This item has the following downloads:
1 BIO ACCUMULATION OF METALS IN AN INSULAR POPULATION OF FLORIDA COTTONMOUTH SNAKES ( AGKISTRODON PISCIVORUS CONANTI ) By JOEL GEORGE WIXSON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIRE MENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013
2 2013 Joel George Wixson
3 This thesi s is dedicated to my family: George and Jeannie Wixson Mary Wixson Heather McDowell, and Angelica Garcia Thank you for everything
4 ACKNOWLEDGMENTS There are many people I would like to thank for their help in my graduate education. First I would like to thank my advisor Harvey Lillywhite for his help in improving my writing, for taking me on numerous research trips, providing guidance on this project, and for giving me a chance to study at the U niversity of F lorida I am thank ful to my committee member s Max Nickerson and Darryl Heard, for their comments and editorial h elp with this thesis, help with the collecting of samples, and guidance. I would like to thank Jake Ferguson for his help with statistics. I am grateful to the numerous people who help ed me develop my ideas and collect s amples I thank my stu dents for constantly reminding me of the reasons why I love science. Their interest s ideas, and patienc e helped me in many ways I am grateful for the opportunity to watch them grow intellectually. I thank Rolando Quesada for his help with the s urgical techniques used in this study and R yan McCleary for collecting many of the samples used in this study I thank Fred Thompson for help identifying mulluscs. I thank the University of Florida for support via a Teaching Assistantship, and for the use of e quipment while in Gainesville, and on Seahorse Key. I thank the Seahorse Key Marine Laboratory for the use of facilities and boat transportation that assisted my collecting of organisms and data. I thank an anonymous private donor for funding. I am inde bted to my parents George and Jeannie Wixson for their love and constant support, even when my ideas were a little crazy I am also grateful to my friend s and family for their support and guidance Lastly, I would like to thank Angelica Garcia whose lov e and support guided me through the completion of this project, graci a s bonita te amo.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 SNAKES: SENTINELS FOR ENVIRONMENTAL CONTAMINATION? .................. 12 Introduction ................................ ................................ ................................ ............. 12 Review Methods ................................ ................................ ................................ ..... 18 TICs in Snakes ................................ ................................ ................................ ....... 18 Tissue Concent rations ................................ ................................ ...................... 19 Laboratory Manipulation ................................ ................................ ................... 21 Effects of Contaminants ................................ ................................ ................... 22 Snakes as Model Study Organisms ................................ ................................ ........ 26 Exposure to Pollutants ................................ ................................ ..................... 27 Moveme nt of Contaminants Within and Between Trophic Levels ..................... 27 Abundance and Ease of Sampling ................................ ................................ ... 28 Home Range and Habitat ................................ ................................ ................. 29 Conclu sions and Future Perspective ................................ ................................ ...... 29 Population Variables ................................ ................................ ........................ 30 Non lethal Sampling ................................ ................................ ......................... 31 Snakes as Model Organisms ................................ ................................ ............ 32 2 BIO ACCUMULATION OF METALS IN AN INSULAR POPULATION OF FLORIDA COTTONMOUTH SNAKES ( AGKISTRODON PISCIVORUS CONANTI ) ................................ ................................ ................................ .............. 40 Introduction ................................ ................................ ................................ ............. 40 Methods and Materials ................................ ................................ ............................ 43 Animal Collection ................................ ................................ .............................. 43 Blood samp les ................................ ................................ ................................ .. 44 Liver samples ................................ ................................ ................................ ... 45 Tissue Samples from Fish ................................ ................................ ................ 46 Sample Preparation and Analysis ................................ ................................ .... 47 Statistical Analysis ................................ ................................ ............................ 47 Conversion of Wet Mass to Dry Mass ................................ .............................. 49 Results ................................ ................................ ................................ .................... 49 Discussion ................................ ................................ ................................ .............. 52 Metal Summary and Comparison to Other Organisms ................................ ..... 54
6 Arsenic ................................ ................................ ................................ ....... 55 Cadmium ................................ ................................ ................................ .... 59 Selenium ................................ ................................ ................................ .... 61 Mercury ................................ ................................ ................................ ...... 63 Size Differences between Mainland and Insular Cottonmouths ....................... 67 Observed Effects of TICs on snakes living at Seahorse Key ............................ 68 Fish sampling and bio concentration ................................ ................................ 70 Future research ................................ ................................ ................................ 70 Conclusion ................................ ................................ ................................ .............. 72 3 METAL CONCENTRAT IONS IN MARINE AND TERRESTRIAL MOLLUSCS FROM SEAHORSE KEY, FLORIDA ................................ ................................ ..... 103 Introduction ................................ ................................ ................................ ........... 103 Methods and Materials ................................ ................................ .......................... 105 Sample Collection ................................ ................................ .......................... 105 Sample Pre paration ................................ ................................ ........................ 107 Analysis of Samples ................................ ................................ ....................... 108 Data from Mussel Watch ................................ ................................ ................ 108 Statistical Analysis ................................ ................................ .......................... 109 Wet Mass and Dry Mass Conversions ................................ ........................... 110 Results ................................ ................................ ................................ .................. 110 Discussion ................................ ................................ ................................ ............ 112 Arsenic in Florida Waters ................................ ................................ ............... 115 Metal Contaminants and the Cedar Key Community ................................ ...... 116 Conclusion ................................ ................................ ................................ ...... 118 LIST OF REFERENCES ................................ ................................ ............................. 130 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 140
7 LIST OF TABLES Table page 1 1 Metal concentrations from the most recent studies regarding snakes, 2011 2013.. ................................ ................................ ................................ ................. 34 2 1 Tissue concentrations of 29 metals for the Florida Cottonmouth Snake ( Agkistrodon piscivorus conanti ) from four locations in Florida ........................... 73 2 2 Hepatic tissue concentrations for four species of fish that were opportunistically collected under bird rookeries on Seahors e Key, FL. .............. 79 2 3 Compilation of representative toxicology studies from around the world that measured arsenic levels in tissues of vertebrates. ................................ ............. 81 2 4 Compilation of representative toxicology studies from around the world that measured cadmium lev els in tissues of vertebrates. ................................ .......... 83 2 5 Compilation of representative toxicology studies from around the world that measured selenium lev els in tissues of vertebrates ................................ ........... 85 2 6 Compilation of representative toxicology studies from around the world that measured mercury levels in tissues of vertebrates. ................................ ............ 87 3 1 Summary of the results of tissue analysis for 29 trace elements in three species of molluscs. ................................ ................................ ........................ 121
8 LIST OF FIGURES Figure page 1 1 A simplified food chain illustrating how a contaminant (+) is bioaccumulated, bi oconcentrated, and biomagnified ................................ ................................ ..... 37 1 2 The number of snake Toxic Inorganic Contaminant (TIC) publications published over the l ast 63 years.. ................................ ................................ ....... 38 1 3 Lead tissue concentration (ppm) as a function of tissue type.. ........................... 38 1 4 Schematic showing possible exposure routes and sour ce of contaminants for snakes. ................................ ................................ ................................ ............... 39 2 1 Simplifie d food chain for Seahorse Key. ................................ ............................. 89 2 2 Collection sites on Seahorse Key for the Florida Cottonmouth Snake ( Agkistrodon piscivorus conanti ) and fi sh during the years 2008 2010. ............. 90 2 3 Location of Florida Cottonmouth Snake ( Agkistrodon piscivorus conanti ) collec tion sites throughout Florida. ................................ ................................ ..... 91 2 4 Snout vent length and total length as a function of mass for all Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled in this study. ... 92 2 5 Snout vent length (SVL) as a function of mass for all Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) at each sampling location. .................. 92 2 6 Arsenic levels in two tissue types in Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) from four locations in Florida,. ......................... 93 2 7 Cadmium levels in two tissue types in Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) from four locations in Florida ........................... 94 2 8 Selenium levels in two tissue types in Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) from four locations in Florida. .......................... 95 2 9 Mercury levels in two tissue types in Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) from four locations in Florida. .......................... 96 2 10 Arsenic tissue concentration as a function of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus cona nti ) sampled on Seahorse Key ................................ ................................ ................................ ..................... 97 2 11 Cadmium tissue concentration as a function of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled on Seahorse Key ................................ ................................ ................................ .................... 97
9 2 12 Selenium tissue concentration as a function of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled on Seahorse Key.. ................................ ................................ ................................ ................... 98 2 13 Mercury tissue concentration as a function of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled on Seahorse Key. ................................ ................................ ................................ .................... 98 2 14 F ish length as a function of fish mass for all regurgitated fish and dropped fish. ................................ ................................ ................................ ..................... 99 2 15 Tissue concentration of arsenic (As), cadmium (Cd), mercury (Hg), and selenium (Se) in liver samples taken from fish regurgitated by nesting waterbirds on Seahorse Key.. ................................ ................................ ........... 100 2 16 Mercury and selenium concentrations for the Florida Cottonmouth Snake ( Agkistrodon piscivorus conanti ) from four locations in Florida. ........................ 101 2 17 Correlation between combined mercury and selenium concentrations for the Florida Cottonmouth Snake ( Agkistrodon piscivorus conanti ) from four locations in Florida and two tissue types. ................................ ......................... 102 3 1 Location of sampling sites: Black Point and Seahorse Key. ............................ 124 3 2 Average tissue concentration for the Marsh Periwinkle Snail ( Littoraria irrorata ) and the Southern Flatcoil Snail ( Polygyra cereolus ) ........................... 125 3 3 Average tissue concentration for the Northern Quahog Clam ( Mercenaria mercenaria ) ................................ ................................ ................................ ...... 126 3 4 Mussel Watch data downloaded from the CCMA website. Data points are average concentration of samples of the Eastern Oyster ( Crassostrea virginica ). ................................ ................................ ................................ .......... 127 3 5 Mussel Watch data downloaded from the CCMA website. Data points are single samples of sediment. ................................ ................................ ............. 128 3 6 Mercury concentration as a function of selenium concentration in three mollusc species collect ed on and around Seahorse Key. ................................ 129
10 Abstract o f Thesis Pr esented to the Graduate School of the University o f Florida i n Partial Fulfillme nt of the Requirements for the Degree of Master o f Science BIO ACCUMULATION OF METALS IN AN INSULAR POPULATION OF FLORIDA COTTONMOUTH SNAKES ( AGKISTRO DON PISCIVORUS CONANTI ) By Joel George Wixson August 2013 Chair: Harvey Lillywhite Major: Zoology Understanding coastal e nvironmental pollution is a The overall objective of the research reported here is to add to a growing body of knowledge regarding metal contamination in coastal waters. The system outlined here is possibly unique in that it involves insular snakes and their trophic link with nestin g water birds via allochthonous resources from the Gulf of Mexico. In Chapter 1 I review the toxicology literature that is relevant for the insular system investigated in this study In Chapter 2 I focus on the ecotoxicology of an insular population of F lorida C ottonmouth S nakes ( Agkistrodon piscivorus conanti ) that scavenge fish carrion dropped by colonial water birds nesting on Sea horse Key, Levy County, Florida In Chapter 3 I focus on the ecotoxicology of molluscs inhabiting Seahorse Key and the surrounding Gulf of Mexico. Levels of 29 contaminant metals were measured in snakes, fish, and molluscs collected from Seahorse Key. Tissue concentrations of metals measured in i nsular cottonmouths were compared to those in mainland populations Statistically significant differences between populations were found for arsenic (As). Mercury (Hg) and selenium (Se) correlation in the cottonmouths suggests a trend towards high levels of mercury compared to
11 selenium ; this correlation is the reversal of what is seen in other vertebrate species. Concentrations of metals in molluscs, discussed in Chapter 3, further support the hypothesis that the likely source of contaminants for cottonmo uths inhabiting Seahorse Key is the Gulf of Mexico. Implications for the aquaculture industry around Seahorse Key are also discussed in Chapter 3.
12 CHAPTER 1 SNAKES: SENTINELS FOR ENVIRONMENTAL CONTAMINATION? Introduction Toxicology is the branch of science that seeks to understand the adverse effects of poisons and toxins on biological systems (Zakrzewski 1997, Crosby 1998) Toxic means poisonous. The American He r itage Dictionary ( 1982) Organisms on earth are surrounded by toxins and poisons. The terms poison and toxin are sometimes used interchangeably and there seems to be much debate regarding their definition (Zakrzewski 1997, Crosby 1998, Fowler et al. 2007) For the purpose of this review I will define toxins as a poisonous substance produced within living cells or o rganisms (Heritage 1982) Every day animals and plants wage biological warfare with each other in an endless arms race to develop the next best toxin and resistance to that toxin Humans have been affected by poisons and have used poisons for centuries usually in the form of toxins Until relatively recently, humans have only been concerned with the short term effects of these poisons For example, a poison tipped dart will do a pretty good job of killing dinner, but whether the chemicals in the poiso n will cause cancer later in life is of little concern to the hunter. This mindset has changed dramatically in the last hundred years. Since the industrial revolution humans have created a vast number of new compounds. These new compounds have been int entionally or accidentally released into the environment (Zakrzewski 1997, Crosby 1998) The consequences of these actions are only beginning to be understood. In the process of creating these new compounds various others have also been released into th e environment. These
13 include organochlorides, radionucleotides, metals, and other compounds used for economic gain (Zakrzewski 1997) Humans have realized the use of poisons can have beneficial and negative effects. This is evident in the countless tons of pesticides, herbicides, fertilizers, and heavy metals released into the environment every year, and the corresponding legislation for the control and use of these chemicals (Zakrzewski 1997, Crosby 1998, Fowler et al. 2007) Research on the widespread release of these contaminants has indicated dire short and long term consequences. Humans are exposed to these compounds through the food we eat, the water we drink, the products we use and the air we breathe (Zakrzewski 1997) Pick almost any middle ag ed person living in a modern society and you will most likely find elevated levels of organochlorides, heavy metals, pesticides, and countless other chemicals. The effect of such widespread contamination from industrialized nations is only beginning to be understood for humans and for the environment. What makes a compound poisonous? The sixteenth century German alchemist scientist Theophrastus, also called Paracelsus, defines poison? All things are poison, and nothing is without poison. Only the dose makes the p (Crosby 1998) The same can also be said of a pollutant or contaminant, the dose in this case being reflected in the amo unt of a particular contaminant in living tissue that is acquired from the environment. These compounds can be synthetic man made compounds such as numerous insecticides, pestic ides, and herbicides that are intentionally released into the environment or trace compounds such as heavy metals, metals, and metalloids which are naturally found in the environment, but usually in lower
14 levels than at a contaminated site (heavy metals at mining sites, shooting ranges, etc.) (Zakrzewski 1997) If the dose is impo rtant, how do we know a particular dose will affect a population? Finding the answer requires contaminant studies be focused on specific environments or types of contamination, as well as the dose. To determine these details the science of toxicology is split into many sub disciplines. This review is concerned with two sub disciplines: environmental toxicology and ecotoxicology. Environmental t detection of toxic substances in the environm ent and in any environmentally exposed (Crosby 1998) the release of toxic pollutants into the environment, their distribution and fate in the biosphere and especially in food chains, a nd qualitative and quantitative measurement of toxic responses in ecosystems and ecosystem (Crosby 1998) These two sub disciplines are very similar, but subtly dif ferent. The key difference is e cotoxicology is concerned with food chains and the movement of contamina n ts between and within trophic systems and between and wit hin populations. Environmental t oxicology is concerned with detecting, observing, and determining the effect of these poisons on the environment and organisms. Many toxico logy studies focus on heavy metals. The term heavy metals has been used by many researchers to refer to a group of metals or metalloids that are contaminants or considered toxic (Duffus 2002) T his classification is confusing. A lthough some heavy metals are toxic there are some that are used in metabolic processes and in low concentration s are not toxic but essential and other metals which
15 but are extremely toxic In addition, there is no consensus among specific gravity of the substance is important although no one can agree on what Bjerrum 's Inorganic Chemistry ( 1936) element s with a specific gravity of more than 7g/cm 3 (Cornelis and Nordberg 2007) Other researchers have suggested metals with a minimum specific gravity of 3.5 to 5 g/cm 3 suggested the atomic number of an element is a good definition of what make s a metal there is no consensus on which atomic number (Cornelis and Nordberg 2007) For this review I define metals with a specific gravity above 5 g/cm 3 to (Jrup 2003) Thu misleading and disapprove of its use (Crosby 1998, Duffus 2002, Cornelis and Nordberg 2007, Appenroth 2010) A more appropriate term for the group of compounds discussed in this review is Toxic Inorganic Chemicals or Compounds (TIC s ) (Crosby 1998) TICs include the heavy metals (defined in this review as elements that have a specific density of at least 5g/cm 3 Jrup 2003 ), the metalloids, the transition metals, and certain nonmetallic elements (halogens, phosphorus, and sulfur) (Crosby 1998) All of these compounds are potentially toxic to organisms in the right doses. TICs remain in the environment for long periods of time where they have the potential to bioconcentrate, bioaccumul ate, and biomagnify in living tissue. There is much confusion about the definition of bioconcentration and bioaccumulation. Their definitions depend on the organism studied (i.e. either an aquatic or terrestrial
16 organism). For this review bioconce ntrati on will be considered the concentration of a contaminant in living tissue, usually not by dietary uptake (Crosby 1998) Bioaccumulation will be defined as the accumulation of contaminants over time in living tissue through ingestion, absorption, and respi ration (Crosby 1998) A classic example of bioconcentration is a fish accumulating an aqueous solute through its gills or dermis and c oncentrating the solute in its tissue (Figure 1 1 ). For some organisms ( reptiles) concentrating contaminants in this way is unlikely (Grillitsch and Schiesari 2010) The time interval for tissue concentrations to reach equilibrium with the surrounding environment through bioconcentration depends o n the compound. For some contaminants, such as DDT, it can take months and f or others it can happen muc h more quickly. Using the example of the fish mentioned above the term bioaccumulation would refer to the chemicals the fish concentrates, plus chemicals absorbed through ingestion of prey items (Figure 1 1 ). Snakes are relatively misunderstood and traditionally under represented in the toxicological literature (Burger 1992, Hopkins 2000a, Campbell and Campbell 2001, Grillitsch and Schiesari 2010) Although, with the publication of two review papers in the early part of the century (Campbell and Campbell 2001, 2002) and the publication of Ecotoxicology of Amphibians and Reptiles (Sparling et al. 2000, 2010a) and Toxicology of Reptiles (Gardner and Oberdrster 2006) toxicology research on snakes has increased dramatical ly over the last decade. Figure 1 2 demonstrates the relatively recent rise in TIC publications involving snakes The increased use of snakes in toxicology research is important because until the publication of Campbell and Campbell ( 2001 ), they had been largely ignored in the toxicology literature. Reasons for
17 their prolonged absence in the toxicology literature are largely due to specific life history characteristics that some researchers may not find ap pealing in a toxicology study. Snakes belong to t he order Squamata which contains more than 9,200 species (as of February 2013 Reptile Database ). Snakes (suborder Ser pentes) contain approximately 24 families with 3,432 species recognized (as of February 2013, Reptil e Database) The majority of species of snakes (approximately 52 percent) belong to the family Colubridae and c onsequently the majority of toxicological studies involve snakes from this family (Grillitsch and Schiesari 2010) Colubrid s nakes tend to be non venomous, ubiquitous, and easily kept in a laboratory environment. When compared to other vertebrate groups very little is known about the effects of pollutants on snakes. It has been suggested for decades snakes would be ideal can didates for indicators of environmental contamination (Bauerle et al. 1975, Stafford et al. 1976) but research into their toxicology has lagg ed behind that of other vertebrates This is unfortunate because snakes are key components of most ecosystems an d are upper trophic level carnivores and scavengers that keep populations of small rodents and insects in check (Pough et al. 1998) Snakes also provide food for predatory birds and mammals. Thus snakes should be considered for inclusion in any environme ntal assessment or impact statement. The purpose of this review is to familiarize the reader with the toxicolog ical literature for both snakes and TICs, discuss the importance of including snakes in ecotoxicology research, and suggest future research direc tions Organic contamina n ts and radiation/radionuclides will not be discussed For more information regarding these contaminants refer to Campbell and Campbell ( 2001 ) and Wood et al. ( 2010 )
18 Review Methods A comprehensive literature search was conducted to find available toxicology literature on snakes. The primary database used to find scientific literature was the ISI Web of Knowledge, specifically the Web of Science available through the University of Fl orida libraries. Search engines such as Google Scholar were also utilized. The following key terms were used to find literature: snake, Serpentes, TIC, heavy metals, contamination, Squamata, toxicology, ecotoxicology, cottonmouth, individual metals (i.e. lead, mercury, manganese, etc), reptiles, Reptilia, and pollution. The key terms were used in every combination, and every year between 1950 and 201 3 was searched individually. With every new paper that was found, reference lists were scanned for potent ial literature that was not found while searching the databases alone. Recently, two comprehensive reptile toxicology books have been published: Ecotoxicology of Reptiles and Amphibians (Sparling et al. 2010a) and Toxicology of Reptiles (Gardner and Oberd rster 2006) Literature cited from these books was also utilized in this review as long as it was relevant to snakes and TICs. TICs in Snakes Research in toxicology using snakes can be divided into tissue concentration, laboratory manipulation, and con taminant effects. Research on tissue concentration involves investigators capturing snakes tak ing a tissue sample and report ing the concentration of specific contaminant s Laboratory manipulation involves investigators capturing snakes, bringing them b ack to the laboratory, and exposing them to experimental conc entrations of a contaminant. These approaches to toxicology research can indicate tolerance levels in different species for a particular TIC. S tudies on the effects of contaminants are conducte d to determine whether a contaminant alters
19 morphology, physiology, or ability to survive in the wild. Effects can vary from altered physical appearance (i.e. tumors, body deformities, etc ) or metabolism to change s in behavior and act ivity Tissue Concentrations Reports of metal contaminant concentrations in different tissue s are the most numerous (38/43) of toxicolog ical studies involving snakes Two comprehensive reviews, Grillitsch and Schiesari ( 2010) and Campbell and Campbell ( 2001) summarize the result s from all snake publications up until t he point of their publication. Thi s review is meant to supplement Grillitsch and Schiesari ( 2010) and Campbell and Campbell ( 2001) by summarizing the results from the most recent publicatio ns on metal contaminants involving snakes (Table 1 1) Since the publication of Grillitsch and Schiesari ( 2010) three publications involving metal con taminants of snakes have been published These are summarized and discussed in detail below (Table 1 1) TICs accumulate in each tissue differently (Ohlendorf et al. 1988, Burger 1992, Campbell and Campbell 2001, Burger et al. 2005, Wylie et al. 2009, Rezaie Atagholipour et al. 2012) For example, Burger et al. ( 2005 ) showed marked variation in lead conc entrations between blood, kidney, liver, muscle, testes, s kin and ovarian tissues in the Northern Water S nake ( Nerodia sipedon ) (Figure 3). Lead (Pb) concentrations in tissues ranged from 0.025 ppm wet mass in the ovarian t issue to 0.106 ppm wet mass in the skin. Similarly, Rezaie Atagholipour et al. ( 2012) collected muscle, liver, kidney, skin, and blood from the Annulated Sea Snake ( Hydrophis cyanocinctus ) and found large differences in tissue concentrations. Pb concentrations in tissues r anged fro m 309 ng/ g dry mass in the muscle to 1081 ng/ g dry m ass in the kidney (Table 1 1).
20 Given the v ariation in the amount of metal stored in each tissue location interpreting how effects from these metals relate to bioaccumulation becomes difficult. Thus spec ies comparison of tissue concentrations should only be made between similar tissue types. Most toxicological studies involve sampling of contaminants primarily in blood, liver and skin tissues (Campbell and Campbell 2001, Sparling et al. 2010) Thus, s et ting these tissues as standards in the field will go a long way toward making comparisons between snakes in diffe rent locations more meaningful. There also seems to be strong evidence that TICs can accumulate differently among species (Campbell and Campbel l 2001, Sparling et al. 2010) For example, Drewett et al. ( 2013) collected tail samples from four species of snakes along a mercury contaminated river in Virginia. Mercury levels varied from a mean of 0.26 mg/kg dry mass in a Western Rat Snake ( Pantherophis obsoletus ) to 5.60 mg/kg dry mass in a Northern Water Snake ( Nerodia sipedon ). Mercury levels found in the other two species, Queen Snake ( Regina septemvittata ) and Common Garter Snake ( Thamnophis si r talis ) were 4.59 mg/kg dry mass, and 1.28 mg/kg dry mass, respectfully (Drewett et al. 2013) Drewett et al. ( 2013) associated this difference in tissue concentration between species as a function of their food and habitat selection. For example the more aquatic snake (Northern Water Snake) con sumes more aquatic prey (fish), and has higher levels of Hg than the more arboreal species ( Western Rat Snake) which consumes more arboreal prey (eggs and bird chicks). This difference is primarily due to Hg being more bioavailable and concentrated in aqu atic systems (Fowler et al. 2007)
21 In a similar study involving Northern Water Snakes collected from the same study site (the South River, Virginia), Chin et al. ( 2012) compared tissue concentrations of Hg in gra vid females and neonate young. T a il clips w ere taken from gravid females following capture Female snakes were taken back to the laboratory and allowed to give birth. F ollowing birth three neonates from each litter were chosen for whole body homogenization. Other parameters were also measured a nd include presence of malformations, dea th during development, stillbirths, r ates of infertility and runts. Average concentration of Hg from the tail clips was 5.78 mg/kg dry mass in the contaminated site and 0.42 mg/kg dry mass in the control site. Average concentration of Hg from the whole body homogenizations of the neonates was 3.42 mg/kg dry mass for the contaminated site and 0.20 mg/kg dry mass for the reference site. These whole body concentrations from the neonates were positively cor related with the maternal tail tissue, with r 2 =0.84. Although they found high correlation with gravid female tissue and neonate tissue, they found no clear evidence of effects on reproduction and neonate survival (Chin et al. 2012) Laboratory Manipulati on Snakes have been exposed to experimental levels of TICs in the laboratory (Hopkins et al. 2002, 2004a, Ganser et al. 2003, Jones and Holladay 2006) For two years, Hopkins et al. ( 2002) fed juvenile Banded W ater S nakes ( Nerodia fasciata ) in the labora tory prey items collected from a site contaminated with coal ash These prey contained elevated levels of arsenic (As), cadmium (Cd), copper (Cu), selenium (Se), strontium (Sr), and vanadium (V). Although the prey contained high levels of these contamina nts and the snakes accumulated these contaminants, all survived the stud y with no abnormal development. However, Ganser et al. ( 2003) later found evidence of
22 liver damage and fibrosis in these animals Interestingly, these snakes accumulated contaminant s, especially Se (24 g/g dry mass) i n much higher concentrations than found in other vertebrates (Hopkins et al. 2002) See Table 2 5 for co mparison to other vertebrates. The use of l aboratory manipulation to investigate the tolerances and effects of met als on snakes requires further research. There are many possible explanations for the absence of research utilizing snakes for toxicological studies in the laboratory. One reason is the mistaken belief snakes are difficult to maintain in a laboratory. T his belief has been shown to be false many times, and snakes can actually be kept in the laboratory with relative ease, requiring much less care than many commonly studied mammals. Another reason is the relatively long life of snakes Due to this, r esearchers have speculated that accumulation and long term effects of contaminants could be hard to replicate in the laboratory. However some studies have shown that snakes can be kept in a laboratory and they can accumulate significant amounts of contami nants in as little as two years (Hopkins et al. 2002, 2005, Hopkins 2006, Jones and Holladay 2006) Effect s of Contaminants Quantifying the effects of TICs requires monitoring snake populations over an extended period of time, or a direct comparison of sn akes from a contaminated and non contaminated site. Hopkins et al. ( 1999) found that Banded Water S nakes ( Nerodia fasciata ) from a coal combustion waste site had an elevated standard metabolic rate when compared to water snakes from a non contaminated sit e. Elevated standard metabolic rates in an organism that feeds infrequently could have a compounded effect on the fitness of the animal. A snake with an increased metabolic rate would require increased foraging time and resources, which could put that
23 in dividual snake at a higher risk of predation. A higher metabolic rate would also require an organism to feed more often and in greater quantities before a hibernation and reproduction event. Another study used snakes from the above contaminated site in a laboratory experiment that tested the effects of metals ingeste d in artificial diets. Banded Water S nakes ( Nerodia fasciata ) were fed a diet contaminated with As, Cd, Se, Sr, and V for two years. The authors found about 30% of the snakes showed signs of liver fibrosis caused by the proliferation and infiltration of collagen fibers into vascular and other inter hepatocyte areas (Ganser et al. 2003) They also found this damage could occur within a year. Snakes from the one year and two year treatments ha d similar frequencies of fibrosis (Ganser et al. 2003) The authors mentioned prolonged exposure to the contaminants, especially Se, could result in snakes having a higher than normal tissue load of contaminants resulting in a higher metabolism and decrea sed fitness, similar to snakes in the Hopkins et al. ( 1999) study. Other effects of contaminants include abnormal growth or proliferation of cells. Oros et al. ( 2009) found abnormal tissue growth on a captive 17 year old female Saharan Horned V iper ( Ceras tes cerastes ). A small tissue sample was biopsied and determined a fibrosarcoma (Oros et al. 2009) The snake was returned to its owner but later died. During postmortem examination Oros et al. found seven subcutaneous masses spread throughout the snake ranging from 1 to 6 cm in diameter. After the postmortem they diagnosed the cancer as metastatic fibrosarcoma with metastasis to the lung, right atrium, ovaries, and the coelomic wall (Or os et al. 2009) They took liver samples to determine the cause and found that this snake had unusually high levels of
24 cadmium, equal to 223.15 39.51 mg/kg wet mass (Oros et al. 2009) According to the literature fibros a rcomas are one of the more comm on types of neoplasms in aged captive snakes ( Garner 2005 Gumber et al. 2010 personal communication Darryl Heard PhD 2013 ) Although this is the first reported case of metastatic fibrosarcoma associated with high levels of cadmium in snakes, this is onl y a single observation and the correlation could be a coinc idence. Compounds containing cadmium have been known to be carcinogenic (Fowler et al. 2007) See Table 2 4 for comparison to other vertebrates. TICs can have a prolonged half life in living ti ssue and there are reports of female snakes transferring contaminant loads to their young (Campbell and Campbell 2001, 2002, Hopkins et al. 2004b, Gardner and Oberdrster 2006, Chin et al. 2012) For example, Hopkins et al. ( 2004b) fed gravid female B row n H ouse S nakes ( Lamprophis fuliginosus ) Se laced mice over a 10 month period. The snakes accumulated significant amounts of Se in their tissues, especially the kidney, liver, and ovaries. These snakes also transferred significant amounts of Se to their y oung. Hopkins et al. ( 2004b) reports eggs from clutches laid by snakes in their highest Se treatment (20 g/g) had levels of 22.65 0.49 g/g dry mass Mothers from the same treatment contained up to 32 g/g dry mass in the kidney and up to 21 g/g dry mass in the liver and ovary (Hopkins et al. 2004b) The s e level s ha ve been known to be reproductively toxic to fish and birds (Hopkins et al. 2004b) although it is difficult to interpret without knowing the le vels of other ele ments and the form of selenium. See Table 2 5 for c omparison to other vertebrates.
25 Snakes, compared to other vertebrates (T able 2 3 through Table 2 6) appear able to tolerate high levels of TICs in both tissues and in food (Hopkins et al 2002, Burger et al. 2007, Oros et al. 2009, Grillitsch and Schiesari 2010) Researchers have suggested this is due to lower metabolic rates and feeding frequencies when compared to endothermic vertebrates (Campbell and Campbell 2001, Grillitsch and Schi esari 2010) off skin during ecdysis (Burger 1992) This is supported by higher TIC levels in shed skins compared to body tissues (Burger 1992) Jones and Holladay ( 2006) evaluated whether shed skins can be used as reliable indicator s of TIC concentrations in Corn S nakes ( Pantherophis guttat us ). They fed the snakes mice that were laced with mercuric chloride, cadmium chloride, and lead acetate prepared in a phosphate buffered saline (Jones and Holladay 2006) The animals were fed every other week for 34 weeks; each snake received 2 mg/kg of each metal during a feeding. They foun d, depending on the metal, the C orn S nakes were shedding between 0.121% and 0.035% of each metal during each ecdysis (Jones and Holladay 2006) These numbers suggest a gradual loss of metals happens during each shedding event and over time snakes could rid themselves of all TICs if placed in an uncontaminated environment. With such low percentages of metals shed during each cycle, however, the amount of time required to completely remove the metal load would be considerable. Contaminants are also know n to cause abnormalities in erythrocyte nuclei, thus providing a means for detecting their presence. This technique has been used in vertebrates with nucleated erythrocytes, e.g. fish with great success (Strunjak Perovic et al. 2010) Strunjak Perovic e t al. ( 2010) evaluated the erythrocyte abnormalities in
26 the Balkan W hip S nake ( Hierophis gemonensis ) to determine genotoxic effects. They knew that erythrocytic abnormalities can occur spontaneously in fishes (Strunjak Perovic et al. 2010) thus they want ed to determine whether erythrocytic abnormalities can occ ur spontaneously in the Balkan W hip S nake. They found nuclear abnormalities (irregular margins) positively correlated with the frequency of micronuclei, and vacuolated nuclei. Nuclear abnormalitie s were most common in the winter and least common in the spring, and vacuolated nuclei more common in the autumn. The results suggest that the presence of some erythro c ytic abnormalities are influenced by season (Strunjak Perovic et al. 2010) This study demonstrates the potential utility of blood as an indicator of environmental contamination. Further research is needed to develop a reliab le and efficient technique to quantitatively monitor snakes using erythrocytic abnormalities Snakes as Model Study Organisms Snakes have generally b een excluded from toxicological studies for a variety of reasons, although their use has been advocated for many decades (Bauerle et al. 1975, Stafford et al. 1976) This exclusion is because of their small clutch sizes, long interbrood periods, long generation time, and in some s pecies, large size (Hopkins 2006, Burger et al. 2006) However using snakes in toxicological studies could provide us eful information. For example s nakes are exposed to pollutants in many ways, they can be used as bioindicators in different trophic levels, they are found on ever y continent, except Antarctica, and some snakes are easily sampled. For these reasons, snakes can be utilized in toxicology research.
27 E xposure to Pollutants Snakes can acquire pollutants through ingestion, respiration, absorption through the skin, and maternal transfer (Hopkins et al. 2004b, Snodgrass et al. 2008, Grillitsch and Schiesari 2010) Traditionally, accumulation of contami nants through skin was thought to be negligible. Due to the high variability of skin permeability in reptiles, however, contaminant concentration due to skin exposure could be extensive (Lillywhite 2006, Grillitsch and Schiesari 2010) T his is an area of snake toxicology that requires further research. The assimilation of contaminants through respiration could be a significant route of exposure for snakes (Snodgrass et al. 2008) This has yet to be shown experimentally but given the permeability of re spiratory membranes, metal accumulation across these structures could be significant. Maternal transfer has been found to be a significant exposure pathway, especially for snakes inhabiting contaminated sites (Chin et al. 2012) Ingestion has been the mo st fully studied exposure pathway, and significant amounts of contaminants can be a ssimilated in tissue is this way (Hopkins et al. 2002, Jones and Holladay 2006) Figure 1 4 illustrates the possible routes of exposure and trophic transfer of contaminants for snakes Movement of C ontaminants W ithin and B etween Trophic L evels Most snakes are c arnivores or scavengers at high trophic levels (Pough et al. 1998, Campbell and Campbell 2002) They feed on amphibians, invertebrates, other herpetofauna fish, birds, and mammals all of which have the potential to bio accumul ate. The bio accumulation of metals in snakes could provide information about the metal accumulation in other vertebrates that occupy the same trophic level (Weir et al. 2010 Chapter 2 th is document ). Using snakes as sentinels can provide useful
28 information when studying endangered species and the direct capture and sampling of the endangered species is impossible. This requires a researcher to understand the trophic level on which the s nake resides and the trophic level of the endangered species (Drewett et al. 2013) Because snakes bio accumulate contaminants they can be useful for studying contaminant cycling events within the same trophic level and the transfer of contaminants betwe en levels (Hopkins 2006) Abundance and Ease of S ampling Many snakes can be captured and easily kept in a laboratory (Hopkins 2006) S nakes often require very little mainten ance for the upkeep of their cages and they can be fed food items that are easily acquired. Non venomous snakes can be transported with ease and handled by newly trained technicians. Large fami lies, like the Colubridae, are found on every continent except Antarctica and in habit similar habitat s Colubrid snakes of the genera Nerodia and Thamnophis have been commonly used in toxicology research. Current studies have found that tissue samples can be collected from snakes in non lethal ways (blood, tail clip, shed skin) (Burger 1992, Burger et al. 2005, Gardner and Oberdrster 2006, Ho pkins 2006, Jones and Holladay 2006) Some snakes can be abundant, and the ease of capturing these snakes suggests they could be captured, ta gged, measured, and sampled in the field without having to kill or transport them to a laboratory (Hopkins et al. 2001, Gardner and Oberdrster 2006, Hopkins 2006) Most snakes, however, can be difficult to find and capturing large numbers can prove difficult. Conversely, snakes can be found in most environments and their inclusion in toxicology studies can be info rmative and provide researchers with a more complete picture of TICs in a particular study site.
29 Home Range and Habitat Snakes can be found in a variety of habitats, with some snake species ranging over wide distances. For example the cottonmouth ( Agkis trodon piscivorus ) is found throughout the southeast, although individual snakes can show remarkable site fidelity (Lillywhite and McCleary 2008) This is beneficial for a researcher comparing a contaminated versus an uncontaminated site. For example, Hopkins et al. ( 1999) compared Nerodia fasciata from two sites in Georgia. Results from this study could be compared to Nerodia fasciata collected elsewhere. Thus, the individual snake shows site fidelity, and the results from that individual snake could be compared to other snakes within the same species but isolated at a different location These comparisons can be informative for investigators who may want to infer effects from a particular contaminant but in different locations This type of compari son can be difficult in animals that migrate large distances, e.g. birds or large mammals. Conclusions and Future Perspective In conclusion, toxicologic al research using snakes is useful and has increased over the last decade (Figure 1 2) Although this research has been limited some trends have emerged. For example, tissue levels appear to increase with age and size (Hopkins et al. 2004b, Rainwater et al. 2005, Burger et al. 2006, 2007, Wylie et al. 2009, Sparling et al. 2010b Chapter 2 this documen t ) and females seem to have less contaminants than males (Hopkins et al. 2004b, Rainwater et al. 2005, Wylie et al. 2009, Sparling et al. 2010b, Rezaie Atagholipour et al. 2012, Drewett et al. 2013) Snakes also seem able to tolerate higher tissue contami nant levels than do other vertebrates (Hopkins et al. 2002, Campbell et al. 2005, Burger et al. 2006, 2007, Sparling et al. 2010b) and are at a higher risk for bio accumulated contaminants due to
30 their upper trophic level status (Hopkins et al. 1999, Gibbo ns et al. 2000, Jones and Holladay 2006, Sparling et al. 2010b, Reading et al. 2010) Understanding the effects of contaminants in snakes and other animals with similar characteristics can provide a better understanding of environmental contamination. Thi s understanding can help regulators and managers make more informed decis ions regarding contamination. Further research is needed to better understand the movement and effects of contaminants within ecosystems, populations, trophic levels, and individuals Using snakes in ecotoxicology research can give us meaningful insight into the cycling of TICs in the environment. Below is a discussion of the current and future research needs, and ideas on how to solve them (Hopkins 2006, Selcer 2006) Population Va riables Most toxicological research in reptiles has been centered on the individual, and includes concentration levels of a known contaminant in tissue samples (Campbell and Campbell 2001) These levels are then compared to environmental levels and the predicted effects on the organism are inferred This comparison does not provide a useful determination of the effect of a contaminant on the organism. Future research should incorporate as many population variables (i.e. growth, reproduction size, age a nd age specific mortality) as possible (Selcer 2006) Understanding population variables would help research ers answer whether snakes are affected by contaminants as much as endothermic vertebrate s of similar size. To determine the effects of a contaminan t on a snake population requires a multilevel sampling approach that includes a mixture of laboratory and field observations and manipulations (Hopkins 2006) The first step is the initial field
31 observation and sampling of the snake population of interes t (Hopkins 2006) Baseline concentration data should be established for the species as well as population and life history data (sex, size, age, reproduction, etc ) (Selcer 2006) The second step includes laboratory manipulation (Hopkins 2006) which wou ld involve exposing snakes to above are affected negatively by the pollutant. The third and final step would be field manipulations (Hopkins 2006) which could involve tra cking a snake, or transplanting it into a contaminated environment, and then monitoring the snake for e ffects. The combination of these approaches along with monitoring a population over a long time interval would greatly increase our knowledge of contami nants and their role in a population or community that includes snakes. Non lethal S ampling To advance the study of snake populations outlined in the last section an efficient and non lethal way of collecting snake tissue needs to be used. Various method s for obtaining tissue samples in a non lethal manner have been developed (Campbell and Campbell 2001, Burger et al. 2005) and the implementation of these methods should be used when designing and conducting field research in toxicology using snake s Researchers have shown that samples collected using non lethal methods have comparable levels of contaminants as samples from more invasive approaches (Burger et al. 2005, Hopkins 2006, Jones et al. 2009) However, r esearchers need to be aware of the p ossibility that some TICs will accumulate in specific tissue more than in other tissues. Thus it is always important to understand the contaminant under study and to sample the appropriate tissue. Lethal methods should
32 only be considered when all other o ptions (i.e. non lethal sampling) have been considered. Snakes as Model O rganisms Reptiles, specifically snakes, have no standard toxicology model unlike birds, mammals, and fish (Hopkins 2006, Selcer 2006) There is a need to standardize snake toxicol ogy by finding a few species that are prospective model organisms. Luckily this has already unknowingly been done. Researchers have picked certain species of water snakes ( Nerodia spp.) because of their availability, abundance, relative frequency in the toxicology literature, longevity, late maturation and ease of sampling (Campbell and Campbell 2001, Grillitsch and Schiesari 2010, Sparling et al. 2010b) All of these characteristics, as well as a large species range, are ideal for the selection of a mo del toxicology species (Selcer 2006) Other ideal model species, primarily because of their large distribution, include garter snakes ( Thamnophis spp.) (Wylie et al. 2009) and vipers (family Viperidae ) (Hopkins 2000a, Rainwater et al. 2005) Having a stan dard snake toxicology model is important for supplying general reptile toxicology data for use between similar species, and surrogate sampling when trying to understand threats to threatened or endan gered species. Although having model organism s is import ant for comparison between similar species, it is also important to recognize the alternative of studying available species to give researchers the ability to discover novel things. Although publications on environmental contaminants concerning snakes are few, it does not necessarily mean snakes are not affected by these pollutants As more studies are published, and more research is done on the susceptibility of snakes to environmental toxins, we may find they are more sensitive to contaminants than once
33 thought. Research in the last decade has uncovered some interesting insights into the effects of contaminants on snakes. Much more research is needed and finding effects will become increas ingly more important as declines of reptiles and the need for in formed decisions regarding contaminant effects on snake populations, increase (Gibbons et al. 2000, Reading et al. 2010)
34 Table 1 1 M etal concentrations from the most recent studies regarding snakes 2011 2013. For a detailed description of results from snake studies prior to 2011 consult Campbell and Campbell ( 2001 ) and Grillitsch and Schiesari ( 2010) Species Year Location N Wet/Dry Units Metal Tissue Conc entration Source Comments Annulated Sea Snake ( Hydrophis cyanocintus ) 2012 Hara Protected area, Persian gulf 13 Dry ng/g Pb Muscle 309 29 Rezaie Atagholipour et al. ( 2012) All samples are Mean Standard Error Liver 390 64 Kidney 1081 134 Skin 368 62 Blood 380 49 Cd Muscle 72 29 Liver 339 120 Kidney 232 60 Skin 117 82 Blood 78 55
35 Table 1 1 : Continued. Species Year Location N Wet/Dry Units Metal Tissue Conc entration Source Comments Annulated Sea Snake ( Hydrophis cyanocintus ) 2012 Hara Protected area, Persian gulf 13 Dry ng/g N Muscle 355 66 Rezaie Atagholipour et al. (2012) All samples are Mean Standard Error Liver 575 178 Kidney 429 49 Skin 2212 1256 Blood 389 110 V Muscle 640 59 Liver 851 200 Kidney 2034 206 Skin 501 55 Blood 1826 795
36 Table 1 1 : Continued. Species Year Location N Wet/Dry Units Metal Tissue Conc entration Source Comments Northern Water Snake ( Nerodia sipedon ) 2012 Middle and South Rivers, Virginia 9 Dry Mg/kg Hg Tail 0.42 0.09 Chin et al. (2012) reference site adult females Blood 0.03 0.30 adult females 12 Dry Mg/kg Hg Whole body 0.20 0.11 reference site, neonate offspring 22 Dry Mg/kg Hg Tail 5.78 0.55 contaminated site adult females Blood 1.72 5.32 contaminated site adult females 12 Dry Mg/kg Hg Whole body 3.42 0.45 contaminated site neonate offspring Northern Water Snake ( Nerodia sipedon ) 2013 South River, Virginia 36 Dry Mg/kg Hg Tail 5.60 0.40 Drewett et al. (2013 ) Contaminated site South River 25 Dry Mg/kg Hg Tail 0.49 0.07 Reference site Queen Snake ( Regina septemvittata ) 2013 South River 9 Dry Mg/kg Hg Tail 4.59 0.38 Drewett et al. (2013 ) Contaminated Site Garter Snake ( Thamnophis sirtalis ) 2013 South River 7 Dry Mg/kg Hg Tail 1.28 0.32 Drewett et al. (2013 ) Contaminated Site Rat Snake ( Elaphe obsolete ) 2013 South River 10 Dry Mg/kg Hg Tail 0.26 0.09 Drewett et al. (2013 ) Contaminated Site
37 Figure 1 1 A simplified food chain illustrating how a contaminant (+) is bioaccumulated, bioco ncentrated, and biomagnified. The fish near letter A show bioconcentration of a contaminant (+) not by the process of ingestion. The fish near letter B show bioaccumulation of a contaminant (+) with the arrows showing ingestion. The letters B D show biomagnification or the stepwise increase of tissue conc entrations of a contaminant (+) across trophic levels. All arrows indicate ingestion. All clipart taken from openclipart.org.
38 Figure 1 2 The number of snake Toxic Inorganic Contaminant (TIC) publications published over the last 6 3 years. Publications that u tilized snakes with radionucleotide or radiation poisoning and organic compounds were not included in this tally. Figure 1 3 Lead tissue concentration (ppm) as a function of tissue type. The tissues are from Nerodia sipedon a nd show the variation in l ead concentration depending on the type of tissue sampled Concentrations are from wet mass samples. Data taken from Burger et al. ( 2005) 0 5 10 15 20 25 # of publications Toxic Inorganic Contaminant publications between 1950 and 2013 involving snakes 0 0.02 0.04 0.06 0.08 0.1 0.12 Liver Kidney Blood Muscle Skin Testis Egg Concentration (ppm) Tissue Type Comparison of Lead Tissue Concentrations
39 Figure 1 4 Schematic show ing possib le exposure routes and source of contaminants for snak es. Transport pathways are arrows and potential uptake routes are indicated by the arrows terminating at the small snake. Boxes provide descriptions of the sources of contaminants. This is a simplified diagram and m ay not show all possible rout es of contamination. Figure adapted from Snodgrass Joel 2008 Urban Herpetology (Page 181, Figure 3) Society for the Study of Amphibians and Reptiles publication, Salt Lake City, Utah
40 CHAPTER 2 BIO ACCUMULATION OF METALS IN AN INSULAR POPULATION OF FL ORIDA COTTONMOUTH SNAKES ( AGKISTRODON PISCIVORUS CONANTI ) Introduction Environmental contamination by Toxic Inorganic Compounds (TICs) is a major global problem (Grillitsch and Schiesari 2010) TICs, which include the heavy metals the metalloids, the transition metals, and certain nonmetallic elements have concerned researchers and the general public for decades (Crosby 1998) Primary reasons for this concern are the unique abilities of TICs to remain in the environment for long periods of time, the potential for TICs to bioconcentrate, bioaccumulate, and biomagnify up trophic levels, and the toxicity of these compounds (Zakrzewski 1997, Crosby 1998) TICs in the right doses can be extremely toxic to organisms and the presence o f TICs in the environmen t should be monitored closely. Biomagnification is the increased co ncentration of a contaminant as it moves up trophic levels (Crosby 1998) This is a concern for upper trophic level consumer s including humans, because of the very high probability that the food being consumed could contain high levels of a TIC. Even small amounts of a TIC found in the environment have the potential to bi oaccumulate up trophic levels. Snakes can be ideal bioindicators of TIC contamination because they are upper trophic level consumers, long lived, and are found in similar environments as o ther more sensitive species, e.g birds (Burger et al. 2007) Many snakes consume a wide variety of prey, and are found in complicated trophic systems. Thus it can be difficult to pinpoint sources of contamination, via food re sources, when studying snakes. Conversely, i nsular snakes can be important for the study of biomagnification in trophic systems because of the simplified trophic systems and limited prey variety. This
41 study focuses on an insular population of Florida C ottonmouth S nakes ( Agkistrodon piscivorus conanti ) on Seahorse Key, Florida. Food resources on the island are limited except for the large amounts of allocht h onous resources brought t o the island via the nesting water birds (Wharton 1969, Lillywhite and McCleary 2008) This influx of allocht h onous resources enriches the insular food supply but also provides the from the Gulf of Mexico. S eahorse Key is a small island (~67 hectares) located in Levy County Florida approximately eleven kilometers from the mainland and six kilometers from the coastal town of Cedar Key, Florida. Seahors e Key is part of the Cedar Key National W ildlife R efuge established in 1929 to protect the unique environment of the island and surrounding waters (Gude 2000) Seahorse Key is an important nesting habitat for colonial bird species, with numbers of nesting birds ranging from 10,000 20,000 individuals (Lillywhit e and McCleary 2008) Cottonmouths on the island concentrate foraging mainly during the night near dense bird rookeries on the western end of the island and scavenge fresh or regurgitated fish the nesting birds drop while feeding their young (Lillywhite a nd McCleary 2008) These fish provide the cottonmouths with abundant and accessible food for almost nine months of the year. The amount of fish dropped is unknown but appears large based on the numbers and relative size of snakes on the island ( Lillywhite and McCleary 2008). The relationship between the cottonmouth snakes and the nesting water birds on Seahorse Key is possibly unique because in most snake bird systems, the snakes feed on t he birds as a primary prey item.
42 The abundance of snake s on the island appears to vary from one e nd of the island to the other with the greater concentrations near bird rookeries. Lillywhite and McCleary (2008) found snake density was much higher on the western end compared to the eastern end of the island. The most evident reason for this is the availability of fish carrion. Snakes do inhabit the eastern portion of the island where they can supplement their diet with the invasive Black R at ( Rattus rattus ). Cottonmouths inhabiting the east ern end of the isla nd tend to be smaller and have a lower body condition than cottonmouths on the west end of the island (Lillywhite and McCleary 2008) Because the fish carrion is most likely coming from the Gulf of Mexico, i nformation regarding the trophic transfer of TIC s from the Gulf of Mexico can be derived by looking at island systems and simplified food chains, like Seahorse Key (Figure 2 1). The principle objective of this study was to examine the tissue concentrations of cottonmouth snakes from Sea horse Key a nd thr ee mainland populations. The hypotheses and o bjectives were: Hypothesis 1: Seahorse Key cottonmouth snakes have higher levels of metals than cottonmouth snakes living on the mainland of Florida. Objective: To determine the concentration of metals in ti ssues collected from Florida Cottonmouth Snakes living on Seahorse Key and Florida Cottonmouth Snakes living in the Lower Suwannee Wildlife Refuge, Paynes Prairie State Preserve, and Big Cyprus National Monument. Hypothesis 2: Florida Cottonmouths inhabi ting Seahorse Key have similar or greater levels of TICs in their tissues when compared to tissues from fish dropped by the nesting water birds. Objective: Collect freshly regurgitated fish and dropped fish from under nesting water bird rookeries and measu re TIC levels in these fish.
43 Presumably, mainland populations of cottonmouths consume different prey sources, thus any differences in tissue concentrations between locations should likely be due to trophic resources namely the allochthonous food source s available to the cottonmouths on Seahorse Key. Insofar the cottonmouth snakes and the water birds nesting on the island both consume fishes from the Gulf of Mexico, the results from this study have important implications regarding sources of TICs and th eir potential impact on the insular ecology of nesting birds, as well as resident cottonmouths. Methods and Materials All animal studies were conducted with approval of the University of Florida Institutional Animal Care and Use Committee (IACUC) (Study # 200801755 ). All animal handling and care approved by IACUC was followed Animal Collection During 2008 2010, 47 Florida Cottonmouth S nakes were captured for liver and blood extraction, from four sites in Florida; Seahorse key (20 snakes) Big Cyprus Nat ional Park (2 snakes) Paynes Prairie State Park (8 snakes) and the Lower Suwannee National Wildlife Refuge (7 snakes) ( Figure 2 2 and Figure 2 3 ). Snakes were captured during searches that begun shortly after sunset. Areas searched for snakes included r iver banks, under logs, under palm fronds, the base of bird rookeries, and around ponds. C ottonmouth s were captured with a snake hook or tongs and placed in a five gallon bucket. All c aptured snakes were taken back to the laboratory for sampling. In ad dition to the above snakes, 65 whole blood samples were collected from snakes in a concurrent study of venom at the University of Florida Veterinary Teaching Hospital (UFVTH) These snakes were collected from the same sampling sites and
44 include 24 from Pa ynes Prairie, five from Big Cyprus National Park, 20 from the Lower Suwannee National Wildlife Refuge, and 26 from Seahorse Key. The rationale for taking blood and liver samples was blood will reflect short term storage of TICs and liver will reflect lon g term storage of TICs (Fowler et al. 2007) In many toxicological studies the liver is considered the metabolic center of the organism and measurements taken from liver c an be compared with data for other vertebrates. Blood and liver samples are also co mmonly used in toxicology studies, especially those involving snakes (Campbell and Campbell 2001) Dropped or regurgitated fish were collected opportunistically from Seahorse Key d uring the years 2008 2010 (Figure 2 2). F ish were placed in plastic bags, labeled, brought to the laboratory, and placed in the freezer to await dissection and analysis The fish were visually identified using (Kells and Carpenter 2011) and included: Spotted S eatrout ( Cynoscion nebulosus ), Striped M ullet ( Mugil cephalus ), Atlantic Thread H erring ( Opisthonema oglinum ), and Hardhead C atfish ( Ariopsis felis ) Fish were collected in order to compare TIC concentrations between a food source (fish) and a consumer (snakes). Blood samples Blood samples were collected from li vin g snakes in the laboratory. Blood samples were opportunistically taken from snakes anesthetized for a concurrent venom study at the UFVTH. Approximately 0.5 1.0 mL of blood was collected from either the ventral coccygeal (caudal) vein using a 25 ga uge ne edle and 3 mL syringe or directly from the heart (Bush and Smeller 1978) The syringe and needle were pre heparinized by drawing heparin into the syringe then expelling all of the liquid; the remaining heparin was sufficient to prevent coagulation. Blood taken from the snakes was stored in
45 labeled cryo genic tubes ( Fisher ) and placed in a refrigerator at 8 o C. All of the blood samples were given a unique number, organized, and shipped to a commercial laboratory for analysis (Utah Veterinary Diagnostic Labo ratory in Logan, Utah). Liver samples For liver biopsy e ach cottonmouth snake was restrained in an acrylic tube and positioned in dorsal recumbency A nesthesia was induced with propofol ( 5 10 mg/kg, Abbott Laboratories, North Chicago, Illinois 60064, USA) either IV in the ventral coccy geal vein, or intracardiac, using a 25 ga x 2.3 in butterfly catheter. Once in a surgical plane of anesthesia, the snake was positioned in the right lateral recumbency, with the surgeon facing the dorsal surface of the snake. The liver was located by transcutaneous coelomic palpation. An area over the mid length of the liver was aseptically prepared. A linear local anesthetic block was done in the area to be incised us ing a mixture of 1:4 lidocaine 2% (Lidoject @, Butler Animal Health Supple, Dublin, Ohio, 43017, USA) and bupivicaine 0.5% 2 mg/kg in tradermally and subcutaneously. A longitudinal incision approximately 3 5 cm was made 2 3 r ows of scales dorsal to the ven tral scutes. The skin was reflected ventrally, and a ventral paramedia l incision was made through the ventral abdominal muscles. The coelomic membrane was incised with scissors. The ventral muscles were reflected ventrally to expose the internal organs. The liver was identified and a small triangular wedge was clamped in the edge of the right liver lobe using small mosquito forceps Care was tak en to avoid the ventral hepatic vein and hepatic portal vein dorsally. The forceps were slowly and gradually c lamped to avoid inadvertent crushing and crumbling of the tissue without hemostasis. The forceps were clamped for three minutes The tissue wedge (~100mg) was then sharply dissected using a scalpel blade lateral to the forceps jaws.
46 A triangular sample of tissue was then collected and preserved accordingly for toxicolog ical analysis. The clamps were removed and the area observed for hemostasis. A small patch of hemostatic compressed collagen gelatin sponge (Gelfoam, Pharmacia & Upjohn, New York, New Yo rk, 10017, USA) or knitted oxidized cellulose fabric (Surgicell, Johnson & Johnson, Sommerville, New Jersey, 08876, USA) was used to control residual hemorrhage. Closure of the body wall was done in two layers (coelomic membrane and muscle) with 3 0 or 4 0 polydioxanone suture (PDS, Ethicon Inc, Sommerville, New Jersey, 08876, USA) and skin was closed in a continuous horizontal mattress pattern using 3 0 or 4 0 polydioxanone. Following surgery the snake was placed inside a cage in the approved animal facil ity to recover. After recovery from anesthesia, ea ch snake was observed for 1 2 weeks, during which time it was allowed to eat and defecate one meal. This was done to insure gastrointestinal function was normal. After two weeks each snake was released b ack to the location of its capture. The collected tissue sample was divided intra operatively into equal portions One section was weighed and frozen for further contaminant analysis. The mass of this section was recorded, and if the target mass of 100 mg was not achieved, a further piece of the remaining section was obtained until 100 m g of tissue was obtained for freezing. The remaining tissue section was preserved in 10% neutral buffered formalin but was not weighed. Tissue S amples from Fish Fish that were collected from Seahorse Key were removed from the freezer and allowed to thaw at room temperature. Fish were visually identified to species using a
47 field guide written by Kells and Carpenter ( 2011) placed on a dissecting tray, weighed and measured. An incision was made along the ventral surface to access the body cavity. The liver was identified and a small piece was removed and placed on a weighing tray. Once 100mg of liver was obtained for analysis, the remaining tissue and fish were placed back into the labeled bag and placed in the freezer. The 100mg of liver was placed in a vial, given a unique number, and placed in the freezer for shipment Sample Preparation and Analysis Liver and blood samples from the snakes, and liver samples from the fish were removed from the freezer and refrigerator, given uniq ue sample numbers, packed in a s tyrofoam cooler with ice packs, and shipped overnight to the Utah Veterinary Diagnostic Laboratory (UVDL) in Logan, Utah. For analysis, t he samples w ere digested in trace mineral grade nitric acid under he at The digests were then diluted with ultra pure water to a final nitric acid content of 5%, which provided a matrix match for the analytical standards. The prepared samples were analyzed by Induct ively Coupled Plasma Mass Spectrometry (ICP MS) and assessed against concentration curves of known mineral standards (personal communication UVDL 2013). Standard curves and quality contr ol samples were analyzed every five samples. Twenty nine elements we re measured in parts per million (ppm) for each sample (Table 2 1) The diagnostic laboratory in Logan, Utah could only detect metals with concentrations greater than 0.01 ppm. Statistical Analysis Data from snake tissue samples were entered into excel and organized to facilitate analysis. Mean, and standard deviation were calculated for each metal (Table
48 2 1). Samples with values < 0.01 were changed to 0, omitting many metals from further analysis because there were below the minimum detectable level M etals were chosen for statistical analysis based on potential persistence in the environment, toxicity, and previous published studies These m etals were arsenic (As), selenium (Se), mercury (Hg), and cadmium (Cd) (Table 2 1) L inear regression analysis was used to compare the snout vent length (SVL) and the total length (TL) to the mass of the snake. Snout vent length was then compared to the mass of each snake from each location. This was done in order to determine whether there is a linear relationship between the length and mass of snakes Graphing each location separately helped define the size differences A one way ANOVA test was used to compare snake masses between locations (McKillup 2005) Snake masses were used due to the tight correlations between snout vent length and the mass of the snake (Figure 2 4 and Figure 2 5 ). T he size of the snakes at Seahorse Key were si gnificantly different (p value <0.01) than that of snakes at the other locations T hus a n ANCOVA was used to control for snake mass while comparing the tissue concentrations between locations. After each ANCOVA was calculated, a Tukey HSD (Honestly Significant Difference) was used to do pair wise compar isons between each location to determine how each location differed from each other Differences were considered statistically significant with p values <0.05. The R statistical program and package Multcomp was used for the ANCOVA and Tukey HSD analysis (Hothorn et al. 2008, R Development Core Team 2011) Linear regression and graphing was completed using Microsoft e xcel
49 D ata from fishes were entered into excel and organized to facilitate analysis. M ean, and standard deviation were calcul ated for eac h metal (Table 2 2) Metals chosen for further analysis were the same metals chosen for the snakes Using Microsoft excel, linear regression analysis was completed on all fish samples to compare the total length (TL) of the fish to the tissue concentrati on for the selected metals. Conversion of Wet Mass to Dry Mass The toxicology literature reports tissue concentr ations in both wet and dr y m ass Thus it is important to be able to convert between the wet and dry mass samples. In order to do this, whole livers were removed opportunistically from three road killed cottonmouth snakes. Portions of the liver that were damaged from the impact of a vehicle were not used. The liver was then divided i nto smaller (~1g) pieces, weigh ed initially ( wet mass ) and placed in an oven set at 68 C Six s amples were weigh ed every twenty f our hours until the mass of the sample did not change over a sampling period. This final mass was considered the dry mass and was used to calculate the moisture content of an average liver sample [(W M D M )/W M = MC]. Where W M = Wet Mass D M = Dry Mass and MC = Moisture Content. Conversion Factor (CF) was calculated by subtracting MC from one [1 MC = CF]. The wet mass tissue concentration (W M C) was converted to dry mass tissue concent ration (DM C) using the following equation: D M C = WM C /CF Conversions were completed for all snake liver samples. Results Results from the snake samples analyzed for 29 metals are summarized in Table 2 1. Included in Table 2 1 are the sample size, arithmetic means and standard
50 deviation (S.D .) values of each location and tissue type. Standard deviation was used to display the variation in the sample data. Wet masses were converted to dry masses using the conversion factor of 0.537, and are includ ed in Table 2 1 along with standard deviation Six samples of liver taken from three cottonmouth snakes were used to calculate the conversion factor. The mean wet mass of the liver samples was 1.331 g ( 0. 211 S. D .) and the average dry mass was 0.716 g ( 0.054 S.D .) Thus MC was 0.462 and the CF was 0.537. Figure 2 4 represents the linear regression analysis between the masses and lengths of all snake samples. Tight correlations (R 2 = 0.852 total length and R 2 = 0.8467 for snout vent length) were found between variables thus mass was used in the ANCOVA described above. Mass was significantly different betw een populations Arsenic (As), selenium (Se), cadmium (Cd), and mercury (Hg) were chosen for further analysis. Controlling for m ass, r esults from the ANCOVA show snake liver tissue from Seahorse Key had significantly higher levels of As when compared to the mainland locations ( p values <0.01 ) Pairwise comparisons using a Tukey HSD test showed significant differences in liver concentra tions for As between each mainland population and Seahorse Key (p values <0.01) Blood tissue samples were not significantly different between populations for As, with p values > 0.05. Liver and blood concentrations of selenium, mercury, and cadmium wer e not significantly different among locations, with p values >0.05. Concentrations of As, Cd, Se and Hg for each location and tissue type are summarized in Figure 2 6 through Figures 2 9. Se and Hg concentrations in liver samples (ppm wet ma ss) from sna kes on Seahorse Key correlated positively with mass with R 2 values of 0.92 and 0.95,
51 respectfully (Figure 2 12 and 2 13). As and Cd c oncentrations in liver samples (ppm wet mass) from snakes on Seahorse Key had slight positive correlations with mass with R 2 values of 0.083 and 0.059, respectfully (Figure 2 10 and Figure 2 11) As, Cd, Se, and Hg concentrations in blood samples (ppm wet mass) from snakes on Seahorse Key had slight positive correlations with mass with R 2 values of 0.11, 0.000004, 0.0066, a nd 0.0044, respectfully (Figures 2 10 through 2 13). Liver concentrations from all of the snakes sampled for the other 2 5 metals in the array had ranges of arithmetic means of concentration (ppm wet mass) as follows: Ag (0.00 0.01), Al (0.96 2.78), B ( 0.05 0.11), Ba (0.01 0.09), Be (0.00), Ca (107 196), Co (0.03 0.07), Cr (0.39 0.67), Cu (6.72 10.11), Fe (794 .44 1326 .77 ), K (1848 .83 2964 .47 ), Li (0.01), Mn (0.46 0.89), Mo (0.30 1.04), Na (1224 .44 1886 .93 ), Ni (0.01 0.03), P (18 79 2868), Pb (0.00 0.04), Sb (0.02 0.07), Si (18.98 31.38), Sn (0.00 0.02), Sr (0.10 0.18), Tl (0.00 0.01), V (0.06 1.92) and Zn (25.01 36.61). Most of these metals had ranges less than ten ppm wet mass with many less than one ppm wet mas s On average for the 29 metals blood tiss ue had lower concentrations than t he liver tissue in snakes sampled in all locations Exceptions include, Ca (ranges of 109 140 ppm blood wet mass, 107 196 ppm liver wet mass), Li (ranges of 0.00 1.27 ppm bl ood wet mass, 0.001 ppm liver wet mass), and Na (ranges of 2802 3093 ppm blood wet mass, 1224 1886 ppm liver wet mass). High levels of Fe, P, Na, K, and Ca were found in the liver and blood samples from all snakes. All other metals were found to be i n concentrations under 70 ppm wet mass liver, and under 18 ppm wet mass b lood.
52 Results from the fish samples analyzed for twenty nine metals are summarized in Table 2 2. Conversion factors for the fish samples cannot be calculated, and all results are i n wet mass. The results for the mass length regression in fish are shown in Figure 2 1 4 Mass is positively correlated to fis h length with a R 2 value of 0.8963. C oncentrations of As, Se, Cd, and Hg for fish collected at Seahorse Key are shown in Figures 2 9 2 12. Tissue concentrations as a function of fish le ngth are shown in Figure 2 15 and all samples were poorly correlated with R 2 values around 0.00. Mercury and selenium levels are compared in Figure 2 16 with mercury levels on average higher than se lenium levels for all locations and tissue types. A strong correlation was found between selenium and mercury with a R 2 value of 0.9186 (Figure 2 17) Discussion As discussed in Chapter 1 there are four exposure pathways for snakes; transcutaneous uptake maternal transfer, respiration, and ingestion. Transcutaneous uptake in snakes is unlikely due to the keratinized structure of the skin (Snodgrass et al. 2008) although this statement has not been experimentally shown. Maternal transfer could be a sig nificant exposure pathway but insofar as birth is a one time event, this exposure is relatively limited (Hopkins et al. 2004b) Respirat ion could be a significant route (Snodgrass et al. 2008) but this would require the metals to be aerosols and absorbed across membranes in the lungs Although absorption across the membranes in the lungs could happen in any population of snakes, the most likely route of exposure is ingestion. Digestion of TIC contaminated prey has been shown to be a primary sourc e of c ontamination in snakes, and large amounts of consumed contaminants can be
53 assimilated into tissue (Hopkins 2000b, Hopkins et al. 2002, Jones and Holladay 2006) Cottonmouths inhabiting Seahorse Key are assimilating these metals in their liver tissues, and these levels are greater than the concentrations in fish for cadmium (Cd) and selenium (Se) ( Figure 2 7 and Figure 2 8 respectfully). Arsenic levels were lower than levels found in fish (Figure 2 6). Lower levels of arsenic could be due to the excretion of arsenic by the kidneys because this has been found to be an elimination route in other vertebrates (ATSDR 2007) These results suggest that fish living around the island are bio concentrating the metals in their tissues. High levels of metals are found in the fish, and even small fish are bio concentrating large amounts of TICs (Figure 2 1 5 ). Thus co ttonmouths can be exposed to contaminants from all sizes of fish and can start to accumu late TICs from a very young age The cottonmouths from Seahorse Key had significantly higher liver tissue concentrations of arsenic than those from four locations on the mainland of Florida. The difference in tissue concentrations suggests that levels of con taminants in insular versus mainland prey resources are different (Wharton 1966, 1969, Lillywhite and McCleary 2008) Cottonmouths on the mainland are consuming many different types of prey items and these items are not coming directly from the Gulf of Mexico. This includes carrion, amphibians, small mammals, birds, and some fish (Burkett 1966) The snakes on Seahorse Key are consuming primarily fish dropped by the nesting water birds (Wharton 1969, Lillywhite and McCleary 2008) which suggests there is an indirect trophic link between snakes and the Gulf of Mexico. Because s nakes and birds are consuming the same prey, we would expect the birds on the island to have similarly elevated levels of the above contaminants. At the time of t his writing, a direct
54 comparison of tissue concentrations between cottonmouths and nesting waterbirds has not been done. However, given that birds are endothermic and consume larger quantities of fish, one would expect similarly elevated ti ssue concentrat ions. The simplified food web described above from Seahorse Key seemingly provides a clear example of bio accumulation and trophic transfer (Figure 2 2) The results from this study are concerning because humans also consume fish that are caught in the Gu lf of Mexico. Ce dar key is a small community eleven km to the east of Seahorse Key and the people living here have a long history of fishing the local waters around Seahorse Key. Sport fishing is conducted around the island as well as clamming and shrimp ing activities. The high levels of contaminants found in the cottonmouths on Seahorse Key should be cautionary for people living near Seahorse Key and who consume a lot of fish in their diet (Fowler et al. 2007) Metal Summary and Comparison to Other O r ganisms In order to fully understand the implications of the results from this study, one must understand the source of these elements, the background levels of these elements, the toxicity to other organisms, and the levels reported in snakes and other gr oups of animals. Below I briefly summarize the current state of knowledge regarding toxicology for each of the four metals chosen in this study. Insofar as arsenic was the only metal found to be significantly different between the snakes on Seahorse Key and the mainland, it will b e discussed in greater detail. In order to provide some form of baseline to compare t issue concentrations in snakes, s tudies tha t report tissue concentrations in birds, mammals, fish, and other reptiles were chosen for comparison to the levels in the cottonmouth snakes. Invertebrate and sediment sample data from the Gulf of Mexico will be discussed in Chapter 3.
55 Arsenic Arsenic levels in the tissue of Seahorse Key C ottonmouths were significantly different than arsenic levels found in snakes on the mainland (Figure 2 6) The difference is likely attributable to differences in food resources Arsenic levels reported in other publications using snakes are similar or less than the levels reported in the Seahorse Key cotto nmouths (Table 2 3). A notable exception is a population of Banded Water Snakes sampled from a contaminated coal ash settling basin (Hopkins et al. 1999) with levels reported in the liver of 132 ppm dry mass (Table 2 3). When comparing the level s of ars enic found in the cottonmouth to levels found in other organisms, there are noticeable differences (Table 2 3). For example, Wickliffe and Bickman ( 1998) collected Brown Pelican ( Pelecanus occidentalis ) eggs from a dredge spoil site in North Carolina and found concentrations of arsenic of just 0.01 g/g wet mass this is much lower than the levels found in the cottonmouth. This might be due to the type of sample i.e. sampling an egg instead of a portion of the liver. Wiemeyer et al. ( 1987) sampled liver s from the Osprey ( Pandion haliaetus ) in New Jersey and found tissue concentrations of arsenic of 3.2 g/g wet mass This concentration is very similar to the 3.89 ppm wet mass concentration found in the Seahorse Key cottonmouths. In a review of arsenic tissue levels Eisler ( 1988) reported ranges of 2 5 mg/kg wet mass in liver sample s collected from marine fish (Eisler 1988) Marine mammals also have similar levels of arsenic. Kunito et al. ( 2008) conducted a comprehensive literature review and found that the highest arsenic in marine mammals, except whales, was the harp seal ( Pagophilus groenlandicus ) with liver samples of 7. 68 g/g dry mass This is very similar to the cottonmouth liver samples of 7.25 ppm dry mass In the same review Kunito et al. ( 2008) compared levels of hepatic arsenic between terrestrial,
56 coastal, and marine birds and found that marine birds had much higher levels of arsenic specifically the Black footed A lbatross ( Phoebas tria nigripes ) with levels as high as 10 g/g dry mass This species feeds entirely at sea and its primary food source is marines fishes, which is similar to the cottonmouths on Seahorse Key whose primary food source is also marine fishes. Although the hepatic arsenic levels in cottonmouths were high compared to other vertebrates, they were not the highest reported in reptiles. Burger et al. ( 2000) collected liver samples from the American Alligator ( Alligator mississippiensis ) and found concentrations as high as 41 g/g wet mass This is much higher than levels reported in other organisms (Table 2 3) and is concerning not only for the health and survival of the American A lligator but for humans that consume alligator meat. According to the standards o f arsenic consumption set by the U.S. D epartment of Health, a person consuming meat from this alligator would be very close to the acute exposure limit of 0.005 mg/kg d of arsenic (ATSDR 2007) If the person consumed this meat regularly during his or her lifetime, then they would have exceeded the chronic exposure limit of 0.0003 mg/kg d (ATSDR 2007) These standards are set based on the lethal dose of arsenic, which according to ATSDR ( 2007) the minimum lethal acute dose of arsenic is around 130 mg or about 2 mg/kg for humans Death in humans has resulted from chroni c exposure of much lower levels. F or example five children between the ages of 2 and 7 years died from chronic arsenic poisoni ng after drin king water with 0.05 0.1 mg A s/kg/d (ATSDR 2007) Arsenic (As) is a metalloid with a standard atomic weight of 74.92 and an atomic number of 33 (ATSDR 2007)
57 found in conjunction wit h sulfur and other metals. Some areas of the United States have unusually high levels of As in rock, and this leads to higher levels in the soil and drinking water. For example, the state of Florida has unusually high levels of arsenic in the soil and in drinking water (Chen et al. 2001) Arsenic can form both inorganic and organic compounds. Inorganic compounds are arsenic compounds that do not b ind with carbon, these include a rsin ite and a rsinate. Arsinite is a chemical compound that contains an ars enic oxoanion, where the arsenic has an oxidation state of +3, for example As0 3 3 Arsinates are chemical compounds that contain As0 4 3 Arsenates are salt or esters of arsenic acid (H 3 AsO 4 ). Arsenic acid is the basis for most arsenic contamination of g roundwater. Arsenic contamination of groundwater is a major concern for people living in locations with high levels of arsenic in the soil. This is a major concern for states like Florida, which has many cities above the US department of Health minimum c hronic exposure levels (Chen et al. 2001) Organic a rsenic compounds are chemical compounds that contain a chemical bond between arsenic and carbon. Organic a rsenic has also been used in pesticides, herbicides, insecticides, and as a chemical weapon in World War I main use is manufacturing, specifically for strengthening other metals such as copper, or lead and in the treatment of wood (Fowler et al. 2007) Using a rsenic to treat wood has been common practice for almost a century (Fowler et al. 2007) Wood treated with arsenic comes in the form chromated copper a rsenate (Khan et al. 2006) Once the treated wood has been placed in the environment, the arsenic slowly leeches out and begins to contaminate the local water supply (Kha n et al. 2006, Fowler et al. 2007) The docks around Cedar Key most likely were treated with chromated c opper a rsenate and
58 the high levels of a rsenic observed in the organisms on and around Seahorse Key are possibly the r esult of the treated wood in the docks. Arsenic was used extensively in the past to treat diseases including syphilis, amoebic dysentery, and trypanosomiasis (Eisler 1988) In 1938, it was established that arsenic could be used to counteract selenium toxicity ( Eisler 1988) Today a rsenic is used in very few medications. The production of organoarsenic compounds has decreased in the last few decades because of environmental concerns, especially concerns regarding drinking water. Arsenic can be very toxic in t he right dose and speciation. Some species of arsenic have acute toxicity levels of 2 mg/kg d, and other species of arsenic have acute toxicity levels that are much higher (i.e. >100mg/kg d). Arsenic disrupts ATP production in the cells through many diff erent mechanisms, but primarily it competes with phosphate in the step of glycolysis that produces 1,3 bisphosphoglycerate from glyceraldehyde 3 phosphate (ATSDR 2007) Instead of 1, 3 bisphosphoglycerate, 1 arseno 3 phosphoglycerate is produced, which is unstable and quickly hydrolyzes and forms the next intermediate 3 phophoglycerate. Thus the ATP molecule that would have been produced in this step is lost and essentially arsenic is a decoupler o f glycolysis (ATSDR 2007) Arsenic in high enough doses can cause vomiting, diarrhea, and gastrointestinal hemorrhaging. Death ensue s from fluid loss and circulatory system collapse. When inorganic arsenic and its compounds enter the food chain, they are progressively converted to a less toxic state through the process of methylation. Methylation is the process of adding a methyl group to a substrate, in this case a methyl
59 group to inorganic arsenic. The addition of this methyl group converts inorganic arsenic to organic arsenic, and slowly through the add ition of additional methyl groups arsenic become less toxic. This process occurs through the use of enzymes, and the natural process in the environment called bio methylation. Thus arsenic converted in this way can become much less toxic (Cullen and Reim er 1989) Arsenic intake in humans usually occurs via oral consumption of food or drinking water and most arsenic is eliminated in the urine (ATSDR 2007) Cadmium Cadm ium (Cd) levels in the Florida C ottonmouths at Seahorse Key were not significantly hi gher than those in t he mainland populations (Figure 2 7 ) Cadmium levels reported in other snake species from the literature are similar or less than the levels reported here for the Florida Cottonmouth Snake (Table 2 4, Chapter 1 this document). Levels reported in other vertebrates are similar to levels reported in the cottonmouths in this study (Table 2 4). For example Bruehler and De Peyster ( 1999) collected hepatic cadmium samples from the brown pelica n and found concentrations of 0.58 g/g dr y mass This is similar to t he 0.37 g/g dry mass hepatic samples from the Seahorse Key C ottonmouths. All hepatic levels outlined i n T able 2 4 except the snakes, are higher with highest levels found in American A lligators collected from lake Apopka in F lorida. Burger et al. ( 2000) collected hep atic samples from the American A lligator in Florida and found wet mass levels of 127 g/g. Studies that sampled hepatic tissue in fish ( Salaria basilisca Phocoenoides dalli ) found l evels of cadmium as high as 42.738 g/g wet mass and 2.9 g/g wet mass respectfully (Ikemoto et al. 2004, Messaoudi et al. 2009) The U S Department of
60 Health has set o ral consumption standards of cadmium for humans at 0.00 0 5 mg/kg d (ATSDR 2012) Cadmium (Cd) is a soft bluish white metal with a standard atomic weight of 112.411, and an atomic number of 48. Cadmium is chemically similar to zinc and mercury. Cadmium is usually found in zinc ores and is a by product of zinc production (Fowler et al. 2007) It has been used as a paint pigment and a corrosion resistant compound on steel. It is currently used in batteries, electroplating, and in some types of solar panels. Most cadmium is emitted into the atmosphere from smelting of metal, and the in cineration of coal, oil and garbage (Fowler et al. 2007) Cadmium is also used as an alloy in some metals and as a fertilizer (Fowler et al. 2007) Thus runoff from rain gutters and fertilized fields can bring cadmium into fresh and salt water environmen ts. Humans can acquire cadmium from smoking cigarettes (cadmium is absorbed much easier in the lungs, than through the digestive tract), drinking water, or eating foods contaminated with cadmium (ATSDR 2012) Cadmium acts primarily as a catalyst in form ing reactive oxygen species, it also increases lipid production, depletes glutathione and protein bound sulfhydryl groups (ATSDR 2012) The reduction of glutathione and protein bound sulfhydryl groups can have major consequences for organisms. These grou ps are important in many metabolic and biochemical reactions which include DNA synthesis and repair, protein synthesis, and amino acid transport. Thus clinical signs of cadmi um poisoning in humans, are flu like symptoms, respiratory and renal damage, bone loss, and muscle weakness (ATSDR 2012) Cadmium exposure has also been linked to certain types of cancers, and itai itai disease, which is a softening of the bones and kidney damage due
61 to cadmium poisoning (Fowler et al. 2007) Death from cadmium overd ose can occur with doses of 25 mg Cd/kg of cadmium iodide or 1,840 mg Cd/kg of cadmium chloride (ATSDR 2012) Acute oral LD 50 (Lethal dose, 50% kill) for rats and mice range from 100 to 300 mg/kg (ATSDR 2012) C admium levels ha ve been measured in snakes from a variety of environments (Hopkins et al. 1997, Campbell and Campbell 2001, Burger et al. 2005, 2007, Campbell et al. 2005, Jones and Holladay 2006) and Oros et al. ( 2009) found high levels of cadmium and fibrosarcomas in a single African Horned Viper (Chapter 1) but a s of this writing, cadmium has not been definitively found to have any effect in snakes. Selenium Selenium (Se) levels in the Florida C ottonmouths at Seahorse Key were not significantly higher than thos e in the mainland populations ( Figure 2 8 ) Selenium hepatic tissue samples from other snakes species are generally lower than the levels found in the cottonmouth snakes from this study (Table 2 5). One notable example is Hopkins et al. ( 1999) who found hepatic selenium levels in a population of Banded Water Snakes inhabiting a coal ash settling basin to have average hepatic tissue concentrations of 140 ppm dry mass. Level s reported in other vertebrates are much higher than levels reported in the cottonm ouths in this study ( Table 2 5 ). For example Bruehler and De Peyster (1999) collected hepatic selenium samples from the Brown P elican and found concentrations of 16 g/g dry mass These levels of selenium have been known to cause immune suppression in other birds (Bruehler and De Peyster 1999) These levels are much higher than the 6.51 g/g dry mass hepatic samples from the Seahorse Key C ottonmouths. All hepa tic levels outlined in Table 2 5 are higher with the highest levels found in a Bottle nosed Dolphin ( Tursiops truncates ) collected from
62 Italy. Concentrations in this mammal were 4330 g/g dry mass (Nigro and Leonzio 1996) Higher Se levels than the Seahorse Key cottonmouths have been reported in other reptiles. For example, Burger et al. ( 2000) collected hepatic samples from the American A lligator in Florida and found wet mass levels of 429 g/g. The U S Department of Health has set o ral consumption standards of selenium fo r humans at 0.005 mg/kg d (ATSDR 2012) Selenium (Se) is a metalloid (sometime s considered non metal) with a standard atomic weight of 78.96 and an atomic number of 34. Selenium is rarely found in its elemental state and is usually found in metal sulfide cores. Selenium can be toxic in large amounts but is an essential trace element used for cellular function. Selenium is used in the manufacture of solar cells and in alloys (Fowler et al. 2007) Most of the selenium found as a pollutant in the environme nt is a result of burning fossil fuels, especially coal. Because selenium is bound with sulfur metals, it can be a by product of burning coals that are high in sulfur (Fowler et al. 2007) Burned coal releases sulfur into the atmosphere from where it fal ls slowly into both fresh and salt water systems. Ocean concentrations of dissolved selenium can range from 0.5 nmol/L on the surface layers to 1.0 1.5 nmol/L in the deeper layers of the ocean (Fowler et al. 2007) Soil concentrations of selenium can be variable, which can lead to deficiencies in animal nutrition in some part s of the world and to selenium toxicity in other parts of the world (Fowler et al. 2007) Symptoms of a selenium overdose can include a garlic odor on the breath, hair loss, sloughi ng of nails, fatigue, and neurological damage in humans (Fowler et al. 2007) In snakes excess selenium can lead to a 30% rise in resting metabolic rate (Hopkins et al. 1999)
63 Selenium is an essential trace element that functions as a cofactor for the reduction of antioxidant enzymes including glutathione peroxidases (ATSDR 2003) Glutathione peroxidases are a family of enzymes that catalyze reactions that remove h ydrogen perox ide a nd organic hydroperoxides. Thus selenium plays an important role in preventing oxidative damage caused to the brain and endocrine tissues. Selenium is important in the functioning of the thyroid gland, and in cells that use thyroid hormones. Seleni um is a cofactor for three of the four known types of thyroid hormone deiodinases. Increased selenium levels have also been found to reduce the effects of mercury toxicity because of the high binding affinity between mercury and selenium (Berry and Ralsto n 2008) Selenium poisoning has occurred accidentally with few reported deaths, doses associated with the deaths are not known (ATSDR 2003) One example of selenium poisoning is a 15 year old female who ingested 22 mg Se/kg of sodium selenite. She survi ved primarily because she was forced to vomit, her symptoms were a garlic odor of the breath and diarrhea (ATSDR 2003) In nonhuman animals, LD 50 values for selenium vary depending on the species of selenium. S odium selenite LD 50 values were reported as 3.2 mg/kg for mice, whereas LD 50 values for L selenocystine were reported as 35.9 mg Se/kg for mice (ATSDR 2003) In snakes, a rise in resting metabolic rate has been correlated with high levels of selenium (Hopkins et al. 2004a) Mercury Mercury (Hg) lev els in the Florida cottonmouths at Seahorse Key were not significantly high er than those in the mainland populations (Figure 2 9 ) M ercury levels reported in the literature from other snakes are much lower than the hepatic levels reported in this study (T able 1 1 and 2 6). The concentrations reported here are over five times greater than hepatic tissue samples from other cottonmouth studies. For
64 example Rainwater et al. (2005) measured hepatic tissue concentrations in cottonmouth snakes from Texas and fo und average level s of mercury ranging from 0.408 1.19 g/g wet mass. As of this writing hepatic tissue concentration of mercury on Seahorse Key are the highest reported in the Florida Cottonmouth Snake and in the snake toxicological literature (Table 2 6) Levels reported in other vertebrates tend to be much higher than levels reported in the cottonmouths in this study ( Table 2 6). For example (Sepulveda et al. 1998) collected hepatic mercur y samples from the Double Crested C ormorant ( Phalacrocorax auritus ) and found concentrations of 48 g/g wet mass These levels are much higher than the 5.84 ppm wet mass hepatic samples from the Seahorse Key C ottonmouths Levels of mercury similar to the cottonmouth were found in the flight feather s of Osprey ( P andion haliaetus ) and hepatic samples of the Brown P elican with 5.25 g/g wet mass and 4.10 g/g wet mass respectfully (Blus et al. 1977, Cahill et al. 1998) All other hepa tic levels outlined in T able 2 6 are higher Hepatic mercury levels in the Bottl e nosed Dolphin ( Tursiops truncates ) collected from Italy were 174 g/g dry mass (Nigro and Leonzio 1996) Hepatic mercury levels in the Amer ican A lligator in Florida were 543 g/g wet mass the highest reported in T able 2 6 (Burger et al. 2000) The US Department of Health has set oral consumption standards of mercury for humans at 0.00 02 mg/kg d (ATSDR 1999) Mercury (Hg) is a transition metal with a standard atomic weight of 200.59 and an atomic number of 80. It occurs in deposits around the world as cinnabar (mercury sulfide), and the inhalation or ingestion of cinnabar is highly toxic. Mercury is used in a variety of applications; these include use in thermometers, barometers, manometers,
65 float valves, mercury switch es mercury replays, an d fluorescent lamps (ATSDR 1999) thermometers and sphygmomanometers. Mercury is still used in amalgam material for dental restoration in some countries Large amounts of mercury are released into the environment through the burning of foss il fuels (especially coal), the produc tion of chlorine, and in some locations the mining for gold (Fowler et al. 2007) Mercury toxicity comes primarily from the irreversible inhibition of selenium dependent enzymes and inactivation of S adenosyl methionine. S adenosyl methionine is necessary for the catecholamine catabolism by catechol o methyl transferase. Without catechol o methyl transferace cells are unable to degrade catecholamin es. This explains a common symptom of mercury poisoning; profus e sweating, tachycardia, increas ed salvation and hypertension (ATSDR 1999) Selenium dependent enzymes include the selenoenzymes and one of their numerous functions is to prevent and reverse the oxidative damage to the brain and endocrine organs (ATSDR 1999) If a person is exposed to mercury for long periods of time, the inhibition of the selenoenzymes could cause permanent oxidative damage to the brain. The high rate of oxygen consumption in the brain makes these tissues very susceptible to damage by selenoenzyme inhibition. Mercury levels in excess of selenium levels can lead to neurological damage, especially in young children and pregnant women. Prevention of mercury poisoning can be a chieved by the use of dietary selenium and this is suggested for women who consume large amounts of fish and are pregnant (ATSDR 1999, Berry and Ralston 2008)
66 When one compares the mercury levels to the selenium levels in the snakes and the fish at Seahorse Key (Table 2 1 and Table 2 2) and the studies outlined in Tables 2 5 and Table 2 6, one notices a reversal of the selenium mercury concentrations in the cott onmouths from this study For the other organisms outlined in Table 2 5 and 2 6, mercury levels are much lower than selenium levels. This is the same for the fish collected from around Seahorse Key. The cottonmouths from all locations seem to have highe r levels of mercury than selenium (Figure 2 16 ) Based on what is known about the mechanisms of mercury toxicity, you would expect the snakes to be experiencing mercury poisoning. This would include the irreversible inhibition of selenoenzymes and consequently neurological damage. These effects were not observed on the island, and this is an a rea that needs further research. As far as the author is aware, this is the first case of a reversal of mercury and selenium concentrations in snakes. Humans are exposed to mercury primarily through the consumption of fish. Mercury in fish flesh is primarily in the form of methylmercury and due to biomagnification, long lived fish and fish in higher trophic levels (i.e. sharks, tuna, swordfish, etc.) h ave much higher levels of methylmercury than short lived fish and fish from lower trophic levels (Boening 2000) Since fish bioaccumulate large amounts of mercury, the US Environmental Protection Agency (EPA) and the Food and Drug Administration (FDA) sug gest people, especially woman and children, avoid consuming large amounts of fish (ATSDR 1999) Mercury levels from the fish collected in this study are low (Table 2 2) compared to levels found in the literature (Table 2 6).
67 Clinical signs of mercury po isoning include peripheral neuropathy, skin discoloration, swelling and desquamation (ATSDR 1999) A n infamous example of mercury poisoning occurred in Minamata, Japan. In the 1950s, inhabitants of Minamata started noticing strange behavior in animals, an d people started to show signs of mercury poisoning. Fishing in the bay was banned and an investigation later found that a petrochemical company was dumping me r cury compounds into the bay. This pollution event killed around 5,000 people and up to 50,000 more were poisoned (ATSDR 1999) Size Differences between Mainland and Insular C ottonmouths Seahorse key has been historically known for producing unusually large snakes. Wharton (1969) documented many large snakes inhabiting the island. The large size o f the snakes on the island can be attributed to the abundance of food resources. S nakes take advantage of the abundant and assessable prey source of dead fish that are dropped by the nesting shorebirds (Lillywhite et al. 2002, Lillywhite and McCleary 2008 ) The abundance and accessibility of prey items on Seahorse Key, allow the cottonmouths to spend very little time foraging for food on the island (Wharton 1969) With less time and energy spent on searching for food, presumably more energy can go into g rowth (Wharton 1966) The mass of cottonmouths between locations was compared using a one way ANOVA statistical test. The results show that there is a significant difference between the size of the snakes on Seahorse key and the mainland populations. I nsofar as this research is concerned with Toxic Inorganic Contaminants and their presence in tissue, it becomes important to account for the size differences between these populations.
68 Differences in tissue concentrations could be attributed to sampling bias by the collectors. There seems to be little sampling bias in this study. Because the researchers gathered snakes by walking along the beach of Seahorse Key at night, this would mean that any snake out would have the opportunity to be captured. Researchers were also not targeting large snake s However there could be a size bias from one end of the island to the other. Personal observation by the author from numerous visits to Seahorse Key suggests that older, larger cottonmouths are more common on the western end of the island and younger, smaller cottonmouths are more common on the eastern end. This could suggest intraspecific competition is occurring on the island for the more productive foraging sites (Wharton, 1969) but much more likely is the distribution of snakes on the island could be due to dispersal by the young. It would appear that snakes sampled on Seahorse Key are on average much larger than snakes sampled on the mainland, although this has never been demonstrated quantitatively Observed Effects of TIC s on snakes living at Seahorse Key Effects from the consumption of Toxic Inorganic Contaminants are far ranging and can include behavioral and physiological changes in vertebrates (Fowler et al. 2007) These changes can include a c hange in metabolic rate, the presence of tumors, loss of reproductive ability, and/or death (Zakrzewski, 1997; Crosby, 1998) Since snakes have long been absent from the toxicology literature, we are only beginning to understand the types of effects that TICs Hopkins et al (1999) found snakes exposed to selenium had as much as a 30% increase in standard metabolic rate. This effect ma y not be occurring on Seahorse K ey. McCue and Lillywhite (2002) measured oxygen consumption in cottonmouths from
69 Seahorse K ey to determine their standard metabolic rate (SMR). They found no significant difference between this SMR and those from cottonmouths from central Florida (McCue and Lillywhite 2002) Although this study is over a decade o ld the cottonmouths on Seahorse K ey have been consuming the same prey for over 40 years (Wharton 1969) This information suggests the elevated levels of TI Cs in the snakes from Seahorse K ey are not altering the SM R of the snakes on the island. Other sign s that an organism is affected by a particular TIC can be neurological changes, behavioral changes, the presence of tumors, loss or reduction of reproductive ability, and death (Zakrzewski, 1997 ; Crosby, 1998) Insofar as most of these indicators of intoxi cation are better measured in a laboratory setting, the following is only a speculation on the effects of TICs on the population of insular cottonmouths at Seahorse key. Neurological changes, behavioral changes, and tumors are extremely difficult to obse rve in the natural environment. This is primarily because if an organism has any of these effects, their fitness could be lower ed and they would most likely be removed from the population either by predation or starvation. Also due to the low level, long term accumulation of TICs effects would likely only be seen in very large or very old individuals, unless there was a pollution event that exposed younger snakes to contaminants. Since very large or very old individuals are inherently rare, the odds of finding neurological changes, behavioral changes, and tumors are equally as rare. Thus none of these effects were found in the cottonmouth population on Seahorse Key. Reproductive effects from the presence of TICs can be difficult to identify in the field. In such a remote setting and with limited long term data, it would be difficult to
70 quantify the reproductive effects from TICs at Seahorse Key. Death is also difficult to quantify on the island without the use of tags or long term monitoring. Th us there could be reproductive effects or death from the presence of TICs on the island, but further data are needed Fish sampling and bio concentration Fish sampled on the island showed h igh levels of TICs, especially arsenic, s ele nium and c admium ( Figu res 2 3 through Figure 2 6) The fish sampled for metal analysis were all captured by the nesting water birds presumably from the Gulf of Mexico. This is likely the source of fish since all of the fish sampled in this study were marine fishes commonly fo und in the coastal waters around Seahorse Key (Reid 1954, Kells and Carpenter 2011) Th us th e high levels of arsenic, selenium, and cadmium found in the tissue of the fishes are likely representative of the water surrounding Seahorse Key. The location of Seahorse Key might contribute to these higher levels of contaminants found in the fish. Seahorse key is located between two large rivers; the Wa ccasassa R iver with a drainage area of 773 km 2 and th e Suwannee R iver with a drainage area of 3910 km 2 The l arge sizes of these river drainages, means that TICs from far upriver could eventually end up in the Seahorse Key area. This becomes important for the local fishing community as well as the wildlife that inhabit the Seahorse Key are a. Future research T he results from this study are a first step in fully understanding the presence and movement of TICs on Seahorse Key and in cottonmouth snakes living in Florida. The results provide background information on the amount of contaminants present in the snake s from four different locations in Florid a with an emphasis on Seahorse K ey.
71 To more fully understand the movement of conta minants in and around Seahorse K ey more research needs to be conducted. This includes sampling other organisms, i.e. invertebrates (aquatic and terrestrial), and possibly the birds on the island. Invertebrate samples from the island will be discussed i n further detail in Chapter 3. T oxicology research on Seahorse K ey needs to go past sampling background levels of contaminants and mov e into laboratory manipulations and long term monitoring programs. This includes capturing snakes, tagging them and taking periodic blood or tissue samples over an extended period of time. I t is not practical to remove a section of liver every time a res earcher needs a sample, but drawing blood can be a simple way of monitoring contamination levels in individual snakes. Other forms of sampling have been developed and they include sampling shed skin or scales, and clipping the tail (Burger et al. 2005, 20 07) Maternal transfer needs to be investigated in further detail because it could be a significant source of contamination for very young snakes. Overall the research regarding toxicology in cottonmouth snakes at Seahorse Key is in its infancy and furth er research is needed in order to better understand the behavior of TICs on Seahorse Key. Toxicological research involving snakes becomes important when one considers the unique environment on Seahorse Key, the proximity of the island to human habitation the designation of the island as a national wildlife refuge, and the trophic interaction the snakes have with nesting water birds. The cottonmouths on the island could provide a n insight into processes that transcend the local environment and affect pop ulations of birds and humans on a global level.
72 Cedar Key has a large aquaculture community, which includes Eastern O yster ( Crassostrea virginica ) and Northern Quahog C lam ( Mercenaria mercenaria ) farming, and fish harvesting for human consumption. Oyste r s, clams, and fish from Cedar Ke ys are being consumed in the local area as well as in restaurants and home s thousands of miles away. Essentially the collection and shipment of seafood from Cedar Key links the welfare of many people to a small location in the Gulf of Mexico. Thus, having information about the types and levels of TIC s in and around Seahorse Key can have a l asting i mpact on the welfare of numerous people. Conclusion In conclusion, snakes inhabiting Seahorse key are accumulating significan tly higher levels of a rsenic (As) than are snakes that are part of mainland populations in Florida. These differences appear to arise primarily from the large amount of fish carrion that cottonmouths consume on Seahorse Key. Therefore, the snakes on Seah orse Key are likely to be bio accumulating large amounts of TICs from the food they consume, which could also result in biomagnification in the associated food chain. This is concerning because of the ubiquity of the TICs in question, the locatio n of Seah orse key in the Cedar K ey National Wildlife Refuge, and the proximity of the island to human habitation (or coastal communities). The elevated levels of TICs in the liver and blood of cotto nmouths from Seahorse K ey could be a harbinger of high levels of t he same TICs in other primary consumers in the Gulf of Mexico, including protected bird species and humans.
73 Table 2 1 Tissue concentrations of 29 metals for the Florida Cottonmouth S nake ( Agkistrodon piscivorus conanti ) from four locations in Florida. Arithmetic mean s tandard d eviation are presented for each metal in the 29 metal array, Mass, Snout Vent Length (SVL), and Total length. Tissue concentrations are presented in ppm. Wet mass samples were converted to dry mass using the conversi on factor of 0.537. Thus Dry Mass = Wet Mass / 0.537. Locations are SHK = Seahorse Key, LS = Lower Suwannee Wildlife Refuge, BC = Big Cypress Swamp, and PP = Paynes Prairie State Preserve. Loc. Type N Mass g SVL mm Total mm Ag Al As SHK Liver Wet 20 1322.29 835.61 1001.62 182.46 1195.62 210.27 0.04 0.03 1.71 0.86 3.89 1.67 Dry 20 0.07 0.06 3.19 1.60 7.25 3.11 SHK Blood Wet 46 1206.87 761.57 956.44 202.62 1137.19 236.18 0.00 0.00 0.07 0.12 0.98 0.76 LS Liver Wet 7 297.31 93.63 648.40 39.49 781.80 48.82 0.01 0.01 1.75 2.53 0.35 0.51 Dry 7 0.02 0.02 3.26 4.72 0.64 0.94 LS Blood Wet 27 546.96 323.30 796.11 141.87 914.32 103.80 0.00 0.00 0.05 0.04 0.07 0.24 BC Liver Wet 2 753.55 119.75 839.50 110.5 1006.50 143.5 0.00 0.00 0.96 0.54 0.10 0.06 Dry 2 0.00 0.00 1.79 1.01 0.19 0.10 BC Blood Wet 7 499.24 60.72 747.14 47.25 894.71 56.61 0.00 0.00 0.17 0.20 0.28 0.62 PP Liver Wet 8 508.99 169.58 805.57 124.17 966.00 144.67 0.01 0.01 2.78 3.70 0.09 0.09 Dry 8 0.01 0.00 5.18 6.89 0.17 0.16 PP Blood Wet 32 631.72 208.03 812.88 94.60 974.55 114.71 0.00 0.00 0.06 0.05 0.07 0.23
74 Table 2 1 C ontinued. Loc. Type N B Ba Be Ca Cd Co SHK Liver Wet 20 0.11 0.04 0.04 0.03 0.00 0.00 107.81 39.45 0.20 0.11 0.07 0.03 Dry 20 0.20 0.08 0.08 0.06 0.00 0.00 200.77 73.46 0.37 0.21 0.12 0.06 SHK Blood Wet 46 0.06 0.03 0.01 0.01 0.00 0.00 109.66 13.13 0.00 0.00 0.01 0.00 LS Liver Wet 7 0.09 0.09 0.03 0.03 0.00 0.00 164.82 56.82 0.03 0.05 0.04 0.03 7 0.16 0.17 0.05 0.06 0.00 0.00 306.93 105.11 0.05 0.09 0.07 0.05 LS Blood Wet 27 0.03 0.02 0.01 0.02 0.00 0.00 117.23 61.46 0.00 0.00 0.01 0.00 BC Liver Mean 2 0.05 0.03 0.04 0.01 0.00 0.00 196.24 25.93 0.01 0.00 0.03 0.02 2 0.10 0.05 0.08 0.02 0.00 0.00 365.44 48.28 0.03 0.01 0.06 0.03 BC Blood Mean 7 0.04 0.02 0.03 0.04 0.00 0.00 140.33 94.03 0.00 0.01 0.01 0.00 PP Liver Mean 8 0.06 0.02 0.09 0.09 0.00 0.00 175.31 46.16 0.07 0.05 0.06 0.03 0.11 0.05 0.16 0.16 0.00 0.00 326.47 85.96 0.12 0.10 0.11 0.05 PP Blood Mean 32 0.03 0.02 0.01 0.01 0.00 0.00 120.52 40.39 0.00 0.00 0.02 0.01
75 Table 2 1 C ontinued. Loc. Type N Cr Cu Fe Hg K Li SHK Liver Wet 20 0.67 0.09 10.11 4.87 1326.77 533.35 5.84 4.55 2964.47 373.66 0.01 0.00 Dry 20 1.25 0.18 18.82 9.08 2470.71 993.20 10.88 8.47 5520.43 695.83 0.02 0.01 SHK Blood Wet 46 0.29 0.07 0.61 0.15 187.65 50.36 0.15 0.11 855.47 161.26 0.78 3.47 LS Liver Wet 7 0.65 0.65 8.73 8.88 794.44 631.81 2.45 1.98 1848.83 590.05 0.01 0.00 Dry 7 1.21 1.21 16.26 16.53 1479.40 1176.55 4.57 3.69 3442.88 1098.79 0.01 0.01 LS Blood Wet 27 0.26 0.06 0.50 0.10 186.21 66.02 0.31 0.37 866.30 289.89 1.27 5.62 BC Liver Wet 2 0.39 0.03 6.72 2.83 1180.38 1024.41 3.59 1.46 2372.23 144.18 0.01 0.00 Dry 2 0.72 0.05 12.51 5.26 2198.11 1907.65 6.68 2.71 4417.57 268.49 0.02 0.01 BC Blood Wet 7 0.24 0.06 0.52 0.13 160.96 42.95 0.12 0.09 809.23 207.19 0.01 0.01 PP Liver Wet 8 0.47 0.05 7.52 2.53 1007.57 960.71 0.79 1.18 2505.47 271.45 0.01 0.00 Dry 8 0.87 0.09 14.01 4.71 1876.30 1789.03 1.48 2.20 4665.67 505.50 0.02 0.01 PP Blood Wet 32 0.29 0.06 0.56 0.16 189.49 46.10 0.04 0.07 894.62 150.44 0.00 0.00
76 Table 2 1 C ontinued. Loc. Type N Mg Mn Mo Na Ni SHK L Wet 20 168.07 15.36 0.83 0.15 1.04 0.36 1224.44 249.24 0.03 0.01 Dry 20 312.98 28.61 1.55 0.28 1.94 0.67 2280.15 464.14 0.05 0.02 SHK B Wet 46 66.61 13.62 0.02 0.01 0.00 0.00 2802.05 267.62 0.01 0.02 LS Liver Wet 7 126.42 36.12 0.54 0.34 0.47 0.34 1886.93 504.46 0.03 0.04 Dry 7 235.42 67.26 1.01 0.64 0.87 0.63 3513.83 989.40 0.06 0.08 LS Blood Wet 27 73.87 21.92 0.02 0.01 0.00 0.00 2848.32 357.64 0.01 0.01 BC Liver Wet 2 157.85 28.44 0.46 0.21 0.42 0.25 1546.58 103.49 0.01 0.00 Dry 2 293.95 52.97 0.86 0.39 0.79 0.46 2880.03 192.73 0.02 0.00 BC Blood Wet 7 65.30 13.28 0.02 0.01 0.00 0.00 3093.92 303.00 0.03 0.04 PP Liver Wet 8 177.24 32.32 0.89 0.25 0.30 0.26 1338.88 242.34 0.02 0.02 Dry 8 330.05 60.18 1.66 0.47 0.56 0.48 2493.25 451.29 0.05 0.03 PP Blood Wet 32 77.72 11.83 0.02 0.01 0.00 0.00 2845.11 219.19 0.01 0.01
77 Table 2 1 C ontinued. Loc. Type N P Pb Sb Se Si Sn SHK Liver Wet 20 2868.76 220.36 0.01 0.01 0.07 0.07 3.50 2.16 31.38 5.57 0.02 0.01 Dry 20 5342.19 410.36 0.02 0.01 0.12 0.13 6.51 4.02 58.43 10.38 0.03 0.02 SHK Blood Wet 46 629.44 139.67 0.01 0.01 0.00 0.00 0.30 0.08 17.34 3.97 0.00 0.00 LS Liver Wet 7 1879.57 718.82 0.04 0.06 0.03 0.03 0.84 0.44 18.98 6.73 0.00 0.00 Dry 7 3500.13 1338.58 0.07 0.11 0.05 0.05 1.56 0.83 35.34 12.54 0.00 0.00 LS Blood Wet 27 640.50 250.43 0.00 0.00 0.00 0.00 0.19 0.07 17.35 3.71 0.00 0.00 BC Liver Wet 2 2532.64 248.46 0.00 0.00 0.02 0.01 0.73 0.39 17.55 2.70 0.00 0.00 Dry 2 4716.28 462.68 0.01 0.00 0.04 0.02 1.35 0.73 32.67 5.02 0.01 0.00 BC Blood Wet 7 566.00 175.59 0.01 0.01 0.00 0.00 0.19 0.04 15.66 3.32 0.00 0.00 PP Liver Wet 8 2608.62 329.32 0.03 0.03 0.04 0.03 0.71 0.35 22.47 5.75 0.00 0.00 Dry 8 4857.76 613.26 0.06 0.05 0.08 0.05 1.33 0.64 41.84 10.70 0.00 0.00 PP Blood Wet 32 660.90 90.92 0.01 0.01 0.00 0.00 0.19 0.06 19.06 4.64 0.00 0.00
78 Table 2 1 Continued. Loc. Type N Sr Tl V Zn SHK Liver Wet 20 0.16 0.09 0.00 0.00 1.92 1.93 35.21 15.61 Dry 20 0.31 0.16 0.00 0.00 3.57 3.57 65.57 29.07 SHK Blood Wet 46 0.18 0.06 0.00 0.00 0.03 0.01 6.84 1.68 LS Liver Wet 7 0.10 0.07 0.01 0.01 0.06 0.05 25.01 11.08 Dry 7 0.19 0.12 0.01 0.02 0.11 0.08 46.57 20.64 LS Blood Wet 27 0.06 0.05 0.00 0.00 0.02 0.01 7.28 2.03 BC Liver Wet 2 0.10 0.03 0.00 0.00 0.09 0.05 26.76 3.53 Dry 2 0.19 0.06 0.01 0.00 0.17 0.09 49.84 6.58 BC Blood Wet 7 0.10 0.06 0.00 0.00 0.02 0.01 6.60 1.24 PP Liver Wet 8 0.18 0.21 0.01 0.00 0.07 0.06 36.61 10.40 Dry 8 0.34 0.38 0.01 0.00 0.13 0.11 68.17 19.36 PP Blood Wet 32 0.05 0.05 0.00 0.00 0.03 0.01 7.59 1.67
79 Table 2 2 Hepatic tissue concentrations for four species of fish that were opportunistically collected under bird rookeries on Seahorse Key, FL Arithmetic means standard deviation is shown for each metal in the array All samples are wet mass samples, and units are ppm. M = Mass (g), L = Length (mm), and N = samp le size. Species are O.o. = Opisthonema oglinum A.f. = Ariopsis felis M.c. = Mugil cephalus and C.n. = Cynoscion nebulosus Species Type N M L Ag Al As B Ba O.o. Liver Wet 16 36.16 18.03 159.50 25.23 0.00 0.00 36.66 92.02 7.60 3.65 1.24 1.76 1.71 2.27 A.f. Liver Wet 1 0.00 1.10 7.00 0.49 0.15 M.c. Liver Wet 3 1.11 0.98 5.87 3.79 3.35 0.80 1.46 0.41 0.08 0.0 4 C.n. Liver Wet 1 0.00 0.26 1.79 0.39 0.01 Table 2 2 Continued. Species Type N Be Ca Cd Co Cr Cu O.o. Liver Wet 16 0.01 0.00 2121.50 1938.94 0.08 0.03 0.02 0.02 0.40 0.64 1.37 0.74 A.f. Liver Wet 1 0.00 170.05 0.14 0.05 2.49 480.41 M.c. Liver Wet 3 0.00 0.00 262.08 60.67 0.18 0.13 0.07 0.05 9.41 7.95 209.30 131.73 C.n. Liver Wet 1 0.00 251.88 0.13 0.03 3.50 250.23 Table 2 2 Continued. Species Type N Fe Hg K Li Mg Mn O.o. Liver Wet 16 97.11 162.61 0.12 0.07 5004.05 1580.44 0.10 0.18 625.32 345.55 10.40 22.27 A.f. Liver Wet 1 0.60 0.26 2984.66 0.04 297.50 0.37 M.c. Liver Wet 3 209.30 131.73 0.15 0.10 2771.46 181.65 0.05 0.01 358.22 79.96 1.06 0.54 C.n. Liver Wet 1 0.67 0.21 3288.92 0.03 364.16 0.16
80 Table 2 2 Continued. Species Type N Mo Na Ni P Pb Sb O.o. Liver Wet 16 0.46 0.66 1062.32 253.30 0.32 0.56 3348.42 1303.32 0.07 0.15 0.00 0.00 A.f. Liver Wet 1 2007.38 1.43 0.03 3561.04 0.14 0.00 M.c. Liver Wet 3 0.52 0.34 1427.57 150.16 0.11 0.03 2375.06 413.60 0.06 0.03 0.00 0.00 C.n. Liver Wet 1 1368.05 1.25 0.05 2685.22 0.04 0.00 Table 2 2 Continued. Species Type N Se Si Sn Sr Tl V Zn O.o. Liver Wet 16 2.20 1.13 31.72 14.46 0.01 0.01 9.84 12.88 0.00 0.00 1.27 2.65 21.48 9.55 A.f. Liver Wet 1 29.81 0.01 3.17 2.42 0.00 0.70 394.43 M.c. Liver Wet 3 2.42 1.48 36.72 3.40 0.01 0.01 1.38 0.04 0.00 0.00 0.28 0.14 37.44 11.64 C.n. Liver Wet 1 28.73 0.01 2.64 1.26 0.00 0.03 29.84
81 Table 2 3 Compilation of representative toxicology studies f r om around the world that measured arsenic levels in tissues of vertebrates Organisms from major vertebrate taxonomic groups are presented here for comparison to the Florida Cottonmouth ( Agkistrodon piscivorus conanti ) snake collected from Seahorse Key US Department of Health Minimal Ri sk Levels (MRL) for human consumption are also included including exposure times. All concentrations are arithmetic mean standard error. Species Year Wet/Dry Units Tissue Location Con c entration Source Comments Florida Cottonmouth ( Agkistrodon piscivorus conanti ) 2013 Wet ppm Liver Seahorse Key 3.89 Chapter 2 2013 Dry ppm Liver Seahorse Key 7.25 Chapter 2 Banded Water Snake ( Nerodia fasciata ) 1999 Dry ppm Liver South Carolina 132 30 Hopkins et al. (1999) Coal ash settling basin Northern Water Snake ( Nerodia sipedon ) 2007 Wet g/g Liver New Jersey 0.039 Burger et al. ( 2007) Raritan Canal 2007 Wet g/g Liver South Carolina 0.089 0.023 Burger et al. (2007) Savannah River site California Giant Garter Snake ( Thamnophis gigas ) 2009 Wet g/g Liver California 6.48 2.25 Wylie et al. (2009) Sacramento Valley US Department of Health consumption standards 2007 mg/kg d U.S.A. 0.005 ATSDR (2007) Oral exposure in humans (MRL) Acute exposure 2007 mg/kg d U.S.A. 0.0003 Chronic exposure Brown Pelican ( Pelecanus occidentalis ) 1994 Wet g/g Egg South Carolina 0.01 Wickliffe and Bickman (1998) Taken from a dredge spoil site Double crested Cormorant ( Phalacrocorax auritus ) No entries found Osprey ( Pandion haliaetus ) 1975 1982 Wet g/g Liver New Jersey 3.20 Wiemeyer et al. (1987) Dead or Moribund birds
82 Table 2 3 Continued. Species Year Wet/Dry Units Tissue Location Conc entration Source Comments Marine finfishes 1988 Wet mg/kg Liver 2 5 Eisler (1988) Range of marine fish liver concentrations Harp Seal ( Pagophilus groenlandicus ) 2001 Dry g/g Liver 7.68 Kunito et al. (2008) Alligator ( Alligator mississippiensis ) 2000 Wet g/g Liver Lake Apopka 41 Burger et al. (2000) Yearling Alligators from three lakes in Florida 2000 Wet g/g Liver Lake Orange and Lake Woodruff 8.89
83 Table 2 4 Compilation of repr esentative toxicology studies f r om around the world that measured cadmium levels in tissues of vertebrates Organisms from major vertebrate taxonomic groups are presented here for comparison to the Florida Cottonmouth ( Agkistrodon piscivorus conanti ) snake collected from Seahorse Key. US Department of Health Minimal Risk Levels (MRL) for human consumption are also included, including exposure times. A ll concentrations are arithmetic mean standard error. Species Year Wet/Dry Units Tissue Location Conc entration Source Comments Florida Cottonmouth ( Agkistrodon piscivorus conanti ) 2013 Wet ppm Liver Seahorse Key 0.20 Chapter 2 2013 Dry ppm Liver Seahorse Key 0.37 Chapter 2 Banded Water Snake (Nerodia fasciata) 1999 Dry p pm Liver South Carolina 0.5 0.15 Hopkins et al. (1999) Coal ash settling basin Northern Water Snake ( Nerodia sipedon ) 2007 Wet g/g Liver New Jersey 0.037 Burger et al. ( 2007) Raritan Canal 2005 Wet g/g Liver Tennessee 0.041 0.007 Campbell et al. (2005b) Little River California Giant Garter Snake (Thamnophis gigas) 2009 Wet g/g Liver California 0.02 0.003 Wylie et al. (2009) Sacramento Valley US Department of Health consumption standards 2012 mg/kg d 0.0005 ATSDR (2012) Oral exposure in humans (MRL) Acute exposure 2012 mg/kg d 0.0001 Chronic exposure Brown Pelican ( Pelecanus occidentalis ) 1999 Wet g/g Liver Salton Sea 0.58 Bruehler and De Peyster (1999) Double crested Cormorant ( Phalacrocorax auritus ) 1995 Wet g/g Liver Eastern Canada 0.2 0.8 Kuiken et al. (1999) Osprey ( Pandion haliaetus ) 1975 1982 Wet g/g Liver New Jersey <1 Wiemeyer et al. (1987) Dead or Moribund birds
84 Table 2 4 Continued Species Year Wet/Dry Units Tissue Location Concentration Source Comments Salaria basilisca 2009 Wet g/g Liver Tunisia 42.738 Messaoudi et al. (2009) Oxidative stress after a 28 day period ( Phocoenoides dalli ) 2004 Wet g/g Liver Japan 2.9 Ikemoto et al. (2004) Alligator ( Alligator mississippiensis ) 2000 Wet g/g Liver Lake Apopka, FL 127 Burger et al. (2000) Yearling Alligators from three lakes in Florida 2000 Wet g/g Liver Orange Lake and Lake Woodruff, FL 122
85 Table 2 5 Compilation of repr esentative toxicology studies f r om around the world that measured s elenium levels in tissues of vertebrates Organisms from major vertebrate taxonomic groups are presented here for comparison to the Florida Cottonmouth ( Agkistrodon pisc ivorus conanti ) snake collected from Seahorse Key. US Department of Health Minimal Risk Levels (MRL) for human consumption are also included, including exposure times. All concentrations are arithmetic mean standard error. Species Year Wet/Dry Units T issue Location Conc entration Source Comments Florida Cottonmouth ( Agkistrodon piscivorus conanti ) 2013 Wet ppm Liver Seahorse Key 3.50 Chapter 2 2013 Dry ppm Liver Seahorse Key 6.51 Chapter 2 Banded Water Snake (Nerodia fasciata) 1999 Dry ppm Liver South Carolina 140 30 Hopkins et al. (1999) Coal ash settling basin Northern Water Snake ( Nerodia sipedon ) 2007 Wet g/g Liver New Jersey 1.09 Burger et al. (2007) Raritan Canal 2005 Wet g/g Liver Tennessee 1.31 0.101 Campbell et al. (2005b) Little River California Giant Garter Snake (Thamnophis gigas) 2009 Wet g/g Liver California 0.773 0.07 Wylie et al. (2009) Sacramento Valley US Department of Health consumption standards 2003 mg/kg d 0.005 ATSDR (2003) Oral exposure in humans (MRL) Chronic exposure Brown Pelican ( Pelecanus occidentalis ) 1999 Wet g/g Liver Salton Sea 16 Bruehler and De Peyster (1999) Levels known to cause immune suppression in other birds. Double crested Cormorant ( Phalacrocorax auritus ) 1994 1997 Wet g/g Liver Southern Florida 19 Sepulveda et al. (1998) Sick and injured birds Osprey ( Pandion haliaetus ) 1998 Wet g/g Flight Feathers West coast of US 3.0 3.4 Cahill et al. (1998) Feathers were not dried
86 Table 2 5 Continued. Species Year Wet/Dry Units Tissue Location Conc entration Source Comments Cormorant ( Phalacrocorax carbo ) 1996 Dry g/g Liver Italy 55 Nigro and Leonzio (1996) Tuna ( Thunnus thynnus ) 1996 Dry g/g Liver Favignana Island 10 Nigro and Leonzio (1996) Bottle nosed Dolphin ( Tursiops truncates ) 1996 Dry g/g Liver Italy 4330 Nigro and Leonzio (1996) Alligator ( Alligator mississippiensis ) 2000 Wet g/g Liver Lake Apopka, FL 429 Burger et al. (2000) Yearling Alligators from three lakes in Florida 2000 Wet g/g Liver Orange Lake and Lake Woodruff 102
87 Table 2 6 Compilation of repr esentative toxicology studies f r om around the world that measured m ercury levels in tissues of vertebrates Organisms from major vertebrate taxonomic groups are presented here for comparison to the Florida Cottonmouth ( Agkistrodon piscivorus conanti ) snake collected from Seahorse Key. US Department of Health Minimal Risk Levels (MRL) for human consumption are also included, including exposure times. All concentrations are arithmetic mean stand ard error. Species Year Wet/Dry Units Tissue Location Conc entration Source Comments Florida Cottonmouth ( Agkistrodon piscivorus conanti ) 2013 Wet ppm Liver Seahorse Key 5.84 Chapter 2 2013 Dry ppm Liver Seahorse Key 10.88 Chapter 2 Cottonmouth ( Agkistrodon piscivorus ) 2005 Wet g/g Liver Texas 1.19 Rainwater et al. (2005) Central Creek 2005 Wet g/g Liver Texas 0.859 Rainwater et al. ( 2005) Goose Prairie Creek 2005 Wet g/g Liver Texas 0.408 Rainwater et al. ( 2005) Harrison Bayou Northern Water Snake ( Nerodia sipedon ) 2005 Wet g/g Liver Tennessee 0.750 0.076 Campbell et al. (2005b) Little River California Giant Garter Snake (Thamnophis gigas) 2009 Wet g/g Liver California 0.393 0.11 Wylie et al. (2009) Sacramento Valley Northern Water Snake ( Nerodia sipedon ) 2007 Wet g/g Liver New Jersey 1.09 Burger et al. ( 2007) Raritan Canal US Department of Health consumption standards 1999 mg/kg d Chronic 0.0002 ATSDR (1999) Oral exposure in humans (MRL) Brown Pelican ( Pelecanus occidentalis ) 1977 Wet g/g Liver Florida 4.10 Blus et al. (1977) Double crested Cormorant ( Phalacrocorax auritus ) 1994 1997 Wet g/g Liver Southern Florida 48 Sepulveda et al. (1998) Sick and injured birds
88 Table 2 6 Continued. Species Year Wet/Dry Units Tissue Location Conc entration Source Comments Osprey ( Pandion haliaetus ) 1998 Wet g/g Flight Feathers West coast of US 5.25 Cahill et al. (1998) Feathers were not dried Cormorant (Phalacrocorax carbo) 1996 Dry g/g Liver Italy 18 Nigro and Leonzio (1996) Tuna (Thunnus thynnus) 1996 Dry g/g Liver Favignana Island 1 Nigro and Leonzio (1996) Bottle nosed Dolphin (Tursiops truncates) 1996 Dry g/g Liver Italy 174 Nigro and Leonzio (1996) Alligator ( Alligator mississippiensis ) 2000 Wet g/g Liver Lake Apopka, FL 108 Burger et al. (2000) Yearling Alligators from three lakes in Florida 2000 Wet g/g Liver Orange Lake and Lake Woodruff, FL 543
89 Figure 2 1 Simplified food chain for Seahorse Key. Thickness of the arrows indicate magnitude of food source. Solid arrows indicate direct links and dashed arrows indicate links via an intermediate in this case nesting water birds.
90 Figure 2 2 Collection sites on Seahorse Key for the Florida Cottonmouth Snake ( Agkistrodon piscivorus conanti ) and fish during the years 2008 2010 Habitat types are listed by color. Base layer is a 2011 High resolution aerial photograph downloaded from www.labins.org Map made by the author u sing ArcGIS version 9.2.
91 Figure 2 3 Location of Florida C ottonmouth Snake ( Agkistrodon piscivorus conanti ) collection sites throughout Florida Snakes were collected during the years 2008 2010. Green arrow is Paynes Prairie Preserv e State Park, black arrow is the Lower Suwannee National Wildlife Refuge, red arrow is Seahorse Key, and the blue arrow is Big Cypress National Park. Map is downloaded free from the One World Nations Online, OWNO Project. Editor Klaus Kastle, 2013, www .nationsonline.org/oneworld/.
92 Figure 2 4 Snout vent length and total length as a function of mass for all Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled in this study. The data points include cottonmouth snakes from Paynes Prairie Preserve State park, Seahorse Key, Big Cyprus National Preserve, and Lower Suwannee National Wildlife Refuge. Regression equation s and R 2 value s are displayed on the figure. Figure 2 5 Snout vent length (SVL) as a function of mass for all Flo rida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) at each sampling location. The data points include cottonmouth snakes from Paynes Prairie Preserve State park (PP), Seahorse Key (SHK), Big Cyprus National Preserve (BC), and Lower Suwannee National Wildlife Refuge (LS). y = 0.2621x + 637.38 R = 0.8467 y = 0.3006x + 768.15 R = 0.852 0.00 200.00 400.00 600.00 800.00 1000.00 1200.00 1400.00 1600.00 1800.00 0.00 500.00 1000.00 1500.00 2000.00 2500.00 3000.00 Length (mm) Mass (g) Snout vent length and total length of Florida Cottonmouth snakes as a function of Mass SVL mm Total mm 0 200 400 600 800 1000 1200 1400 0 500 1000 1500 2000 2500 3000 Snout Vent Length SVL (mm) Mass (g) Snout Vent Length as a Function of Mass (g) SVL mm SHK SVL mm PP SVL mm LS SVL mm BC
93 Figure 2 6 Arsenic levels in two tissue types in Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) from four locations in Florida, and from fishes collected at Seahorse Key. Liver samples and blood samples were taken an d all samples are wet mass. Locations are: SHK = Seahorse Key, LS= Lower Suwannee River National Wildlife Refuge, BC= Big Cyprus National Preserve, and PP = Paynes Prairie Preserve State Park. The error bars are Standard Error calculated as equal to the Standard Deviation ov er the square root of n samples SHK Fish are liver samples collected from regurgitated fish dropped by nesting water birds on Seahorse Key. Blood samples were not taken for the fish, liver samples are wet mass, and measured in ppm. Fish samples contain the following fish species: O.o. = Opisthonema oglinum A.f. = Ariopsis felis M.c. = Mugil cephalus and C.n. = Cynoscion nebulosus Sample sizes are indicated above each column. Arsenic in SHK snake liver samples were significantly different (*) than mainland samples (p value <0.01). 20 7 2 8 21 46 27 7 32 0 1 2 3 4 5 6 7 8 SHK LS BC PP SHK-Fish Tissue Concentration (ppm) Location Arsenic (As) Liver Blood
94 Figure 2 7 Cadmium levels in two tissue types in Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) from four locations in Florida, and from fishes collected at Seahorse Key. Liver samples and blood samples were taken and all samples are wet mass. Locations are: SHK = Seahorse Key, LS= Lower Suwannee River National Wildlife Refuge, BC= Big Cyprus National Preserve, and PP = Paynes Prairie Preserve State Park. The error bars are Standard Error calculated as equal to the Standard Deviation over the square root of n samples. Tissue concentrations for liver and blood between the four locations were not significantly different. SHK Fish are liver samples collected from regurgita ted fish dropped by nesting water birds on Seahorse Key. Blood samples were not taken for the fish, liver samples are wet mass, and measured in ppm. Fish samples contain the following fish species: O.o. = Opisthonema oglinum A.f. = Ariopsis felis M.c. = Mugil cephalus and C.n. = Cynoscion nebulosus Sample sizes are indicated above each column. 20 7 2 8 21 46 27 7 32 0 0.05 0.1 0.15 0.2 0.25 SHK LS BC PP SHK-Fish Tissue Concentration (ppm) Location Cadmium (Cd) Liver Blood
95 Figure 2 8 Selenium levels in two tissue types in Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) from four locations in Florida, and from fishes collected at Seahorse Key. Liver samples and blood samples were taken and all samples are wet mass. Locations are: SHK = Seahorse Key, LS= Lower Suwannee River National Wildlife Refuge, BC= Big Cyprus Nat ional Preserve, and PP = Paynes Prairie Preserve State Park. The error bars are Standard Error calculated as equal to the Standard Deviation over the square root of n samples. Tissue concentrations for liver and blood between the four locations were not significantly different. SHK Fish are liver samples collected from regurgitated fish dropped by nesting water birds on Seahorse Key. Blood samples were not taken for the fish, liver samples are wet mass, and measured in ppm. Fish samples contain the fol lowing fish species: O.o. = Opisthonema oglinum A.f. = Ariopsis felis M.c. = Mugil cephalus and C.n. = Cynoscion nebulosus Sample sizes are indicated above each column. 20 7 2 8 21 46 27 7 32 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 SHK LS BC PP SHK-Fish Tissue Concentration (ppm) Location Selenium (Se) Liver Blood
96 Figure 2 9 Mercury levels in two tissue types in Florida C ottonmouth S nakes ( Agkistrodon piscivorus conanti ) from four locations in Florida and from fishes collected at Seahorse Key L iver samples and blood samples were taken and a ll samples are wet mass Locations are: SHK = Seahorse Key, LS= Lower Suwannee River National Wildlife Refuge, BC= Big Cyprus National Preserve, and PP = Paynes Prairie Preserve State Park. The error bars are Standard Error calculated as equal to the Standard Deviation over the s quare root of n samples. Tissue concentrations for liver and blood between the four locations were not significantly different. SHK Fish are liver samples collected from regurgitated fish dropped by nesting water birds on Seahorse Key. Blood samples were not taken for the fish, liver samples are wet mass, and measured in ppm. Fish samples contain the following fish species: O.o. = Opisthonema oglinum A.f. = Ariopsis felis M.c. = Mugil cephalus and C.n. = Cynoscion nebulosus Sample sizes are indicate d above each column. 20 7 2 8 21 46 27 7 32 0 1 2 3 4 5 6 7 8 SHK LS BC PP SHK-Fish Tissue Concentration (ppm) Location Mercury (Hg) Liver Blood
97 Figure 2 10 Arsenic tissue concentration as a f unction of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled on Seahorse Key. Two tissue types were sampled, blood (blue) and liver (red) all samples are wet mass Regression equation s and R 2 value s are displayed on the figure. Figure 2 11 Cadmium tissue concentration as a function of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled on Seahorse Key. Two tissue types were sampled, blood (blue) and liver (red) all samples are wet mass Regression equation s and R 2 value s are displayed on the figure. y = 0.0003x + 0.58 R = 0.1109 y = 0.0006x + 3.1302 R = 0.0831 0 2 4 6 8 10 12 0 500 1000 1500 2000 2500 3000 Tissue Concentration (ppm) Snake Mass (g) Arsenic (As) Seahorse Key Blood Liver y = 1E 09x + 9E 05 R = 4E 06 y = 3E 05x + 0.1567 R = 0.0592 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.00 500.00 1000.00 1500.00 2000.00 2500.00 3000.00 Tissue Concentration (ppm) Snake Mass (g) Cadmium (Cd) Seahorse Key Blood Liver Linear (Liver)
98 Figure 2 12 Selenium tissue concentration as a function of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled on Seahorse Key. Two tissue types were sampled, blood (blue) and liver (red) all samples are wet mass Regression equation s and R 2 value s are displayed on the figure. Figure 2 13 Mercury tissue concent ration as a function of mass for all of the Florida Cottonmouth Snakes ( Agkistrodon piscivorus conanti ) sampled on Seahorse Key. Two tissue types were sampled, blood (blue) and liver (red) all samples are wet mass Regression equation s and R 2 value s are displayed on the figure. y = 9E 06x + 0.291 R = 0.0066 y = 0.0014x + 1.6833 R = 0.2812 0 2 4 6 8 10 12 0 500 1000 1500 2000 2500 3000 Tissue Concentration (ppm) Snake Mass (g) Selenium (Se) Seahorse Key Blood Liver y = 9E 06x + 0.1341 R = 0.0044 y = 0.0026x + 2.4555 R = 0.2209 0 2 4 6 8 10 12 14 16 18 20 0 500 1000 1500 2000 2500 3000 Tissue Concentration (ppm) Snake Mass (g) Mercury (Hg) Seahorse Key Blood Liver
99 Figure 2 14 Fish length as a function of fish mass for all regurgitated fish and dropped fish gathered from under bird rookeries on Seahorse Key FL. Fish samples contain the following fish species: O.o. = Opisthonema oglinum A.f. = Ariopsis felis M.c. = Mugil cephalus and C.n. = Cynoscion nebulosus Sample size is fourteen fish. Regression equation and R 2 value are displayed on the figure. y = 1.3246x + 111.61 R = 0.8963 0 50 100 150 200 250 0 10 20 30 40 50 60 70 80 90 Fish Mass (g) Fish Length (mm) Fish Length as a Function of Fish Mass
100 Figure 2 1 5 Tissue concentration of arsenic (As), cadmium (Cd), mercury (Hg), and s elenium (Se) in liver samples taken from fish regurgitated by nesting waterbirds on Seahorse Key. Fish samples contain the following fish species: O.o. = Opisthonema oglinum A.f. = Ariopsis felis M.c. = Mugil cephalus and C.n. = Cynoscion nebulosu s All samples are wet mass Regression equations and R 2 values As and Se are displayed on the figure. The regression equation for Hg was y=0.0011x + 0.0549 and R 2 was 0.1475. The regression equation for Cd was y=0.00039x + 0.018 and R 2 was 0.0801. y = 0.0318x + 12.673 R = 0.0485 y = 0.0023x + 2.5664 R = 0.0027 0.000 2.000 4.000 6.000 8.000 10.000 12.000 14.000 16.000 125 145 165 185 205 225 245 Tissue Concentration (ppm) Length (mm) Tissue Concentration as a Function of Fish Length As Cd Hg Se
101 Figure 2 1 6 Mercury and s elenium concentrations for the Florida C ottonmouth S nake ( Agkistrodon piscivorus conanti ) f rom four locations in Florida. Locations are: SHK = Seahorse Key, LS= Lower Suwannee River National Wildlife Refuge, BC= Big Cyprus National Preserve and PP = Paynes Prairie Preserve State Park. All tissue samples are wet mass and ppm 0 1 2 3 4 5 6 7 SHK LS BC PP Tissue Concentration (ppm) Location Mercury (Hg) Selenium (Se) Hg-Liver Se-Liver Hg-Blood Se-Blood
102 Figure 2 1 7 Correlation between combined mer cury and selenium concentrations for the Florida Cottonmouth Snake ( Agkistrodon piscivorus conanti ) from four locations in Florida and two tissue types Locations are: SHK = Seahorse Key, LS= Lower Suwannee River National Wildlife Refuge, BC= Big Cyprus N ational Preserve, and PP = Paynes Prairie Preserve State Park. Tissue types are blood and liver samples. All samples are wet mass and ppm. Regression equation and R 2 value are displayed on the figure y = 0.4962x + 0.1816 R = 0.9186 0.00 2.00 4.00 6.00 8.00 10.00 12.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 Selenium Concentration (ppm) Mercury Concentration (ppm) Selenium Mercury Concentrations all snakes
103 CHAPTER 3 METAL CONCENTRATIONS IN MARINE AND TERRESTRIAL MOLLUSCS FROM SEAHORSE KEY, FLORIDA Introduction Environmental pollution is one of the most important issues of our time (Crosby 1998) Since the industrial revolution the effects humans have had on the l ands cape and the environment have been extreme. The burning of fossil fuels, the increase in mining to access natural resources, and an increas ing human population have all contribute d to e nvironmental pollution. Th e human population has reached 7 billion wi th models predicting 10 billion by 2050 (Bloom 2011) This drives the need to understand and monitor types and amounts of pollut ants emitted in to the environment. In general, pollution comes from two sources; point and non point. Point source pollution comes from an easily identifiable source (Fowler et al. 2007) For example, finding oil sludge in a bay after a tanker collides with a rock or oil sludge in the Gulf of Mexico after the Horizon disaster. Because the source is easi ly identified, steps ca n be taken to quickly fix the problem. Non point pollution does not come from an easily identifiable source (Fowler et al. 2007) making it more difficult to identify. It is primarily the result of small events that may combine into a large pollution pro blem. For example, the ease of the movement of contaminants after a storm can lead to an increase of non point source pollution such as storm water containing large amounts of r esidential lawn fertilizer. Metals in particular are of high concern because of their ability to bioconcentrate in tissue, and bioaccumulate through the food chain (Zakrzewski 1997, Crosby 1998, Allinger et al. 2001, Fowler et al. 2007) Most metals are found in low concentrations i n the environment and high levels are usually the result of anthropogenic actions. These include, but are not limited to, mining, the burning of fossil fuels, and industrialization (Crosby 1998) Metals also have the ability to stay in the environment fo r long periods of time. For example, Lefcort et al.
104 ( 2010 ) found even after 75 years the macro invertebrate community in a high mountain stre am in northern Idaho was still a ffected by high levels of cadmium (Cd), and zinc (Zn). They found correlations be tween a smaller percentage of Plecoptera (stoneflies) and higher levels of cadmium and zinc (Lefcort et al. 2010) This study illustrates the need to have long term monitoring programs to study the effects of meta l on animal communities Because the pot ential risk of pollution is likely to increase there is an urgent need to monitor contaminant levels in the environment. Molluscs, specifically gastropods (Class Gastropoda) and bivalves (Cla ss Bivalvia) appear to be good indicators of environmental con tamination (Berger and Dallinger 1993, el Sikaily et al. 2004, Rittschof and McClellan Green 2005) Gastropods and bivalves are found in aquatic and terrestrial environments and on every continent except Antarctica (Rittschof and McClellan Green 2005) I n addition, their limited home range, constant contact with soil/sediment and their short life span make them valuable for chronologically assessing heavy metal contamination. M olluscs can be used for ecotoxicological assessment of the environment in se veral ways Viard et al. ( 2004 ) used soil deposits, autochthonous plants and the vis c era of snails Helix aspersa to quantify the levels of heavy metal contamination near highways in France. Initiated in the 1970s by Edward Goldberg (Goldberg 1986) the United States National Oceanic and Atmospheric Association (NOAA) and the Center for Coastal Monitoring and Assessment (CCMA) runs and maintains an online database called Mussel Watch Mussel Watch is a program that collects molluscs from coastal an d inland sites throughout the United States and analyzes them for a wide range of contaminants (Apeti et al. 2012) In another study Blanvillain et al. ( 2007 ) used marsh Periwinkles ( Littoraria irrorata ) to measure environmental levels of mercury in an e stuarine system in South Carolina, and Georgia. They found methyl mercury concentrations in the snails were dependent on size.
105 The soft tissue in molluscs easily absorbs trace elements which makes molluscs efficient accumulators of trace elements (Dalli nger 1994) Metal uptake in molluscs occurs though multiple pathways including facilitated diffusion, active transport, endocytosis, and absorption through the digestive epithelium (Dallinger 1994, Marigmez et al. 2002) Most molluscs have specialized s torage mechanisms for the compartmentalization of contaminants (Berger and Dallinger 1993, Viard et al. 2004) These include the binding of the metals to metallothioneins, incorporation into lysosomes, and the formation of mineralized granules (Dallinger 1994, Marigmez et al. 2002) The principle objective of this study was to examine the tissue concentrations of molluscs living on and around Sea horse Key. The hypotheses and o bjectives were: Hypothesis 1: Marine snails have higher levels of metals tha n terrestrial snails living on Seahorse Key. Hypothesis 2: Marine bivalves have higher metal levels than marine and terrestrial snails. Objective: To determine the concentration of metals in tissues collected from marine bivalves and gastropods collected from the Gulf of Mexico, and terrestrial gastropods collected from Seahorse Key. Hypothesis 3: The soft tissue of bivalves i.e. viscera, concentrate mor e metals and at higher levels than the hard tissues i.e. the shell. Objective: To determine the concentration of metals taken from the shells of bivalves and from the viscera of bivalves living around Seahorse Key. Hypothesis 4: Tissue metal levels from Seahorse Key molluscs are lower than Environmental Protection Agency ( EPA ) stand ards. Objective: To compare the tissue concentrations from molluscs to standards published by the EPA. Methods and Materials Sample C ollection Invertebrates were collected from Seahorse Key and the surrounding area during the years 2008 2012. Thirteen Northern Quahog clams ( Mercenaria mercenaria ) with average
106 length of 48.11 mm ( 5.16 SD ) were donated in 2010 by a local clam farm er. T hese clams were raised on a local Cedar Key clam farm lease within a 1 5 km radius of Cedar Key ( Figure 3 1 ) Twelve M arsh Periwinkle s ( Littoraria irrorata ) with an average length of 21.0 mm ( 1.61 S D ) were collected from the marshes around Seahorse K ey ( Figure 3 1 ). Fourteen Southern Flatcoil ( Polygyra cereolus ) snails with an average width of 6.1 mm ( 0.0 9 S D ) were c ollected from the coastal h ammock on Seahorse Key ( Figure 3 1 ) Evening surveys were conducted during fall 2011 and spring 2012 to collect Southern Flatcoil and Marsh Periwinkle snails. The search for terrestrial snails involved walking through the hamm ock, overturning logs, lifting palm fronds, and visually searching the understory for snails. Terrestrial searches typically lasted 90 120 minutes. The search for marine snails ( Marsh Periwinkles ) involved walking through the saltmarsh and collecting sna ils directly from the Smooth Cordgrass ( Spartina alte r niflora ). Marsh Periwinkles were collected from various locations around Seahorse Key Captured snails were placed in a labeled Ziploc bag and locations were logged on a GPS unit (Garmin e trex). Sna ils were taken back to the laboratory and identified to species with help from Dr. Fred Thompson (Curator of Non m arine Malacology, University of Florida) Once in the laboratory, snails were placed in the freezer ( 18 C) for further preparation and analysis. From August 2011 to June 2012 night searches were made on Seahorse Key to collect terrestrial snails for comparison with marine snails. The targeted species were the Rosy Wolf Snail ( Euglandina rosea ) which is a terres trial snail that reaches sizes of up to 60 or 70mm (Capinera and White 2011) and the Marsh Periwinkle ( Littoraria irrorata ) which is a marine snail that reaches sizes up to 30mm (Silliman and Bertness 2002) The large size of the Rosy Wolf Snail made it an ideal candidate for dissection and comparison of soft versus shell tissue. Other terrestrial snails were also sought during this time and collected via the
107 sampling methods outlined above. During this time the Cedar Key area experienced a severe droug ht (Smith 2012) which made snail collecting on Seahorse Key difficult. Searches were conducted multiple times during this period and no Rosy Wolf Snails were found on Seahorse Key for analysis. On one collecting event fourteen Southern Flatcoil ( Polygyr a cereolus ) Snails were found. These were collected and used in the analysis in this report. Unfortunately the sample size for the terrestrial snails was unavoidably small. However the Marsh Periwinkle remained abundant and samples for this snail were e asily acquired from the locations shown in Figure 3 1 and Figure 2 2. The Southern Flatcoil is a much smaller snail, and although we were able to collect fourteen snails, only two samples could be acquired. To obtain a sample large enough for analysis (> 1 00mg), seven Southern Flatcoil S nails had to be combined for one sample. This was an unfortunate and unavoidable result from the field season of collecting snails at Seahorse Key. Sample Preparation In summer 2012 s amples were removed from the freezer and allowed to thaw to room temperature (25 C ). Excess water was removed from the samples with a paper towel Viscera and half of the shell of each clam were removed and placed on separate drying trays Whole periwinkle snails and flatcoil snails were placed on individual trays. All samples were weighed (initial wet mass), then dried in an oven set at 68 C. Samples were weighed daily and dry mass recorded when the samples did not change by more than 0.01 g during a 24 hour period. Dried samples wer e homogenized using a mortar and pestle contain ing ten mL of liquid nitrogen. Samples were crushed by first tapping the sam ple to break the shell, then ground vigorously to obtain approximately 100 mg of material with a consistency similar to baking flour. A dditional liquid nitrogen was added as needed. R emaining fragments of snails and clams were placed in separate Ziploc bags, labeled and stored for future use. S ample s
108 weighing under 100 mg were combined with samples from snails of the same species to make up the difference. All snail ( Southern F latcoil and Marsh P eriwinkle) samples were homogenizations of the entire snail. For each Northern Q uahog samples from the soft tissue (viscera mass) and one half of th e shell were used in th e analyse s. Once 100 mg of material was collected from each sample, it was placed in a unique 1.5 mL cryo genic vial ( Nalgene, Rochester, NY ) and shipped cold to the Utah Veterinary Diagnostic Laboratory (UVDL) in Logan, Utah, for analysis. Analysis of Sam ples For elemental analysis t he samples were digested in trace mineral grade nitric acid under heat. The digests were then diluted with ultra pure water to a final nitric acid content of 5%, which provided a matrix match for the analytical standards. The prepared samples were analyzed by Inductively Coupled Plasma Mass Spectrometry (ICP MS) and assessed against concentration curves of known mineral standards (personal communication UVDL 2013) Standard curves and quality contr ol samples were analyzed eve ry five samples. Twenty nine elements were measured in parts per million (ppm) for each sample with minimum detectable levels of 0.001 ppm. Data from Mussel Watch Trace element concentration data were obtained from the Mussel Watch program database on the Center for Coastal Monitoring and Assessment (CCMA) website (NOAA 2013) This program samples molluscs and sediments around the country to quantify the amount of con taminants in the United States c oastal and inland waters. The closest collection cit e to Cedar Key was Black Point (lat: 29.206104, long: 83.070674) a small outcropping located approximately ten km north of Cedar Key (Figure 3 1) Data available from the CCMA website were sediment samples and Eastern O yster ( Crassostrea virginica )
109 soft tissue samples. Samples were reported in dry mass and ppm These data were only available for the samples years 1986 2007. For a complete description of sampling methods, refer to NOAA (1986 ) Statistical Analysis Mean, standard deviation, and standar d error were calculated for each metal and each sample. Metals were chosen for statistical analysis based on the following criteria: persistence to remain in the environment, toxicity, and presence i n the toxicological literature. Metals were chosen in t his manner to facilit ate comparison to the toxicological literature and Chapter 2 All metals in the twenty nin e metal array are presented in T able 3 1 T he metals chosen for further statistical analysis were arsenic (As), selenium (Se), mercury (Hg), an d cadmium (Cd) (Chapter 2). The UVDL could only detect metals with concentration greater than 0.001ppm. A ll samples with concentrations <0.001 were changed to 0. Shell and viscera samples for the Northern Quahog were compared using a one way ANOVA statistical test (McKillup 2005) Whole snail homogenizations of the Marsh P eriwinkle and the Southern F latcoil were compared using a one way ANOVA statistical test. Mean ti ssue concentrations were compared between the viscera of the Northern Quahog, the whole body homogenization of the Marsh P eriwinkle snails and the who le body homogenizations of the S outhern F latcoil using a one way ANOVA statistical test. If a statistical ly significant difference was found between three or more samples, then a post hoc Tukey HSD (Honestly Significant Difference) te st was completed in order to do a pair wise comparison between sample means. All statistical tests were completed in Microsoft excel, except the Tukey HSD analysis which was completed in the R statistical program (R Development Core Team 2011) All statistical tests were considered significant with p values 0.05.
110 Wet Mass and Dry Mass Conversions T issue concentrations can be reported as either wet or dry mass (Zakrzewski 1997, Crosby 1998) Sampl es were weighed initially (wet mass ) and every twenty four hours until the mass did not change. This final mass was considered the dry mass and was used to calculate the mois ture co ntent of each sample [(WM DM)/WM = MC]. Where W M = Wet Mass DW = Dry Mass, and MC = Moisture Content. A conversion f actor (CF) was calculated by subtracting MC from one [1 MC = CF]. The wet mass tissue concentration (WM C) was converted to dry mass tissue concentration (DM C) u sing the following equation: DMC = WM C/CF. CF was calculated for each sample and is presented in addition to the calculated wet mass tissue concentrations Results The a rithmetic mean and standard deviation for each sample and each metal are presented in Table 3 1. Wet and dry masses are presented in Table 3 1 The conversion factor used to calculate wet mass for each species is included in Table 3 1. Statistical analysis was completed on arsenic (As), mercury (Hg), selen ium (Se), and c admium (Cd). Average tissue concentrations of As, Hg, Se, and Cd for the Marsh Periwinkle, the Southern Flatcoil, and the Northern Q uahog are presented in Figure 3 1 and 3 2 Standard error was used because of the difference in sample size s and is presente d as error bars on the graphs Thirteen Northern Q uahog s ( viscera and shell), twelve periwinkle s ( whole body homogenizations), and two S outhern F latcoil (seven whole snail homogenizations) samples were analyzed Viscera tissue contained significantly higher levels of As, Hg, Se, and Cd, than shell tissue taken from the same Northern Quahog with arithmetic mean SD viscera samples of 17.99 3.25 0.16 0.04 4.00 0.54 and 0.66 0.19 ppm dry mass respectful ly and shell samples of 0.00 0.00 0.00 0.00 0.15 0.13 and 0.01 0.01 ppm dry mass respectfully
111 (Table 3 1) Whole body homogenizations taken from the Marsh Periwinkle and Southern Flatcoil snails were not significantly different for As, Hg, and Se. The Southern Flatcoil samples contained significantly higher levels of Cd than the Marsh Periwinkle with arithmetic mean SD levels of 0.93 0.17 ppm dry mass, and 0.22 0.06 ppm dry mass, respectfully (Table 3 1) Mean tissue concentrations betw een the viscera of the Northern Quahog, the whole body homogenization of the periwinkle snails and the whole body homogenizations of the southern flatcoil were significantly different for As (17.99 3.25 0.04 0.10 and 0.00 0.00 ppm dry mass, respect fully), Hg (0.16 0.04 0.01 0.00 and 0.01 0.00 ppm dry mass, respectfully), and Se (4.00 0.54 0.35 0.10 and 0.25 0.11 ppm dry mass, respectfully), but not for Cd (0.66 0.19 0.22 0.06 and 0.93 0.17 ppm dry mass, respectfully) (Table 3 1) Mercury and selenium concentration for all three mollusc species was highly correlated with R 2 values of 0.9691 (Figure 3 6). Trace metal concentration from the Mussel Watch program for the Eastern O yster and sediment samples were retrieved from the CCMA website for Black Point Florida. The s e data are summarized in Figure 3 3 and Figure 3 4 respectfully. D ata for the Eastern Oyster show a trend for arsenic from ten ppm (dry mass) in 1986, increasing to thirty ppm in 1990, and then oscill a ting around fifteen ppm until it reaches its 2007 level of nine ppm. Sediment samples taken during the same time period but less frequently show sediment levels of arsenic around 5.6 ppm (dry mass) in 1986, then peaking at eleven ppm in 1987 and falling t o 2.1 ppm in 2007. Cd and Se follow similar trends as As with peaks in oyster tissue of four ppm (dry mass) occurring in 1989 then oscillating around three ppm until 2007. Hg concentration stayed near zero ppm for the entire sample period (1986 2007). Sediment sampled during the sample time period show Se around one ppm (dry mass) in 1987 and 1997 and near zero ppm for 1986 and 2007. Hg and Cd concentration are near zero ppm for
112 the entire sampling period (1986 2007). Tissue samples from the Eastern O yster were not compared to tissue samples from the Northern Quahog because the Mussel Watch data had a sample size of one. Discussion Significant ly higher levels of Cd were found in the terrestrial gastropod when compared to the marine gastropod, while tissue concentrations of As, Hg, and Se were not significantly different. Significant differences were found between the marine bivalves and terrestrial gastropods for As, Hg, and S e with the highest levels of As, Hg, and Se found in the viscera l tissue of the Northern Quahog. Significantly higher levels of Cd were found in the Southern Flatcoil when compared to the Marsh Periwinkle and Northern Quahog Significantly higher levels of As, Hg, Se, and Cd were found in the viscera of the Northern Quahog when compared to the shell of the Northern Quahog. The differences observed in the mollusc samples seem to suggest that cadmium is more prevalent in terrestrial species, and arsenic is more prevalent in marine species. However these differences could be attributed to different metal species, inter specific differences in the assimilation of metals by molluscs, or different feeding strategies. The two species of gastropods used in th is study have very different habitat requirements and life history characteristics. The Southern F latcoil is a terrestrial species that feeds on plants, and has a n average size of 8mm in diameter with a range of 7 18mm (Capinera and White 2011) The Sout hern F latcoil is found throughout Florida and the southeastern United States and typically inhabits soil, detritus, and dead wood (Capinera and White 2011) The Southern Flatcoil snail is usually found i n large number s with clusters having many hundreds o f individuals (Capinera and White 2011) The Marsh P eriwinkle is a marine snail inhabiting tidal saltwater marshes in the southeastern United States. The n umber s of periwinkles in a salt marsh can be quite large
113 and can be the determining factor for mar sh vegetation density (Bertness and Silliman 2008) Marsh P eriwinkles are fungivores, and grow and maintain a fungus colony on salt marsh vegetation by scra ping the vegetation, then maintaining the colony (Silliman and Newell 2003) Given these two life h istories, we would expect the terrestrial snail to have higher levels of cadmium due to the elemen t s natural occurrence in soils. The S outhern F latcoil is in contact with the soil much longer than the Marsh P eriwinkle and has significantly higher levels of Cd (0.93 ppm dry mass, 0.22 ppm dry mass, respectfully) On the other hand, t he Marsh P eriwinkle is in contact with the Gulf of Mexico for a greater amount of time than the flatcoil and thus we would expect, given t he results from Chapter 2, that this species would have higher levels of arsenic, which it does (0.04 ppm dry mass, 0.00 ppm dry mass, respectfully) Selenium and mercury levels are similar between the Southern Flatcoil and the M arsh Periwinkle with sele nium levels around 0.25 and 0.35 ppm dry mass, respectfully and mercury levels of 0.01 ppm dry mass for each. Higher levels of selenium than mercury levels are reported and this is comparable to what is found in other organisms, i.e. vertebrates (Table 2 5 and Table 3 1 ). The tissue levels of each contaminant (As, Se, Cd, and Hg) are much lower than those reported in other studies For example, Berger and Dallinger ( 1993b) collected Arianta arbustorum from multiple sample sites in Austria and found cadmiu m levels of 6.5 g/g dry mass T he samples collected from Seahorse Key are much lower ( cadmium levels o f 0.9 ppm in Southern F latcoil and 0.22 ppm in Marsh P eriwinkle ) Other studies report tissue concentrations of As, Hg, and Se with dry mass soft tissu e concentrations ranging from 2 5 g/g (Rittschof and McClellan Green 2005) These ranges are much higher than the levels found in the Southern Flatcoil and Marsh Periwinkle from Seahorse Key with As, Hg, and Se levels less than 0.3 g/g dry mass.
114 Storag e locations in gastropods depend on the type of contaminant accumulated. For exam ple, in a recent review of subcellular storage locations of metals in molluscs, Marigmez et al. ( 2002) found that Littorina littorea would store Cd in the hemocytes Hg in the digestive cells, and As and Se in the digestive gland. He also found that some metals are stored in multiple locations. For example, Littorina littorea would also store Cd in the gill epithelial cells and the digestive cells (Marigmez et al. 2002) Two species of bivalves collected from the Gulf of Mexico were analyzed for metal concentration. These were the Northern Q uahog ( Mercinaria mercinaria ) and the Eastern O yster ( Crassostrea virginica ). D ata for the Eastern Oyster were collected f r o m the National (NOAA 2013) The E astern O yster was collected at a site roughly ten k ilometers north of Cedar Key (Figure 3 1) Alth ough the oysters and the clams were not collected at Seahorse Key, they were collected in close proximity to the island and most likely encounter similar conditions. Since the linear distance from the collection site and Seahorse Key is relatively short, bivalves can provide useful information about the con centration of metals in the waters arou nd Seahorse Key and Cedar Key. Both of these species are harvested extensively for human consumption (Rittschof and McClellan Green 2005) The primary source of income for the residents of Cedar Key is clam farming and oyster collection (McCarthy 2006) Concentrations of Hg, Se, and Cd in the molluscs and in the soils are comparable to levels found in other countries and similar habitats (Elder and Mattraw 1984, De Wolf et al. 2001, Viard et al. 2004, Rittschof an d McClellan Green 2005, Notten et al. 2006) For example, samples taken from the mussel watch program along the coast of the Gulf of Mexico show similar levels of As (~15 ppm dry mass) Hg (~0.1 ppm dry mass) Se (~3.5 ppm dry mass) and Cd (~0.51 ppm dry mass) (NOAA 2013) Hg levels were much lower in Cedar Key when compared to other sites in Florida, especially sites located near large cities (Apeti
115 et al. 2012) Data from the mussel watch program showed concentrations of metal in the Gulf of Mexico is metals around cities (Apeti et al. 2012) Data from the oysters provide long term information regarding contaminants. The mussel watch program started in the early 1980s and the earliest sample we have from the site near Cedar Key is 1986. Thus looking at Figure 3 4 one notices an interesting trend in the data. Tissue concentrations from the Eastern O yster rises to almost 30 ppm in 1987, the n oscillates between 10 ppm and 20 ppm for the next two decades. The overall trend of the data is decreasing with the most recent (2007) data showing a concentration of 10 ppm. This trend seems to follow that seen in s ediment samples from the same t ime period. The data indicate a consid erable amount of arsenic was present in the water surrounding Cedar Key in the 1980 s and since then has slowly decreased. Absorption of metals in bivalves can occur via facilitated diffusion, active transport, or endocytosis. Storage locations depend o n the type of contaminant accumulated. For example, in a recent review of subcellular storage locations of metals in molluscs, Marigmez et al. ( 2002) found that Mercinaria mercinaria would store Cd in the gills, Hg in the foot, and As and Se in the digestive gland. As in the Gastropods, the same metals are also stored in multiple locations (Marigmez et al. 2002) A rsenic in Florida Waters Looking at the data from Chapter 3 and Chap ter 2 a general trend emerges; a rsenic levels are significantly higher for organisms living in the Gulf of Mexico, or who receive their prey source from the Gulf of Mexico when compared to organisms who live or forage elsewhere Chen et al. ( 2001 ) in an effort to classify background concentrations of Arsenic in the state of Florida, did a meta analysis of soil samples taken from the Florida Soil Survey Program and fou nd that soils on average have a rsenic concentrations of around 0.5 ppm dry
116 mass. To qua ntify Arsenic levels locally, we would need long term monitoring of soil sediment or water samples to determine the local trend. Luckily, the Mussel Watch program also collected soil sediment samples since 1986, although not as frequently as the bivalve s ampling (Figure 3 5 ). A rsenic levels from Black Point, FL have fluctuated over time. At the start of the data set, arsenic was near 6ppm, rose to 11ppm in 1987, and has since continued to drop to 2ppm in th e last sample year ( 2007 ) The s e data provide a clear baseline to compare the levels in the molluscs and other organ isms living near Seahorse Key. Metal Contaminants and the Cedar Key Community Arsenic levels are high er in the Cedar Key area when compared to other locations in the Gulf of Mexico (Wilson et al. 1992, Chen et al. 2001, NOAA 2013) This trend has been observed since the start of the Mussel Watch program in 1986. Reasons for these elevated levels could be agricultural runoff decomposing docks, or the higher levels found in Florida soils (Chen et al. 2001) Regardless, t he levels of arsenic in the soft tissue samples (viscera of bivalves) are much higher than the arsenic in the sediment samples, thus a rsenic is bioaccumulating in the soft tissue of the E astern Oyster and the N o rther n Q uahog (Figure 3 4 and Figure 3 5) Both species are major food sources and sources of revenue for the residents of Cedar Key. Because the tissue that is being consumed is the soft tissue, or the viscera, it be comes imperative to know what the results mean for people who regularly consu me shellfish raised in the waters surrounding Cedar Key The Environmental Protection Agency (EPA) has set standards for arsenic consumption based on the No Observable Adverse Effect Level (NOAEL). The NOAEL is the lev el of exposure of an organism to a toxic substance at which there is no statistically significant adverse effect when compared to an appropriate control (EPA 1997) This standard or reference dose for chronic toxicity (RfD) is set at 3.0 x 10 4 mg/kg d an d the oral cancer potency is 1.5 per mg/kg d (EPA 1997) This level is based on two studies which
117 showed that consuming more inorganic arsenic than this could cause an increase of skin lesions and hyperpigmentation (EPA 1997) The oral cancer potency is defined as the proportion (of a population) affected per mg of substance/ kg body mass day (EPA 1997) This is usually used to calculate risk in low dose exposures. Similar standards have been set by the U.S. Department of Health, this includes an acute exposure limit of 0.005 mg/kg d of arsenic and a chronic exposure limit of 0.0003 mg/kg d (ATSDR 2007) These standards are set based on the lethal dose of arsenic, which according to ATSDR ( 2007) the minimum lethal acute dose of arsenic is around 130 mg or about 2 mg/kg for humans. Death in humans has resulted from chronic exposure of much lower levels, for example five children between the ages of 2 and 7 years died from chronic arsenic poisonin g after drinking water with 0.05 0.1 mg As/kg/d (ATSDR 2007) Considering the information above, and the samples collected in this study, are people who consume clam s from Cedar Key at risk? If we calculate the amount of arsenic consumed for one person wh o weighs 50 kilograms, and consumes 1 oyster a day the answer is no. This person would be consuming 0.0009 mg of arsenic per day. After dividing this numbe r by the mass of the person you would get a daily dose of 1.8 x 10 5 mg/kg d, which is well below the reference dose set by the EPA. Although this amount is below the minimum dose, one must remember that arsenic is a non essential trace element and the consumption of any arsenic could have negative effects on your health. Thus the EPA recommends that people, especially pregnant women and children, limit the amount of seafood consumed to a few times per month (EPA 1997) The daily consumption levels for Hg, Se, and Cd found in the northern quahog and eastern oyster are well below the Environmental Pro tection Agencies (EPA) list for human consumption (EPA 1997)
118 Conclusion In conclusion, molluscs can be useful for the long term collection of toxicological data. Molluscs have high site fidelity as adults and can be easily sampled for analysis. Data col lected from molluscs can be compared to different locations and conclusions can be drawn. L ong te rm trends observed in the Mussel Watch dataset, indicate clearly the importance of using molluscs in toxicology research. Molluscs collected in and around S eahorse Key were used to compare tissue concentrations between aquatic and terrestrial environments. Marine and terrestrial molluscs were sampled at Seahorse Key and significant levels of arsenic were found in the tissue of marine molluscs compared to ter restrial molluscs. Arsenic level increased from the highly terrestrial species (Southern Flatcoil, 0.00 ppm dry mass) to the obligate marine mollusc (Northern Quahog, 17.99 ppm dry mass). This increase in arsenic levels suggests that the Northern Quahog is bioaccumulating arsenic from the Gulf of Mex ico. Thus other organisms living in the Gulf of Mexico, could also be accumulating arsenic in their tissues. This includes the fishes collected from the island (Chapter 2) and indirectly the Florida Cottonmo uth Snake. This information is important for humans inhabiting the Cedar Key area, becaus e like the cottonmouths they can be exposed to potentially high levels of arsenic. When one considers the entire Seahorse Key system, it is important to consider all aspects of the environment. This includes the origin of the TICs (i.e. mining, agriculture runoff, burning of fossil fuels, etc), the transport of these metals (nearby rive rs, the wind, or weathering of rocks), bio concentration (the fish and mulluscs) and bioaccumulation (the snakes and birds). The entire environment is important to fully understand the fate and movement of contaminants in the Seahorse Key system or any sy stem in which TICs are a concern.
119 Snakes are important aspects of any ecosystem and should be included in any environmental impact assessment. Novel insight can be found regarding snakes and the use of snakes as bioindicators can have clear advantages f or researchers. These include long term bioaccumulation data, site specific data, and the long term effects of TICs on organisms. The research outlined here is another contribution to the field of toxicology usi ng snakes as study organism. The cottonmout hs from Seahorse Key had significantly higher tissue concentrations of arsenic than those from four locations on the mainland of Florida. The difference in tissue concentrations suggest that levels of contaminants in insular versus mainland prey resources are different. Cottonmouths on the mainland are consuming many different types of prey items and these items are not coming directly from the Gulf of Mexico. Whereas, t he snakes on Seahorse Key are consuming primarily fish dropped by the nesting water bird s which suggests there is an indirect trophic link between snakes and the Gulf of Mexico. Because s nakes and birds are consuming the same prey, we would expect the birds on the island to have similarly elevated levels of the above contaminants. A di rect comparison of tissue concentrations between cottonmouths and nesting waterbirds has not been done. However, given that birds are endothermic and consume larger quantities of fish, one would expect similarly elevated ti ssue concentrations. The resul ts from Chapter 2 are a first step in fully understanding the presence and movement of TICs on Seahorse Key and in cottonmouth snakes living in Florida. In order to more fully understand the movement of contaminants in and around Seahorse Key molluscs we re sampled and their tissues analyzed. The levels of contaminants reported in Chapter 3, correspond directly with the levels reported in Chapter 2, because the source of the contaminants is the same. The location of Seahorse Key (between two large rivers ) and the tidal flows around Seahorse Key all contribute to an environment that has similar levels of
120 contaminants. Thus finding elevated levels of arsenic in sediment, molluscs, fish, and snakes indicate pretty clearly that the source of the conta minant is the Gulf of Mexico. The toxicological research outlined in this thesis becomes important when one considers the unique environment on Seahorse Key, the proximity of the island to human habitation, the designation of the island as a national wildlife re fuge, and the trophic interaction the snakes have with nesting water birds. The cottonmouths on the island as well as the molluscs sampled in and around the island could provide insight into processes that transcend the local environment and affect popula tions of birds and humans on a global level. Cedar Key has a large aquaculture community, which includes Eastern Oyster ( Crassostrea virginica ) and Northern Quahog Clam ( Mercenaria mercenaria ) farming, and fish harvesting for human consumption. Oysters, c lams, and fish from Cedar Keys are being consumed in the local area as well as in restaurants and home thousands of miles away. Essentially the collection and shipment of seafood from Cedar Key links the welfare of many people to a small location in the G ulf of Mexico. Thus, having information about the types and levels of TICs in and around Seahorse Key can have a lasting impact on the welfare of numerous people.
121 Table 3 1 Summary of the results of tissue analysis for 29 trace elements in three spec ies of molluscs Arithmetic Mean Standard Error are show for tissue concentration of each metal in the array. N= sample size, CF is the C onversion Factor calculated using the eq uation in the methods section, s ample type is the type o f tissue that was sampled, W = whole body homogenization, W(7) is whole body homogenization of seven snails to make on e sample, V = the v iscera or soft tissue, and S = the shell of the mollusc. DM = Dry Mass, and WM = Wet Mass. Wet mass was calculated usin g the following equation WM = DM*CF Length is measured as the greatest width anywhere on the sample. MP = Marsh Periwinkle Snail ( Littoraria irrorata ), SF = Southern Flatcoil Snail ( Polygyra cereolus ), and NQ = the Northern Quahog Clam ( Mercenaria mercen aria ) Marsh Periwinkle and Southern Flatcoil Snails were collected from Seahorse Key and Northern Quahogs were donated by a local clam farmer and were grown in waters around Cedar Key. All tis sue concentrations are in ppm. Species CF N Type Length (mm) Ag Al As B Ba MP 0.85 12 W D M 21.04 1.61 0.00 0.00 9.83 5.32 0.04 0.10 2.65 0.39 1.39 0.15 W M 0.00 0.00 8.33 4.53 0.04 0.08 2.25 0.33 1.18 0.13 SF 0.75 2 W(7) D M 6.14 0.0 9 0.00 0.00 4.78 1.14 0.00 0.00 8.07 0.64 3.14 1.09 W M 0.00 0.00 3.59 0.86 0.00 0.00 6.05 0. 48 2.36 0.82 NQ 0.23 13 V D M 48.11 5.16 0.00 0.00 34.85 24.59 17.99 3.25 10.87 2.18 1.62 0.48 W M 0.00 0.00 8.02 5.66 4.14 0.75 2.50 0.50 0.37 0. 11 1.00 13 S D M 48.11 5.16 0.01 0.04 6.62 1.06 0.00 0.00 3.98 0.85 6.98 1.83 W M 0.01 0.04 6.62 1.06 0.00 0.00 3.98 0.85 6.98 1.83
122 Table 3 1: Continued. Species Be Ca Cd Co Cr Cu Fe MP 0.00 0.0 1 362803.99 34254.41 0.22 0.06 0.32 0.03 0.09 0.05 3.61 0.77 1110.85 39.48 0.00 0.0 1 307294.98 29116.25 0.18 0.05 0.27 0.03 0.08 0.05 3.06 066 940.89 33.56 SF 0.03 0.01 331502.22 4876.96 0.93 0.17 0.35 0.01 0.19 0.07 7.69 0.29 1040.09 24.82 0.02 0.01 248626.67 3657.72 0.70 0.13 0.26 0.01 0.14 0.05 5.77 0.22 780.07 18.61 NQ 0.01 0.03 6267.67 1849.04 0.66 0.19 0.92 0.14 2.13 0.35 9.18 1.69 162.87 52.18 0.00 0.01 1441.56 425.28 0.15 0.04 0.21 0.03 0.49 0.08 2.11 0.39 37.46 12.00 0.01 0.01 399671.20 18891.76 0.01 0.01 0.36 0.03 0.02 0.01 0.36 0.06 1240.49 138.76 0.01 0.01 399671.2 0 18891.76 0.01 0.01 0.36 0.03 0.02 0.01 0.36 0.06 1240.49 138.76 Table 3 1: Continued. Species Hg K Li Mg Mn Mo MP 0.01 0.00 710.95 65.85 0.26 0.04 1276.08 116.10 6.98 2.87 0.06 0.06 0.01 0.00 602.18 55.97 0.22 0.03 1084.66 98.69 5.91 2.44 0.05 0.05 SF 0.01 0.00 1884.27 8.01 0.10 0.01 1445.52 73.66 29.78 3.60 0.30 0.09 0.01 0.00 1413.20 6.01 0.07 0.01 1084.14 55.25 22.34 2.70 0.23 0.06 NQ 0.16 0.04 6373.74 659.00 0.60 0.11 6202.77 681.55 29.21 20.69 0.53 0.12 0.04 0.01 1465.96 151.57 0.14 0.02 1426.64 156.76 6.72 4.76 0.12 0.03 0.00 0.00 169.66 22.15 0.33 0.05 226.79 17.20 3.27 1.71 0.04 0.03 0.00 0.00 169.66 22.15 0.33 0.05 226.79 17.20 3.27 1.71 0.04 0.03
123 Table 3 1: Continued. Species Na Ni P Pb Sb Se Si MP 7561.73 488.51 0.42 0.12 243.80 55.99 0.07 0.05 0.01 0.01 0.25 0.11 4 0.69 22.90 6404.78 415.23 0.35 0.10 206.50 47.59 0.06 0.05 0.01 0.01 0.21 0.09 34.46 19.47 SF 1477.62 29.64 0.24 0. 13 2338.80 165.89 0.07 0.0 2 0.03 0.01 0.35 0.10 19.50 0.13 1108.22 22.23 0.18 0. 10 1754.10 124.42 0.05 0.0 1 0.02 0.01 0.26 0.0 8 14.62 0. 10 NQ 32572.47 3499.08 3.13 0.76 8155.27 450.29 0.43 0.22 0.01 0.02 4.00 0.54 104.05 14.68 7491.67 804.79 0.72 0.17 1875.71 103.57 0.10 0.05 0.00 0.01 0.92 0.12 23.93 3.38 5795.4 7 282.46 0.39 0.06 41.23 17.57 0.07 0.02 0.01 0.0 1 0.15 0.13 39.14 22.64 5795.47 282.46 0.39 0.06 41.23 17.57 0.07 0.02 0.01 0.01 0.15 0.13 39.14 22.64 Table 3 1: Continued. Species Sn Sr Tl V Zn MP 0.01 0.01 1669.18 91.65 0.00 0.01 0.04 0.01 3.03 1.93 0.01 0.01 1413.80 77.90 0.00 0.00 0.03 0.01 2.56 1.64 SF 0.03 0.02 1416.79 1 54.25 0.03 0.01 0.08 0.01 19.93 3.89 0.02 0.01 1062.59 115.69 0.02 0.01 0.06 0.00 14.95 2.91 NQ 0.01 0.02 70.14 12.28 0.01 0.02 1.00 0.20 144.61 30.62 0.00 0.01 16.13 2.83 0.00 0.00 0.23 0.05 33. 26 7.04 0.01 0.01 1404.99 100.96 0.00 0.01 0.08 0.02 0.91 0.26 0.01 0.01 1404.99 100.96 0.00 0.01 0.08 0.02 0.91 0.26
124 Figure 3 1 Location of sampling sites : Black Point and Seahorse Key Eastern Oyster ( Crassostrea virginica ) and sediment samples were collected from Black Point by the NOAA Mussel Watch program from 1986 2007. Marsh Periwinkle ( Littoraria irrorata ) and Southern Flatcoil ( Polygyra cere olus ) snails were collected from Seahorse Key during 2011 2012. Northern Quahog Clams ( Mercenaria mercenaria ) were donated by a Cedar Key clam farmer in 2010 and were grown on a clam lease within 15km of Cedar key. Map created by the author using ArcGIS version 9.2.
125 Figure 3 2 Averag e tissue concentration for the Marsh P eriwinkle Snail ( Littoraria irrorata ) and the Southern F latcoil Snail ( Polygyra cereolus ) for a rsenic (A s), cadmium (Cd), mercury (Hg), and s elenium (Se) from Seahorse Key, Florida. Error bars are calculated using Stand ard Error, sample size for the M arsh Periwinkle was twelve and for the Southern F latcoil w as two. All tissue concentrations are dry mass and ppm. bet ween Marsh Periwinkle Snail and Southern Flatcoil Snail samples with p values < 0.05. 0 0.2 0.4 0.6 0.8 1 1.2 As Cd Hg Se Concentration (ppm) Metal Type Tissue concentration of As, Cd, Hg, Se in two snail species on Seahorse key Periwinkle Flatcoil
126 Figure 3 3 Averag e t issue concentration for the N orthern Quahog Clam ( Mercenaria mercenaria ) for a rsenic (As), cadmium (Cd), mercury (Hg), and s elenium (Se) Clams were donated by a local clam farmer from Cedar Key, Florida, and all clams were grown on local clam leases around Cedar Key. Error bars are calculated using Stand ard Error, and sample size for the Northern Quahog was thirteen ind ividual clams. All tissue concentrations are dry mass and ppm samples with p values < 0.05. * * 0 2 4 6 8 10 12 14 16 18 20 As Cd Hg Se Concentration (ppm) Metal Type Tissue concentration of As, Cd, Hg, Se for the Northern Quahog Clam-Viscera Clam-shell
127 Figure 3 4 Mussel Watch data downloaded from the CCMA website. Data po ints are average concentration of samples of the Eastern O yster ( Crassostrea virginica ) taken from Black Point, Florida from 1986 2007. Th ese data were produced by the U.S. National Oceanic and Atmospheric Administration through its National Status and Trends Program (NOAA 2013) All measurements are dry mass and ppm. 0 5 10 15 20 25 30 35 Concentration (ppm) Black Point, Cedar Key, FL Arsenic Cadmium Mercury Selenium
128 Figure 3 5 Mussel Watch data downloaded from the CCMA website. Data points are single samples of sediment taken from Black Point, Florida from 1986 2007. Th ese data were produced by the U.S. National Oceanic and Atmospheric Administration through its National Status and Trends Program (NOAA 2013) All measurements are dry mass and ppm 0 2 4 6 8 10 12 1986 1987 1997 2007 Concentration (ppm) Black Point, Cedar Key, FL Sediment samples Arsenic Cadmium Mercury Selenium
129 Figure 3 6 Mercury concentration as a function of selenium concentration in three mollusc species collected on and around Seahorse Key. Species are the Northern Quahog ( Mercenaria mercenaria ), the Marsh Periwinkle ( Littoraria irrorata ), and the Southern Flatcoil ( Polygyra cereolus ). y = 23.787x + 0.1444 R = 0.9486 0 1 2 3 4 5 6 0.00 0.05 0.10 0.15 0.20 0.25 Selenium Concentration (ppm) Mercury Concentration (ppm) Mercury Concentration as a Function of Selenium Concentration
130 LIST OF REFERENCES Allinger, R. D., B. B. Erger, and R. T. R. hler. 2001. Soil Biology and Ecotoxicology. Pages 489 525 in G. M. Barker, editor. The Biol ogy of Terrestrial Molluscs First edition. CAB international. Apeti, D. A. G. G. Lauenstein, and D. W. Evans. 2012. Recent status of total mercury and methyl mercury in the coastal waters of the northern Gulf of Mexico using s mussel watch program. Marine Pollution B ulletin 64:2399 408. Appenroth K. 2010. Soil Heavy Metals 19:19 30. ATSDR. 1999. Toxicological Pro file for Mercury. Pages 1 676. Agency for Toxic Substances and Disease Registry. Atlanta, GA, USA. ATSDR. 2003. Toxicological Profile for Selenium. Pages 1 457. Agency for Toxic Substa nces and Disease Registry. Antlanta, GA, USA. ATSDR. 2007. Toxicological Profile for Arsenic. Pages 1 559. Agency for Toxic Substances and Disease Registry. Atlanta, GA, USA. ATSDR. 2012. Toxicological Profile for Cadmium. Pages 1 487. Agency for Toxic Substances and Disease Registry. Atlanta, GA, USA. Bauerle, B., D. L. Spencer, and W. Wheeler. 1975. The use of snakes as a pollution indicator species. Copeia 1975:366 368. Berger, B., and R. Dallinger. 1993. Terrestrial snails as quanti tative indicators of environmental metal pollution. Environmental Monitoring and A ssessment 25:65 84. Berry, M. J., and N. V. C. Ralston. 2008. Mercury toxicity and the mitigating role of selenium. EcoHealth 5:456 459. Bertness, M. D., and B. R. Silliman. 2008. Consumer control of salt marshes driven by human disturban ce. Conservation biology : The J ournal of the Society for Conservation Biology 22:618 23. Third edition. Danish. Blanvillain, G., J. Schwente r, R. Day, D. Point, S. Christopher, W. Roumillat, and D. Owens. 2007. Diamondback terrapins, Malaclemys terrapin, as a sentinel species for monitoring mercury pollution of estuarine systems in South Carolina a nd Georgia, USA. Environmental. 26:1441 1450.
131 Bloom, D. E. 2011. 7 Billion and Counting. Science (New York, N.Y.) 333:562 9. Blus, L. J., B. S. Needy Jr, T. G. Lamont, and B. Mulhern. 1977. Residues of organochlorides and heavy metals in tissues and eggs of brown pelicans. Journal of Pesticide Manage ment 11:40 53. Boening, D. W. 2000. Ecological Effects, transport, and fate of mercury: a general review. Chemosphere 40. Bruehler, G., and A. de Peyster. 1999. Selenium and other Trace Metals in Pelicans dying at the Salton Sea. Bulletin of Environmental Contamination and T oxicology 63:590 597. Burger, J. 1992. Trace element levels in pine snake hatchlings: tissue and tem poral differences. Archives of Environmental C ontamination an d T oxicology 22:209 13. Burger, J., K. R. Campbell, T. S. Campbell, T. Shukl a, C. Jeitner, and M. Gochfeld. 2005. Use of skin and blood as nonlethal indicators of heavy metal contamination in northern water snakes (Nerodia sipedon). Archives of Environmental Contamination and Toxicology 49:232 238. Burger, J., K. R. Campbell, S. M urray, T. S. Campbell, K. F. Gaines, C. Jeitner, T. Shukla, S. Burke, and M. Gochfeld. 2007. Metal levels in blood, muscle and liver of water snakes (Nerodia spp.) from New Jersey, Tennessee and South Carolina. Science of the Total Environment 373:556 563. Burger, J., M. Gochfeld A. A. Rooney, E. F. Orlando, A R. Woodward, and J. L. J. Guillette. 2000. Metals and Metalloids in Tissues of American Alligators in Three Florida Lakes. Archives of Environmental Contamination and Toxicology 38:501 508. Burger J., S. Murray, K. F. Gaines, J. M. Novak, T. Punshon, C. Dixon, and M. Gochfeld. 2006. Element levels in snakes in South Carolina: Differences between a control site and exposed site on the savannah river site Environmental Monitoring and A ssessment 112 :35 52. Burkett, R. D. 1966. Natural History of Cottonmouth Moccasin, Agkistrodon piscivorus (Reptilia). Pages 435 491 (R. Hall, H. Fitch, and F. Cross, Eds.) University of Kansas University of Kansas, Lawrence Bush, M., and J. Smeller. 1978. Blood col lection & injection techniques in snakes. Veterinary m edicine, small animal clinicia. V eterinary Medicine. SAC 73:211 4. Cahill, T. M., D W. Anderson, R. A. Elbert, B. Perley, and D. R. Johnson. 1998. Elemental profiles in feather samples from a Mercury Contaminated lake in Central California. Archives of Environmental Contamination and Toxicology 35:75 81.
132 Campbell, K. R., and T. S. Campbell. 2001. The accumulation and effects of environmental contaminants on s nakes: A review. Environmental Monitoring and A ssessment 70:253 301. Campbell, K. R., and T. S. Campbell. 2002. A logical starting point for developing priorities for lizard and snake ecotoxicology: A review of available data. Environmental Toxicology and Chemistry 21:894 898. Campbell, K. R., T. S. Campbell, and J. Burger. 2005. Heavy metal concentrations in northern water snakes (Nerodia sipedon) from East Fork Poplar Creek and the Little River, Ea st Tennessee, USA. Archives of Environmental Contamination and T oxicology 49:239 48. Capinera, J., and J. White. 2011. Terrestrial snails affecting plants in Florida. Featured Creatures:1 12. Chen, M., L. Ma, C. G. Hoogeweg, and W. G. Harris. 2001. Arsenic Background C oncentrations in Florida, U.S.A. Surface Soils: Determination and Interpretation. Environmental Forensics 2:117 126. Chin, S. Y., J. D. Willson, D. A. Cr istol, D. V. V Drewett, and W. A. Hopkins. 2012. High levels of maternally transferred mercury do not a ffect reproductive output or embryonic survival of northern water snakes (N erodia sipedon). Environmental Toxicology and C hemistry / SETAC. Cornelis, R., and M. Nordberg. 2007. General Chemistry, Sampling, Analytical Methods, and Speciation. Pages 11 38 in G. F. Nordberg, B. A. Fowler, and M. Nordberg, editors. Handbook on the Toxicology of Metals Third editio n Crosby, D. G. 1998. Env ironmental T oxi cology and C hemistry. Oxford University Press, New York. Cullen, W. R., and K. J. Reimer. 1989. Arsenic speciati on in the environment. Chemical Reviews 89:713 764. Dallinger, R. 1994. Invertebrate organisms as biological indicators of heavy metal pollution. Applied Biochemistry and B iotechnology 48:27 31. Database, T. R. 2013. Species Numbers by Higher Taxa. http:// www.reptile database.org/db info/SpeciesStat.html. Drewe tt, D. V. V, J. D. Willson, D. A. Cri stol, S. Y. Chin, and W. A. Hopkins. 2013. Inter and intraspecific variation in mercury bioaccumulation by snakes inhabiting a contaminated r iver floodplain. Envi ronmental Toxicology and C hemistry / SETAC 32:1178 86.
133 and applied chemistry 74:793 807. Eisler, R. 1988. Arsenic Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Re view. Biological Report USFWS 85:1 65. Elder, J. F., and H. C. Mattraw. 1984. Accumulation of trace elements, pesticides, and polychlorinated biphenyls in sediments and the clam Corbicula manilensis of the Apalachico la River, Florida. Archives of Environmental Contamination and T oxicology 13:453 69. E l Sikaily, A., A. Khaled, and A. el Nemr. 2004. Heavy metals monitoring using bivalves from Mediterranean Sea and Red Sea. Environmental Monitoring and A ssessment 98:41 58. EPA, U. E. P. A. 1997. Guida nce for Assessing Chemical Contaminant Data for Use in Fish Advisories. Vol. 2 Risk Assessment and Fish Consumption Limits 2. Fowler, B., G. Nordberg, M. Nordberg, and L. Friberg (Eds.). 2007. Handbook on the Toxicology of Metals. Third edition. Elsevier I nc. Ganser, L. R., W. A Liver histopathology of the southern watersnake, Nerodia fasciata fasciata, following chronic exposure to trace element contaminated prey from a coal ash disposal sit e. Journal of Herpetology 37:219 226. Gardner, S. C., and E. Oberdrster. 2006. Toxicology of Reptiles. (S. C. Gardner and E. Oberdrster, Eds.)Fi rst. CRC press, Boca Raton Fla. Pensacola, Fla. Garner, M. M. 2005. Trends in Reptilian Neoplasia: A Diagnosti cian's Perspective. The North American Veterinary Conference 2005 Proceedings:1278 1280. Gibbons, J. W., D. E. Scott, T. J. Ryan, K. A. Buhlmann, T. D. Tuberville, B. S. Metts, J. L. Greene, T. Mills, Y. Leiden, S. Poppy, and C. T. Winne. 2000. The globa l decline of reptiles, Deja Vu amphibians. Bioscience 50:653 666. Goldberg, E. 1986. The mussel watch concept. Environmental Monitoring and Assessment 7:91 103. Grillitsch, B., and L. Schiesari. 2010. The Ecotoxicology of Metals in Reptiles. Pages 337 448 in D. W. Sparling, G. Linder, C. A. Bishop, and S. K. Krest, editors. Ecotoxicology of Amphibians and Reptiles S econd edition SETAC. Gude, A. 2000. Cedar Keys National Wildlife Refuge Fact S heet. United States Fish and Wildlife Service. http://www.wilderness.net/printFactSheet.cfm?WID=108
134 Gumber, S., J. G. Nevarez, and D. Y. Cho. 2010. Endocardial Fibrosarcoma in a Reticulated Python (Python Reticularis). Journal of Veter inary Diagnostic Investigation 22:1013 1016. Heritage, D. A. 1982. Revised edition of the American Heritage Dict ionary of the English Language New College Edition. Hopkins, W. A., B. P. Staub, J. A. Baionno, B. P. Jackson, and L. G. Talent. 2005. Transfer of selenium from prey to predators in a simulated terrestrial food ch ain. Environmental P ollution (Barking, Essex : 1987) 134:447 56. Hopkins, W. A. 2000a. Reptile toxicology: challenges and opportunities on the last frontier in vertebrate ecotoxicology. E nvironmental Toxicology and Chemistry 19:2391 2393. Hopkins, W. A. 2000b. Letter to the Editor, Reptile Toxicology: Challenges and Opportunities on the last front Frontier in Vertebrate Ecotoxicology. Environmental Toxicology and Chemistry 19:2391 2393. Hopkins, W. A. 2006. Use of Tissue Residues in Reptile Ecotoxicology: A Call for Integration and Experimentalism. Pages 35 62 in S. C. Gardner and E. Pensacola, Fla, USA. Hopki ns, W. A ., J. H. Roe, T. Philippi, and J. D. Congdon. 2004a. Standard and digestive metabolism in the banded water snake, Nerodia fasciata fasciata. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 137:141 149. Hopkins, W. A., J. H. Roe, J. W. Snodgrass, B. P. Jackson, D. E. Kling, C. L. Rowe, and J. D. Congdon. 2001. Nondestructive indices of trace element exposure in squamate reptiles. Environmental Pollution 115:1 7. Hopkins, W. A., J. H. Roe, J. W. Snodgrass, B. P. Stau b, B. P. Jackson, and J. D. Congdon. 2002. Effects of chronic dietary exposure to trace elements on banded water snakes (Nerodia fasciata). Environmental Toxicology and Chemistry 21:906 913. Hopkins, W. A., C. L. Rowe, and J. D. Congdon. 1997. Elevated mai ntenance costs in banded water snakes, Nerodia fasciata, exposed to coal combustion wastes. American Zoologist 37:122A. Hopkins, W. A., C. L. R owe, and J. D. Congdon. 1999. Elevated trace element concentrations and standard metabolic rate in banded water s nakes (Nerodia fasciata) exposed to coal combustion wastes. Environmental Toxicology and Chemistry 18:1258 1263.
135 Hopkins, W. A., B. P. Staub, J. a Baionno, B. P. Jackson, J. H. Roe, and N. B. Ford. 2004b. Trophic and maternal transfer of selenium in brown house snakes (Lamprophis f uliginosus). Ecotoxicology and Environmental S afety 58:285 93. Hothorn, T., F. Bretz, and P. Westfall. 2008. Simultaneous Inference in General Parametric Models. Biometrical Journal 50:346 363. Ikemoto, T., T. Kunito, H. Tanaka, N Baba, N. Miyazaki, and S. Tanabe. 2004. Detoxification mechanism of heavy metals in marine mammals and seabirds: interaction of selenium with mercury, silver, copper, zinc, and cadmium in liver. Archives of E nvironmental Contamination and T oxicology 47:4 02 13. Jrup, L. 2003. Hazards of heav y metal contamination. British Medical B ulletin 68:167 182. Jones, D. E., and S. D. Holladay. 2006. Excretion of three heavy metals in the shed skin of exposed corn snakes (Elaphe guttata). Ecotoxicology and Environmental S afety 64:221 5. Jones, D. E., G. Magnin Bissel, and S. D. Holladay. 2009. Detection of polycyclic aromatic hydrocarbons in the shed skins of corn snakes ( Elaphe guttata). Ecotoxicology and Environmental S afety 72:2033 2035. Kells, V., and K Carpenter. 2011. A Field Guide to Coastal Fishes from Maine to TexasFirst. The Johns Hopkins University Press, Baltimore, Maryland, USA. Khan, B. I., H. M. Solo Gabriele, T. G. Townsend, and Y. Cai. 2006. Release of arsenic to the environment from CCA tr eated wood. 1. Leaching and speciation during service. Environmental Science and T echnology 40:988 93. Kuiken, T., G. A. Fox, and K. L. Danesik. 1999. Bill Malformations in double crested Cormorants with low exposure to organochlorines. Environmental Toxic ology and Chemistry 18:2908 2913. Kunito, T., R. Kubota, J. Fujihara, T. Agusa, and S. Tanabe. 2008. Arsenic in marine mammals, seabirds, and sea turtles. Reviews of Environmental Contamination and T oxicology 195:31 69. Lefcort, H., J. Vancura, and E. L. L ider. 2010. 75 Years After Mining Ends Stream Insect Diversity Is Still Affected By Heavy Metals. Ecotoxicology (London, England) 19:1416 25. Lillywhite, H. B. 2006. Water relations of tetrapod integument. Journal of Experimental Biology 209:202 226.
136 Lilly white, H. B., and R. J. R. McCleary. 2008. Trophic ecology of insular cottonmouth snakes: review and perspective. South American Journal of Herpetology 3:175 185. Lillywhite, H. B., C. M. Sheehy, M. D. McCue, and C. M. S. Iii. 2002. Scavenging behaviors of cottonmouth snakes at island bird rookeries. Herpetological Review 33:259 261. Marigmez, I., M. Soto, M. P. Cajaraville, E. Angulo, and L. Giamberini. 2002. Cellular and subcellular distribution of metals in molluscs. Microscopy Research and T echnique s. 56:358 92. McCarthy, K. M. 2006. Cedar Key, Florida: An Illustrated History. Pages 1 244 F irst edition Nature Coast Publishing House. McCue, M. D., and H. B. Lillywhite. 2002. Oxygen consumption and the energetics of island ddwelling Florida cottonmouth snakes. Physiological and Biochemical Zoology 75:165 178. McKillup, S. 2005. Statistics Explained: An Introductory Guide for Life Scientists F irst edition Cambridge University Press, Cambridge, U.K. Messaoudi, I., S. Barhoumi, K. Sad, and A. Kerken. 200 9. Study on the sensitivity to cadmium of marine fish Salaria basilisca (Pisces: Blennidae). Journal of Environmental Sciences 21:1620 1624. Nigro, M., and C. Leonzio. 1996. Intracellular storage of mercury and selenium in different marine vertebrates. Mar ine Ecology Progress Series 135:137 143. NOAA. 1986. National Status and Trends: Mussel Watch Project Analyte Chemistry File Science. http://ccma.nos.noaa.gov/about/coast/nsandt/metada ta/MW_ChemistryMetadata.m ht. NOAA. 2013. Mussel Watch Data Portal. http://ccma.nos.noaa.gov/about/coast/nsandt/download.aspx. Notten, M. J. M., A. J. P. Oosthoek, J. Rozema, and R. Aerts. 2006. Heavy metal pollution affects consumption and reproduction of the landsnail Cepaea nemoralis fed on naturally polluted Urtica dioica leaves. Ecotoxicology (London, England) 15:295 304. Ohlendorf, H. M., R. L. Hothem, and T. W. Aldrich. 1988. Bioaccumulation of Selenium by Snakes and Frogs in the San Joaquin Valley California. Copeia 1988:704 710.
137 Oros, J., P. Monagas, M. Andrada, P. Calabuig, and J. Pether. 2009. Metastatic fibrosarcoma in a captive Saharan horned viper (Cerastes cerastes) with high hepatic levels of cadmium. Veterinary Record 164:690 692. Pough, F. H., R. M. Andrews, J. E. Cadle, M. L. Crump, A. H. Savitsky, K. P. Wells, a nd J. Savage. 1998. Herpetology Prentice Hall Upper Saddle River (NJ). R Development Core Team, R. 2011. R: A Language and Enviro nment for Statistical Computing R Foundation for Statistical Computing. Rainwater, T. R., K. D. Reynolds, J. E. Caas, G. P. Cobb, T. a Anderson, S. T. McMurry, P. N. Smith, and J. E. Canas. 2005. Organochlorine pesticides and mercury in cottonmouths (Agkistrodon piscivorus) from northeastern Texas, USA. Environmental Toxicology and Chemistry 24:665 673. Reading, C. J., L. M. Luiselli, G. C. Akani, X. Bonnet, G. Amori, J. M. Ballouard, E. Filippi, G. Naulleau, D. Pearson, and L. Rugiero. 2010. Are snake populations in wides pread decline? Biology L etters 6 :777 80. Reid, G. K. 1954. An ecological study of the Gulf of Mexico fishes, in the vicinity of Cedar Key, Florida. Bulletin of Marine Science 4. Rezaie Atagholipour, M., A. Riyahi Bakhtiari, M. Sajjadi, C. K. Yap, S. Ghaffari, Z. Ebrahimi Sirizi, and P. G hezellou. 2012. Metal concentrations in selected tissues and main prey species of the annulated sea snake (Hydrophis cyanocinctus) in the Hara Protected Area, northeastern coast of the Persia n Gulf, Iran. Marine pollution B ulletin 64:416 421. Rittschof, D. and P. McClellan Green. 2005. Molluscs as multidisciplinary models in environment toxicology. Marine Pollution B ulletin 50:369 73. Selcer, K. W. 2006. Reptile ecotoxicology: Studying the effects of contaminants on populations. Toxicology of Reptiles.CRC, Boca Raton, FL, USA:267 297. Sepulveda, M., R. Poppenga, J. Arrecis, and L. Quinn. 1998. Concentrations of mercury and selenium in tissues of double crested cormorants (Phalacrocorax auritus) from southern Florida. Colonial Waterbirds 21:35 42. Silliman, B. R., and M. D. Bertness. 2002. A trophic cascade regulates salt marsh primary production. Proceedings of the National Academy of Sciences of the United States of America 99:10500 5. Silliman, B. R., and S. Y. Newell. 2003. Fungal farming in a snail. Pro ceedings of the National Academy of Sciences of the United States of America 100:15643 8.
138 Smith, C. 2012. Drought has Cedar Key Scrambling to Provi de Fresh Water. Gainesville Sun. http://www.gainesville.com/article/20120620/ARTICLES/120629937 Snodgrass, J. W., R. E. Casey, J. A. Simon, and K. Gangapura. 2008. Ecotoxicology of amphibians and reptiles in urban environments: an overview of potential exposure rou tes and bioaccumula tion. Urban H erpetology. Society for the Studyof Amphibians and Reptiles. Salt Lake City, Utah. Pp 177 196 Sparling, D. W., G. Linder, C. A. Bishop, and S. K. Krest (Eds.). 2000. Ecotoxicology of Amphibians and Reptiles F irst edition SETAC, Pensacola, Florida, USA. Sparling, D. W., G. Linder, C. A. Bishop, and S. K. Krest (Eds.). 2010a. Ecotoxicology of Amphibians and Reptiles Second edition SETAC, Pensacola, Florida, USA. Sparling, D. W., G. Linder, C. A. Bishop, and S. K. Krest. 2010b. Recent Advancement in Amphibians and Reptile Ecotoxicology. Pages 1 12 in D. W. Sparling, G. Linder, C. A. Bishop, and S. K. Krest, editors. Ecotoxicology of Amphibians and Reptiles S econd edition CRC press, Pensacola, Florida, USA. Stafford, D. P., F. W. Plapp, and R. R. Fleet. 1976. Snakes as indicators of environmental contamination: relation of detoxifying enzymes and pesticide residues to species occurrence in three a quatic ecosystems. Archives of Environmental Contamination and T oxicology 5:15 27. Strunjak Perovic, I., D. Lisicic, R. Coz Rakovac, N. Topic Popovic, M. Jadan, V. Benkovic, and Z. Tadic. 2010. Evaluation of micronucleus and erythrocytic nuclear abnormalities in Balkan whip snake Hierophis gemonensis. Ecotoxicology (London, England) 19:1460 5. Viard, B., F. Pihan, S. Promeyrat, and J. C. Pihan. 2004. Integrated assessment of heavy metal (Pb, Zn, Cd) highway pollution: bioaccumulation in soil, Graminaceae and land snails. Che mosphere 55:1349 59. Weir, S. M., J. G. Suski, and C. J. Salice. 2010. Ecological risk of anthropogenic pollutants to reptiles: Evaluating assumptions of sensitivi ty and exposure. Environmental P 606. Wharton, C. 19 66. Reproduction and growth in the cottonmouths, Agkistrodon piscivorus Lacepede, of Cedar Keys, Florida. Copeia 1966:149 161. Wharton, C. H. 1969. The Cottonmouth Moccasin on Sea Horse Key Florida. Bulletin of the Florida State Museum Biological Sciences 14:227 272. Wickliffe, J. K., and J. W. Bickman. 1998. Flow cytometric analysis of hematocytes from brown pelicans (Pelecanus occifentalis) exposed to planar halogenated
139 hydrocarbons and heavy metals. Bulletin of Environmental C ontamination and T oxicology 61:239 246. Wiemeyer, S. N., S. K. Schmeling, and A. Anderson. 1987. Environmental pollutant and necropsy data for ospreys from the Eastern United States, 1975 1982. Journal of Wildlife Distribution 23:279 291. Wilson, E. A., E. N. Powell, T. L. Wade, R. J. Taylor, B. J. Presley, and J. M. Brooks. 1992. Spatial and temporal distributions of contaminant body burden and disease in scale climatic controls 235:201 235. De Wolf, H., S. a Ulomi, T. Backeljau, H. B. Pratap, and R. Blust. 2001. Heavy metal levels in the sediments of four Dar es Salaam mangroves. Accumulation in, and effect on the morphology of the periwinkle, Littoraria scabra (Moll usca: Gastropoda). Environment I nternational 26:243 9 Wood, M. D., N. A. Beresford, D. V Semenov, T. L. Yankovich, and D. Copplestone. 2010. Radionuclide trans fer to reptiles. Radiation and E nvironmental B iophysics 49:509 530. Wylie, G. D., R. L. Hothem, D. R. Bergen, L. L. Martin, R. J. Taylor, and B. E. B russee. 2009. Metals and Trace Elements in Giant Garter Snakes (Thamnophis gigas) from the Sacramento Valley, California, USA. Archives of Environmental Contamination and Toxicology 56:577 587. Zakrzewski, S. F. 1997. Principl es of environmental toxicology Sigmund F. Zakrzewski. American Chemical Society, Washington, DC.
140 BIOGRAPHICAL SKETCH Joel Wixson was bo rn in Hayden Colorado in 1983. He rece ived his bachelor of science from Colo rado State University, where he major ed in fishery b iology and minored in m athematics. While attending Colorado State University, Joel would spend the summers working for the Colorado Division of Wildlife (CDOW). While working for CDOW, he gained valuable experience collecting data, o bserving wildlife, sampling fish and amphibians, and thinking like a scientist. After graduating in 2005, he worked for the Nevada Department of Fish and Wildlife (NDOW) concerning native threatened and endangered fish and amphibian species. After workin g for NDOW he moved to Oregon and worked for the Oregon Department of Fish and Wildlife (ODOW) this time working with coastal fisheries. Most of his work in NDOW, CDOW, and ODOW involved collecting data in the field, analyzing that data, writing reports and conducting GIS analysis. After working for various private organizations, Joel applied and was accepted to the University of Florida. He entered the Zoology program in 2010 in order to e arn a Master of Science degree.