperature control effects of nesting material has been studied sporadically, with

most studies looking at homogeneous habitats (Bara, 1972; Chabreck, 1973, 1975;

Goodwin and Marion, 1978; Dietz and Hines, 1980; Wilkinson, 1983; Joanen and

McNease, 1987, 1989). These studies of nest environment are from relatively small

geographic locales and do not investigate the diversity of habitat encountered by the

nesting alligator.

Due to the varying habitats found throughout the state of Florida, the potential

effect of the nesting habitat on reproductive success cannot be ignored. Moreover,

this variation in nesting ecology allows the testing of hypotheses concerning egg

viability and nesting ecology.


Determination of fertility










In order to accurately evaluate embryonic mortality, a precise method of

determining fertility must be established. A fertile egg is one in which the pronuclei

of the sperm and egg unite and produce a zygote. The fertility of an intact

post-ovipositional egg can be detected indirectly by the "banding" pattern of the egg

as described by Ferguson (1982) (see above for description). This method allows for

recognition of fertility only if the embryo survives to attach to the inner shell

membrane and therefore, produces a visible spot or band. Another successful, but

indirect method used in crocodilians is the presence of sub-embryonic fluid flowing

from an opened egg (Webb and Manolis, 1987). It is sometimes possible to see the

fluid movement within a viable egg using a horizontal candler (Masson, unpubl. data).

The origin of the sub-embryonic fluid is not known, but has been described as "being

secreted under the embryo and vitelline membrane of an egg containing an embryo "

and is used as a method to determine fertility for Crocodylus porosus and C. johnsoni

(Webb and Manolis, 1987). From these accounts, it appears two phenomena are

being described as a single event. That is, the fluid first identified as sub-embryonic

is from the gastrocoele of the 'young' gastrula-staged embryos. Secondly, the fluid

draining from an opened egg is probably derived from the water and waste by-

products in the allantois of later-staged embryos.

Embryonic death occurs at various stages but its causes still have not been

defined for any crocodilian. For that matter, very few studies have identified when

and under what conditions embryos die for any reptile (Limpus et al., 1979). The

"viability" rates of Florida alligators were thought to be lower than other regions










where the alligator occurs naturally, such as South Carolina and Louisiana

(Wilkinson, 1983). When the data are analyzed using the same criteria, however, t

he populations found in Florida are comparable to those of South Carolina and

Louisiana. In the past, infertility was considered a major problem with wild alligator

production in Florida. For example, fertility rates were reported as approximately 70

% for the years 1984-1988 from selected lakes in central Florida (Woodward et al.,

1989). The problem with ascertaining fertility rates is partly due to problems in

defining fertility as described above. The term "infertility" has been used previously

to describe those eggs without a readily identifiable embryo and/or the absence of

banding on the egg (Ferguson, 1982). For example, one method of determining

fertility involved laying questionable eggs on a counter and waiting two weeks to see

if the egg rots, apparently indicating the presence of an embryo that could rot as a

result of bacterial breakdown (P. Cardeilhac, pers. comm.). Unfortunately, eggs with

early stage embryos may not rot due to the antifungal and antibacterial proteins found

within the albumen. However, a more detailed examination, macroscopically and

microscopically, of the eggs for a zygote may determine the actual infertility rate.

The ability to identify true infertility is hampered only by our technology and lack of

information on alligator reproductive biology.

Fertilization in alligators has not been studied. Such basic information as the

biology of the acrosome reaction of the sperm or the number of sperm required to

fertilize the ovum are unknown. No mechanism for sperm storage has been found

(Guillette and Palmer, unpubl. data), and the reproductive tract of the female appears










to contain sperm only during the mating season in May and June. The reproductive

tract of the female can be separated into the ostium, tube, and two regions of the

uterus the fiber region and calcium region. We know nothing of sperm transport in

the alligator and presume that fertilization occurs in the upper regions of the tube as

occurs in birds (van Tienhoven, 1983). Once the ovum traverses the mid and lower

tube, it has acquired a thick layer of albumen which would presumably make

fertilization very difficult (see Palmer, 1990; Palmer and Guillette, 1991, 1992).



Embryonic mortality



Embryonic mortality may occur at any time after the pro-nuclei of the egg and

sperm unite. The majority of mortality studies to date have been on mammals (see

Putney, 1988). Early embryonic mortality ranged from 54.1% to 65.1% for bovids

(Putney, 1988). Early embryonic mortality was defined as prior to day 34 of

gestation. This time period includes when the genome of the embryo is activated

(prior to day 20) and implantation occurs (day 20-22). Both of these phases of

embryonic development are particularly vulnerable (Putney, 1988). Large mammals

generally produce large young that are precocial and have a greater chance of survival

after parturition than smaller mammals (van Tienhoven, 1983). The tradeoff is for

size, not numbers of offspring. These larger animals are less susceptible to

hypothermia and predation. The cost associated with this for the female is the

additional impact of maintaining a larger embryo and a longer gestation period. The











literature addressing embryonic mortality in birds is comprised of post-hatching

phenomenon in relation to life history strategies (see Steams, 1992). The production

of larger clutches has been argued to be an insurance against loss of an egg or

fledgling (Tuomi et al., 1983).

Few descriptions of embryonic mortality have been published for the American

alligator (Alligator mississippiensis) and very few data are available for any reptile

(Limpus et al., 1979; Blanck and Sawyer, 1981; Magnusson, 1982; Webb, 1987;

Woodward, 1987; Brandt, 1989; Joanen and McNease, 1989; Woodward et al., 1989;

Carbonneau and Chabreck, 1990). Each of these studies has discussed the natural

mortality in the nesting reptiles from predation, flooding, desiccation, or exposure.

However, only Blanck and Sawyer have addressed mortality during embryogenesis.

Their emphasis was to ascertain if mortality was attributable to movement and/or

incubation techniques in hatchery programs for the loggerhead sea turtle (Caretta).

Embryonic mortality after oviposition can be attributed to factors either

external and internal to the egg (Joanen, 1977; Wilkinson, 1983). Depredation of the

nest is a major cause of mortality with estimations of > 30 % occurrence (Goodwin

and Marion, 1978; Deitz and Hines, 1980). The major predators of nests are

raccoons, wild hogs, black bears and humans (Deitz and Hines, 1980; Magnusson,

1982). In addition, the construction and placement of the nest could adversely affect

viability. The nesting media may have an influential role in temperature regulation,

gas content and exchange, and humidity within the egg chamber. Currently, there are

no data available to indicate whether nest composition influences viability or not.










Placement of a nest where shading is minimal would cause a temperature elevation

within the nest cavity compared to a shaded nest. The nesting material is extremely

variable in Florida and includes construction materials where greater than 80 % of the

construction material can be rock (most as large as the eggs) and mud whereas other

nests are composed of up to 100 % sawgrass (Percival et al., 1991). These

conditions obviously influence temperature, gas exchange, and moisture content of the

clutch cavity. Incubation temperature is critical for crocodilian embryonic

development with low or high extremes being lethal (Webb et al., 1987; Ferguson,

1987, Deeming and Ferguson, 1990). That is, incubating temperatures below 28 0 C

or above 35 0 C are lethal to crocodilian embryos (Ferguson, 1985). Additionally,

embryonic temperature is critical for sex determination in all crocodilian species

studied (Ferguson, 1982). Flooding, whether from incidental rainfall seepage or from

submersion of a floating nest by the weight of the female has been documented as a

factor decreasing the viability of the eggs (Joanen, 1977; Kushlan, 1980). Embryonic

mortality may be attributed to flooding of the clutch cavity for as little as four or

more hours (Joanen and McNease, 1977). The signs of flooding are often readily

discernible, as the tannin from the nesting material discolor the eggs. Disturbance of

the nest and consequently the eggs, may also influence the occurrence of mortality.

This disturbance can be due to human activity or for example, by turtles using the

alligator nest as their oviposition site (Jackson, 1988). Likewise, other alligators may

use the same nest for egg deposition, thus disturbing the original nest.










The quality of the egg components (shell, albumen, yolk) can also affect

embryonic viability. Stress may be a major factor modifying egg component quality.

Stress as defined by Selye (1937) is the normal response of an organism to

environmental modifications, whether good or bad. Stress can, however, become a

pathological condition that modifies greatly the natural functioning of many organ

systems, including the reproductive system. High population densities and poor or

inappropriate nutrition may suppress or preclude reproductive activity in alligators as

it does in other vertebrates. Environmental pollution may modify reproductive tract

function as reported in the 1960's and 1970's for DDT contamination and eggshell

calcium in birds (Anderson and Hickey, 1972; Wiemeyer et al., 1986). For example,

there is an association between the porosity of the eggshell, determined by the

thickness of the shell, and egg viability in alligators. The eggs with fewer pores have

a higher incidence of early embryonic mortality when compared to eggs with greater

pore densities (Wink, et al., 1990). We know nothing about the formation of pores in

crocodilian eggs, although recent work has begun to determine the general

phenomenon of egg formation in reptiles (Palmer and Guillette, 1991). In fact, pore

formation in any hard shelled vertebrate egg (e.g., birds, turtles) is not understood

and yet, pores are critical for embryonic survival (see Packard and DeMarco, 1991).

The contents of the egg (yolk, albumen) also affects viability (Ferguson and

Deeming, 1990). The constituents of the yolk and albumen provide all the nutrient

needs, except for water, for the development and production of viable hatchlings

(Manolis et al., 1987; Palmer and Guillette, 1991). Utilization of these components









33
by the embryo and the appropriate concentrations of gases and water are necessary for

embryonic development. Factors which can affect the composition and distribution of

the egg components have not been examined in detail in any reptile. This is

especially true for environmental pollutants.

The presence of contaminants in the egg can cause or contribute to embryonic

mortality in alligators as it does in birds and fish (see Colborn and Clement, 1992).

However, there are few data available concerning contaminants in crocodilian eggs.

Contaminants including mercury, DDE, DDD and other metabolites of DDT have

been found in alligator tissues, fish and water obtained from several study sites in

central and southern Florida (Percival, 1989; Heinz et al.; 1990, Hord, 1990).

Elevated mercury and selenium were found in the shells of non-viable eggs from

Lakes Griffin, Apopka, and Water Conservation Area 3A obtained in 1990 (Hansen,

unpubl. data). There are elevated levels of DDE and DDD in alligator eggs taken

from lake Apopka (Percival, 1989; Heinz et al., 1990). In fact, the levels recorded

would have detrimental effects on bird embryonic development. Another study of

"infertile" eggs from several lakes was inconclusive in determining the extent of

heavy metals within the yolk of alligator eggs from lake Apopka (Cardeilhac, 1988).

Unfortunately, without causal data, the biological importance of pesticides, heavy

metals, and other contaminants on embryonic development of alligators cannot be

determined. A database, however, is growing rapidly showing that many

environmental pollutants do modify or preclude embryonic development in many

vertebrates (Colburn and Clement, 1992).










The influence of natural environmental factors such as drought, high

population density, and poor nutrition on the production of a 'good' egg must also be

determined. Moreover, the normal composition of viable alligator eggs must be

understood to evaluate natural and contaminant-induced variation. This work has only

just begun in reptiles ( Palmer and Guillette, 1991). Female stress may be associated

with inadequate production of egg components during vitellogenesis and immediately

after ovulation. Stress in alligators, as in other reptiles, is associated with elevated

plasma corticosterone concentration (Elsey et al., 1990). Follicular development and

vitellogenesis are dependent on elevated plasma estrogen concentrations. Stress

causes a decline in circulating E2 concentrations in female turtles (Mahmoud et al.,

1989) that could influence vitellogenesis. Similar studies have not been performed in

alligators or any crocodilian. Modification of circulating E2 could produce aberrant

eggs. The proportions of the major egg constituents (yolk lipids) and their changes

during incubation have been examined (Deeming and Ferguson, 1990). However, the

relationship of these changes to viability or normal egg production have not been

examined. Moreover, it had been assumed that the major function of albumen within

crocodilian eggs, as presumed for all reptile eggs, was the control of osmotic

gradients for embryonic water maintenance (Webb and Manolis, 1987). It has also

been assumed that the albumen proteins of all amniotes were similar. Palmer (1990)

found the albumen protein components of the alligator were similar to that of birds

and other reptiles but differed in proportion and there was an additional protein not

seen in other birds or reptiles (see Palmer and Guillette, 1991). The nature of the









35
albumen components can be directly related to antifungal and antibacterial roles. The

complexity and variety of compound s within the albumen suggest it has the ability to

provide an immunological barrier as well as trace elements. Other factors, such as

MRNA and/or growth factors, may be necessary and present within the egg to

facilitate and continue embryonic development (L. Guillette, pers. comm). The

incorporation of growth factors in teleost, amphibian and bird eggs all derived from a

maternal source (ovary or oviduct) has been described and is necessary for normal

embryonic development (see Colborn and Clement, 1992). The role of such chemical

mediators (for example: triiodothyronine (T3); insulin-like growth factor I (IGF-I)

and II (IGF-II) secreted by the ovary or reproductive tract is not known in reptiles. It

has been hypothesized that the early developmental importance attributed to these for

avians and mammals would be similar for the alligator (H. Bern, pers. comm.).

IGF-I was recently immunolocalized in the alligator reproductive tract (Cox and

Guillette, 1991). Furthermore, the presence of IGF-I and IGF-II has been

demonstrated in albumen from alligator eggs exhibiting early gastrulation (Guillette

and Williams, 1991). Embryogenesis may require not only these factors to be present

but also to be in appropriate concentrations. The synthesis of all these factors appears

to be dependent on elevated plasma E2 concentrations. Restriction of these factors by

such stressors as malnourishment, overcrowding or even environmental contaminants

could result in the reduction or absence of these products in the follicle s or albumen.

Further research on the physiological mechanisms associated with embryonic viability

are needed.










Finally, female behavior may adversely affect embryonic mortality. The

ability of the female to thermoregulate while eggs are in utero influences their

development rate. That is, temperature is known to influence embryonic development

in alligators, with cooler temperatures slowing down embryogenesis and warmer

temperatures accelerating it (Ferguson, 1982, 1987, 1989; Manolis et al., 1987;

Whitehead, 1987). It has been hypothesized that a reduction in salt-water crocodile

egg viability is due to females staying in the water for extended periods, thus,

reducing her temperature and subsequently that of the eggs (G. Webb, pers. comm.).

Conversely, social stresses also may keep the female out of water allowing the eggs to

become overheated.














CHAPTER 3

VARIATION IN CLUTCH VIABILITY OF WILD AMERICAN ALLIGATORS IN

FLORIDA



Mortality is high during embryogenesis in many vertebrates (Caughley, 1977).

In Florida, populations of the American alligator, (Alligator mississippiensis) have

been reported to exhibit embryonic mortality as high as 62% (Woodward et al.,

1989). This is somewhat higher than the reported 52% in Louisiana populations

(Joanen and McNease, 1989). However, these mortality percentages include losses

from flooding, infertility, and predation. The combination of infertility and

post-ovipositional mortality may cause the loss of 87 % of the eggs laid annually

(Woodward et al., 1989). High infertility rates (up to 30%) have been reported for

alligators in Florida (Jennings et al., 1987, Woodward et al., in press) with similar

values (24.6%) reported for animals from the Rockefeller Wildlife Refuge in

Louisiana (Joanen and McNease, 1975).

Alligator clutch viabilities have been described for several specific regions in

Florida including north-central (Goodwin and Marion, 1978; Dietz and Hines, 1980)

and the everglades ( Kushlan and Kushlan, 1980). Each of these studies has provided

data on viability and natural nesting conditions but have not addressed regional

differences. A recent study by Jennings et al., (1988) did investigate regional

37










variation. They found there were no identifiable patterns or trends for their four

study sites. The data from Jennings et al., (1988) were used as a pilot study by the

Florida Cooperative Fish and Wildlife Research Unit to address feasibility and

potential for a longer study.

Variability was observed in clutch viability and hatch percentages among four

alligator populations in Florida from 1983-1986 (Jennings et al., 1987; Woodward et

al., 1989, Woodward et al., in press). Cause for this variation was unknown but

different nest materials, nest temperatures, diets, adult densities, age structure of the

female population, and pesticide/heavy metal contaminants were suggested as possible

candidates (Woodward et al., in press). Recent reviews of the literature on

crocodilian embryology and nesting (Ferguson 1985, 1987; Webb et al., 1987;

Deeming and Ferguson, 1990), embryonic respiration (Whitehead 1987), and egg

chemistry (Grigg, 1987; Manolis et al., 1987) include no definitive clues regarding

possible causes of variation in alligator clutch viability in Florida. In fact, it is

unknown if the viability percentages observed are normal or aberrant.

The influence of the nesting environment on incubating crocodilian eggs has

been studied with varying interpretations (Barros, 1966; Joanen, 1969; Goodwin and

Marion, 1978; Bing Hua et al., 1981a,b; Ferguson, 1982; Jennings, 1986;

Carbonneau, 1987; Jennings and Abercrombie, 1987; Kofron, 1989; Maskey, 1989;

Schulte and Chabreck, 1990). Throughout the range of the American alligator, it is

known that they use available nesting materials for the construction of their nests.

Nest site location and viability were addressed for Louisiana by Schulte and Chabreck









39
(1990), although they only examined a total of six nests from three macrohabitat types

(levee, high and lowland marsh). However, quantification and analyses of different

vegetation usage have not been undertaken. Brief descriptions of general nest

formation were made with a basic classification of macrohabitat types by Wilkinson

(1984), Jennings (1986), Jennings and Abercrombie (1987). These classifications

were for nests located on floating islands, levees, wooded swamps, islands, and

emergent marsh. Clutch cavity dimensions have been examined in relation to the

effects of temperature on sex ratios and circumstantially in relation to viability

(Carbonneau, 1987; Ferguson, 1987). It has been hypothesized that the nesting

vegetation could raise the carbon dioxide content and decrease oxygen concentrations

in the nest cavity thereby decreasing viability (Ferguson, 1987; Schulte and Chabreck,

1990). In addition, the specific nesting material may not maintain its rigidity

throughout the incubation period and the female attending the nest and lying on top of

it she would crush the eggs. The nesting material acts as an insulator against the high

and low temperatures of the day and night (Chabreck, 1973).

In order to determine the effect of nest characteristics on viability, the

following null hypotheses were tested. Ho 1: Different habitats within the state of

Florida have no effect on alligator egg viability; Ho 2: There is no annual effect on

alligator egg viability within these habitats; Ho 3: Nesting material used during nest

construction does not affect the viability of alligator eggs during incubation. These

hypotheses were tested by: (1) determining if clutch viability varied among seven

wetlands in three separate drainages; (2) determining egg viability trends over years










within wetlands; and (3) examining the relationship of various environmental and

biological variables with clutch viability percentages.



Methods



Study Sites



Three drainages from north central to south Florida were chosen. These were

the Oklawaha, St. Johns, and Everglades. These drainages represented diverse

wetlands from open marsh and canals (Everglades) to a mixture of wooded swamp,

open marsh, and levees (Oklawaha, St. Johns). Additionally, each area has had

preliminary work on clutch viability and there were sufficient nests produced so that

the Florida Game and Fresh Water Fish Commission could permit the collection of 35

clutches without exceeding a removal of greater than 50% of all nests from those

areas. The Oklawaha River Drainage study sites were Lakes Griffin (Gr) and Apopka

(Ap); the St. Johns River Drainage included Lakes Jessup (JE) and George (GE); the

Everglades drainage included Indian Prairie (ON) and Observation Shoals (OS)

marshes of Lake Okeechobee and Conservation Areas 2A and 3A (CA) (Figure 3.1).











Egg Collections



A targeted sample size of 35 clutches was collected each year during June or

July from 1988 to 1992 from Lakes Griffin (Gr), Apopka (Ap), Jessup (Je) and

George (Ge) and Indian Prairie (ON) and Observation Shoals (OS) marshes of Lake

Okeechobee and Conservation Areas 2A and 3A (CA). Due to nesting densities

and/or logistical costs, targeted sample sizes were not met for Lake Apopka for any

of the collection years and for Lake George in 1988 and 1989. Egg collections were

conducted by field crews directed to the most convenient access point of targeted

nests by an observer in a helicopter. In a few instances, nests obscured by vegetation

from above were located by ground crews. Upon identifying a specific nest, a

general location within the lake and a nest number was recorded. Each nesting site

was categorized using the following general habitat descriptions: wooded swamp,

emergent marsh, island, levee or floating island. Estimates on the amount of shade

on each nest during the entire day and time of arrival at the nest were recorded. A

brief description of the vegetation within ten meters of the nest was described. Nest

composition was listed as a gross percentage of nesting material. For analysis the

nest was placed into a specific category only if greater than 50% of the nesting

material was of a single type. Field crews recorded 18 different classes of vegetation

for nest construction. The following eight categories were used for analysis: grasses,

peat, phragmites, typha, sagitarria, sawgrass, spartina. Some classes were

represented by from one to five nests and were placed into a category called other










(including; mud, gator weed, smart weed, poison ivy, maidencane, bullrush, rocks,

vines, woody and herbaceous material). Because of the varying knowledge of native

vegetation by field crews some of the vegetation classifications were listed as

unknown, in those cases the data were not used for analysis.

A sketch of each nest and the "microhabitat" (area within three meters of the

nest) was made. This diagram included prominent physical factors such as trees,

rocks, guard pools, trails and vegetation types with location of these factors in

reference to the nest. Identification and numbers of any additional species of eggs

(reptilian or amphibian) were recorded. The sketches were used in verifying

estimates of percent shading and to identify any nests which had trees or large rocks.

Nests that were constructed around trees or other obstructions were not used in

calculating nest volumes.

At each nest, the following data were recorded: nest height, longest and

shortest diameters, ambient temperature, clutch cavity temperature, clutch depth

(depth to the upper most egg), status (dry, partially flooded or flooded), female

presence, female behavior, nesting habitat and general comments. Nests were

classified as dry, moist or wet depending on an estimate of the water content. Dry

nests were those in which no water could be squeezed from the material, whereas wet

nests dripped water and the category in between was moist nests. The height of the

nest was established from a line parallel to the nest site substrate and at the level of

the highest point of the nest. The clutch cavity temperature was taken by excavating

a small opening into the top of the nest until the eggs were located and then










immediately placing a thermometer within the cavity and covering it. The ambient

temperature was recorded adjacent to the nest in the shade. Each thermometer was

calibrated to a standardized calibration thermometer prior to its use.

Each nest was completely searched because multiple nesting (two or more

females ovipositing in a single nest) had been observed previously (Percival et al.,

1991). Prior to removal from the nest, alligator eggs were marked on their

uppermost surface, avoiding the opaque band, with a felt-tipped waterproof marking

pen to indicate their relative orientation. Individual eggs less than 25 % inundated

with water (termed dry) were marked with black; eggs inundated 25-75% (termed

partially flooded) were marked red; and eggs inundated over 75% (termed flooded)

were marked blue. A nest was determined to be high and dry if greater than 75 % of

the eggs were termed dry. All clutches used in the analyses described in this study

were from high, dry nests. Large sample sizes allowed for elimination of nests with

any partially or flooded eggs, except on Okeechobee North and Okeechobee South in

1988, when some partially flooded nests were accepted in order to complete the

required sample sizes. Flooded eggs from those clutches were deleted from analysis.



Incubation



Eggs were positioned in plastic pans (61 cm X 36 cm X 13 cm) on 5 cm of

natural nest material with additional nest material cushioning layers of eggs when

required. Eggs along the pan perimeter were cushioned with 2-3 cm of nesting









44
material. An identifying plastic tag was affixed to a hole in the comer of the pan rim.

Nest number, date, time, and crew members were recorded on the identifying tag.

These data were important in matching data sheets to appropriate clutches when errors

were made. Eggs were placed within the pans in a serpentine pattern, in order from

the uppermost egg in the cavity to the lowest egg and covered with natural nesting

media (2-4 cm). The uppermost egg always was placed in the same corner as the

identifying tag. Parts of eggs or eggs which were found crushed were placed in a

sealable plastic bag, labeled with the nest number, and placed on top of the last layer

of eggs in the pan. Care was taken to remove rocks, sticks, or other hardened objects

from the nesting material used for insulation and cushioning in the pans.

During transportation of clutches, care was taken to avoid excessive vibration

or shock which could cause detachment of the embryonic membranes from the shell

and subsequent mortality (Limpus et al., 1979; Ferguson 1985, 1991). Woodward et

al., (1989) found that with care, handling and transport of alligator egg s avoided

such mortality. The pans were held level and in such a manner as to avoid contact

with any part of the airboat or the body except hands while in transit to a 4.9 m

aluminum, egg transport airboat. Clutches were padded (Woodward et al., 1989)

within the transport boat which carefully was maneuvered within sheltered waters as

much as possible to avoid additional vibration and shock. Eggs generally were

transported to an incubator facility in Gainesville, FL on the same day of collection in

the covered bed of a pickup truck, and padded similarly to the transport boat

(Woodward et al., 1989). In a very few instances, logistics demanded delivery on the











second day after collection. Temperatures were monitored to insure that eggs were

maintained between 28 and 34 o C during transport to the incubator.

At the incubator facility, the pans, with and without the eggs, were weighed to

calculate a clutch weight. Eggs which were crushed or broken and had been placed in

plastic bags at the time of collection w ere not included in the clutch weight.

Extreme care was taken not to bump or rotate any of the viable eggs. All intact eggs

were transilluminated (candled) to check for viability and egg band development

(Woodward et al., 1989). Measurements of the long axis and widest region

perpendicular to the long axis of eggs were taken using vernier calipers (to the nearest

0.1 mm). Eggs which could not be accurately measured because of physical defects

or swelling and cracking due to advanced embryonic age were not included in the

analyses. Each egg was measured in 1988, and after preliminary analysis it was

determined that a smaller subset of eggs could be measured. That is, variation among

eggs in a clutch was so small (P < 0.001, d.f. = 288) that a subset of measurements

would adequately describe egg size. The following three years, ten eggs from each

clutch were randomly chosen for length and width measurements. The exception to

this, was in the case of abnormally large clutch size where the possibility of multiple

clutches existed; each egg was measured in these clutches. Anomalous eggs (those

containing two yolks, yolkless or with aberrant calcified regions) were recorded and

when possible were also measured. Measurements of the eggs were taken

independent of their viability status.











Two incubation facilities were used: (1) the primary incubator housing

clutches was a modified 2.4 m x 4.8 m x 1.8 m chicken hatcher (INC) (Robbins,

Denver, CO.), with controls for temperature, humidity and circulating air. Two

blades from each of its 2 fans were removed to reduce air movement and decrease

evaporative water loss from the incubating medium and subsequently, the eggs.

Humidity was enhanced by dripping water onto a burlap partition placed in front of

the fans. Mean relative humidity readings for three hour intervals were 95 % + 1 %

(measured by wet bulb thermometer). Additionally, atomized sprinklers were placed

between the egg racks and turned on for 20 minutes each morning and early afternoon

to minimize desiccation of the incubation medium and eggs. Individual trays within

each rack were inspected twice per day to insure adequate moisture content for each

clutch. Target temperature within the incubator was 30.5 0 C (a mid-range

temperature for predicted optimal survival, Joanen and McNease, 1987) and was

monitored with mercury thermometers and a multi-probe thermocouple system (Atkins

Technical, Gainesville, FL.) designed for this purpose. Eight probes were positioned

randomly within the nesting medium of sphagnum and in the general circulation of the

incubator. Temperatures were printed hourly and averaged 30.7 + 0.4 0 C.

The second incubation system used was a modified 7.3 m x 3.7 m portable

building (Bi) (Lark Industries, FL.). The building was layered for insulation with

2.54 cm blown-in styrofoam insulation, 7.6 cm fiberglass batting, a 0.3 mm plastic

vapor barrier an d 2.54 cm styrofoam sheeting. Humidity and temperature were

controlled by a Hawkhead International heat/cool humidifier unit. This unit utilized











mercury switches for temperature and wet-bulb humidity monitoring. Electric fans

circulated the warm humidified air throughout the building. A thermostatically

controlled fan and louvers provided additional cooling, if necessary. The relative wet

bulb humidity in the building was 94-96% and mean temperature within the nests and

in the building was 30.6 0 C + 0.5.

The gas samples from the incubator were not different from the control sample

of outside air (P > 0.05). The analyses of the incubation gas environment did not

reveal any differences within regions of the incubator (P > 0.5, d.f. = 27). Oxygen

(> 14%) and carbon dioxide (< 7%) levels were measured from samples taken

periodically within nesting medium and were within limits described for other

reptilian nests (Thompson, 1989).

Eggs were maintained within their clutch in the same order as collected. A

maximum of three clutches per tray, separated by a board, nested in sphagnum moss

were placed in the INC. Remaining clutches were placed in sphagnum in pans and

shelved in B1. Each tray and pan was covered with 50% shade cloth to allow air

circulation but maintain the integrity of the clutch and neonates after hatching.

Clutches were placed in trays or pans with a minimum of 2.54 cm of sphagnum

insulation on the sides, top and bottom. To minimize premature hatching of alligators

by audible cuing from adjacent clutches, clutches with similar hatch dates were

grouped together.

Neonatal alligators were maintained in an insulated building (Lark Industries,

FL) in Gainesville. The floor was covered with water resistant acrylic latex and was









48
mopped daily with a dilute bleach solution. Accessory heating was provided by two

5,000 BTU radiant electrical heaters. Lighting and secondary heating consisted of

two fluorescent lights and two infrared heat lamps. Cooling was regulated by a

thermostatically controlled exhaust fan and louvres. The building contained 18

galvanized steel tanks separated into 72 individual 0.6 m x 0.7 m compartments.

These were sealed from each other with rivets and latex caulking. To prevent cross

contamination all compartments were individually rained. Each compartment was on

a 3.8 cm slope to provide a dry surface at one end and water (maintained at 29-30 0

C) at the other end. To increase the dry surface area and minimize stress from

overcrowding, an elevated mesh platform was added to each compartment (2.54 cm

squares made of molded plastic-coated heavy gauge wire screen). Tanks were

cleaned with a hot (40 0 C) 3 % bleach solution then rinsed thoroughly. A maximum

of 30 hatchlings were kept in each compartment. Whenever possible, clutch integrity

was maintained and hatchlings from a given clutch were kept within a single

compartment. Personnel were required to wash their hands and equipment with a

Betadine surgical scrub solution prior to beginning work and between each clutch or

compartment to eliminate possible cross contamination.

Most hatchlings were retained until they were actively feeding (approximately

10 days). After yolk absorption was complete and edematous swelling of the feet was

not noticeable, hatchlings were web-tagged with sequentially numbered #1 Monel tags

between the 3rd and 4th toes of the right rear foot. Data on physical abnormalities,

weakness, and death were recorded.











Dead embryos and hatchlings were preserved for analysis of anomalies.

Mortality was recorded for all clutches collected since the 1987 nesting season.

Weight, total length and snout vent length have been recorded on all live hatchlings in

1987 and 1988 and for a large sample of hatchlings in 1989 and 1990.



Analyses



Clutch size was the total of all shelled and unshelled eggs found in a clutch

cavity. Clutch viability was calculated as the number of hatchlings surviving > 1 day

divided by clutch size. Since the main interest of the study was in the inherent

viability of eggs, clutches from flooded or disturbed nests (by predators, turtles,

humans, or other alligators) were excluded from viability analyses. However, only

clutches from disturbed nests were excluded from analyses of clutch size. Each

clutch was the experimental unit in analyses of viability and clutch size. Analyses

involving clutch weights included only clutches having greater than 2 eggs and valid

clutch weights. Valid clutch weights did not include any crushed eggs collected in

plastic bags.

Viability percentages were calculated as the ratio of hatchlings (surviving > 1

day) from all the eggs within a clutch (viability percent = (hatchlings / total clutch) X

100)). The hatch percentages were calculated as the ratio of hatchlings (surviving >

1 day) from only those eggs that were incubated (hatch percent = (hatchlings /

number incubated) X 100)). The viability and hatch percent data were arcsine










transformed for statistical analyses (Sokal and Rohlf, 1981). Data were analyzed

using computer programs (SAS 6.01 and DBase III Plus for the IBM whereas

SuperAnova and Statview were used on a Macintosh SE/30).

Study design was a 4 year by 7 area fixed effects factorial, but differences

between Okeechobee subareas were tested for possible pooling into 1 area. An effect

was considered significant (P < 0.05) if 1 subanalysis so indicated. If either no

AREA x YEAR interaction was detected, or an interaction with a significant area

mean effect, then comparisons were made of all possible pairs of area means.

Because analysis of trends among areas was a principle study objective, the AREA x

YEAR linear effect was tested if the full AREA x YEAR interaction was significant.

In this circumstance, linear contrast of year within areas was tested and compared all

possible pairs o f contrast estimates. Mean or trend pairs were declared significantly

different when P < 0.05/k (where k was the number of comparisons to be made), to

guarantee that the probability of falsely declaring significant greater than 1 of the k

comparisons was no greater than 5 % (Bonferroni adjustment). Viability was

regressed on area, year, clutch size, and clutch weight to determine the relationship of

clutch parameters to clutch viability. Analyses were made of nesting parameters (nest

height, nesting material, egg cavity depth, egg cavity temperature) to determine if the

physical environment prior to collection had a significant effect upon the viability

percent. Nest characteristic data were standardized for ontogenetic differences in

metabolic heat production of the clutch by using embryonic age as a covariate. It was

realized that the temperature data were for a single fixed point in time when clutch











cavity temperature may be a dynamic condition responding to solar heating, rain,

shade etc. Therefore, time of day was used as a covariate to standardize egg cavity

temperatures and reduce bias due to solar heating. Additionally, for my comparisons

it was assumed degradation percents of a given nesting material would affect viability

equally among clutches, within an area and among areas.

Although Woodward et al. (1989) found no difference in incubators used in

earlier studies of egg viability, it needed to be confirmed for the incubators used in

this study. The INC was used to incubate most of the eggs each year. All clutches

from lakes Okeechobee North, Okeechobee South, George, and the Conservation

Area and 35 clutches from lakes Jessup and Griffin. Because > 35 clutches were

collected from lakes Jessup and Griffin, the excess above the minimum sample size of

35 were incubated in B1. All lake Apopka eggs were incubated in B1 principally

because all space in the hatcher was consumed by the time lake Apopka eggs were

collected (generally the last collection date of the field season).

Clutch cavity and ambient temperatures were tested for correlation using a

linear regression model (SuperAnova). Time of day was used as a covariate with

analysis of covariance (ANCOVA) in determining relationships between ambient and

clutch cavity temperatures. Three way analysis of variances (ANOVA) were

performed on nesting data for determining the affects of lake, macrohabitat and

vegetation type on viability percentages.











Results



Variation and Trends in Clutch Viability



Clutch size



A total of 1258 alligator clutches (52,592 eggs) were collected out of 1215

nests from 7 Florida study areas during the 1988-91 nesting seasons (Table 3.1). The

difference in total clutches and nests is the sum of multiple clutches obtained from

nests. Mean collection date for all clutches was 19.1 + 0.4 days post- oviposition

(Table 3.2). Mean clutch sizes ranged from a low of 31.5 eggs per clutch from

Conservation Area in 1991 to a high of 49.3 from lake Apopka in 1988 (Table 3.2)

with an overall mean (all years and all lakes) of 43.2 eggs per clutch. The mean

clutch size for the Everglades drainage was significantly (P < 0.05) lower than the

Oklawaha or St. Johns drainages (Table 3.2). Individual clutch sizes ranged from 1

(n = 2) to 71 (n = 1) (Figure 3.2). Mean clutch sizes by lake showed the

Conservation Area and Lake Okeechobee as having significantly smaller clutch sizes

than Lakes Apopka, George, Griffin and Jessup (Table 3.3). The mean clutch size

within each study site was not significantly different among years.

Clutch sizes > 65 probably represent contributions from 2 females. That is

most nests which contained greater than 65 eggs could be separated into two or more

clutches by egg size. Clutches were considered distinct if found in discrete clutch










cavities within the nest, between clutch egg sizes were distinctive, and/or egg

deposition dates were different. In three instances, the variable sizes of eggs within a

large clutch precluded exact clutch size, and these were not included in the analyses.

These were 2 clutches from lake Jessup (66 eggs, 1989 and 71 eggs, 1990) and 1

from lake Griffin (70 eggs, 1991). The possibility also exists that nest disturbance

went undetected and some proportion of the small clutches were only partial clutches.

Also a confounding observation is that 5 of 32 females, which were collected at their

nests in 1989 and 1990, had retained 2 to 5 intact, and apparently normal eggs in the

oviduct after ovipositing 36 to 51 eggs (Masson and Guillette, unpubl. data). Actual

clutch size for these females could have ranged from 38 to 54. Thus, some

proportion of all nests may be considered partial clutches.

Clutch weight varied among study areas (P = 0.05) (Table 3.1). The lightest

clutches (2405 gm) were from Okeechobee South in 1990 and the heaviest (4025 gm)

from lake George in 1989. A positive (P = 0.0001) linear relationship existed

between clutch size and clutch weight. Hall (1990) found some evidence, with a

small sample size, that clutch weight may be related to female size.



Viability



Viability and hatch percentages of lakes Jessup and Griffin clutches incubated

in the hatcher (INC) were tested against those incubated in B1. No differences in

hatch percent or the viability percent could be detected between the 2 incubator types.











Further evidence that the incubators had no effect on hatch percent or the viability

percent was a significant correlation at the 0.05 level (r2 = 0.798) between viability

percent (independent variable) and hatch percent (dependent variable). No differences

(P = 0.484) in hatch percentages were found between incubation facilities 1988-91.

Therefore, data from both incubators are combined in the viability and hatch percent

analyses.

Viability varied greatly among the 7 study areas with the lowest annual mean

of 3.9% on lake Apopka in 1988 and the highest of 71.0% on the conservation areas

in 1990 (Table 3.1). No trends (linear or polynomial) have been detected (P > 0.1)

for any area. However, lake Apopka has shown a positive increase in viability for

the 1988-1991 study period. Even with this increase, lake Apopka has had consistent

and significantly lower viability percentages than the other study areas (F = 15, d.f.

= 6; P < 0.05). A test for homogeneity of slopes was performed for trend analysis

with no significant difference between any of the study sites and a slope of zero (P >

0.05), excluding Lake Apopka (Figure 3.3). However, if data for 1992 (Masson,

Guillette and Percival unpubl. data) are added to the 1988-1991 lake Apopka mean

viability percentages there is no significant change in any of the sites when compared

to a slope of zero.

Overall mean viability for the study period 1988-1990 was 44.1% (Table 3.1).

Viability percentages were normally distributed using g I and g2 tests (Sokal and

Rohlf, 1979). A nested ANOVA detected a significant year effect (P = 0.0001), area

effect (P = 0.0001), and a year by area interaction (P = 0.0007). Using the











Tukey-Kramer test, viability varied significantly among all years (P < 0.05:

1988-1991) when data from all lakes are grouped. When data on clutch viability from

lake Apopka is excluded, no difference was found between 1989 and 1990.

Therefore, the overall increase in viability from the grouped data was attributable to

the rather dramatic increase in clutch viability observed for lake Apopka in 1990.

Mean viability for lake Apopka was lower (P < 0.05) than those for all other study

areas; Conservation Area was higher (P < 0.05) than all other areas except

Okeechobee North, and Okeechobee North was higher (P < 0.05) than lakes Apopka

and Jessup.

Annual variation of viability within each area did not exhibit a common pattern

(Figure 3.4). Mean viability for all study sites was (P < 0.05) higher only between

1988 and 1990, 1991 in lake Apopka, Conservation Area, and Okeechobee South.

Differences (P < 0.05) in mean viability between 1988 and 1990; and between 1989

and 1990, 1991 occurred only in lake George. Mean viability was higher (P < 0.05)

in 1990 and 1989 when compared to 1988 only in lake Griffin. No annual variation

in viability was detected in lake Jessup or Okeechobee South. Mean hatch percent

ranged from 36.5% on lake George in 1989 to 89.9% on the Conservation Area in

1990 (Table 3.1). Mean hatch percent on lake George in 1988 (51.2 %) was different

from the hatch percent in 1989 (Hatch percent = 36.5%) and 1990 (Hatch percent =

66.0%). The mean hatch percent did not vary between 1989 and 1990 (Hatch percent

= 63.6%). For all years combined hatch percent from Conservation Area was higher

(P < 0.05) than the hatch percent on lakes Griffin, George, Jessup, and Apopka; the










hatch percent on lake Apopka and lake Jessup was lower (P < 0.05) than

Okeechobee South and Okeechobee North.



Relationship between Clutch Environment and Clutch Viability



The materials most frequently used for nesting were sawgrass (Cladium

jamaicensis), cattail (Typha spp.) and cord grass (Spartina spp.) (Figure 3.5). Mean

nest heights were higher (P < 0.05) on Okeechobee North and Okeechobee South

than for the other study areas (Figure 3.6). Lakes Apopka and Griffin had lower (P

< 0.05) nest heights and mean clutch cavity depths. There was no correlation

between ambient temperature and clutch cavity temperature (P < 0.05)). The nest

temperatures were rather constant between 27 and 35 0 C (Figure 3.7) and were not

different (P < 0.05) between systems, among areas or within an area. The time of

day was a covariate for the temperature analyses. Nest height and cavity depth were

directly attributable to the materials comprising the nest. However, the Sagittaria

latifolia nests which had lower nest heights and cavity depths, also had lower clutch

cavity temperatures (P < 0.05). No relationship between nest height and viability (P

> 0.05) nor between egg cavity depth and viability (P > 0.05) were detected.

Changes in the natural incubation media during the full natural incubation event could

not be addressed within the scope of this project.

Shade was not correlated with clutch cavity temperature (Figure 3.7). Lakes

George and Apopka had the highest percent shade and the greatest amount of wooded











swamp nests, however, no relationship with clutch viability could be established.

Conservation Areas and lake Okeechobee North and South had the largest nest

volumes and these nests were almost uniformly Typha spp. in composition. These

nests were also the least shaded, both of these factors directly related to the open

marsh conditions occurring at these study sites. There was a significant correlation

between nest height and egg cavity depth (r2 = 0.83; P < 0.05). Depth to the egg

cavity was proportional to the height of the nest (Figure 3.8).



Discussion



Significant variation existed in viability percent and hatch percent among

populations. Viability percentages ranged from a low of 3.9% on lake Apopka to a

high of 71.0% from the Conservation Areas. There was not a discernible pattern of

viability for any of the study sites nor was there a predictive pattern when study sites

were combined. Variation of clutch viability within each study site and between years

within a study site suggest a need to examine individuals within a population to

determine their reproductive contribution over time. Additionally, the relationship of

reproductive females within a population needs to be identified to factor out

confounding variable such as senescence, interbreeding periods, recruitment of prima

parous and the impact from their contributions.

The extremely variable, low viability and hatch percentages for Lake Apopka

confirms the suggestion that a severe environmental problem exists on this lake











(Woodward et al. in press). Lake Apopka has been described as one of the most

polluted lakes within Florida (Heinz et al., 1990). Pollution of this lake had initially

been described as due to the eutrophication and water stability. Speculation had been

that other factors such as draining the marsh and back-pumping of irrigation water

(the practice of using water for irrigation, collecting the drainoff and pumping it back

into the source) were major contributors. Pesticides found within alligator eggs were

first thought to have been derived from the back-pumping. However, in 1990 it was

learned that there had been a major chemical spill at Gourd Neck Springs in 1980.

This spill consisted of approximately 2500 gallons of dicofol (a pesticide almost

identical to DDT in its structure) and approximately 5000 gallons of concentrated

sulfuric acid. The Environmental Protection Agency declared Gourd Neck Springs

area of lake Apopka as a "Superfund" site in 1982. Without knowledge of a chemical

spill and its impact on the alligator population in this lake, research has been kept in

the discovery phase. That is, explanations for low clutch viability have not been

found while continued documentation of a serious problem has been ongoing.

Knowing that pesticides such as dicofol have estrogenic affects (Colborn and Clement,

1992) has revealed another avenue for further research on lake Apopka, the

examination of adult males and females for contaminants and possible immigration

and emigration.

The variation among viability and hatch percentages on the other study areas

was not so easily classified. All of the systems have been subjected to environmental

degradation (Heinz et al., 1991), but not so blatantly obvious as that apparently











occurring on Lake Apopka. The variation observed in viability and hatching on the

study areas (except lake Apopka) occurred generally within a more limited range.

However, viability was still lower than that reported for other crocodilians. For

example, the gharial (Gavialis gangeticus) has > 75% egg viability (Maskey, 1989).

Gharial work in Nepal is cited as an example because of the gharial's severely

endangered status, limited technical support, thus greater possibility of handling shock

in transportation of eggs, more primitive incubation facilities and techniques. Thus, it

is apparent that egg viability is depressed for the populations of the American

alligators studied and is due to some extrinsic force other than collection methods.

Recording and measuring the effect and determining the cause, however, are two

different aspects of a study.

The analyses presented here have begun to eliminate some potential causes of

depressed viability such as nest composition, nest location, and percent shade.

Viability could be affected by such factors as: nest composition; nest location; female

nest attendance during the full natural incubation event. There was no incubator

effect, therefore viability was determined prior to collection.

Likewise, mean collection date was 19.1 + 0.4 days post-ovipositional, so

nest material apparently had a minimal opportunity to affect viability (certain nest

materials such as 'flag' (S. latifolia) may demonstrate an effect because of the

likelihood of flooding, possible reduction of gas exchange and elimination of space for

gases in the clutch cavity. Additional analyses are needed to investigate nest material

effects directly on gas exchange, heat exchange and tendency to flood during late









60

embryonic development. Regardless, parameters associated with the nest could affect

only post-ovipositional components of viability since 37% of embryonic mortality

occurred prior to oviposition); coincidentally 65% of mortality occurred prior to the

6th day of incubation (see Chapter 4). It is important to note that because no affect

of the nesting material was detected on egg viability prior to day 19, does not mean

that an affect might not be observed during later stages of embryonic development.

Possible critical times would be during sex determination (4-6 weeks of incubation,

Ferguson, 1991) or the last ten days of incubation when gas exchange and water

uptake increase (Thompson, 1989). Due to degradation of nesting material, nest

height would be reduced over the 65 day incubation period and the relationships

between nesting material and viability in the wild would undoubtedly change. The

rate of decomposition of the materials, moisture content and compaction from female

attendance would be a few of the variables which would change during incubation.

Sagittaria is an extreme example of nesting material which has a high water content

and decomposition and compaction appear to progress more rapidly than for other

materials.

The viability percentages were erratic within each lake, between lakes and

among years. Data obtained on viability in 1984-1987 (Percival, Jennings,

Woodward unpubl. data) demonstrate the extreme variation for those years. The data

for 1988-1991 which include two additional areas (lake George and the Conservation

Area), show this variation is a common theme in Florida. However, it is obvious that

lake Apopka has had difficulties with low viability. Coincidentally, the decline in










viability and population density parallels a large chemical spill into Gourd Neck

Springs which brought about the declaration of lake Apopka as an EPA super fund

site (Woodward et al., in press). The positive trend and possible recovery for lake

Apopka is lessened to a line with a slope of zero when the data obtained by Percival,

Jennings and Woodward during 1984-1987 is added to the current data set. Mean

viability for lake Apopka has not increased appreciably and remains at 27.7% for

1992 (Masson, Percival and Guillette unpubl. data). The viability percentages for

each lake do not indicate any specific pattern and could not be correlated with any

parameters yet measured. Mean viability for each clutch size (over 20 eggs) is not

correlated (r2 = 0.4).

Reproductive potential of alligators is based on the number of neonates which

are produced from a system, lake and nest. The viability percent for each lake is the

starting point for survival curves. Variation in egg viability suggests that alligators

have a great deal of elasticity in trying to obtain an optimal clutch size. Although

clutch sizes were variable and the areas with smaller clutch sizes appeared to have

higher viability, as of yet no correlation has been found between clutch size and

viability. Therefore, instead of ascribing alligators an optimal clutch size based on

the Lack clutch theory it would be more prudent to list a range of clutch sizes. It is

imperative to gather data on female nutritional state, senescence, nesting potential and

status within the population to establish a possible link between clutch size and

viability. Then possible fluctuations of the alligator population in Florida can be

extrapolated from the viability percentages.









62
The preponderance of data used to develop hypotheses concerning reproductive

potential and viability in vertebrates has come from either birds or mammals (Steams,

1992). These data may lead to interpretations which are not appropriate for animals

that have clutches or litters larger than 4 or 5. Since alligators produce an average

clutch of 42.3 eggs, they afford an opportunity to examine intra clutch variation.

Larger clutches also lend themselves to looking at the establishment of the viability

percent for embryos (onset of a survival curve) by examining embryonic mortality.

Moreover, embryonic mortality can be examined in order to determine if there is a

particular time at which mortality has a significant affect on viability curve. Those

embryos not surviving to hatching or birth can be used to quantify problems within

the environment or female and possibly predict populational changes. Analyses of

these dead embryos can be made for contaminants, genetic anomalies or deficiencies

metabolic constituents.








63



Figure 3.1: Alligator study areas






m



0>



(/0

C











Table 3.1: Mean Clutch Parameters for Alligators from 1215 Nests
from 7 Study Areas in Florida from 1988-1991.


No. of Mean Mean
Clutches Clutch Clutch Viabil.a Hatchb
Area (eggs) Size Weight Rate% Rate%
Apopka
1988 21 (1036) 49.3 3575 3.9 38.8
1989 22 (950) 43.2 3386 9.8 39.6
1990 25(1127) 45.1 3643 30.8 63.0
1991 23(1125) 46.6 4151 28.6 44.2
Conservation Areas 2A & 3A
1988 35 (1148) 32.8 2667 53.7 68.1
1989 37(1319) 35.6 2791 66.9 85.7-
1990 35(1162) 33.2 2500 71.0 89.9
1991 35(1103) 31.5 2356 64.1 80.5
George
1988 23 (1026) 44.6 3544 40.4 51.2
1989 26 (1218) 46.9 4025 24.2 36.5
1990 34 (1552) 45.6 3822 60.5 66.0
1991 35 (1637) 46.8 4043 56.3 76.1
Griffin
1988 82(3914) 47.7 3940 31.1 49.0
1989 95 (4255) 44.8 3816 44.7 72.4
1990 96 (4463) 46.5 3958 55.8 76.4
1991 98 (4500) 45.9 3894 40.9 62.7
Jessup
1988 49 (2396) 48.9 3799 37.1 53.9
1989 50(2321) 46.4 3849 44.0 58.4
1990 56 (2644) 47.2 3799 41.0 60.0
1991 60 (2869) 47.8 3906 51.0 68.9
Okeechobee North
1988 36 (1522) 42.3 3062 48.4 59.0
1989 35(1499) 42.8 3282 51.0 77.1
1990 35 (1297) 37.0 2747 55.4 74.1
1991 35 (1337) 37.1 2776 64.2 80.5
Okeechobee South
1988 32 (1239) 38.7 3033 35.4 57.1
1989 36 (1392) 38.7 2897 66.9 84.9
1990 34 (1136) 33.4 2405 49.7 68.0
1991 35 (1328) 37.9 2880 58.9 81.2
TOTAL 1215 (52515) 43.2 3494 46.7 64.3


a Viabil : (Hatchlings / Total Clutch) X 100
b Hatch: (Hatchlings / Incubated Eggs) X 100

























Table 3,2 Alligator Clutch Size and Age


YEAR N CLUTCH INC VIAB
SIZE TIME RATE


AGE AT
CANDLING


1988 243 44.5 66.0 41.7 21.3

1989 229 43.6 66.7 55.6 19.7

1990 253 43.1 69.4 60.7 19.8

1991 227 43.1 63.6 55.6 17.8

Mean 709 43.2 66.7 57.4 19.1











66













300 -



250-

x- 43.2
n 1049
# 200

C
L
U
T 150
C
H
E
S
100



50-




0 -,
6-10 16-20 26-30 36-40 46-50 56-60 )65

CLUTCH SIZE


Figure 3.2: Frequency distribution of alligator clutch
size for 1049 clutches throughout Florida
















80


60


40



20-


0
Sgrass Phrag Typha Spart Peat Sagit Vine Grass Mud

] Ap42 CA 66 r Ge 54 Gr 170
Je 92 0 ON65 OS 64


Figure 3.3: Alligator Nest Composition











Nesting Material Usage


100%



75%



50%



25%



0%


Sawgrass
Sagitarria


Phragmitee M Typha
E Peat M Grasses


Figure 3.4: Alligator nesting material usage by study
area.


Ap CA Ge Gr Je ON OS


Spartina
Other











NEST PARAMETERS


Ap CA 6 .r JS O O.


Height


Figure 3.5: Mean alligator nest height in inches


Inches


69


20

18

16












37

35

33

31

29

27

25L
Ap


CA Ge Gr Je ON


4- Ambient Temp.


9 Clutch Cavity Temp.


Comparison of ambient with clutch cavity temperatures
of alligator nests from seven study sites in Florida during
1988 1991. Temperatures have been corrected for time of day.
Ap = Lake Apopka
CA = Conservation Area
Ge = Lake George
Gr = Lake Griffin
Je = Lake Jessup
ON = Lake Okeechobee North
OS = Lake Okeechobee South


Figure 3.6 "

























Mean Egg Cavity Depth


inches
12


1 1 .5


11 -


10,5-


10-


9,5-


9-


8.5-


8-


I i I I I I I
Ap CA Ge Gr Je ON OS


Figure 3.7: Depth from the top of the nest to the top of
the egg cavity by study area.


.............. .. ...



























mg cu cm
4000 4000

3600

.......................3 6 0 0

303400
. . .... .. ... ...... . . . . . . . . . ..... ... ..... 3 4 0 0





.............. ....... 2 6 0 0
2000 ............. 3000

-2800


......... .. 2 4 0
...... 2 00
1000-


.2200


0 2000
Ap CA Ge Gr Je ON OS

M0lutch Weight EMlutch Vol,

Figure 3.8: Mean clutch weight and mean clutch volume of
alligator eggs from seven study areas in Florida.























NEST PARAMETERS


25


20-


Ap CA Ge Gr Je ON OS


ECav Depth


M Height


Figure 3.9: Alligator nest mean height and mean depth of cavity
from seven study areas in Florida and 1218 nests.














CHAPTER 4

EMBRYONIC MORTALITY IN FLORIDA ALLIGATORS



Very few studies exist examining the timing and causes of embryonic mortality

in reptiles. Those studies that are available have dealt with mortality as a

consequence of environmental factors (Ewert, 1979; Fowler, 1979; Magnusson, 1982;

Limpus, 1987; Jennings et al., 1988; Deeming and Ferguson, 1991; Horikoshi,

1992). These environmental factors have included predation and flooding of the nest

cavity as well as disturbance by other nesting animals (intra and inter-specific) and

exposure of the eggs. Classification of post-ovipositional mortality has been generally

separated by size of the embryo as an approximate age of development (Blanck and

Sawyer, 1981). This does not provide an accurate description of the timing of

embryonic mortality in crocodilians as variation in size of embryos of the same age is

considerable (Webb and Manolis 1987a,b; Webb et al., 1987; Deeming and Ferguson,

1990).

Post-ovipositional embryonic mortality in the American alligator (Alligator

mississippiensis) in Florida has been reported as high as 62%, and if combined with

infertility as high as 87% (Woodward et al., 1991; see Chapter 3 ).

Post-ovipositional embryonic mortality has been attributed to both external and

internal factors. Externally, depredation and flooding have been cited as major causes

74











of crocodilian mortality (Joanen, 1977; Deitz and Hines, 1980; Magnusson, 1982;

Webb and Manolis, 1987). Flooding of short duration may have an impact on the

eggs by not only lowering the oxygen content of the cavity but by reduction of the

incubation temperature (Magnusson, 1982). Flooding for more than 12 hours has an

adverse effect on viability (Kam Yeong Choy, pers. comm.) whereas, reduced times

and partial submergence of the eggs produces a less definitive outcome. Nesting

medium also influences temperature regulation, gas content and exchange, and

humidity with in the egg chamber. Nest disturbance by ovipositing turtles (Jackson,

1988), the nesting female, or other alligators (Percival et al., 1991) may also

contribute to mortality by damaging, rotating or exposing the eggs.

Production of egg components such as yolk and albumen, and their

incorporation into the egg are internal factors influencing viability. The ability of a

female to produce complete and viable eggs can be compromised by stress either

natural such as overly aggressive males or by man induced (e.g., environmental

contaminants). Eggs recovered from the wild have been found in varying states of

calcification, shapes and instances of eggshell and shell membrane encompassing only

albumen. There is an association between the porosity and thickness of the eggshell

and embryonic viability, as those eggs with fewer pores and thinner shells exhibit a

higher frequency of early embryonic mortality (Wink et al., 1990). The premise

Wink et al., (1990) worked with is more pores represented the ability to transport

water and gases to the embryo in quantities to maintain development. Additionally, a

thicker shell would provide a calcium reservoir for the developing alligator. Content










and metabolism of lipids and proteins during incubation can also affect viability of

reptilian eggs ( Manolis et al., 1987; Noble et al., 1989; Noble et a-I., 1990). The

proper constituents of the yolk and albumen are necessary for viable hatchlings

(Whitehead, 1987; Manolis and Webb, 1987). Utilization of these nutritive

components by the embryo and the appropriate concentrations of gases and water are

necessary for embryonic development (Thompson, pers. comm.).

Distinguishing the potential number of fertile from "infertile" eggs is an initial

step in evaluating causes of embryonic mortality. During this assessment, it is also

necessary to establish when during embryonic development mortality occurs.

Previous reports have provided basic guidelines for determining which

post-ovipositional eggs are fertile. "Fertility" has previously been defined as an egg

having a readily identifiable embryo, and/or the presence of banding, external, visible

evidence of the attachment between the embryonic and egg shell membranes

(Ferguson, 1982). Presence of sub-embryonic fluid was proposed as a technique to

identify fertility in unbanded eggs of saltwater crocodiles (Crocodylus porosus; Webb

et al, 1987). This technique has also been used in alligators as it becomes difficult to

visualize the band by transillumination when embryonic development is 30 35 days

post-ovipositional as the band has progressed to the poles (Ferguson, 1982). Joanen

and McNease, (1975), Magnusson, (1982) and Wink et al., (1990) all suggested that

determination of fertility earlier than evidence of banding was difficult. Thus, in

previous studies all unbanded alligator eggs have been considered infertile. Infertility

rates at the Rockefeller Wildlife Refuge were reported to average 24.6% (Joanen and









77
McNease, 1975). High percentages of infertility (up to 30%) have also been reported

for alligators in Florida (Jennings et a-I., 1987; Woodward et al., in press). Infertility

will be addressed in Chapter 5.

My objective for this study was to test the following hypothesis. Ho:

Mortality is not different among nests, nor among years or among embryonic stages.

This first involved the development of a reliable technique to identify

pre-ovipositional embryos in unbanded alligator eggs. Additionally, it was necessary

to establish an embryonic staging sequence which would be accurate in reference to

alligators in Florida. Finally, using these two methods, an assessment of patterns of

embryonic mortality was obtained.



Methods



Eggs used in this study were collected as described in Chapter 3. Each egg,

determined to be nonviable by candling, was opened and inspected for evidence of an

embryo. Eggs were opened by removing that part of the shell which was determined

at collection to be the uppermost surface. Due to the limited resources and the

volume of eggs present, morphometric analysis of the embryos was not performed.

However, relative size, condition and presence of anomalies were recorded for each

embryo. Eggs which contained dead embryos and were found during the second or

third candling were opened and the age of the embryo recorded. If an age could not

be determined accurately these embryos were classified as unknown mortality and










excluded from analyses. Embryos were placed in four general mortality categories:

unknown, < 20 days post-oviposition, 20-40 days post-oviposition, and > 40 days

post-oviposition (Table 4.1). Subsequent to techniques developed by Masson et al.,

(1991), in utero mortality was added as a category in 1989, 1990, and 1991 (see

Chapter 5).

Analyses of embryonic mortality and viability were done by ANOVA and

either Scheffes or Duncan's New Multiple Range Test post hoc (Sokal and Rohlf,

1981).



Results



Like viability, proportions of mortality in the various categories appear to vary

among years, among study areas and by category within study areas. A total of

52,515 eggs were collected from 1988 to 1991, in which 19,711 dead embryos were

classified as to stage of embryonic development using the criteria found in Table 4.1.

Of the total eggs collected, evidence of embryonic mortality could not be found in

3,635 eggs. Of this total, 2,371 were categorized as infertile whereas. 1364 were

crushed or addled and a determination of fertility was not possible. Early mortality

(including in utero mortality) appeared most prevalent with greater than 75 % of the

observed mortality occurring before day 20 of embryonic development (Table 4.1); a

more detailed analysis revealed that 63 % of the mortality actually occurred prior to

the tenth day of incubation (Figure 4.1). The mean date of egg collection was 19.1











+ 0.4 days post-oviposition as determined by embryonic staging. This is

significantly later than the period of maximal embryonic mortality indicating that the

majority of the observed mortality occurred prior to egg collection. This also

provided additional evidence that the date of egg collection and incubation techniques

had no observable effect on hatching or viability. Substantially less mortality

occurred during late embryonic stages (see Figure 4.1). Mortality occurring in the

20-40 day category represented only 5 % of the total and in the > 40 day category

15 % (Figure 4.1). Seven percent of the total eggs examined were not able to be

categorized due to advanced putrification and damage. The unknown fertility

category contained eggs that were crushed and/or addled, but showed no evidence of

fertility. However, not all crushed or addled eggs were categorized as "unknown"; as

many still contained enough evidence of embryonic development to properly

categorize them. For example, crushed, empty eggs sometimes still showed evidence

of attachment of shell and embryonic membranes.

There was one incidence of human introduced embryonic mortality and that

was for lake George in 1989. Extreme weather conditions during transport of the

eggs from the collection boat to shore artificially increased 20-40 day

post-ovipositional embryonic mortality. Evidence of mechanical damage to the eggs

was determined during the transillumination phase, where the non-viable eggs were

opened and through visual examination it could be verified that the majority (61 %) of

dead embryos had lost their attachment to the inner shell membrane.










There was no relationship between clutch size and the number of embryos

which died in that clutch (Figure 4.1). Embryonic mortality occurred in the same

proportions (Figure 4.2) independent of the study site. The only difference between

sites was the total number of embryonic deaths recorded (Table 4.2).



Discussion



Hatch percent and viability percent were so closely correlated that it was

concluded viability percent was established very early in embryonic development.

This was corroborated by the occurrence of such a high incidence (66.7%) of

mortality during the in utero period and the first five days after oviposition.

Early embryonic mortality accounts for the majority of mortality and

apparently for much of the depressed viability rates reported for wild alligators in

Florida. The pattern of post-ovipositional embryonic mortality within areas, among

areas and between years was consistent with > 75 % of the mortality occurring before

day 20 of incubation. Moreover, 63% of embryonic mortality occurred before day 10

of incubation providing strong evidence that the majority of mortality is not in

response to changes in the incubation medium over time.

Graphically, the pattern of embryonic mortality for the alligator closely

resembles a type III survival curve. A direct correlation between embryos dying and

clutch size cannot be found (Figure 4.2). Further investigation into the relationship

between egg quality and clutch size is necessary to ascertain if there is an optimum










clutch size or if there is an evolutionary compensatory mechanism set up for

overproduction of eggs. Elevated mortality for early embryos could reflect the

embryos developmental pattern from differentiation to a phase of exponential growth.

Environmental influences during early embryonic mortality can not be ruled out.

That is, low temperatures within the nest may increase mortality or at least retard

development. Moisture content and compaction may fluctuate as the female lays on

the nest and degradation commences. However, a reduced oxygen content within the

nest cavity may not have a significant affect since overall oxygen consumption is low

at this time (Thompson, 1989). Temperature fluctuations within the nest early in

embryonic development could contribute to increased mortality. Studies on nest

construction and materials have not addressed relationships to embryonic mortality but

in the broadest sense (Deeming and Ferguson, 1990; Schulte and Chabreck, 1990).

Those studies conducted to date have only analyzed nesting affects on gross viability

and have not examined stages of embryonic mortality and possible relationships to the

nest. With limited sample size it was concluded nests from levees produced more

young but they were also subject to a higher incidence of predation. Additionally,

there was no relationship between the placement of an egg in the clutch cavity and the

likelihood of its hatching. The data from this study suggest that early embryonic

mortality is independent of nesting material. Although the observed mortality

appeared random from clutch to clutch within a lake, between lakes and among years

a similar pattern of high early mortality was consistent.










Late embryos coordinate their emergence (pipping) from the eggs by a

vocalization termed "yerking." If there were embryos developmentally retarded they

may be encouraged to pip prematurely from the vocal stimulation and possibly die due

to their inability to use their lungs for respiration. In order for the neonates to use

their lungs for respiration there must be a sufficient quantity of surfactant present

without which the lungs could not expand properly. A thick egg shell and tough

inner shell membrane undoubtedly inhibit some individuals/clutches to hatch. Without

egg shell degradation, from use of calcium by the embryo in addition to the

possibility of water and carbon dioxide forming a carbonic acid medium, pipping and

hatching are greatly reduced. Degradation of the eggshell producing longitudinal

cracking and therefore weakening of the egg shell does not always occur. With the

primary source of embryonic calcium coming from the egg shell (Ferguson, 1985)

there is an increase in pore size and numbers (Wink et al., 1990) as well as reduction

in shell thickness. Both of these properties allow for more water and gases to be

transported across the shell during a period of high embryonic metabolic activity.

This increase transport of water may cause swelling of the egg resulting in

longitudinal cracking of the shell and providing less resistance for the alligator to

hatch. Thicker egg shells have been noted on eggs from several study sites,

unfortunately logistics did not allow quantification of egg shell thickness nor hatching

success. Pollutants as well as other stressors may influence the female to retain eggs

longer than necessary, thereby increasing calcium deposition and thickening the shell.

The increased rate of late embryonic over mid-incubation mortality could be











indicative of a lack of the correct ratio of lipids being present (Deeming and

Ferguson, 1990).

Mortality patterns for mammals and birds appear similar to that described her

for alligators with a high incidence of early mortality and an increase at birth or

hatching (Wilson, 1980; van Tienhoven, 1983; Putney, 1988). The difficulty in

drawing parallels or differences with mammals and birds is their low egg (embryo)

clutch sizes in comparison to alligators. Studies concentrating on embryonic mortality

within amniotes are constrained by the small clutch size and are devoted to

post-hatching mortality. Other reptiles may exhibit the pattern of mortality observed

in alligators but the majority of reptiles have not been investigated in conjunction with

embryonic mortality.

The reproductive potential of the female alligator appears to be limited by the

amount of embryonic mortality occurring. An increase in embryonic mortality

reduces the number of available neonates for future propagation. Conversely, a

reduction in embryonic mortality increases the neonatal production. Mortality is

constrained to early developmental stages. Embryonic development is known to be at

mid to late gastrulation at oviposition (Ferguson, 1987). Since the majority of

embryonic mortality occurs early a certain amount of that mortality must be present

before oviposition. Moreover, embryonic mortality combined with infertility can be

as high as 87% in Florida (Woodward et al., in press). What percentage of infertility

is really in utero mortality? A distinction of true infertility from in utero mortality

needs to be made. Do alligators have a high incidence of infertility which accounts









84

for a reduced reproductive potential, or do alligators resemble mammals and birds in

having infertility rates of 2-3% (Austin and Short, 1984).













Table 4.1
Embryonic Mortality for Seven Populations of
Alligators in Florida from 1989-1991.


AREA Na EGGSb ZYG- 010d 11-20 20-40 >40 UNKe
MORTC
Ap 36 3202 299 245 68 60 127 294
CA 67 .. 3584 156 185 48 21 71 95
Ge 1 67 4407 315 192 128 29 135 358
Gr 152 13218 779 630 332 108 330 974
Je 75 7834 339 549 343 37 258 333
ON 61 4133 206 173 53 36 130 146
OS 61 3856 328 286 68 13 91 164
Tot. 634 40324 3456 3217 1351 455 1373 3103


aN = clutches
bEGGS = total eggs per area
cZYG-MORT = zygote mortality
do- 10 = Days of Incubation
eUNK = unknown mortality









86

Figure 4.2: Schematic of survivorship curve for alligators





Embryonic
Mortality




Survivorship
Curve


-86 days 10 15 years


Time
















Figure 4.3: Per cent mortality of alligator embryos by lake




% Mortality by Lake
1.07
0.9-
0.8-
0.7-
0"7, Unknown
0.6 >40
0.5 0 20-40
0.4 11-20
0 0-10
0.3 U Zymort
0.2
0.1
0.0
Ap CA Ge Gr Je CN 0










Figure 4.4: Total numbers of eggs and disposition by lake and year
Total numbers of eggs and embryonic mortalities by stages for 1990-1991.

Area N Eggs In Utero < 20 days 20-40 days > 40 days Unknown
Mortality
Ap 36 2252 299 (0.13) 312 (0.14) 60 (0.027) 127 (0.06) 294 (0.13)
CA 67 2265 156 (0.69) 233 (0.10) 21 (0.009) 71 (0.03) 95 (0.04)
Ge 67 3189 315 (0.98) 320 (0.10) 29(0.009) 135(0.04) 358 (0.11)
Gr 152 8963 779 (0.87) 962 (0.11) 108 (0.012) 330 (0.04) 974(0.11)
Je 75 5513 339 (0.61) 892 (0.16) 37 (0.007) 258 (0.05) 333 (0.06)
ON 61 2634 206 (0.78) 226 (0.09) 36 (0.013) 130 (0.05) 146 (0.06)
OS 61 2464 328 (0.13) 354 (0.14) 13 (0.005) 91 (0.04) 164 (0.07)
Total 634 24646 3456 (0.14) 3299 (0.13) 455 (0.018)1 1373 (0.06) 3103 (0.12)


Ap = Apopka, CA = Conservation Area, Ge = George, Gr = Griffin
Je = Jessup, ON = Okeechobee North, OS = Okeechobee South














CHAPTER 5

IN UTERO MORTALITY OF ALLIGATORS IN FLORIDA



Infertility within reptiles has not been thoroughly investigated. In order to

ascertain causes or identify problems stemming from embryonic mortality it is

essential to determine fertility. In many vertebrates causes of infertility have been

attributed to poor sperm, poor sperm transport low sperm count or no sperm/egg

contact due to an abnormal zona pellucida (Austin and Short, 1987).

Distinguishing the potential number of fertile from infertile eggs is the initial

step in evaluating infertility in alligator eggs. Fertile alligator eggs have previously

been defined as those eggs having readily identifiable embryo(s) and/or the presence

of banding external, visible evidence of the attachment between the embryonic and

egg shell membranes (Ferguson, 1982). It was thought that embryos must be present

in some proportion of these unbanded eggs, but methods to detect early development

have only recently been reported. Cardeilhac (pers. comm.) suggests that intact eggs

that did not begin decomposition after 2 weeks at 21 0 C were infertile. Moreover,

the eggs left at 21 0 C may not decompose due to antifungal and antibacterial agents

found within the albumen (Palmer and Guillette, 1991). It may be misleading to

surmise that infertile eggs are those that are not banded, do not rot, or do not produce

sub-embryonic fluid.










Infertility rates at the Rockefeller Wildlife Refuge were reported to average

24.6% (Joanen and McNease, 1975). High percentages of infertility (up to 30%)

have also been reported for alligators in Florida (Jennings et al., 1987; Woodward et

al., in press). Post-ovipositional embryonic mortality in the American alligator

(Alligator mississippiensis) has been reported to be as high as 62% in Florida if

combined with infertility and mortality can be as high as 87% (Woodward et al., in

press). Previous reports have provided basic guidelines for determining which

post-ovipositional eggs are fertile. Presence of sub-embryonic fluid was proposed as

a technique to identify fertility in unbanded eggs of saltwater crocodiles (Crocodylus

porosus; Webb et al., 1987). Identification of fertility by the presence of

sub-embryonic fluid has also been used in alligators as it becomes difficult to visualize

a band by transillumination when embryonic development is 30 35 days

post-ovipositional because the band has progressed to the poles (Ferguson, 1982).

However, the sub-embryonic fluid of C. porosus was found in larger volumes than in

alligators (2 ml to less than 1 ml respectively, G. Webb pers. comm.). Joanen and

McNease (1975), Magnusson (1982) and Wink et al. (1990) all suggested that

determination of fertility earlier than evidence of banding was difficult. Thus, all

unbanded alligator eggs have generally been considered infertile.

My objective for this study was to test the following hypothesis. Ho:

Infertility is not the cause of poor alligator egg viability in Florida ecosystems. Also,

an objective of this study was to develop a reliable technique to identify

pre-ovipositional embryos in unbanded alligator eggs and to determine what











proportion of the unbanded eggs that, previously would have been reported as

infertile, did in fact have evidence of embryonic development.



Methods



Alligator eggs for this study were collected from natural nests as part of a

larger study examining egg viability (see Chapter 3). Those eggs classed as infertile

in 1988 (N = 2,115) were examined macroscopically as were 376 eggs from 6

different farms in 1989. All of the farm eggs were determined to be non-viable by

the farmers and transported to our laboratory. In prior years all these eggs would

have been classed as infertile.

Each egg determined to be non-viable by candling was opened and inspected

for evidence of an embryo. Eggs were opened by removing that part of the shell

which was determined at collection to be the uppermost surface. Exposing the

dorsum of the egg, a small, 2-4 mm white disc was often visible, between the dorsal

aspect and one pole of the yolk (Figure 5.1). A representative sample of 130 eggs

containing a disc were preserved for microscopic examination (115 wild and 15 farm

eggs). Prior to fixation the albumen was dissected away from the yolk and each yolk

was fixed in 10% neutral buffered formalin (10% NBF) for histological examination.

After fixation, the disc and a small (1-2 cm) section of surrounding yolk were

excised, dehydrated in an ascending alcohol series, paraffin embedded, serially

sectioned at 10 um and mounted (Humason, 1979). Every third slide was stained










with Alcian blue then counter stained with hematoxylin and eosin for examination

under light microscopy (Humason, 1979). The remaining slides were kept for future

analysis.



Results



Microscopic examination of the fixed discs revealed, in all cases, cells that

exhibited characteristics of mitosis. These discs were areas of previous active

embryonic mitosis and were similar to blastodiscs described for other amniote species

having megalecithal eggs (Gilbert, 1988). Analysis indicated blastodiscs vary in size

and generally contained from 80 to 200 cells and exhibited differing stages of mitotic

activity. Since alligator eggs are oviposited after reaching mid to late gastrulation

(Ferguson, 1981), it was determined that mortality at the blastodisc stage occurred in

utero and the term "in utero mortality" was used for embryonic mortality at this

stage.

The wild eggs (Percival et al., 1991) and farm eggs showed the same general

patterns of embryonic mortality; that is greater than 96% of the non-viable eggs

showed evidence of embryonic activity. Of the 376 farm eggs that could be analyzed

only six of the 376 did not have a macroscopically visible blastodisc present.

However, those 6 eggs were among 67 eggs which were kept at incubation

temperatures before they could be delivered to the research facility for analysis and

their yolk surfaces were covered in bacteria and/or fungus. The remaining 61 eggs










had an identifiable blastodisc. Therefore, the most appropriate status for those 6

specimens is unknown, not infertile.



Discussion



Female alligators are reported to retain their eggs for approximately 3 weeks

after ovulation in order to secrete and lay down the albumen, shell fiber and egg shell

prior to oviposition (Lance, 1987). There has been little, if any, research completed

during this time on embryonic activity or the viability and status of the egg.

One of the indices used to distinguish viable eggs from non-viable is the

presence of a "band" on the egg. The band appears as an opaque chalky spot on the

uppermost surface of the egg and the embryo and its attachment is always directly

beneath it. The band progresses ventrally around the waist and toward the poles with

development. It is known that the initial spotting occurs within 24 hours of natural

oviposition and progresses around the waist of the egg completing the band around the

circumference within 3 days (Ferguson, 1982). A pearl white band and light rose tint

of the non-banded areas (by transillumination) aid in determining viable from

non-viable eggs. As embryonic development progresses, the band traverses the length

and circumference of the egg and the diminishing area becomes a darker red.

Therefore, "dead" eggs can be readily visualized and removed during incubation.

Presence or absence of factors within the egg itself also influence viability.

Demands on the female for calcium mobilization and stress associated with mating










(for example: territorial defense and overly aggressive males) may reduce the

female's ability to produce adequate eggs. There is an association between the

porosity and thickness of the eggshell, with the eggs having fewer pores and thinner

shells having higher incidence of early embryonic mortality (Wink et al., 1990).

Content and metabolism of lipids during incubation may affect viability of reptile eggs

(Thompson, 1981; Manolis et al., 1987; Noble et al., 1989; Noble et al., 1990). The

proper constituents of the yolk and albumen are necessary for viable hatchlings

(Jenkins, 1975; Manolis and Webb, 1987). Utilization of these components by the

embryo and the appropriate concentrations of gases and water are necessary for

embryonic development (Thompson, 1981).

Presence of blastodiscs at one pole and at the uppermost surface suggest a time

frame for migration. The movement of the embryo from the pole to the uppermost

surface is enabled by sub-embryonic fluid (Manolis and Webb, 1987). Although,

blastodiscs were found at either the uppermost surface or a pole in eggs that did not

band, suggesting that mortality of zygotes occurs at different stages of development.

Early embryonic mortality accounts for the majority of mortality and for much

of the depressed viability rates of wild Florida alligators. The proportion of all

embryonic mortality occurring before day 10 is 75 %. Furthermore, in utero mortality

accounts for 35% of the overall mortality. Although based on a smaller sample size,

a high incidence of in utero mortality appears to have occurred on alligator farms.

The observation of in utero mortality defines an area for concentrated research need;

we must determine what relationships exist between low hatch rates and physiological










and/or ecological correlates. Although these data suggest the problem is common

throughout the state of Florida further sampling is necessary to evaluate the incidence

in varying wetland habitats. Additional embryological research is needed to further

define the stages of in utero mortality. Ultimately, the causes of in utero and

post-ovipositional mortality must be ascertained to affect management of either wild

or captive stocks.

The incidence of infertility in wild alligators is not known, but can now be

assumed to be less than 4% of the total eggs. Entire clutches have been found that

were not banded, but of the 1259 clutches collected from this study, every clutch had

at least one egg with evidence of embryonic development. Captive animals have

produced eggs without the presence of a male and presumably infertile as no evidence

of sperm storage has been observed (A. Woodward, pers. comm.). These females

oviposited eggs in their enclosure and had no nesting opportunity. It is not known if

they would construct a nest and oviposit the eggs. Spontaneous mitotic activity may

account for a proportion of the eggs classified as in utero mortality. Therefore, some

percentage of the eggs found in the wild may be truly infertile and some may have

been misidentified as in utero mortality.

The proportion and identification of in utero mortality within most vertebrate

classes is not known. Estimates for mammals have been given generally for domestic

species. These estimates of in utero mortality have been from 25-30% in horses (M.

LeBlanc, pers. comm.). Pregnancy rates as low as 15 % in domestic cattle have been