View on the World Porifera Database

Home , Agelas schmidti, Aplysina, Aplysina archeri, Aplysina fistularis, Biemnida, Callyspongia plicifera

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2018 Shifting Distributions of Marine Sponges and the Ecology of an Endemic Species Kathleen Kaiser

Follow this and additional works at the DigiNole: FSU's Digital Repository. For more information, please contact [email protected] FLORIDA STATE UNIVERSITY

COLLEGE OF ARTS AND SCIENCES

SHIFTING DISTRIBUTIONS OF MARINE SPONGES AND THE ECOLOGY OF AN ENDEMIC

SPECIES

KATHLEEN KAISER

A Thesis submitted to the Department of Biological Science in partial fulfillment of the requirements for the degree of Master of Science

2018 Kathleen Kaiser defended this thesis on June 19, 2018. The members of the supervisory committee were:

Janie L. Wulff Professor Directing Thesis

Don R. Levitan Committee Member

Sophie J. McCoy Committee Member

The Graduate School has verified and approved the above-named committee members, and certifies that the thesis has been approved in accordance with university requirements.

This thesis is dedicated to my amazing mother, Mary Kaiser, who, despite a world of obstacles and setbacks, always cheers me on and inspires me to keep striving.

iii

ACKNOWLEDGMENTS

I would like to thank Janie Wulff for her amazing guidance and collaboration throughout this project and many others, as well as her continued encouragement. I thank Don Levitan, and Sophie McCoy for discussion on project development, data analysis, and writing. Special thanks to Gregg Hoffman for facilitating and encouraging in-situ experimentation on Halichondria corrugata, as well as Diver Mike for his guidance and aid in finding and collecting sponges in the Cedar Key and Tarpon Springs regions. Special thanks, also, to Patrick Erwin for his assistance in cholorphyl analysis. I especially want to thank the Florida State Coastal and Marine Lab for the use of their facilities, as well as Chris Peters and the Academic Dive Program for use of reseach gear and assistants. I thank the TH Stone Memoral St. Joseph Peninsula State Park and the St. Joseph Bay State Buffer Preserve Center. I also want to thank all of my amazing field and laboratory assistants for their help in project development and implementation including Sandra Brook, Kevin Olson, Jackson Powell, Abbey Engleman, Jose Moscoso, Kate Hill, Alex Strawhand, Kelly Vasbinder, and Jonathan Eskeldson. I would also like to thank the Undergraduate Reseach Opportunity Program at FSU for providing materials grants and for facilitating undergraduate research, without which I would not have many amazing students that aided in field experiments, data analysis, and spicule measurements and preparations including Shannon Conley, Anna Wuest, Alivia Schnoering, Kristie Dick, Kelsey Alter, Michael Swain, and Matthew Norton. I also want to thank the other undergraduates who aided in data collection and analysis including Connor O’Halloran and Dakota Owings. This research was funded by the PADI Foundation (2016 Award, Application number 21857) and the Florida State Coastal and Marine Laboratory Graduate Scholarship Fund, as well as the UROP materials grant. Special thanks to Kate Hill for her never-ending support, advice, and encouragement as a fellow lab mate and friend. A big thank you also to the enthusiastic academic and social support from the biology graduate student community past and present, particularly EERDG, Natalie Ramirez-Bullon, Katie Pearson, Jose Moscoso, Alex Hooks, Jackson Powell, Alex Strawhand, and many others. Finally, thank you to my family and friends for their unconditional encouragement, and for their sincere interest in my research.

TABLE OF CONTENTS

List of Tables ...... vii

List of Figures ...... viii

Abstract ...... xi

CHAPTER 1 INTRODUCTION ...... 1

CHAPTER 2 SHIFTING DISTRIBUTIONS OF NORTHEASTERN GULF OF MEXICO SPONGES AND COMPARISONS WITH CARIBBEAN FAUNA ...... 4

Introduction ...... 4

Methods...... 7

Sampling locations, collection, and species identification ...... 7

Comparisons and confirmation of museum specimens ...... 8

Northeastern Gulf of Mexico shifting distributions of common species ...... 8

Wider-scale faunal comparisons between the Northeastern Gulf of Mexico, Gray’s Reef (GA, U.S.A), and the Caribbean Sea ...... 9

Results ...... 10

Common species reported in the Northeastern Gulf of Mexico from the present study ...... 10

Northeastern Gulf of Mexico shifting distributions of common species ...... 20

Wider-scale faunal comparisons between the Northern Gulf of Mexico, Gray’s Reef (GA, U.S.A), and the Caribbean Sea ...... 27

Discussion ...... 36

Northeastern Gulf of Mexico shifting distributions of common species ...... 36

Wider-scale faunal comparisons between the Northeastern Gulf of Mexico, Gray’s Reef (GA, U.S.A), and the Caribbean Sea ...... 39

CHAPTER 3 ASPECTS OF ECOLOGY THAT ALLOW A SESSILE SPONGE WITH A LIMITED DISTRIBUTION TO THRIVE IN TWO DISSIMILAR HABITATS ...... 42

Introduction ...... 42 v

Methods...... 45

Study area...... 45

Clonal diversity and recruitment ...... 46

Susceptibility to predation ...... 48

Predator avoidance strategies ...... 48

Seasonal abundance ...... 49

The effect of shading on specific growth and symbiont abundance ...... 49

Results ...... 51

Clonal diversity and recruitment ...... 51

Susceptibility to predation ...... 51

Predator avoidance strategies ...... 52

Seasonal abundance ...... 52

The effect of shading on specific growth and symbiont abundance ...... 54

Discussion ...... 55

CHAPTER 4 CONCLUSION...... 59

Appendix A Supplementary Information ...... 62

References ...... 71

Biographical Sketch ...... 77

LIST OF TABLES

Table 1: Common species in the Northeastern Gulf of Mexico (present study) and past studies. Previously common species reported by Little, de Laubenfels, and Storr from 1950-1963 and currently common species from this study’s collections. Red boxes denote species missing or less common in a study. ¹ Newly reported species. ² Cryptic species Lissodendoryx spinulosa revealed through voucher specimens of L. isodictyalis. ³Missapplication of species name ...... 20

Table 2: Comparison of common Caribbean fauna. Caribbean species reported in Freeman et. al (2007) and Hopkinson et al (1991) at Gray’s Reef (GA, U.S.A) and the Northeastern Gulf of Mexico (present study) and those reported commonly throughout the Caribbean. Red boxes denote species missing from the location. Numbers are references for Caribbean species locations[1:(Alvarez et al. 1990); 2: (Wulff and Swain 2004); 3: (Villamizar et al. 2014); 4: (Alcolado 1990); 5: (Wulff 2006a); 6: (Schmahl 1990); 7: (Wulff 2006c); 8: Keys (personal observation); 9: (Wulff 2013); 10: (Wulff, Personal communication); 11: (Wulff 2009); 12:(Diaz et al. 1993); 13:(de Laubenfels 1936); 14:(Wiedenmayer 1977); 15:(Diaz 2005); 16:(Rützler et al. 2014); 17:(Hartman 1955)] ...... 29

Table 3: North West Atlantic species reported in Freeman et. al (2007) and Hopkinson et al (1991) at Gray’s Reef (GA, U.S.A) and the Northeastern Gulf of Mexico (present study). Red boxes denote species missing from the location...... 34

Table 4: Caribbean fauna comparisons between species reported in Freeman et. al (2007) and Hopkinson et al (1991) at Gray’s Reef (GA, U.S.A) and the Northeastern Gulf of Mexico (present study). Red boxes denote species missing from the location...... 34

Table 5: Collection site information from past and current collections...... 62

vii

LIST OF FIGURES

Figure 1: Study collection sites past and present. Yellow markers denote Little (1963) sampling sites in the northern Gulf of Mexico. Green markers denote de Laubenfels (1953) sampling sites along the Florida coast. Current study collections represented by purple and blue markers. Kaiser collections (purple) were performed in multiple habitat types. Paulay/Wulff collections were performed in March 2011 by Brendan Biggs, Anna Strimaitis, and Nicole Roca in a cruise collaboration...... 6

Figure 2: Storr (1964) historical collection sites along the Florida Gulf coast sites and range map for common species denoting the northern and southern boundaries reported by Storr (1976) based on habitat, current, and temperature patterns...... 7

Figure 3: Wider-scale comparisons. Comparison areas represented by pink shaded regions. Northeastern Gulf of Mexico are represented by the present study collections. Gray’s Reef off the Georgia coast are represented by Freeman et al. (2007) and Hopkinson et al (1991). Common species in the Caribbean Sea were based on common fauna reported by 17 studies from various regions of the Caribbean...... 10

\Figure 4: Distribution of Niphates erecta along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 22

Figure 5: Distribution of Haliclona curacaoensis along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 23

Figure 6: Distribution of Lissodendoryx sigmata along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 23

Figure 7: Distribution of Cinachyrella alloclada along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 24

Figure 8: Distribution of Geodia gibberosa along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 25

Figure 9: Distribution of Ircinia campana along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 26

viii

Figure 10: Distribution of Cliona varians along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 27

Figure 11: Distribution of Stelletta grubii along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present...... 27

Figure 12: Experimental locations. The western seagrass sites in St. Joseph Bay, FL and the eastern dock site in Shell Point, FL marked with stars...... 46

Figure 13: Tissue fusion experimental locations. At the seagrass sites (a) tissue grafts were performed within a site (purple markers) and between the sites shown with arrows. At the dock site (b) tissue grafts were preformed between dock slips and within a dock slip shown with arrows...... 47

Figure 14: A) Average time of E. spinulosus approach. The average time for E. spinulosus to approach a sponge to attempt feeding did not significantly differ whether the sponge was attached to a pen shell or a decorator crab (P=0.0297). B) Average time for E. spinulosus to begin feeding. The average time for E. spinulosus to successfully begin feeding on a sponge attached to a decorator crab was significantly higher than those attached to a pen shell (P<0.001). Error bars represent standard error...... 52

Figure 15: Seasonal changes of organism abundance, average water temperature, and grass cover from census site in St. Joseph Bay. Top graph displays monthly fluctuations in decorator crab, sponge Halichondria corrugata, and predatory sea star Echinaster spinulosus abundance. Middle and bottom graphs display fluctuations in average temperature and average % grass cover within the census site. Error bars represent standard error...... 53

Figure 16: Difference in chlorophyll a amounts between shaded and unshaded treatments. Shaded sponges experienced significantly higher loss of chlorophyll a (P=0.012843). Error bars represent standard error...... 54

Figure 17: Difference in chlorophyll a amounts between shaded and unshaded initial and final treatments. Letters denote significant differences (P<0.005). Error bars represent standard error ...... 55

Figure 18: Specific growth was significantly higher for unshaded sponges (n=17) than for shaded sponges (n=17) (Paired t-test, P=0.04). Error bars represent standard error...... 55

Figure 19: Modern sea surface temperature map of the Gulf of Mexico and Florida Atlantic coast. Image and data retrieved from NOAA OSPO (http://www.ospo.noaa.gov/data/sst/contour/gulfmex.cf.gif) ...... 65

Figure 20: Current and past distribution of Geodia gibberosa along the Florida coast. Green circles denote current collections. Previous collections are represented by gray diamonds. Gray shaded areas represent zones this species occurs as reported by Storr (1976)...... 66

Figure 21: Current and past distribution of Ircinia campana along the Florida coast. Green circles denote current collections. Previous collections are represented by gray diamonds. Gray shaded areas represent zones this species occurs as reported by Storr (1976)...... 67

Figure 22: Past distribution of Cliona varians along the Florida coast. Previous collections are represented by gray diamonds. Gray shaded areas represent zones this species occurs as reported by Storr (1976). This species was not reported in the present study...... 68

Figure 23: Probability of presence of Geodia gibberosa as a function of latitude in the past (A) and the present (B). The probability of successfully occurring at varying latitudes is represented as the solid red line. The bars represent the frequency of presence or absence at a given latitude. Probability of presence shows no significant effect of latitude in the past (P=0.7917,df=35 , F=0.0698), however there was an effect of latitude from the present (P=0.005617 ,df=43 , F= 7.6693)...... 69

Figure 24: Probability of presence of Ircinia campana as a function of latitude in the past (A) and the present (B). The probability of successfully occurring at varying latitudes is represented as the solid red line. The bars represent the frequency of presence or absence at a given latitude. Probability of presence shows no significant effect of latitude in the past (P=0.1752,df=35 , F= 1.9168), however there was an effect of latitude from the present (P= 0.002119 ,df=43 , F= 10.728)...... 69

Figure 25: Probability of presence of Spheciospongia vesparium as a function of latitude in the past (A) and the present (B). The probability of successfully occurring at varying latitudes is represented as the solid red line. The bars represent the frequency of presence or absence at a given latitude. Probability of presence shows no significant effect of latitude in the past (P=0.6188 ,df=35, F= 0.2521), however there was an effect of latitude from the present (P=0.002986, df=43, F= 8.8159)...... 70

Figure 26: Probability of presence of Ircinia felix as a function of latitude in the past (A) and the present (B). The probability of successfully occurring at varying latitudes is represented as the solid red line. The bars represent the frequency of presence or absence at a given latitude. Probability of presence shows no significant effect of latitude in the past (P= 0.1747 ,df=35, F=1.9219), however there was an effect of latitude from the present (P= 0.006367,df=43, F=7.4433)...... 70

ABSTRACT

The Florida Gulf of Mexico coast extends over both tropical and subtropical zones resulting in an intermingling of fauna typical to both zones. Cold winter water temperatures historically limited the distribution of many tropical species and allowed sub-tropical species to thrive. In the past 60 years, average winter sea surface temperatures have increased 2-3°C in the Northern Gulf of Mexico (NOAA OSPS), potentially allowing range expansion of tropical species and local extinction of subtropical species. While poleward range expansion is possible for species along the Atlantic coast, species in the Gulf of Mexico face land barriers that prevent northward movement. Distribution patterns of sponges in the Northeastern Gulf of Mexico have changed since previous studies in the 1950’s and 1960’s. A little over half of the common species (56%) are still widespread throughout the Northeastern Gulf of Mexico. Eleven species (44%) previously reported as widespread are either missing entirely or are no longer widespread throughout the region. Two species are newly reported in the Northeastern Gulf of Mexico (Niphates erecta and Haliclona curacaoensis) and eight other species are more widespread than before. Caribbean species make up most of the common species composition of the Northeastern Gulf of Mexico (de Laubenfels 1953, Little 1963, Storr 1976, current study). Due to the limited latitudinal range of the Northeastern Gulf of Mexico it is necessary to look at a wider scale to determine species’ northern range limits. When compared to the wide-scale Caribbean fauna, it appears that only a handful of Caribbean species occure in the Gulf of Mexico and the North West Atlantic. Of 90 common Caribbean species only 27% (24 species) were found in the Northeastern Gulf of Mexico and 38% (34 species) at Gray’s Reef in Georgia U.S.A. while 53% (48 species) were not found in either the Northeastern Gulf of Mexico or Gray’s Reef (Hopkinson et al. 1983, Freeman et al. 2007). While the species composition of Caribbean sponges at Gray’s Reef and in the Northeastern Gulf of Mexico is similar, each contain unique species that are not found at the other location. This may be due to a combination of biotic and abiotic factors that differ between the Atlantic and Gulf coasts of Florida, aiding the migration of some sponge species while hindering others.

One species commonly found in the Northeastern Gulf of Mexico is endemic to the region. Halichondria corrugata Diaz, van Soest & Pomponi, 1993 is the most abundant sponge in shallow seagrass beds and on dockside fouling communities in the Northeastern Gulf of Mexico. The variation in abiotic conditions between these habitats and H. corrugata’s ubiquity make it an interesting system to test what aspects of ecology allow it to be so successful. H. corrugata utilizes different ecological strategies in seagrass and dock habitats. H. corrugata is flexible in its reproductive strategy, utilizes predator avoidance techniques when needed, and seems to be resilient to light variation despite being reliant on photosynthetic symbionts for supplemental nutrition.

xii

CHAPTER 1

INTRODUCTION

Sponges provide a variety of different ecosystem services in their habitats. They provide shelter for many organisms such as juvenile spiny lobster (Butler et al. 1995) and can serve as a “living hotel” for a multitude of small invertebrates (Pearse 1950, Ribeiro et al. 2003). Sponges also perform geological roles on coral reefs by protecting and stabilizing the exposed calcium carbonate undersides of corals from boring organisms (Wulff and Buss 1979). Sponges are potent filter feeders – they are one of the only organisms that can filter bacteria sized particles directly from the water column and they accomplish this through their specialized feeding cells, choanocytes, that are lined with a fine microvilli collar to trap fine food particles from the water column (Reiswig 1971, 1975). The efficiency of sponge filtering has been linked to controlling phytoplankton blooms in Florida Bay (Peterson et al. 2006).

Sponges have high species diversity with over 8500 recognized species worldwide, 945 of which are tropical western Atlantic species (Van Soest et al. 2012), Every year there are 35-87 new species described worldwide (Van Soest et al. 2012). This continual update in taxonomic understanding has made sponge identification challenging. This is compounded with species of different genera and families often sharing physical characteristics such as body form and color, making field identification difficult. Skeletal analysis of silica spicules and spongin fibers is critical in order to properly identify a sponge (Rützler 1978). The arduous task of sponge identification has resulted in a gap in the knowledge of species distributions.

The Florida Gulf of Mexico coast extends over both tropical and subtropical zones resulting in an intermingling of fauna typical to both zones. Cold winter water temperatures were predicted, historically, to limited the distribution of many tropical species and allow sub-tropical species to thrive (Storr 1976). In the past 60 years, average winter sea surface temperatures have increased 2-3°C in the Northern Gulf of Mexico (NOAA OSPS), potentially allowing range expansion of tropical species and local extinction of subtropical species. While poleward range 1 expansion is possible for species along the Atlantic coast, species in the Gulf of Mexico face land barriers that prevent northward movement.

Tropicalization of temperate and sub-tropical ecosystems seems to be occurring around the world (Vergés et al. 2014). Tropical herbivores have increased in sub-tropical regions, like the Northern Gulf of Mexico (Fodrie et al. 2010), due to increased temperatures. This in turn causes increased grazing pressure and phase shifts (Vergés et al. 2014). Determining which tropical Caribbean sponge species have expanded their ranges and which sub-tropical North West Atlantic species have retained their ranges could shed light on our understanding of how environmental factors affect species distributions. It can also provide information on individual species’ abilities to undergo latitudinal shifts in response to increasing ocean temperatures and persist in new habitats.

The Northern Gulf of Mexico is particularly under-researched despite the high abundance of sponges in the area (Rutzler et al. 2009). Only three studies have focused on species distribution patterns for sponges (de Laubenfels 1953, Little 1963, Storr 1976). Thus, I ask how those distributions may have changed with changing temperatures. Using historical studies as a reference, I compare the current distribution of common species with reported distribution patterns 60 years prior. The goals of the first chapter of this dissertation are to 1) compare common sponge species distributions in the Northeastern Gulf of Mexico from the past and present and 2) compare sponge communities among the Northeastern Gulf of Mexico, Gray’s Reef (GA, U.S.A), and the Caribbean Sea to see wider-scale patterns of species range limits.

Other abiotic and biotic factors such as habitat availability, habitat continuity, and water current patterns were also predicted to influence sponge species range limits (Storr 1976). Smaller scale habitat distribution patterns can be determined by habitat availability, competition, and predator abundance (Wulff 2017). In order to test what ecological aspects explain habitat and geographic distributions and abundance of species, I selected a species with a limited geographic distribution but a wide habitat distribution between dissimilar habitats. One species commonly found in the Northeastern Gulf of Mexico is endemic to the region. Halichondria 2 corrugata Diaz, van Soest & Pomponi, 1993 has only been found in shallow waters in the Northeastern Gulf of Mexico. The limited geographic distribution of this species combined with its ubiquity in the Northeastern Gulf of Mexico make this an ideal species to understand flexibility in strategies that may explain distribution patterns. Unlike most sponge species that specialize their strategies to be successful in a particular habitat, H. corrugata is abundant in both seagrass meadows and dockside fouling communities. The overall goal of the second chapter was to document what role reproduction strategy, predation, and light availability play for H. corrugata in two dissimilar habitats where it is abundant.

CHAPTER 2 SHIFTING DISTRIBUTIONS OF NORTHEASTERN GULF OF MEXICO SPONGES AND COMPARISONS WITH CARIBBEAN FAUNA

Introduction

Demonstrations of shifting species ranges have primarily been in temperate and polar regions, with only a handful of tropical marine studies (Parmesan et al. 1999, Bellard et al. 2012, Vergés et al. 2014, Poloczanska et al. 2016). Ranges for individual species can shift entirely e.g poleward range shift since the 1960’s of the blue mussel Mytilus edulis shifting both its northern and southern range limit poleward (Jones et al. 2010), or they can expand while retaining historical locations e.g. mangroves expanding into salt marshes due to freeze-free winters (Guo et al. 2013) and reef corals off the coast of Japan expanding poleward corresponding to a 0.5°C increase in average winter sea surface temperatures (Yamano et al. 2011). With increasing water temperatures we would predict tropical species to expand their ranges polewards, assuming habitable conditions still exist at their historic ranges.

The Florida Gulf of Mexico coast extends over both tropical and subtropical zones, resulting in an intermingling of fauna typical to both zones. While poleward range expansion is possible for species along the Atlantic coast, species in the Gulf of Mexico face land barriers that prevent northward movement. The dramatic increase in winter sea surface temperatures of 2-3°C since the 1950’s (NOAA OSPO) could potentially allow northward range expansion for tropical species. Another consequence could be local extinction of sub-tropical species due to either abiotic or biotic factors, as well as the potential extinction of species endemic to the Northern Gulf of Mexico. Even if extinction is not occurring there could be dramatic effects on local processes and population drivers for species with a limited range (Vergés et al. 2014). Determining which tropical Caribbean species expand their ranges and which sub-tropical North West Atlantic species can retain their range could shed light on our understanding of environmental factors affecting species ranges and individual species’ abilities to shift their ranges and persist in new habitats.

The Northern Gulf of Mexico is particularly under-researched, with only three studies on species distributions for sponges. Previous studies documenting species distributions were performed in 1948, 1955-1956, and 1963. De Laubenfels (1953) sampled from 22 stations, in 1948, which ranged from the Ochlocknee River along the Big Bend to the Dry Tortugas (Figure 1). Storrs (1976) study consisted of three extensive dive surveys, from 1955-1956, and communication with local fishermen from the ten thousand islands to St. Marks (Figure 2). Little’s survey in 1963 focused on the Northern Gulf of Mexico, specifically Apalachee Bay (Little 1963), including 18 stations near and offshore (Figure 1). These three studies comprise our understanding of the historical distribution of sponges along the Florida coast. Storrs’ study (1976) also recorded the sponge species distributions in a north-south gradient along the Gulf Coast of Florida from St. Marks to Tampa Bay. Storr distinguished unique zonation patterns for each sponge species that reflect water current patterns, temperature, and the availability of hard substratum that change as one moves north or south along the Gulf Coast (Storr 1976, Figure 2). Cold winter water temperatures limit the distribution of many tropical species by affecting sponge reproduction and size (Storr 1964). The rapid decline in mean low water temperature as you move Northward in the Gulf of Mexico was suggested to explain the trend in decreasing species diversity from South Florida Bay to the Northern Gulf of Mexico (Storr 1976).

The goals of this project are to 1) document current Northeastern Gulf of Mexico common species patterns, and compare those to historical occurrences, in order to determine if species have shifted their distributions and 2) document wider geographic scale faunal comparisons between the Northeastern Gulf of Mexico, Gray’s Reef (GA, U.S.A) on the Atlantic coast, and the Caribbean Sea to determine the northern range limit of different species common in the Caribbean, and to see if northern range limits differ between the Florida Gulf and Atlantic coasts. To accomplish the first goal I have re-surveyed historical and additional collection locations and mapped the current common and widespread species to compare with previous findings by Storr, Little, and de Laubenfels. For the second goal, I have compared my findings from the first aim with surveys conducted at Gray’s Reef along the Atlantic coast in Georgia, U.S.A (Hopkinson et al. 1983, Freeman et al. 2007) and the wider Caribbean Sea where I recorded 1) how many Caribbean species are found in the Gulf of Mexico and up the Atlantic

5 coast, 2) how many North West Atlantic species are present in both sites, and 3) how the two locations are similar and how they are different in respect to species composition.

Methods

Sampling locations, collection, and species identification

To document the current species ranges I collected from historical collection sites recorded in previous studies (Figure 1, Figure 2), and additional sites to incorporate artificial reefs and a variety of habitats (Figure 1). Site locations and characteristics are reported in Table 5. These additional sites were the results of independent sampling by myself and collections from a research cruise conducted in March 2011 by Brendan Biggs, Anna Strimaitis, and Nicole Roca in a combined cruise performed by the Gustav Paulay and Janie Wulff research labs (Figure 1).

At each location, sponge species were photographed and a small piece of each species was collected and stored in 75% ethanol for identification. Once all the species at a location were collected, sweep dives were performed over the site to ensure no species had been missed in the original collection. Identifications based on photos and skeletal analysis were made for the most common species by survey collectors including myself, Janie Wulff, and Brendan Biggs.

Collection specimens will be stored at the University of Miami Marine Invertebrate Museum collection (code UMML).

Comparisons and confirmation of museum specimens

To ensure new recorded locations are not an artifact of updated species discriminations, I verified species identifications of voucher specimens from the Little, Storr, and de Laubenfels collections. These specimens were obtained from the University of Miami Marine Invertebrate Museum collection (code UMML) and the Peabody museum collection (code YPM).

Northeastern Gulf of Mexico shifting distributions of common species

Common species were identified and documented at each site. Species were assigned as common or widespread if they occurred at high abundance at three or more sites in the Northeastern Gulf of Mexico. Abundance was visually estimated at a site based on the number of individuals collected and documented in photo sweeps. Species determined as widespread were reliably found at multiple locations. Species from de Laubenfels (1953) and Little’s (1963) collections were assigned to be common if they occurred in three or more sites and authors remarked on their high abundance at a given site. Storr’s (1976) samples were categorized into rare, widely dispersed and common within a given range. Species that were common within multiple zones and those described as widely dispersed by Storr were included as common Northeastern Gulf of Mexico Species. Past and present common species lists were then compared to determine which species are still widespread, which species may be currently missing or less common, and which species have newly colonized the Northeastern Gulf of Mexico or are more widespread than reported previously. Species in the newly reported or newly widespread and the less widespread or missing categories were then cross checked with literature and current collections to determine if and how the distribution has changed since previous studies.

Further analysis was performed for 10 species in this category to determine patterns of ranges shifts that may not be captured by general categorization. To see if the probability of occurrence changed with latitude in the past and present studies, I used a generalized linear model (GLM) with a binomial distribution, where the response variable is presence of a species, and the explanatory variable is latitude, which is continuous. Since I’m treating it as continuous I used a regression to analyze my data using the glm function in R. To test if there was a change in the effect of latitude over time, I used a generalized linear model (GLM) with a binomial distribution, where the response variable is presence of a species, and the explanatory variable is latitude as a function of time. All analyses were conducted using the free software R. (Core Team 2015), and all models were checked for overdispersion and changed to a quasibinomial distribution if necessary. Four species were not analyzed due to a lack of previous records on specific locations either due to survey reporting method, or later discovered cryptic specietion.

Wider-scale faunal comparisons between the Northeastern Gulf of Mexico, Gray’s Reef (GA, U.S.A), and the Caribbean Sea

Results from survey collections in the Northeastern Gulf of Mexico were compared with surveys of Gray’s Reef by Freeman et al. (2007) and Hopkinson et al (1991). Gray’s Reef was chosen for comparison since, even though it is at a higher latitude, there are similar temperature regimes between the Northeastern Gulf of Mexico and Gray’s Reef due to the diagonal isotherms (Figure 3, Figure 19). These two locations were also compared with a list of common Caribbean species I gathered from surveys throughout the Caribbean. 90 species were included based on common fauna reported by 17 studies from various regions of the Caribbean (Table 2). Species from the Northeastern Gulf of Mexico and Gray’s Reef were classified as North West Atlantic species if their reported distribution was between 26.75N, 79.91W and 41.25N, 70.33W or as Caribbean species if reports of the species distributions were confined to the Caribbean Sea.

Fauna from each of the three regions (Figure 3) were then compared to see how many and which Caribbean species are found in the Northeastern Gulf of Mexico and Gray’s Reef. The number and identity of North West Atlantic species at Gray’s Reef and the Northeastern Gulf of

Mexico were also compared. Finally the fauna between Gray’s Reef and the Northeastern Gulf of Mexico were compared.

Results

Common species reported in the Northeastern Gulf of Mexico from the present study

Axinella polycapella de Laubenfels, 1953. Commonly called orange devil’s fingers, this is a vibrant orange to red branching sponge. Spicule sizes and types match A. polycapella reported by Alvarez, van Soest, and Rützler (1998). This species was found on natural limestone reefs and artificial reefs in the Northeastern Gulf of Mexico in the recent survey by myself at site code 40,

43, and 54; and in the past studies by Little (1963) at site 12 (site code 12) and by de Laubenfels (1953) at site 1, 4, 5, 6, 9, 10, 11, 13, 18, 19, 20, 21, and 22 (site codes 15, 18, 19, 20, 23, 24, 25, 27, 32, 33, 34, 35, and 36) and in regions C, D, E, G, H, and I (Figure 2) by Storr (1976) (Table 5).

Axinella pomponiae Alvarez, van Soest, & Rützler, 1998. A red branching Axinella, commonly called red devil’s fingers. Samples were also compared with A. waltonsmithi, and A. polycapella, however styles matched A. pomponiae as did the color description reported by Alvarez, van Soest, and Rützler (1998). This species was described as Homaxinella rudis by Storr and de Laubenfels as noted by skeletal analysis by Alvarez et al (1998). This species was found on artificial reefs and mixed limestone reef and seagrass sites. This species was reported at site code 44, 46, 53, 54, 57, 59, 70, and 72 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by de Laubenfels (1953) at station 4 (site code 18) and as widely distributed by Storr (1976) (Table 5).

Chondrilla (cf) caribensis Rützler, Duran & Piantoni, 2007. Described as Chondrilla nucula Schmidt, 1862 by Little (1963) this species has been attributed as a misapplication of the name on the World Porifera Database. The specimens described by Little and de Laubenfels match the specimens collected by the present study. This sponge is often called the “chicken liver” sponge due to its color and texture. This is a flat lobate sponge that differs in texture from most sponges. The surface is smooth and slippery with the color varying from white to a light tan or brown, often with molted patches of both colors. There are no megascleres, only spherasters. Spheraster sizes matched those reported by Little (1963) with two size classes resulting in the overall range reported. The larger size class averaged about 25µm in diameter and the second averaged about 15 µm in diameter. This sponge was found on natural limestone and artificial reefs, often cryptic within the reef frame and under ledges. This species was reported at site code 41, 49, 51, 53, 54, 55, and 67 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 10 (site code 10) (Table 5).

Cliona celata Grant. 1826. This common yellow boring sponge was commonly found encrusting on natural limestone and artificial reefs throughout the Northeastern Gulf of Mexico. When the 11 species outgrows its burrow it forms tall, massive, cylindrical, papillate chimneys as described by Little (1963). Large individuals currently found were roughly 25cm in maximum height. This species is commonly found in the Adriatic and Mediterranean Sea, so it is possible the specimens in the Northeastern Gulf are a cryptic species. This species has not been reported in the wider Caribbean Sea. Past and current samples fit within the size ranges for C. celata and C. viridis, however color and morphology descriptions support C. celata (Little 1963). This species was reported at site code 39, 40, 41, 43, 44, 46, 47, 52, 53, 55, and 80 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 1, 2, and 3 (site codes 1, 2, and 3) (Table 5).

Spheciospongia vesparium (Lamarck 1815). This sponge is commonly referred to as the loggerhead sponge and is usually a massive black sponge. The yellow version of the species, Spheciospongia vesparium pallida Vicente, Rützler & Carballeira, 1991, was also found. De Laubenfels remarked on finding a bright yellow version of the loggerhead sponge and believed there to be a yellow race in northern gulf regions (Vicente et al. 1991). Vicente and Rützler also remarked on the differences within Spheciospongia vesparium. They described two forms: the typical form and a yellow form dubbed pallida. Both were collected from Puerto Rico and they commented that the yellow version mostly differs in external morphological characteristics and in habitat. S. papillosa is the only other similar sponge species reported in the Gulf of Mexico (Rützler et al. 2009) which did not match our specimens based on close scrutiny of the spirasters. This species was reported at site code 40, 43, 46, 48, 49, 50, 52, 53, 54, 55, and 74 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 1, and 10 (site codes 1, and 10), by de Laubenfels (1953) at 6, 8, 10, 13, 20, 21 (site codes 20, 22, 24, 27, 34, and 35) and widely distributed by Storr (1976) (Table 5).

Placospongia intermedia Sollas, 1888. Placospongia species are typically encrusting with a hard, stony outer tissue layer containing cracks and ridges. Specimens found matched descriptions of P. carinata identified by Little (Little 1963). This species is no longer accepted for the Gulf of Mexico based on inaccurate records, however, according to the Word Porifera Database. P. melobesoides, which is a species de Laubenfels recorded (de Laubenfels 1953), was also evaluated, but the spicule sizes are much too large and is typically found in deeper waters. 12

This species was reported at site code 43, 44, 48, 68, 74 and 75 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 10 (site codes 10) as P. carinata, and P. melobesoides was reported by de Laubenfels (1953) at 8 (site code 22) (Table 5).

Spirastrella cf. coccinea (Duchassaing & Michelotti, 1864). There are three Spirastrella species that we focused on in the Caribbean but all of them have different color descriptions. S. mollis and S. hartmani are salmon brown, and S. coccinea is a vermillion red color, none of which seem to match our specimens, which are typically red-orange. Current specimens closely resemble spicule sizes reported for S. coccinea by Boury-Esnault et al. (1999). Duchassaing & Michelotti first described this species as a dramatic red color, however other reports have called it an orange color. Little (1963) describes S. coccinea as orange in life, and remarked on differences in spicule ranges but left his specimen as S. coccinea due to more similarities than differences. S. coccinopsis was also reported by de Laubenfels (1953) but not found by Little (1963), who remarked how de Laubenfels believed it to be similar to S. coccinea since it differs primarily in color and size. Spirastrella coccinopsis, coccinea, and phyllodes are the only species in the Rützler chapter on sponges in the Gulf of Mexico (Rützler et al. 2009) and S. phyllodes is currently under review on the World Porifera Database. We use S. coccinea with the caveat that the spicules are a close but not great match, and the color is different than the original description for the species. This species was reported at site code 43, 44, 46, 51, and 68 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 7 (site code 7), S. coccinopsis was reported by de Laubenfels (1953) at site 20 (site code 34) and in region G (Figure 2) by Storr (1976) (Table 5).

Igernella notabilis (Duchassaing & Michelotti, 1864). Previously known as Darwinella joyeuxi Topsent, 1889, this sponge is a reddish-pink color with a soft and spongy, commonly found throughout the Northeastern Gulf of Mexico. One of the other potential species we looked into was Darwinella mulleri as described by de Laubenfels in 1950. Little suggested that these two species may in fact be one species (Little 1963). Cross-checking with the World Porifera Database revealed an inaccurate recording of D. mulleri in the Gulf of Mexico and a genus transfer for D. joyeuxi, so we are referring to our species as I. notabilis. This species was 13 reported at site code 39, 40, 41, 43, 44, 45, 46, 49, 50, 51, 53, 54, 63, and 73 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 4, 8, and 13 (site codes 4, 8, and 13), by de Laubenfels (1953) at station 20 (site code 34) and in regions A, C, and F (Figure 2) by Storr (1976) (Table 5).

Hippospongia lachne (Laubenfels, 1936). The “sheepswool sponge”, commonly collected in commercial sponge fishing, occurs along the big bend region in mixed limestone and seagrass habitats. This species was reported at site code 49, 50, 51, and 54 in the recent survey by myself in the Cedar Key and Tarpon Springs regions; and in the past studies by de Laubenfels (1953) at station 18, and 22 (site code 32, and 36) and classified as “widely distributed” by Storr (1976) (Table 5).

Ircinia campana (Lamarck, 1814). A distinctive vase-shaped Ircinia that appears grey, sometimes with reddish patches. Current collections matched those described by Little (1963) and de Laubenfels (1953). This species was abundant at most collection locations and widely distributed in limestone and mixed seagrass habitats. This species was reported at site code 39, 40, 41, 43, 44, 45, 46, 47, 49, 50, 51, 52, 53, 54, 55, 59, 67, 74, and 75 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 1, 4, 10, and 12 (site codes 1, 4, 10, and 12), by de Laubenfels (1953) at stations 5, 6, 7, 8, 10, 11, 13, 14, 17, 19, 20, and 21 (site codes 19, 20, 21, 22, 24, 25, 27, 28, 31, 33, 34, and 35) and in regions C, D, and E (Figure 2, Figure 21) by Storr (1976) (Figure 9, Table 5).

Ircinia felix (Duchassaing & Michelotti, 1864). Specimens from the current study were brownish-white in color. The shape varied from massive to lobate at all locations. This species was commonly found on limestone reefs, artificial reefs and seagrass habitats throughout the study area. Previous collections identified this species as Ircinia fasciculata (Pallas, 1766) sensu de Laubenfels, 1949 which has since been accepted as I. felix according to the World Porifera Database. This species was reported at site code 39, 40, 41, 43, 44, 46, 51, 54, 55, and 67 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 1, 2, 10, 11, and 12 (site codes 1, 2, 10, 11, and 12), by de Laubenfels (1953) at stations 4,

13, 14, 19, 20, and 22 (site codes 18, 27, 28, 33, 34, and 36) and classified as “widely distributed” by Storr (1976) (Table 5).

Callyspongia (Cladochalina) vaginalis (Lamarck, 1814) The tube sponge is white in appearance, occasionally with tan or grey brown patches, often growing in a ramose-like fashion, unlike larger individuals in the Caribbean. Spicule sizes matched those reported by Little (1963). Some specimens harbored zooanthids in the outer tissue layers. This species was found on limestone reefs, artificial reefs, and in mixed limestone and seagrass habitats. This species was reported at site code 40, 43, 45, 49, 50, 51, 53, 54, 64, 70, 72, and 73 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 4, and 12 (site codes 4, and 12), by de Laubenfels (1953) at stations 10, 14, 16, and 20 (site codes 24, 28, 30, and 34) and in regions C, D, E, G, H, and I (Figure 2) by Storr (1976) (Table 5).

Haliclona (Rhizoniera) curaҫaoensis (van Soest, 1980). This is a light purple-blue-gray sponge consisting of all oxea from 85-138µm. Our particular specimen was found frequently, encrusting on pen shells in shallow seagrass beds, and on limestone and artificial reefs. Five Haliclona species were evaluated for our specimens based on spicule shape and relative sizes (de Weerdt 2000); H. curaҫaoensis, H. melana, H. vermeuleri, H. lehnerti, and H. mucifibrosa,. We narrowed the list to H. lehnerti, H. vermeuleri, and H. curaҫaoensis after further scrutiny of shape and size along with color. H. curaҫaoensis was the best match. Our sponge had the same texture and slime-like, almost cotton candy-like way of being pulled apart, along with the characteristic color, and oscule shape. This species was reported at site code 38, 39, 40, 41, 43, and 44 in the recent survey by myself in the Northeastern Gulf of Mexico (Figure 5, Table 5). No record of this specimen was found from the past collections or from further scrutiny of museum collections.

Niphates erecta Duchassaing & Michelotti, 1864. This grey-white sponge was commonly found in the Northeastern Gulf of Mexico. Ramose individuals were commonly found with zooanthids embedded within the other tissue layer. Measurements of our specimens (172-227µm) matched those recorded by Sven Zea (161-247 µm) and those of van Soest (154-232 µm) (van Soest 1980, Zea 1987). Previous reports of this specimen in the Gulf of Mexico are in Veracruz 15

Mexico (Green 1977), and the Dry Tortugas (de Laubenfels 1936) (Rutzler et al. 2009). Museum samples of Haliclona variabilis collected by Storr (1976) were analyzed to ensure no previous samples matched Niphates erecta. The current study appears to be the farthest northern confirmed report of Niphates erecta in the Gulf of Mexico. This species was reported at site code 40, 41, 43, 46, 49, 50, 51, 53, 54, 60, 62, 70, and 72 in the recent survey by myself and the Paulay/Wulff cruise (Figure 4, Table 5). No confirmed record of this specimen was found from the past collections or from further scrutiny of museum collections.

Clathria (Clathria) prolifera (Ellis & Solander, 1786). Previously known as Microciona prolifera (Ellis & Solande, 1786), our specimens were lobate and red to orange in color. Clathria prolifera was the only species described, of the evaluated potential species, as red and not entirely an encrusting species (van Soest 1984). The spicules were also a close match though our isochela were smaller, averaging 9.12µm. Toxa and acanthostyles were also larger, averaging 82.2µm and 95.9µm respectively. Current sample subtylostyles ranged from 221 to 372µm. Van Soest’s (1984) reported measurements had subtylostyles 160-252.8-342 and smooth styles from 141-387, acanthostyles 56-74.2-86, palmate isochela 12-14.9-17, and small thin toxa 15-20.6-27. This species was commonly found on limestone reefs. Storr (1976) reports Microciona juniperina as a widespread species, however the identification of these specimens could not be verified at the present time. This species was reported at site codes 39, 40, 49, 51, and 53 in the recent survey by myself in the Northeastern Gulf of Mexico; and in the past studies by Little (1963) at stations 1, 2, and 9 (site codes 1, 2, and 9) (Table 5).

Lissodendoryx (Lissodendoryx) spinulosa Rützler, Piantoni & Díaz, 2007. Specimens found were massive sponges that ranged in color from blue-green to yellow-orange. Spicules consisted of Subtylostyle 132-139-168µm by 1.6-3.68-4.8µm, tylotes 177-194-213µm, two size classes of sigma with I:36-39.47-44µm and II:11-12.57-16µm, and two size classes of isochela I:31.2- 36.87-42.3µm and II:8.8-10.67-12µm Little (1963) reported finding Lissodendoryx isodictyalis, though with a rather large size range for the microscleres, reporting 16-28µm. Our measurements were close to some that Little (1963) reported, however Rützler et al. (2007) made clear the distinctions between L. isodictyalis, L. carolinensis, and L. spinulosa after the time of Little’s (1963) survey. Museums specimens from Little’s (1963) collections could not be re-identified, 16 but analysis of one of Storr’s (1976) samples of L. isodictyalis was confirmed as L. spinulosa by myself. In comparison to L. spinulosa we recorded our specimens with two size classes of anisochela, instead of three. Though the range of means reported has an overlap of the first and second size classes, some specimens do not report the second size class. In the diagnosis, Rützler et al. (2007) defines it with two size classes of isochela and sigmas. L. spinulosa was most similar to our specimens in size range and shape for all the spicule types we found. Many subtylostyles found in our samples displayed small spines on the heads which match the description made by Rützler et al. (2007). This species was reported at site code 44, 46, and 47 in the recent survey by myself in the Northeastern Gulf of Mexico; and in the past studies by Little (1963) reports L. isodictyalis at station 10 (site codes 10), and in regions C, D, E, G, H, and I (Figure 2) by Storr (1976) as L. isodictyalis. (Table 5).

Tedania (Tedania) ignis (Duchassaing & Michelotti, 1864). This is a dull orange sponge with characteristic hairs on the tylote heads. We compared the size ratio between the style and tylotes to differentiate between the species T. ignis and T. klausi. On average our ratios were closer to those of T. ignis so we identified this species as such, with the caveat that the spicules on the whole are smaller than those found in other parts of the Caribbean (Wulff 2006b) with the average Style length of 228µm and the average tylote length of 206µm compared to the Belize average style length of 261µm and the average tylote length of 226µm (Wulff 2006). This species was reported at site code 39, 40, 41, 43, 44, 46, 47, 48, 53 and 54 in the recent survey by myself in the Northeastern Gulf of Mexico; and in the past studies by Little (1963) at station 10, and 11 (site codes 10, and 11), and in regions F, G, H, and I (Figure 2) by Storr (1976) (Table 5).

Lissodendoryx (Anomodoryx) sigmata (de Laubenfels, 1949). This was one of the most abundant species in the present collections. Individuals were typically massive and somewhat encrusting with bright orange coloration that sometimes ranged from a soft yellow or yellow brown to orange. All the specimens lacked styles and had the characteristic large sigma with two size classes of sigma and chela. The descriptions and spicule sizes matched those recorded by Rützler et al. (2007). This species was reported at site code 39, 40, 41, 43, 44, 46, 47, 49, 50, 51, 52, 53, 54, 62, 63, 67, and 75 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 1 (site code 1) as Xytopsene sigmatum (Figure 6, Table 5). 17

Phorbas amaranthus Duchassaing & Michelotti, 1864. This sponge ranged from a fire red to darker maroon in color. Commonly found on limestone reefs this sponge was often encrusting and lobate, with conspicuous arcolate porefields on the surface. The speculation of tornotes, acanthostyles, and arcurate isochela is distinctive (Little 1963). This species was reported at site codes 41, 43, 51, and 53 in the recent survey by myself in the Northeastern Gulf of Mexico; and in the past studies as Merriamium tortugaensis by Little (1963) at station 4 and 7 (site code 4 and 7), and in region C (Figure 2) by Storr (1976) (Table 5).

Halichondria (Halichondria) corrugata Diaz, van Soest & Pomponi, 1993. This is a light green to a darker dull green sponge. Specimens occasionally had spire-like formations. Little reported this sponge as H. panicea and remarks it resembles H. bowerbanki (Little 1963). This sponge was also commonly called the “bread crumb sponge” which H. bowerbanki is sometimes called. Diaz et al. (1993) reports on H. corrugata and describes Halichondria panacea from Little (1963) as a synonymous sponge, describing how it differs from H. panicea in its characteristic grooved oscular chimneys. Both the spicules and the skeleton structure matched those reported by Diaz et al. (1993). This species was reported at site codes 38 and 42 in the recent survey by myself in the Northeastern Gulf of Mexico as well as many additional shallow docks; and in the past studies by Little (1963) at station 2 and 11 (site code 2 and 11) as Halichondria panacea (Table 5).

Tethya sp.. Specimens found were bright orange, spherical, tuberculate sponges that were about 4-6cm in diameter and height. Individuals were found on Limestone reefs in the Northeastern Gulf of Mexico. The species identification for our specimens could not be finalized. Previous Tethya species reported in the region include 2 separate names, each unique to a particular study. Tethya aurantium was reported by Little (1963), and Tethya diploderma by de Laubenfels (1953). Tethya aurantium has been classified as a misidentification by the World Porifera database, since the normal occurrence of this species is in Europe and the Mediterranean. This species was reported at site codes 43, 49, and 51 in the recent survey by myself in the Northeastern Gulf of Mexico; and in the past studies by Little (1963) at station 1 and 10 (site code 1 and 10) as Tethya aurantium and by de Laubenfels (1953) at station 10 (site code 24) as Tethya diploderma (Table 5). 18

Geodia gibberosa Lamarck, 1815. This sponge is often found with other sponges growing on it. Specimens were yellow in color, with a distinctive sieve in the cortex. This sponge was found thinly encrusting and massive or globular on Limestone and artificial reefs. The spicule complement of this sponge firmly put it in the Geodia genus and within that genus we looked at three species; Geodia gibberosa, Geodia papyracea, and Geodia neptuni. Of these, the spicule measurements seemed to closely match G. gibberosa, which is the basis for our identification choice (Hajdu et al. 1992). This species was reported at site code 40, 41, 43, 49, 51, and 53 in the recent survey by myself in the Northeastern Gulf of Mexico; and in the past studies by Little (1963) at station 4, 7, and 11 (site codes 4, 7, and 11), by de Laubenfels (1953) at station 8, 10, 15, and 20 (site codes 22, 24, 29, 34) (Figure 8, Figure 20) ( Table 5).

Cinachyrella alloclada (Uliczka, 1929). Our specimens were yellow and typically ball-shaped and very distinct, with spicules sticking out and small balls of sponge tissue on the middle and ends of the visible spicules projecting from the surface. These characteristics suggested Cinachyrella, and after looking at the surface of the specimens we narrowed it to C. alloclada or C. kuekenthali but the specimen was much too small for C. kuekenthali and at closer inspection was confirmed to C. alloclada. This species was reported at site code 40, 43, 46, 47, 49, 52, 53, 54, 74, and 80 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 13 (site code 13) (Figure 7, Table 5). Cinachyra cavernosa was reported in regions E, G, H, and I (Figure 2) by Storr (1976),

Aplysina (cf) fulva (Pallas, 1766). This common sponge is known for its finger-like branches that are a golden yellow color. A. fulva also has characteristic golden spongin fibers. The specimens we found were the same texture, color and shape as A. fulva, however there are several differences between our samples, which match the description of Little (1963)’s Verongia sp., and those found commonly in the Caribbean. Branch diameters were about 1cm on average and the length exceeded 30cm for some branches, often found intertwined. The sponge surface has a thick outer dermis with irregularly scattered oscules that can appear a dark purple when packed with cyanobacterial symbionts. Pith percentage ranges from 20%-50% for our specimens. Little (1963) reports a long thin ramose verongia that is light brown on the upper surface and a dull yellow on the lower surface, when touching air, the tissue turns black, similar to our specimens. 19

Little (1963) remarked the similarity to de Laubenfels’ (1948) Verongia fulva (=aurea) but found its living coloration, conule arrangement and spacing, dermal thickness, and location of oscules resemble Verongia longissima more closely (Little 1963). This species was reported at site code 40, 43, 46, 47, 49, 52, 53, 54, 74, and 80 in the recent survey by myself and the Paulay/Wulff cruise; and in the past studies by Little (1963) at station 7, 10, 11, and 12 (site code 7, 10, 11, and 12) (Figure 7, Table 5) as Verongia sp. de Laubenfels reports 2 verongids (Verongia longissima and Verongia sp.) at Sites 19 and 21 (site code 33, and 35), however these specimens could not be re-identified to confirm modern taxonomic placement. Aplysina longissima was reported in regions C, D, E, G, H, and I (Figure 2) by Storr (1976). Overall the taxonomic record for this species is muddled with 2 different genus names and three different species names, requiring further analysis to confidently determine characteristics and differences between the current Aplysina fulva, Aplysina cauliformis, and our specimens in the Northeastern Gulf of Mexico that match A. fulva in many, but not all characteristics.

Northeastern Gulf of Mexico shifting distributions of common species

Order Previously common species Currently common species Axinellida Axinella polycapella Axinella polycapella Axinellida Axinella pomponiae Axinella pomponiae Axinellida Axinella waltonsmithi Chondrillida Chondrilla caribensis Clionaida Cliona caribbaea Clionaida Cliona celata Cliona celata Clionaida Cliona varians Clionaida Spheciosongia vesparium Spheciospongia vesparium Clionaida Placospongia (cf) intermedia Clionaida Spirastrella coccinea Dendroceratida Igernella notabilis Igernella notabilis 20

Table 1 - continued

Order Previously common species Currently common species Dictyoceratida Dysidea variabilis Dictyoceratida Hippospongia lachne Hippospongia lachne Dictyoceratida Ircinia campana Ircinia campana Dictyoceratida Ircinia felix Ircinia felix Dictyoceratida Ircinia strobilina Haplosclerida Amphimedon viridis Haplosclerida Callyspongia vaginalis Callyspongia vaginalis Haplosclerida Haliclona curacaoensis¹ Haplosclerida Niphates erecta¹ Poecilosclerida Clathria prolifera Clathria prolifera Poecilosclerida Clathria vasiformis Poecilosclerida Lissodendoryx isodictyalis Poecilosclerida Lissodendoryx spinulosa² Lissodendoryx spinulosa² Poecilosclerida Tedania ignis Tedania ignis Poecilosclerida Lissodendoryx sigmata Poecilosclerida Phorbas amaranthus Suberitida Halichondria corrugata Tethyida Tectitethya crypta Tethyida Tethya sp. Tetractinellida Geodia gibberosa Geodia gibberosa Tetractinellida Stelletta grubii³ Tetractinellida Cinachyrella alloclada Verongiida Aiolochroia crassa Verongiida Aplysina (cf.) fulva Aplysina (cf.) fulva

Newly reported and more widespread species: Based off collections for the 36 sites and reported zones from past studies (de Laubenfels 1953, Little 1963, Storr 1976) and the 44 sites from the current collections, two species that are now widespread have not been previously reported (Table 1). Previous confirmed reports of Niphates erecta are located in the southern Gulf of Mexico around Veracruz Mexico and in Cuba (Alvarez et al. 1990, Schmahl 1990, Wulff and Swain 2004, Wulff 2006a, Villamizar et al. 2014). A handful of records in the Northern Gulf report specimens that have later been attributed to Niphates erecta. Haliclona variabilis (Dendy, 1890) collected by Storr (1976) is reported 21 along the Florida coast, however when I analyzed the voucher museum samples it revealed this specimen is not Niphates erecta. Gelliodes ramosa (Carter,1882), a name that has been applied to sponges we now know as Niphates erecta, was reported by Teerling (Teerling 1975) in the Northwestern Gulf of Mexico. Since Teerling (1975) was never published, records of the species and specific documented locations could not be verified. With the caveat that Teerling’s (1975) samples could not be confirmed, our reporting of Niphates erecta seems to be the first confirmed documentation in the Northeastern Gulf of Mexico. The high abundance of this species in many locations in the Northeastern Gulf of Mexico (Figure 4) and its absence in previous collections by Little, de Laubenfels, and Storr suggest this species could be a new addition to the Northeastern Gulf of Mexico in the last 55 years. Haliclona curacaoensis is also commonly found (Figure 5) however due to the cryptic and encrusting nature found in the Northeastern Gulf this species could have been overlooked in previous studies. Species that are now commonly found in multiple sites but were previously only reported in one or two locations include Lissodendoryx sigmata (Figure 6), Spirastrella coccinea, Phorbas amaranthus, Halichondria corrugata, Cinachyrella alloclada (Figure 7), Placospongia (cf) intermedia, Tethya sp, Chondrilla caribensis (Table 1).

Figure 4: Distribution of Niphates erecta along the Florida coast in the past (A) and present (B). Open circles denote survey locations where the species was absent and closed circles denote survey locations where the species was present.

Previously and currently widespread species: Of the commonly found species reported by de Laubenfels, Little, and Storr in the Northeastern Gulf of Mexico 56% (14/25) are still found throughout the area (Table 1). These include Axinella polycapella, Callyspongia vaginalis, Geodia gibberosa (Figure 8), Cliona celata, Clathria prolifera, Igernella notabilis, Ircinia campana (Figure 9), Ircinia felix, Spheciospongia vesparium, Aplysina (cf.) fulva, Axinella pomponiae, Tedania ignis. Lissodendoryx spinulosa and Hippospongia lachne.

Analysis of logistic regressions suggests that the probability of occurrence does not change with latitude in the past studies or the current studies for Axinella polycapella (past data: P= 0.4704,df=35, F=0.5329; current data: P= 0.07228,df=43, F= 3.2305), or for Callyspongia vaginalis (past data: P= 0.2282 ,df=35, F= 1.4522 ; current data: P=0.258 ,df=43, F= 1.3151), and there is no interaction between latitude and time (P=0.05143 ,df=77, F= 3.7941; and P=0.1067 ,df=77 , F= 2.6645 respectively). The probability of occurrence is influenced by latitude in both the past and present studies for Clathria prolifera (past data: P= 0.02909,df=35, F=4.7624; current data: P=0.02044,df=43, F= 5.3738), Cliona celata (past data: P=0.04577,df=35, F= 3.9902; current data: P= 0.0001616 ,df=43, F= 14.232), and Igernella notabilis (past data: P= 0.02062 ,df=35, F= 5.3588; current data: P=0.0009921 ,df=43, F=12.535), and there is no interaction between latitude and time (P=0.3975,df=77, F= 0.716; P=

0.7307,df=77, F= 0.1185; and P= 0.5642 ,df=77, F=0.3324 respectively). Three other species show a change in the influence of latitude over time. The probability of occurrence is influenced by latitude in the present studies, but not in the past, for Geodia gibberosa (past data: P= 0.7917 ,df=35 , F=0.0698 ; current data: P= 0.005617 ,df=43 , F= 7.6693)(Figure 23), Ircinia campana (past data: P= 0.1752 ,df=35 , F= 1.9168 ; current data: P= 0.002119 ,df=43 , F= 10.728)(Figure 24), and Spheciospongia vesparium (past data: P= 0.6188 ,df=35, F= 0.2521; current data: P=0.002986, df=43, F= 8.8159)(Figure 25), all of which show an interaction between latitude and time (P= 0.01511 ,df=77, F=5.9031 ; P= 0.0004362 ,df=77 , F= 12.37; and P=0.01028, df=77, F= 6.9274 respectively). The probability of occurrence is influenced by latitude in the present study, but not in the past for Ircinia felix as well (past data: P= 0.1747 ,df=35, F= 1.9219; current data: P= 0.006367,df=43, F= 7.4433)(Figure 26), however there is not a significant interaction between latitude and time (P= 0.2525, df=77, F=1.3296). The probability of occurrence does not change with latitude in the current study for Aplysina fulva, but it may have in the past (past data: P= 0.01669 ,df= 35 , F= 5.7289; current data: P= 0.2174 ,df=43 , F= 1.5682), There is no interaction between latitude and time (P= 0.08773 ,df=77 , F=2.9921 ) for A. fulva and the effect of latitude in the past is likely due to unconfirmed identification of several past verongid samples by de Laubenfels.

Missing or less common species: Of the species previously reported as common 44% (11/25) are no longer widespread or might be missing entirely (Table 1). Aiolochroia crassa, Cliona caribbaea, Cliona varians (Figure 10), and Lissodendoryx isodictyalis were not found in the present study and have not reported by other recent studies or surveys in the Northeastern Gulf of Mexico (Rutzler et al. 2009) Axinella waltonsmithi, Clathria vasiformis, Dysidea variabilis, Ircinia strobilina, Amphimedon viridis and Tectitethya crypta have not been confirmed by skeletal analysis by the present study but are likely still found based on field identifications. These species were not commonly found however and would not be classified as widespread by this study. Stelletta grubii, denoted as the “oyster sponge” has been attributed as a misapplication of the name and instead is recognized as Stelletta kallitetilla (Laubenfels, 1936). The oyster sponge documented by Little (1963) has not been found in the present study (Figure 11). Gulf Specimen Marine Lab, a local invertebrate supply company, routinely found this species at a variety of sites in the 1990’s and early 2000’s but have not found this species in the last 5-10 years (Debbi Clifford, personal communication). This could suggest that the species Little (1963) reported has only recently disappeared from the region.

Wider-scale faunal comparisons between the Northern Gulf of Mexico, Gray’s Reef (GA, U.S.A), and the Caribbean Sea

Northern range limit of species common in the Caribbean: Species found in the present study were compared to those reported at Gray’s Reef (Freeman et al. 2007, Hopkinson et al 1991) and 90 common Caribbean species from a total of 16 orders. 24 Caribbean species from 10 orders were found in the Northeastern Gulf of Mexico

27 while 34 Caribbean species from 14 orders were reported in Gray’s Reef (Table 2, Freeman et al. 2007, Hopkinson et al 1991). 48 Caribbean species (53% of those evaluated) from 12 orders were not reported in Gray’s Reef nor the present study (Table 2).

Southern range limit of North West Atlantic species: Gray’s Reef has 10 North West Atlantic species from 7 orders, while 4 species from 3 orders are in the Northeastern Gulf of Mexico (Table 3). Cliona celata, and Clathria prolifera were both common in the Northeastern Gulf of Mexico but were not found south of Sarasota, FL in the present study. Axinella pomponiae was found in the Marco Island region in South Florida and has been reported along Florida Atlantic coast as well as the Southeastern US (Alvarez et al. 1998) but has not been reported in the wider Caribbean. Axinella waltonsmithi was not one of the common Northeastern Gulf of Mexico species but was found at one site in the current study. A. waltonsmithi, like A. pomponiae has not been reported in the wider Caribbean (Alvarez et al. 1998).

Range limit differences and similarities between Gray’s Reef and the Northeastern Gulf of Mexico: Both the Northeastern Gulf of Mexico and Gray’s Reef had 16 Caribbean species from 10 orders (Table 2). For the North West Atlantic fauna, 4 species from 3 orders are found in both the Northeastern Gulf of Mexico and Gray’s Reef (Table 3).

For Caribbean species unique to either Gray’s reef or the Northeastern Gulf of Mexico, 18 Caribbean species from 10 orders were only found at Gray’s Reef whereas 7 species from 4 orders were only found in the Northeastern Gulf of Mexico (Table 4). Of the three Caribbean species evaluated in the Scopalinida order, all are missing from the Northeastern Gulf of Mexico but one species was found at Gray’s Reef. For the North West Atlantic fauna 6 species from 5 orders were reported at Gray’s Reef that are not reported in the Northeastern Gulf of Mexico while all 4 of the species found in the Northeastern Gulf of Mexico were also reported at Gray’s Reef (Table 3).

Order Widespread in the Gray's Reef NE Gulf Caribbean Caribbean Caribbean fauna fauna Agelasida Agelas clathrodes (2) Agelasida Agelas conifera (1,2) Agelasida Agelas dispar (1,2,5) Agelasida Agelas sceptrum (2,5) Agelasida Agelas schmidti (2) Agelasida Agelas sventres (2) Agelasida Agelas wiedenmayeri (2) Axinellida Axinella corrugara (1) Axinellida Axinella polycapella (17) Axinella polycapella Axinella polycapella Axinellida Dragmacidon reticulatum Dragmacidon (14, 15,16) reticulatum Axinellida Ectyoplasia ferox (1,2,4,6) Axinellida Ptilocaulis walpersi (2) Ptilocaulis walpersi Ptilocaulis walpersi (uncommon) Biemnida Neofibularia nolitangere Neofibularia (2,5) nolitangere Chondrillida Chondrilla caribensis Chondrilla Chondrilla (2,4,5) caribensis caribensis. Chondrosiida Chondrosia collectrix Chondrosia (14,15) collectrix Clionaida Cliona aprica (4)

Table 2 - continued

Order Widespread in the Gray's Reef NE Gulf Caribbean Caribbean Caribbean fauna fauna Clionaida Cliona caribbaea (2) Cliona caribbaea Clionaida Cliona delitrix (1,2) Clionaida Cliona varians (2,4,5) Cliona varians Clionaida Cliona vesparium (4) Clionaida Placospongia intermedia Placospongia (5) intermedia Clionaida Spheciospongia vesparium Spheciospongia Spheciospongia (8, 10) vesparium vesparium Clionaida Spirastrella coccinea (4) Spirastrella Spirastrella coccinea coccinea Clionaida Spirastrella hartmani (2,5,7) Clionaida Spirastrella mollis (16) Spirastrella mollis Dendroceratida Igernella notabilis * Igernella notabilis Igernella notabilis Dictyoceratida Dysidea etheria (4,5,8) Dysidea etheria (uncommon) Dictyoceratida Hippospongia lachne (13) Hippospongia lachne Dictyoceratida Hyrtios proteus (4) Dictyoceratida Hyrtios violaceus Hyrtios violaceus (13,14,17) Dictyoceratida Ircinia campana (9) Ircinia campana Ircinia campana Dictyoceratida Ircinia felix (1,4,7) Ircinia felix Ircinia felix Dictyoceratida Ircinia strobilina (2,5) Ircinia strobilina Ircinia strobilina (uncommon) Dictyoceratida Smenospongia aurea (4,5)

Table 2 - continued

Order Widespread in the Gray's Reef NE Gulf Caribbean Caribbean Caribbean fauna fauna Dictyoceratida Smenospongia Smenospongia cerebriformis (16) cerebriformis Dictyoceratida Spongia tubulifera (14,15) Spongia tubulifera Haplosclerida Amphimedon compressa Amphimedon (2,5,7) compressa (uncommon) Haplosclerida Callyspongia fallax (1) Callyspongia fallax Haplosclerida Callyspongia plicifera (2,3,5) Haplosclerida Callyspongia vaginalis Callyspongia (1,2,4,6,7) vaginalis Haplosclerida Chalinula molitba (8) Chalinula molitba Haplosclerida Cribrochalina dura (2) Haplosclerida Cribrochalina vasculum (2) Haplosclerida Haliclona curacaoensis Haliclona (11) curacaoensis Haplosclerida Haliclona implexiformis (4) Haplosclerida Haliclona manglaris (4) Haplosclerida Neopetrosia rosariensis (7) Haplosclerida Niphates amporpha (6) Haplosclerida Niphates digitalis (2,3,4,5,6) Haplosclerida Niphates erecta (1,2,3,6,7) Niphates erecta Niphates erecta Haplosclerida Oceanapia bartschi (2)

Table 2 - continued

Order Widespread in the Gray's Reef NE Gulf Caribbean Caribbean Caribbean fauna fauna Haplosclerida Siphonodictyon coralliphagum (2) Haplosclerida Xestospongia muta (2,3,5) Homosclerophorida Plakortis (2) Homosclerophorida Plakortis angulospiculatus (1) Poecilosclerida Clathria coriacea (4) Clathria coriacea Poecilosclerida Clathria curacaoensis Clathria (15,16) curacaoensis Poecilosclerida Clathria curacaoensis (2) Poecilosclerida Desmapsamma anchorata Desmapsamma (13,15) anchorata Poecilosclerida Iotrochota birotulata (1,2,4,5,7) Poecilosclerida Lissodendoryx colombiensis (7) Poecilosclerida Lissodendoryx isodictyalis (4) Poecilosclerida Lissodendoryx sigmata Lissodendoryx Lissodendoryx (10) sigmata sigmata Poecilosclerida Lissodendoryx spinulosa* Lissodendoryx spinulosa Poecilosclerida Monanchora arbuscula (7) Poecilosclerida Mycale laevis (1,4,5,7) Poecilosclerida Mycale laxissima (5,7) Poecilosclerida Mycale microsigmatosa (4) 32

Table 2 - continued

Order Widespread in the Gray's Reef NE Gulf Caribbean Caribbean Caribbean fauna fauna Poecilosclerida Phorbas amaranthus (2) Phorbas amaranthus Phorbas amaranthus Poecilosclerida Tedania ignis (4,8) Tedania ignis Scopalinida Scopalina ruetzleri (1,2,4,5,6) Scopalinida Scopalina ruetzleri Scopalina ruetzleri (14,15,16) Scopalinida Svenzea zeai (2) Suberitida Axinyssa ambrosia (12,13) Axinyssa ambrosia Suberitida Halichondria melanadocia (4) Tethyida Tethya crypta (4,6) Tethya sp. Tethya sp. Tetractinellida Cinachyrella kuekenthali (2) Tetractinellida Cinachyrella alloclada (1) Cinachyrella Cinachyrella alloclada alloclada Tetractinellida Erylus formosus (4,5) Tetractinellida Geodia gibberosa * Geodia gibberosa Geodia gibberosa Tetractinellida Geodia neptuni (2,5) Tetractinellida Stelletta kallitetilla (4) Verongiida Aiolochroia crassa Aiolochroia crassa (1,2,3,5) Verongiida Aplysina archeri (1,2,3) Verongiida Aplysina cauliformis (2,5,6) Verongiida Aplysina fistularis (2,4,5) Aplysina fistularis Verongiida Aplysina fulva (7) Aplysina fulva Aplysina fulva

Table 2 - continued

Order Widespread in the Gray's Reef NE Gulf Caribbean Caribbean Caribbean fauna fauna Verongiida Aplysina lacunosa (1,2) Verongiida Verongula gigantea (2,5) Verongiida Verongula rigida (1,2,5)

Order Gray's Reef North West Atlantic fauna NE Gulf North West Atlantic fauna Axinellida Axinella pomponiae Axinella pomponiae Axinellida Axinella waltonsmithi Axinella waltonsmithi (uncommon) Clionaida Cliona celata complex Cliona celata Haplosclerida Haliclona oculata Poecilosclerida Clathria prolifera Clathria prolifera Poecilosclerida Mycale fibrexilis Suberitida Ciocalypta gibbsi Suberitida Halichondria bowerbanki Tetractinellida Stelletta carolinensis Verongiida Aplysilla longispina

Order Gray's Reef Caribbean fauna NE Gulf Caribbean fauna Axinellida Axinella polycapella Axinella polycapella Axinellida Dragmacidon reticulatum Axinellida Ptilocaulis walpersi Ptilocaulis walpersi (uncommon) Biemnida Neofibularia nolitangere Chondrillida Chondrilla caribensis Chondrilla caribensis

Table 4 - continued

Order Gray's Reef Caribbean fauna NE Gulf Caribbean fauna Chondrosiida Chondrosia collectrix Clionaida Cliona caribbaea Clionaida Cliona varians Clionaida Spheciospongia vesparium Spheciospongia vesparium Clionaida Spirastrella coccinea Spirastrella coccinea Clionaida Spirastrella mollis Clionaida Placospongia intermedia Dendroceratida Igernella notabilis Igernella notabilis Dictyoceratida Hyrtios violaceus Dictyoceratida Ircinia campana Ircinia campana Dictyoceratida Ircinia felix Ircinia felix Dictyoceratida Ircinia strobilina Ircinia strobilina (uncommon) Dictyoceratida Smenospongia cerebriformis Dictyoceratida Spongia tubulifera Dictyoceratida Dysidea etheria (uncommon) Dictyoceratida Hippospongia lachne Haplosclerida Callyspongia fallax Haplosclerida Chalinula molitba Haplosclerida Niphates erecta Niphates erecta Haplosclerida Amphimedon compressa (uncommon) Haplosclerida Callyspongia vaginalis Haplosclerida Haliclona curacaoensis Poecilosclerida Clathria coriacea Poecilosclerida Clathria curacaoensis Poecilosclerida Desmapsamma anchorata Poecilosclerida Lissodendoryx sigmata Lissodendoryx sigmata Poecilosclerida Phorbas amaranthus Phorbas amaranthus Poecilosclerida Lissodendoryx spinulosa Poecilosclerida Tedania ignis Scopalinida Scopalina ruetzleri Suberitida Axinyssa ambrosia Tethyida Tethya sp. Tethya sp. Tetractinellida Cinachyrella alloclada Cinachyrella alloclada Tetractinellida Geodia gibberosa Geodia gibberosa Verongiida Aiolochroia crassa Verongiida Aplysina fistularis Verongiida Aplysina fulva Aplysina fulva

Discussion

Northeastern Gulf of Mexico shifting distributions of common species

Distribution patterns of sponges in the Northeastern Gulf of Mexico have changed since previous studies in the 1950’s and 1960’s. A slight majority of the once common species (56%) are still widespread throughout the Northeastern Gulf of Mexico, as reported by previous studies (de Laubenfels 1953, Little 1963, Storr 1976). Eleven of these species are commonly found throughout the Caribbean and three are found in the North West Atlantic. Upon closer analysis there appears to be an effect of latitude on the likelihood of occurrence for several of these species. Clathira prolifera and Cliona celata are more likely to occur at more poleward latitudes, which is not surprising given that both of these are sub-tropical species found along the North West Atlantic. Igernella notabilis also occurred in higher abundance at more poleward latitudes, which is surprising given this species is found in the Caribbean. All three of these species do not change ther patterns over time, which contradicts my prediction that sub-tropical species may retreat given increased water temperatures. Two species were no more or less likely to occcur at any particular latitude. Callyspongia vaginalis and Axinella pollycapella were found all along the Florida coast in both the past and present study. It appears these species distribution patterns are also unchanged over time. Three species show no effect of latitude influencing the probability of occurrence based on the past studies, but had latitudinal influences in the present study. Geodia gibberosa, Ircinia campana, and Spheciospongia vesparium are more likely to occur at more polward latitudes currently than in the 1960’s. Ircinia campana and Speciospongia vesparium are both large and distinctive species, common throughout the Caribbean and easily identified in the field, so it is unlikely they were overlooked by previous or current studies. Geodia gibberosa is typically cryptic and allows other sponges to overgrow it in the field (personal observation) so it is possible this species could have been missed by previous or current collectors.

Eleven other species (44%) previously reported as widespread are either missing entirely or are no longer widespread throughout the region, though many of them are commonly found in the Caribbean. Two species are newly reported in the Northeastern Gulf of Mexico (Niphates

36 erecta and Haliclona curacaoensis) and eight other species are more widespread than before, all of which are commonly found in the Caribbean.

Some of the species in the newly widespread category may have just been missed in previous sampling efforts. Several of the “newly widespread” species are encrusting or cryptic species and one other is only found in very shallow waters. Halichondria corrugata is commonly found in very shallow environments and is not found in waters deeper than 5m. Previous sampling efforts have focused on offshore reefs and seagrass beds with the exception of 8 sites in Little’s (1963) survey, so it is possible this species seems more abundant now due to a higher number of shallow sites in the present study. Other species may have been missed due to their cryptic or encrusting nature in the Northeastern Gulf of Mexico. Encrusting sponges are often less common in surveys since they are typically thin layers of tissue on the reef that are easily missed compared to the other growth forms. Meanwhile, cryptic species are often only found hidden within the reef structure. Haliclona curacaoesnsis is a common mangrove sponge in the Caribbean (Wulff 2009) however in the Northeastern Gulf of Mexico it is only found as a thinly encrusting sponge on hard substrates. Placospongia (cf) intermedia was also commonly found encrusting on hard substrate but only reported at one site by Little (1963) and reported as rare by Storr (1976). Chondrilla caribensis is often a cryptic species in the Gulf, being found at 7 stations hidden inside artificial reef balls and under limestone ledges. This species was previously only reported at one site but could have easily been missed at others. Three other species, Spirastrella coccinea, Phorbas amaranthus, and Tethya sp. were found at more sites than previously reported, however the difference between the current and previous number of sites was not dramatically higher with 2, 1, and 1 additional sites respectively.

Two species have increased their distribution in the Northeastern Gulf of Mexico. Lissodendoryx sigmata, is currently found at 13 stations but was only reported by Little (1963) at one site. This particular species is unlikely to have been overlooked. L. sigmata is often massive and was one of the most commonly found species at almost every site. Along with distinctive skeletal characteristics this species is unlikely to be mistaken and therefore has likely increased in abundance since previous studies. Cinachyrella alloclada is another species that likely has increased in abundance. This species was currently reported at 8 sites and is also unlikely to be 37 overlooked. C. alloclada was reported by Little (1963) at one site and C. cavernosa was reported by Storr (1976) south of Cedar Key. Based on analysis of a museum voucher specimen Storr’s species name could be inaccurate. A museum specimen from Storr’s survey was found for C. alloclada within the region reported for C. cavernosa which could suggest that and C. cavernosa reported by Storr (1976) was actually C. alloclada, highlighting the need to verify previous reports with voucher specimens.

Comparisons of this nature would not be possible without voucher specimens obtained through museum collections. By analyzing specimens obtained from the University of Miami Invertebrate Museum and the Yale Peabody Museum, species descriptions for many of the historic samples have been updated. Some include a report of Lissodendoryx spinulosa being revealed by analyzing samples of Lissodendoryx isodictyalis collected by Storr showing a previously unknown presence of L. spinulosa in the Northeastern Gulf of Mexico. Niphates erecta’s new documentation was also substantiated by analyzing specimens of Haliclona variabilis, a species name sometimes used for Niphates erecta, by Storr that did not match Niphates erecta.

Many comparisons were hindered or impossible due to a lack of voucher specimens and species descriptions. As sponge systematics is continually updated, it is critical that specimens are kept to allow future comparisons, especially in light of increased anthropogenic stressors that have been attributed to many shifts in species distributions (Cahill et al. 2012). Successful verification of past records is key to understanding how species distribution patterns are changing.

Overall we are seeing a loss of some Caribbean species and an increase in other Caribbean species, while the sub-tropical North West Atlantic species remain relatively unchanged. This contradicts the original prediction that sub-tropical species would disappear and Caribbean species would migrate north or increase their distribution. Instead we see a variety of trends for species commonly found throughout the Caribbean.

Wider-scale faunal comparisons between the Northeastern Gulf of Mexico, Gray’s Reef (GA, U.S.A), and the Caribbean Sea

The northern range limit differs for many species commonly found throughout the Caribbean. It appears that only a handful of Caribbean species occur in the Gulf of Mexico and the North West Atlantic. Of 90 common Caribbean species only about 27% (24 species) were found in the Northeastern Gulf of Mexico and about 38% (34 species) at Gray’s Reef in Georgia U.S.A. Meanwhile 53% (48 species) were not found in either the Northeastern Gulf of Mexico or Gray’s Reef. Of the 8 Western Caribbean species analyzed in the Agelasida order, none were found north of the Caribbean Sea. One other order, Homosclerophorida, was also not found north of the Caribbean Sea however only 2 species were examined so this restriction should be further evaluated. Overall, however, there does not seem to be a general pattern of growth form or order that explains why some Caribbean species were able to persist in the more northern locations while others were excluded.

There is a much smaller representation of North West Atlantic fauna represented in the present study and at Gray’s Reef compared to the Caribbean fauna, 4 and 10 species respectively. North West Atlantic fauna made up roughly 12.5% (3 common species) of the common Northeastern Gulf of Mexico species and 23% (10 species) of the Gray’s Reef species. Of these fewer are found in the Northeastern Gulf of Mexico than at Gray’s Reef, and all 4 species found in the Northern Gulf of Mexico are also found at Gray’s Reef. This could be due to southern temperature limits restricting the range north of the Florida Keys, thus isolating the Northeastern Gulf of Mexico. The two Axinella species reported are also found along the southern Florida coasts but Cliona celata and Clathria prolifera seem to be restricted to the more northern locations. More investigations should be conducted to determine if and for how long these populations have been isolated.

When determining if northern range limits differ between the Florida Gulf and Atlantic coasts, in respect to the Caribbean fauna, the Northeastern Gulf of Mexico and Gray’s Reef seem to overlap in a slight majority of the species but each contain Caribbean species that are absent at

39 the other location. Roughly 47% (42 species) of the 90 examined Caribbean species were reported in either the Northeastern Gulf of Mexico, Gray’s Reef, or both locations. Of these species 38% (16 species) occur at both Gray’s Reef and in the Northeastern Gulf of Mexico. For species located either at Gray’s Reef of the Northeastern Gulf of Mexico roughly 17% (7 species) were only found in the Northeastern Gulf of Mexico while 43% (18 species) were only found at Gray’s Reef.

There does not seem to be a pattern based on order or growth form to explain why some species are able to persist at Gray’s Reef, but not the Northeastern Gulf of Mexico and vica- versa. It is possible that abiotic conditions such as river outflow locations and habitat connectivity could explain some of the discontinuities. Sand-flats and river outflow, along with temperature, have been the hypothesized explanation behind historic species boundaries along the Florida Gulf coast (Storr 1976). Water temperatures might also explain some of the inconsistencies. Cold winter water temperatures was one of the main predictors of northern limits for many of the species along the Florida Gulf coast, however in the last 60 years there has been an increase of 2-3°C in the average winter sea surface temperatures (NOAA OPS data) which could have allowed Caribbean species to move northward along the Gulf coast but not along the Atlantic coast. Placospongia intermedia is among the species that is found only in the Northeastern Gulf of Mexico and is also among the species that seems to have increased their distribution in the Northeastern Gulf of Mexico (referenced above), though there is not a dramatic increase for this species. Haliclona curacaoensis is also among those Caribbean species only found in the Northeastern Gulf of Mexico and is newly documented in this region. Several of the Caribbean species found only in the Northeastern Gulf of Mexico are common mangrove species in the Caribbean. Tedania ignis, Haliclona curacaoensis, and Dysidea etheria are three of the common mangrove species found in both Panama and Belize (Wulff 2009). T. ignis and D. etheria are also commonly found on mangrove roots in the Florida Keys (personal observation). Mangrove species tend to be highly successful at recruitment and growing quickly (Wulff 2004, 2005) which could explain why these species are so successful. Tedania ignis in particular is known to overgrow other species on recruitment surfaces (Sutherland 1980, Wulff 2005) and is one of the most rapid growing mangrove sponges (Wulff 2005). Overall the reason some species are able to successfully move to some locations while other closely related species are hindered 40 is unknown. It is likely that a combination of abiotic and biotic factors influence each species distribution and how flexible a species is in its ecological strategies could illuminate why some species are widespread while others are localized to a specific site or region.

The northern range limit is north of the Caribbean Sea for only a fraction (47%) of the common species evaluated. Species’ northern range limits also differ between Gray’s reef and the Northeastern Gulf of Mexico. It is surprising that some species range limits differ between the Atlantic and Gulf coasts, with more species existing farther north along the Atlantic coast despite our winter temperature increases. This may suggest that other factors, besides temperature, play a major role in determining sponge species ranges.

CHAPTER 3 ASPECTS OF ECOLOGY THAT ALLOW A SESSILE SPONGE WITH A LIMITED DISTRIBUTION TO THRIVE IN TWO DISSIMILAR HABITATS

Introduction

Sponges are commonly found in seagrass meadows, coral reefs, fouling communities and mangrove roots. Each of these habitats has distinct biotic and abiotic factors that species must deal with and many which provide trade-offs. Coral reefs, for example, have a variety of predators that sessile sponges must avoid and lower amounts of picoplankton in the water column (Wulff 2017). Mangroves, on the other hand, have high food availability and sponges are relatively protected from predators however competition for space prevents many species from persisting in the habitat (Wulff 2009, 2017). Due to these conflicting biotic and abiotic factors, sponge species will often specialize to a particular habitat.

Sponges are relatively simple compared to other animals and this allows them to have an impressive variety of life history and ecological strategies to thrive in different habitats. For example, sponges can reproduce via asexual fragmentation or sexual reproduction (e.g. broadcast spawning), but often favor one mode of reproduction depending on their habitat or morphology (Maldonado and Riesgo 2008). Branching reef sponges that specialize in asexual fragmentation lower their gamete production and increase their attachment rate (Wulff 1997, Tsurumi and Reiswig 1997). Sponges also have astonishing biochemical diversity and can produce chemical deterrents specific to the predator in their habitat (Wulff, 2017). Alternatively, sponges lacking chemical defenses against the local predator can associate with other sponges that have chemical defenses (Wulff 2008, Ramsby et al. 2012) or other organisms that allow them to avoid predation (e.g. decorator crabs). Finally, sponges are recognized for their impressive ability to filter bacteria-sized particles out of the water column at high efficiency (Reiswig 1971, 1975), but some species also have the ability to receive supplemental nutrition from cyanobacterial symbionts (Erwin and Thacker 2008). If a species is reliant on cyanobacterial symbionts, light

42 availability may play a major role in sponge metabolism and consequently restrict a sponge to a habitat with sufficient light for photosynthesis.

Since sponges typically specialize to a particular habitat, it is uncommon to find species that are successful in more than one habitat. One sponge species in the Northern Gulf of Mexico is an exception. Halichondria corrugata Diaz, van Soest & Pomponi, 1993 is the dominant sponge in shallow seagrass beds and on dockside fouling communities in the Northeastern Gulf of Mexico. The variation in abiotic conditions between the habitats and H. corrugata’s success makes it an interesting system to test what aspects of ecology allow it to be so successful.

Seagrass beds and dock fouling communities differ in their biotic and abiotic stressors. Seagrass meadows provide continuous habitat but hard substrate (e.g. pen shells, rocks) can be limited. Meanwhile dockside fouling communities are discontinuous habitats. Within a dock site, platforms are often separated and distance between docks and dock sites can widely vary. Physical disturbance occurs in both habitats, but differs in frequency and duration. Seagrass meadows are prone to physical disturbance by storms resulting in suspension of sediment into the water column. Docks, on the other hand, experience extreme low tides resulting in potential grounding and sediment smothering. During extreme low tides, fouling organisms would need to withstand sediment smothering for several hours over multiple days. Not all dock segments are beached at extreme low tides, so there is the possibility of recolonizing the dock from the regional pool of individuals. Predation risk from benthic crawling predators, such as sea stars, is virtually non-existent in the floating docks, but could be a major biotic stressor in seagrass habitats causing species to be restricted from this habitat (Wulff 2005). Contrarily, competition for space is a major factor in dock fowling communities (Jackson 1977) but may be less of a factor in seagrass meadows. Light accessibility and intensity also varies between the habitats. The amount of light would vary dramatically between the sides and undersides of the dock. This variation could be a major hindrance for organisms reliant on photosynthetic symbionts for nutrients. Seagrass meadows, however, would have little light variation.

Seagrass meadows and dockside fouling communities require sponges to utilize alternative strategies to overcome the different influences of 1) habitat continuity, 2) physical 43 disturbance, 3) predation, 4) competition for space, and 5) light availability between the habitats. Continuous habitat could facilitate asexual fragmentation but hinder sexual reproduction as suitable substrate is available for fragments to reattach but potentially not for larvae to recruit. Whereas in the discontinuous dock habitats, fragmentation would be a poor strategy because it is unlikely a sponge would be able to reattach to the underside of a dock quickly enough before it sank to the bottom and was smothered by sediment. On the other hand, sexual reproduction is a good strategy for sponges in dockside fouling communities as larvae would be able to colonize bare space. Physical disturbance by storms would require seagrass sponges to quickly reattach after being dislodged (e.g. the reef sponge Aplysina fulva) or have a large basal attachment to the substrate to resist breakage (e.g. the loggerhead sponge Spheciospongia vesparium) (Wulff 1997, 2006a). Predation risk in the seagrass would require associational or chemical defenses, and similar to mangrove root communities (Wulff 2009), the ability to recruit and retain space would be critical with high competition for space in the dock habitat. In the seagrass meadows, the high light intensity would allow sponges with photosynthetic symbionts to be successful, whereas species in the dock habitat that utilize photosynthetic symbionts would need to be flexible in their feeding strategy to compensate for shaded conditions.

For H. corrugata to be successful in both seagrass meadows and dockside fouling communities it would need to be flexible in the strategies it utilizes. In seagrass meadows fragmentation would be the ideal reproductive strategy for H. corrugata whereas in the fouling community larval recruitment would be ideal. H. corrugata would need to avoid predators in the seagrass meadows via chemical deterrence or associational strategies, but these defenses would not be necessary in the dockside fouling community. If H. corrugata is lacking chemical defenses against Echinaster spinulosus, then this species could be utilizing associational defenses to avoid predation. H. corrugata was always observed attached to decorator crabs in St. Joseph Bay when predators were abundant, and only found on the substrate when predators were absent (personal observation). Finally, the low light conditions in dockside fouling communities may influence competitive outcomes or growth of H. corrugata which depends on its photosynthetic symbionts as a source of nutrition.

The overall goal of this chapter was to describe the different strategies H. corrugata uses to overcome potential selection pressures between the two habitats. To test the reproductive strategies of H. corrugata, a shallow seagrass site in St. Joseph Bay, FL and a dockside fouling community in Shell Point, FL were selected to document the larval recruitment ability and fragmentation ability through clonal diversity. To test the role predation plays, long term census observations were performed in St. Joseph Bay along with laboratory feeding trials and predator avoidance trials with the local seagrass sponge predator Echinaster spinulosus. To evaluate the role of light availability in sponge growth and photosynthetic symbiont abundance, in-situ shading experiments were performed at a dockside fouling community in Shell Point, FL. This three-pronged approach highlights the different ecological aspects that could allow H. corrugata to be successful in two habitats that differ dramatically in biotic and abiotic factors.

Methods

Study area

Populations of H. corrugata at two Thalassia seagrass meadows roughly 3km apart. (1) 29.74803, -85.38989; and 2) 29.77552, -85.4007) in St. Joseph Bay, FL and one dock site, roughly 112km east, in Shell Point, FL (30.06547, -84.28625) were used (Figure 12). Seagrass sites had a high abundance of sponge eating sea stars Echinaster spinulosus with little variation in natural light between grass beds and within a grass bed. No other sponges at either site were found during the course of the study. The dock fouling community site is a floating dock unconnected to the substrate thus protecting sponges from any possible predation by E. spinulosus, while also providing natural variations in light based on the dock surface the sponge settles on.

Figure 12: Experimental locations. The western seagrass sites in St. Joseph Bay, FL and the eastern dock site in Shell Point, FL marked with stars.

Clonal diversity and recruitment

To approximate clone membership, somatic tissue grafts were performed. Tissue fusion experiments have successfully been used to identify clone mates (Wulff 1986), however certain conditions must be met. Successful tissue fusion implies that individuals either share all alleles at a loci governing tissue compatibility, or only a subset of the alleles (Wulff 1986). Whether or not the sharing of alleles could equate to clonal identity depends on knowledge of genetic requirements for fusion, propagule dispersal, and relative dispersal abilities of fragments (Wulff 1986). Due to these limitations, this study will be used as a preliminary indicator of clonal membership and should be followed up by a population genetic study to confirm genetic identity.

Tissue grafts were performed on spatially separate individuals within both seagrass habitats and dockside fouling communities. In St. Joseph Bay, tissue grafts were performed with 10 individuals (5 from each site) within a grass bed and between grass beds roughly 3.2km apart (Figure 13a) resulting in 33 pairwise comparisons. At Shell Point, tissue grafts were performed with 5 individuals from separate dock platforms resulting in 15 pairwise comparisons, and 4 individuals within a platform (Figure 13b) resulting in 10 pairwise comparisons.

Fragments of spatially separate individuals were attached to CPVC pipes with the two individuals’ ectosome in contact. The fragments were monitored over the course of two weeks and checked for tissue fusion. Fragments that readily fuse might be asexual fragments of the same genotype, whereas those that retain separation are likely separate larval recruits. The number and identity of successfully fused individuals within a habitat was recorded and compared with other tissue grafts to determine which experimental individuals are potential clone mates. Fragments from all experimental fusions were stored in 90% ethanol for future molecular analysis.

Recruitment pipes were deployed in seagrass habitats and dockside fouling communities in June 2017 to determine the potential for larval recruitment for a given habitat. Sanded CPVC pipes with beaded zip tie attachments were used to provide amenable recruitment surfaces. For dockside fouling communities pipes were hung from the dock and regularly checked. Half the 47 pipes were regularly cleared of fouling barnacles to ensure clear recruitment surfaces. For the seagrass communities pipes were mounted to cinderblocks and deployed in St. Joseph Bay Florida. Recruitment pipes were monitored for three months and potential recruits were collected and identified by spicule preparation.

Susceptibility to predation

Trials were performed to assess how often H. corrugata was consumed by E. spinulosus. Each individual sea star was given live and intact pieces of sponges. A total of 22 species were tested but only specific results for H. corrugata are reported here. Three to five species were used in each trial. Sea star feeding was monitored over five days. A sponge was considered “eaten” if a sea star remained on that sponge for more than 45 minutes and exhibited feeding behavior by everting its stomach, or if the sponge had tissue loss from sea star feeding. After each feeding trial, the skeletons of the sponges in the trial were examined for further evidence of feeding such as tissue discoloration at the feeding site or the appearance of skeleton. The data from the feeding observations and the post-trial feeding evidence were combined to form a binary data set which revealed the feeding preferences of E. spinulosus. A non-parametric G-test was used to determine if there was a statistical feeding preference with a 50/50 null expectation.

Predator avoidance strategies

To test whether this association with decorator crabs reduces predation caged pairings with approach time and successful feeding time were conducted. H. corrguata was attached to either a pen shell or to a decorator crab to simulate what E. spinulosus would naturally encounter. Treatments consisted of one decorator crab and sponge with one sea star or one sponge attached to a pen shell with one sea star. Four trials of each treatment were performed. For both treatments the time it took E. spinulosus to approach and contact the sponge was recorded, as well as the time until successful feeding. Successful feeding was determined by both E. spinulosus behavior and sponge tissue loss or discoloration as described above. A t-test was used

48 to determine if there was a statistical difference in the approach time between treatments, as well as the time until successful feeding between treatments.

Seasonal abundance

Seasonal abundance of decorator crabs, sponges, and sea stars was measured in a 100m² census site in St. Joseph Bay, FL over the course of 15 months (May 2016-July 2017). Light and temperature recorders (HOBO Pendant Temperature/Light Data Loggers) were deployed to measure hourly light and temperature values over the course of the census. Average percent cover of seagrass was determined monthly from February to July by evaluating cover of each square meter for six rows resulting in percent cover estimates of 60m² within the 100m² census site.

The effect of shading on specific growth and symbiont abundance

Halichondria corrugata harbors photosynthetic cyanobacteria that may aid in growth through supplemental nutrition as seen in other species (Erwin and Thacker 2008). To determine the effect of the photosymbiont abundance on sponge growth I performed in-situ shading experiments with 20 individuals at Shell Point and lab flow-through experiments at the Florida State University Coastal and Marine Laboratory in St. Teresa FL with 20 individuals collected from Shell Point. Lab experimental individuals were transported from Shell Point then left to attach and recover for one month in flow-through tanks before trials began.

In order to determine if cyanobacterial symbionts were affected by shading, I compared the change in chlorophyll a between shaded and unshaded treatments in flow-through tanks for 20 individuals. Shaded individuals experienced an average light intensity of 12.96 lum/ft² (SE 4.52 lum/ft²), while unshaded sponges experienced an average of 447.4 lum/ft² (SE 31.04 lum/ft²). Replicate genotypes were attached to CPVC pipes and divided into two treatment tanks. The shaded treatment tank was covered with a tarp. Light and Temperature loggers were

49 deployed in both treatment tanks. Chlorophyll a was measured using methods outlined in Erwin and Thacker (2008). 0.25g of ectosome for each sponge piece was extracted in 10ml of 90% acetone. Samples were then wrapped in foil and held at 4°C overnight to prevent photodegradation and allow adequate extraction. Then, 1.5mL of extract was centrifuged and 1mL of the supernatant was used for absorbance measures. Absorbance values were measured at 750, 664, 647, and 630 nm. Parsons et al. (1984) equations were then used to estimate chlorophyll a concentrations. After two months overall differences between control and shaded treatments were assessed using a t-test. An ANOVA was used to determine differences between initial and final measurements between treatments. When significant, differences among groups were compared post-hoc using Tukey's Honest Significant Difference (HSD).

To document the effect of shading on sponge growth, in-situ shading experiments were performed with the assumption the effect is due to changes in photosymbiont abundance. Replicate genotypes of 20 individuals were attached to CVPC pipes and hung off the dockside. The sponges were monitored for 2 weeks to ensure attachment and health then a shading tarp was set up that floats over one set of replicate genotypes to create a shaded treatment, without impeding water flow or food availability between treatments. Shaded individuals experienced an average light intensity of 4.12 lum/ft² (SE 0.50 lum/ft²), while unshaded sponges experienced an average of 58.4 lum/ft² (SE 8.75 lum/ft²). Sponges were monitored for two months to record changes in specific growth between treatments. Light and temperature recorders (HOBO Pendant Temperature/Light Data Loggers) were attached in both shaded and unshaded treatments to ensure differential light treatments without differences in temperature between treatments. Images of the pipes were taken biweekly and sponge volume was measured by breaking the sponge into geometric solids and measuring the shape dimensions using ImageJ software. Sponge specific growth was calculated and compared between treatments for replicate genotypes. A t-test was used to determine statistical differences in specific growth rate between treatments.

Results

Clonal diversity and recruitment

From the 33 pairwise comparisons at the seagrass sites (Figure 13a), successful fusion between all individuals within and between sites occurred. Every individual fused with every other individual at both the northern Eagle Harbor site and the southern Hammock trail site. No larvae recruits were found on deployed recruitment pipes. For the seagrass habitat there was one clonal group consisting of individuals 3km apart and no evidence of larval recruitment or individuals brooding larvae.

At the dock site no fusion between any individuals within or between slips (Figure 13b) was observed so there were no fusion groups displayed at this site. Successful self-fusion for all individuals was observed. Larval recruits were found on recruitment pipes in August two months after deployment. Individuals brooding embryos were also found in May and April. From analysis of one individual’s 35.450mm2 tissue section, larvae density was 3.7 larvae/mm2 each larvae being roughly 0.011mm2, resulting in roughly 4% or the tissue area represented as larvae.

Susceptibility to predation

H. corrugata was consumed in 70% of trials (31/44). There was a preference for H. corrugata (G=7.4989, df=1, p<0.01), suggesting this species is chemically undefended against E. spinulosus. Consumption of the other 21 species ranged from 86% to 0% of positive consumption in trials. In total five species, including H. corrugata, were preferred, seven species showed no preference or avoidance, and ten species were avoided or rejected.

Predator avoidance strategies

Decorator crabs hindered E. spinulosus from successfully feeding on sponges (P<0.001, Figure 14b). The average time for E. spinulosus to approach a sponge did not statistically differ whether the sponge was attached to a pen shell or a decorator crab (P=0.0297), Figure 14a).

Seasonal abundance

Prior to the start of the census, predators were abundant and isolated H. corrugata individuals were not seen, however predators disappeared from the survey site after back to back red tide events. The census surveys were started after the die-off to record how the natural abundance may change with the absence and return of predators.

Seasonal abundance of H. corrugata tracked decorator crab abundance and monthly temperature changes in the absence of predators (Figure 15). Individuals persisting on the substrate seemed to disappear when predators returned (Figure 15), though they were still found

52 in association with decorator crabs (personal observation). During the census all sponges persisting on the substrate had decorator crab molts embedded in them, suggesting that the sponges likely arrived via decorator crabs. Organism abundance fluctuated with seasonal change in grass cover and temperature (Figure 15).

The effect of shading on specific growth and symbiont abundance

In fouling communities, symbiont abundance dramatically decreased with shading. Unshaded sponges experienced some loss in Chlorophyll a, however shaded sponges experienced significantly more loss (P=0.013, Figure 16). Loss of symbionts in the unshaded treatment was likely due to accidental partial shading by algae. Sponges and tanks were cleaned twice a week, however algae growth still resulted in partial shading. Tukey Post-Hoc analysis showed no significant difference between shaded and unshaded initial chlorophyll a amounts, and unshaded final and initial chlorophyll a amounts (P=0.824, and P=0.196 respectively, Figure 17). H. corrugata specific growth was strongly influenced by light availability. Shaded individuals grew less than unshaded individuals over one month (P= 0.04, Figure 18).

Figure 17: Difference in chlorophyll a amounts between shaded and unshaded initial and final treatments. Letters denote significant differences (P<0.005). Error bars represent standard error.

Figure 18: Specific growth was significantly higher for unshaded sponges (n=17) than for shaded sponges (n=17) (Paired t-test, P=0.04). Error bars represent standard error.

Discussion

H. corrugata has an impressive flexibility1 in life history and ecological strategies to thrive in two habitats that differ markedly in abiotic and biotic factors. In seagrass meadows, H. corrugata propagates asexually and avoids predation by sea stars by associating with decorator crabs while in dockside fouling communities, H. corrugata larvae recruit to open substrate on the dock and tolerate shading despite their association with photosynthetic symbionts. This

1 The term flexibility is used since I was unable to determine whether the differences in strategies were due to local adaptation or to plasticity. To distinguish between these, molecular analysis of the two populations and reciprocal transplants of genetic replicates would need to be performed. 55 flexibility has allowed this species to become the most abundant sponge in both seagrass meadows and dockside fouling communities in the Northeastern Gulf of Mexico.

In the St. Joseph Bay seagrass meadows, fusion experiments revealed one clonal group consisting of individuals that are kilometers apart and no evidence of larval recruitment, suggesting that asexual fragmentation is the primary form of reproduction at this site. This was further supported through observations that each individual sponge in the seagrass contained a decorator crab molt, suggesting the decorator crabs facilitate asexual propagation. When a decorator crab molts, it collects a piece of sponge from its old carapace and attaches it to its new carapace (e.g Wicksten 1975; personal observation), thus creating two fragments of the same genotype. At the dockside fouling community at Shell Point, there was no fusion between individuals. Evidence of larval recruitment and observations of H. corrugata brooding larvae suggest that each individual resulted from larval recruitment. This flexibility in reproductive strategies gives H. corrugata the advantage in the discontinuous dock habitats, where larval recruitment is the only viable strategy, and continuous seagrass habitats, where fragmentation is a viable strategy.

H. corrugata lacks inherent defenses against E. spinulosus predation, however the sponge is able to coexist with E. spinulosus because it associates with decorator crabs. Decorator crabs fend off and flee from E. spinulosus carrying H. corrugata away from the would-be sponge predator. Associational defenses are not uncommon among sponges lacking chemical defenses. Lissodendoryx colombiensis and Geodia vosmaeri lack chemical defenses against predators within their habitats and avoid predation by allowing sponge species that produce chemical defenses against the local predator to overgrow them (Wulff 2008, Ramsby et al. 2012). When predators are absent, H. corrugata is able to persist independently on the substrate and fluctuates in abundance with seasonal variations in temperature.

Cyanobacterial symbionts enhance growth in H. corrugata. Though the chlorophyll a amount for this species is lower compared to other species reported by Erwin and Thacker (2007), the loss of symbiont abundance or function with shading does result in a decreased specific growth rate. This suggests that, despite the low chlorophyll a levels, the symbionts are a 56 key source of nutrition for H. corrugata. In the dockside fouling community, cyanobacterial symbiont abundance and light availability at a particular settlement location could play a major role in determining an individual’s success. If an individual settles in a shaded location, it would experience significantly less growth compared to an individual in direct light and runs the risk of being overgrown; however, individuals that settled in shaded conditions persisted throughout the experiment despite their lower growth rates.

Though this kind of life history and ecological flexibility is uncommon in sponges, there is a similar example in one species. The reef sponge Lissodendoryx colombiensis is able to live in seagrass meadows by associating with other sponges that protect it from predation by the sea star Oreaster reticulatus (Wulff 2008). The key to L. colombiensis’ success is its rapid growth which allows the sponge to recover tissue lost to predation and prevents complete overgrowth by seagrass sponges (Wulff 2008).

Similar to L. colombiensis, H. corrugata also has a fast growth and attachment rate which is enhanced by its cyanobacterial symbionts. H. corrugata’s fast growth allows decorator crabs to successfully harvest and attach this species to their carapace. Fast growth and the aid of symbionts may also prevent complete smothering in dockside fouling communities. Individuals in treatment pipes had to compete with algae in unshaded conditions however many individuals quickly covered the pipe surface, securing their spot on the substrate and preventing algae growth. While this species is commonly colonized by bryozoans and ascidians as epibionts, its survival seems uninfluenced by them.

Though Halichondria corrugata is successful in both habitats there are costs associated with each habitat. By living on decorator crabs in seagrass meadows, H. corrugata is able to avoid predation, but its population is limited by decorator crab abundance. In the dockside fouling community, H. corrugata can avoid competition with algae by growing on the undersides of docks in lower light conditions, but this results in lower specific growth rates for the sponge. While H. corrugata is protected from predation on the dock, larval recruitment is the only possible source of reproduction. Each habitat has ecological costs and benefits for H.

57 corrugata, but the flexibility of this species in different aspects of its ecology allow this sessile organism with a limited distribution to be highly successful in dissimilar habitats.

CHAPTER 4

CONCLUSION

The Gulf of Mexico has a unique blend of sub-tropical and tropical species. The seasonal temperatures of the Northern Gulf of Mexico were predicted historically to restrict northward range boundaries for many Caribbean sponge species (Storr 1976). Likewise, many sub-tropical species were predicted to be limited in their southern range boundary due to warm water temperatures (Storr 1976). Range boundaries for many species may have shifted due to increasing winter temperatures since average winter temperatures in the Northeastern Gulf of Mexico have increased 2-3°C since the range limits were originally documented (Biasutti et al. 2012, NOAA OSPO). This potential decay of temperature induced range limits could allow tropical species to expand northward, while also forcing sub-tropical species northward. The movement northward of sub-tropical species is limited, however, due to land barriers. The shrinkage of habitable sub-tropical waters could result in local extinction of species that cannot withstand warmer tropical water temperatures. Results from this study show the Northeastern Gulf of Mexico sponge species composition has changed since previous studies in the 1950’s and 1960’s (de Laubenfels 1953, Little 1963, Storr 1976). Only a little over half of the previously common species are still widespread throughout the Northeastern Gulf of Mexico and eleven species previously reported as widespread are either missing entirely or are no longer widespread throughout the region. Two species are newly reported in the Northeastern Gulf of Mexico (Niphates erecta and Haliclona curacaoensis) and eight other species are more widespread than before. When looking at a wider geographic scale, species range limits differ between the Northwestern Gulf and Atlantic coastline. Temperature alone does not seem to predict species range limits or we would see consistent patterns of occurrence for species commonly found in the Caribbean. Instead it appears some species are disappearing and others are increasing their abundance.

Ecological aspects may explain habitat and geographic distribution and abundance of a species. From my case study of Halichondria corrugata, flexibility in its reproductive strategy and its ability to associate with decorator crabs to avoid predation, allows this species to have a 59 wide habitat distribution but does not seem to extend the species overall range. That this species seems to be endemic to the Northeastern Gulf of Mexico, despite its multiple strategies and flexible ecological traits, could suggest that temperature may play a role in limiting this species range. If warming conditions persist, this species could be in danger despite its ability to thrive in different abiotic and biotic conditions. Other ecological factors that have been demonstrated to influence species distributions, and could influence H. corrugata’s distribution, are habitat availability, temperature, water current patterns, and picoplankton availability (Maldonado 2006, Van Soest et al. 2012, Wulff 2012, 2017). Habitat availability could play a role since this species seems to require very shallow habitats. Temperature seems to affect the seasonal abundance of this species but more work would need to be done to test thermal tolerances. Water current patterns could hinder larva movement to additional locations (Maldonado 2006). Finally picoplankton availability has been observed to affect overall size of sponges (Wulff 2017).

When we look in the literature of distribution changes related to climate change it always seems there is a single factor, usually temperature, that explains why species are shifting their ranges (Yamano et al. 2011, Guo et al. 2013, Feary et al. 2014, Gorman et al. 2016). These studies, however, are typically limited to one species or a smaller group of species such the Acropora corals (Yamano et al. 2011). Results from the second chapter did not indicate any single explanatory factor. When looking at ecological aspects it appears that all the species that have shifted their range are commonly found in the Caribbean Sea; however they differ in their growth forms, taxonomic orders, and habitats. Since sponges often specialize to a particular habitat it is likely that the habitat differences could also reflect different recruitment strategies, competitive abilities, and growth and recruitment rates (Wulff 1997, 2009, 2017, Maldonado 2006). Overall it seems there is not one definitive strategy that can easily explain why sponge species ranges shift. The lack of common patterns among species may not be surprising given that “sponges” represent at least 20 different orders that have been distinct for hundreds of millions of years (Erpenbeck et al. 2007). It should not be surprising then that species differ, sometimes dramatically, in ecological requirements, vulnerabilities, and strategies reflecting long separate evolutionary histories. One example is seen in how fresh water influx affects different sponge species. Some species like many bath sponges seem to benefit from increased nutrients 60 associated with small amounts of fresh water (Storr 1964), whereas other species on mangrove roots are restricted from root sections near the surface due to their intolerance to salinity shifts caused by rain (Wulff 2009).

It is clear from my study that species distributions are influenced by ecological factors beyond temperature and there is not one clear factor that explains species range limits and distribution patterns. Species, such as Halichondria corrugata, can have multiple strategies that allow them to live in two dissimilar habitats but still experience a limited geographic distribution. Meanwhile other species can have a wide geographic distribution perhaps by utilizing only one strategy. Therefore my results of each species reacting and changing independently is not surprising given the idiosyncratic nature of different sponge species.

APPENDIX A SUPPLEMENTARY INFORMATION

Table 5: Collection site information from past and current collections.

Collector site Site site depth collection Time /study number code location habitat lat long (m) method Apalachee limestone and - Past Little 1 1 Bay sand 29.92667 84.44167 3.5 mixed Apalachee intertidal sand - Past Little 2 2 Bay shells 29.90500 84.43333 1 mixed Apalachee - Past Little 3 3 Bay oyster bar 29.90833 84.38333 1 mixed Apalachee - Past Little 4 4 Bay rock and sand 29.78500 84.32500 14 mixed Apalachee - Past Little 5 5 Bay sand 29.82889 84.27167 12 mixed Apalachee - Past Little 6 6 Bay sand 29.85000 84.19000 8 mixed Apalachee - Past Little 7 7 Bay rock and sand 29.82500 84.12528 9.5 mixed Apalachee - Past Little 8 8 Bay oyster bar 30.08333 84.19167 2.5 mixed Apalachee sand and - Past Little 9 9 Bay seagrass 30.07500 84.18333 1 mixed Apalachee limestone and - Past Little 10 10 Bay seagrass 30.05000 84.08333 2.5 mixed Apalachee sand and - Past Little 11 11 Bay seagrass 29.95833 83.91667 2.5 mixed Apalachee - Past Little 12 12 Bay beach 29.89889 84.33333 0.5 mixed Apalachee - Past Little 13 13 Bay muddy 29.92167 84.23667 6 mixed Apalachee - Past Little 14 14 Bay sand/mud 29.77917 84.70333 6 mixed de - Past Laubenfels 1 15 Clearwater N/A 27.81667 82.88333 5.5 Dive de - Past Laubenfels 2 16 Clearwater N/A 27.80000 82.96667 9 Dive de - Past Laubenfels 3 17 Bradenton N/A 27.41667 82.75000 10 Dive de - Past Laubenfels 4 18 Fort myers N/A 26.66667 82.45000 14.5 Dive de - Past Laubenfels 5 19 Fort myers N/A 26.38333 82.23333 11.5 Dive de south of - Past Laubenfels 6 20 Naples N/A 25.96667 81.91667 12 Dive de Marco - Past Laubenfels 7 21 island N/A 25.66667 81.91667 14 Dive de - Past Laubenfels 8 22 Keys N/A 24.71667 81.96667 14 Dive de - Past Laubenfels 9 23 Keys N/A 24.68333 81.96667 13 Dive de - Past Laubenfels 10 24 Keys N/A 24.70000 82.11667 12.5 Dive

Table 5 - continued

Collector site Site site depth collection Time /study number code location habitat lat long (m) method de - Past Laubenfels 11 25 Keys N/A 24.65000 82.21667 11 Dive de - Past Laubenfels 12 26 Keys N/A 24.60000 82.45000 6 Dive de - Past Laubenfels 13 27 Keys N/A 24.56667 82.61667 19 Dive de Dry - Past Laubenfels 14 28 Tortugas N/A 24.58333 82.91667 12 Dive de Dry - Past Laubenfels 15 29 Tortugas N/A 24.63333 83.03333 17 Dive de Dry - Past Laubenfels 16 30 Tortugas N/A 24.71667 82.98333 20 Dive de Dry - Past Laubenfels 17 31 Tortugas N/A 24.68333 82.91667 7 Dive de Apalachee - Past Laubenfels 18 32 Bay N/A 29.65000 83.93333 14 Dive de Panama - Past Laubenfels 19 33 city N/A 30.26667 86.06667 18 Dive de - Past Laubenfels 20 34 Dog Island N/A 29.83333 84.53333 12.5 Dive de - Past Laubenfels 21 35 St. Marks N/A 29.98333 84.08333 6.5 Dive de Apalachee - Past Laubenfels 22 36 Bay N/A 29.65000 83.93333 14.5 Dive limestone and - Northern concrete 30.12124 6 to 85.73440 Current Kaiser 1 37 Gulf boulders 12 Dive Northern - 29.74803 Current Kaiser 2 38 Gulf seagrass 85.38989 1 Snorkel Northern - 29.70595 Current Kaiser 3 39 Gulf artificial reef 84.62369 11 Dive Northern - 29.76533 Current Kaiser 4 40 Gulf artificial reef 84.52808 10.5 Dive Northern - 29.76786 Current Kaiser 5 41 Gulf artificial reef 84.52463 10.5 Dive Northern - 29.83333 Current Kaiser 6 42 Gulf sand/mud 84.53333 6.5 Dive Northern - 29.79368 Current Kaiser 7 43 Gulf limestone reef 84.48942 11 Dive Northern - 29.66610 Current Kaiser 8 44 Gulf artificial reef 84.36919 13 Dive Northern - beach 29.89888 Current Kaiser 9 45 Gulf beach shore 84.33333 0 collection Northern mixed hard - 29.98333 Current Kaiser 10 46 Gulf ground 84.08333 6.7 Dive Northern - 29.95833 Current Kaiser 11 47 Gulf seagrass 83.91666 3 Dive Northern seagrass with - 30.05000 Current Kaiser 12 48 Gulf some limestone 84.08333 3 Snorkel mixed hard - 29.10643 Current Kaiser 13 49 Cedar Key ground 83.29200 7.5 Dive mixed hard - 29.04091 Current Kaiser 14 50 Cedar Key ground 83.19391 7.5 Dive

Table 5 - continued

Collector site Site site depth collection Time /study number code location habitat lat long (m) method mixed hard - 29.03705 Current Kaiser 15 51 Cedar Key ground 83.17910 9.5 Dive mixed hard - Current Kaiser 16 52 Cedar Key ground 29.04025 83.14929 6.5 Dive Tarpon - 28.49978 Current Kaiser 17 53 Springs artificial reef 82.97392 9 Dive Tarpon mixed hard - 28.22576 Current Kaiser 18 54 Springs ground 82.87244 9 Dive Tarpon mixed hard - 28.20019 Current Kaiser 19 55 Springs ground 82.81837 5 Dive Tarpon mixed hard - 28.18824 Current Kaiser 20 56 Springs ground 82.80721 3.5 Snorkel mixed hard - Current Paulay/Wulff 2 57 Sarasota ground 27.24830 82.73190 11 N/A - Current Paulay/Wulff 3 58 Fort myers sargassum 26.41020 82.40340 15.2 N/A 17.07 - - Current Paulay/Wulff 4 59 Fort myers sand bottom 26.41020 82.40340 18.29 dive - Current Paulay/Wulff 5 60 Fort myers N/A 26.43260 82.30200 12.2 dive - Current Paulay/Wulff 6 61 Fort myers N/A 26.28617 82.34767 N/A trawl - Current Paulay/Wulff 7 62 Fort myers N/A 26.23383 82.37600 N/A trawl sand bottom with large - Current Paulay/Wulff 8 63 Fort myers octocorals 26.23983 82.33800 16.2 dive - Current Paulay/Wulff 9 64 Fort myers N/A 26.04000 82.30833 N/A trawl - "mud Current Paulay/Wulff 10 65 Keys mud bottom 24.69033 82.11783 N/A dive" - sargassum Current Paulay/Wulff 11 66 Keys sargassum 24.65700 82.00000 N/A dipnet - Current Paulay/Wulff 12 67 Keys eel grass beds 24.67950 81.76113 7 dive - Current Paulay/Wulff 13 68 Keys hard bottom 24.73750 81.79100 11 dive - dipnet/ Current Paulay/Wulff 14 69 Keys N/A 24.73750 81.79100 N/A nighlight Marco - Current Paulay/Wulff 15 70 Island N/A 25.76167 82.01900 N/A trawl Marco - Current Paulay/Wulff 16 71 Island N/A 25.80933 82.08800 N/A trawl Marco - Current Paulay/Wulff 17 72 Island N/A 25.76167 82.01900 18.3 trawl - Current Paulay/Wulff 18 73 Fort myers N/A 25.99350 82.33183 21.3 trawl - Current Paulay/Wulff 19 74 Fort myers N/A 26.28150 82.71867 30.5 dredge - 40.84 Current Paulay/Wulff 20 75 Fort myers N/A 26.31968 83.01483 -41.45 dredge - Current Paulay/Wulff 21 76 Fort myers N/A 26.40297 83.28358 51.8 dredge 64

Table 5 – continued

Collector site Site site depth collection Time /study number code location habitat lat long (m) method - Current Paulay/Wulff 22 77 Fort myers N/A 26.47940 83.49535 58 dredge - 55.47- Current Paulay/Wulff 23 78 Sarasota N/A 27.07127 83.62000 59.13 dredge - 57.30- Current Paulay/Wulff 24 79 Sarasota N/A 27.08760 83.70958 57.91 dredge - bottom Current Paulay/Wulff 25 80 Sarasota N/A 27.09113 83.70485 N/A grab

Figure 19: Modern sea surface temperature map of the Gulf of Mexico and Florida Atlantic coast. Image and data retrieved from NOAA OSPO (http://www.ospo.noaa.gov/data/sst/contour/gulfmex.cf.gif)

REFERENCES

Alcolado, P. 1990. General features of Cuban sponge communities. Pages 351–357 Rützler K (ed.). New perspectives in sponge biology. Smithsonian Institution Press, Washington DC.

Alvarez, B., M. C. Diaz, and R. A. Laughlin. 1990. The sponge fauna on a fringing coral reef in Venezuela, I. Composition, distribution, and abundance. Pages 358–366 New Perspectives in Sponge Biology (ed. K. Rützler). Smithsonian Insti- tution Press, Washington DC.

Alvarez, B., R. van Soest, and K. Rutzler. 1998. A Revision of Axinellidae of the Central West Atlantic Region SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY • NUMBER 598. Smithsonian Contributions to Zoology.

Bellard, C., C. Bertelsmeier, P. Leadley, W. Thuiller, and F. Coutchamp. 2012. Impacts of climte change on the future of biodiversity. Ecology Letters 15:365–377.

Biasutti, M., A. H. Sobel, S. J. Camargo, and T. T. Creyts. 2012. Projected changes in the physical climate of the Gulf Coast and Caribbean. Climatic Change 112:819–845.

Boury-Esnault, N., M. Klautau, C. Bézac, J. Wulff, and A. M. Solé-Cava. 1999. Comparative study of putative conspecific sponge populations from both sides of the Isthmus of Panama. J. Mar. Biol. Ass. U.K. 79:39–50.

Butler, M., J. Hunt, W. Herrnkind, M. Childress, and R. Bertelsen. 1995. Cascading disturbances in Florida Bay, USA: cyanobacteria blooms, sponge mortality, and implications for juvenile spiny lobsters Panulirus argus. Marine Ecology Progress Series 129:119–125.

Cahill, A. E., M. E. Aiello-lammens, M. C. Fisher-Reid, X. Hua, C. J. Karanewsky, H. Y. Ryu, G. C. Sbeglia, F. Spagnolo, J. B. Waldron, O. Warsi, J. John, H. Yeong Ryu, G. C. Sbeglia, F. Spagnolo, J. B. Waldron, O. Warsi, and J. J. Wiens. 2012. How does climate change cause extinction? Proceedings of the Royal Society B: Biological Sciences 280:20121890– 20121890.

Core Team, R. 2015. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Asutria.

Diaz, M. C. 2005. Common sponges from shallow marine habitats from Bocas del Toro region, Panama. Caribbean Journal of Science 41:465–475.

Diaz, M. C., S. A. Pomponi, and R. W. van Soest. 1993. A systematic revision of the central West Atlantic Halichondrida (Demospongiae, Porifera). Part III: Description of valid species. Scientia Marina 57:283–306.

Erpenbeck, D., S. Duran, K. Rützler, V. Paul, J. N. a. Hooper, and G. Wörheide. 2007. Towards a DNA taxonomy of Caribbean demosponges: a gene tree reconstructed from partial 71

mitochondrial CO1 gene sequences supports previous rDNA phylogenies and provides a new perspective on the systematics of Demospongiae. Journal of the Marine Biological Association of the UK 87:1563–1570.

Erwin, P. M., and R. W. Thacker. 2007. Incidence and identity of photosynthetic symbionts in Caribbean coral reef sponge assemblages. Journal of the Marine Biological Association of the UK 87:1683–1692.

Erwin, P., and R. Thacker. 2008. Phototrophic nutrition and symbiont diversity of two Caribbean sponge–cyanobacteria symbioses. Marine Ecology Progress Series 362:139–147.

Feary, D. A., M. S. Pratchett, M. J Emslie, A. M. Fowler, W. F. Figueira, O. J. Luiz, Y. Nakamura, and D. J. Booth. 2014. Latitudinal shifts in coral reef fishes: Why some species do and others do not shift. Fish and Fisheries 15:593–615.

Fodrie, F. J., K. L. Heck, S. P. Powers, W. M. Graham, and K. L. Robinson. 2010. Climate‐ related, decadal‐scale assemblage changes of seagrass‐associated fishes in the northern Gulf of Mexico. Global Change Biology 16:48–59.

Freeman, C. J., D. F. Gleason, R. Ruzicka, R. W. M. Van Soest, and A. W. Harvey. 2007. A biogeographic comparison of sponge fauna from Gray ’ s Reef National Marine Sanctuary and other hard-bottom reefs of coastal Georgia , U . S . A . Ocean Science 28:319–325.

Gorman, D., A. Turra, E. R. Bergstrom, and P. A. Horta. 2016. Population expansion of a tropical seagrass (Halophila decipiens) in the southwest Atlantic (Brazil). Aquatic Botany 132:30–36.

Green, G. 1977. Ecology of toxicity in marine sponges. Marine Biology 40:207–215.

Guo, H., Y. Zhang, Z. Lan, and S. C. Pennings. 2013. Biotic interactions mediate the expansion of black mangrove (Avicennia germinans) into salt marshes under climate change. Global change biology 19:2765–74.

Hajdu, E., G. Muricy, M. Custodio, C. Russo, and S. Peixinho. 1992. Geodia Corticostylifera (Demospongia, Porifera) New Astrophorid From the Brazilian Coast (Southwestern Atlantic). Bulletin of Marine Science 51:204–217.

Hartman, W. D. 1955. A collection of sponges from the west coast of the Yucatan Peninsula with descriptions of two new species. Bulletin of Marine Science of the Gulf and Caribbean 5:161–189.

Hopkinson, C. S., D. Robert, B. Jansson, and J. P. Schubauer. 1983. l at Gray ’ s Reef , a hard bottom habitat in the Georgia Bight *.

Jackson, J. B. C. 1977. Competition on Marine Hard Substrata: The Adaptive significance of solitary and colonial strategies. The American Naturalist 111:743–767. 72

Jones, S. J., F. P. Lima, and D. S. Wethey. 2010. Rising environmental temperatures and biogeography: Poleward range contraction of the blue mussel, Mytilus edulis L., in the western Atlantic. Journal of Biogeography 37:2243–2259. de Laubenfels, M. W. 1936. A Discussion of the Sponge Fauna of the Dry Tortugas in Particular and the West Indies in General, with Material for a Revision of the Families and Orders of the Porifera. Carnegie Institute of Washington Publication (Tortugas:1–287. de Laubenfels, M. W. 1953. Sponges from the Gulf of Mexico. Bulletin of Marine Science 2:511–557.

Little, F. J. J. 1963. The sponge fauna of the St. George’s Sound, Apalache Bay, and Panama City Regions of the Florida Gulf Coast. Tulane Studies in Zoology 11:31–71.

Maldonado, M. 2006. The ecology of the sponge larva. Canadian Journal of Zoology 84:175– 194.

Maldonado, M., and A. Riesgo. 2008. Reproduction in the phylum Porifera: a synoptic overview. Treballs de la SCB 59:29–49.

Parmesan, C., N. Ryrholm, C. Stefanescu, J. K. Hill, C. D. Thomas, H. Descimon, B. Huntley, L. Kaila, J. Kullberg, T. Tammaru, W. J. Tennent, J. A. Thomas, and M. Warren. 1999. Poleward shift in georgaphical ranges of butterfly species associated with regional warming. Nature 399:579–583.

Parsons, T., Y. Maita, and C. Lalli. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon Press, New York.

Pearse, A. 1950. Notes on the Inhabitants of Certain Sponges at Bimini. Ecology 31:149–151.

Peterson, B. J., C. M. Chester, F. J. Jochem, and J. W. Fourqurean. 2006. Potential role of sponge communities in controlling phytoplankton blooms in Florida Bay 328:93–103.

Poloczanska, E. S., M. T. Burrows, C. J. Brown, J. García Molinos, B. S. Halpern, O. Hoegh- Guldberg, C. V. Kappel, P. J. Moore, A. J. Richardson, D. S. Schoeman, and W. J. Sydeman. 2016. Responses of Marine Organisms to Climate Change across Oceans. Frontiers in Marine Science 3:1–21.

Ramsby, B., A. Massaro, E. Marshall, T. Wilcox, and M. Hill. 2012. Epibiont-basibiont interactions: Examination of ecological factors that influence specialization in a two-sponge association between Geodia vosmaeri (Sollas, 1886) and Amphimedon erina (de Laubenfels, 1936). Hydrobiologia 687:331–340.

Reiswig, H. M. 1971. Particle Feeding in Natural Populations of Three Marine Demosponges Author ( s ): Henry M . Reiswig Published by : Marine Biological Laboratory Stable URL : http://www.jstor.org/stable/1540270. Biological Bulletin 141:568–591. 73

Reiswig, H. M. 1975. Bacteria as food for temperate-water marine sponges. Canadian Journal of Zoology 53:582–589.

Ribeiro, S. M., E. P. Omena, and G. Muricy. 2003. Macrofauna associated to Mycale microsigmatosa (Porifera, Demospongiae) in Rio de Janeiro State, SE Brazil. Estuarine, Coastal and Shelf Science 57:951–959.

Rützler, K. 1978. Sponges in coral reefs. Coral reefs: research methods. Monographs on oceanographic methodology 5:299–313.

Rützler, K., C. Piantoni, and M. . Díaz. 2007. Lissodendoryx: rediscovered type and new tropical western Atlantic species (Porifera: Demospongiae: Poecilosclerida: Coelosphaeridae). Journal of the Marine Biological Association of the UK 87:1491–1510.

Rützler, K., C. Piantoni, R. W. M. Van Soest, and M. . Díaz. 2014. Diversity of sponges (Porifera) from cryptic habitats on the Belize barrier reef near Carrie Bow Cay. Zootaxa 3805:1–129.

Rutzler, K., R. van Soest, C. Piantoni, D. L. Felder, and D. K. Camp. 2009. Gulf of Mexico origin, waters, and biota - Volume 1: Biodiversity. Harte Research Institute for Gulf of Mexico Studies series. 1:v.

Schmahl, G. P. 1990. Community structure and ecology of sponges associated with four southern Florida coral reefs. Pages 376–382 K. Rützler, editor. New Perspectives in Sponge Biology. Smithsonian Institute Press, London.

Van Soest, R. W. M., N. Boury-Esnault, J. Vacelet, M. Dohrmann, D. Erpenbeck, N. J. De Voogd, N. Santodomingo, B. Vanhoorne, M. Kelly, and J. N. a Hooper. 2012. Global diversity of sponges (Porifera). PloS one 7:e35105. van Soest, R. W. 1980. Marine sponges from Curaçao and other Caribbean localities. Part II. Haplosclerida. Hummelinck, P.W. & Van der Steen, L.J. (Eds), Uitgaven van de Natuurwetenschappelijke Studiekring voor Suriname en de Nederlandse Antillen. No. 104. Studies on the Fauna of Curaçao and other Caribbean Islands 62:1–173. van Soest, R. W. 1984. Marine sponges from Curaçao and other Caribbean localities. Part III. Poecilosclerida. Hummelinck, P.W. & Van der Steen, L.J. (Eds), Uitgaven van de Natuurwetenschappelijke Studiekring voor Suriname en de Nederlandse Antillen. No. 112. Studies on the Fauna of Curaçao and other Caribbean Islands 62:1–173.

Storr, J. F. 1964. Ecology of the Gulf of Mexico Commercial Sponges and its Relation to the Fishery. Special Scientific Report Fisheries:i–iv, 1-73.

Storr, J. F. 1976. ECOLOGICAL FACTORS CONTROLLING SPONGE DISTRIBUTION IN THE GULF OF MEXICO AND THE RESULTING ZONATION. Aspects of Sponge Biology:261–276. 74

Sutherland, J. P. 1980. Dynamics of the epibenthic community on roots of the mangrove Rhizophora mangle, at Bahia de Buche, Venezuela. Marine Biology 58:75–84.

Teerling, J. 1975. A survey of sponges from the northwestern Gulf of Mexico. University of Southwestern Louisiana. Lafayette, Louisiana.

Tsurumi, M., and H. M. Reiswig. 1997. Sexual versus asexual reproduction in an oviparous rope-form sponge, Aplysina cauliformis (Porifera; Verongida). Invertebrate Reproduction & Development 32:1–9.

Vergés, A., P. D. Steinberg, M. E. Hay, A. G. B. Poore, A. H. Campbell, E. Ballesteros, K. L. Heck, D. J. Booth, M. A. Coleman, D. A. Feary, W. Figueira, T. Langlois, E. M. Marzinelli, T. Mizerek, P. J. Mumby, Y. Nakamura, M. Roughan, E. van Sebille, A. Sen Gupta, D. A. Smale, F. Tomas, T. Wernberg, and S. K. Wilson. 2014. The tropicalization of temperate marine ecosystems : climate-mediated changes in herbivory and community phase shifts The tropicalization of temperate marine ecosystems : climate-mediated changes in herbivory and community phase shifts. Proceedings of The Royal Socienty B 281:1–10.

Vicente, V. P., K. Rützler, and N. M. Carballeira. 1991. Comparative morphology, ecology, and fatty acid composition of West Indian Spheciospongia (Demospongea). Marine Ecology 12:211–226.

Villamizar, E., M. C. Díaz, K. Rützler, and R. De Nóbrega. 2014. Biodiversity, ecological structure, and change in the sponge community of different geomorphological zones of the barrier fore reef at Carrie Bow Cay, Belize. Marine Ecology 35:425–435. de Weerdt, W. H. 2000. A Monograph of the Shallow-water Chalinidae (Porifera, Haplosclerida) of the Caribbean. Beaufortia 50:1–67.

Wicksten, M. K. 1975. Observations on decorating behavior following molting in Loxorhynchus crispatus stimpson (Decapoda, Majidae). Crustaceana 29:315–316.

Wiedenmayer, F. 1977. Shallow-water sponges of the western Bahamas. Experientia Supplementum 28:1–287.

Wulff, J. 2004. Sponges on mangrove roots, Twin Cays, Belize: Early stages of community assembly. Atoll Research Bulletin:1–10.

Wulff, J. 2006a. Resistance vs recovery: morphological strategies of coral reef sponges. Functional Ecology 20:699–708.

Wulff, J. 2012. Ecological interactions and the distribution, abundance, and diversity of sponges. Page Advances in marine biology.

Wulff, J. 2013. Recovery of sponges after extreme mortality events: morphological and taxonomic patterns in regeneration versus recruitment. Integrative and comparative biology 75

53:512–23.

Wulff, J., and L. Buss. 1979. Do sponges help hold coral reefs together? Nature 281:474–475.

Wulff, J. L. 1986. Variation in clone structure of fragmenting coral reef sponges. Biological Journal of the Linnean Society 27:311–330.

Wulff, J. L. 1997. Mutualisms among species of coral reef sponges.

Wulff, J. L. 2005. Trade-offs in resistance to competitors and predators, and their effects on the diversity of tropical marine sponges. Journal of Animal Ecology 74:313–321.

Wulff, J. L. 2006b. Sponge Systematics by Starfish : Predators Distinguish Cryptic Sympatric Species of Caribbean Fire Sponges , Tedania ignis and Tedania klausi n . sp . ( Demospongiae , Poecilosclerida ):83–94.

Wulff, J. L. 2006c. Rapid diversity and abundance decline in a Caribbean coral reef sponge community. Biological Conservation 127:167–176.

Wulff, J. L. 2008. Collaboration among sponge species increases sponge diversity and abundance in a seagrass meadow. Marine Ecology 29:193–204.

Wulff, J. L. 2009. Sponge Community Dynamics on Caribbean Mangrove Roots: Significance of Species Idiosyncrasies. Proceedings of the Smithsonian Marine Science Symposium 38:501–514.

Wulff, J. L. 2017. Bottom-up and top-down controls on coral reef sponges : disentangling within- • habitat processes 98:1130–1139.

Wulff, J., and T. Swain. 2004. Sponges of Navassa Sponges of Navassa:1–7.

Yamano, H., K. Sugihara, and K. Nomura. 2011. Rapid poleward range expansion of tropical reef corals in response to rising sea surface temperatures. Geophysical Research Letters 38.

Zea, S. 1987. Esponjas del Caribe colombiano. Editorial Catálogo Científico, Bogotá.

NOAA OSPO: Current Sea Surface Temperature data retrieved from NOAA OSPO (HTTP://WWW.OSPO.NOAA.GOV/PRODUCTS/OCEAN/SST/CONTOUR/) Historic data retrieved from: Storr (1976) : U.S. Department of Commerce 1955 Surface Water Temperatures at tide stations, Atlantic coast, North and South America. U.S. Dept. Commerce, Coast and Geog. Survey, Spec. Publ. No. 278, 5th ed.

BIOGRAPHICAL SKETCH

Kathleen Kaiser was born on May 6, 1992 in a small Kansas town. She graduated with a B.S. in Biological Science from Florida State University in 2014. During her undergraduate studies, she joined the research team of Janie Wulff through the Women in Math Science and Engineering program. There she assisted in many projects learning the intricacies of sponge species identification and field work. For her undergraduate thesis she documented patterns of sponge species distributions in Apalachee Bay and looked at predatory preferences of sea stars. After graduation, she traveled then worked for Scott Burgess at Florida State University as a field and lab research assistant. In 2015 she joined the Wulff Lab as a graduate student at Florida State University to expand her project started as an undergraduate and increase her quantitative and analytical skills to become an expert in sponge species distribution patterns and ecology.