<%BANNER%>

Conservation Genetics of the Florida Manatee, Trichechus manatus latirostris

Permanent Link: http://ufdc.ufl.edu/UFE0021345/00001

Material Information

Title: Conservation Genetics of the Florida Manatee, Trichechus manatus latirostris
Physical Description: 1 online resource (190 p.)
Language: english
Creator: Pause, Kimberly Christ
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: conservation, diversity, dna, florida, genetic, identification, individual, manatee, microsatellite, parentage, pedigree, population, structure
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Small population sizes, a low reproductive rate, and habitat degradation have resulted in Florida manatees (Trichechus manatus latirostris) being a protected species for many decades. Previous genetic studies using allozymes and mitochondrial DNA have been unable to resolve the population structure within Florida. This study used nuclear microsatellite DNA markers to visualize population structure at a fine scale. Eleven new polymorphic microsatellite loci were identified for Florida manatees. These loci were used in combination with seven previously published loci to create a panel of markers for individual identification. The individuals? unique multilocus genotypes generated from this panel of markers were used to create a DNA database, the Manatee Individual Genetic-identification System (MIGS). The MIGS was developed as a model to be used in the future for capture-recapture population modeling applications in concert with the Manatee Individual Photo-identification System (MIPS). The multilocus genotype data were also used to examine the fine-scale population structure of manatees in Florida. Manatees have the ability to travel great distances within their lifetime, as well as migrate long distances within a single year. Each winter, Florida manatees migrate to natural and artificial warm-water sources to thermoregulate. Population genetic analyses suggest that there is subtle, but statistically significant, population structure that corresponds to the previously designated management units. This subtle signal of population structure is evident in the winter season, but not in warmer seasons. The genetic results correspond to the winter site fidelity patterns observed by photo-identification, telemetry, and mortality studies. Reconstruction of a captive Florida manatee pedigree was attempted using the suite of 18 markers. Family units were clearly distinguishable using allele sharing methods. Maximum likelihood methods were used to estimate family relationships, but were not completely successful. Complete pedigree reconstruction will likely be possible with the inclusion of a few additional polymorphic microsatellite loci. The genetic studies on Florida manatees to date provide a thorough analysis of the population structure and genetic variability of manatees in Florida using three different types of molecular markers. Continued genetic monitoring of the species should focus on pedigree reconstruction to create baseline data for examining the level of inbreeding and genetic health of the Florida manatee population.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kimberly Christ Pause.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: McGuire, Peter M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021345:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021345/00001

Material Information

Title: Conservation Genetics of the Florida Manatee, Trichechus manatus latirostris
Physical Description: 1 online resource (190 p.)
Language: english
Creator: Pause, Kimberly Christ
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: conservation, diversity, dna, florida, genetic, identification, individual, manatee, microsatellite, parentage, pedigree, population, structure
Genetics (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Small population sizes, a low reproductive rate, and habitat degradation have resulted in Florida manatees (Trichechus manatus latirostris) being a protected species for many decades. Previous genetic studies using allozymes and mitochondrial DNA have been unable to resolve the population structure within Florida. This study used nuclear microsatellite DNA markers to visualize population structure at a fine scale. Eleven new polymorphic microsatellite loci were identified for Florida manatees. These loci were used in combination with seven previously published loci to create a panel of markers for individual identification. The individuals? unique multilocus genotypes generated from this panel of markers were used to create a DNA database, the Manatee Individual Genetic-identification System (MIGS). The MIGS was developed as a model to be used in the future for capture-recapture population modeling applications in concert with the Manatee Individual Photo-identification System (MIPS). The multilocus genotype data were also used to examine the fine-scale population structure of manatees in Florida. Manatees have the ability to travel great distances within their lifetime, as well as migrate long distances within a single year. Each winter, Florida manatees migrate to natural and artificial warm-water sources to thermoregulate. Population genetic analyses suggest that there is subtle, but statistically significant, population structure that corresponds to the previously designated management units. This subtle signal of population structure is evident in the winter season, but not in warmer seasons. The genetic results correspond to the winter site fidelity patterns observed by photo-identification, telemetry, and mortality studies. Reconstruction of a captive Florida manatee pedigree was attempted using the suite of 18 markers. Family units were clearly distinguishable using allele sharing methods. Maximum likelihood methods were used to estimate family relationships, but were not completely successful. Complete pedigree reconstruction will likely be possible with the inclusion of a few additional polymorphic microsatellite loci. The genetic studies on Florida manatees to date provide a thorough analysis of the population structure and genetic variability of manatees in Florida using three different types of molecular markers. Continued genetic monitoring of the species should focus on pedigree reconstruction to create baseline data for examining the level of inbreeding and genetic health of the Florida manatee population.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kimberly Christ Pause.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: McGuire, Peter M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021345:00001


This item has the following downloads:


Full Text

PAGE 1

1 CONSERVATION GENETICS OF THE FLORIDA MANATEE, Trichechus manatus latirostris By KIMBERLY CHRISTINA PAUSE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Kimberly Christina Pause

PAGE 3

3 To my little brother, Bill. You can accomplish anything that yo u put your mind to! I love you!

PAGE 4

4 ACKNOWLEDGMENTS The McGuire Lab group has been extremely help ful throughout my graduate school career. They have listened to countless presentations, pr oofread innumerable versions of documents and posters, and have given much support to me thro ugh the years. My mentor and advisor, Dr. Peter McGuire, has been very pr otective of his graduate students and our dissertation projects. He also has made it possible for us to attend se veral meetings, seminars, and workshops so that we could make valuable contac ts and learn new, exciting things. Our many talks always reminded me why I chose this projectto help the manatees. Maggie Ke llogg, has not only been supportive as a fellow graduate stude nt, but as a close friend. I am also very thankful for Frank Bouchard who volunteered his time to isolate some of the samples for this study. Coralie Nourisson was involved in the primer screening pr ocess, and also as a collaborator and friend. I am greatly indebted to the USGS, Sirenia Project. They provided all of the manatee tissue and blood samples for my project, and also much of the projects funding and my financial support. All samples were collected and anal yses were performed under the USGS, Sirenia Project permit (USFWS Wildlife Research Permit MA791721/4). Th ey allowed me to tag along for field work in Puerto Rico and Floridae xperiences that I will cherish forever! Bob Bonde has been integral to this projec t. It was through his vision that the genetics project was started with cookies and other archived samples almost two decades ago. I am very thankful that he has been one of my mentors and sh ared his expertise with me. I am also very thankful for Dr. Timothy King with the USGS Leetown Science Ce nter, who gave me many helpful suggestions for analysis of the microsate llite data and on a manuscript. The UF Marine Mammal and Aquatic Animal Health Program provided the first few years of financial support for the project.

PAGE 5

5 The UF ICBR Genetic Analysis Laboratory generously allowed me to utilize the laboratory space and equipment to complete this pr oject. I am very thankful for Ginger Clark, who has helped me throughout my graduate career. Other labs in the UF ICBR have also been of great assistance to me during this time. The Se quencing Core processed a ll of my microsatellite sequences, and I have attended many useful wo rkshops held by the Education Core. I am also thankful for Dr. Tara Paton and Si mone Russell at the Genetic Analysis Facility at the Hospital for Sick Children in Ontario, Cana da. They processed the bulk of my samples for fragment analysis with great speed and excellent quality. Pandora Cowart of the UF Health Science Ce nter Information Tec hnology Training Center was very helpful with the development of th e MIGS database through her Microsoft Access courses and one-on-one office hours. My committee members have been very help ful and supportive thr oughout my education. I would like to thank Dr. Madan Oli for all of his helpful sugg estions; Dr. Roger Reep for his encouragement; Dr. Margaret Peggy Wallace for al l of her assistance, e xpertise, and a listening ear; and especially Cathy Beck from the USGS, Sirenia Project for the many occasions where we discussed MIPS and genetics. Cathys help with editing my dissertation was invaluable, and she has been very encouraging every step of the way! Our graduate secretary, Joyce Conners, has been extremely helpful throughout the time that I have been at UF. She always keeps her students up on deadlines a nd has helped me in any way that she could. Many thanks go to my IDP friends, who have shared in all of the trials and tribulations of graduate school with me, especially Jenny Joseph Heather Rossi, Michelle and Patrick Thiaville,

PAGE 6

6 Chriss Gavegnano, and my adopted lab mates from the Gulig Lab who have kept my spirits up and shared the good and bad times with me. I am so happy that I have been able to ma intain my long-distance friendships throughout this process. I am grateful to still have Tracy Wheeler, Jackie Owen, Heather Townsend, and Trisha Spears in my life. I am glad that we have remained friends throughout the years. Thanks go to my boyfriend, Matt Tucker, who has listened to me ramble on and on, helped me with advice, and has been there for me when I was at my worst. He has been a tremendous support for me. Many thanks go to my family who have always been there to listen to me when I needed someone to talk to, especially my Mom and Da d, and my step-parents, Sheila and Don. Thank you for the extra financial assist ance and emotional suppor t in my time of cri tical need. Also, thank you for understanding that I could not vi sit as often as I would have liked.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES................................................................................................................ .......11 ABSTRACT....................................................................................................................... ............12 CHAPTER 1 INTRODUCTION..................................................................................................................14 The Biology of the Florida Manatee.......................................................................................14 Studies of the Genetic Divers ity of the Florida Manatee.......................................................16 Specific Aims and Hypotheses...............................................................................................20 Specific Aim 1: Individual Identif ication of Florida Manatees.......................................20 Specific Aim 2: Fine-scale Populati on Structure of Florida Manatees...........................21 Specific Aim 3: Pedigree Analysis..................................................................................21 2 CHARACTERIZATION OF POLYMO RPHIC MICROSATELLI TE DNA MARKERS AND GENETIC INDIVIDUAL IDENTIFICATION OF FLORIDA MANATEES.............25 Introduction................................................................................................................... ..........25 Materials and Methods.......................................................................................................... .26 Sample Collection...........................................................................................................26 DNA Isolation.................................................................................................................26 Related Species Primer Screening...................................................................................27 Manatee-specific Micr osatellite Identification and Screening........................................28 Genotyping..................................................................................................................... .32 Identity Determination.....................................................................................................33 Development of the Manatee Individual Genetic-identification System (MIGS)...........34 Results........................................................................................................................ .............35 Related Species Primer Screening...................................................................................35 Manatee-specific Micros atellite Screening.....................................................................35 Identity Determination.....................................................................................................36 Discussion..................................................................................................................... ..........37 3 SEASONAL POPULATION GENETIC STRUCTURE OF MANATEES IN FLORIDA........................................................................................................................ .......48 Introduction................................................................................................................... ..........48 Materials and Methods.......................................................................................................... .50 Sample Collection and DNA Isolation............................................................................50 Genotyping..................................................................................................................... .51

PAGE 8

8 Statistical Analyses..........................................................................................................51 Genetic Diversity Statistics......................................................................................51 Individual Level Analyses........................................................................................52 Population Level Analyses.......................................................................................55 Results........................................................................................................................ .............56 Genetic Diversity Statistics.............................................................................................56 Individual Level Analyses...............................................................................................57 Population Level Analyses..............................................................................................59 Discussion..................................................................................................................... ..........60 4 FLORIDA MANATEE PEDIGREE RECONSTRUCTION.................................................77 Introduction................................................................................................................... ..........77 Materials and Methods.......................................................................................................... .78 Sample Collection and DNA Isolation............................................................................78 Genotyping..................................................................................................................... .78 Statistical Analyses..........................................................................................................78 Results........................................................................................................................ .............79 Discussion..................................................................................................................... ..........82 5 CONCLUSIONS AND IMPLICAT IONS FOR MANAGEMENT.......................................90 Conclusions.................................................................................................................... .........90 Implications for Management.................................................................................................91 Anthropogenic Threats That May Affect the Population Genetic Structure...................91 Future Directions for Florida Manatee Genetics.............................................................93 Genetics and Photo-identification............................................................................93 Population Genetic Stru cture of Manatees...............................................................94 Genetic Health of the Fl orida Manatee Population..................................................95 APPENDIX A LIST OF ABBREVIATIONS.................................................................................................99 B LOCALITY INFORMATION.............................................................................................103 C MANATEE INDIVIDUAL GENETIC-IDENTI FICATION SYSTEM (MIGS) HELP FILE........................................................................................................................... ...........124 D SUPPLEMENTAL INFORMATION..................................................................................144 LIST OF REFERENCES.............................................................................................................177 BIOGRAPHICAL SKETCH.......................................................................................................190

PAGE 9

9 LIST OF TABLES Table page 2-1. Microsatellite loci character ized for related species selected from the literature..............42 2-2. Characteristics of the eleven polymorphic microsatellite loci developed for the Florida manatee ( T. m. latirostris ).....................................................................................43 2-3. Cross-species amplification of the pol ymorphic primers developed for Florida manatees ( T. m. latirostris )................................................................................................44 2-4. Optimized conditions for seven polymorphic primer pairs previously published for Florida manatees............................................................................................................... .44 2-5. Probabilities of identity as estimated by GENECAP...........................................................46 2-6. Probability of two randomly chosen unrelated individuals or fullsiblings having two or more genotypic mismatches gi ven genotypic information from L loci.........................46 2-7. Diversity statistics for the loci as calculated by MM-DIST................................................47 3-1. Manatees used for the calculation of diversity and F-statistics.........................................67 3-2. Manatees used in sub-sampled dataset for the winter season individual genetic distance-based Nei ghbor-Joining trees..............................................................................68 3-3. Diversity statistics for the 18 lo ci on the set of 362 manatees...........................................69 3-4. Average measures of genetic variation over 18 microsatellite loci for Florida manatees....................................................................................................................... ......70 3-5. Average measures of genetic variation over 18 microsatellite loci for Florida manatees....................................................................................................................... ......70 3-6. Average measures of genetic variation over 18 microsatellite loci for Florida manatees....................................................................................................................... ......70 3-7. STRUCTURE results for inferring the number of distinct population subgroups for all Florida samples and for the dataset subdi vided by season of sample collection...............72 3-8. Pairwise FST and RST values among the management unit populations.............................76 3-9. Pairwise FST and RST values between Floridas east and west coast populations..............76 4-1. Pedigree reconstruction results usi ng the maximum likelihood approach.........................87 4-2. Identification information for samples from captive manatees.........................................88

PAGE 10

10 B-1. Sample information..........................................................................................................103 D-1. Diversity statistics for the 362 manatees ove r all loci for when data are grouped as the east and west coast populations.................................................................................153 D-2. Diversity statistics for all loci over th e 362 manatees where data are grouped as management unit populations..........................................................................................154 D-3. Diversity statistics for all loci for when da ta are grouped as the wi nter east and west coast populations..............................................................................................................164 D-4. Diversity statistics for all loci for when data are grouped as winter management unit populations.................................................................................................................... ...165 D-5. Diversity statistics for all loci for when data are grouped as the summer east and west coast populations..............................................................................................................175 D-6. Diversity statistics for all loci for wh en data are grouped as summer management unit populations............................................................................................................... .176

PAGE 11

11 LIST OF FIGURES Figure page 1-1. Range of the Florida manatee............................................................................................22 1-2. Phylogenetic relationship of the Si renia to other mammalian taxa...................................23 1-3. Mitochondrial haplotype distributi on for the West Indian manatee..................................23 1-4. Florida manatee management units as established by the USFWS...................................24 2-1. Scar pattern accumulation of a Florida manatee over time................................................40 2-2. Tissue sampling method for free-ranging manatees..........................................................41 2-3. MIGS database organization..............................................................................................45 2-4. Diagrammatic representation of the MI GS DNA database structure and match process. ...................................................................................................................... ........47 3-1. Warm-water aggregation..................................................................................................67 3-2. Florida manatee population distribution am ong regions used to sub-sample the dataset. ...................................................................................................................... ........68 3-3. STRUCTURE bar plots..........................................................................................................71 3-4. Neighbor-Joining trees for samples from i ndividuals collected during the winter season.. ...................................................................................................................... ........73 3-5. Neighbor-Joining trees for samples from i ndividuals collected during the summer season.. ...................................................................................................................... ........74 3-6. Neighbor-Joining trees for populations coll ected during the different seasons.................75 4-1. Herd of mating manatees...................................................................................................85 4-2. Neighbor-Joining tree based on the propor tion of shared alleles (PSA)...........................86 4-3. Actual pedigrees of captive manatees................................................................................89

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CONSERVATION GENETICS OF THE FLORIDA MANATEE, Trichechus manatus latirostris By Kimberly Christina Pause August 2007 Chair: Peter McGuire Major Department: Medical Sciences -Genetics Small population sizes, a low reproductive rate, and habitat degradati on have resulted in Florida manatees ( Trichechus manatus latirostris ) being a protected species for many decades. Previous genetic studies usi ng allozymes and mitochondrial DNA have been unable to resolve the population structure within Florida. This st udy used nuclear microsatellite DNA markers to visualize population structure at a fine scale. Eleven new polymorphic microsat ellite loci were identified for Florida manatees. These loci were used in combination with seven previo usly published loci to create a panel of markers for individual identification. The individuals unique multilocus genotypes generated from this panel of markers were used to create a DNA database, the Manat ee Individual Geneticidentification System (MIGS). The MIGS was devel oped as a model to be used in the future for capture-recapture population modeli ng applications in concert with the Manatee Individual Photo-identification System (MIPS). The multilocus genotype data were also used to examine the fine-scale population structure of manatees in Florida. Manatees have the ability to travel great distances within their lifetime, as well as migrate long distances within a single year. Each winter, Flor ida manatees migrate to natural and artificial warm-water sources to thermoregulate. P opulation genetic analyses suggest

PAGE 13

13 that there is subtle, but statis tically significant, population st ructure that corresponds to the previously designated management units. This su btle signal of population st ructure is evident in the winter season, but not in warmer seasons. Th e genetic results correspond to the winter site fidelity patterns observed by photo-identific ation, telemetry, and mortality studies. Reconstruction of a captive Fl orida manatee pedigree was attempted using the suite of 18 markers. Family units were clearly distingu ishable using allele sharing methods. Maximum likelihood methods were used to estimate fam ily relationships, but were not completely successful. Complete pedigree reco nstruction will likely be possibl e with the inclusion of a few additional polymorphic microsatellite loci. The genetic studies on Florida manatees to date provide a thorough analysis of the population structure and genetic variability of mana tees in Florida using th ree different types of molecular markers. Continued genetic monito ring of the species should focus on pedigree reconstruction to create baseline data for examin ing the level of inbreeding and genetic health of the Florida manatee population.

PAGE 14

14 CHAPTER 1 INTRODUCTION The Biology of the Florida Manatee The Florida manatee ( Trichechus manatus latirostris ) is a subspecies of the West Indian manatee ( T. manatus ) that is indigenous to the coastal wate rs and rivers of Florida, the northern limit of the species winter range (Figure 1-1). Manatees were originally listed in 1967 by the Endangered Species Preservation Act and are protected by the Endangered Species Act of 1973 and the Marine Mammal Protection Act of 1972. They are currently listed at the federal level as endangered, although their status is under re view (FWC 2006; USFWS 2007). The minimum population size based on synoptic aerial survey data is estima ted to be approximately 3,000 individuals (Florida Fish and Wild life Research Institute (FWRI) 2007). West Indian manatees belong to the order Sirenia, which includes the West African manatee ( Trichechus senegalensis ), the Amazonian manatee ( Trichechus inunguis ), the dugong ( Dugong dugon ) and the extinct Stellers sea cow ( Hydrodamalis gigas ). Sirenians closest terrestrial relatives are the elepha nt and hyrax (Figure 1-2) (Murphy et al. 2001, Kellogg et al 2007). Sirenians are a unique group with many beha vioral and physiologi cal features that distinguish them from other marine mammals. Th e main social unit for manatees is a female (cow) and her calf. The cow-calf unit remains int act for approximately one to two years. During this time the calf learns locations of feeding and resting areas and travel routes to winter warmwater refuges. Other social groupings, such as mating herds and warm-water aggregations, appear to be transient (R eynolds III and Odell 1991). Sirenians are the only herbivorous marine mamm als. Extant sirenians have a tropical to subtropical distribution, and this restriction to warm waters refl ects their inability to maintain

PAGE 15

15 body heat in cooler temperatures. They have a low metabolic rate, only 15-22% of what would be expected for a terrestrial mammal of similar size (O'Shea and Reep 1990). This is a necessary adaptation to minimize energy expenditure due to their low-energy diet, including seagrasses such as turtle grass ( Thalassia testudinum ), manatee grass ( Syringodium filiforme ), and shoal grass ( Halodule wrightii ). They also consume freshwater plants such as Hydrilla verticillata water hyacinth ( Eichhorina crassipes ), and have been known to consume leaves from overhanging terrestrial branches and other ba nk vegetation, as well as algae (Reynolds III and Odell 1991). Substantial quantities of these low-energy foods are required for survival. Sirenians have adapted a large body size, which allo ws for the processing of such large quantities of plant material (O'Shea and Reep 1990). Their fibrous, herbivorous diet has resu lted in other unique adaptations, including hindmolar progression. In this process, the mola rs move forward as the tooth surface wears, and as new molars erupt at the back of each tooth row. As the ch ewing process stimulates forward movement of the teeth, the fibrous plants and in cidentally ingested sand wear down the teeth. Manatees can produce an unlimited number of teeth, and hindmolar progression continues throughout their lifetime (Domning 1983). The manatee brain has also evolved to meet th e needs of its herbivorous lifestyle. When their brains were first examine d, it was thought that manatees were not as intelligent as other mammals. This is because the surface of the manatee brain has very few convolutions, unlike the brains of dolphins and humans (O'Shea and Reep 1990). Recent studies of manatee neuroanatomy established that manatees have a large cerebral cortex, a region of the brain dedicated for higher order information proces sing (Reep and Bonde 2006). This region is abnormally thick in manatees and exhibits la yering patterns comparable to carnivores and

PAGE 16

16 primates (Marshall and Reep 1995; Reep et al. 1989). The cerebral cort ex is thought to be the area of the brain that acts as the repository for many aspects of long-term memory (Purves et al. 2001). It is necessary for manatees to rememb er where food, fresh water, and winter warmwater refuges are located year after year. This un ique feature is related to the manatees winter site fidelity and other dis tinct migratory patterns. The sensory system of manatees has also evolved to complement their herbivory and environment. It has been determined through beha vioral and histological studies that manatees have poor eyesight (Bauer et al. 2003; Harper 2004). Corneal va scularization has been observed in the eyes of all age classes of manatees. In humans, corneal vascular ization is a pathologic condition associated with an infection or other in jury to the eye. Although manatees have poor eyesight, they have compensated evolutionarily with tactile hairs that are located along their bodies. These hairs are hypothesized to act analogously to the lateral line of fishes and allow the manatee to sense movement in the water (Reep et al. 2002). Manatees often occupy turbid, murky waters. Primarily relying on eyesight in these conditions coul d be detrimental. Additionally, manatees spend a la rge proportion of time resting. Th e ability to sense movement in the surrounding areas while at rest is beneficial for de fensive or elusive maneuvering. Studies of the Genetic Diversity of the Florida Manatee The genetic diversity of the Florida ma natee population has been studied on the chromosomal level and by using three very differe nt genetic markers: allozymes, mitochondrial DNA, and microsatellite DNA. These methods have provided a survey of the genetic variability and structure of the manatee population. Chromosomal analysis was very popular method of genetic anal ysis before the advent of molecular genetic techniques. Standard karyotyp ing is the process of visualizing and organizing an individuals chromosomes. Cells in metapha se are placed onto a microscope slide, and are

PAGE 17

17 forced to burst, spreading the chromosomes. The chromosome preparations are then stained through a variety of available methods and visual ized through microscopy. Differences can exist from individual to individual in the form of chromosome duplicati ons, translocations, and deletions, but these changes typical ly suggest a disease state (Str achan and Read 1999). Isolated populations can also show differe nt arrangements of chromosome segments or banding patterns (Avise 2004). Karyotyping is not a very efficient method for a ssaying population variability, and is more useful for answering evolutionary ques tions. Additionally, sample must be harvested from living animals or immediat ely following euthanasia or na tural death to produce blood or tissue cultures for cytogenetic preparations. This is not feasible for many protected or endangered species. Allozymes, also known as isozymes, are pr otein isoforms that can vary between individuals and populations, and can be used to infer genetic divers ity. Allozymes are useful for measuring genetic diversity among individuals and populations because of their unique electrophoretic mobilities, which reflect gene sequence variation. Allozyme-based analyses of genetic variation, while useful, have several drawbacks that in clude the following: large amounts of relatively fresh tissue are required to obtain sufficient am ounts of active protein for the analyses; the analyses provide an indirect measur e of variation based on sets of genes that are likely under selection; and the mutations underlyi ng the allozyme changes are unresolved (Avise 2004). For these reasons, conser vation geneticists have shifted to direct analysis of DNA using Polymerase Chain Reaction (PCR) techniques such as sequencing and microsatellite fragment analysis. PCR requires a small quantity of samp le and can be used to amplify DNA isolated from non-invasive samples (hairs, feces) and degraded tissues.

PAGE 18

18 Genetic diversity may be investigated mo re directly using mitochondrial DNA sequence data. Mitochondrial DNA analyses can provide insight into th e structure of populations, but these analyses also have limitations. Mitochond rial DNA is inherited maternally, which limits the information gained to female dispersal or migr ation. It is possible th at the males and females of a species have completely different patterns of dispersal. Analysis of mitochondrial DNA can help researchers infer populati on structure, but other methods should be used to confirm the findings (Bowen et al. 2005; Frankham et al. 2002). Higher-resolution, highly variable, nuclear DNA markers, such as microsatelli tes, might be more useful indi cators of genetic variation in isolated populations like the Florida manatee. Nuclear microsatellite DNA markers are short re peated sequences that are typically two to nine basepairs (bp) in length. Microsatellites are also known as short tandem repeats (STRs) or simple sequence repeats (SSRs). The number of repeats can vary am ong individuals and is thought to arise from errors in DNA replicati on or unequal crossing-ove r (Strachan and Read 1999). The high variability of microsatellites ma kes them ideal markers for endangered species, which are likely to have reduced genetic divers ity due to small, fragmented populations (Wright and Bentzen 1994). Additionally, microsatellites are designed to be short in length and easily amplified using PCR. This feature makes them especially useful in cases where the DNA sample may be of low quality or quantity. Microsatellites are frequently used in conservation genetics to answer population genetic questions for ma ny endangered and domestic species (Bowen et al. 2005; DeSalle and Amato 2004; King et al. 2006). Florida manatees have 48 chromosomes (o r 24 pairs of homologous chromosomes) and follow the standard mammalian XY sex determination system (Gray et al. 2002; White et al. 1976; White et al 1977). Gray et al examined chromosomes from ten captive and wild

PAGE 19

19 individuals, but did not note any significan t differences or chromosomal abnormalities among them (2002). Karyotyping of individual blood prepar ations from additional Florida manatees has not identified any noteworthy diffe rences among individuals or popul ations (K. C. Pause and M. E. Kellogg, University of Florida, Unpublished Data). Allozymes were used to study the Florida manatees genetic diversity identified 10 polymorphic allozyme loci from 24 total loci (41. 67% polymorphic) in a sample of 59 manatees (McClenaghan and O'Shea 1988). These authors reported that the manatee population did not suffer from an excess of homozygosity and was li kely not suffering from inbreeding depression. The genetic distances calculated among the populations from the a llozyme data di d not correlate with the manatees geographic proximity. This indicated that the popula tion structure may not correspond with the geographic units identified for manatees. The genetic diversity of Florida manatees has also been examined using mitochondrial DNA (Garcia-Rodriguez et al. 1998; Vianna et al. 2006). Analyses of sequences amplified from a 410 bp region of the highly variable D-loop were used to determine the relationship between the two subspecies of West Indian manatees, the Florida manatee ( T. m. latirostris ) and the Antillean manatee ( T. m. manatus ). The Florida manatee popul ation possessed one haplotype, while the Antillean manatee population possessed greater haplotypic diversity throughout the range (Figure 1-3) (Garcia-Rodriguez et al. 1998; Vianna et al. 2006). Based on these findings, it was suggested that there were two separate migration events or iginating from South America: one toward Mexico, and one toward Puerto Rico and Florida. Th e single haplotype identified in the Florida population was hypothesized to be the result of a founder effect from a small number of migrants that colonized the norther n limit of the range (Garcia-Rodriguez et al. 1998).

PAGE 20

20 A total of 14 polymorphic microsatellite loci wa s identified for West Indian manatees that were capable of amplifying across sirenian taxa (Garcia-Rodriguez et al. 2000). These loci had low allelic diversity in Florida manatees, only ei ght of which were identified as polymorphic for the subspecies. Therefore, additional polymorphic microsatellite loci for Florida manatees have been characterized to address the spec ific aims and hypotheses of this study. Specific Aims and Hypotheses Specific Aim 1: Identify a Panel of Polymo rphic Microsatellite Markers for Individual Identification of Florida Manatees. Currently, individual identific ation of Florida manatees is accomplished through photoidentification. The Manatee Indi vidual Photo-identification Syst em (MIPS) consists of an extensive catalog of photographs for individual identification and sighting history data for individual Florida manatees (Beck and Reid 1995). Identifying marks can be natural, but in the majority of cases, they are scars and mutilations from watercraft collisions. Photo-identification is an excellent tool for capture-recapture studies and has been used in many species of marine mammals (Araabi et al. 2000; Auger-Mth and Whitehead 2007; Hammond et al. 1990; Langtimm et al. 2004; Pennycuick 1978; Pettis et al. 2004). Unfortunately, animals that are not uniquely identifiable cannot be included in the database. Carcasses in advanced stages of decomposition are also often unidentifiable through this method. Linking genetic-identification with photo-identification could he lp alleviate these issues. Genetic capture-recapture methods have the advantage of using a tag that does not change over time. Genetic profiles are unique to all individu als and are permanent so that matches could be made many years later (Allendorf and Luikart 2007). A model DNA database, the Manatee Individual Genetic-identification System (MIGS), has been develope d as a part of this study to

PAGE 21

21 archive this genotypic information and aid in the future developmen t of a large-scale centralized genetics database. Specific Aim 2: Examine the Fine-scale Po pulation Structure of Florida Manatees. Manatees have been divided into four mana gement units along coastal Florida based on migratory patterns from telemetry studies, photo-id entification data, and mo rtality data (Figure 1-4). It is unknown whether these units actua lly correspond to the gene tic structure of the species because other studies could not detect th e level of migration and interbreeding among the units. Genetic markers, such as microsatellites, can be used to examine population structure and genetic connectivity, and have been used succes sfully for other species. As previously mentioned, mitochondrial DNA cannot be used for th is purpose for manatees. The D-loop is a non-coding region of mitochondrial DNA that is typically highly variable, but only one haplotype of the 410 bp region examined was id entified for the Florida individuals sampled (Garcia-Rodriguez et al 1998), and for approximately 50 add itional individuals (A. M. Clark, University of Florida Interdisciplinary Center for Biotechnology Resear ch (UF ICBR), Personal Communication). Mitochondrial DNA is maternally inherited, and does not show the male contribution to the population structure. A dditionally, mitochondrial DNA has a low mutation rate (Avise 2004) and is more useful for det ecting historical pattern s of variation. Highly variable nuclear microsatellite DNA markers are very useful for examining the connectivity and gene flow of populations (Selkoe and Toonen 2006) These markers can be examined at the population and individual levels for a high-re solution analysis of population structure. Specific Aim 3: Test the Microsatellite Markers for Their Utility in Pedigree Analysis. Pedigree analysis requires more sensitivity than allozymes can provide. Mitochondrial DNA can identify only the maternal lineage becaus e mitochondria in the sperm are destroyed by the oocyte after fertilization (Sutovsky et al. 1999). By using a panel of microsatellite markers

PAGE 22

22 for individual identification, a genetic fingerprint for each individual can be determined. When manatee calves are sampled for genetics, the moth er is usually documented and identified. The genotypes of both mother and offspr ing can be compared to those from a pool of putative fathers to determine the paternal lineage. Microsatel lite genotyping is routinel y used for parentage determination in humans (Buckelton et al 2005) and other animal sp ecies, including bison (Schnabel et al. 2000), dogs (Ichikawa et al. 2001), and koalas (Houlden et al. 1996). Known individuals from three captive family units were used to determine if the markers were sufficient for reconstruction of three known captive pedigrees. Figure 1-1. Range of the Florida manatee. The wint er range is fragmented, but the general area is outlined in yellow, and the typical summer ra nge is outlined in red. Most manatees stay within the coastal waters of Florida, but some have been sighted as far west as Texas and as far north as Rhode Island. Figure created with Google Maps and Google Earth Mapping Service (Google 2007).

PAGE 23

23 Figure 1-2. Phylogenetic relationship of the Sireni a to other mammalian taxa. Sirenians closest terrestrial relatives are the elephant and th e hyrax; the order is not closely related to other marine mammals. Figure adapted from Murphy et al. (2001). Figure 1-3. Mitochondrial haplot ype distribution for the West I ndian manatee. The Florida manatee is the only group represented with only one haplotype. This low diversity might be attributed to a founder effect fr om more southerly regions. Figure from Vianna et al. (2006).

PAGE 24

24 Figure 1-4. Florida manatee management units as established by the USFWS (2001). The four regions are the Northwest, Southwest, Uppe r St. Johns River, and Atlantic coast. These units were designated based on da ta from telemetry, photo-identification, mortality studies, and the di fferent anthropogenic threats to manatees in the regions. Figure from the Florida Fish and Wild life Conservation Commission (FWC 2006).

PAGE 25

25 CHAPTER 2 CHARACTERIZATION OF POLYMORP HIC MICROSATELLITE D NA MARKERS AND GENETIC INDIVIDUAL IDENTIFI CATION OF FLORIDA MANATEES Introduction Individual identification of manatees is cu rrently accomplished using photo-identification. The Manatee Individual Photo-iden tification System (MIPS) consis ts of an extensive catalog of photographs and observational data documenting Florida manatees that is managed by the Florida Fish and Wildlife Research Institute (FWRI) for the Southwest region and the United States Geological Survey (USG S), Sirenia Project for the rema ining regions. Distinguishing marks, primarily scars from boat-induced injuri es, are used to differentiate among individual manatees (Beck and Reid 1995). Researchers mu st photographically docum ent the body and tail of an individual for positive identification. Sc ar patterns can accumulate or be obscured by new injuries, requiring frequent monitoring and updati ng of identification information (Figure 2-1). To improve individual identifica tion and the estimation of survival and life history parameters, it is proposed to incorporate a method to link photographic-identif ication and geneticidentification. Previous genetic analyses of the Florida ma natee population were unable to reveal finescale population genetic st ructure (Garcia-Rodrigu ez 2000; Garcia-Rodriguez et al. 1998; McClenaghan and O'Shea 1988). Additionally, the microsatellite markers previously developed for the species were not sufficiently polymorphi c to provide unique multilocus genotypes for individual Florida mana tees (Garcia-Rodriguez et al. 2000). The purposes of this study were to develop a pa nel of microsatellite markers that would be sufficient for individual identification and to de velop a model for a DNA database that could be linked to the MIPS at a later time. Microsatellite loci developed for closely related species were tested for cross-species amplif ication and polymorphism on Florid a manatees. Florida manatee-

PAGE 26

26 specific microsatellite loci were also characte rized and screened on a sample of the Florida manatee population. The eleven new polymorphic micr osatellite loci described herein were used in combination with seven prev iously published polymorphic microsatellite loci for Florida manatees (Garcia-Rodriguez et al. 2000) for genetic indivi dual identification. Materials and Methods Sample Collection All samples used in this study were collect ed by the USGS, Sirenia Project and obtained from their tissue archive, courtesy of R. K. Bonde (USFWS W ildlife Research Permit MA791721/4, issued to the USGS Sirenia Project). Skin samples were collected from the tail region of dead manatees examined at necropsy or from the tails of live animals using a cattle ear notcher (Figure 2-2). All samples were stored in a high salt buffer (S ED buffer, 0.24 M EDTA, pH 7.5, 2.99 M NaCl, 20% DMSO) at room temp erature (Amos and Hoelzel 1991; Proebstel et al. 1993). A total of 450 samples was obtained for genetic analysis, greater than 10% of the current minimum population estimate of a pproximately 3,000 individuals (FWRI 2007). Geographic location and date of sample coll ection, sex, and other basic information were obtained from biologists' fiel d notes or the FWRI online manatee mortality database (http://research.myfwc.com/manatees/search_summary.asp ), and are listed in Appendix B. DNA Isolation DNA was isolated from 166 live animal skin samples using the QIAGEN DNeasy Tissue Kit protocol, eluting the DNA in 50 l AE buffer (10 mM Tris-Cl, 0.5 mM EDTA, pH 9.0) in the final step (QIAGEN, Valencia, CA ). Standard phenol/chloroform isolations were performed to recover DNA from the 284 necropsy tail samples (Hillis et al. 1996). In the final step of the phenol/chloroform isolations, the pellets were re-suspended in 50-75 l of TE buffer (10 mM Tris-HCl, pH 7.4, 1 mM EDTA, pH 8.0) de pending on the size of the pellet (Sambrook et al.

PAGE 27

27 1989). All isolations were quantified using a spectrophotometer (BioPhotometer, Eppendorf, Hamburg, Germany). Each DNA isolation was dilu ted to a final concentration of 10 ng/l for use in PCR. Related Species Primer Screening Microsatellite loci can be conserved across closely relate d species, and in many cases primers developed for one species may be usef ul in related species. Sirenians closest evolutionary relationships are with el ephants and hyraxes (Figure 1-2) (Murphy et al. 2001). A literature search yielded polymorphic microsat ellite primers developed for two species of elephants ( Elephas maximus Loxodonta africana ) and for hyraxes ( Procavia johnstoni ). One polymorphic microsatellite locu s was identified for dugongs ( Dugong dugon ) (Moore et al. 1998). A microsatellite repeat found in the bov ine insulin-like growth factor-1 gene ( IGF-1 ) was also examined because it is known to be a highly polymorphic locus for other marine mammals (T. R. Frasier, Trent University, Ontario, Cana da, Personal Communication). The loci selected from the listed species were tested for polymorphism on the Florida manatee. Temperature gradient PCRs were performe d using a Biometra T-gradient (WhatmanBiometra, Gttingen, Germany) to determine optimal reaction conditions with manatee samples. Reactions were performed with an initial denaturation step of 94C for 5 minutes; 34 cycles of 94C for 1 minute, the optimized temperature (Tab le 2-1) for 1 minute, and 72C for 1 minute; and a final elongation of 72C for 10 mi nutes. The concentration of MgCl2 was altered to optimize primer binding (Table 2-1), and 0.4 mg /ml bovine serum albumin (BSA) was required for the reactions with the loci LafMS02, LafMS07, LaT26, BtaIGF1, Hy-D32, and Hy-D49. Once optimal conditions were determined, tests for polymorphism were conducted for the loci identified from related species using twelve manatee DNA samples collected from various sites around the state of Florida.

PAGE 28

28 Manatee-specific microsatellite identification and screening Several libraries were devel oped and screened to produce ma natee-specific microsatellite primers. Two microsatellite libraries were pr oduced by Genetic Identification Services, Inc. (GIS, Chatsworth, CA (http://www.genetic-id-services.com/ )). The libraries prepared by GIS used a magnetic bead capture technology to enrich for CA, ATG, and AGAT repeat motifs. These libraries were screened, and 52 sequences containing micr osatellites were identified. Thirty microsatellite loci were optimized for PCR, of which six were determined to be polymorphic. Libraries enriched for (CA)n were produced using protocols from the Molecular Markers: Tools for Developing Enriched Microsatellite Libraries Workshop (MMW) cosponsored by the University of Florida Interdisciplinary Center for Biotechnology Resear ch Genetic Analysis Laboratory (UF ICBR GAL) and th e Education Core (Brazeau et al. 2005). Techniques for producing the enriched libraries using the MMW protocols are br iefly described below. Whole genomic DNA was digested using Dpn II restriction enzyme, an isoschizomer of Sau 3AI (New England Biosystems, Ipswich, MA). The dige sted product was size-selected for fragments greater than 400 bp using Chro ma Spin+400 TE columns (Clontech Laboratory, Palo Alto, CA). Linkers with ends complementary to the Dpn II restriction site (SauLA: 5-GCG GTA CCC GGG AAG CTT GG-3, SauLB: 5-GAT CCC AAG CTT CCC GGG TAC CGC-3) were attached using T4 DNA ligase (Fis her Scientific, Fair Lawn, NJ). Excess linkers were removed using Chroma Spin+400 TE columns. Samples were PCR amplified in a 50 l reaction using the free strand of the linker, SauLA, as the pr imer to produce a whole genome library. The reaction mix consisted of 0.2 mM each dN TP (0.8 mM total dNTPs), 3 mM MgCl2, 1X Sigma PCR buffer (10 mM Tris-HCl pH 8.3, 50 mM KCl, 0.001% gelatin), 1 unit of Sigma JumpStart Taq polymerase (Sigma-Aldrich, Fair Lawn, NJ), and 0.5 M SauLA primer.

PAGE 29

29 Approximately 50 ng of DNA from the purified li gation reaction were added to the PCR mix. Reactions were performed with an initial denaturation step of 94C for 3 minutes, 25 cycles of 94C for 1 minute, 68C for 1 minute, and 72C fo r 1 minute, with a final elongation step at 72C for 10 minutes. Enrichment for (CA)n repeats was completed by hybridizing the fragments to a biotinylated (CA)15TATAAGATA probe. The biotinylated products were bound to an Avidin matrix (VECTREX Avidin D, Vector Laborato ries, Burlingame, CA), allowing the fragments not containing (CA)n repeats to be removed through a seri es of washes, as described in the MMW protocol. A second PCR amplification using the free strand of the linker as the primer produced the enriched library. The enriched PCR products we re cloned into pCR2.1-TOPO vector using the TOPO TA Cloning Kit and transformed into E. coli Top-10 cells (Invitrogen, Carlsbad, CA). Screening for colonies containing repeat frag ments of interest was done on Luria-Bertani (LB) plates containing 0.1 mg/ml ampicillin and 0.032 mg/ml 5-bromo-4-chloro-3-indolyl-betaD-galactopyranoside (X-gal) for blue-white sc reening. Colonies were lifted onto nylon membranes and cellular debris was removed thr ough a series of washes (Wash 1: 10% SDS; Washes 2 and 3: denaturing solution (0.5 M NaOH, 1.5 M NaCl); Wash 4: neutralizing solution (0.5 M Tris, pH 7.4, 1.5 M NaCl); Wash 5: (2X SSC). The DNA was UV-crosslinked to the membranes using the auto-crosslink setting on a UV Stratalinker 1800 (Stratagene, La Jolla, CA). The Phototope-Star Kit (New England Biosys tems, Ipswich, MA) uses chemiluminescence to detect colonies containing seque nces of interest. The biotinylat ed probes used for enrichment of the library were also used to detect colo nies containing repeated sequences through the

PAGE 30

30 chemiluminescence protocol. The probes were h ybridized to the membranes with a 30 minute wash at 55C in the prehybridization buffer (6 X SSC, 5X Denhardts so lution (Sambrook et al. 1989), 0.5% SDS), followed by a 60 minute hybridiz ation at 55C in th e hybridization buffer (6X SSC, 0.5% SDS), which included 50 ng of the biotinylated probe for every 100 cm2 of nylon membrane. The Phototope-Star Kit Maximum Volume protocol was followed for the colony lifts, where the CDP-Star reagent was diluted 1:50 0 with 1X CDP-Star diluent in the final step. Blue-white screening was used in combination with chemiluminescence to minimize unnecessary sequencing. The white colonies th at produced a chemiluminescent signal visible on X-ray film were chosen for furt her screening. These were grown by shaking overnight at 37C in 3 ml LB cultures containing 0.1 mg/ml ampicillin. The cultures were centrifuged at 8000 rpm (4293 x g ) for 5 minutes to pellet the bacteria, and th e plasmids were isolated using the QIAprep Miniprep Kit (QIAGEN, Valencia, CA). Dot blots are a form of Southern blotting in which DNA is dotted onto a nitrocellulose or nylon membrane and probed for sequences of intere st. These blots were performed to screen for clones that contained the repeated fragments. Serial d ilutions were made of each plasmid isolate (1:100, 1:500, and 1:2500). Each plasmid isol ate and its corresponding three dilutions were dotted (1 l each) onto a nylon membrane and UV-c rosslinked in the same manner as previously described for the colony lifts, w ithout the first wash in 10% SDS. The membranes were then screened for the repeated sequence fragments using the chemiluminescence protocol previously described for the colony lifts. The final incubati on for the dot blots was less stringent than for the colony lifts. The Maximum Sensitivity protocol was followed, which calls for the CDPStar reagent to be diluted 1:100 w ith the 1X CDP-Star diluent.

PAGE 31

31 A hybridization signal across al l of the dilutions indicated that the clones contained fragments with microsatellite re peat sequences. These plasmid isolates were digested with Eco RI to determine if the fragments would likel y have sufficient flanking sequence for primer design. The 10 l digestion mix consisted of 1X Prom ega Buffer H (9 mM Tris-HCl, pH 7.5, 1 mM MgCl2, 5 mM NaCl), 12 units Promega Eco RI enzyme (Promega Corporation, Madison, WI), and 2 l plasmid DNA. These reactions were visu alized on 1% agarose gels to determine approximate sizes of the fragments. Isolates with fragments 300 bp or larger were sent for sequencing. The UF ICBR Sequencing Core comple ted all of the sequenci ng reactions. Twentyfour clones with sufficient flanking sequence we re identified, and three were identified as polymorphic. PRIMER3 (Rozen and Skaletsky 2000) was used to design primers using candidate clone sequences from all of the libraries. Two loci (TmaSC5 and TmaSC13) were developed from an une nriched library designed in 1994 (A. M. Clark, UF ICBR GAL, Personal Comm unication) and were generously made available for this study. Conditions for PCR were optimized using a gr adient thermocycler (Biometra T-gradient, Whatman-Biometra, Gttingen, Germany). Polymo rphism was initially determined for each Florida manatee-specific locus using 12 sample s from around the state of Florida. Reactions were performed in a total volume of 12.5 l containing 10 ng DNA, 0.8 mM dNTPs, 1X Sigma PCR buffer, 0.04 units Sigma JumpStart Taq polymerase, and 0.24 M each primer. The concentration of MgCl2 was 3 mM for all reactions except TmaKb60 (2 mM) and TmaH23 (1.8 mM), and 0.4 mg/ml BSA was added to reactions for TmaE1, TmaE4, TmaE7, TmaE14, and TmaH23. The PCR program consisted of 94C fo r 5 minutes; 34 cycles of 94C for 1 minute, the optimized temperature (Table 2-2) for 1 minute, and 72C for 1 minute; and a final

PAGE 32

32 elongation of 72C for 10 minutes. If polymorphism was initially detected using 12 samples, one primer per pair was fluorescently labeled with ei ther 2, 4, 4, 5, 7, 7-hexachlorofluorescein (HEX, green) or carboxyfluorescein (FAM, blue ). The polymorphic Florida manatee-specific primers were also tested for cross-species amp lification on three related sirenian groups (Table 2-3). PCR products were run with th e GeneScan-400HD [ROX] (ROX, carboxy-xrhodamine) size standard (Applied Biosystems (AB I), Foster City, CA) and analyzed using an ABI Prism 377 Automated DNA Sequencer and GENESCAN, version 3.1.1 software (ABI). The alleles were sized using GENOTYPER, version 2.5 software (ABI). Some samples were also analyzed on the MegaBACE 1000 (Amersham, Sunnyvale, CA), and scored using GENEMARKER, version 1.3 (Soft Genetics LLC, State Coll ege, PA). A subset of samples was analyzed on both the ABI and MegaBACE to validate allele sizes across platforms. Genotypes from 78 individuals were used for pre liminary screening of the loci. Data were exported from GENOTYPER and GENEMARKER and organized in Microsoft Excel. CONVERT (Glaubitz 2004) was used to format the data to create input files for the genetics software packages. ARLEQUIN, version 3.1 (Excoffier et al. 2005a) was used to estimate the observed and expected heterozygosities ( HO and HE) and to test for Hardy-Weinberg equilibrium. The effective number of alleles ( NE) was calculated using POPGENE, version 1.31 (Yeh and Boyle 1997). Linkage disequilibri um was tested using GENEPOP, version 3.2 (Raymond and Rousset 1995). MICRO-CHECKER (Van Oosterhout et al. 2004) was used to test for the presence of null alleles. Genotyping Amplifications were attemp ted for all 450 tissue sample s. A total of eighteen microsatellite loci was included in the panel: TmaA02, TmaE02, TmaE08, TmaE11, TmaE26, TmaF14, and TmaM79 (Garcia-Rodriguez et al. 2000), TmaE1, TmaE4, TmaE7, TmaE14,

PAGE 33

33 TmaH13, TmaK01, TmaJ02, TmaKb60, TmaSC5, TmaSC13 (Pause et al. 2007), and TmaH23 (C. Nourisson, ECOSUR, Personal Communicati on). An additional polymorphic locus, TmaA03 (Garcia-Rodriguez et al. 2000), was available but ex cluded from the analyses. Preliminary screening of that locus showed non-specific amplif ication, which pe rsisted after optimization attempts. PCR was performed in 20 l reactions containing 0.8 mM dNTPs, 1X Sigma PCR buffer, 0.08 units Sigma JumpStart Taq polymerase, 0.24 M each primer, and 20 ng genomic DNA, and reaction conditions for the el even new loci are described earlier. The annealing temperatures for the primers from the literature (Garcia-Rodriguez et al. 2000) were re-optimized (Table 2-4) and 0.2 mg/ml BSA was added to the reactions with TmaM79. Fragment analysis was completed by the Geneti c Analysis Facility at the Hospital for Sick Children (Ontario, Canada) on an ABI 3730 xl (Applied Biosystems, Foster City, CA) using the GeneScan 500-LIZ size standard. Fragment data were analyzed using GENEMARKER, version 1.4 (Soft Genetics, State College, PA). Allelic information was stored in a Microsoft Access database maintained by the author and the McGuire laboratory group at the Un iversity of Florida. Identity Determination Genotypic data from a total of 443 individual manatees collected from around the state of Florida were used for the following calculations. GENECAP, version 1.2.1 (Wilberg and Dreher 2004) wa s used to calculate the probability of identity in two ways. The probability of identity is the probability that two individuals chosen at random from the same population share the sa me multilocus genotype. First, the standard probability of identity (HW P(ID) ) assumes Hardy-Weinberg equi librium, and was calculated using Equation 2-1 (Paetkau and Strobeck 1994):

PAGE 34

34 iii j j i ip p p ID P2 4) 2 ( ) ( HW (2-1) where pi and pj are the frequencies of the ith and jth alleles. A more conservative estimate of the probability of identity was also calculated using Equation 2-2 (Evett and Weir 1998): ) 25 0 ( ] ) ( 5 0 [ ) 5 0 ( 25 0 ) ( Sib4 2 2 2 i i ip p p ID P (2-2) The Sib P(ID) accounts for the possibility of close relativ es in the population. This provides a conservative upper boundary to the probability of id entity estimate. The actual probability of identity lies somewhere be tween the two values (Waits et al. 2001). Another method to estimate the usefulness of the panel for individual identification is to calculate the mismatch probabilities. The MM-DIST software package (Kalinowski et al. 2006a) was used to calculate the mism atch probability distributions given unrelated or full-sibling relationships. Results were gr aphed in Microsoft Excel. MM-DIST was also used to calculate the mismatch probability, or the probability th at two individuals have the same genotype at L loci with a given relationship (unrelated or full-si bling). To accomplish this, the heterozygosity values were calculated for each locus, and the loci were ordered from the greatest expected heterozygosity value to the smallest. The mismat ch probability was calculated for the two loci with the greatest heterozygosity and agai n for each increasing number of loci. Development of the Manatee Individual Ge netic-identification System (MIGS) A database to archive the genot ypic information for individual manatees was developed in Microsoft Access (Figure 2-3). This was done in an effort to minimize human error and to expedite the creation of data input files for population genetics software packages. Through collaboration with MIPS researchers at the USGS, common fields betw een the two databases were designated to enable linking them at a late r date. As a preliminary step, a MIPS# field was added in the MIGS database as well as including all other field sampling identification

PAGE 35

35 numbers in the FieldID# field. A thorough he lp file is included in the MIGS database distribution package. This file describes the database setup and common user functions (Appendix C). Results Related Species Primer Screening Based on the set of twelve manatee samples, ten of the fifteen primer pairs that were developed for related species am plified products using manatee DNA. Nine of the ten primers with successful amplification yi elded products that were in th e expected size range, but none showed polymorphism for Flor ida manatees (Table 2-1). Manatee-specific Microsatellite Screening Most of the microsatellite lo ci designed for this study were determined to be in HardyWeinberg equilibrium, indicating their utility fo r population genetic analys es (Table 2-2). To minimize type I statistical errors, the sequentia l Bonferroni correction was used to determine significance for multiple simultaneous statistical tests (Allendorf and Luikart 2007; Rice 1989). Before a sequential Bonferroni correction, 6 of the 11 loci met Hardy-Weinberg expectations, and following the Bonferroni, TmaE14 was estimat ed to meet equilibrium expectations for a total of seven of the eleven loci in equilibrium. The number of alleles per locus ranged from two to seven and single locus heterozygosities rang ed from 20.8% to 68.0%. The eleven new polymorphic loci had an average of 4.0 alleles per locus and an average heterozygosity of 47.5% across all loci. GENEPOP calculated 55 pairwise comparisons, ( n ( n -1)/2), of the loci to test for linkage disequilibrium. No evidence for linkage was observed after a sequential Bonferroni correction. MICRO-CHECKER suggested that two loci (TmaK b60 and TmaE1) presented evidence of null alleles. Null alleles re sult in a general excess of homoz ygotes, which was the signal that MICRO-CHECKER used to identify loci w ith null alleles (Van Oosterhout et al. 2004). The

PAGE 36

36 presence of PCR products when the loci were amplified in related species indicates their potential utility in studies of ot her sirenian species (Table 2-3). Identity Determination Two methods were used to estimate the ut ility of the panel of markers for Florida manatees, the probability of identity and th e mismatch probability. Single locus HW P(ID) estimates ranged from 0.0486-0.4599, with the overall 18 locus HW P(ID) estimated as 4.124x10-13. Single locus Sib P(ID) estimates ranged from 0.3558-0.6988, and the overall Sib P(ID) was estimated to be 5.719x10-6 with all 18 loci included (Table 2-5). This is equivalent to saying that there is between a 1 in 175,000 and 1 in 2 trillion chance of randomly choosing two Florida manatees with identical multilocus genotypes. Based on an estimated population size of approximately 3,000 manatees (FWRI 2007), this is sufficient for individual identification. Loci were ordered by descending heterozygosity values. In some cases, individuals with greater numbers of alleles had smaller heterozygo sities. This occurred when there were rare alleles or increased homozygotes at a locus due to null alleles. For unrel ated individuals, the probability of identity is approximately 10-13 (Table 2-5), and there is a high probability (P 0.9999, or a 99.99% chance) that the individuals will differ by mo re than two loci (Table 26). The estimated probability of identity is greater for siblings (approximately 10-6) (Table 2-5), and the mismatch probability is slightly smaller (P 0.9991, or a 99.91% chance) (Table 2-6). The two most heterozygous loci were determin ed to be TmaSC5 and TmaE11 (Table 2-7). If only these two loci were used to distinguish unrelated individuals, there would be a 75.91% chance that the individuals genot ypes would differ. If they were full-siblings, there would be a 32.72% chance that their genotypes would differ (T able 2-7). If a subset of the 15 most heterozygous loci were chosen, the chance their genotypes would differ at more than two loci increases to 99.99% for unrelated individuals and to 99.82% if they are full-siblings (Table 2-6).

PAGE 37

37Discussion The panel of markers developed for this study is sufficient for indivi dual identification of Florida manatees using microsatellite DNA fingerp rints. A model DNA da tabase was developed using the data generated with th ese 18 markers. Genetic studie s of other sirenians may benefit from the development of the sy stem; however, those species may require a different panel of markers for individual iden tification. Allele frequencies can be different in separate populations, with some loci being polymorphic in one popul ation and monomorphic in another (Buckleton 2005). This is because some alleles become fixe d in a population and lost in other populations when there is little or no ge netic connectivity due to gene tic drift (Gillespie 2004). Some of the loci described herein were iden tified as monomorphic when applied to the Puerto Rican population of Antillean manatees, T. m. manatus (M. E. Kellogg, University of Florida, Personal Communication). One of the eleven new polymorphic loci, TmaSC5, did not successfully amplify in some of the other sirenian species examined (Table 2-3). These results are not unexpected, since Florida manatees, T. m. latirostris are not breeding with the other species tested, and the species are assumed to be genetically separate. In addition, none of the microsatellite loci that were originally is olated from related speci es amplified polymorphic products when screened on Florid a manatees. Ascertainment bias is a well-known problem that occurs when microsatellite markers ar e transferred across species (Primmer et al. 1996, Webster et al. 2002). A recent meta-analysis of mamma lian genetic variability suggested that ascertainment bias can cause a 10 to 13% reducti on in the estimation of average heterozygosity in the transferred species (Garner et al 2005). When attempting to identify indi viduals from other species or subspecies of sirenians, a different panel of markers will likely be required due to differences in allele frequencies among populations. It is known from human forensics that population structure and individuals with

PAGE 38

38 mixed ancestry may affect the allele frequency distributions and the probability of identity estimates (Buckleton 2005; Rowold a nd Herrera 2005; Serre and Pbo 2004). The database that has been developed is a model for a large-scale centralized DNA system. In the future, the MIGS could be used for cap ture-recapture estimates of population parameters in concert with the MIPS. The MIGS will store genetic profiles from known MIPS individuals and other individuals that are not photographically identifiable. Genetic matches could be made from a known MIPS individual to a previously anal yzed sample from the same MIPS individual, from an unknown sample to a known MIPS individual, and from an unknown sample to a previously analyzed sample in the MIGS database (Figure 2-4). Eighteen additional polymorphic microsatellite markers have been developed for Florida manatees by Florida Fish and Wildlife Conser vation Commissions (FWC) Florida Fish and Wildlife Research Institute (FWRI) and Mote Marine Laboratory (M. D. Tringali, FWRI, Personal Communication). Additional micros atellite markers developed for dugongs ( Dugon dugong ) amplified products when tested on a Florida manatee (Broderick et al. 2007), and are being screened for polymorphism on Florida mana tees (J. M. Lanyon, University of Queensland, Personal Communication). Although the panel of markers presented he re is sufficient for genetic individual identification of Florida manatees, new markers are a welcome addition to this panel. Loci containing triand tetranucleotide repeat motifs w ill be especially useful because they facilitate allele scoring. There are severa l advantages to using triand te tranucleotide repeat loci. One advantage is the elimination of most stutter pe aks. Stutter peaks are a result of polymerase slippage during PCR amplification, and are more likely to be problematic for allele scoring for dinucleotide repeat loci. It woul d be beneficial to in clude loci that are hi ghly polymorphic, have

PAGE 39

39 high heterozygosities, and show no ev idence of null alleles. A null al lele occurs when an allele fails to amplify due to a mutation in the priming site. A heterozygote for a null allele appears the same as a true homozygote during fragment analys is, and a homozygote for a null allele will not amplify a PCR product (Wagner et al. 2006). This results in a general excess of homozygotes and produces problems for pedigree analysis. Including loci with the above qualities in the panel will reduce the number of loci required for a positive identification. This is important to reduce costs and increase efficiency. If more polymorphic and heterozygous loci ar e identified and includ ed in the new panel, it can be used for individual identificat ion, as well as pedigr ee reconstruction. Pedigr ee reconstruction requires greater sensitivity than individual identificati on due to the presence of close relatives with similar genotypes in the analyses. Collaboration among laboratories is now essential. The panel of loci should be established quickly to prevent the developm ent of incompatible datasets among laboratories. Additionally, allelic ladders should be created for each locus to st andardize allele scor ing among the different laboratories involved (LaHood et al. 2002; Moran et al. 2006). Allelic ladde rs are developed to represent the spread of alleles present for a given locus in the populat ion of interest (LaHood et al. 2002). They are important for standardizing allele scoring due to differences in equipment and reagents among laboratories and w ithin the same laboratory over time. Collection of genetic data for the wild ma natee population is th e next step in the implementation of a genetic capture-recapture syst em. Ideally, while the genetic data are being collected, the database ca n be refined and the statistical met hods for integrating the MIGS data with the MIPS data should be developed.

PAGE 40

40 Figure 2-1. Scar pattern accumulation of a Fl orida manatee over time. All photographs ar e from the same individual. In March 1991, the individual photographed had a scar on the left side of the tail. By Marc h 1996, the animal incurred an additional feature, a mutilation from a boat propeller wound. If this animal did not have othe r distinguishing marks or had not been observed and photographed before the regi on containing the initial scar was rem oved, the photograph taken in February 1999 could not have correctly identified this individual. (Photographs courtesy of USGS, Sirenia Project.)

PAGE 41

41A B C Figure 2-2. Tissue sampling method for free-ra nging manatees. A) Cattle ear notcher used for collec ting tissue samples. B) Ma natees are periodically captured for health assessment and telemetry st udies. Tissue samples are taken for genetics using the cattle ear notcher. The device is also used to sa mple calves in the water. C) The sampling device leaves a small scar in the tail, indicated by the red arrowhead. The locati on of this scar can help to determine th e year that it was sampled. (Photographs courtesy of R. K. Bonde USGS Sirenia Project.)

PAGE 42

42Table 2-1. Microsatellite loci characterized fo r related species selected from the literature. Tm, optimized annealing temperature; MgCl2 (mM) optimal concentration of MgCl2; Result A No amplification, B, Amplification not specific with any conditions attempted, C Product amplified, not of expected size, D Product of expected size, not polymorphic. Taxa Locus Name GenBank Accession Primer Sequences (5 3) Reference Tm MgCl2 (mM) Result Elephant ( E. maximus ) LafMS02 DQ198478 F: GAAACCACAACTTGAAGGG R: TCGCTTGTAAGAAGGCGTG (Nyakaana and Arctander 1998) 58 3 D Elephant ( L. africana ) LafMS06 AY817498 F: AGCTGTCCTAAGTCATAAATACACA R: ACAGCCACTGAAACCCCATGGA (Nyakaana et al. 2005) A Elephant ( L. africana ) LafMS07 AY817499 F: TACCCCACTCCAATTCCATGTCT R: GGACAGGCAGAATCTAGTGGAGG (Nyakaana et al. 2005) 60 3.5 D Elephant ( L. africana ) LafMS09 AY817501 F: CTGGGGCAGTAAGCTGTATTTATC R: ACGAGGATGACAGACCAGGCAACA (Nyakaana et al. 2005) 59 2 D Elephant ( E. maximus ) EMX-1 AF352833 F: AGGACTTATTTGCTTAGATGG R: AGGCAATGTTTCGTTCTGT (Fernando et al. 2001) B Elephant ( L. africana ) LaT05 AY172173 F: CACCACCCATCCATCTGT R: TGGCTTCTGTGAGTTCACC (Archie et al. 2003) 56 2 D Elephant ( L. africana ) LaT07 AY172175 F: CCTGAGCCATTTTCTTGAG R: GATGGAGAGACAGATTTGCTAG (Archie et al. 2003) B Elephant ( L. africana ), ( E. maximus ) LaT08 AY172176 DQ198483 F: ATGGACAGGCAGAAAGATTT R: TCCCAATAACAGGATAGCATT (Archie et al. 2003) B Elephant ( L. africana ), ( E. maximus ) LaT26 AY172184 DQ198488 F: AACCCAGGATAAAGCACCAA R: TTTCCTGCTTGAGAGCCAAA (Archie et al. 2003) 58 3 D Elephant ( L. africana ) LA4 AF311673 F: GCTACAGAGGACATTACCCAGC R: TTTCCTCAGGGATTGGGAG (Eggert et al. 2000) B Cow ( Bos taurus ) BtaIGF1 AF404761 F: GGGTATTGCTAGCCAGCTGGT R: CATATTTTTCTGCATAACTTGAACCT (Barendse et al. 1994) 60 3 D Dugong ( D. dugon ) NCAM AF025990 F: AAAGTGACACAACAGCTTCTCCAG R: AACGAGTGTCCTAGTTTGGCTGTG (Moore et al. 1998) 59 3.5 C Hyrax ( P. johnstoni) Hy-D32 AF268564 F: TGATGATAATCAGCAATGAG R: TGCCTTCATATAAATCTGTC (Gerlach et al. 2000) 50 5.5 D Hyrax ( P. johnstoni) Hy-D48 AF268567 F: AACAGAATCCACAGACCGTG R: GTTCATATCCATTGCTCGTG (Gerlach et al. 2000) 50 2 D Hyrax ( P. johnstoni) Hy-D49 AF268568 F: TGTGTACTGATTGTTAAATGAG R: CACAGGTCTTGCTCTCAAG (Gerlach et al. 2000) 58 3.5 D

PAGE 43

43Table 2-2. Characteristics of the eleven polymorphic microsat ellite loci developed for the Florida manatee ( T. m. latirostris ). Tm, optimized annealing temperature; size range observed size range of alleles in bp; N number of reactions attempted/number of animals successfully genotyped; NA, number of alleles in the Florida population; NE, effective number of alleles in the Flor ida manatee population; HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibrium, those in italics are not in equilibrium after a sequential Bonferroni correction is performed. Locus Name GenBank Accession Repeat motif Primer Sequences (5 3) Tm Size Range N NA NE HO HE P HWE 1TmaK01 EF191340 (CT)10(CA)11 F: CTATCAAGCGGCATGTTCAA R: AGCTTGGGATCGTGTTTGTT 58 184-202 78/ 67 4 1.9564 0.6716 0.4924 0.0008 1TmaJ02 EF191341 (AC)9 F: CACCATTGCCTCACAATCAG R: TGGTGGTTAACTTTCTGTGCAA 62 224-230 78/ 76 2 1.5831 0.3816 0.3708 1.0000 1TmaKb60 EF191342 (TG)4(CG)1(TG)12(CG)6 F: TAGACACAGGCAAGCAGTGG R: AAGAGTGAGCGGAGATGTGG 62 215-219 78/ 78 3 1.8909 0.3846 0.4742 0.00001 2TmaSC5 EF191349 (AC)5G(CA)18 F: ATTACCCAGCTCAGCATGTAC R: GCAACCTCTTCTGTTTTCAAA 60 124-146 78/ 78 7 3.2284 0.6795 0.6947 0.0002 2TmaSC13 EF191348 (GT)21 F: GGTTTGAAGAATCAGTTTGAA R: AATAAAGTTCTTCGTGTGCC 56 107-129 78/ 78 6 2.3257 0.5897 0.5737 0.1344 3TmaE1 EF191343 (CA)14 F: ATGGGTGAGTTTTGCT R: TGAAGAAATAGTGATGGTGT 55 272-280 78/ 77 6 2.1766 0.4286 0.5441 0.0043 3TmaE4 EF191344 (GT)2(AT)1(GT)15 F: CCTGACCAGTCCCTTTCC R: GGCTTTTCGGTTTCAACATA 55 246-250 78/ 71 2 1.4816 0.3521 0.3273 0.7186 3TmaE7 EF191345 (TG)14(T)1(TG)15 F: GCAAGCTTACATGTGTGTATGTG R: GTGGCTGACTTCTTGGAAGC 56 190-196 78/ 72 3 2.7141 0.6111 0.6360 0.3745 3TmaH13 EF191347 (TCTA)13 F: GCATCTTGGAAGATTTTTCCTT R: CACTGACAGATACGTGGTGGA 54 237-253 78/ 78 3 1.5317 0.3205 0.3494 0.3840 3TmaE14 EF191346 (AT)7(GT)14(TT)1 (GT)5 F: TTTTGGTAGTGGGATGACCA R: GTGGAGTAGGGTGGACCAGA 56 238-254 78/ 74 6 2.7109 0.5946 0.6354 0.0090 3TmaH23 (AGAT)5AAAT (AGAT)4GGAT (AGAT)5 55 222-230 79/ 72 2 1.2630 0.2083 0.2097 1.0000 indicates the clone sequence has not yet been submitted to GenBank, (C. Nourisson, ECOSUR, Personal Communication). Primer sequences designed by: 1K. C. Pause (UF), 2A. M. Clark (UF ICBR), and 3C. Nourisson (ECOSUR)

PAGE 44

44 Table 2-3. Cross-species amp lification of the polymorphic pr imers developed for Florida manatees ( T. m. latirostris ). Primers were tested for amplification with Antillean manatee ( T. m. manatus ), the Amazonian manatee ( T. inunguis ), and the dugong ( Dugong dugon ) samples. N number of individuals sampled above the diagonal and the number of successful amplifi cations are indicated below; NA, observed number of alleles; NP no product observed from the animals tested; NT not tested. T. m. manatus T. inunguis D. dugon Locus Name N NA N NA N NA TmaK01 17/14 2 8/2 1 3/1 1 TmaJ02 17/13 2 8/8 6 3/3 1 TmaKb60 17/9 2 8/8 4 3/3 3 TmaSC5 17/16 3 8/8 3 3/0 NP TmaSC13 17/15 3 8/4 3 3/1 1 TmaE1 17/10 5 8/7 2 3/3 1 TmaE4 17/8 1 8/7 3 3/1 2 TmaE7 17/14 4 8/5 1 3/1 1 TmaH13 17/15 3 8/8 3 3/3 1 TmaE14 17/8 3 8/8 4 3/3 1 TmaH23 0/0 NT 0/0 NT 0/0 NT Table 2-4. Optimized conditions for seven polym orphic primer pairs previously published for Florida manatees (Garcia-Rodriguez et al. 2000). Tm, Re-optimized annealing temperatures. Locus Name Tm TmaA02 56 TmaE02 58 TmaE08 60 TmaE11 58 TmaE26 58 TmaF14 58 TmaM79 54

PAGE 45

45 A B Figure 2-3. MIGS database organi zation. A) Main menu with op tions for querying the database for genotypic information from Florida or ot her regions. Locality Data Input Form is open. This form includes all fields in the Tbl Locality Data table. B) Option to query a single unit within Florida (Get Data from a Unit). In this screen capture, the database was queried for all genotypic information for the Atlantic management unit.

PAGE 46

46 Table 2-5. Probabilities of identity as estimated by GENECAP (Wilberg and Dreher 2004). HW P(ID) represents the standard probability of identity (Paetkau and Strobeck 1994), and Sib P(ID) is a conservative estimate of th e probability of identity, allowing for siblings and close relatives in the population (Evett and We ir 1998). Loci are ordered by descending hetero zygosity value. Locus Name HW P(ID) Sib P(ID) TmaSC5 0.0916 0.3905 TmaE11 0.1256 0.4314 TmaE7 0.1301 0.4261 TmaE08 0.1979 0.4821 TmaKb60 0.1535 0.4496 TmaE14 0.0486 0.3558 TmaE1 0.1431 0.4595 TmaE02 0.3204 0.5700 TmaK01 0.1867 0.4919 TmaM79 0.2648 0.5318 TmaSC13 0.2283 0.5135 TmaJ02 0.3204 0.5965 TmaF14 0.3892 0.6311 TmaA02 0.3475 0.6085 TmaH13 0.1684 0.4908 TmaE4 0.3164 0.6005 TmaE26 0.4599 0.6988 TmaH23 0.3402 0.6226 Overall 4.124x10-13 5.719x10-6 Table 2-6. Probability of two randomly chosen unr elated individuals or full-siblings having two or more genotypic mismatches gi ven genotypic information from L loci. The loci were selected to maximize expected hetero zygosity (e.g. indicates that the two most heterozygous loci were chosen, TmaS C5 and TmaE11. indicates that the three most heterozygous loci were ch osen, TmaSC5, TmaE11, and TmaE7.). Number of loci ( L ) Unrelated Siblings 2 0.75912698 0.32723321 3 0.93911689 0.58368162 4 0.98479556 0.75202165 5 0.99603498 0.85423743 6 0.99902252 0.91620683 7 0.99970930 0.94849545 8 0.96764455 0.99996607 9 0.99996607 0.97984990 10 0.99998717 0.98725538 11 0.99999540 0.99202934 12 0.99999818 0.99469191 13 0.99999920 0.99637734 14 0.99999962 0.99747843 15 0.99999983 0.99822488 16 0.99999992 0.99865522 17 0.99999995 0.99890502 18 0.99999998 0.99909128

PAGE 47

47 Table 2-7. Diversity statistics fo r the loci as calculated by MM-DIST (Kalinowski et al. 2006a). Loci were ordered based on their heterozygosity value. N represents the number of manatees successfully genotyped at a locus, and NA represents the number of alleles observed at the locus. Locus Name Heterozygosity N NA TmaSC5 0.7263 410 11 TmaE11 0.6925 437 10 TmaE7 0.6468 399 4 TmaE08 0.6272 438 11 TmaKb60 0.6093 402 5 TmaE14 0.6056 305 7 TmaE1 0.5210 377 7 TmaE02 0.5097 438 5 TmaK01 0.4995 391 7 TmaM79 0.4972 410 3 TmaSC13 0.4967 401 8 TmaJ02 0.4217 425 7 TmaF14 0.4063 433 6 TmaA02 0.3844 416 8 TmaH13 0.3602 349 5 TmaE4 0.2929 388 3 TmaE26 0.2171 409 4 TmaH23 0.1977 375 3 Figure 2-4. Diagrammatic representation of the MIGS DNA database structure and match process. Genetic profiles contained in th e MIGS can be matched in three ways: A) Genetic profiles from known MIPS individuals matched/recaptured as the same MIPS individual, B) Genetic prof iles an unidentified indivi dual matched to a known MIPS individual, or C) Genetic profiles already contained in the MIGS are recaptured through their genetic profiles. Figure ad apted from Walsh and Buckleton (2005).

PAGE 48

48 CHAPTER 3 SEASONAL POPULATION GENETIC STRUCTURE OF MANATEES IN FLORIDA Introduction Florida manatees ( Trichechus manatus latirostris ) live in the coastal waters and rivers of Florida and the southeastern United States. The Florida manatee population has been divided into four management units for the conserva tion and recovery of the species (FWC 2006; USFWS 2001). These units were established based on di stribution patterns estimated from aerial surveys, photo-identification, telemetry resear ch, and mortality studies. The regions were designated to reflect the winter site fidelity, migratory pa tterns, as well as the variety of anthropogenic and natural threats to manatees in each geographic region; however they are not meant to imply isolation of th e sub-populations (Figure 1-4). The MIPS catalog has documented strong winter site fidelity for many individual manatees year afte r year. For example, between 1978 and 1991, 93% of females and 87% of males from the Northwest management unit in the photo-identification catalog returned each year (USFWS 2001). These observations led to the hypothesis that genetic substructuring may exist in the Florida population. Previous surveys of the geneti c diversity of Florida manatees suggest that there is an average-to-low level of divers ity in the Florida population. Analyses with mitochondrial DNA identified only one haplotype in th e Florida population (Garcia-Rodriguez et al. 1998; Vianna et al. 2006). Analyses with allozymes (McClenagha n and O'Shea 1988) and eight microsatellite markers (Garcia-Rodriguez 2000) id entified somewhat greater divers ity, and suggested that there may be a slight differentiation between manatees that inhabit the east and west coasts of Florida. These results are reexamined here using eleven additional microsatellite loci and additional analytical techniques. It is well-established th at endangered species typi cally have lower genetic variation than populations of closelyrelated species (Avise 2005; Frankham et al. 2002).

PAGE 49

49 Genetic diversity is a form of biodiversity, and can be thought of as the raw material for adaptation and evolutionary change (DiBattista 2007). Reduced genetic diversity has been suggested to decrease a populations evoluti onary potential or ability to respond to environmental change, the mean individual and population fitness, and the long-term population persistence and adaptability (Hedrick 2001; Garner et al 2005; Lacy 1997). Population geneticists use neutral variation (regions of the genome not affected by natural selection) to infer the overall diversity of a populat ion. Microsatellite marker s are found throughout the genome, and are treated as neutral markers, providing multilocus estimates of diversity. Some have cautioned on using a single class of molecula r markers to infer overall population diversity (Avise 2004), and with the comp letion of this study, there are now three different, independent surveys of the Florida manatees genetic diversity. The Florida manatee subspecies inhabits the northern limit of the ra nge of West Indian manatees ( T. manatus ). Seasonal changes in water temperature are an important factor in their distribution patterns. In the winter months, they must migr ate to warm-water areas to thermoregulate (Figure 3-1). Floridas natural sp rings, power plant effluent s, and natural thermal basins in south Florida serve as the winter habitat for manatees Manatees disperse throughout the state in the warm summer months. Many inst ances when manatees have been found outside Florida waters occurred during the summer season (Fertl et al. 2005; CNN.com 2006a; CNN.com 2006b; Reid 2000; USGS 2001). These observations led to the hypothesis that the signature of population structur e would be different depending on the season of sample collection, and that the p opulation genetic structure would be more distinct in the winter than in the summer due to the strong site fide lity observed during the colder months.

PAGE 50

50 An alternative possibility is that high rates of gene flow and few geographic barriers to migration would result in very little genetic subdivision among the populations. Marine organisms tend to have more sub tle patterns of population genetic structure. Few barriers to migration result in high rates of gene flow with weak signals of genetic differentiation (Waples 1998). Although the behavioral ev idence supports substructuring in the population, manatees are long-lived animals with long generation times. Th ey are capable of migrating great distances, and could begin life in one management unit and spe nd their adult life in a nother. In addition, breeding occurs throughout much of the warmer months, when th ere may be extensive mixing of individuals (Rathbun et al. 1995). This would allow for high ra tes of gene flow around the state. The purpose of this study was to survey th e nuclear genetic dive rsity and examine the population structure of Florida manatees using multilocus micr osatellite genotypes. These genotypes were examined at both the individual and population leve ls. In order to examine the hypothesis of population sub-struct ure, the dataset was examin ed as a whole. A second hypothesis, that migration in res ponse to seasonal changes in wate r temperature would result in population substructure, was exam ined by partitioning the genotypi c data into subsets based on the season of sample collection and sex. Materials and Methods Sample Collection and DNA Isolation Manatee skin samples used for genetic analysis were collected by the USGS, Sirenia Project from free-ranging calves and manatees capt ured for health assessment, and by the FWRI from necropsy specimens. All samples were obtaine d from the USGS tissue archive, courtesy of R. K. Bonde (USFWS Wildlife Research Pe rmit MA791721/4, issued to the USGS, Sirenia Project); the samples used here were collected from 1992 to 2005. DNA was isolated from the

PAGE 51

51 manatee skin samples using standard phenol/chl oroform methods or the QIAGEN DNeasy kit as previously described (Chapter 2). Genotyping PCR was performed as previously described (C hapter 2). Fragment data were analyzed using GENEMARKER, version 1.4 (Soft Genetics, LLC, State College, PA). Allelic information was stored in a Microsoft Access database de veloped for Florida manatees (Chapter 2). A total of 362 manatees was ge notyped successfully at twelve or more loci (individuals with six or more loci that did not amplify were not included in th e analyses). Data were exported from GENEMARKER to Microsoft Excel for conversion of th e fragment length values to integers. CONVERT (Glaubitz 2004) was used to format the data as input files for the genetics software packages. Statistical Analyses Genetic Diversity Statistics Genetic diversity statistics were calculated fo r the entire dataset and for the dataset when subdivided by management unit or coast. ARLEQUIN, version 3.1 (Excoffier et al. 2005a) was used to calculate the observed and expected heterozygosities ( HO and HE) and to test for HardyWeinberg equilibrium. Linkage disequilibrium was tested using GENEPOP, version 3.2 (Raymond and Rousset 1995), and MICRO-CHECKER (Van Oosterhout et al. 2004) was used to identify loci with evidence of null alleles. Data were grouped by season of sample co llection to test th e hypothesis that the distribution of manatees in the wa rm months may be different than in the winter subpopulations. The seasons were designated as winter (N ovember through February) and summer (March through October) based on the amou nt of migration and site fi delity observed during specific months (R. K. Bonde, USGS Sirenia Project, Personal Communication) The 362 individuals

PAGE 52

52 were sorted by sampling location, date, and sex and were retained for the seasonal datasets (Table 3-1). Diversity statistic s were calculated as described above for the seasonal datasets when grouped by either management unit or coast. Individual Level Analyses Several methods were available at the time of this study for defini ng population structure at the individual level using multilocus microsatellite genotypes. The methods that were chosen to visualize the relationships among individuals were model-ba sed Bayesian clustering, as implemented by STRUCTURE, version 2.1 (Pritchard et al. 2000), microsatellite genetic distancebased Neighbor-Joining trees, and Multidimensional Scaling (MDS) analysis of genetic structure, the latter tw o based on individual pairwi se genetic distances. STRUCTURE was used in an attempt to assign in dividuals to inferre d genetic clusters without a priori designation of the geographi c collection location. STRUCTURE accomplishes this using the principles of Hardy-Weinberg e quilibrium and linkage equilibrium to create subpopulations of genotypes that meet the equi librium expectations. The multilocus genotypes were used to identify the most probable number of genetic clusters in a gi ven dataset. This was accomplished by estimating the probability of the data, Pr(X|K) A prior value of K (number of clusters) was chosen, and the probability was estimated over a number of Monte Carlo Markov Chain (MCMC) iterations. The posterior probabilities, Pr(K|X) can be calculated following Bayes rule. The value of K with the highest posterior probabili ty is the most likely number of clusters. A more detailed explanation can be found in the STRUCTURE help documentation (Pritchard and Wen 2004). All STRUCTURE analyses used the admixture model, which allows individuals of mixed ancestry to be assigned to one or more inferred genetic clusters, with their assignment scores corresponding to the probability of ancestry in a given cluster. In this study, the data were

PAGE 53

53 examined as a whole, and examined again using the seasonal subs ets. The value of K was set from 1 to 12. A burn-in period of 200,000 iterations was used. Following the burn-in period, 1,000,000 MCMC iterations were used, and the program was run without a priori designation of the collecti on location. Although the STRUCTURE documentation suggests that fewer iterations are necessary to obtain proper results (P ritchard and Wen 2004), other simulation studies have shown that longer runs are required in populations with high gene flow (Waples and Gaggiotti 2006). Three independent runs of this scheme were simulated for each value of K The three values for the estimated ln( Pr ( X | K )) were averaged, from which the posterior probabilities were calculated. The value for K with the greatest posterior probability ( Pr 1.0000) was identified as the optimum number of subpopulations. To attempt to reveal relationships among indivi duals, individual pair wise genetic distances were calculated and visualized w ith microsatellite genetic distan ce-based Neighbor-Joining trees. Analyses were completed for the whole dataset a nd the seasonal subsets. In an attempt to improve clarity and resolution in the output figures and to prev ent bias in the Neighbor-Joining trees, the winter dataset was s ubsampled to represent the estim ated manatee population size in the management units. MIPS observations, telemetr y studies, aerial surveys, and mortality data suggest that the Atlan tic and Southwest management units support the majority of Florida manatees (USFWS 2001; USFWS 2007). To preven t over representing a re gion, the previously estimated percentages of the population in each management unit winter subpopulation were used for subsampling of the winter dataset (Figur e 3-2). The number of manatees included from each management unit was calculated based on the unit with the fewest relative samples (Table 3-2). Data were subsampled by randomly ordering the samples and then removing every other genotype from a given unit until th e subset contained the proper number of individuals. The

PAGE 54

54 analyses were completed with males and female s from each season separately and together. Only the trees with the males and females separated were presented to improve clarity and resolution for the reader. Estimates of individual pair wise genetic distances were calculated as the Proportion of Shared Alleles, the ln(PSA) as in Equation 3-1, D f f PSAka j a i a ) min(, (3-1) where the PSA is equivalent to the sum over all loci and all alleles where fa,i and fa,j are the frequencies of allele a in population i or j and D is the number of loci. Individual pairwise genetic distances were also calculated as th e Cavalli-Sforza and Edwards chord distance, DC as in Equation 3-2, r j m i ij ij Cjy x r D ) 1 ( 2 ) / 2 ( (3-2) where r is the number of loci and xij and yij are the frequencies of ith allele in the jth locus. Both genetic distances were calculated by the program MICROSAT, version 1.5d (Minch 2007). These genetic distance matrices were subjected to the NEIGHBOR algorithm in PHYLIP (Felsenstein 2004) for the development of the Neighbor-Joini ng trees. The tree files generated by the NEIGHBOR program were visualized using TREEVIEW, version 1.6.6 (Page 1996). The tree images were imported to Microsoft PowerPoint to add a color key for designation of the individuals from the four management units. The individual pairwise DC distances were also utilized for MDS using the SYSTAT statistical software package, version 12 (SYSTAT Software, Inc., San Jose, CA). The data were examined for each seasonal subset using all of the available regression settings for the Square Similarities model in SYSTAT.

PAGE 55

55 Population Level Analyses Neighbor-Joining trees were also used to ex amine the genetic structure at the population level. The datasets we re bootstrapped using the SEQBOOT algorithm in the PHYLIP package (Felsenstein 2004). Estimates of pa irwise Cavalli-Sforza and Edwards DC genetic distance among populations were calculated by the GENDIST algorthim in the PHYLIP package. This generated distance matrices which were subjected to PHYLIPs NEIGHBOR algorithm to develop Neighbor-Joining trees. The consensu s trees were generated by the CONSENSE algorithm in PHYLIP and were visualized using TREEVIEW, version 1.6.6 (Page 1996) The image files were imported to Microsoft PowerPoint to add a color key for visualization of the units. To test the hypothesis that the established management units for Florida manatees correspond to the genetic population structure, the genotypes for i ndividuals were grouped into their respective management units based on the ge ographic sampling location. Previous studies demonstrated subtle genetic structure between Fl oridas east and west co asts (Garcia-Rodriguez 2000; McClenaghan and O'Shea 1988). This hypothe sis was also tested by grouping individuals based on the coast from which the sample was obtained. To examine structure at the population leve l, comparisons of genotype frequencies between populations were performed. Global FST values were calculated using MICROSATELLITE ANALYSER (MSA), version 4.05 (Dieringer and Sc hltterer 2003), which follows the methodologies of Wier and Co ckerham (1984). Pairwise pop ulation differentiation was examined by calculating pairwise FST values and associated significances using MSA. Significance was determined thr ough permutation of the dataset. Permutation of the genotypes between populations produced a null distribution of pairwise FST values. The p-value of the test represents the proportion of these permutati ons that led to an FST larger than the observed value. This was done to test if the populations are significantly di fferent (Excoffier et al 2005b).

PAGE 56

56 Microsatellite studies should also calculate RST, an analog of FST, because it was developed to account for the high mutation rate of microsat ellites (Balloux and G oudet 2002). Pairwise RST values and associated signifi cances were calculated using ARLEQUIN, version 3.0 (Excoffier et al. 2005a). Results Genetic Diversity Statistics Genetic diversity statistics were calculated fo r the entire dataset (T able 3-3), and for the dataset grouped by management unit or coast of collection (Appendix D). When the entire dataset was examined, 2 tests for Hardy-Weinberg equi librium showed departures from equilibrium expectations at many of the loci. Of the 18 loci, only two met Hardy-Weinberg expectations after a sequential Bonferroni correction for simulta neous tests was performed. Bonferroni corrections reduce the type I error rate (i.e., falsely rejecting the null hypothesis), and were used for accepting or rejecting the null hypo thesis of equilibrium based on the p-value, P HWE in the tables (Rice 1989). Pair wise comparisons of loci by GENEPOP to test for linkage equilibrium showed that seve ral loci pairs were in dise quilibrium. After 153 pairwise comparisons ( n ( n -1)/2) were performed, three pairs of loci, TmaSC5 and TmaE1, TmaSC5 and TmaE7, and TmaSC5 and TmaK01, were estimated to be in linkage disequilibrium after a sequential Bonferroni co rrection (Rice 1989). MICRO-CHECKER suggested that 6 of 18 loci demonstrated evidence of null alleles, based on a general excess of homozygotes. The average observed heterozygosity ( HO) for the entire dataset was 45.0% w ith an average of 5.3 alleles per locus ( NA) observed across all 18 loci. The diversity statistics were calculated for the 362 samples grouped by their management unit or coast of collection, which were comp iled for comparison with the seasonal datasets

PAGE 57

57 (Appendix D, Tables D-1 and D-2) When the 362 samples were partitioned by the management unit or coast of collection, more lo ci met Hardy-Weinbe rg expectations. Average heterozygosity values and average numbe r of alleles did not he lp to differentiate among the regions. Overall values were very sim ilar for each unit or coast overall (Table 3-4). Partitioning the data by season also did not show great differences in the average values (Tables 3-5 and 3-6). Some individual lo ci showed differences between popul ations. In the analysis of the overall dataset, five loci showed differences in observed heterozygosity ( HO) of 0.1 or greater. In the summer subset th ere were eight loci, and in the winter subset there were seven loci with differences of 0.1 or greater. The loci that differed most frequently were TmaE26, TmaF14, TmaSC5, TmaE1, and TmaK01 (Appendix D, Tables D-1, D-2, D-3, D-4, D-5, and D6). The presence of private or rare alleles and examining allele frequency histograms were not informative among the units or coasts overall or when the genotypes were grouped by season. Although some rare alleles were present, their appearance did not help to identify a region, which can be observed in the allele frequenc y histograms produced for the overall dataset and the seasonal subsets (Appendix D) Loci TmaE11 and TmaJ02 were estimated to be in linkage disequilibrium in the winter subs et, and no pairs of loci were in disequilibrium in the summer subset. Several loci showed evidence of null alleles in the seasonal subsets, with the locus TmaKb60 identified most frequently for evidence of null alleles. Individual Level Analyses STRUCTURE was used in an attempt to assign in dividuals to genetic clusters without a priori designation of the population stru cture. The estimated posterior probabilities were used to calculate K the optimal number of clusters. When the entire dataset was analyzed in STRUCTURE, the optimal value for K was suggested to be four. When the dataset was subdivided

PAGE 58

58 by season, the optimal value for K in the winter was estimated as three (Table 3-7). Bar plots created by STRUCTURE for each analysis showed a high degr ee of admixture within individuals and populations (Figure 3-3). For the summer se ason, the optimal number of clusters was one (Table 3-7). Neighbor-Joining trees were produced for the su bdivided datasets from pairwise genetic distance matrices based on the ln(PSA) and the DC distances. Both distance measures produced trees with similar topology. The DC distance was chosen for pres entation based on previous simulation studies from the literatu re. It has been shown that the DC distance produces microsatellite trees with prope r topology with even a small nu mber of microsatellite loci (Takezaki and Nei 1996). All of the individual-based ge netic distance trees that were produced had similar topology (Figures 3-4 and 3-5). The trees for the males a nd females in the winter showed sligh tly stronger clustering patterns (Figure 3-4), bu t neither set of trees showed di stinct clustering of individuals from a region. The MDS analysis was completed in an attemp t to visualize patterns in the data without forcing them into a bifurcating system, as with the Neighbor-Joining tree s. MDS is a powerful set of techniques that computes coordinates fo r a set of points in space where the distances between the pairs of points co rrespond to the measured dissimilarities between pairs of individuals or objects (SYSTA T Software, Inc. 2007). The MDS analyses of the manatee genetic distance matrices for the complete dataset and for the seasonal datasets were not robust. The stress values for the manatee genetic distance data were not optimal, and no distinct clusters were observed from the MDS plots. The configurations were cal culated to be no better than

PAGE 59

59 arbitrary (Quinn and Keough 2002), and therefore results from the MDS are not presented herein. Population Level Analyses The population level genetic distance-based Ne ighbor-Joining trees fo r the winter showed males and females from each unit clustering w ith high bootstrap support (Figure 3-6A). The clustering of males and females from the same regions and the clusteri ng of management units from the same coast suggest that subtle geneti c differences may exist within and between the coasts. The population level genetic distan ce-based Neighbor-Joining trees for the summer season did not show robust clustering patterns and had low bootstrap sup port, which indicates that the clustering configurati on was not strong (Figure 3-6B). FST and RST values were calculated for the dataset to examine the level of genetic subdivision among the regions. Global FST values were calculated to estimate the proportion of genetic variation in the populations due to s ubdivision and genetic differentiation among them. The global FST values for the summer were 0.00735 when the data were subdivided by coast, and 0.00668 when the data were subdivide d by management unit. The global FST for the winter subset were higher with values of 0.01465 when the data were subdivided by coast, and 0.02261 when the data were subdivided by management unit. These values indicate that the proportion of variation in the population due to genetic subdivision is low; howeve r there is greater subdivision in the winter than in the summer. Pairwise FST and RST values were calculated among the four management units and between the east and west coasts in the two seasons to estimate relationships among the regions. For the winter subset, fi ve of the six pairwise FST values among the management units were significant after 10,000 permutations of th e data (Table 3-8), and the pairwise FST value was also significant between the coasts af ter 10,000 permutations of th e data. Only the pairwise FST

PAGE 60

60 between the Atlantic and the Southwest manage ment units was significant during the summer, whereas almost all pairwise FST values between management un its were signifi cant during the winter. Pairwise FST values between the coasts were significant in both seasons, but RST was only significant between the coasts in the winter. RST was significant among the Atlantic and Northwest and the Atlantic and Southwest management units only in the winter. Although some values were significant, all values were very low, representing high levels of gene flow and low genetic differentiation. Discussion When the dataset was examined as a whole, departures from equili brium (Hardy-Weinberg and linkage) were apparent. A lthough departures from equilibr ium and excesses of homozygotes could signal null alleles and issues with the loci or the dataset, they can also be signals of a Wahlund effect. A Wahlund effect results from including individuals from multiple populations in an analysis intentionally or due to cryptic, or unidentified, population structure. Departures from equilibrium, such as an excess of homozygo tes, can also be signals of inbreeding in the population. When the data were divided into the seasonal subsets and partitioned by management unit or coast, many loci met the equ ilibrium expectations. This signaled that the departures from equilibrium may have been caused, at least in part, by population substructure. This study presents further evidence that Florid a manatees have a low to average level of genetic variation. Recent meta -analyses of microsatellite gene tic variation data suggest that healthy mammalian populations have an average heterozygosity of 0.6 to 0.7, whereas demographically challenged mammalian populations have an average heterozygosity of 0.5 to 0.6 (DiBattista 2007; Garner et al .2005). This study estimated that Florida manatees have an average microsatellite heterozygosity of 0.45 over 18 loci. The number of alleles per locus is another measure of genetic varia tion that represents the genetic potential of a population, but is

PAGE 61

61 more sensitive to small samples sizes than hetero zygosity. It has been es timated that the average number of alleles in undisturbed, healthy mamma lian populations approaches 9 alleles per locus, and just over 6 for challenged populations (DiBa ttista 2007). The average number of alleles per locus across the 18 polymorphic loci screened on Fl orida manatees was only 5.3. Both average heterozygosity and the average number of allele s per locus were sma ller than the average observed for healthy and demogr aphically challenged mammalian species. This corresponds to the other surveys of the manatees genetic dive rsity, and could be a re sult of a l ong-term or historical disturbance which caused a population decline or du e to a founder effect (GarciaRodriguez et al. 1998; Garcia-Rodriguez et al. 2000; McClenaghan and O'Shea 1988). Longterm disturbances are thought to cause steeper de clines in genetic variation (Lande 1988) and historical disturbances have been associated with lower levels of genetic variation than more recent disturbances (DiBattista 2007). There is a growing debate ove r the importance of genetic va riation to the survival of populations (Avise 2004; Caro and Laurenson 1994) and whether es timates of genetic diversity from neutral molecular marker s reflect adaptive differences am ong population and their ability to respond to environmental changes (DiBattist a 2007; Reed and Frankha m 2001). Some species seem to be unaffected by extremely low genetic variation (Northern El ephant Seals (Weber et al 2004)), whereas other species show many detrimen tal effects of inbreeding (Florida panthers (Roelke et al. 1993); cheetahs (OBrien 1994)). Florida manatees have a fairly large population size for an endangered species, and do not seem to be currently suffering from the effects of inbreeding, but further studies should focus on assessing the genetic health of the population, including assessing neonatal/calf mortality and examining the extent to which in breeding is occurring in the wild.

PAGE 62

62 The genetic data are in agreement with the observations from MIPS, telemetry, and other studies that documented the Florida manatees winter site fidelity. The population-level Neighbor-Joining trees show cluste ring of groups of males and fe males from each unit and coast with strong bootstrap support (Fig ure 3-6A). Subtle, but stat istically signif icant population genetic structure exists during the cooler months, as indicated by pairwise tests for FST and RST (Tables 3-8 and 3-9). Little gene tic subdivision exists during warm er months when there is more migration and mixing among the regions, indicated by the weak bootstrap support and poor clustering in the population level Neighbor-Joi ning trees (Figure 3-6B). Pairwise FST and RST values were not significant among the manageme nt units in the summ er, but the pairwise FST value was significant between the co asts in the summer. This coul d suggest there is a possibility of very weak genetic differentia tion between the coasts in the su mmer, as well. Although some values were significant, all pairwise FST and RST values were very low. This is further evidence of the high gene flow and low genetic differentiation among the regions. If no clear geographic barriers to migration exist, delineation of the population structure using the multilocus genotypes without a priori designation of subpopulations should be attempted first. Tests that are based on a priori samples can overlook cryptic population structure if there are actually more populations or the populations have different boundaries than were designated by the investig ator (Waples and Ga ggiotti 2006). Although several methods exist for examining the dataset on the individu al level, many traditi onal population genetics statistics require a priori designation of different subpopulations to allow for the testing of specific hypotheses. Analyses were performed for this study at the individual and population level, but because the signal of population structure was so weak for Florida manatees, the individual-based methods showed high genetic connectivity among the regions.

PAGE 63

63 STRUCTURE is a widely cited and popular prog ram for estimating the number of subpopulations and identifying popul ation clusters at the indi vidual level. Although STRUCTURE suggested that there were multiple population cl usters for the whole dataset and the winter subset, the authors of the software caution that certain signals could indicate the results are actually an artifact of the softwares assumptions of linkage equilibrium and Hardy-Weinberg equilibrium. For example, if no individuals are strongly assigned to a certain cluster, the result could be an artifact of the program attempting to divide the individuals from one population into several clusters. Another artif actual signal is when th e proportion of individuals assigned to the different population clusters is approximately proportional to the number of assumed population clusters (1/ K ) (Pritchard and Wen 2004). In each case for the Florid a manatees, the individuals were approximately divided into 1/ K clusters and all individu als showed a high degree of admixture (Figure 3-3). Simulation studies have demonstr ated that clustering methods are not effective for species with moderate to high rates of gene flow, even with very long runs of the program (up to 4,000,000 MCMC iterations) (Waples and Gaggiotti 2006). The type of analysis that STRUCTURE uses may not be ideal for attempting to delineate populations with high genetic connectivity within a subspecies, and may be more appropriate fo r delineating between subspecies (T. L. King, USGS Leetown Science Center, Personal Communication). Manatees are large animals that can have ve ry long migratory routes There have been many documented instances of manatees traveling great distances (Fertl et al. 2005). Through MIPS and telemetry studies, it is known that manatees can travel between the Southwest and Atlantic management units through the canals as sociated with Lake Okeechobee and through the Florida Keys (C. A. Beck, USGS Si renia Project, Pers onal Communication).

PAGE 64

64 There have been several documented instances of manatees trave ling great distances outside of their typical range. In 1994, Chessie, an adul t male manatee, was rescued from Chesapeake Bay. He was radio-tagged and returned to manatee habitat off Floridas east coast. He wintered in the Port Everglad es Power Plant discharge, and trav eled north again in the spring. He traveled as far north as R hode Island by the summer, and returned back to the Florida waters by the fall (Deutsch et al. 2003; Reep and Bonde 2006; USGS 2001). In 1996, Sweetpea was captured in Texas, reha bilitated, radio-tagged, and released in Homosassa River, Florida. In the year that sh e carried the transmitter, Sweetpea traveled from the Homosassa River northwest to the Apalach ee and Apalachicola Bays of the Florida Panhandle. She remained there through the sp ring and summer, and then headed south to Marathon in the Florida Keys in November. She spent the wint er farther northeast in south Miami. By March 1997, Sweetp ea had traveled to Brevard C ounty on the central east coast of Florida. She was last tracked in th e Banana and Indian rivers (Deutsch et al. 2003; Weigle et al. 2001). These individuals are just a few examples of the long-distance migr ation capabilities of manatees. In the summer of 2006, confirmed mana tee sightings were report ed in New York City and other locations in the northeastern United States (CNN.com 2006a). There are also many Florida manatees that have been sighted in areas west of Florida (Fernandez and Jones 1990; Fertl et al. 2005; Gunter 1944; Schiro and Fertl 1995) Recently, a known adult female Florida manatee and her calf were sighted in Cuba. This individual had an extensive MIPS sighting history in the Crystal River region since 1979 (Alvarez-Alemn et al. 2007). Another MIPSknown Florida manatee became resident in the Ba hamas (Reid 2000). Some dispersal may also be due to hurricanes and tropical storms. Decreased su rvival estimates for Florida manatees in

PAGE 65

65 years with increased storm activity could be due to manatees being carried by currents to places they could not regularly disp erse (Langtimm and Beck 2003). Manatees have also been transported around the state of Florida through human interaction. In 1971, Sewer Sam was rescued from the sewers of Miami and rehabilitated at the Miami Seaquarium. Jacques Cousteau and his team released the large male; however, instead of releasing him near th e area he was found, he was rel eased in northwest Florida in Three Sisters Spring, Crystal River (Cousteau and Kaufman 1989). Other rescued/rehabilitated manatees have been released in Blue Spring State Park after periods of captivity. Three individuals that are included in the captive pedigree reconstr uction study (Chapt er 4) were recently released in Blue Spring State Park. G ene, rescued from Brevard County, mated with another manatee while in captivity to produ ce Dundee. Both Gene and Dundee were released in Blue Spring in the Upper St. Johns management unit in February 2007. The third individual from the genetic pedigr ee reconstruction study that was released at the same time and location was Stoneman. Stoneman is the inbred offspring of Romeo and his daughter, Aurora. Releasing inbred mana tees is unlikely to be favorable for maintaining the genetic health of the population. The frequent long migrations, as well as hum an interaction, have allowed manatees to reproduce with animals from various locations around the state. The winter site fidelity is strong enough to produce subtle population genetic structure among the management units and between the coasts during the cold months. Marine organisms tend to have greater gene tic connectivity, resulting in more subtle patterns of genetic structure (Waples 1998). Th e population genetic resu lts from the Florida manatee study are not unusual for marine organism s. Genetic studies of the worlds population

PAGE 66

66 of Hooded seals found in the Northwest Atlantic s uggest relative panmixia. Hooded seals in this region are likely the product of a recent recol onization event following th e last glacial period (Coltman and Slate 2003), which, as in manatees may not have been sufficient evolutionary time for significant genetic differentiation to occur. The manatee social structure is more similar to that of many whale species than phocids. Bel uga whales migrate seas onally and stocks have been identified based upon their summer dist ribution (Reeves and Mitchell 1989; Smith and Hammill 1986). Genetic studies of the Beluga whale population show genetic substructuring that corresponds to the summer distribution and genetic homoge neity, or mixture, during the winter months (Brown Gladden et al. 1999). Recent studies of Bowhead whales in Alaskan waters have revealed genetic va riability, but have been unable to resolve population structure, possibly due to high levels of gene flow and a sub-optimal sampling scheme (Jorde et al. 2007). Manatees offer a unique paradigm for marine organisms. Decades of research have produced a vast body of knowledge describing thei r natural history, de mography, and population distribution. Many of the locati ons manatees frequent are acce ssible by humans, allowing for the collection of samples. By understanding the ge netic interactions of Florida manatees more thoroughly, we can apply the techniques and compare the results to other, less accessible species. Investigations of the genetic diversity of the sp ecies should continue in order to monitor the genetic health of the Florida population and to monitor changes in the population genetic structure due to alterations to the manatees natural habitat.

PAGE 67

67 Figure 3-1. Warm-water aggrega tion. Florida manatees must tr avel to warm-water sources to thermoregulate during the wint er. (Photograph courtesy of USGS, Sirenia Project.) Table 3-1. Manatees used for the calculation of di versity and F-statisti cs. A total of 362 individuals were used for the calculations. Individuals were partitioned by Season the season of sample collection and Sex sex of the individual. Total the total number included in each seasonal dataset; Units the number of indi viduals included from each management unit ( S Southwest; N Northwest; J Upper St. Johns River; A Atlantic). Season Total Sex Units (S) 18 (N) 32 (J) 23 (F) 96 (A) 23 (S) 23 (N) 42 (J) 17 Winter 201 (M) 105 (A) 23 (S) 32 (N) 3 (J) 4 (F) 71 (A) 32 (S) 24 (N) 5 (J) 11 Summer 161 (M) 90 (A) 50

PAGE 68

68 Figure 3-2. Florida manatee popu lation distribution among regions used to sub-sample the dataset. Colors correspond to the locations on the map. The majority of the manatee population in Florida inhabits the Atlantic and Southwest management units. The number of manatees that live on the east and west coasts is approximately equal. The estimated population composition was base d on the highest mini mum winter counts for the regions from 1996-1999 (USFWS 2001). Table 3-2. Manatees used in sub-sampled da taset for the winter season individual genetic distance-based Neighbor-Joi ning trees. Individual s were subdivided by sex sex of the individual. Total the total number included in the dataset; Units the number of individuals included from each management unit ( S Southwest; N Northwest; J Upper St. Johns River; A Atlantic). Total Sex Units (S) 18 (N) 6 (J) 2 (F) 49 (A) 23 (S) 20 (N) 7 (J) 3 102 (M) 53 (A) 23

PAGE 69

69 Table 3-3. Diversity statistics for the 18 loci on th e set of 362 manatees. Those loci with evidence of null alleles are indicated with an *. N Number of indi viduals genotyped; NA, number of alleles observed, HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibr ium, values in italics are not in equilibrium after a sequential Bonferroni correction. Locus Name N NA HO HE P HWE TmaA02 358 5 0.3436 0.3811 0.0017 TmaE02* 362 5 0.4282 0.5122 0.0000 TmaE08 362 7 0.5774 0.6042 0.0000 TmaE11 362 8 0.6492 0.6865 0.0000 TmaE26* 351 3 0.1624 0.1971 0.0022 TmaF14 362 4 0.3812 0.4148 0.0009 TmaM79 357 3 0.4958 0.4849 0.0003 TmaSC5 339 11 0.6844 0.7139 0.0000 TmaSC13 346 7 0.5058 0.4937 0.0002 TmaE1* 335 7 0.4627 0.5265 0.0003 TmaE4 343 3 0.2857 0.3071 0.0000 TmaE7* 342 4 0.5614 0.6445 0.0001 TmaH13 305 4 0.3541 0.3499 0.6731 TmaE14 268 7 0.5560 0.6035 0.0001 TmaH23 314 2 0.2006 0.2058 0.5866 TmaK01 333 7 0.6426 0.4952 0.0000 TmaJ02* 357 4 0.3698 0.4145 0.0000 TmaKb60* 344 4 0.4390 0.6170 0.0000

PAGE 70

70 Table 3-4. Average measures of genetic vari ation over 18 microsatellite loci for Florida manatees. Results are provided for Flor ida samples from both seasons. The measures of variation (mean 1 s.d.) are: HO, average observed heterozygosity, HE, average expected heterozygosity, and NA, average number of alleles per locus. Population HO HE NA All Florida samples 0.450 0.146 0.481 0.149 5.28 2.231 East coast 0.457 0.146 0.481 0.126 4.56 1.921 West coast 0.442 0.157 0.475 0.18 4.83 2.007 Atlantic 0.461 0.148 0.484 0.123 4.00 1.563 St. Johns River 0.450 0.155 0.472 0.140 4.00 1.667 Northwest 0.451 0.183 0.468 0.185 4.00 1.563 Southwest 0.436 0.150 0.474 0.179 4.33 1.826 Table 3-5. Average measures of genetic vari ation over 18 microsatellite loci for Florida manatees. Results are provided for Florid a samples from the winter season. The measures of variation (mean 1 s.d.) are: HO, average observed heterozygosity, HE, average expected heterozygosity, and NA, average number of alleles per locus. Population HO HE NA All winter samples 0.455 0.158 0.481 0.156 4.89 2.079 East coast 0.464 0.154 0.477 0.132 4.33 1.732 West coast 0.450 0.172 0.477 0.184 4.39 1.830 Atlantic 0.469 0.168 0.475 0.135 3.61 1.253 St. Johns River 0.459 0.150 0.479 0.131 3.89 1.560 Northwest 0.449 0.186 0.467 0.186 3.94 1.615 Southwest 0.454 0.176 0.478 0.183 3.83 1.607 Table 3-6. Average measures of genetic vari ation over 18 microsatellite loci for Florida manatees. Results are provided for Florid a samples from the summer season. The measures of variation (mean 1 s.d.) are: HO, average observed heterozygosity, HE, average expected heterozygosity, and NA, average number of alleles per locus. Population HO HE NA All summer samples 0.443 0.136 0.478 0.143 4.39 1.768 East coast 0.452 0.146 0.481 0.126 3.94 1.615 West coast 0.430 0.142 0.469 0.179 3.94 1.810 Atlantic 0.456 0.148 0.486 0.123 3.89 1.595 St. Johns River 0.427 0.207 0.456 0.179 2.72 0.803 Northwest 0.48 0.206 0.474 0.201 2.67 1.000 Southwest 0.421 0.146 0.467 0.181 3.83 1.772

PAGE 71

71 Figure 3-3. STRUCTURE bar plots. A) All Florida ma natee samples from all seasons, K = 4. B) All Florida manatee samples from the winter season, K = 3. Individuals are represented by vertical bars, and the different colo rs in a bar represent the proportion of admixture, or ancestry, from a certain infe rred genetic cluster. The Y-axis represents Q the proportion of admixture. In this analysis, each vertical bar consists of every color, which represents an i ndividual being assigned to all of the populations. This demonstrat es the high amount of admixt ure within individuals and among the different clusters.

PAGE 72

72 Table 3-7. STRUCTURE results for inferring the number of distinct popul ation subgroups for all Florida samples and for the dataset subdivi ded by season of sample collection. K number of clusters, lnPr(X|K) average value over the thr ee independent runs for the probability of the data given by STRUCTURE, Pr(K|X) posterior probability of the data. All Florida Winter Summer K lnPr(X| K ) Pr( K |X) lnPr(X| K ) Pr( K |X) lnPr(X| K ) Pr( K |X) 1 -11125 0.0000 -6094 0.0000 -4647.9 1.0000 2 -11031 0.0000 -6116 0.0000 -4668.4 0.0000 3 -11026 0.0000 -6043 1.0000 -4708 0.0000 4 -10997 1.0000 -6090 0.0000 -4786 0.0000 5 -11111 0.0000 -6183 0.0000 -4881.5 0.0000 6 -11228 0.0000 -6354 0.0000 -5114.7 0.0000 7 -11285 0.0000 -6458 0.0000 -5155.9 0.0000 8 -11321 0.0000 -6708 0.0000 -5203 0.0000 9 -11388 0.0000 -6673 0.0000 -5190.6 0.0000 10 -11504 0.0000 -6731 0.0000 -5501.7 0.0000 11 -11746 0.0000 -6907 0.0000 -5398.2 0.0000 12 -12022 0.0000 -7029 0.0000 -5490.8 0.0000

PAGE 73

73 Figure 3-4. Neighbor-Joining trees for sample s from individuals collected during the winter season. The trees are based on ind ividual pairwise Cavalli-Sforza and Edwards DC distance for A) males and B) females. Colors represent the different management units, Red Atlantic, Black Northwest, Green Southwest, Blue Upper St. Johns River. The scale bar located at the bottom left of each tree represents DC distance units.

PAGE 74

74 Figure 3-5. Neighbor-Joining trees for samp les from individuals collected during the summer season. The trees are based on individual pairwi se Cavalli-Sforza and Edwards DC distance for A) males and B) females. Colors represent the different management units, Red Atlantic; Black Northwest; Green Southwest; Blue Upper St. Johns River. The scale bar located at the bottom left of each tree represents DC distance units.

PAGE 75

75 Figure 3-6. Neighbor-Joining trees for populatio ns collected during the different seasons. The trees are ba sed on individual p airwise Cavalli-Sforza and Edwards DC distance for the A) winter, B) summer seasons. The management units are abbreviated as follows: StJR Upper St. Johns River; ATL Atlantic; NW Northwest; and SW Southwest. The popul ations were further subdivided into males, M and females, F

PAGE 76

76 Table 3-8. Pairwise FST and RST values among the management unit populations. FST values are below the diagonal, and RST values are above the dia gonal. Values that were significant after a Bonferr oni correction with 10,000 permutations are denoted by italics. Winter Atlantic St. Johns River Northwest Southwest Atlantic 0.0137 0.0274 0.0273 St. Johns River 0.0066 0.0074 0.0032 Northwest 0.0252 0.0239 0.0177 Southwest 0.0221 0.0316 0.0307 Summer Atlantic St. Johns River Northwest Southwest Atlantic -0.0109 -0.0167 0.0016 St. Johns River 0.0026 -0.0322 0.0052 Northwest 0.0151 0.0090 0.0057 Southwest 0.0063 0.0089 0.0009 Table 3-9. Pairwise FST and RST values between Floridas east and west coast populations. FST values are below the diagonal, and RST values are above the diagonal. Values that were significant after a Bonf erroni correction with 10,000 permutations are denoted by italics. Winter East West East 0.0080 West 0.0147 Summer East West East 0.0003 West 0.0074

PAGE 77

77 CHAPTER 4 FLORIDA MANATEE PEDIGREE RECONSTRUCTION Introduction Florida manatee family relationships are cu rrently constructed us ing photo-identification sighting data. These pedigrees can identify only the mother-offspr ing (cow-calf) relationships. Once calves are weaned and independent, they are usually no longer id entifiable unless they acquired scars or other distinguishing marks while associated with their mothers. Identification of the father can also prove problematic. The manatee mating system has been termed scramble promiscuity (Anderson 2002), whic h stems from the observation that female manatees in estrous can be pursued by a herd of up to 22 males. Any of these males may successfully copulate with her over the cour se of up to a month (Rathbun et al. 1995). This mating system makes it difficult to establish wild manatee br eeding pedigrees (Figure 4-1). If samples for genetics (skin biopsy, hair, feces) are collected from the males in the breeding herd, and the female was photographically identifiable, it woul d be possible to subject the individuals to paternity analysis once the offspring is born. Microsatellites are commonl y used for pedigree analysis in humans (Buckleton et al. 2005) and domesticated species such as bison (Schnabel et al. 2000) and dogs (Ichikawa et al. 2001). Pedigree analysis has also been used in popula tions of wild animals, including harbor seals (Hayes et al. 2006) and koalas (Houlden et al. 1996). The purpose of the work reported here was to determine if the panel of 18 microsatellite markers for individual identif ication (Chapter 2) is suffi cient for breeding pedigree reconstruction. This chapter describes the testing of statis tical methods on genotypic data collected from manatees of known pedigree to test the u tility of the markers in future pedigree reconstruction. Allele shari ng methods were used to delineate family units and maximum

PAGE 78

78 likelihood methods were used to estimate possi ble family relationships. Accurate pedigree reconstruction would permit reconstruction of multi-generational pedigrees, estimation of inbreeding depression, and determination of the genetic health of the wild Florida manatee population. Materials and Methods Sample Collection and DNA Isolation A total of 15 blood samples from individuals of known captive-bred family units was obtained from the USGS, Sireni a Project (USFWS wildlife res earch permit MA791721/4, issued to the USGS, Sirenia Project). The blood samp les were stored in blood lysis buffer (100 mM Tris-HCl, 100 mM EDTA, pH 8.0, 10 mM NaCl, 1% SDS) at room temperature (White and Densmore 1992). Each sample was labeled with a unique identifier, blindi ng the experimenter to eliminate bias. Standard phenol/chloroform isolations were performed to obtain DNA from the blood samples (Hillis et al. 1996). Genotyping PCR was performed in 20 l reactions as previo usly described (Chapter 2). Fragment data were analyzed using GENEMARKER, version 1.4 (Soft Genetics, St ate College, PA). Allelic information was stored in a Microsoft Access data base developed for Florida manatees, which is maintained by the author and the Mc Guire laboratory group (Chapter 2). Statistical Analyses An allele sharing test was used to identify possible fam ily groups. The proportion of shared alleles (PSA) was the allele sharing measure used for this study. These pairwise genetic distances were calculated as the ln(PSA) by MICROSAT, version 1.5d (Minch 2007). MICROSAT generated a genetic distance matrix, which was subjected to the NEIGHBOR algorithm in the program PHYLIP, version 3.66, for the developm ent of Neighbor-Joining trees (Felsenstein

PAGE 79

79 2004). The tree files generated by P HYLIP were visualized using TREEVIEW, version 1.6.6 (Page 1996). Maximum likelihood methods were used to iden tify possible relations hips. The program, ML-RELATE (Kalinowski et al. 2006b) was used to attempt to reconstruct the relationships without a priori knowledge of the pedigree. ML-RELATE calculated relatedness coefficients from the data and employed maximum likelihood algorithms to account for the presence of null alleles and genotyping errors in the data set (Wagner et al. 2006). The primer pairs for loci TmaE1 and TmaKb60 were previously identified as having possible nu ll alleles (Chapter 2). To account for this, two sets of calculations were perfo rmed. First, the loci with null alleles were not specified, and second, those loci with null alleles were identified, allowing for modifications to the algorithms by the ML-RELATE software. ML-RELATE was also used to calculate confidence sets for the different relationships through simulation. The number of random genotypes created was set to 1000 for each simulation completed. The simulations were perf ormed to produce a list of relationships that would be possible within a 0.05 significance level. Results The allele sharing method allowed for the delinea tion of three discrete family units within this captive-bred group. These tests were em ployed because they are somewhat insensitive to genotyping errors and quickly provi de easily interpreted results. Individuals who are related to each other will share a certain proportion of allele s. Those who are unrelated will share fewer alleles than expected, whereas i ndividuals who are inbred will sh are more alleles than expected (Blouin 2003). These proportions ar e relative to the type of marker s used, the particular set of markers used, and the study population of animal s. The allele sharing analysis clearly distinguished three cluste rs (Figure 4-2). Family Unit 1 is comprised of two smaller clusters

PAGE 80

80 containing a total of eight i ndividuals. Family Unit 2 cont ains three closely clustered individuals. Family Unit 3 also contains three individuals, but with a weaker association than Family Unit 2. Maximum likelihood methods are more sensitiv e than allele sharing methods, but still allow for a dataset in which all of the relations hips are unspecified. With this approach, the relationships among individuals were not intrinsically obvious (Table 4-1). The identities of the individual samples were made available afte r the maximum likelihood tests were performed (Table 4-2), and the reconstruc ted genetic relationships were compared to the true pedigree (Figure 4-3) to validate the genetic approach. The maximum likelihood approach correctly identifi ed related and unrelated pairs, but specific relationships were not al ways clear. The greatest numb er of incorrectly identified relationships was in Family Unit 1. In these pairs, the true relationship was not within the 0.05 significance level (Table 4-1). Two cases of note were relationship estimates between the grandmother and grandchild (e. g. multiple generations) of two separate consanguineous mating events between daughters and their father. The relatedness coefficient for a grandparentgrandchild relationship is equi valent to a half-s ibling relationship (B louin 2003). The MLRELATE program does not have the option to reconstruct multi-generational pedigrees, but the half-sibling relationship categor y would be the equivalent of a grandparent-grandchild relationship and was considered a correct relatio nship assignment. There were other instances where inbreeding was a factor that did not yield proper relationship assignments. For example, a consanguineous mating between P-3 (Romeo) and P-5 (Lorelei) produced P-21 (Hugh). Although the designated relations hips among Romeo, Lorelei, and Hugh were correct, Hughs relationships with his other i nbred half-siblings were consis tently estimated as unrelated,

PAGE 81

81 although the half-sibling relationshi p was usually within the 0.05 significance level. Another important discrepancy was between the known mo ther-offspring pair, P4 (Juliet) and P-20 (Hurricane). The maximum likelihood approach excluded the parent-offspr ing relationship as a possibility. For Family Unit 2, it was assumed that P-60 (Gene) and P-130 (Rita) were not immediately related. They mated and pr oduced P-23 (Dundee). Dundee and Gene were correctly identified as parent-offspring. Dund ee and Ritas most likely relationship was not parent-offspring; however the parent-offspri ng relationship was within the 0.05 significance level. In Family Unit 3, P-35 (Ocean Reef) was the mother of twins P-33 (Patch, male) and P-34 (Pumpkin, female). The relationship between Oc ean Reef and Pumpkin was parent-offspring in every simulation. Ocean Reef and Patchs most likely relationship was not identified as parentoffspring, although the parent-offspring relationshi p was within the 0.05 significance level. The fraternal twins, Patch and Pumpkin, were expe cted to be full-siblings. The most likely relationship between these two indi viduals was identified as half-s ibling or unrelated in many of the simulations; however the full-sibling relatio nship was within the 0.05 significance level. The modification of the algorithm to accommodate null alleles did not significantly change the results, and therefore only the results withou t nulls are presented. There were only two differences between the two sets of analyses, both in Family Unit 1. First, the most likely relationship for individu als P-273 (Stoneman) and P-26 (Buffett ) was changed from unrelated to half-sibling with the adjustment for null alleles. Secondly, although the li kely relationship for P3 (Romeo) and P-26 (Buffett) was still identifie d as unrelated, the true relationship (parentoffspring) was possible within th e 0.05 significance level with th e adjustment for null alleles.

PAGE 82

82 Discussion The data indicate that the current panel of 18 microsatellite markers is sufficient for distinguishing known family units. Pedigree rec onstruction was attempted, but not completely successful. In all the po ssible relationships, thos e individuals who were in different family units were identified as unrelated, although other rela tionships were sometimes significant. It is possible that the discrepancies of the actual relationships and the maximum likelihood estimated relationships can be explained by genotyping error, undetected null alleles, or the power of the current panel of markers. Inbr eeding can also create a number of problems for identification of genetic relationships. Some of the individuals sampled were products of consanguineous mating (Figure 4-3). The number of alleles identical by descent, and therefor e shared between close relatives, is increased in cases of inbreeding. If the inbred i ndividuals are not specified when analyses are completed, the estimated relationshi ps can be skewed. For example, a full-sibling would be expected to share approxi mately half of his alleles with his brother. In the case of inbreeding between a daughter and fa ther, the daughter already shares alleles that are identical by descent with her father. Their offspring will shar e more alleles that are identical by descent from the father than expected with his family members (mother/sister other siblings, etc.) (Blouin 2003). The maximum likelihood approach used in this study is insufficient for resolving cases of inbreeding, especially without base line normal relationship information. One discrepancy of note is be tween the dizygotic, or frater nal twins P-33 (Patch) and P-34 (Pumpkin). It is clear that Pa tch and Pumpkin are dizygotic twin s because Patch is a male and Pumpkin is a female. Monozygotic, or identical, twins must be of the same sex. Dizygotic twins are known to have the same mother, which could re present a half-sibling (d ifferent fathers) or full-sibling (same father) relationship. The mo st likely relationship be tween Patch and Pumpkin was consistently identified as ha lf-sibling in the simulations pe rformed. Twins are thought to

PAGE 83

83 occur in between 1.4-4% of births in the wild to Florida manatees (Rathbun et al. 1995). The proportion of these twin births that are monozygotic and dizygotic is unknown, and it is also unknown if multiple paternity can occur in the case of dizygotic twins in manatees. Based on morphological evidence, it has b een suggested that manatees may exhibit sperm competition (Reynolds III et al. 2004). Sperm competition is a characte ristic of males in promiscuous mating systems. When mating occurs between a female and multiple males, the ejaculate of these males can remain in the females reproductive tract. The sperm will then essentially compete for fertilization of the ova (Dean et al. 2006). Multiple paternity coul d be a possibility in manatees, which is consistent with the half-sibling relationship identified for Patch and Pumpkin. However, the odd result for Patch and Pumpkin coul d also be due to the ty pical sources of error (genotyping error, presence of undete cted null alleles). Further inve stigation into th e genetics of this family unit is necessary. Additional family units containing twins should also be examined to investigate further the possibility of multiple paternity in Florida manatees. Similar approaches to the ones presented here can be used in future studies attempting to identify related manatees from field samples. For example, the allele sharing method might be useful in the identification of extended fa mily groups, and the maximum likelihood method could be used for preliminary tests of relationships. Additional methods can be employed when the mother and offspring genotypes are known a nd a pool of putative fathers is supplied. Parentage exclusion could be used to exclude putative parents with alleles that could not have contributed to the offspring genot ype. The drawback of exclusion methods is that they are very stringent and do not typica lly allow for genotyping error, null alleles, or mutations (Jones and Ardren 2003). In addition, it is not always possible to exclude all but two candida te parents. Therefore, likelihood approaches mu st be used to identify the mo st likely, non-excluded parents.

PAGE 84

84 The pedigree information could be useful if captive breeding programs were to be established. Successful captive breeding programs are already es tablished for dolphins (Cornell et al. 1987; Robeck et al. 1994), orcas (Duffield et al. 1995), and other marine mammals. Currently, captive breeding is discouraged for Florida manatees, and mating in captivity is a serendipitous event (C. A. Beck, USGS, Sirenia Project, Personal Communi cation). Aquaria in Europe have been breeding Antillean manatees ( Trichechus manatus manatus ) from South America for several decades. After only a few gene rations of inbreeding, the offspring are sterile (R. K. Bonde, USGS, Sirenia Project, Persona l Communication). The genetic information from pedigree reconstruction could be useful to minimize the effects of inbreeding in captive populations. The ultimate goal of a pedigree study is not simply pedigree reconstruction. Once the pedigree information has been verified in the captive manatee pedigree, the methods can be used to reconstruct wild manatee pedigrees. This information will be of great value for many different population studies, includi ng estimating the level of inbreeding in the wild population. To study the genome-wide effects of inbreeding, the traditional method of calculating inbreeding coefficients from known pedigree s should be used (Pemberton 2004). Future studies for Florida manatees should focus on reconstructing pedigr ees using genetic data. Relatively few generations are required for accurate estimation of inbreeding coefficients (Coltman and Slate 2003; Keller et al. 2002; Pemberton 2004), and many sample s currently exist from well-studied family groups in the wild (R. K. Bonde, US GS, Sirenia Project, Pe rsonal Communication). The pedigree information will also be useful in estimating the reproductive success of males in the Florida population and the number of effective breeders in the population. The

PAGE 85

85 question of multiple paternity in di zygotic twins could also be reso lved. These future studies can give us clues about the health of the wild Florida manatee population. Figure 4-1. Herd of mating manat ees. Photographic-identification of individuals in the herd is difficult. If a copulatory event was observed, it is not necessarily the event that led to conception. If the female is photo-identifie d, she could be associated with the calf after birth. If skin scrapings, biopsy punches, or hair sa mples are taken from individuals in the herd, the samples could be subjected to genetic pedigree analysis once the calf is born and sampled. (Photogr aph courtesy of USGS, Sirenia Project.)

PAGE 86

86 Figure 4-2. Neighbor-Joining tree ba sed on the proportion of shared alleles (PSA). The branch length is proportional to the genetic distan ce, which was calculated as, ln(PSA). The names of the individuals correspond to those given in Table 4-3. Confirmed family units cluster together, as indicate d by the circles in the diagram.

PAGE 87

87 Table 4-1. Pedigree reconstructi on results using the maximum likelihood approach. Family Units are indicated by the dividing lines within the table. Ind1 Ind2 Individual identification numbers of the two indivi duals with relationship in question; Likely R relationship with the highest likelihood; LnL(R) log likelihood of r for the relationship with the highest likelihood; Delta Ln(L) how much lower the log likelihoods are for the other possible relati onships, 9999 indicates the relationship is not possible; 0.05 R other relationships possible at the 0.05 level of significance; True R actual relationship from observational data; D X indicates a discrepancy between the maximum likelihood estimate and the actual relationship. Delta Ln(L) Ind1 Ind2 Likely R LnL(R) U HS FS PO 0.05 R True R D P-13 P-20 PO -21.1 2.45 1.08 0.21 HS, FS, PO FS P-13 P-4 PO -25.9 1.35 0.49 1.07 U, HS, FS, PO PO P-21 P-13 U -28.02 1.11 3.82 9999 U, HS IHS P-21 P-20 U -28.11 0.41 3.35 9999 U, HS IHS P-21 P-26 U -29.21 2.41 6.76 9999 U IHS X P-21 P-273 U -30.15 0.07 1.39 9999 U, HS, FS IHS P-21 P-3 PO -24.86 3.07 1.26 1.04 HS, FS, PO IPO P-21 P-4 U -31.81 2.26 6.27 9999 U GMA X P-21 P-5 PO -24.46 3.97 1.73 0.8 HS, FS, PO IPO P-26 P-13 U -29.79 0.07 0.01 9999 U, HS, FS FS P-26 P-20 HS -24.44 0.3 0.72 9999 U, HS, FS FS P-26 P-4 PO -25.89 2.55 1.03 0.88 HS, FS, PO PO P-273 P-13 HS -28.14 0.39 0.4 9999 U, HS, FS IHS P-273 P-20 U -25.68 0.27 1.33 9999 U, HS, FS IHS P-273 P-26 U -32.9 0.33 1.79 9999 U, HS, FS IHS P-273 P-4 FS -28.05 1.33 0 9999 U, HS, FS GMA P-3 P-13 HS -26.29 0.03 1.34 0.26 U, HS, FS, PO PO P-3 P-20 U -23.47 0.29 1.89 0.86 U, HS, FS, PO PO P-3 P-26 U -30.69 0.73 3.06 9999 U, HS PO X P-3 P-273 HS -29.01 0.43 0.83 0.28 U, HS, FS, PO IPO P-3 P-4 U -27.17 1.49 4.44 9999 U, HS U P-4 P-20 HS -27.14 0.2 0.06 9999 U, HS, FS PO X P-5 P-13 U -25.84 0.51 1.83 1.3 U, HS, FS, PO FS P-5 P-20 U -23.97 0.14 1.13 0.54 U, HS, FS, PO FS P-5 P-26 U -30.21 1.19 4.07 9999 U, HS FS X P-5 P-273 U -28.95 0.49 2.72 1.48 U, HS, PO IHS P-5 P-3 FS -23.76 2.98 1.59 0.61 HS, FS, PO PO P-5 P-4 U -27.67 0.52 2.85 1.46 U, HS, PO PO P-23 P-130 FS -28.01 3.04 1.24 0.28 HS, FS, PO PO P-23 P-60 PO -28.11 3.77 1.42 0.61 HS, FS, PO PO P-60 P-130 HS -30.81 0.01 0.91 9999 U, HS, FS U P-34 P-33 HS -45.38 0.37 1.38 9999 U, HS, FS FS P-35 P-33 HS -35.17 0.2 1.87 1.02 U, HS, FS, PO PO P-35 P-34 PO -34.21 3.31 0.95 0.19 HS, FS, PO PO Relationship categories: U unrelated; HS half-sibling; FS full-sibling; PO parent-offspring; IPO inbred parent-offspring; IHS inbred half-sibling; GMA grandmother-grandchild.

PAGE 88

88 Table 4-2. Identification information for samples from captive manatees. Family Unit identification number that corr esponds to the family group; Pedigree ID lab identification number for the blinded study; Name name of the captive manatee; Sex male (M) or female (F); Location location at time of publication; Status living (L), deceased (D), unknown (U). Family Unit Pedigree ID Name Sex Location Status 1 P-3 Romeo M Miami Seaquarium L P-4 Juliet F Miami Seaquarium L P-5 Lorelei F Homosassa Springs Wildlife State Park L P-13 Foster M Released, Monroe County U P-20 Hurricane M Lowry Park Zoo L P-21 Hugh M Mote Marine Laboratory L P-26 Buffett M Mote Marine Laboratory L P-273 Stoneman M Released, Blue Springs State Park L 2 P-23 Dundee M Released, Blue Springs State Park L P-60 Gene M Released, Blue Springs State Park L P-130 Rita F SeaWorld Florida L 3 P-33 Patch M Miami Seaquarium D P-34 Pumpkin F Miami Seaquarium L P-35 Ocean Reef F Miami Seaquarium L

PAGE 89

89 Figure 4-3. Actual pedigrees of captive manatees Individuals are identified by their Family Unit number and Pedigree ID number (Table 4-3). Samples from two individuals were not provided for the genetics study, X and Y X is Aurora, a captive manatee from Family Unit 1, and Y is the unidentified father of the fraternal twins in Family Unit 3. The dotted lines in Family Unit 1 indicate the same individual (P-3) placed in several locations in the pedigree to accommodate consanguineous mating events, which are indicated by the solid double lines.

PAGE 90

90 CHAPTER 5 CONCLUSIONS AND IMPLICATIONS FOR MANAGEMENT Conclusions In this study, eleven additiona l polymorphic microsatellite loci were developed specifically for Florida manatees (Pause et al. 2007). These eleven loci were combined with seven previously published lo ci (Garcia-Rodriguez et al. 2000), screened on over 450 manatee tissue and blood samples, and were determ ined to be sufficient for indi vidual identifica tion. A model DNA database was developed to store the genetic information (Chapter 2). The 18 loci were used to create unique multilocus genotypes for individual Florida manatees and to examine their population genetic st ructure. The datasets were separated by sex and season of collection for a thorough survey of the genetic structure. Very subtle, but statistically significant population structure exis ts in the Florida populat ion during the winter corresponding to the previously designated ma nagement units. The microsatellite data correspond to photo-identifica tion and telemetry observati ons, supporting the winter subpopulations and suggesting vagility duri ng warmer months (Chapter 3). Although genetic reconstruction of a captive pe digree was not completely successful, a few additional highly polymorphic loci may allow for complete reconstruction of captive and wild pedigrees. Additional polymorphi c microsatellite loci from la boratories at the FWRI and the University of Queensland (Broderick et al. 2007) are now available a nd should be screened on the captive pedigree samples to determine if the markers will be sufficient for reconstruction of wild Florida manatee pe digrees (Chapter 4).

PAGE 91

91 Implications for Management Anthropogenic Threats That May Aff ect the Population Genetic Structure Conflict occurs because humans and manatees utilize the same ecosystem. Ecotourism, in the form of swimming with mana tees, has become very popular in certain areas (Northwest Florida), but the effects on the manatee populati on are not yet known. There may be negative impacts on the use of optimal natural areas by co w-calf pairs to avoid hu man aggregations (King and Heinen 2004; Sorice et al. 2006). Alternatively, public inte raction with manatees in their natural habitat increases the awar eness of manatee-related issues and may help in conservation efforts. Ecotourism is relatively benign compared to human overexploitation of Floridas springs, which may become a significant problem in the near future. Reduced spri ng water flow can have many detrimental effects. For example, from 1932 to 1994, spring flow at Blue Spring in Volusia County has decreased by 13%, and it is es timated that flow coul d decrease another 16% by 2010 (Laist and Reynolds III 2005a). Blue Spring provides essentia l winter warm-water habitat for approximately 170 manatees, and unch ecked declining spring flows could reduce the size of thermal plumes to a poi nt where the current subpopulation of manatees can no longer be supported (Laist and Reynolds III 2005a). Without the high flow of fresh water, many of the areas that are now fresh could become more sa line (Williams 2006). This could alter the flora and fauna of the region, changing resource availab ility for manatees. Additionally, the increased water input from the oceans or rivers will decreas e average winter temperatures in the natural spring areas, making them unsu itable habitat for manatees. Development of coastal and spring areas for commercial and residential purposes also poses a threat to manatees (U SFWS 2001). The increased huma n population leads to increased pollution, increased boat traffic, a nd fragmentation of coastal habitat. Watercraft collisions are a

PAGE 92

92 major cause of mortality for Florida ma natees (USFWS 2001; USFWS 2007; Wright et al 1995), and increasing the amount of traffic on the waterways without sufficient protections could be detrimental to the long-term survival of manatees (Langtimm et al. 1998; Marmontel et al. 1997). In addition to the degradation and fragmenta tion of the manatees natural habitat, the landscape of artificial warm-water winter refuge s will change dramatically within the upcoming decades. Warm-water discharges from power pl ants are major winter aggregation sites for manatees in Florida, and have probably been highly significant in their recovery. It has been estimated that at least 60% of manatees use out falls from ten Florida power plants (Laist and Reynolds III 2005a). It is unknown if sufficient manatee habitat exists in Floridas natural warm-water sources. These warm-water sour ces consist of many fi rstand second-order magnitude springs; only four of these springs are currently us ed by manatees due to human modification (Laist and Reynolds III 2005b). Over the next fe w decades, many power plants will be retired and replaced with more environm entally-friendly facilities designed to reduce thermal pollution. This will eliminate nearly half of the warm-water areas currently frequented by manatees (Laist and Reynolds III 2005b). These artificial wa rm-water sources are not only important as manatee winter hab itat, but they also act as temporary habitats between natural warm-water refuges. Maintaining and improving the natural wintering areas is critical to the survival of Florida manatees. As the artificial warm-water sources are removed, it is even more important to be maintaining and restoring the natural spring habitats. Over the next few decades, the distribution of the manatee population in Florid a during the winter will change due to these habitat alterations. The changing landscape of Fl orida will change manatee distribution patterns, possibly affecting the population ge netic structure. Habitat fragme ntation that disrupts natural

PAGE 93

93 migration and connectivity is known to affect the genetic structur e of populations (Dixon et al. 2007). Management strategies should focus on maintaining current na tural spring habitat, restoring modified springs, creating additional thermal basins in southern Florida, and attempting to wean manatees from artificial warm-water sour ces before they are reti red. The alteration of the winter aggregation sites may change the populat ion genetic structure of manatees in Florida. Future Directions for Florida Manatee Genetics Genetics and Photo-identification Implementing a genetic capture-recapture syst em should be the next step for Florida manatee population modeling. The model DNA data base (MIGS) described here may work in concert with the MIPS for the es timation of Florida manatee populat ion parameters. The genetic markers identified in this study were determin ed to be sufficient for genetic individual identification. New markers can be included or substituted to increase the power of the panel. A genetic-identific ation system can be informative in wa ys that supplement and complement the present photographic-identification system. It ma y be possible to obtain genetic samples from locations where photographic sampli ng is difficult. Genetic sampling can be very informative in the identification of carcasses in advanced stages of decomposition, adding to the power of the survival estimates. The MIPS/MIGS combination has the potential to be a very powerful tool for population modeling applications. There are many groups of researchers that use either photographic identification (Araabi et al. 2000; Auger-Mth and Whitehead 2007; Beck and Reid 1995; Graham and Roberts 2007) or genetic-identific ation (Peakall et al. 2006; Richard et al. 1996; Taberlet and Luikart 1999; Woods et al. 1999) methods to estimate population parameters, but few have used both methods in concert (Anderson et al. 2001). The current statistical methods used for MIPS may be modified to account for th e assumptions of the genetics database. The

PAGE 94

94 next step in the development of a MIPS/MIGS database would requir e collaboration among the MIPS researchers, geneticists, population modelers, and database developers to determine the needs and specifications of th e first manatee prototype photo /genetics database. Although genotyping necropsy samples is valuable for popul ation genetics studies, building the database with field samples from living manatees of all ag es is more important for implementation of the MIGS. Intensive field sampling for genetics of MIPS individuals is cr itical for validating the system. Population Genetic Structure of Manatees One of the questions addressed here is whether or not the established management units for manatees actually correspond to th e population genetic structure. It is reasonable to include genetics as a contributing factor in designati ng units for conservation; however some authors have cautioned against basing management units solely on genetics due to the importance of anthropogenic threats (Taylor and Dizon 1999). In the context of genetic s, management units have been defined as, populations with significant divergence of allele frequencies at nuclear or mitochondrial loci, regardless of the phylogenetic distinctiveness of the alleles (Moritz 1994). The currently recognized management units represent populations with su btle, but statistically significant differen ces in genotype frequencies duri ng the winter season, capturing the winter site fidelity, as was the intention fo r the management unit desi gnations. It is likely that the level of migra tion in the warm months is high, re sulting in a mixed population; whereas in the winter months, the site fidelity results in the subtle genetic structure that was observed. In the case of manatees, it is of the utmost importance to maintain the critical winter warm-water habitat, but it is also important to maintain and protect the summer dispersal areas and manatee travel corridors to maintain ge netic connectivity among the regions.

PAGE 95

95 Future directions for manatee population gene tic research should focus on refining our understanding of how Florida manatees interact with Antil lean manatees ( T. m. manatus ) with respect to gene flow and migrati on. The genotypic data produced for this study can be used with genotypic data collected for manatees from other regions if the raw data are analyzed following a standard methodology. Standardi zation among researchers and laborat ories is integral to largescale studies of this kind. Population genetic studies of ma natees within the state of Fl orida should also continue, but should focus on samples collected from free-rangi ng manatees or manatees captured for health assessment. It is important to utilize the vast number of samples collected at necropsy for examination of the population structure, but ex amining a large number of live and necropsy specimens obtained during the same year, or seas on, would provide a great deal of information about how the population structur e is changing over time, as well as through capture-recapture population modeling and survival estimates. Genetic Health of the Florida Manatee Population The Florida manatee population consists of an estimated 3,000 individuals (FWRI 2007). It is thought that there may ha ve been an expansion in the population size over time (O'Shea 1988), and that the manatee population in Florida may be a product of a recent colonization (or re-colonization) event within the past 20,000 years (Garcia-Rodriguez et al. 1998). Manatees are long-lived animals with long generation times. Telemetry and photo-identif ication studies show that their migratory routes can cover great distances in a single season and that individuals can travel many miles in only a few days (Deutsch et al. 2003; Fertl et al. 2005; Reid et al. 1991). The genetic diversity of a sp ecies has been correlated w ith the overall health of the population. Maintenance of gene tic heterogeneity is thought to be beneficial for long-term survival of the population by increasing populatio n fitness and reducing disease susceptibility

PAGE 96

96 (Munson et al. 2005; Pearman and Garner 2005; Quattro and Vrijenhoek 1989; Wayne et al. 1991). Based on the neutral microsatellite ma rkers, the non-coding re gion of mitochondrial DNA analyzed (Garcia-Rodriguez et al. 1998), and the allozyme markers (McClenaghan and O'Shea 1988), it appears that manatees in Florid a have a low to aver age level of genetic variation. Floridas single mitochondrial DNA haplotype can easily be explained by the mutation rate of DNA. Overall, mitochondrial DNA mutates at only about 2% (or two bases per 100 bases) per 1 million years (Avise 2004). The m itochondrial studies of West Indian manatees examined only 410 bp of the typically highly variable, non-coding m itochondrial D-loop. Although 410 bp was a common number of bases for phylogeographic analysis at that time, it was not sufficient for resolving the fine-scale population genetic structure of the recently ( 20,000 years) established Florid a subspecies. Comprehensive genetic analyses of other species have included great er sequence lengths (King et al 2006). Additional gene sequencing could be performed to increase the number of bases in the phylogeograp hic analysis; however, it is likely that sufficient evolu tionary time has not yet passed to allow for the divergence of multiple mitochondrial DNA haplotypes for manatees wi thin the Florida waters. It is known that sirenians have resided in the Florida peninsul a for millions of years (Domning 2005; MacFadden et al. 2004), but the T. m. latirostris subspecies is a more recent derivative of T. manatus (Domning 2000; Garcia-Rodriguez et al. 1998). For these reasons, the low level of genetic differentiation among the Florida management units and coasts is not surprising. If additional DNA sequencing is attempted, Y-chromosome gene s and nuclear genes should be included in the analyses to increase the resolving power and give a thorough phyl ogeographic analysis. Genetic reconstruction of wild pe digrees will give us insight in to the health of the Florida manatee population. A major concern for mana gers of any endangered species should be

PAGE 97

97 estimating the level of inbreeding in the wild. The tools presented herein are useful for delineating family units and for reconstructing so me pedigree relationships. By refining the panel of markers and using analysis tools similar to those presented, reconstruction of the captive pedigree should be relatively straightforward. Once the tools and analyses have been validated on the captive samples, they should be applie d to the wild population. With only a few generations of pedigrees, accurate estimates of inbreeding coefficients can be calculated (Coltman and Slate 2003; Keller et al. 2002; Pemberton 2004). The traditional method for estimating inbreeding is by calcu lating inbreeding coefficients from reconstructed breeding pedigrees. Although statistical methods exist for estimating inbreeding using the level of heterozygosity in a population, several author s caution against using them, suggesting that estimating inbreeding with the trad itional methodology is best (Balloux et al. 2004; Pemberton 2004; Slate et al. 2004). Manatees presen t an interesting paradigm to study pedigrees and inbreeding in the wild. Seve ral decades of research and population monitoring through photoidentification studies provide geneticists with va st sources of information for the production and genetic reconstruction of multi-generational pedigrees. A comprehensive management plan incorporates the best available science in many areas, including demographics, habitat, and genetic co nsiderations (Avise 2004; Lande 1988). Much is known about the demographic and lif e history traits of Florida manatees and the anthropogenic threats in the regions around the state. The Florida manatee popul ation is thought to be stable throughout its range, with the ex ception of the Southwest manage ment unit where estimates are less precise due to incomplete da ta (USFWS 2007). The studies of the Florida manatee genetic diversity suggest that there is a low to average le vel of genetic variation in the Florida subspecies (Garcia-Rodriguez 2000; Garcia-Rodriguez et al. 1998; Garcia-Rodriguez et al. 2000;

PAGE 98

98 McClenaghan and O'Shea 1988; Pause et al. 2007). Genetic monitori ng of this unique species should continue so that any changes in the gene tic diversity of the spec ies over time can be identified. The habitat issues that will be facing this imperile d species in the upcoming decades will be of great importance, and will be in tertwined with their conservation and population genetics.

PAGE 99

99 APPENDIX A LIST OF ABBREVIATIONS Organizations, etc. ABI: Applied Biosystems FWC: Florida Fish and Wild life Conservation Commission FWRI: Florida Fish and Wildlife Research Institute GIS: Genetic Identification Services, Inc. MIGS: Manatee Indivi dual Genetic-identification System MIPS: Manatee Indi vidual Photo-iden tification System UF: University of Florida o UF ICBR: University of Florida Interd isciplinary Center for Biotechnology Research o UF ICBR GAL: University of Florida Interd isciplinary Center for Biotechnology Research Genetic Analysis Laboratory o MMW: Molecular Markers Workshop, co-sponsored by the UF ICBR GAL and UF ICBR Education Core USFWS: United States Fish and Wildlife Service USGS: United States Geological Survey Laboratory Reagents and Equipment BSA: Bovine Serum Albumin bp: Basepairs DMSO: Dimethyl Sulfoxide DNA: Deoxyribonucleic acid dNTP: Deoxynucleotide Triphosphate EDTA: Ethylenediamine Tetraacetic acid FAM: carboxyfluorescein (blue)

PAGE 100

100 HEX: 2, 4, 4, 5, 7, 7-h exachlorofluorescein (green) LB: Luria-Bertani MgCl2: Magnesium Chloride NaCl: Sodium Chloride NaOH: Sodium Hydroxide PCR: Polymerase Chain Reaction SDS: Sodium Dodecyl Sulfate SSC: Saline-Sodium Citrate Tm: Annealing temperature for PCR TM#: Prefix for the laboratory sample identi fication number for manatees at the UF ICBR GAL UV: Ultraviolet X-gal: 5-bromo-4-chloro-3-indo lyl-beta-D-galactopyranoside Population Genetics HE: Expected heterozygosity: Expected pr oportion of heterozygot es in the population based on the principle of Hardy-Weinberg equilibrium. HO: Observed heterozygosity: The actual pr oportion of heterozygot es in the population calculated from the dataset. HWE: Hardy-Weinberg equilibrium: State wh ich describes an idealized population of diploid organisms which reproduce sexually in a random fashion w ith non-overlapping generations. This population of infinite size experiences no mutation, no migration, and no selection. In population genetic data analysis, this principle st ates that allele frequencies will remain the same, but genotype frequencies will change over time. HO and HE values are used to test if allele fr equencies meet HWE expectations. P HWE : P-value for Hardy-Weinberg Equilibrium; typically a 2 test is performed. P ( ID ): Probability of Id entity: probability of choosing two individuals from a population that have identical genotypes at the loci examined. o HW P ( ID ): Standard probability of id entity, assuming Hardy-Weinberg equilibrium (Waits et al. 2001):

PAGE 101

101 iii j j i ip p p ID P2 4) 2 ( ) ( HW o Sib P ( ID ): Sibling probability of identity, accounting for close relatives in the dataset (Waits et al. 2001): ) 25 0 ( ] ) ( 5 0 [ ) 5 0 ( 25 0 ) ( Sib4 2 2 2i i ip p p ID P F-statistics: Introduced by Wri ght (Wright 1951) as possibiliti es for describing the genetic structure of populations in term s of allelic correla tions. The following are equations that are used to describe departur es from Hardy-Weinberg equi librium. Definitions adapted from Avise (2004). o FIS: The correlation between homologous allele s within individuals with reference to the local population. o FIT: The corresponding allelic correlati on with referen ce to the total population. o FST: The proportion of genetic variat ion distributed among subdivided populations. An approach for estimating gene flow and population subdivision. RST: Analogous to FST, but thought to be more appropriate for microsatellite datasets to account for the high mutation rate of microsatellites (Slatkin 1995). Genetic Distance: o DC: Cavalli-Sforza and Edwards ch ord distance (Takezaki and Nei 1996) r j m i ij ij Cjy x r D ) 1 ( 2 ) / 2 ( Where r is the number of loci and xij, and yij are the frequencies of ith allele in the jth locus. o PSA: Proportion of shared allele s, typically calculated as ln ( PSA ) (Dieringer and Schltterer 2003): D f f PSAka j a i a ) min(, The PSA is equivalent to the sum over all loci and all alleles where fa,i, and fa,j are the frequencies of allele a in pop i or j, and D is the number of loci.

PAGE 102

102 MCMC: Monte Carlo Markov Chain: Statistical technique used to estimate the posterior distribution of all possible comb inations of parameter values by only sampling a subset of the likelihood space instead of an exhaustiv e exploration of the whole parameter space (Excoffier and Heckel 2006). MDS: Multidimensional scaling analysis: A powerful set of multivariate statistical techniques for the visualization of patterns with in a dataset. The method utilizes a distance matrix (genetic, geographic, etc.) and pe rforms procedures to place points in multidimensional space that correspond to the actual distances among individuals (Quinn and Keough 2002). N : Number of samples. NA: Number of alleles observed. NE: Effective number of alleles: measure of allelic diversity that takes into account the homozygosity as well as the numbe r of alleles (Yeh and Boyle 1997). Specific to the STRUCTURE software package (Pritchard et al. 2000): o K : Number of genetic population clusters. o LnPr(X| K ): Log probability of the data. o Pr( K |X): Posterior probability of the data; th e greatest posterior probability of the data represents the most likely number of genetic clusters. o Q : Proportion of admixture: Admixture w ithin an individual corresponds to an individual with mixed ancestry from diffe rent populations. Admixture within a population corresponds to many indi viduals having mixed ancestry. SSRs: Simple Sequence Repeats; term for microsatellite. STRs: Short Tandem Repeats; term for microsatellite.

PAGE 103

103 APPENDIX B LOCALITY INFORMATION Table B-1. Sample information. TM# UF ICBR GAL laboratory identification number, FieldID# identification numbers located on sample container, Locality specific location of sample collection, Notes available notes from USGS biologists field notes or FWRI necropsy database, including listed cau se of death, County Florida county, MU management unit, ATL represents At lantic, STJR represents St. Johns River, NW represents Northwes t, SW represents Southwest, Sample Type sample from live animals as tissue (Live) or from carcasses (Necropsy), Sex sex of the animal if known (M, male; F, female) or U if unknown, MIPS# MIPS identification number (if available), Date date of sample collection as MM/DD/YYYY.

PAGE 104

104Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM128 92-001-1 Blue Spring Volusia STJR Live M BS126 01/01/1992 TM129 92-002-1 Blue Spring Volusia STJR Live F BS120 01/02/1992 TM130 92-017-1 Crystal River Calf of CR087 Citrus NW Live F 01/17/1992 TM131 94-063-1 Blue Spring Doyle Volusia STJR Live M BS150 03/04/1994 TM132 94-064-1 Blue Spring Rachael Volusia STJR Live F BS151 03/05/1994 TM133 96-338-1 Blue Spring Calf of Lily (BS004) Volusia STJR Live M 12/03/1996 TM134 96-338-2 Blue Spring Calf of Success (BS054) Volusia STJR Live M 12/03/1996 TM135 96-338-3 Blue Spring Calf of Lola (BS073) Volusia STJR Live U 12/03/1996 TM136 98-026-1 Blue Spring Whiskers Volusia STJR Live M BS181 01/26/1998 TM137 98-026-2 Blue Spring Levy Volusia STJR Live F BS182 01/26/1998 TM141 92-044-1 Blue Spring Bartram, Calf of BS107, twin of Bertram Volusia STJR Live M BS128 02/13/1992 TM142 92-061-1 Blue Spring Bertram, Calf of BS107, twin of Bartram Volusia STJR Live M BS127 03/07/1992 TM143 96-045-1 FL Power and Light Calf of Adair (TBC38 ) Broward ATL Live F 02/14/1996 TM144 93-138-1 Banana River Robbie, Calf of Betty (TBC24) Brevard ATL Live M 05/19/1993 TM145 95-110-1 Banana River C-Cow (TBC09), large female, mother of E-cow Brevard ATL Live F 04/20/1995 TM146 92-126-1 Banana River E-cow (TBC35), adult, calf of CCow Brevard ATL Live F 05/08/1992 TM148 93-036-2 Blue Spring Precious, Calf of BS173 Volusia STJR Live F BS140 02/05/1993 TM149 93-036-3 Blue Spring Volusia STJR Live F BS130 02/05/1993

PAGE 105

105Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM152 98-307-1 Miami Dakota (TMI04), adult, rehab/release Dade ATL Live M 11/03/1998 TM153 98-352-5 Homosassa River Citrus NW Live M 12/18/1998 TM154 98-352-1 Crystal River Citrus NW Live M 12/18/1998 TM155 96-330-1 Crystal River Calf of CR123 Citrus NW Live F 11/26/1996 TM156 96-330-2 Crystal River Citrus NW Live M 11/26/1996 TM157 95-027-1 Crystal River Citrus NW Live F 01/27/1995 TM158 95-031-2 Crystal River Citrus NW Live M 01/31/1995 TM500 MNE9919 St. Johns River Watercraft Duval ATL Necropsy M 06/19/1999 TM501 MNE9921 Cross Florida Barge Canal Undetermined, decomposed Putnam STJR Necropsy M 08/20/1999 TM502 MNE9923 Ortega River Perinatal Duval ATL Necropsy F 09/21/1999 TM503 MNE9924 St. Johns River Undetermined, decomposed Putnam STJR Necropsy M 09/26/1999 TM504 MNE9925 Bulow Creek Natural Volusia ATL Necropsy M 10/13/1999 TM505 MEC0013 Grand Canal Undetermined, decomposed Brevard ATL Necropsy F 03/04/2000 TM506 MEC0014 Indian River Watercraft Brevard ATL Necropsy M 03/04/2000 TM507 MEC0015 St. Johns River Watercraft Lake STJR Necropsy F BS156 03/05/2000 TM508 MEC0016 Indian River Human, other Brevard ATL Necropsy F 03/07/2000 TM509 MEC0017 Indian River Watercraft Brevard ATL Necropsy F 03/11/2000 TM510 MSE0004 Biscayne Creek Watercraft Dade ATL Necropsy F 01/29/2000 TM511 MSE0006 Ft. Everglades Basin Watercraft Broward ATL Necropsy F 02/02/2000 TM512 MSE0009 Intracoastal Waterway Watercraft Palm Beach ATL Necropsy F 02/13/2000 TM513 MSE0010 Lake Worth Undetermined, other Palm Beach ATL Necr opsy M 02/14/2000 TM514 MSE0012 Lake Wyman Perinatal Pa lm Beach ATL Necr opsy M 02/20/2000 TM515 MNW0001 Kings Bay Perinatal Citrus NW Necropsy M 01/17/2000 TM516 MNW0002 Gulf of Mexico Human, ot her Citrus NW Necropsy F CR294 01/18/2000

PAGE 106

106Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM517 MNW0003 Suwannee River Undetermined, decomposed Dixie NW Necropsy M 01/17/2000 TM518 MNW0004 Blackwater River Cold stress Santa Rosa NW Necropsy M 01/22/2000 TM519 MNW0005 Gulf of Mexico Watercraft Levy NW Necropsy M 02/12/2000 TM520 MSW0001 Tamiami Canal Natural Collier SW Necropsy F 01/01/2000 TM521 MSW0003 Imperial River Watercraft Lee SW Necropsy M 01/03/2000 TM522 MSW0006 Ten Mile Canal Cold stress Lee SW Necropsy M 01/20/2000 TM523 MSW0007 Caloosahatchee River Natural Lee SW Necropsy F 01/22/2000 TM524 MSW0008 Imperial River Natural Lee SW Necropsy M 01/23/2000 TM600 99-061-1 Crystal River Citrus NW Live M 03/02/1999 TM601 99-061-2 Crystal River Citrus NW Live M 03/02/1999 TM602 99-061-3 Crystal River Citrus NW Live M 03/02/1999 TM603 98-352-2 Crystal River Citrus NW Live M 12/18/1998 TM604 02-350-1 Crystal River Citrus NW Live M 12/18/1998 TM605 02-350-2 Crystal River Citrus NW Live U 12/16/2002 TM606 02-340-1 Port Of the Islands Juvenile, TNP19 Collier SW Live M 12/06/2002 TM607 02-340-2 Port Of the Islands Juvenile, TNP20 Collier SW Live M 12/06/2002 TM608 02-340-3 Port Of the Islands 266cm, TNP21 Collier SW Live F 12/06/2002 TM609 03-028-4 Warm Mineral Springs 277cm, TSW047 Sarasota SW Live F 01/28/2003 TM610 03-028-3 Warm Mineral Springs 277 cm, TSW045 Sarasota SW Live F 01/28/2003 TM611 02-347-1 Tampa Electric Company 290cm, TTB090 Hillsborough SW Live F 12/13/2002 TM612 02-347-2 Tampa Electric Company 259cm, TTB091 Hillsborough SW Live F 12/13/2002 TM613 02-347-3 Tampa Electric Company 309cm, TTB092 Hillsborough SW Live M 12/13/2002 TM614 03-017-5 Blue Spring State Park Lucretia Volusia STJR Live M 01/17/2003 TM615 03-017-6 Blue Spring State Park Lola Volusia STJR Live M 01/17/2003

PAGE 107

107Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM616 03-017-7 Blue Spring State Park Dana, Twin? Volusia STJR Live F 01/17/2003 TM617 03-017-8 Blue Spring State Park Dianne Volusia STJR Live M 01/17/2003 TM618 03-017-9 Blue Spring State Park Jessica Volusia STJR Live F 01/17/2003 TM619 03-017-10 Blue Spring State Park Dana, Twin? Volusia STJR Live M 01/17/2003 TM620 03-170-1 Blackpoint CMI-03-01, 263cm Dade ATL Live M 06/19/2003 TM621 98-307-2 Miami Same as 98-307-1, Dakota, TMI04 Dade ATL Live M 11/03/1998 TM622 99-145-1 Miami Adult, TMI06 Dade ATL Live M 05/27/1999 TM623 99-011-4 Port Everglades Calf of Betty/TBC24 Broward ATL Live U 01/11/1999 TM624 93-020-1 Crystal River Calf of CR041 Citrus NW Live F 01/20/1993 TM625 93-020-2 Crystal River Calf of CR104, Twin or adoption? Citrus NW Live M 01/20/1993 TM626 93-029-1 Crystal River Citrus NW Live F 01/29/1993 TM627 93-032-1 Tampa Bay TTB21 Hillsborough SW Live F 02/01/1993 TM628 93-032-2 Tampa Bay CTB93-01 H illsborough SW Live M 02/01/1993 TM629 93-033-1 Tampa Bay CTB93-02 H illsborough SW Live M 02/02/1993 TM630 93-033-2 Tampa Bay CTB93-03 Hillsborough SW Live F 02/02/1993 TM631 02-008-1 Warm Mineral Springs Calf of TSW-036, CSW039 Sarasota SW Live M 01/08/2002 TM632 03-013-2 Warm Mineral Springs 259cm, TSW044 Sarasota SW Live M 01/13/2003 TM633 02-341-1 Port Of the Islands 285cm, TNP22 Collier SW Live M 12/07/2002 TM634 02-341-2 Port Of the Islands 335cm, TNP23 Collier SW Live F 12/07/2002 TM635 99-110-1 Banana river Juvenile, CBC99-01 Brevard ATL Live M 04/22/1999 TM636 00-021-11 Blue Spring Calf of Julie Volusia STJR Live M 01/21/2000 TM637 00-021-13 Blue Spring Calf of Cora Volusia STJR Live F 01/21/2000 TM638 00-021-9 Blue Spring Calf of Donna Volusia STJR Live M 01/21/2000

PAGE 108

108Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM639 00-021-12 Blue Spring Calf of Phyllis Volusia STJR Live F 01/21/2000 TM640 00-032-7 Blue Spring Volusia STJR Live 02/02/2000 TM641 00-021-6 Blue Spring Calf of Laurie Volusia STJR Live M 01/21/2000 TM642 00-021-8 Blue Spring Calf of Debora Volusia STJR Live F 01/21/2000 TM643 00-021-14 Blue Spring Calf of Lucy Volusia STJR Live F 01/21/2000 TM644 01-087-3 Blue Spring Lacy, Calf of Lily Volusia STJR Live M 03/29/2001 TM645 01-025-1 Blue Spring Calf of Lola Volusia STJR Live U 01/21/2001 TM646 01-025-2 Blue Spring Doom, Calf of Destiny Volusia STJR Live M 01/25/2001 TM647 99-054-2 Riviera Beach Power Plant Adult Palm Beach ATL Live F 02/23/1999 TM648 MSE9906 Biscayne Bay Watercraft Dade ATL Necropsy M MI066 02/14/1999 TM649 MSE9908 Bessey Creek Perinatal Martin ATL Necropsy M 02/19/1999 TM650 MSE9833 Hendry Creek Canal Gate/lock Okeechobee ATL Necropsy M 12/02/1998 TM651 MSE9943 Lake Okeechobee Gate/lock Okeechobee ATL Necropsy M 10/19/1999 TM652 MSE9948 Tamiami Canal Gate/lock Dade ATL Necropsy F 12/29/1999 TM653 MSE9826 Loxahatchee River Perinatal Martin ATL Necropsy F 07/28/1998 TM654 MSE9828 Biscayne Canal Gate/lock Dade ATL Necropsy M 09/21/1998 TM655 MSE9829 St. Lucie Canal Gate/lock Martin ATL Necropsy M 09/26/1998 TM656 MSE9831 St. Lucie Canal Gate/lock Martin ATL Necropsy M 10/23/1998 TM657 MSE9832 South Fork, St. Lucie River Natural Martin ATL Necropsy M 10/31/1998 TM658 MSE9916 Florida Bay Perinatal Monroe ATL Necropsy M 03/25/1999 TM659 MSE9918 South Fork New River Undetermined, decomposed Broward ATL Necropsy M 04/15/1999 TM660 MSE9919 Manatee Pocket Watercraft Martin ATL Necropsy F 04/15/1999 TM661 MSE9920 Port EvergladesIntracoastal Waterway Watercraft Broward ATL Necropsy M MI048 04/19/1999 TM662 MSE9921 C-10 Canal Perinatal Broward ATL Necropsy M 05/14/1999 TM663 MSE9922 Tamiami Canal Human, other Dade ATL Necropsy F 06/03/1999

PAGE 109

109Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM664 MSE9923 Tamiami Canal Human, other Dade ATL Necropsy M 06/03/1999 TM665 MSE9924 Miami Canal Human, other Dade ATL Necropsy M 06/04/1999 TM666 MSE9925 North New River Canal Undetermined, decomposed Palm Beach ATL Necropsy U 06/10/1999 TM667 MSE9901 Tarpon Basin Channel Watercraft Monroe ATL Necropsy M 01/02/1999 TM668 MSE9902 Port EvergladesIntracoastal Waterway Natural Broward ATL Necropsy F 01/29/1999 TM669 MSE9904 Biscayne Bay, Intracoastal Waterway Natural Dade ATL Necropsy M 02/07/1999 TM670 MSE9905 Lake Worth Creek Watercraft Palm Beach ATL Necropsy F 02/07/1999 TM671 MSE9907 Snake Creek Canal (C-9) Natural Dade ATL Necropsy M 02/15/1999 TM672 MSE9909 Indian River Natural Martin ATL Necropsy F 03/05/1999 TM673 MSE9911 Largo Sound, Atlantic Ocean Watercraft Monroe ATL Necropsy M 03/12/1999 TM674 MSE9913 South Fork New River Perinatal Broward ATL Necropsy M 03/16/1999 TM675 MSE9915 Indian River Natural Martin ATL Necropsy M 03/24/1999 TM678 02-324-1 Crystal River Citrus NW Live M 11/20/2002 TM679 02-324-2 Crystal River Citrus NW Live M 11/20/2002 TM680 02-324-3 Homosassa River Citrus NW Live F 11/20/2002 TM681 02-324-4 Homosassa River Citrus NW Live F 11/20/2002 TM682 02-330-1 Crystal River Citrus NW Live M 11/26/2002 TM683 02-330-3 Crystal River Citrus NW Live M 11/26/2002 TM684 02-330-2 Crystal River Citrus NW Live M 11/26/2002 TM685 02-336-2 Crystal River Citrus NW Live M 12/02/2002 TM686 02-336-3 Crystal River Citrus NW Live M 12/02/2002 TM687 02-336-4 Crystal River Citrus NW Live M 12/02/2002 TM688 02-336-1 Crystal River Citrus NW Live M 12/02/2002 TM689 02-336-5 Crystal River Citrus NW Live M 12/02/2002

PAGE 110

110Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM690 02-336-6 Crystal River Citrus NW Live F 12/02/2002 TM691 02-337-1 Crystal River Citrus NW Live F 12/03/2002 TM692 02-337-2 Crystal River Citrus NW Live M 12/03/2002 TM693 02-337-3 Crystal River Citrus NW Live F 12/03/2002 TM694 02-337-4 Crystal River Citrus NW Live F 12/03/2002 TM695 02-337-5 Crystal River Citrus NW Live M 12/03/2002 TM696 02-337-6 Crystal River Citrus NW Live F 12/03/2002 TM697 02-337-7 Homosassa River Citrus NW Live F 12/03/2002 TM698 02-346-1 Crystal River Citrus NW Live F 12/12/2002 TM699 02-346-2 Crystal River Citrus NW Live M 12/12/2002 TM700 02-329-1 Crystal River Citrus NW Live M 11/25/2002 TM701 02-350-3 Crystal River Citrus NW Live M 12/16/2002 TM702 02-350-4 Crystal River Citrus NW Live F 12/16/2002 TM703 99-076-2 Blue Spring State Park Michael, Calf of Michelle Volusia STJR Live M 03/17/1999 TM704 99-076-3 Blue Spring State Park Lovis, Calf of Lily/BS004 Volusia STJR Live M 03/17/1999 TM705 99-076-4 Blue Spring State Park Calf of Lillith/BS098 Volusia STJR Live U 03/17/1999 TM706 99-076-5 Blue Spring State Park Eddie, Juvenile Volusia STJR Live M 03/17/1999 TM707 99-089-1 Blue Spring State Park Calf of Georgia/TBS02 Volusia STJR Live M 03/31/1999 TM708 00-021-10 Blue Spring State Park Calf of Hortenese Volusia STJR Live M 01/21/2000 TM709 02-009-1 Blue Spring State Park Volusia STJR Live F 01/09/2002 TM710 02-009-2 Blue Spring State Park Calf of Lucy Volusia STJR Live F 01/09/2002 TM711 02-009-3 Blue Spring State Park Calf of Ester Volusia STJR Live M 01/09/2002 TM712 02-009-4 Blue Spring State Park Calf of Phyllis Volusia STJR Live F 01/09/2002

PAGE 111

111Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM713 02-017-3 Blue Spring State Park Calf of Della Volusia STJR Live F 01/17/2002 TM714 02-017-4 Blue Spring State Park Calf of Amanda Volusia STJR Live F 01/17/2002 TM715 03-030-1 Blue Spring State Park Calf of Lily Volusia STJR Live M 01/30/2003 TM716 03-030-2 Blue Spring State Park Calf of Julie Volusia STJR Live M 01/30/2003 TM717 03-030-3 Blue Spring State Park Calf of Calista Volusia STJR Live F 01/30/2003 TM718 03-030-4 Blue Spring State Park Calf of Georgia/TBS02 Volusia STJR Live F 01/30/2003 TM719 03-017-11 Blue Spring State Park Calf of Judith Volusia STJR Live M 01/17/2003 TM720 03-017-12 Blue Spring State Park Calf of Laurie Volusia STJR Live M 01/17/2003 TM721 04-022-1 Blue Spring State Park Volusia STJR Live 01/22/2004 TM722 04-022-2 Blue Spring State Park Volusia STJR Live 01/22/2004 TM723 04-022-3 Blue Spring State Park Volusia STJR Live 01/22/2004 TM724 04-022-4 Blue Spring State Park Volusia STJR Live 01/22/2004 TM725 04-022-5 Blue Spring State Park Volusia STJR Live 01/22/2004 TM726 04-022-6 Blue Spring State Park Volusia STJR Live 01/22/2004 TM727 04-072-1 Blue Spring State Park Volusia STJR Live 01/22/2004 TM728 MSW9945 Naples Bay Watercraft Collier SW Necropsy F 06/11/1999 TM729 MSW9946 Matlacha Pass Watercraft Lee SW Necropsy M 06/16/1999 TM730 MSW9947 Wiggins Pass Watercraft Collier SW Necropsy F 06/22/1999 TM731 MSW9948 Estero Bay Perinatal Lee SW Necropsy M 06/23/1999

PAGE 112

112Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM732 MSW9949 Estero Bay Perinatal Lee SW Necropsy F 07/03/1999 TM733 MSW9950 Florida Bay Watercraft Monroe SW Necropsy F 07/03/1999 TM734 MSW9951 Caloosahatchee River Gate/lock Lee SW Necropsy F 07/04/1999 TM735 MSW9952 Whidden Creek Watercraft Charlotte SW Necropsy M 07/05/1999 TM736 MSW9955 Lake Hicpochee Undetermined, decomposed Glades SW Necropsy M 07/22/1999 TM737 MSW9956 Placida Harbor Perinatal Charlotte SW Necropsy F 07/23/1999 TM738 MSW9957 Caloosahatchee River Perinatal Lee SW Necropsy U 07/27/1999 TM739 MSW9958 Alligator Creek Perinata l Charlotte SW Necropsy F 08/03/1999 TM740 MSW9959 San Carlos Bay Watercraft Lee SW Necropsy F 08/15/1999 TM741 MSW9960 Phillippi Creek Watercraft Sarasota SW Necropsy F 08/28/1999 TM742 MSW9961 Peace River Perinatal Ch arlotte SW Necrops y M 09/02/1999 TM743 MSW9962 Pine Island Sound Undetermined, decomposed Lee SW Necropsy M 09/04/1999 TM744 MSW9963 Lemon Bay Watercraft Sarasota SW Necropsy M 09/11/1999 TM745 MSW9964 Okeechobee Waterway Undetermined, decomposed Glades SW Necropsy F 09/25/1999 TM746 MSW9965 Caloosahatchee Canal Watercraft Glades SW Necropsy M 10/02/1999 TM747 MSW9966 Rookery Bay Watercraft Collier SW Necropsy M 10/03/1999 TM748 MSW9967 Shackett Creek Undetermined, decomposed Sarasota SW Necropsy F 10/03/1999 TM749 MSW9968 Lemon Bay Natural Charlotte SW Necropsy F 10/11/1999 TM750 MSW9969 Lemon Bay Watercraft Charlotte SW Necropsy F 10/11/1999 TM751 MSW9970 Placida Harbor Undetermined, decomposed Charlotte SW Necropsy M 10/15/1999 TM752 MSW9971 Lemon Bay Watercraft Charlotte SW Necropsy F 10/16/1999 TM753 MSW9920 Matlacha Pass Natural Lee SW Necropsy F 02/17/1999 TM754 MSW9921 Myakka River Undetermined, decomposed Sarasota SW Necropsy F 02/23/1999

PAGE 113

113Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM755 MSW9922 Caloosahatchee River Undetermined, decomposed Lee SW Necropsy M 02/23/1999 TM756 MSW9923 Forked Creek Undetermined, decomposed Sarasota SW Necropsy F 02/27/1999 TM757 MSW9924 Caloosahatchee River Undetermined, decomposed Lee SW Necropsy M 02/28/1999 TM758 MSW9926 Charlotte Harbor Waterc raft Lee SW Necropsy F 03/16/1999 TM759 MSW9927 Lake Okeechobee Watercraft Glades SW Necropsy M 03/23/1999 TM760 MSW9928 Buttonwood Harbor Perinatal Sarasota SW Necropsy F 03/24/1999 TM761 MSW9929 Pine Island Sound Perinatal Lee SW Necropsy M 04/14/1999 TM762 MSW9930 Lake Okeechobee Undetermined, decomposed Glades SW Necropsy M 04/18/1999 TM763 MSW9931 Blackburn Bay Natural Sarasota SW Necropsy M 04/18/1999 TM764 MSW9932 Lemon Bay Natural Sarasota SW Necropsy F 04/19/1999 TM765 MSW9933 Rookery Bay Watercraft Collier SW Necropsy F 04/21/1999 TM766 MSW9934 Kesson Bayou Perinatal Lee SW Necropsy M 04/27/1999 TM767 MSW9935 Peace River Watercraft Charlotte SW Necr opsy M 05/07/1999 TM768 MSW9936 Gulf of Mexico Undetermined, decomposed Collier SW Necropsy F 05/11/1999 TM769 MSW9937 Gulf of Mexico Watercraft Sarasota SW Necropsy M 05/11/1999 TM770 MSW9938 Big Slough Perinatal Sarasota SW Necropsy F 05/26/1999 TM771 MSW9939 Caloosahatchee River Watercraft Lee SW Necropsy M 05/28/1999 TM772 MSW9940 Naples Bay Watercraft Collier SW Necropsy F 05/31/1999 TM773 MSW9941 Pine Island Sound Watercraft Lee SW Necropsy M 06/01/1999 TM774 MSW9942 Estero Bay Watercraft Lee SW Necropsy F 06/02/1999 TM775 MSW9943 Naples Bay Perinatal Collier SW Necropsy M 06/07/1999 TM776 MSW9944 Gordon River Undetermined, decomposed Collier SW Necropsy F 06/09/1999 TM777 MNW9902 Tampa Bay Watercraft H illsborough SW Necropsy M 01/06/1999 TM778 MNW9903 St. Joseph Bay Natural Gulf NW Necropsy F 01/16/1999

PAGE 114

114Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM779 MNW9904 Withlacoochee River Cold stress Levy NW Necropsy F 01/20/1999 TM780 MNW9905 Little Manatee River Watercraft Hillsborough SW Necropsy M 02/08/1999 TM781 MNW9906 Tampa Bay Watercraft Pinellas SW Necropsy M 02/15/1999 TM782 MNW9909 Old Tampa Bay Cold stress Pinellas SW Necropsy F 03/06/1999 TM783 MNW9910 Manatee River Perinatal Manatee SW Necropsy F 03/07/1999 TM784 MNW9911 Cross Florida Barge Canal Gate/lock Levy NW Necropsy M 03/10/1999 TM785 MNW9912 Placido Bayou Undetermined, decomposed Pinellas SW Necropsy F 03/18/1999 TM786 MNW9913 Anna Maria Sound Perinatal Manatee SW Necropsy M 03/19/1999 TM787 MNW9914 Hillsborough Bay Watercraft Hillsborough SW Necropsy M 03/29/1999 TM788 MNW9915 Palma Sola Bay Perinatal Manatee SW Necropsy F 04/27/1999 TM789 MNW9916 Bowles Creek Perinatal Manatee SW Necropsy M 05/01/1999 TM791 MNW9918 Aucilla River Not recovered Taylor NW Necropsy U 05/16/1999 TM792 MNW9919 Sarasota Bay Watercraft Manatee SW Necropsy M 05/19/1999 TM793 MNW9921 Pithlachascotee River Watercraft Pasco SW Necropsy F 06/07/1999 TM794 MNW9922 Dixie Bay Perinatal Citrus NW Necropsy F 07/01/1999 TM795 MNW9923 Suwannee River Perinatal Gilchrist NW Necropsy F 07/01/1999 TM796 MNW9924 Crystal River Perinatal Citrus NW Necropsy M 07/06/1999 TM797 MNW9925 Braden River Natural Manatee SW Necropsy F 07/08/1999 TM798 MNW9927 Terra Ceia Bay Manatee SW Necropsy F 07/25/1999 TM799 MNW9928 Little Manatee River Perinatal Hillsborough SW Necropsy M 08/17/1999 TM800 MNW9929 Tampa Bay Watercraft H illsborough SW Necropsy F 08/23/1999 TM801 MNW9930 Tampa Bay Undetermined, decomposed Pinellas SW Necropsy F 08/29/1999 TM802 MNW9931 Palma Sola Bay Watercraft Manatee SW Necropsy M 09/06/1999 TM803 05-039-1 Blue Spring Alice, Calf of Amy Volusia STJR Live F 02/08/2005

PAGE 115

115Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM804 05-039-2 Blue Spring State Park Calf of Phalcon Volusia STJR Live F 02/08/2005 TM805 05-039-3 Blue Spring State Park Calf of Laurie Volusia STJR Live M 02/08/2005 TM806 05-049-1 Salt Springs Calf of Clover/BS195 Marion STJR Live M 02/18/2005 TM858 MEC9852 St. Johns River Watercraft Volusia STJR Necropsy F 11/03/1998 TM859 MEC9849 St. Johns River Watercraft Volusia STJR Necropsy F JX032 09/10/1998 TM860 SWFTM9821B St. Johns River Watercraft Volusia STJR Necropsy M 06/07/1998 TM861 00-362-2 Blue Spring Lacy, Calf of Lily/BS004 Volusia STJR Live F 12/27/2000 TM862 02-050-1 Blue Spring State Park Calf of Michelle Volusia STJR Live F 02/19/2002 TM863 01-311-2 Crystal River Citrus NW Live F 11/07/2001 TM864 01-311-3 Crystal River Calf of Lump Citrus NW Live F 11/07/2001 TM865 01-311-4 Crystal River Citrus NW Live M 11/07/2001 TM866 01-311-5 Crystal River Calf of CR436 Citrus NW Live M 11/07/2001 TM867 01-311-6 Homosassa River Citrus NW Live F 11/07/2001 TM868 01-311-7 Homosassa River Citrus NW Live F 11/07/2001 TM869 01-317-1 Crystal River Citrus NW Live M 11/13/2001 TM870 01-317-2 Crystal River Citrus NW Live F 11/13/2001 TM871 01-317-3 Crystal River Citrus NW Live F 11/13/2001 TM872 01-317-4 Crystal River Citrus NW Live M 11/13/2001 TM873 01-317-5 Crystal River Citrus NW Live M 11/13/2001 TM874 01-319-1 Homosassa River Citrus NW Live F 11/15/2001 TM875 02-004-1 Homosassa River Citrus NW Live M 01/04/2002 TM876 02-004-2 Crystal River Citrus NW Live F 01/04/2002 TM877 02-004-3 Crystal River Citrus NW Live F 01/04/2002 TM878 02-007-1 Crystal River Citrus NW Live M 01/07/2002 TM879 02-007-2 Crystal River Citrus NW Live F 01/07/2002 TM880 02-007-3 Crystal River Citrus NW Live M 01/07/2002 TM881 02-007-4 Homosassa River Citrus NW Live M 01/07/2002 TM882 02-007-5 Homosassa River Citrus NW Live F 01/07/2002

PAGE 116

116Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM883 02-007-6 Homosassa River Citrus NW Live M 01/07/2002 TM884 02-007-7 Homosassa River Calf of CR032 Citrus NW Live F 01/07/2002 TM885 02-015-1 Crystal River Calf of CR186 Citrus NW Live M 01/15/2002 TM886 02-016-1 Crystal River Citrus NW Live M 01/16/2002 TM887 02-016-2 Crystal River Citrus NW Live F 01/16/2002 TM888 02-017-1 Crystal River Citrus NW Live M 01/17/2002 TM889 02-017-2 Crystal River Citrus NW Live F 01/17/2002 TM890 02-039-1 Crystal River Citrus NW Live M 02/08/2002 TM891 02-036-1 Crystal River Citrus NW Live M 02/05/2002 TM892 02-0136-2 Crystal River Citrus NW Live TM893 02-158-1 Crystal River, Hunter's Cove Citrus NW Live F 06/07/2002 TM894 02-232-1 Hall's River Calf of CR254 Citrus NW Live M 08/20/2002 TM895 02-303-1 Crystal River Citrus NW Live F 10/30/2002 TM896 02-323-1 Crystal River Citrus NW Live F 11/19/2002 TM897 02-323-2 Crystal River Citrus NW Live F 11/19/2002 TM898 02-323-3 Homosassa River Citrus NW Live M 11/19/2002 TM899 MNE9707 Robinson's Creek Undetermined, decomposed St. Johns ATL Necropsy M 06/08/1997 TM900 MNE9208 Welaka Springs, St. Johns River Perinatal Putnam STJR Necropsy F 04/04/1992 TM901 MNE9209 Welaka Springs, St. Johns River Perinatal Putnam STJR Necropsy M 04/07/1992 TM902 MSW9836 Lake Okeechobee Watercraft Glades SW Necropsy F 07/13/1998 TM903 MSW9837 Mantanzas Pass Watercraft Lee SW Necropsy M 07/14/1998 TM904 MSW9841 Peace River Watercraft Charlotte SW Necr opsy F 07/23/1998 TM905 MSW9843 Peace River Undetermined, decomposed Charlotte SW Necropsy F 08/06/1998 TM906 MSW9844 Caloosahatchee River Watercraft Lee SW Necropsy F 08/27/1998 TM907 MSW9845 Myakka River Natural Sarasota SW Necropsy F 09/16/1998 TM908 MSW9847 Caloosahatchee Canal Gate/lock Glades SW Necropsy M 09/22/1998

PAGE 117

117Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM909 MSW9848 Caloosahatchee Canal Gate/lock Glades SW Necropsy F 09/22/1998 TM910 MSW9849 Caloosahatchee Canal Perinatal Glades SW Necropsy M 09/24/1998 TM911 MSW9851 Pine Island Sound Human, other Lee SW Necropsy F 10/01/1998 TM912 MSW9852 Pine Island Sound Human, other Lee SW Necropsy M 10/01/1998 TM913 MSW9857 Matlacha Pass Natural Lee SW Necropsy M 10/30/1998 TM914 MSW9859 Charlotte Harbor Undetermined, decomposed Lee SW Necropsy M 12/05/1998 TM915 MSW9860 Sarasota Bay Perinatal Sarasota SW Necropsy F 12/06/1998 TM916 MSW9861 Orange River Undetermined, decomposed Lee SW Necropsy M 12/18/1998 TM917 MSW9862 Pine Island Sound Watercraft Lee SW Necropsy M 12/19/1998 TM918 MSW9863 Caloosahatchee River Undetermined, decomposed Lee SW Necropsy M 12/22/1998 TM919 MSW9901 Orange River Undetermined, decomposed Lee SW Necropsy F 01/03/1999 TM920 MSW9902 Pelican Bay Watercraft Lee SW Necropsy F 01/03/1999 TM921 MSW9904 Caloosahatchee Canal Watercraft Glades SW Necropsy F 01/13/1999 TM922 MSW9905 Caloosahatchee River Undetermined, decomposed Lee SW Necropsy F 01/14/1999 TM923 MSW9906 Caloosahatchee River Natural Lee SW Necropsy F 01/13/1999 TM924 MSW9909 Sarasota Bay Natural Sarasota SW Necropsy M 01/17/1999 TM925 MSW9910 Matlacha Pass Human, other Lee SW Necropsy M 01/21/1999 TM926 MSW9911 Charlotte Harbor Natu ral Lee SW Necropsy M 01/26/1999 TM927 MSW9912 Charlotte Harbor Natu ral Lee SW Necropsy M 01/27/1999 TM928 MSW9913 Charlotte Harbor Undetermined, decomposed Charlotte SW Necropsy M 01/29/1999 TM929 MSW9914 Gulf of Mexico Undetermined, decomposed Charlotte SW Necropsy M 01/30/1999 TM930 MSW9915 Redfish Pass Natural Lee SW Necropsy M 02/01/1999

PAGE 118

118Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM931 MSW9917 Matlacha Pass Undetermined, decomposed Lee SW Necropsy F 02/15/1999 TM932 MSW9918 Matlacha Pass Natural Lee SW Necropsy M 02/16/1999 TM933 MSE9930 Port Everglades Turning Basin Watercraft Broward ATL Necropsy F 07/16/1999 TM934 MSE9931 Port Everglades Turning Basin Watercraft Broward ATL Necropsy M 07/16/1999 TM935 MSE9938 Middle River Undetermined, decomposed Broward ATL Necropsy M 09/19/1999 TM936 MSE9940 St. Lucie Canal Gate/lock Martin ATL Necropsy M 10/05/1999 TM937 MSE0005 New River Canal Undetermined, decomposed Broward ATL Necropsy F 02/02/2000 TM938 MSE0007 Little River Gate/lock Dade ATL Necropsy F 02/03/2000 TM939 MSE0008 Tamiami Canal SFWMD C-4 Canal; Human/other Dade ATL Necropsy F 02/09/2000 TM940 MSE0011 Lake Worth Cold stress Palm Beach ATL Necropsy F 02/19/2000 TM941 MSE0014 Hillsboro River Watercra ft Broward ATL Necropsy F 02/26/2000 TM942 MSE0015 North Fork, St. Lucie River Watercraft St. Lucie ATL Necropsy M 03/13/2000 TM943 MSE0016 Biscayne Bay Watercraft Dade ATL Necropsy F 03/14/2000 TM944 MSE0003 Dania Cut-Off Canal Perinatal Broward ATL Necropsy F 01/13/2000 TM945 MNE9206 Mosquito Lagoon Perinatal Volusia ATL Necropsy M 03/27/1992 TM946 MNE9602 St. Johns River Undetermined, decomposed Duval ATL Necropsy F 01/15/1996 TM947 MNE9618 Ortega River Undetermined, decomposed Duval ATL Necropsy M 11/19/1996 TM948 MNE9619 St. Johns River Watercraft Duval ATL Necropsy M 11/24/1996 TM949 MNE9202 Julington Creek Undetermined, decomposed Duval ATL Necropsy M 01/13/1992 TM950 MNE9901 Atlantic Ocean Cold stress St. Johns ATL Necropsy F 01/02/1999 TM951 MNE9902 St. Johns River Perinatal Clay ATL Necropsy 01/08/1999 TM953 MNE9704 Doctors Lake Perinatal Clay ATL Necropsy M 04/20/1997

PAGE 119

119Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM954 MNE9706 Palm Coast Canals Perinatal Flagler ATL Necropsy M 06/06/1997 TM955 MNE9708 Pottsburg Creek Perinatal Duval ATL Necropsy M 06/15/1997 TM956 MNE9713 St. Johns River Watercraft Duval ATL Necropsy M 08/01/1997 TM957 MNE9714 St. Johns River Undetermined, decomposed Duval ATL Necropsy F 08/06/1997 TM958 MNE9715 Flagler BeachIntracoastal Waterway Perinatal Flagler ATL Necropsy F 08/14/1997 TM959 MNE9716 St. Johns River Watercraft Putnam STJR Necropsy M BS119 09/23/1997 TM960 MNE9717 Pablo Creek Watercraft Duval ATL Necropsy M 10/03/1997 TM961 MNE9719 St. Johns River Cold stress Duval ATL Necropsy M 12/22/1997 TM962 MNE9720 Doctors Lake Cold stress Clay ATL Necropsy M 12/26/1997 TM963 MEC9662 Port Canaveral Inlet Gate/lock Brevard ATL Necropsy F 10/16/1996 TM964 MEC9664 Indian River Watercraft Indian River ATL Necropsy F 10/28/1996 TM965 MEC9665 Mullet Creek, Indian River Undetermined, decomposed Brevard ATL Necropsy M 11/01/1996 TM966 MEC9670 Indian River Undetermined, decomposed Indian River ATL Necropsy M 12/10/1996 TM967 MEC9671 Indian River Natural Brevard ATL Necropsy M 12/22/1996 TM968 MEC9701 Indian River Natural Brevard ATL Necropsy F 01/06/1997 TM969 MEC9702 Indian River Undetermined, decomposed Brevard ATL Necropsy M 01/13/1997 TM970 MEC9703 Banana River Watercraft Brevard ATL Necropsy F 01/14/1997 TM971 MEC9704 Indian River, Banana Creek Undetermined, decomposed Brevard ATL Necropsy F MI092 01/15/1997 TM972 MEC9709 Banana River Watercraft Indian River ATL Necropsy M 02/04/1997 TM973 MEC9710 Indian River Natural Indian River ATL Necropsy M 02/05/1997 TM974 MEC9712 Indian River Cold stress Brevard ATL Necropsy F 02/21/1997 TM975 MEC9713 Indian River Watercraft Brevard ATL Necropsy F 03/10/1997 TM976 MEC9715 Banana River Natural Brevard ATL Necropsy F 03/16/1997 TM977 MEC9718 Banana River Watercraft Brevard ATL Necropsy F 04/01/1997

PAGE 120

120Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM978 MEC9719 Banana River Undetermined, decomposed Brevard ATL Necropsy F 04/11/1997 TM979 MEC9720 Atlantic Ocean Watercraft Brevard ATL Necropsy M 04/15/1997 TM980 MEC9721 Indian River Perinatal Brevard ATL Necropsy F 04/16/1997 TM981 MEC9722 Banana River Watercraft Brevard ATL Necropsy M 04/17/1997 TM982 MEC9723 Port Canaveral Barge Canal Watercraft Brevard ATL Necropsy F 04/20/1997 TM983 MEC9728 Indian River Undetermined, decomposed Brevard ATL Necropsy F 05/27/1997 TM984 MEC9729 Indian River Human, othe r Brevard ATL Necropsy M BS116 05/30/1997 TM985 MEC9730 Banana River Natural Brevard ATL Necropsy F 05/31/1997 TM986 MEC9732 Banana River Perinatal Brevard ATL Necropsy M 06/05/1997 TM987 MEC9734 Banana River Natural Brevard ATL Necropsy M 06/15/1997 TM988 MEC9737 Port Canaveral Barge Canal Gate/lock Brevard ATL Necropsy F 06/28/1997 TM989 MEC9738 Banana River, Barge Canal Gate/lock Brevard ATL Necropsy F 06/30/1997 TM990 MEC9742 Tomoka River Not recovered Volusia ATL Necropsy U 07/22/1997 TM991 MEC9746 Banana River Perinatal Brevard ATL Necropsy F 08/01/1997 TM992 MEC9749 Indian River Watercraft Brevard ATL Necropsy M 08/09/1997 TM993 MEC9751 Banana River Perinatal Brevard ATL Necropsy M 08/16/1997 TM994 MNE9802 St. Johns River Cold stress Duval ATL Necropsy F 01/08/1998 TM995 MNE9818 St. Johns River Natural Clay ATL Necropsy M 04/27/1998 TM996 MNE9820 Pablo CreekIntracoastal Waterway Watercraft Duval ATL Necropsy M 05/24/1998 TM997 MNE9821 Trout River Perinatal Duval ATL Necropsy U 05/25/1998 TM998 MNE9822 Matanzas InletIntracoastal Waterway Undetermined, decomposed St. Johns ATL Necropsy F 05/26/1998 TM999 MNE9823 Intracoastal Waterway Watercraft Flagler ATL Necropsy F 05/26/1999

PAGE 121

121Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM1000 MNE9824 Intracoastal Waterway Natural Flagler ATL Necropsy M 06/11/1998 TM1001 MNE9827 Halifax RiverIntracoastal Waterway Watercraft Volusia ATL Necropsy F 07/17/1998 TM1002 MNE9828 Halifax RiverIntracoastal Waterway Watercraft Volusia ATL Necropsy M 07/19/1998 TM1003 MNE9829 North Fork Black Creek Perinatal Clay ATL Necropsy M 07/24/1998 TM1004 MNE9830 St. Johns River Watercraft Clay ATL Necropsy M 07/26/1998 TM1005 MNE9831 Intracoastal Waterway Perinatal Flagler ATL Necropsy M 08/09/1998 TM1006 MNE9833 St. Johns River Watercraft Clay ATL Necropsy F 08/29/1998 TM1007 MNE9835 Ortega River Watercraft Duval ATL Necropsy F 09/27/1998 TM1008 MNE9902 St Johns River Perinatal Clay ATL Necropsy F 01/08/1999 TM1009 MNE9903 Doctors Lake Cold stress Clay ATL Necropsy M 01/20/1999 TM1010 MNE9907 Cross Florida Barge Canal Gate/lock Putnam STJR Necropsy F BC370 02/22/1999 TM1011 MNE9910 Palm ValleyIntracoastal Waterway Natural St. Johns ATL Necropsy M MI088 04/11/1999 TM1012 MNE9914 St. Johns River Natural Duval ATL Necropsy M 04/28/1999 TM1013 MNE9917 San Sebastian River Undetermined, decomposed St. Johns ATL Necropsy M 06/02/1999 TM1014 MSE9808 Alantic Ocean Perinatal Palm Beach ATL Necr opsy M 01/26/1998 TM1015 MSE9816 Intracoastal Waterway Watercraft Palm Beach A TL Necropsy M 05/04/1998 TM1016 MSE9818 Dania Cut-Off Canal Perinatal Broward ATL Necropsy M 05/24/1998 TM1017 MSE9821 Snake Creek Canal Dade ATL Necropsy M 06/21/1998 TM1018 MSE9822 C-111 Canal SFWMD Dade ATL Necropsy F 07/03/1998

PAGE 122

122Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM1019 MSE9823 Dania Cut-Off Canal Perinatal Broward ATL Necropsy F 07/05/1998 TM1020 MSE9926 Boynton Inlet Human, othe r Palm Beach ATL Necropsy F 06/23/1999 TM1021 MSE9928 Lake Worth Human, othe r Palm Beach ATL N ecropsy M 07/07/1999 TM1022 MSE9932 St Lucie Canal Gate/lock Martin ATL Necropsy M 07/25/1999 TM1023 MSE9933 Biscayne Bay Gate/lock Dade ATL Necropsy M 07/30/1999 TM1024 MSE9936 Biscayne Bay Gate/lock Dade ATL Necropsy M 09/02/1999 TM1025 MSE9939 Biscayne Bay Gate/lock Dade ATL Necropsy M 09/27/1999 TM1026 MSE9941 North Fork, St Lucie River Undetermined, decomposed St. Lucie ATL Necropsy M 10/09/1999 TM1027 MSE9942 Okeechobee Canal Gate/lock Martin ATL Necropsy M 10/17/1999 TM1028 MSE9944 North Fork, St Lucie River Perinatal St. Lucie ATL Necropsy F 11/01/1999 TM1029 MSE9945 Lake Okeechobee Gate/lock Okeechobee ATL Necropsy M 12/03/1999 TM1030 MSE9946 Tamiami Canal Gate/lock Dade ATL Necropsy F 12/07/1999 TM1031 MSE0001 Taylor Creek Undetermined, decomposed St. Lucie ATL Necropsy M 01/01/2000 TM1032 MSE0002 Intracoastal Waterway Watercraft Palm Beach ATL Necropsy F 01/09/2000 TM1033 MEC9944 Banana River Perinatal Brevard ATL Necropsy F 09/05/1999 TM1034 MEC9945 Indian River Undetermined, decomposed Brevard ATL Necropsy M 09/22/1999 TM1035 MEC9946 Banana River Undetermined, decomposed Brevard ATL Necropsy F 10/11/1999 TM1036 MEC9947 Banana River Perinatal Brevard ATL Necropsy M 10/20/1999 TM1037 MEC9948 Atlantic Ocean Watercraft Volusia ATL Necropsy M 10/31/1999 TM1038 MEC9949 Banana River Natural Brevard ATL Necropsy F 11/04/1999 TM1039 MEC9950 St. Johns River Perinatal Volusia STJR Necropsy F 11/18/1999 TM1040 MEC9951 Indian River Perinatal Brevard ATL Necropsy F 11/19/1999 TM1041 MEC9952 Florida Power and Light Intake off the Indian River Human, other Brevard ATL Necropsy F 11/20/1999

PAGE 123

123Table B-1. Continued TM# FieldID# Locality Notes County MU Sample Type Sex MIPS# Date TM1042 MEC9953 Port Canaveral Inlet Watercraft Brevard ATL Necropsy M 11/21/1999 TM1043 MEC9955 Cane Creek Natural Brevard ATL Necropsy M 12/04/1999 TM1044 MEC9956 Sebastian River Natural Brevard ATL Necropsy M 12/04/1999 TM1045 MEC9958 Sebastian River Natural Brevard ATL Necropsy M 08/29/1999 TM1046 MEC0001 Indian River Undetermined, decomposed Brevard ATL Necropsy M 01/04/2000 TM1047 MEC0002 Indian River Natural Indian River ATL Necropsy F 01/04/2000 TM1048 MEC0003 Indian River Watercraft Brevard ATL Necropsy M 01/06/2000 TM1049 MEC0004 Indian River Watercraft Brevard ATL Necropsy M 01/10/2000 TM1050 MEC0005 Indian River Cold stress Indian River ATL Necropsy M 01/28/2000 TM1051 MEC0006 Indian River Undetermined, decomposed Brevard ATL Necropsy M 02/11/2000 TM1052 MEC0007 Indian River Watercraft Brevard ATL Necropsy M 02/15/2000

PAGE 124

124 APPENDIX C MANATEE INDIVIDUAL GENETIC-IDENTIFICATION SYSTEM (MIGS) DATABASE HELP FILE I hope that you find this database useful. I have put together the following help file to guide you throu gh using the MIGS database. If you want to learn more about the Access program, Microsoft has a very useful online tutorial that you can work through. If you are at the University of Florida, the Health Science Center In formation Technology Training Center also has great Access courses. (M any thanks to Pandora Cowart!) For any questions that are not answer ed in this help file, email me: kimpause@gmail.com Kimberly Pause 2007 Table of Contents: Overview of MIGS Step 1: Inputting Locality Information Step 2: Importing Your Genotype Data Troubleshooting Step 3: Updating the Main Allele Table Step 4: Getting Your Data Out of MIGS! References Links of Interest

PAGE 125

125Overview of MIGS This database was designed in an effort to minimize human error and to expedite the creation of data input f iles for population genetics software packages. Through collaboration with MIPS research ers at the USGS, common fields are designated in the MIGS for li nking two databases at a later date. These fields are the MIPS# field, which is the unique identifie r in the MIPS database, and the FieldID# field, which contains any other numbers asso ciated with the sample. Data for each locus are divided into indi vidual tables. The data are uploaded either by manually i nputting individual entr ies, or by importing larger tables with matching file names from Microsoft Excel into existing tables in the database. Each table has the TM# field as its primary key, and referential integrity is enforced between the linked tables. The TM# is the laborator y identification number for each sample. Each locus table is linke d to the "Tbl Locality Data" table, which contains information such as: the date of sa mple collection, sex, sample type, and MIPS number. To prevent inputting genotypes for nonexistent samples, i ndividuals and their genotypes cannot be added to a locus table un less their locality information is also included in the table, Tbl Locality Data." This screen capture shows how the individual locus tables ar e linked to the Locality Data Table through the TM# field. The MIGS database is a model for a large-scale centralized DNA database. In the future, a similar database may be used for capturerecapture estimates of population parameters in concert with the MIPS database. The DNA database will store genetic profiles from

PAGE 126

126 known MIPS individuals and other individuals that are not phot ographically identifiable. Genotype matches could be made from an unknown sample to another sample, from an unknown sample to a MIPS individual, and fr om a MIPS individual to the same MIPS individual (See the figure below adapted from Walsh and Buckleton (2005)).

PAGE 127

127So you have new tissue samples Step 1 in this process will always be to check that your samp les are included in the Locality Data Table. If they are not there, you cannot upload your genotypic information It is set up this way to avoid duplicat e samples, as well as inputting data for a sample that does not exist! If you know that your sample is already included in the Locality Data Table, then you may proceed to the next step. If you are not sure, from the Main Menu, click the button: Open Locality Data Report and browse for your sample ID numbers. If you notice a problem, contact an administrator (a.k.a. me ) or change the data through the Locality Data Input Form by clicki ng the button on the MIGS Main Menu. If your sample is new to the database, it must be added! From the MIGS Main Menu, open the Locality Data Input Form. This will open up a form with several fields for you to fill-in. To import your data properly through the input form, you must click on the appropriate buttons. The button on the left will create a new record, and the button on the right will save the new record that you just created. Alternatively, you may us e the buttons provided by Access. At the bottom left corner of th e Locality Data Input Form, you will see the following set of buttons: The RED and GREEN arrows indicate the buttons that allow you to navigate through the data set. The RED arrows point to the buttons th at direct you to the first and last record, whereas the GREEN arrows point to the buttons that let you navigate through the list one-by-one. The YELLOW arrow points to the Create New Record button. You will want to click this before adding a new individual. .. Now that you are ready to add a new indi vidual, begin by adding the TM#, and then add any other information that you have for the sample. You can tab through the different fields, or you ca n use your mouse to click on the appropriate field. If you do not have information for a fiel d, leave it blank, however, the TM# field ABSOLUTELY must be filled in. This is the lab identification number. If this is empty, you will receive an error message. This is the most important item to be included because it links all of the different tables together

PAGE 128

128 Lets go through the fields together: TM# Lab identification number. FieldID# Any identifying numbers found on the tube or bottle containing the sample. Sex Select male (M), female (F), or undetermined (U) from the pull-down menu, if available. Otherwise leave blank. Locality Any locality information that is known for the manatee; basically can be used as a notes field. GPS Include only if coordinates are known for the sampling or carcass location. State FL, GA, etc. (Include PR here although it is not technically a state.) Country USA, PR, Mexico, etc. Mgmt Unit For FL this is the Management Unit field. For other re gions, it is the region or to be left blank. Use the pull-down menu to view the possible selections. SampleType Select Skin, Cookie, Blood etc. from the pull-down menu. MIPS# If the manatee is a known MIPS indi vidual, include the MIPS identification number in this field. Date of Collection include as much information in this field as possible. Try to include at least the month and year of collec tion in the form of MM/DD/YYYY. Season: This is an optional field. For FL manatees, the seasons were designated as Winter from November to February and Summer from March to October.

PAGE 129

129Step 2: Uploading your genotype data. After fragment analysis is complete, the alleles must be rounded to the nearest integer for uploading into the MIGS. It is wise to save your raw data and how you rounded these files in separate Microsoft Excel spreadsheets. It is also important that your files are well-organized (e ither by analysis da te, locus, etc.). This way, you can refer back to the chromatograms and raw da ta when there are discrepancies between alleles for an individual that has been genotyped multiple times. In the future, we may implement a way to save the chromatograms in MIGS for easy recall; but for now, you must archive them on your own. Each locus has its own individual allele table in the MIGS. These are combined in a larger allele table that has all loci that are typically being used. The administrator can add/subtract loci from this allele table at any time, if reque sted. If you have data for a locus not included in the main table, you mu st create a new table through the standard Access protocol (if you need help, you can look in the Microsoft Access he lp file). In the MIGS, each locus table has a certain format th at you should adhere to. There are three fields (equivalent to columns in Excel): T M#, LocusAllele1, and LocusAllele2. (NOTE: For your tables, Locus should be re placed with the actual locus name.) Your new table should be created the same way. If you are uploading data for a locus that are already included in the MIGS, then you must also adhere to these guidelines for your table. Remember, if you have samples that you know are already included in the databa se for that locus, you will get an error message. There are ways to get around this which I will discuss later For now, we will assume that all samples that are being uploaded are new (no genotype data exists for those individuals as of yet, and the animals are included in the Locality Data Table). Close MIGS and SAVE A BACKUP COPY OF THE DATABASE on the desktop of your computer (or anywhere that you can easily navigate back to). Then open the copy of the database and proceed with the following steps

PAGE 130

130 Navigate away from the MIGS Main Menu and go to the Main Database Window. Click on the Tables tab (under O bjects) and then clic k New (as seen in the previous screen capture). Next, click on Import Table and OK. Select your file on the next screen and click the Import button. The next window that pops up is the I mport Spreadsheet Wizard. The Wizards are very useful in Access, and should be us ed whenever possible to avoid typographical and other formatting errors. This is a very important step! This is where you check that the field names are spelled correctly. You n eed them to be exactly the same as in the table you are uploading to, however they are not case-se nsitive. They should be: TM# LocusAllele1 and LocusAllele2 After you have made certain this is correct, click Next >. The Wizard will ask you where you want to import your data. Assuming field names are identical and you are uploading all new data for which there are already

PAGE 131

131 corresponding entries in the Locality Data Table, you may proceed. You will need to import your table In an Existi ng Table. Choose the correct Locus Allele Table. Each table is named Tbl LocusConvert edAlleles. This title is to remind you that only data for the rounded allele value may be uploaded to the database. Se lect the proper Locus Allele Table and click the Next > button. The last step in the Wizard is to confirm that you have the correct table. This is very important! If you do not ha ve the right table, your fiel d names are not the same and you will get an error message. (NOTE: Leav e the two boxes at the bottom of the window unchecked.) If all goes well, you should get a message stating that your da ta have imported successfully. If there are individuals incl uded in your new tabl e that already have genotypic data in the database for that locus, you will get an error message. If you have individuals that are not in cluded in the Locality Data Table, you will get an error message. If you made a typographical error in the field names, you will get an error message If you get an error message CLICK THE BUTTON THAT SAYS YOU DO NOT WANT TO PROCEED WITH THE UPLOAD. Then read wh at I have to say about troubleshooting your error messages If all is well, move on to Step 3

PAGE 132

132ERROR MESSAGE Troubleshooting! If you are getting an error message, you have one of the following four potential problems: 1. You do not have the right locus table and ther efore, your field names are not the same. 2. You do have the right locus table, but your field names are sp elled incorrectly. 3. You have individuals included in your data set that are not included in the Locality Data Table. 4. You have individuals included in your dataset that already have genotypic information in the database at the locu s for which you are attempting to upload new genotypic data. Begin by ruling out problems 1 and 2. These are the easiest problems to solve, since you just will need to go back to your Excel spreadsheet and check the column headings. If you suspect you have problem 3 or 4, upload your table as a NEW table. To do this, we will go through similar steps as previously de scribed in Step 2 of this Help document. Click on Import Table and OK. Select your file on the next screen and click the Import button. The next windo w is the Import Spreadsheet Wizard. Remember to double check that your field headings are: TM# LocusAllele1 and LocusAllele2 After you have made certain this is correct, click Next >.

PAGE 133

133 Import your tables In a New Table and click Next>. Click Next > when you see this window. It is not im portant for our purposes. This window is very important. You w ill want to make sure that your TM# field is your PRIMARY KEY. For our purposes the primary key is defined as the field that links all of the tables t ogether. Select TM# as your primary key in the Choose my own primary key drop-down window and then click Next >.

PAGE 134

134 Name your table however you wish, because you will delete it later. Do not duplicate the name of any existing tables. We will refer to this as the Error Test table. (It might be helpful if you included that in your name.) Then click Finish. You next need to query the database to figure out what is the problem with your upload. Click on the Query ta b and then click New. First, we will go through to check that the animals are include d in the Locality Data Table To do this, choose the Fi nd Unmatched Query Wizard and click OK.

PAGE 135

135 Select your newly create d Error Test table and click Nex t > and then select the Tbl Locality Data and click Next >. As you recall, the TM# is what links all of the tables together. Make sure that these fields are highlighted and then click Next >. You do not need to see all of the details for this step, so just highlight the TM# field in the Available fields: section and click the > butto n to bring it to the Selected fields: section. Then click Finish. A table will now open with the individuals who do not have matches in the Locality Data Table. No w you must decide if those individuals were due to typographical errors or if they need to be added to the Locality Data Table

PAGE 136

136 Once you are finished with your query-ge nerated table, please delete it by clicking the name with the mouse and then clicking the X button at the top middle of the Main Database Window. The table will have the names of both tables that you queried linked by the words without matching. If this solves your problem, delete your Error Test table from the Table s tab in the same manner. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ If the table generated by the Find Unma tched query is empty, then you do not have samples that are not included in the Local ity Data Table. This indicates that your problem is that you are trying to import duplic ate data. This is where your scientific knowledge will come into play! You will need to query the two tables and decide the following: A) Are the genotypes the same for the i ndividual in question? This is the optimal situation it means that you have done good work and your genotypes match! B) Are the genotypes different for the i ndividual in question? This is not what you want to see this means, that you have a genotyping error. It could be a rounding error, a data in putting error, or an error with the fragment analysis. In either situation, the process preceding your decision is the same From the Queries tab, click the New button.

PAGE 137

137 Select Design View from the New Query Wizard and click OK. The Show Table window will open to ask which Tables or Queries are to be queried. We will only be querying Tables for th is, so make sure that the Tables tab is selected. Choose your newly created data ta ble (your Error Test table) and click Add, and then select the Tbl LocusConvertedAllele s data table of the same locus and click Add. Then click Close. Double click on the in the windows. Alternatively, you may click and drag the into the cells at the bottom. Either way, your query should resemble this one.

PAGE 138

138 In the upper right hand corner of the screen, you will see the View button. Click on it. Take a minute to try to figure out the error that was made when attempting to import this table. .. Thats right! I tried to import a table fo r locus TmaE1 into the TmaM79 table. This is an easy fix! Go back and import th e table into the proper Locus Allele Table. When you are done, make sure to delete the Error Test table that you made from the Tables tab in the Main Database Window. What would it look like if the problem we re that the new da ta table already included information for indivi duals in the existing table?

PAGE 139

139 Here you can see that the individuals in the three columns to the left (my new data) have matches in the existing table (the la st three columns). This must be resolved! When I examine these tables, I like to bri ng them into Excel to work with them. To do this, click in the empty gray cell in the top left hand corner of the query window and press Ctrl and A and then Ctrl and C. This will select and copy all of the information in this table. Then paste the ta ble into an Excel spreadsheet. Examine your spreadsheet to determine if your genotypes are the same for both attempts or if they are different. If they are the same, you should crea te a copy of your origin al Excel table file that you were trying to upload. Then, de lete the matched individuals and their corresponding genotypes. Save this file. Now you can import the newly created data table (without matches) into the database as previously described. If you have individuals that have genotypes that do not match You can go ahead and upload the data fo r the other individuals. Do this by returning to your original Excel file and saving it as a copy. Delete the individuals with questionable genotypes, and import the table as previously described. You will have to return to your original chromatograms to decide which set of genotypes you want to save in the database. You might think of it as a drawback that you can only save one set of genotypic information for an individual (not saving multiple possible genotypes). If you need to save ge notypic information for an individual several times (for example, sub-samples or intent ionally repeated samp les), you can create a slightly different name for the sample (remember to include it in the Locality Data Table.) This could get messy, so I caution against doing it frequently. You can always refer back to your Excel spread sheets. This is why organi zing your fragment analysis files is so critical! You will have to go b ack to these files and decide which genotype should be included in the database.

PAGE 140

140 Once you have decided which genotype is correct, you must change it in the database. ONLY DO THIS AFTER firs t checking with the administrator (me ) or a senior lab person (Maggie Kellogg )! DO THIS WITH CAUTION SAVE A BACKUP COPY OF THE DATAB ASE BEFORE YOU DO THIS!! This can be done in Windows by selec ting the icon, right clicking, and copying and pasting the file. Select the table you want to edit, and doubl e click to open. The Table will pop up and you can navigate through it w ith the scroll bar. Find the TM# of the individual you want to fix. Click on the cell in the allele column that you ne ed to alter a nd type in the new value. Click elsewhere in the table or pr ess an arrow key to navigate away from the cell, which will save the change you made. If you mess this up, it will be difficult to go back and locate the proper allelic information. PLEASE REMEMBER TO SAVE A BACKUP COPY OF THE DATABASE BEFORE DOING THIS! If this did not help you to solve your problem, contact an administrator (me ) to help you fix the problem. Now you are done troubleshooting! Proceed to Step 3 .

PAGE 141

141Step 3: Updating the Allele Table You must update the main A llele Table that includes al l of the loci whenever you are done uploading new data. It is not necessary to do this step after every locus you upload; however, before you exit the database you should perform the Update Allele Table query. This is accomplished from the MIGS Main Menu. Simply click the Update Allele Table button, and when asked if you are sure that you want to proceed, click Yes. This will take a few minutes, depending on your computers processing speed and how much data was changed. IT IS IMPORTANT THAT YOU DO NOT DO ANYTHING ELSE WHILE THE TABLE IS BEI NG UPDATED! Your computer could freeze or crash, and it will slow or halt the updating process. When you have completed all of these steps, close the MIGS and change the name of the copied file to MIGS_200X_Month_Day. It is important that the information on the last update is included in the database name because two people could be working on different c opies of the database! Communication among MIGS researchers is critical to avoid creating two databases with different information! Now that the Main Allele Table is update d, you can query the database for the data of interest. Go to Step 4 to learn how to do this

PAGE 142

142Step 4: Getting your data out of MIGS! Youve successfully added a ll of your data into the database, a nd now you want to get it out so you can analyze it using differe nt population genetics software packages. The database is great for generating these file s because it will leav e the cells blank where there is no data! This means that you will not have to go back and do a lot of cutting and pasting from an Excel spreadsheet, which is just asking for trouble with large datasets! Currently, there are several options from the Main Menu for rapid querying of the database. If you need all the data found in the database, you can click the GET ALL DATA button. If you just want the data for a location of interest, you have some options. For Florida data, you can either query a single management unit, or you can request all data for the state of Florida. For Puerto Rico, you can query for all data from the region and all the genotypes with ex tra locality and date information. Options can be added in the future to query for data from Mexico, Belize, Australia (dugong), and any ot her sampling locations for which genotypic data is generated.

PAGE 143

143 Reference: Walsh SJ, Buckleton J (2005) DNA Intelligence Databases. In: Forensic DNA Evidence Interpretation (eds. Buckleton J, Triggs CM, Walsh SJ), pp. 439-469. CRC Press, Boca Raton, FL, USA. Links and Acknowledgements Funding for the genotyping of manatees and deve lopment of the database was provided by the United States Geological Survey and the University of Florida College of Veterinary Medicine Marine Animal Health Program The MIGS database was developed with the assistance of Pandora Cowart at the UF Health Science Center Information Technology Training Center All laboratory work was completed at the UF Interdisciplinary Center for Biotechnology Research Genetic Analysis Core Lab My academic program here at UF is the Interdisciplinary Program in Biomedical Sciences The Florida Fish and Wildlif e Conservation Commissions Fish and Wildlife Research Institute and Mote Marine Laboratory is also working on mana tee genetics projects.

PAGE 144

144 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.00000.25780.74220.0000 StJR 0.00000.00000.27360.72640.0000 NW 0.00000.00610.18900.80490.0000 SW 0.01050.00000.26840.71580.0053 248250252254256 0.0000 0.2000 0.4000 0.6000 0.8000 ATL 0.00000.00390.34770.61720.0312 StJR 0.00000.00910.24550.70910.0364 NW 0.00000.00000.45120.50000.0488 SW 0.00520.00520.34540.60310.0412 169171173175177 APPENDIX D SUPPLEMENTAL INFORMATION Figure D-1. TmaA02 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-2. TmaE02 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green.

PAGE 145

145 Figure D-3. TmaE08 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-4. TmaE11 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 ATL 0.06250.44920.07420.28120.00000.05470.07030.0078 StJR 0.02730.59090.04550.20000.00000.08180.05450.0000 NW 0.06710.44510.00610.31710.00000.14020.00610.0183 SW 0.11860.47940.02060.22680.00520.08760.06190.0000 182184196198200202204206 0.0000 0.2000 0.4000 0.6000 0.8000 ATL 0.00390.35940.00780.10550.00000.49610.0273 StJR 0.02730.30910.00000.04550.00910.59090.0182 NW 0.02440.38410.00000.01830.00000.53660.0366 SW 0.01550.42270.02060.02580.00520.42780.0825 150152154156164166168

PAGE 146

146 Figure D-5. TmaE26 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-6. TmaF14 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.00000.29300.7070 StJR 0.00000.00910.42730.5636 NW 0.00000.00000.29880.7012 SW 0.01030.00000.18560.8041 199203205207 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00410.80740.1885 StJR 0.01850.85190.1296 NW 0.00610.97560.0183 SW 0.00540.94620.0484 203205207

PAGE 147

147 Figure D-7. TmaM79 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-8. TmaSC5 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.01190.62300.3651 StJR 0.02830.60380.3679 NW 0.00610.63410.3598 SW 0.00000.56770.4323 151153155 0.0000 0.1000 0.2000 0.3000 0.4000 0.5000 0.6000 0.7000 0.8000 0.9000 1.0000 ATL 0.00000.00000.02050.34840.00410. 00410.24180.00000.00000.01230.3689 StJR 0.00930.00000.01850.29630.00000. 00930.26850.01850.00000.01850.3611 NW 0.00000.00650.01950.38960.00000. 02600.16230.00000.00000.05190.3442 SW 0.00000.00000.02910.25000.02330. 04070.27330.00000.00580.04650.3314 119121123125127129131133135137139

PAGE 148

148 Figure D-9. TmaSC13 allele freq uency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-10. TmaE1 allele frequency distributio n for both seasons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.39020.00810.00000.00000.00000.01220.5894 StJR 0.34620.01920.00960.00000.00960.00960.6058 NW 0.28210.01920.00000.00000.00640.01920.6731 SW 0.28490.02150.00000.00540.00000.02690.6613 107115119123125127129 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.01330.02210.05750.16810.73450.0044 StJR 0.00000.00000.00000.06730.20190.73080.0000 NW 0.00000.07410.03700.10490.27780.50620.0000 SW 0.01120.05620.03930.05620.18540.63480.0169 266272274276278280282

PAGE 149

149 Figure D-11. TmaE4 allele frequency distributio n for both seasons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-12. TmaE7 allele frequency distributio n for both seasons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.15160.00000.32790.5205 StJR 0.22120.01920.36540.3942 NW 0.23680.00660.28950.4671 SW 0.25540.02170.33700.3859 190192194196 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.02420.77820.1976 StJR 0.00000.78570.2143 NW 0.00000.88270.1173 SW 0.00000.81460.1854 246248250

PAGE 150

150 Figure D-13. TmaH13 allele frequency distributi on for both seasons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-14. TmaE14 allele frequency distributi on for both seasons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.77270.09550.1318 StJR 0.00000.83330.14710.0196 NW 0.00000.87100.04030.0887 SW 0.00610.73780.10370.1524 241245249253 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.63720.00000.02210.09730.00000.2434 StJR 0.00000.66670.02560.02560.15380.01280.1154 NW 0.02000.44000.00000.01000.34000.03000.1600 SW 0.00000.42420.00000.00000.12880.00000.4470 236238240248250252254

PAGE 151

151 Figure D-15. TmaH23 allele frequency distributi on for both seasons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-16. TmaK01 allele frequency distributi on for both seasons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.15220.8478 StJR 0.09000.9100 NW 0.15080.8492 SW 0.05810.9419 222230 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.01740.60870.00000.00870.30000.0652 StJR 0.00000.00000.64290.01020.00000.31630.0306 NW 0.00620.01230.71600.00000.00000.17900.0864 SW 0.00000.01140.69320.00000.00570.21020.0795 188190192194196198200

PAGE 152

152 Figure D-17. TmaJ02 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-18. TmaKb60 allele fre quency distribution for both seas ons. Allele frequencies are visualized by histograms with the associ ated allele frequency values. Each management units is represented separately to show allele frequency shifts between regions. The management units are repr esented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.19840.03970.74210.0198 StJR 0.26360.04550.68180.0091 NW 0.20990.04940.70990.0309 SW 0.16840.03680.78420.0105 224226228230 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.20490.52870.26230.0041 StJR 0.20370.49070.30560.0000 NW 0.15750.42470.41780.0000 SW 0.17370.54210.28420.0000 215217219225

PAGE 153

153 Table D-1. Diversity statistics fo r the 362 manatees over all loci fo r when data are grouped as the east and west coast populations. Loci with possible null alleles on either coast are denoted with *. Total NA, total number of alleles in the population subset; N number of manatees sampled; NA, number of alleles; HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibrium, values in italics are not in equilibrium after a sequential Bonferroni correction. Total East Coast West Coast Locus Name NA N NA HO HE P HWE N NA HO HE P HWE TmaA02 6 181 2 0.3481 0.3882 0.1796 177 5 0.3390 0.3742 0.0110 TmaE02* 5 183 4 0.4262 0.4840 0.0002 179 5 0.4302 0.5354 0.0000 TmaE08 7 183 7 0.5792 0.5995 0.0000 179 7 0.5754 0.6047 0.0000 TmaE11 8 183 7 0.6448 0.6788 0.0008 179 8 0.6536 0.6919 0.0000 TmaE26 4 176 3 0.2557 0.2976 0.0282 175 3 0.0686 0.0774 0.2373 TmaF14 4 183 3 0.4317 0.4493 0.7456 179 3 0.3296 0.3716 0.0014 TmaM79 4 179 3 0.4637 0.4861 0.0007 178 3 0.5281 0.4843 0.1786 TmaSC5 12 176 9 0.7500 0.6940 0.0000 163 9 0.6135 0.7355 0.0000 TmaSC13 8 175 6 0.5143 0.5058 0.0124 171 6 0.4971 0.4755 0.0116 TmaE1* 8 165 6 0.3697 0.4276 0.0007 170 7 0.5529 0.6082 0.0595 TmaE4 4 173 3 0.3006 0.3508 0.0000 170 2 0.2706 0.2599 0.7691 TmaE7* 5 174 4 0.5345 0.6240 0.0016 168 4 0.5893 0.6626 0.0266 TmaH13 5 161 3 0.3292 0.3522 0.2950 144 4 0.3819 0.3475 0.7568 TmaE14* 8 152 6 0.5526 0.5286 0.0071 116 6 0.5603 0.6640 0.0013 TmaH23 3 165 2 0.2182 0.2318 0.4952 149 2 0.1812 0.1763 1.0000 TmaK01 8 164 6 0.7134 0.5224 0.0000 169 6 0.5740 0.4604 0.0000 TmaJ02* 5 181 4 0.4033 0.4277 0.0000 176 4 0.3352 0.4013 0.0000 TmaKb60* 5 176 4 0.3977 0.6166 0.0000 168 3 0.4821 0.6158 0.0000 Mean 6.1 173.9 4.6 0.4574 0.4814 167.2 4.8 0.4424 0.4748

PAGE 154

154 Table D-2. Diversity statistics for all loci over the 362 manatees where data are grouped as management unit populations. Loci with possible null alleles in the corresponding subpopulations are denoted with *. Total NA, total number of alleles in the population subset; N number of manatees sampled; NA, number of alleles; HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibrium, values in italics are not in equilibrium after a sequen tial Bonferroni correction. Total Atlantic St. Johns River Locus Name NA N NA HO HE P HWE N NA HO HE P HWE TmaA02 6 128 2 0.3125 0.3842 0.0399 53 2 0.434 0.4013 0.7314 TmaE02 5 128 4 0.4531 0.4992 0.0054 55 4 0.3636 0.4395 0.0924 TmaE08 7 128 6 0.6250 0.6152 0.0146 55 6 0.4727 0.5571 0.0171 TmaE11 8 128 7 0.6406 0.7044 0.0002 55 6 0.6546 0.6038 0.3782 TmaE26 4 122 3 0.2623 0.3139 0.0365 54 3 0.2407 0.2596 0.1838 TmaF14 4 128 2 0.4141 0.4159 1.0000 55 3 0.4727 0.5043 0.8747 TmaM79 4 126 3 0.4524 0.4803 0.0158 53 3 0.4906 0.5040 0.0450 TmaSC5 12 122 7 0.7459 0.6863 0.0013 54 8 0.7593 0.7151 0.0078 TmaSC13 8 123 4 0.5203 0.5021 0.0165 52 6 0.5000 0.5176 0.3282 TmaE1* 8 113 6 0.3628 0.4301 0.0021 52 3 0.3846 0.4248 0.2359 TmaE4 4 124 3 0.2984 0.3562 0.0000 49 2 0.3061 0.3402 0.6679 TmaE7* 5 122 3 0.5164 0.6011 0.1750 52 4 0.5769 0.6682 0.0274 TmaH13 5 110 3 0.3636 0.3781 0.6266 51 3 0.2549 0.2864 0.4952 TmaE14 8 113 4 0.5664 0.5272 0.0112 39 6 0.5128 0.5238 0.2877 TmaH23 3 115 2 0.2522 0.2592 0.7225 50 2 0.1400 0.1655 0.3258 TmaK01 8 115 5 0.7304 0.5372 0.0000 49 4 0.6735 0.4906 0.0155 TmaJ02 5 126 4 0.3651 0.4096 0.0000 55 4 0.4909 0.4677 0.0024 TmaKb60* 5 122 4 0.4098 0.6122 0.0000 54 3 0.3704 0.6302 0.0000 Mean 6.1 121.8 4 0.4606 0.4840 52.1 4 0.4499 0.4722 121.8 Northwest Southwest Locus Name N NA HO HE P HWE N NA HO HE P HWE TmaA02 82 3 0.3171 0.3184 1.0000 95 4 0.3579 0.4177 0.0065 TmaE02* 82 3 0.3659 0.5474 0.0000 97 5 0.4845 0.5179 0.0003 TmaE08 82 5 0.6342 0.5657 0.0009 97 7 0.5258 0.6334 0.0000 TmaE11 82 7 0.7317 0.6809 0.0257 97 7 0.5876 0.6963 0.0000 TmaE26 82 3 0.0244 0.0481 0.0382 93 3 0.1075 0.1028 1.0000 TmaF14 82 2 0.3781 0.4216 0.4297 97 3 0.2887 0.3205 0.0072 TmaM79 82 3 0.5244 0.4713 0.2192 96 2 0.5313 0.4934 0.5324 TmaSC5* 77 7 0.6104 0.7042 0.0083 86 8 0.6163 0.7521 0.0001 TmaSC13 78 5 0.4615 0.4696 0.1807 93 5 0.5269 0.4829 0.0169 TmaE1* 81 5 0.6296 0.6528 0.6364 89 7 0.4832 0.5575 0.0072 TmaE4 81 2 0.2346 0.2083 0.5885 89 2 0.3034 0.3038 1.0000 TmaE7 76 4 0.5790 0.6461 0.4818 92 4 0.5978 0.6755 0.0446 TmaH13 62 3 0.2581 0.2338 1.0000 82 4 0.4756 0.4242 0.8257 TmaE14* 50 6 0.6800 0.6705 0.0447 66 3 0.4697 0.6083 0.0906 TmaH23 63 2 0.2698 0.2582 1.0000 86 2 0.1163 0.1102 1.0000 TmaK01 81 5 0.5556 0.4504 0.0495 88 5 0.5909 0.4715 0.0000 TmaJ02* 81 4 0.3457 0.4514 0.0000 95 4 0.3263 0.3571 0.0046 TmaKb60* 73 3 0.5206 0.6246 0.2017 95 3 0.4526 0.5983 0.0000 Mean 76.5 4.0 0.4511 0.4680 90.7 4.3 0.4357 0.4735

PAGE 155

155 Figure D-19. TmaA02 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-20. TmaE02 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.01090.32610.64130.0217 StJR 0.01250.23750.70000.0500 NW 0.00000.45270.49320.0541 SW 0.01220.30490.58540.0976 171173175177 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.00000.20650.79350.0000 StJR 0.00000.00000.26320.73680.0000 NW 0.00000.00680.18240.81080.0000 SW 0.02440.00000.34150.62200.0122 248250252254256

PAGE 156

156 Figure D-21. TmaE08 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-22. TmaE11 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.31520.01090.09780.00000.54350.0326 StJR 0.03750.27500.00000.06250.01250.61250.0000 NW 0.02030.37840.00000.01350.00000.54730.0405 SW 0.03660.46340.02440.03660.00000.36590.0732 150152154156164166168 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.09780.41300.08700.25000.00000.07610.06520.0109 StJR 0.01250.65000.06250.15000.00000.07500.05000.0000 NW 0.07430.43240.00680.31080.00000.14860.00680.0203 SW 0.19510.45120.02440.18290.01220.08540.04880.0000 182184196198200202204206

PAGE 157

157 Figure D-23. TmaE26 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-24. TmaF14 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.40220.5978 StJR 0.01250.45000.5375 NW 0.00000.29730.7027 SW 0.00000.21950.7805 203205207 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.80680.1932 StJR 0.02560.84620.1282 NW 0.00000.97970.0203 SW 0.01220.97560.0122 203205207

PAGE 158

158 Figure D-25. TmaM79 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-26. TmaSC5 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.01110.55560.4333 StJR 0.03950.57890.3816 NW 0.00680.62840.3649 SW 0.00000.50000.5000 151153155 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.00000.01190.38100.00000.00000.21430.00000.01190.3810 StJR 0.01250.00000.02500.31250.00000.01250.21250.02500.02500.3750 NW 0.00000.00690.01390.38890.00000.02080.15970.00000.05560.3542 SW 0.00000.00000.01320.19740.02630.01320.26320.00000.05260.4342 119121123125127129131133137139

PAGE 159

159 Figure D-27. TmaSC13 allele fre quency distribution for manatees collected during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-28. TmaE1 allele frequency distributio n for manatees collected during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.47670.01160.00000.00000.01160.5000 StJR 0.32890.01320.01320.01320.01320.6184 NW 0.26430.02140.00000.00710.00710.7000 SW 0.31710.04880.00000.00000.04880.5854 107115119125127129 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.01250.06250.13750.78750.0000 StJR 0.00000.00000.09210.22370.68420.0000 NW 0.07530.02740.10960.26710.52050.0000 SW 0.02560.03850.06410.19230.65380.0256 272274276278280282

PAGE 160

160 Figure D-29. TmaE4 allele frequency distributio n for manatees collected during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-30. TmaE7 allele frequency distributio n for manatees collected during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.17050.00000.26140.5682 StJR 0.22970.02700.33780.4054 NW 0.22060.00740.29410.4779 SW 0.23750.03750.31250.4125 190192194196 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.04350.79350.1630 StJR 0.00000.75000.2500 NW 0.00000.89040.1096 SW 0.00000.83330.1667 246248250

PAGE 161

161 Figure D-31. TmaH13 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-32. TmaE14 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.83330.06940.0972 StJR 0.81580.17110.0132 NW 0.86610.04460.0893 SW 0.75000.11670.1333 245249253 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.60530.00000.01320.15790.00000.2237 StJR 0.00000.68330.03330.03330.11670.00000.1333 NW 0.02380.45240.00000.01190.33330.03570.1429 SW 0.00000.36670.00000.00000.10000.00000.5333 236238240248250252254

PAGE 162

162 Figure D-33. TmaH23 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-34. TmaK01 allele frequency distributi on for manatees collecte d during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.12820.8718 StJR 0.10810.8919 NW 0.15450.8455 SW 0.05710.9429 222230 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.02330.61630.00000.26740.0930 StJR 0.00000.00000.61760.01470.33820.0294 NW 0.00680.01370.70550.00000.17810.0959 SW 0.00000.02860.67140.00000.21430.0857 188190192194198200

PAGE 163

163 Figure D-35. TmaJ02 allele freque ncy distribution for manatees collected during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-36. TmaKb60 allele freque ncy distribution for manatees collected during the winter. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.22220.05560.71110.0111 StJR 0.26250.01250.71250.0125 NW 0.21230.05480.69860.0342 SW 0.19510.04880.75610.0000 224226228230 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.12500.67050.19320.0114 StJR 0.18750.48750.32500.0000 NW 0.17420.38640.43940.0000 SW 0.07500.66250.26250.0000 215217219225

PAGE 164

164 Table D-3. Diversity statistics fo r all loci for when data are groupe d as the winter east and west coast populations. Loci with possible null alleles on either coast are denoted with *. Total NA, total number of alleles in the population subset; N number of manatees sampled; NA, number of alleles; HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibr ium, values in italics are not in equilibrium after a sequentia l Bonferroni correction. Total East Coast West Coast Locus Name NA N NA HO HE P HWE N NA HO HE P HWE TmaA02 6 96 2 0.3750 0.3663 1.0000 116 5 0.3535 0.3893 0.0195 TmaE02* 4 98 4 0.4898 0.4812 0.1352 116 4 0.3966 0.5604 0.0000 TmaE08 7 98 7 0.5816 0.5727 0.0360 116 6 0.6379 0.6013 0.0000 TmaE11 8 98 7 0.6327 0.6587 0.1410 116 8 0.6638 0.7068 0.0000 TmaE26 4 95 3 0.2632 0.2849 0.2307 116 3 0.0259 0.0425 0.0411 TmaF14 3 98 3 0.4898 0.4897 1.0000 116 2 0.3535 0.3973 0.2449 TmaM79 4 95 3 0.5790 0.5151 0.0069 116 3 0.5517 0.4923 0.1544 TmaSC5 11 94 8 0.7447 0.6912 0.0051 111 8 0.6306 0.7115 0.0002 TmaSC13 7 93 6 0.4839 0.5175 0.1933 112 5 0.5268 0.4884 0.0152 TmaE1 7 89 4 0.3708 0.4405 0.1107 113 6 0.5398 0.6068 0.1959 TmaE4 4 92 3 0.2826 0.3376 0.0001 113 2 0.2389 0.2247 0.6908 TmaE7* 5 93 4 0.5161 0.6393 0.0034 109 4 0.5963 0.6543 0.1095 TmaH13 4 85 3 0.2824 0.3153 0.2879 87 3 0.3218 0.3014 0.6642 TmaE14 8 75 6 0.5600 0.5435 0.0237 73 6 0.6027 0.6816 0.0072 TmaH23 3 88 2 0.1705 0.1937 0.2496 91 2 0.2088 0.2053 1.0000 TmaK01 7 89 5 0.7079 0.5240 0.0000 109 5 0.5872 0.4765 0.0001 TmaJ02* 5 97 4 0.4536 0.4291 0.0127 115 4 0.3565 0.4360 0.0000 TmaKb60* 5 96 4 0.3750 0.5817 0.0000 107 3 0.5047 0.6040 0.0310 Mean 5.7 92.7 4.3 0.4644 0.4768 108.4 4.4 0.4498 0.4767

PAGE 165

165 Table D-4. Diversity statistics for all loci for when data are grouped as winter management unit populations. Loci with possible null alle les in the corresponding subpopulations are denoted with *. Total NA, total number of alleles in the population subset; N number of manatees sampled; NA, number of alleles; HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibrium, values in italics are not in equilibrium after a sequential Bonferroni correction. Total Atlantic St. Johns River Locus Name NA N NA HO HE P HWE N NA HO HE P HWE TmaA02 6 46 2 0.3261 0.3313 1.0000 38 2 0.3684 0.3930 0.6893 TmaE02 4 46 4 0.5000 0.4871 0.9230 40 4 0.4000 0.4567 0.2032 TmaE08 7 46 5 0.6087 0.6011 0.2286 40 5 0.5500 0.5506 0.3328 TmaE11 8 46 7 0.6957 0.7477 0.2812 40 6 0.5750 0.5497 0.3694 TmaE26 4 44 2 0.2955 0.3153 0.6434 39 3 0.2308 0.2704 0.2076 TmaF14 3 46 2 0.5000 0.4861 1.0000 40 3 0.4750 0.5149 0.8634 TmaM79 4 45 3 0.6222 0.5091 0.1311 38 3 0.5263 0.5246 0.0663 TmaSC5 11 42 5 0.7619 0.6715 0.3439 40 8 0.7500 0.7234 0.0046 TmaSC13 7 43 4 0.4651 0.5286 0.4440 38 6 0.5263 0.5154 0.2846 TmaE1 7 40 4 0.3500 0.3614 0.0720 38 3 0.4211 0.4797 0.1213 TmaE4* 4 46 3 0.2391 0.3457 0.0004 34 2 0.3824 0.3806 1.0000 TmaE7 5 44 3 0.4773 0.5865 0.3320 37 4 0.5676 0.6772 0.0392 TmaH13 4 36 3 0.3056 0.2954 0.6641 38 3 0.2632 0.3091 0.4120 TmaE14 8 38 4 0.6316 0.5660 0.5693 30 5 0.5333 0.5079 0.2170 TmaH23 3 39 2 0.2051 0.2264 0.4828 37 2 0.1622 0.1955 0.3439 TmaK01 7 43 4 0.7209 0.5458 0.0001 34 4 0.7059 0.5105 0.0563 TmaJ02 5 45 4 0.4222 0.4467 0.0210 40 4 0.4750 0.4285 0.8387 TmaKb60* 5 44 4 0.3182 0.5031 0.0000 40 3 0.3500 0.6294 0.0000 Mean 5.7 43.3 3.6 0.4692 0.4753 37.8 3.9 0.4590 0.4787 Northwest Southwest Locus Name N NA HO HE P HWE N NA HO HE P HWE TmaA02 74 3 0.2973 0.3114 0.7517 41 4 0.4634 0.5020 0.0338 TmaE02* 74 3 0.3649 0.5526 0.0000 41 4 0.4390 0.5616 0.0038 TmaE08 74 5 0.6081 0.5588 0.0023 41 6 0.6829 0.6507 0.0055 TmaE11* 74 7 0.7432 0.6930 0.0238 41 7 0.5366 0.7233 0.0000 TmaE26* 74 2 0.0135 0.0400 0.0206 41 3 0.0488 0.0485 1.0000 TmaF14 74 2 0.3784 0.4207 0.4095 41 2 0.2927 0.3469 0.3656 TmaM79 74 3 0.5405 0.4752 0.1661 41 2 0.5610 0.5062 0.5421 TmaSC5* 72 7 0.5972 0.6989 0.0308 38 7 0.6842 0.7088 0.0044 TmaSC13 70 5 0.4429 0.4428 0.5370 41 4 0.6585 0.5589 0.0076 TmaE1 73 5 0.6164 0.6437 0.5298 39 6 0.4103 0.5355 0.0422 TmaE4 73 2 0.2192 0.1965 0.5884 39 2 0.2821 0.2814 1.0000 TmaE7 68 4 0.5735 0.6411 0.7176 40 4 0.6250 0.6829 0.2333 TmaH13 56 3 0.2679 0.2421 1.0000 30 3 0.4333 0.4130 0.5323 TmaE14 42 6 0.6905 0.6698 0.0439 30 3 0.5000 0.5808 0.6264 TmaH23 55 2 0.2727 0.2637 1.0000 35 2 0.1143 0.1093 1.0000 TmaK01 73 5 0.5753 0.4643 0.0560 35 4 0.6000 0.5023 0.0090 TmaJ02* 73 4 0.3425 0.4659 0.0000 41 3 0.3902 0.3927 0.1340 TmaKb60 66 3 0.5303 0.6321 0.3070 40 3 0.4500 0.4927 0.0489 Mean 68.8 3.9 0.4 0.5 38.6 3.8 0.5 0.5

PAGE 166

166 Figure D-37. TmaA02 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-38. TmaE02 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.35980.60370.0366 StJR 0.00000.26670.73330.0000 NW 0.00000.43750.56250.0000 SW 0.00890.37500.61610.0000 169173175177 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.28660.7134 StJR 0.30000.7000 NW 0.25000.7500 SW 0.21300.7870 252254

PAGE 167

167 Figure D-39. TmaE08 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-40. TmaE11 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00610.38410.00610.10980.00000.46950.0244 StJR 0.00000.40000.00000.00000.00000.53330.0667 NW 0.06250.43750.00000.06250.00000.43750.0000 SW 0.00000.39290.01790.01790.00890.47320.0893 150152154156164166168 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.04270.46950.06710.29880.04270.07320.0061 StJR 0.06670.43330.00000.33330.10000.06670.0000 NW 0.00000.56250.00000.37500.06250.00000.0000 SW 0.06250.50000.01790.25890.08930.07140.0000 182184196198202204206

PAGE 168

168 Figure D-41. TmaE26 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-42. TmaF14 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.23170.7683 StJR 0.00000.36670.6333 NW 0.00000.31250.6875 SW 0.01790.16070.8214 199205207 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00640.80770.1859 StJR 0.00000.86670.1333 NW 0.06250.93750.0000 SW 0.00000.92310.0769 203205207

PAGE 169

169 Figure D-43. TmaM79 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-44. TmaSC5 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.01230.66050.3272 StJR 0.00000.66670.3333 NW 0.00000.68750.3125 SW 0.00000.61820.3818 151153155 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.02500.33120.00630.00630.25620.00000.01250.3625 StJR 0.00000.25000.00000.00000.42860.00000.00000.3214 NW 0.10000.40000.00000.10000.20000.00000.00000.2000 SW 0.04170.29170.02080.06250.28120.01040.04170.2500 123125127129131135137139

PAGE 170

170 Figure D-45. TmaSC13 allele fre quency distribution for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-46. TmaE1 allele frequency distributio n for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.34380.00630.00000.01250.6375 StJR 0.39290.03570.00000.00000.5714 NW 0.43750.00000.00000.12500.4375 SW 0.25960.00000.00960.00960.7212 107115123127129 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.02050.02740.05480.18490.70550.0068 StJR 0.00000.00000.00000.00000.14290.85710.0000 NW 0.00000.06250.12500.06250.37500.37500.0000 SW 0.02000.08000.04000.05000.18000.62000.0100 266272274276278280282

PAGE 171

171 Figure D-47. TmaE4 allele frequency distributio n for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-48. TmaE7 allele frequency distributio n for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.14100.00000.36540.4936 StJR 0.20000.00000.43330.3667 NW 0.37500.00000.25000.3750 SW 0.26920.00960.35580.3654 190192194196 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.01280.76920.2179 StJR 0.00000.86670.1333 NW 0.00000.81250.1875 SW 0.00000.80000.2000 246248250

PAGE 172

172 Figure D-49. TmaH13 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-50. TmaE14 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.00000.74320.10810.1486 StJR 0.00000.88460.07690.0385 NW 0.00000.91670.00000.0833 SW 0.00960.73080.09620.1635 241245249253 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.65330.02670.06670.00000.2533 StJR 0.61110.00000.27780.05560.0556 NW 0.37500.00000.37500.00000.2500 SW 0.47220.00000.15280.00000.3750 238248250252254

PAGE 173

173 Figure D-51. TmaH23 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-52. TmaK01 allele frequency distributi on for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.16450.8355 StJR 0.03850.9615 NW 0.12500.8750 SW 0.05880.9412 222230 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.01390.60420.01390.31940.0486 StJR 0.00000.70000.00000.26670.0333 NW 0.00000.81250.00000.18750.0000 SW 0.00000.70750.00940.20750.0755 190192196198200

PAGE 174

174 Figure D-53. TmaJ02 allele freque ncy distribution for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. Figure D-54. TmaKb60 allele freque ncy distribution for manatees collected during the summer. Allele frequencies are visualized by histogr ams with the associated allele frequency values. Each management units is represen ted separately to show allele frequency shifts between regions. The management units are represented with the following colors and abbreviations: ATL Atlantic, red; StJR St. Johns River, blue; NW Northwest, yellow; SW Southwest, green. 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.18520.03090.75930.0247 StJR 0.26670.13330.60000.0000 NW 0.18750.00000.81250.0000 SW 0.14810.02780.80560.0185 224226228230 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 ATL 0.25000.44870.3013 StJR 0.25000.50000.2500 NW 0.00000.78570.2143 SW 0.24550.45450.3000 215217219

PAGE 175

175 Table D-5. Diversity statistics for all loci for when data are grouped as the summer east and west coast populations. Loci with possible null alleles on either coast are denoted with *. Total NA, total number of alleles in the population subset; N number of manatees sampled; NA, number of alleles; HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibr ium, values in italics are not in equilibrium after a sequentia l Bonferroni correction. Total East Coast West Coast Locus Name NA N NA HO HE P HWE N NA HO HE P HWE TmaA02 3 97 2 0.3505 0.4128 0.1446 62 2 0.3065 0.3434 0.4547 TmaE02 4 97 3 0.4021 0.4933 0.0009 64 3 0.5000 0.4859 1.0000 TmaE08* 7 97 6 0.5773 0.6143 0.0222 64 7 0.4688 0.6193 0.0004 TmaE11 7 97 7 0.6495 0.6826 0.0149 64 6 0.6250 0.6580 0.0078 TmaE26 4 93 3 0.2473 0.3023 0.0324 60 3 0.1500 0.1410 1.0000 TmaF14 3 97 2 0.3814 0.3795 1.0000 64 3 0.2969 0.3225 0.0128 TmaM79 4 96 3 0.3646 0.4571 0.0017 63 2 0.4921 0.4715 0.7914 TmaSC5* 9 94 7 0.7447 0.6947 0.0035 53 8 0.5849 0.7727 0.0006 TmaSC13 6 94 4 0.5319 0.4852 0.0069 60 4 0.4500 0.4559 0.1051 TmaE1 8 87 6 0.3563 0.4351 0.0075 58 7 0.5690 0.6071 0.1898 TmaE4 4 93 3 0.3011 0.3439 0.0053 58 2 0.3276 0.3207 1.0000 TmaE7 5 93 3 0.5484 0.6152 0.5292 60 4 0.5833 0.6741 0.5269 TmaH13 5 87 3 0.3678 0.3898 0.4113 58 4 0.4655 0.4095 0.7505 TmaE14* 6 84 5 0.5238 0.5197 0.0207 44 3 0.4773 0.6392 0.1335 TmaH23 3 89 2 0.2472 0.2509 1.0000 59 2 0.1356 0.1275 1.0000 TmaK01 6 87 5 0.7126 0.5190 0.0000 61 4 0.5574 0.4369 0.0709 TmaJ02 5 96 4 0.3646 0.4211 0.0000 62 4 0.2903 0.3280 0.0297 TmaKb60* 4 92 3 0.4565 0.6465 0.0000 62 3 0.4516 0.6314 0.0001 Mean 5.2 92.8 3.9 0.4515 0.4813 59.8 3.9 0.4295 0.4691

PAGE 176

176 Table D-6. Diversity statistics for all loci for when data are grouped as summer management unit populations. Loci with possible null alle les in the corresponding subpopulations are denoted with *. Total NA, total number of alleles in the population subset; N number of manatees sampled; NA, number of alleles; HO and HE, observed and expected heterozygosities; P HWE p-value for Hardy-Weinberg equilibrium, values in italics are not in equilibrium after a sequential Bonferroni correction. Total Atlantic St. Johns River Locus Name NA N NA HO HE P HWE N NA HO HE P HWE TmaA02* 3 82 2 0.3049 0.4114 0.0285 15 2 0.6000 0.4345 0.2368 TmaE02 4 82 3 0.4268 0.5079 0.0020 15 2 0.2667 0.4046 0.2259 TmaE08* 7 82 6 0.6342 0.6231 0.1987 15 3 0.2667 0.5701 0.0111 TmaE11 7 82 7 0.6098 0.6809 0.0120 15 5 0.8667 0.7058 0.6121 TmaE26* 4 78 3 0.2436 0.3151 0.0230 15 2 0.2667 0.2391 1.0000 TmaF14 3 82 2 0.3659 0.3582 1.0000 15 2 0.4667 0.4805 1.0000 TmaM79* 4 81 3 0.3580 0.4594 0.0019 15 2 0.4000 0.4598 1.0000 TmaSC5 9 80 7 0.7375 0.6967 0.0026 14 3 0.7857 0.6746 0.4656 TmaSC13 6 80 4 0.5500 0.4782 0.0027 14 3 0.4286 0.5370 0.7463 TmaE1* 8 73 6 0.3699 0.4671 0.0083 14 2 0.2857 0.2540 1.0000 TmaE4 4 78 3 0.3333 0.3629 0.0106 15 2 0.1333 0.2391 0.2034 TmaE7 5 78 3 0.5385 0.6069 0.5602 15 3 0.6000 0.6598 0.9133 TmaH13 5 74 3 0.3919 0.4166 0.3702 13 3 0.2308 0.2185 1.0000 TmaE14 6 75 4 0.5333 0.5072 0.0076 9 4 0.4444 0.5752 0.3749 TmaH23 3 76 2 0.2763 0.2767 1.0000 13 2 0.0769 0.0769 1.0000 TmaK01 6 72 5 0.7361 0.5339 0.0000 15 3 0.6000 0.4529 0.6392 TmaJ02 5 81 4 0.3333 0.3901 0.0001 15 3 0.5333 0.5701 0.0017 TmaKb60* 4 78 3 0.4615 0.6496 0.0000 14 3 0.4286 0.6482 0.1582 Mean 5.2 78.6 3.9 0.4558 0.4857 14.2 2.7 0.4267 0.4556 Northwest Southwest Locus Name N NA HO HE P HWE N NA HO HE P HWE TmaA02 8 2 0.5000 0.4000 1.0000 54 2 0.2778 0.3384 0.2228 TmaE02 8 2 0.3750 0.5250 0.5304 56 3 0.5179 0.4841 0.8604 TmaE08* 8 4 0.8750 0.6500 0.2808 56 6 0.4107 0.6186 0.0000 TmaE11* 8 3 0.6250 0.5750 1.0000 56 6 0.6250 0.6717 0.0058 TmaE26 8 2 0.1250 0.1250 1.0000 52 2 0.1539 0.1434 1.0000 TmaF14 8 2 0.3750 0.4583 1.0000 56 3 0.2857 0.3018 0.0121 TmaM79 8 2 0.3750 0.4583 1.0000 55 2 0.5091 0.4764 0.7752 TmaSC5* 5 5 0.8000 0.8222 0.8996 48 8 0.5625 0.7735 0.0008 TmaSC13 8 3 0.6250 0.6417 0.5445 52 4 0.4231 0.4164 0.5420 TmaE1 8 5 0.7500 0.7417 1.0000 50 7 0.5400 0.5780 0.0900 TmaE4 8 2 0.3750 0.3250 1.0000 50 2 0.3200 0.3232 1.0000 TmaE7 8 3 0.6250 0.7000 0.5098 52 4 0.5769 0.6738 0.3666 TmaH13 6 2 0.1667 0.1667 1.0000 52 4 0.5000 0.4341 0.7412 TmaE14 8 3 0.6250 0.7000 0.8076 36 3 0.4444 0.6217 0.0921 TmaH23 8 2 0.2500 0.2333 1.0000 51 2 0.1177 0.1118 1.0000 TmaK01 8 2 0.3750 0.3250 1.0000 53 4 0.5849 0.4548 0.0824 TmaJ02 8 2 0.3750 0.3250 1.0000 54 4 0.2778 0.3311 0.0312 TmaKb60* 7 2 0.4286 0.3626 1.0000 55 3 0.4546 0.6490 0.0004 Mean 7.7 2.7 0.4803 0.4742 52.1 3.8 0.4212 0.4668

PAGE 177

177 LIST OF REFERENCES Allendorf FW, Luikart G (2007) Conservation and the Genetics of Populations Blackwell Publishing, Malden, MA, USA. Alvarez-Alemn A, Powell JA, B eck CA (2007) First report of a Florida manatee documented on the North Coast of Cuba. SireNews (Newsletter of th e IUCN/SSC Sirenia Specialist Group) 47, 9-10. Amos B, Hoelzel AR (1991) Long-term preservation of whale skin for DNA analysis. Report of the International Whalin g Commission Special Issue 13, 99-103. Anderson MJ, Hinten G, Paton D, Baverstock PR (2001) A model for the integration of microsatellite genotyping with photographi c identification of humpback whales. Memoirs of the Queensland Museum 47, 451-457. Anderson PK (2002) Habitat, niche, and e volution of sirenian mating systems. Journal of Mammalian Evolution 9, 55-98. Araabi BN, Kehtarnavaz N, McKinney T, Hillm an G, Wursig B (2000) A string matching computer-assisted system for dolphin photo-identification. Annals of Biomedical Engineering 28, 1269-1279. Archie EA, Moss CJ, Alberts SC (2003) Characteriza tion of tetranucleotide microsatellite loci in the African Savannah Elephant ( Loxodonta africana africana ). Molecular Ecology Notes 3, 244-246. Auger-Mth M, Whitehead H (2007) The use of na tural markings in studies of long-finned pilot whales ( Globicephala melas ). Marine Mammal Science 23, 77-93. Avise JC (2004) Molecular Markers, Natural Hist ory, and Evolution, Second Edn Sinauer Associates, Inc., Sunderland, MA, USA. Balloux F, Amos W, Coulson T (2004) Does heterozygosity estimate inbreeding in real populations? Molecular Ecology 13, 3021-3031. Balloux F, Goudet J (2002) Statis tical properties of population differentiation estimators under stepwise mutation in a finite island model. Molecular Ecology 11, 771-783. Barendse W, Armitage SM, Kossarek LM, Shalom A, Kirkpatrick BW, Ryan AM, Clayton D, Li L, Neibergs HL, Zhang N, Grosse WM, Weiss J, Creighton P, McCarthy F, Ron M, Teale AJ, Fries R, McGraw RA, Moore SS, George s M, Soller M, Womack JE, Hetzel DJS (1994) A genetic linkage ma p of the bovine genome. Nature Genetics 6, 227-235. Bauer GB, Colbert DE, Gaspard JC Littlefield B, Fellner W (2003) Underwater visual acuity of Florida manatees ( Trichechus manatus latirostris ). International Journal of Comparative Psychology 16, 130-142.

PAGE 178

178 Beck CA, Reid JP (1995) An automated photoidentification catalog fo r studies of the life history of the Florida manatee. In: Population biology of the Florida manatee (eds. OShea TJ, Ackerman BB, Percival HF), pp. 120-134. National Biological Service Information and Technology Report 1. Washington, D. C. Blouin MS (2003) DNA-based met hods for pedigree reconstructi on and kinship analysis in natural populations. Trends in Ecology and Evolution 18, 503-511. Bowen BW, Bass AL, Soares L, Toonen RJ ( 2005) Conservation implications of complex population structure: Lessons from the loggerhead turtle ( Caretta caretta ). Molecular Ecology 14, 2389-2402. Brazeau D, Clark G, Norton S, Moraga-Amador D, Liu L (2005) Molecular Markers: Tools for Developing Enriched Microsatellite Libraries University of Florida, Interdisciplinary Center for Biotechnology Res earch, Gainesville, FL, USA. Broderick D, Ovenden J, Slade R, Lanyon JM (20 07) Characterization of 26 new microsatellite loci in the dugong ( Dugong dugon ). Molecular Ecology Notes Online Early. Brown Gladden JG, Ferguson MM, Friesen MK, Clayton JW ( 1999) Population structure of North American beluga whales ( Delphinapterus leucas ) based on nuclear DNA microsatellite variation a nd contrasted with the popul ation structure revealed by mitochondrial DNA variation. Molecular Ecology 8, 347-363. Buckleton J (2005) Populati on Genetic Models. In: Forensic DNA Evidence Interpretation (eds. Buckleton J, Triggs CM, Walsh SJ), pp. 65-112. CRC Press, Boca Raton, FL, USA. Buckleton J, Clayton T, Triggs CM (2005) Parentage Testing. In: Forensic DNA Evidence Interpretation (eds. Buckleton J, Triggs CM, Wa lsh SJ), pp. 341-394. CRC Press, Boca Raton, FL, USA. Caro TM, Laurenson MK (1994) Ecological and ge netic factors in conser vation: A cautionary tale. Science 263, 485-486. CNN.com (2006a) Accidental tourist: M anatee cruises Hudson River Accessed: 7 August 2006. Available from http://www.cnn.com/2006/US/0 8/07/manatee.hudson.river.ap/index.html. CNN.com (2006b) Manatee finds way to Memphis Accessed: 20 November 2006. Available from http://www.cnn.com/2006/US/10/24/ memphis.manatee.ap/index.html. Coltman DW, Slate J (2003) Microsatellite m easures of inbreeding: A meta-analysis. Evolution 57, 971-983. Cornell LH, Asper ED, Antrim JE, Searles SS, Young WG, Goff T (1987) Progress report results of a long-range captive breeding pr ogram for the bottlenose dolphin, Tursiops truncatus and Tursiops truncatus gilli Zoo Biology 6, 41-54.

PAGE 179

179 Cousteau JP, Kaufman J (1989) The Forgotten Mermaids. Pacific Arts Video, Beverly Hills, CA, USA. Dean MD, Ardlie KG, Nachman MW (2006) The fr equency of multiple paternity suggests that sperm competition is common in house mice ( Mus domesticus ). Molecular Ecology 15, 4141-4151. DeSalle R, Amato G (2004) The expa nsion of conservation genetics. Nature Reviews Genetics 5, 702-712. Deutsch CJ, Reid JP, Bonde RK, Easton DE Kochman HI, O'Shea TJ (2003) Seasonal movements, migratory behavior and site fidelity of West Indian manatees along the Atlantic Coast of the United States. Wildlife Monographs 151, 1-77. DiBattista JD (2007) Patterns of genetic variatio n in anthropogenically impacted populations. Conservation Genetics Online Early. Dieringer D, Schl tterer C (2003) MICROSATELLITE ANALYSER (MSA): A platform independent analysis tool for larger microsatellite data sets. Molecular Ecology Notes 3, 167-169. Dixon JD, Oli MK, Wooten MC, Eason TH, McCown JW, Cunningham MW (2007) Genetic consequences of habitat fragmentation and loss: The case of the Florida black bear ( Ursus americanus floridanus ). Conservation Genetics 8, 455-464. Domning DP (1983) Marching teeth of the manatee: Its special ad aptation to an abrasive diet has enabled this aquatic mammal to outdo the dugong. Natural History 92, 8-11. Domning DP (2000) Another view of manatee conservation. SireNews (Newsletter of the IUCN/SSC Sirenia Specialist Group) 34. Domning DP (2005) Fossil Sirenia of the West Atlantic and Caribb ean region. VII. Pleistocene Trichechus manatus Linnaeus, 1758. Journal of Verteb rate Paleontology 25, 685-701. Duffield DA, Odell DK, McBain JF Andrews B (1995) Killer whale ( Orcinus orca ) reproduction at SeaWorld. Zoo Biology 14, 417-430. Eggert LS, Ramakrishnan U, Mundy NI, Woodru ff DS (2000) Polymorphic microsatellite DNA markers in the African elephant ( Loxondonta africana ) and their use in the Asian elephant ( Elephas maximus ). Molecular Ecology 9, 2223-2225. Evett IW, Weir BS (1998) Interpreting DNA Evidence: Statistical Genetics for Forensic Scientists Sinauer Associates, Inc ., Sunderland, MA, USA. Excoffier L, Heckel G (2006) Computer programs for populati on genetics data analysis: A survival guide. Nature Reviews Genetics 7, 745-758. Excoffier L, Laval G, Schneider S (2005a) ARLEQUIN ver 3.0: An integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1, 47-50.

PAGE 180

180 Excoffier L, Laval G, Schneider S (2005b) ARLEQUIN ver 3.1: User Manual Computational and Molecular Population Genetics Lab, Univ ersity of Berne, Bern, Switzerland. Felsenstein J (2004) PHYLIP (Phylogeny Inferen ce Package) version 3.6, Distributed by the author. Department of Genome Sciences, Univer sity of Washington, Seattle, WA, USA. Available from http://evolution.gs.w ashington.edu/phylip/n ewicktree.html. Fernandez S, Jones SC (1990) Manatee stranding on the coast of Texas. Texas Journal of Science 42, 103. Fernando PJ, Vidya TNC, Melnick DJ (2001) Is olation and characteri zation of triand tetranucleotide microsatellite loci in the Asian elephant, Elephas maximus Molecular Ecology Notes 1, 232-233. Fertl D, Schiro AJ, Regan GT, Beck CA, Ad imey N, Price-May L, Amos A, Worthy GAJ, Crossland R (2005) Manatee occurrence in the nort hern Gulf of Mexico, west of Florida. Gulf and Caribbean Research 17, 69-94. Frankham R, Ballou JD, Briscoe DA (2002) Introduction to Conservation Genetics Cambridge University Press, Cambridge, UK. FWC (2006) Final Biological Status Review of the Florida m anatee (Trichechus manatus latirostris) St. Petersburg, FL, USA. FWRI (2007) Manatee Synoptic Surveys St. Petersburg, FL, USA. Accessed 13 February 2007. Available from http://floridamarine.o rg/features/view_article.asp?id=15246. Garcia-Rodriguez AI (2000) Genetic Studies of th e West Indian Manatee (Trichechus manatus). Doctoral Dissertation, University of Florida, Gainesville, FL, USA. Garcia-Rodriguez AI, Bowen BW, Domning DP, Mignucci-Giannoni AA, Marmontel M, Montoya-Ospina RA, Morales-Vela B, R udin M, Bonde RK, McGuire PM (1998) Phylogeography of the West Indian manatee ( Trichechus manatus ): How many populations and how many taxa? Molecular Ecology 7, 1137-1149. Garcia-Rodriguez AI, Moraga-Amador D, Farmer ie W, McGuire P, Ki ng TL (2000) Isolation and characterization of microsatellite DNA markers in the Florida manatee ( Trichechus manatus latirostris ) and their application in se lected sirenian species. Molecular Ecology 9, 2161-2163. Garner A, Rachlow JL, Hicks JF (2005) Patterns of genetic diversity and its loss in mammalian populations. Conservation Biology 19, 1215-1221. Gerlach G, Derschum HS, Martin Y, Brinkma nn H (2000) Characteriza tion and isolation of DNA microsatellite primers in hyrax species ( Procavia johnstoni and Heterohyrax brucei, hyracoidea). Molecular Ecology 9, 1675-1677.

PAGE 181

181 Glaubitz JC (2004) CONVERT: A user-friendly program to refo rmat diploid genotypic data for commonly used population genetic software packages. Molecular Ecology Notes 4, 309310. Gillespie JH (2004) Population genetics: A concise guide, Second edn. Johns Hopkins University Press, Baltimore, MD, USA. Google (2007) Google Earth Mapping Service, 4.0.2737. Mountain View CA, USA. Accessed 29 June 2007. Available from http://earth.google.com/. Graham RT, Roberts CM (2007) Assessing the size, growth rate and structure of a seasonal population of whale sharks ( Rhincodon typus Smith 1828) using conventional tagging and photo identification. Fisheries Research 84, 71-80. Gray BA, Zori RT, McGuire PM, Bonde RK (2002) A first-generation cyt ogenetic ideogram for the Florida manatee ( Trichechus manatus latirostris ) based on multiple chromosome banding techniques. Hereditas 137, 215-223. Gunter G (1944) Texas manatees. Texas Game and Fish 2, 9-11. Hammond PS, Mizroch SA, Donovan GP (1990) I ndividual recognition of cetaceans: Use of photo-identification and other technique s to estimate population parameters. Report of the International Whaling Co mmission Special Issue, 12. Harper JY (2004) Corneal Vascularization in the Fl orida Manatee (Trichechus manatus latirostris). Doctoral Dissertation, University of Florida, Gaines ville, FL, USA. Hayes SA, Pearse DE, Costa DP, Harvey JT, Le Boeuf BJ, Garza JC (2006) Mating system and reproductive success in eastern Pacific harbour seals. Molecular Ecology 15, 3023-3034. Hedrick PW (2001) Conservation ge netics: Where are we now? Trends in Ecology and Evolution 16, 629-636. Hillis DM, Mable BK, Larson A, Davis SK, Zi mmer EA (1996) Nucleic Acids IV: Sequencing and Cloning. In: Molecular Systematics (eds. Hillis DM, Moritz C, Mable BK). Sinauer Associates, Inc., Sunderland, MA, USA. Houlden BA, England P, Sherwi n WB (1996) Paternity exclusi on in koalas using hypervariable microsatellites. The Journal of Heredity 87, 148-152. Ichikawa Y, Takagi K, Tsumagari S, Ishihama K, Morita M, Kanema ki M, Takeishi M, Takahashi H (2001) Canine parentage testi ng based on microsatelli te polymorphisms. Journal of Veterinary Medical Science 63, 1209-1213. Jones AG, Ardren WR (2003) Methods of pa rentage analysis in natural populations. Molecular Ecology 12, 2511-2523.

PAGE 182

182 Jorde PE, Schweder T, Bickham JW, Givens GH, Suydam R, Hunter D, Stenseth NC (2007) Detecting genetic structure in migrating bowhead whales off the coast of Barrow, Alaska. Molecular Ecology 16, 1993-2004. Kalinowski ST, Sawaya MA, Taper ML (2006a) I ndividual identificatio n and distribution of genotypic differences between individuals. Journal of Wildlife Management 70, 11481150. Kalinowski ST, Wagner AP, Taper ML (2006b) ML-RELATE: A computer program for maximum likelihood estimation of relatedness and relationship. Molecular Ecology Notes 6, 576-579. Keller LF, Grant PR, Grant BR, Petren K (2002) E nvironmental conditions affect the magnitude of inbreeding depression in surv ival of Darwin's finches. Evolution 56, 1229-1239. Kellogg ME, Burkett S, Dennis TR, Stone G, Gray BA, McGuire PM, Zori RT, Stanyon R (2007) Chromosome painting in the manat ee supports Afrotheria and Paenungulata. BMC Evolutionary Biology 7. King JM, Heinen JT (2004) An assessment of the behaviors of overwintering manatees as influenced by interactions with touris ts at two sites in central Florida. Biological Conservation 117, 227-234. King TL, Switzer JF, Morrison CL, Eackles MS, Young CC, Lubinski BA, Cryan P (2006) Comprehensive genetic analyses reveal evolutionary distin ction of a mouse ( Zapus hudsonius preblei ) proposed for delisting from the US Endangered Species Act. Molecular Ecology 15, 4331-4359. Lacy RC (1997) Importance of genetic variation to the viability of mammalian populations. Journal of Mammalogy 78, 320-335. LaHood ES, Moran P, Olsen J, Grant WS, Park LK (2002) Microsatellite a llele ladders in two species of Pacific salmon: Prepar ation and field-test results. Molecular Ecology Notes 2, 187-190. Laist DW, Reynolds III JE (2005a) Florida manat ees, warm-water refuges, and an uncertain future. Coastal Management 33, 279-295. Laist DW, Reynolds III JE (2005b) Influence of power plants and other warm-water refuges on Florida manatees. Marine Mammal Science 21, 739-764. Lande R (1988) Genetics and demogr aphy in biological conservation. Science 241, 1455-1460. Langtimm CA, Beck CA (2003) Lowe r survival probabilities for adul t Florida manatees in years with intense coastal storms. Ecological Applications 13, 257-268.

PAGE 183

183 Langtimm CA, Beck CA, Edwards HH, Fick-Child KJ, Ackerman BB, Barton SL, Hartley WC (2004) Survival estimates for Florida ma natees from the photo-identification of individuals. Marine Mammal Science 20, 438-463. Langtimm CA, O'Shea TJ, Pradel R, Beck CA (1998) Estimates of annual survival probabilities for adult Florida manatees ( Trichechus manatus latirostris ). Ecology 79, 981-997. MacFadden BJ, Higgins P, Clemen tz MT, Jones DS (2004) Diets, habitat preferen ces, and niche differentiation of Cenozoic sirenians from Fl orida: Evidence from stable isotopes. Paleobiology 30, 297-324. Marmontel M, Humphrey SR, O'Shea TJ (1997) P opulation viability analysis of the Florida manatee ( Trichechus manatus latirostris ), 1976-1991. Conservation Biology 11, 467-481. Marshall CD, Reep RL (1995) Manatee cerebral co rtex: Cytoarchitecture of the caudal region in Trichechus manatus latirostris Brain, Behavior and Evolution 45, 1-18. McClenaghan LR, O'Shea TJ (1988) Genetic variability in the Florida manatee ( Trichechus manatus ). Journal of Mammalogy 69, 481-488. Minch E (2007) MICROSAT, version 1.5d: A comput er program for calculati ng various statistics on microsatellite allele data. Available fr om http://hpgl.stanford.e du/projects/microsat/. Moore SS, Hale P, Byrne K (1998) NCAM: A po lymorphic microsatellite locus conserved across eutherian mammal species. Animal Genetics 29, 33-36. Moran P, Teel D, LaHood E, Drake J, Kali nowski S (2006) Standardising multi-laboratory microsatellite data in Pacific salmon: An historical view of the future. Ecology of Freshwater Fish 15, 597-605. Moritz C (1994) Defining Evolutionarily Significant Units for conservation. Trends in Ecology and Evolution 9, 373-375. Munson L, Terio KA, Worley M, Jago M, Bagot-S mith A, Marker L (2 005) Extrinsic factors significantly affect patterns of disease in free-ranging and captive cheetah ( Acinonyx jubatus ) populations. Journal of Wildlife Diseases 41, 542-548. Murphy WJ, Eizirik E, Johnson WE, Zhang YP Ryder OA, OBrien SJ (2001) Molecular phylogeny and the origins of placental mammals. Nature 409, 614-618. Nyakaana S, Arctander P (1998) Is olation and characterization of microsatellite loci in the African elephant, Loxodonta africana Molecular Ecology 7, 1436-1437. Nyakaana S, Okello JBA, Muwanika V, Siegismund HR (2005) Six new polymorphic microsatellite loci isolated and characterized from the Afri can savannah elephant genome. Molecular Ecology Notes 5, 223-225.

PAGE 184

184 OBrien SJ (1994) A role for molecular genetics in biological conservation. Proceedings of the National Academy of Sciences of the United States of America 91, 5748-5755. O'Shea TJ (1988) The past, present, and future of manatees in th e southeastern United States: Realities, misunderstandings, and enigmas. In:The Third Southeastern Nongame and Endangered Wildlife Symposium Social Circle, GA, USA. Georgia Department of Natural Resources, Game and Fish Division, Pp. 184-204. O'Shea TJ, Reep RL (1990) Encephalization quotient s and life-history traits in the Sirenia. Journal of Mammalogy 71, 534-543. Paetkau D, Strobeck C (1994) Mi crosatellite analysis of gene tic variation in black bear populations. Molecular Ecology 3, 489. Page RDM (1996) TREEVIEW: An application to display phylogenetic trees on personal computers. Computer Applications in the Biosciences 12, 357-358. Pause KC, Nourisson C, Clark A, Kellogg ME Bonde RK, McGuire PM (2007) Polymorphic microsatellite DNA markers for the Florida manatee ( Trichechus manatus latirostris ). Molecular Ecology Notes Online Early. Peakall R, Ebert D, Cunningham R, Lindenmayer D (2006) Mark-recapture by genetic tagging reveals restricted move ments by bush rats ( Rattus fuscipes ) in a fragmented landscape. Journal of Zoology 268, 207-216. Pearman PB, Garner TWJ (2005) Susceptibility of Italian agile frog popula tions to an emerging strain of Ranavirus parallels population genetic diversity. Ecology Letters 8, 401-408. Pemberton JM (2004) Measuring in breeding depression in the wild: The old ways are the best. Trends in Ecology and Evolution 19, 613-615. Pennycuick CJ (1978) Identificati on using natural markings. In: Animal marking Recognition Marking in Animals in Research (ed. Stonehouse B), pp. 147-159. Macmillan Press Ltd, London, UK. Pettis HM, Rolland RM, Hamilton PK, Brault S, K nowlton AR, Kraus SD (2004) Visual health assessment of North Atlantic right whales ( Eubalaena glacialis ) using photographs. Canadian Journal of Zoology 82, 8-19. Primmer CR, Moller AP, Ellegren H (1996) A wide-ra nge survey of cross-species microsatellite amplification in birds. Molecular Ecology 5, 365-378. Pritchard JK, Stephens M, Donnelly P (2000) In ference of population structure using multilocus genotype data. Genetics 155, 945-959. Pritchard JK, Wen W (2004) Documentation for STRUCTURE software: Version 2 Department of Human Genetics, University of Chicago, Chicago, IL, USA.

PAGE 185

185 Proebstel DS, Evans RP, Shiozawa DK, Williams RN (1993) Preservation of nonfrozen tissue samples from a salmonine fish Brachymystax lenok (Pallas) for DNA analysis. Journal of Ichthyology 9, 9-17. Purves D, Augustine GJ, Fitzpatrick D, Katz LC, LaMantia A-S, McNamara JO, Williams SM (2001) Neuroscience Second edn Sinauer Associates, Inc ., Sunderland, MA, USA. Quattro JM, Vrijenhoek RC (1989) Fitness differences among remnant populations of the endangered Sonoran topminnow. Science 245, 976-978. Quinn GP, Keough MJ (2002) Experimental Design and Data Analysis for Biologists Cambridge University Press, Cambridge, UK. Rathbun GB, Reid JP, Bonde RK, Powell JA (1 995) Reproduction in free-ranging Florida manatees. In: Population Biology of the Florida Manatee (eds. OShea TJ, Ackerman BB, Percival HF), pp. 135-156. National Bi ological Service Info rmation and Technology Report 1. Washington, D. C. Raymond M, Rousset F (1995) GENEPOP (version 1.2): Population ge netics software for exact tests and ecumenicism. Journal of Heredity 86, 248-249. Reed DH and Frankham R (2001) How closely correlated are molecular and quantitative measures of genetic variation? A meta-analysis. Evolution 55, 1095-1103. Reep RL, Bonde RK (2006) The Florida Manatee: Biology and Conservation University Press of Florida, Gainesville, FL, USA. Reep RL, Johnson JI, Switzer RC, Welker WI (1989) Manatee cerebral cort ex: Cytoarchitecture of the frontal region in Trichechus manatus latirostris Brain, Behavior and Evolution 34, 365-386. Reep RL, Marshall CD, Stoll ML (2002) Tactile hairs on th e postcranial body in Florida manatees: A mammalian lateral line? Brain, Behavior and Evolution 59, 141-154. Reeves RR, Mitchell E (1989) Status of white whales, Delphinapterus leucas in Ungava Bay and Eastern Hudson Bay. Canadian Field-Naturalist 103, 220-239. Reid JP (2000) Florida manatee now resident in the Bahamas. SireNews (Newsletter of the IUCN/SSC Sirenia Specialist Group) 33, 7-8. Reid JP, Rathbun GB, Wilcox JR (1 991) Distribution patterns of individually identifiable West Indian manatees ( Trichechus manatus ) in Florida. Marine Mammal Science 7, 180-190. Reynolds III JE, Odell DK (1991) Manatees and Dugongs Facts on File, Inc., New York, NY, USA. Reynolds III JE, Rommel SA, Pitchford ME (2 004) The likelihood of sperm competition in manatees Explaining an apparent paradox. Marine Mammal Science 20, 464-476.

PAGE 186

186 Rice WR (1989) Analyzing tables of statistical tests. Evolution 43, 223-225. Richard KR, Whitehead H, Wright JM (1996) Polymorphic micros atellites from sperm whales and their use in the genetic identification of individuals from natura lly sloughed pieces of skin. Molecular Ecology 5, 313-315. Robeck TR, Curry BE, McBain JF, Kraemer DC (1994) Reproductive-biology of the bottlenosed dolphin ( Tursiops truncatus ) and the potential applicati on of advanced reproductive technologies. Journal of Zoo and Wildlife Medicine 25, 321-336. Roelke ME, Martenson JS, OBrie n SJ (1993) The consequences of demographic reduction and genetic depletion in the e ndangered Florida panther. Current Biology 3, 340-350. Rowold DJ, Herrera RJ (2005) On hum an STR sub-populatio n structure. Forensic Science International 151, 59-69. Rozen S, Skaletsky HJ (2000) PRIMER3 on the WWW for general users and for biologist programmers. In: Bioinformatics Methods and Protocol s: Methods in Molecular Biology (eds. Krawetz S, Misener S), pp. 365-386. Humana Press, Totowa, NJ, USA. Sambrook J, Fritsch E, Maniatis T (1989) Molecular Cloning: A Laboratory Manual, Second Edn Cold Spring Harbor Laboratory Pre ss, Cold Spring Harbor, NY, USA. Schiro A, Fertl D (1995) Mermaids sighted in Galveston Bay. Soundings (Newsletter of the Galveston Bay Foundation) 7, 4-5. Schnabel RD, Ward TJ, Derr JN (2000) Validation of 15 microsatellites fo r parentage testing in North American bison, Bison bison and domestic cattle. Animal Genetics 31, 360-366. Selkoe KA, Toonen RJ (2006) Micr osatellites for ecologists: A practical guide to using and evaluating microsatellite markers. Ecology Letters 9, 615-629. Serre D, Pbo S (2004) Evidence for gradients of human genetic diversity within and among continents. Genome Research 14, 1679-1685. Slate J, David P, Dodds KG, Veenvliet BA Glass BC, Broad TE, McEwan JC (2004) Understanding the relationship between th e inbreeding coefficient and multilocus heterozygosity: Theoretical expe ctations and empirical data. Heredity 93, 255-265. Slatkin M (1995) A measure of popula tion subdivision based on microsat ellite allele frequencies. Genetics 139, 457-462. Smith TG, Hammill MO (1986) Populat ion estimates of white whale, Delphinapterus leucas in James Bay, eastern Hudson Bay, and Ungava Bay. Canadian Journal of Fisheries and Aquatic Science 43, 1982-1987.

PAGE 187

187 Sorice MG, Shafer CS, Ditton RB (2006) Mana ging endangered spec ies within the usepreservation paradox: The Florida manatee ( Trichechus manatus latirostris ) as a tourism attraction. Environmental Management 37, 69-83. Strachan T, Read AP (1999) Human Molecular Genetics, Second Edn BIOS Scientific Publishers, Ltd., Oxford, UK. Sutovsky P, Moreno RD, Ramalho-Santos J, Domi nko T, Simerly C, Shatte n G (1999) Ubiquitin tag for sperm mitochondria. Nature 402, 371-372. SYSTAT Software, Inc. (2007) SYSTAT, ver. 12 help documentation San Jose, CA, USA. Available from http://www.systat.com/about/. Taberlet P, Luikart G (1999) N on-invasive genetic sampling and individual iden tification. Biological Journal of the Linnean Society 68, 41-55. Takezaki N, Nei M (1996) Genetic distances and reconstruction of phylogenetic trees from microsatellite DNA. Genetics 144, 389-399. Taylor B, Dizon A (1999) First policy then sc ience: Why a management unit based solely on genetic criteria cannot work. Molecular Ecology 8, S11-16. USFWS (2001) Florida Manatee Recovery Plan (Trichec hus manatus latirostris), Third Revision United States Fish and Wildlife Service, Atlanta, GA, USA. USFWS (2007) West Indian Manatee (Trichechus m anatus) 5-Year Review: Summary and Evaluation United States Fish and Wildlife Service, Atlanta, GA, USA. USGS (2001) Chessie the manatee on a comeba ck tour after 5-year hiatus Accessed 20 November 2006. Available from http://www.usgs.gov/newsroom/article_pf.asp?ID=443. Van Oosterhout C, Hutchinson WF Wills DPM, Shipley P (2004) MICRO-CHECKER: Software for identifying and correcting genotypi ng errors in microsatellite data. Molecular Ecology Notes 4, 535-538. Vianna JA, Bonde RK, Caballero S, Giraldo JP, Lima RP, Clark A, Marmontel M, Morales-Vela B, De Souza MJ, Parr L, Rodriguez-L opez MA, Mignucci-Giannoni AA, Powell JA, Santos FR (2006) Phylogeogr aphy, phylogeny and hybridization in trichechid sirenians: Implications for manatee conservation. Molecular Ecology 15, 433-447. Wagner AP, Creel S, Kalinowski ST (2006) Es timating relatedness and relationships using microsatellite loci with null alleles. Heredity 97, 336-345. Waits LP, Luikart G, Taberlet P (2001) Estim ating the probability of identity among genotypes in natural populations: Ca utions and guidelines. Molecular Ecology 10, 249-256.

PAGE 188

188 Walsh SJ, Buckleton J (2005) DNA Intelligence Databases. In: Forensic DNA Evidence Interpretation (eds. Buckleton J, Triggs CM, Wa lsh SJ), pp. 439-469. CRC Press, Boca Raton, FL, USA. Waples RS (1998) Separating the wheat from the ch aff: Patterns of genetic differentiation in high gene flow species. Journal of Heredity 89, 438-450. Waples RS, Gaggiotti O (2006) What is a populatio n? An empirical evalua tion of some genetic methods for identifying the number of gene pools and their degr ee of connectivity. Molecular Ecology 15, 1419. Wayne RK, Lehman N, Girman D, Gogan PJP, Gilbert DA, Hansen K, Peterson RO, Seal US, Eisenhawer A, Mech LD, Krumenaker RJ (1991) Conservation genetics of the endangered Isle Royale gray wolf. Conservation Biology 5, 41-51. Weber DS, Stewart BS, Schienman J, Lehman N (2004) Major histoc ompatibility complex variation at three class II loci in the northern elephant seal. Molecular Ecology 13, 711718. Webster MT, Smith NGC, Ellegren H (2002) Mi crosatellite evolution inferred from humanchimpanzee genomic sequence alignments. Proceedings of the National Academy of Sciences of the United States of America 99, 8748-8753. Weigle BL, Wright IE, Ross M, Flamm R (2001) Movements of radio-tagged manatees in Tampa Bay and along Florida's west coast 1991-1996. Florida Marine Research Institute Technical Report 7, 1-156. Weir BS, Cockerham CC (1984) Estimating F -statistics for the analysis of population structure. Evolution 38, 1358-1370. White JR, Harkness DR, Isaacks RE, Duffield DA (1976) Some studies on blood of the Florida manatee, Trichechus manatus latirostris Comparative Biochemi stry and Physiology A 55, 413-417. White JR, Harkness DR, Isaacks RE, Duffield DA (1977) Trichechus manatus latirostris (manatee). Order: Sirenia. Family Trichechidae. In: An Atlas of Mammalian Chromosomes (eds. Hsu TC, Benirschke K), pp. 496. Springer-Verlag, New York, NY, USA. White PS, Densmore LD (1992) Mitochondrial DNA isolation. In: Molecular Genetic Analysis of Populations: A Practical Approach (ed. Hoezel AR), pp. 29-58. Oxford University Press, New York, NY, USA. Wilberg MJ, Dreher BP (2004) GENECAP: A program for analysis of multilocus genotype data for non-invasive sampling and capture -recapture population estimation. Molecular Ecology Notes 4, 783-785.

PAGE 189

189 Williams PB (2006) Managing freshwater infl ow to the San Francisco Bay Estuary. Regulated Rivers: Research and Management 4, 285-298. Woods JG, Paetkau D, Lewis D, McLellan BN, Proctor M, Strobeck C (1999) Genetic tagging of free-ranging black and brown bears. Wildlife Society Bulletin 27, 616-627. Wright JM, Bentzen P (1994) Microsatelli tes: Genetic markers for the future. Reviews in Fish Biology and Fisheries 4, 384-388. Wright S (1951) The genetical structure of populations. Annals of Eugenics 15, 323-354. Wright SD, Ackerman BB, Bonde RK, Beck CA, Banowetz DJ ( 1995) Analysis of watercraftrelated mortality of manatees in Florida, 1979-1991. In: Population biology of the Florida manatee (eds. OShea TJ, Ackerman BB, Percival HF), pp. 259-268. National Biological Service Information and Technology Report 1. Washington, D. C. Yeh FC, Boyle T (1997) Population genetic analysis of co-domin ant and dominant markers and quantitative traits. Belgian Journal of Botany 129, 157.

PAGE 190

190 BIOGRAPHICAL SKETCH Kimberly Christina Pause has always been fa scinated by marine mammals, and dreamed of becoming a marine biologist. She majored in Bi ological Sciences at Florida State University, graduating in 2003. While at Fl orida State, she worked as a laboratory technician for Dr. Michael Blaber from 2001 to 2002. There she learned the basics of laboratory work and good habits. She also volunteered and did an i ndependent study with Dr. Trisha Spears and Dr. Laurence Abele studying the molecular phylogene tics of branchiopods in Northern Floridas ephemeral ponds from 2002-2003. Ms. Pause also enjoys teaching, and had the opportunity to be a teaching assistant for undergraduate biology at Florida State for two seme sters. Additionally, sh e participated in the Saturday at the Sea program, taking middle school students to the FSU Marine Lab to study the estuarine ecosystem. After college, Ms. Pause worked as a Park In terpreter/Naturalist for Island Beach State Park, Seaside Park, NJ. She led several different daily tours for the public including kayak tours and seining tours, as well as answering beac hgoers questions at the Nature Center. While at the University of Fl orida, Ms. Pause has taken ever y opportunity she could to be out in the field, attending as many marine mammal necropsies she could, assisting her lab mate Maggie Kellogg with manatee tissu e sample collections, and taggi ng along with Bob Bonde to Crystal River. She attended workshops, includ ing SEAVET I and Forensic Science for Marine Biologists, to help broade n her educational background.