xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001875641
007 cr mnu|||uuuuu
008 070806s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001652
Hilber, Susan Elizabeth.
Spatial and temporal patterns of feeding and food in three species of Mellitid sand dollars
h [electronic resource] /
by Susan Elizabeth Hilber.
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: Sand dollars are abundant and conspicuous macroinvertebrates in sandy habitats in the Gulf of Mexico. They ingest sediment and associated organisms. Given their high abundances, size, and deposit feeding, they likely are the most important consumer coupling primary productivity with the rest of the food chain in these habitats. Moreover, sand dollars bioturbate sediments and affect species diversity and community structure.Three species of sand dollars in the family Mellitidae, Mellita tenuis, Encope michelini and Encope aberrans, were studied off the West Coast of Florida. Food and feeding of these sand dollar species were studied to understand their relationship to spatial and temporal patterns.Particle size and organic content gut contents, collected from inshore and offshore sites, were analyzed and compared with sediment collected concomitantly. Offshore sediment is coarser and has a higher organic content than inshore sediment. The gut particle sizes are simila r for E. michelini and M. tenuis. The gut contents of E. aberrans have larger particles. All three species have gut particles smaller than the ambient sediment. The gut contents and sediment have higher organic content in the fall than the spring. Additionally, the gut organic content varies dielly, with peaks shifting by date. The gut organic content of the three species was higher than the sediment. The gut organic content of E. aberrans was significantly less than that of co-occurring M. tenuis.The particle sizes and organic content of the gut contents of Mellita tenuis and Encope michelini are similar. However food type ingested may differ. Encope aberrans may coexist with E. michelini and M. tenuis because it occupies a different niche. Inconsistent peak feeding times could reflect differences in food availability and predation pressure. Inverse relationships between particle size and organic content exist for sediment and gut contents. High densities of M. tenuis insho re may deplete the organic content of the sediment and reverse the negative relationship between particle size and organic content.
Thesis (M.A.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 55 pages.
Adviser: John M. Lawrence, Ph.D.
t USF Electronic Theses and Dissertations.
Spatial and Temporal Patterns of Feeding and Food in Three Species of Mellitid Sand Dollars by Susan Elizabeth Hilber A thesis submitted in partial fulfillment Of the requirements for the degree of Master of Science Department of Biology College of Arts and Sciences University of South Florida Major Professor: John M. Lawrence, Ph.D. Susan S. Bell, Ph.D. Florence I. Thomas, Ph.D. Date of Approval: May 26, 2006 Keywords: encope aberrans, encope michelini, mellita tenuis, organi c content, sediment Copyright 2006, Susan Elizabeth Hilber
Acknowledgments I would sincerely like to thank my advi sor, Dr. John Lawrence, for his support, guidance, and willingness to give me a light shove in the right direction. Life in the Lawrence Lab was highlighted by tanks fu ll of echinoderms, ongoing experiments, visiting professors, and access to his persona l echinoderm library that includes numerous articles predating online catalogs that greatly en hanced my research. I am grateful to my committee members; Dr. Susan Bell and Dr. Florence Thomas, who asked me the hard questions so that my research would be better. My time at USF was supported by a Teaching Assistantship in the Biology Depart ment, which afforded me many more hours to work on my research that otherwise would have been impossible. Also, the opportunity to teach other students biology gave me a greater appreciation for the study of life, and how to help others grasp the know ledge and concepts of biology. Without the help of Captain Tom Worle and the crew of R.V. Bellows and a league of research divers, including Bill Dent, Dr. Jim Garey, Dr Joan Herrera, Sofia Hussein, Ben Meister, Jennifer Rhora, and J. P. Swigart, the n ecessary dives and specimen collections would never have been made. So, thank you for your time and commitment. Two undergraduate assistants, Mela Star and Br ett Kaminisky, helped with the unenviable tasks of dissecting sand dollars and removi ng the gut contents thank you for your dedication and commitment. St atistical help was kindly prov ided by Dr. Susan Bell, Dr. Bruce Cowell, and Dr. Earl McCoy.
My family and friends gave inestimable emotional support. My parents, Jean Bronaugh and Rick Hilber, have always encour aged my interest in marine biology even though it has led me far from home. I truly appreciate all the different kinds of support my parents have given me in order that I ma y pursue my dreams. Without them, I would not have done so much or traveled so fa r. Thank you. Also, my mother and sister, Kendra Hilber, provided much needed pairs of fresh eyes to edit and improve the manuscript: thank you. Also, thank you to Joe Piacenza, who is always willing to listen and be supportive. Also, for participating wi th me in diverting a nd recreational activities that are both fun and refreshing after a long day in the lab. Finally, I came into an existing community of biology graduate students that welc omes in the newbies and leads the way for socialization and friendships Thank you to all the friends I made at USF: for the impassioned conversations, both bi ological and not, and the many field trips. I would answer you: Ocean will sa y it the arc of its lifetime is vast as the sea-sand, flawless and numberless. Between cluster and cluster, the bloo d and the vintage, time brightens the flint in the petal, the beam in the jellyfish; the branches are threshed in the skein of coral from the infinite pearl of the horn. -Pablo Neruda from The Engimas, Canto General (1950)
i Table of Contents List of Tables ii List of Figures iii Abstract v Introduction 1 Materials and Methods 11 Collection sites 11 Preservation of Samples 13 Analysis of Gut Contents and Substratum 14 Food Groove Analysis 16 Statistical Analysis 16 Results 18 Description of Populations 18 Particle Analysis 20 Sediment 20 Gut Contents 22 Comparison of Sediment, F ood Grooves and Gut Particles 25 Organic Content Analysis 28 Sediment 28 Gut Contents 30 Discussion 37 Description of Populations 37 Particle Analysis 38 Organic Content Analysis 40 Conclusions 48 References 50
ii List of Tables Table 1 Collection sites for sand dollars 13 Table 2 Mean particle sizes for sediment, food grooves and guts 22 Table 3 Pair wise comparisons of percen t organic content of sediment from Mellita tenuis habitats 29 Table 4 Pair wise comparisons of percent organic content of Mellita tenuis guts 32 Table 5 Pair wise comparisons of percent organic content of Encope michelini guts 34
iii List of Figures Figure 1 Sand dollar collection sites 12 Figure 2 Diagram of internal anatomy of sand dollar 15 Figure 3 Size-frequency distribution of Encope michelini collections 19 Figure 4 Size-frequency distribution of Mellita tenuis collections 19 Figure 5 Comparison of sediment from E. michelini and M. tenuis habitats 21 Figure 6 Photographs of offshore and in shore sediment from a light microscope (40X) 21 Figure 7 Box and whiskers plot of mean particle sizes at sand dollar sites 23 Figure 8 Size-frequency distributi on of particles in the guts of Encope aberrans, Encope michelini, and Mellita tenuis 23 Figure 9 Percent cumulative frequency distributions for sand dollar gut particles 25 Figure 10. Photographs of E. michelini gut contents from a light microscope (100x) 26 Figure 11 Photographs of M. tenuis gut contents from a light mi croscope (100X) 26 Figure 12 Box and whiskers plot of mean part icle sizes for sand dollar collections 27 Figure 13 Comparison of particle size distri bution in food grooves and gut contents 27 Figure 14 Box and whiskers plot of percent or ganic content of sediment from sand dollar habitats 29 Figure 15 Scatter plot of mean particle size and percent organic content at sand dollar sites 31 Figure 16 Scatter plot of mean percent organic content of M. tenuis guts across seasons at Anna Maria Island 32 Figure 17 Scatter plot of mean percent organic content of E. michelini guts across seasons at offshore Charlotte Harbor 34
iv Figure 18 Box and whiskers plot of mean percent orga nic content from two co-existing species off Egmont Key 35 Figure 19 Scatter plot of mean gut particle size and mean per cent organic content for six different collections 35
v Spatial and Temporal Patterns of Feeding and Food in Three Species of Mellitid Sand Dollars Susan Elizabeth Hilber Abstract Sand dollars are abundant and conspicuous macroinvertebrates in sandy habitats in the Gulf of Mexico. They ingest sediment and associated organisms. Given their high abundances, size, and deposit feeding, they li kely are the most important consumer coupling primary productivity with the rest of the food chain in these habitats. Moreover, sand dollars bioturbate sediment s and affect species diversity and community structure. Three species of sand dollars in the family Mellitidae, Mellita tenuis, Encope michelini and Encope aberrans were studied off the West Coast of Florida. Food and feeding of these sand dollar species were studied to understand th eir relationship to spatial and temporal patterns. Particle size and organic content gut cont ents, collected from inshore and offshore sites, were analyzed and compared with sediment collected concomitantly. Offshore sediment is coarser and has a higher organic content than inshore sediment The gut particle sizes are similar for E. michelini and M. tenuis. The gut contents of E. aberrans have larger particles. All three species have gut particles smaller than the ambient sediment. The gut contents and sediment have higher organic content in the fall than the spring. Additionally, the gut organic content va ries dielly, with peaks shifting by date. The gut organic content of th e three species was higher th an the sediment. The gut organic content of E. aberrans was significantly less than that of co-occurring M. tenuis.
vi The particle sizes and organic co ntent of the gut contents of Mellita tenuis and Encope michelini are similar. However food ty pe ingested may differ. Encope aberrans may coexist with E. michelini and M. tenuis because it occupies a different niche. Inconsistent peak feeding times could re flect differences in food availability and predation pressure. Inverse re lationships between particle size and organic content exist for sediment and gut contents. High densities of M. tenuis inshore may deplete the organic content of the sediment and revers e the negative relations hip between particle size and organic content.
1 Introduction In the shallow, warm waters of the Gulf of Mexico, sand dollars in the family Mellitidae can be found in sandy bottoms fr om a depth of 0.5-90 m (Hendler 1995). Sand dollars are deposit-feeding herbivores (a nd incidentally detritivores) that consume benthic microalgae (Lane 1982) and are cons umed by piscine predators, such as triggerfish and elasmobranches (Frazer et al 1991; Kurz 1995). Th erefore, sand dollars couple primary productivity to the higher trophic levels. Gi ven their high densities in sandy bottoms, they are arguably the most im portant consumer of primary producers and microfauna in their habitat. Sand dollars ex hibit behavioral pattern s seasonally (Lane and Lawrence 1982), on a diel scale (Salsman a nd Tolbert), and in regard to substrate preference (Serafy 1979). Know ledge of the spatial and te mporal aspects of food and feeding would clarify their role in the ecosystem. For this reason, Mellita tenuis Encope michelini and Encope aberrans were investigated to bett er understand their feeding ecology. Four species regularly occur in the Gu lf of Mexico, but differ in habitat preferences. Mellita tenuis is commonly found inshore along the eastern Gulf of Mexico in fine sediments (Hendler et al. 1995). Leodia sexiesperforata occurs occasionally in the Gulf, in low abundances (Hendl er et al. 1995), and for this reason was not included in this study. The offshore species, Encope aberrans and Encope michelini, occur on sand
2 plains with crushed shell and qua rtz particles. In some areas, Mellita tenuis densities can reach 74 individualscm -2 (Salsman and Tolbert 1965). Densities of Encope aberrans and E. michelini are generally much less than Mellita tenuis 2.00.53 1.03.00 Encope michelini m -2 and from 2.00.41 0.17 0.46 E. aberrans m -2 off Egmont Key (Lawrence et al. 2005). In order to appreciate the ecological role of sand do llar feeding, an understanding of the physical mechanics of the feeding system is essential. According to Smith (1984), clypeasteroids (irregular, bilaterally symme trical echinoids) evolved from regular echinoids (sea urchins) in the Paleocene to occupy a new habitat, sand bottoms. One group, the sea biscuits, has a re latively thick body. The othe rs, the sand dollars, have a thin, disc-like body. The evolution of tube feet on both the oral and aboral sides allowed sand dollars to feed on particles, a new f ood source for echinoids (Smith 1984). Instead of macroscopic food (algae and animals), sand dollars feed on sand particles and microscopic organisms. Sand dollars are covered with short spin es, which aid in movement and feeding (Telford et al. 1985). Food gr ooves on the oral side work in conjunction with the spines and tube feet to carry food to the mouth. Inte rnally, the lantern, the chewing apparatus of echinoids, has five broadly flattened pyramids and teeth, which grind the food (Hyman 1955). The flattened test of sand dollars ha s greatly influenced the structure and functioning of the inte rnal anatomy. Although the basic echi noid form remains, the gut is modified (Hyman 1955). The esophagus exits the lantern into the capacious stomach, which is nearly concentric to the test margin. The next region, the intestine is narrow and doubles back to meet the anus.
3 Sand dollars have evolved an interesting way for ingestion of food. Mucus traps particles and the spines work in conjunction wi th the tube feet to transport food to the mouth. There are two opposing models for sand dollar feeding: the sieve hypothesis, originally proposed by MacGinitie and MacGinitie (1949) and later supported by Goodbody (1960) and others, and a challenging view proposed by Telford et al. (1985). According to the sieve hypothesis, diatoms, other microalgae, detritus, and sand are filtered through the spines of the sand dollar according to size. Only small grain sizes, generally less than 100 m, fit in the interstices between the spines. In turn, the podia grasp onto the small particles and transport th em via ciliary current s. Once in the food grooves, food is moved to the mouth. Buried sand dollars utilize sediment on both the aboral and oral sides of their bodies. At one point, it was be lieved that the chief function of lunules was to present a short cut for food be ing transported from the aboral side to the mouth (Telford 1981). However, that appe ars to be only a minor aspect of lunule function. In fact, lunules generally serve to deflect lift from wave action and currents (Telford 1981). There are some problems w ith this feeding model. Mainly, sand dollar guts contain particle s greater than 100 m, and the velocity and direction of ciliary currents oppose transport vectors (Telford 1985). Telford (1985) did not believe observations supported this theory. Telford et al. (1985) observed feeding of Mellita tenuis and came to some new conclusions. First, they found sand dollar feed ing is intermittent instead of continuous. Second, instead of ciliary current s creating a smooth flow to the mouth, most currents are weak, in opposing or perpendicula r directions, or centrifugal. According to Telford et al. (1985), feeding begins with barrel-tipped tube feet grasping particles from the sediment.
4 These particles are coated with mucus secreted from the tips of the tube feet. Eventually mucus cords form in the food grooves. The co rds are then moved into the mouth. They found no evidence of particle rejection. In addition, the authors found sand dollars selected diatoms in greater proportion to their occurren ce in the benthos; between 90100% of particles in the food grooves were di atoms. Within the gut, 97% of particles were less than 100 m, which was significantly different from the ambient sediment. Telford et al. (1985) argued th at the lantern is responsible for breaking up larger particles into the fine particles observe d in the gut and that the cilia ry currents primarily serve to ventilate and cleanse the sand dollar surface. This new model does explain the role of the lantern, which the sieve hypothesi s ignored. It seems that the Telford et al. (1985) model does take into account some details that were ignored or missed in the formulation of the sieve hypothesis. Telfords theory for the feeding mechanism followed direct observation and thus seems convincing. Reports differ on food types ingested by sand dollars. Culver (1961) reported only diatoms and other microalgae in the gut of Mellita quinquiesperforata ( =isoforma ). She did not report inorganic matter in th e gut, but Bell and Frey (1969) reported terrigenous detritus in the gut of M. quinquiesperforata. Other reports indicate Mellita tenuis ingests diatoms, foraminifera, dinoflage llates, organic detr itus and sand grains smaller than 500 m (Lane 1977; Lane and Lawrence 1982; Ghiold 1984). The sizes of sand particles in gut contents of Mellita quinquiesperforata are smaller than that of the substratum (Borzone et al. 1997). As Findlay and White (1983) found that M. tenuis also ingests bacteria and non-photos ynthetic microeukaryotes; the di et of sand dollars is not
5 strictly herbivorous. Mellita tenuis ingests diatoms, but Lane (1977) found diatoms can pass through the gut undigested. Evidence exists to support th e idea that several species of sand dollars can coexist in the same habitat due to resource partitio ning of sediment particle sizes (Phelan 1972 and Telford et al. 1987 ). Sister species Encope aberrans and Encope michelini are found together in the Gulf of Mexico; however E. aberrans is larger than E. michelini (Hendler et al. 1995) and have distinctly different food grooves (Mooi 1989). In addition, mixed populations of E. michelini with C lypeaster subdepressus and Leodia sexiesperforata as well as E. aberrans co-occurring with Mellita tenuis, can be found in the Gulf of Mexico (pers. observation; Telford et al. 1987). Based on analysis of particles in the food grooves, Telford et al. (1987) concluded sand dollars feed on different fractions of the sediment as a result of difference in size of food collecting tube feet Although of interest, size of particles in the food grooves and gut may not indicate feedi ng preference. Telford et al. (1985) suggested the accumulati on of fine particles less than 50 m in the gut of Mellita quinquiesperforata resulted from crushing diatoms and fracturing sand grains by the teeth. Their analysis showed no selection of fine particles in the food grooves as they are virtually identical to t hose in the sediment. In cont rast, Telford and Mooi (1986) found Encope michelini has particles greater than 200 m but not less than 100 m in their food grooves. Sand dollars have specific requirements for substratum. Sediments either too fine (<60 m) or too coarse (>1mm) do not support populations of the sand dollar Mellita tenuis (Pomory et al. 1995) Ghiold (1979) found Mellita quinquiesperforata (= isometra ) burrowed most efficiently in 3 sand grains, confirming the observation of
6 Bell and Frey (1969) that individuals could not move on coarse gravel. Generally, fine sediments can suffocate the sand dollars by clogging the intermeshes between their spines and podia whereas coarse sediments are an impediment to movement and feeding because the tube feet cannot grasp the grai ns (Pomory et al. 1995). The quality of sediments not only is important in determini ng distribution of sand dollars, it may play a role in determining if a sand dollar is cap able of movement, and even righting (Kenk 1944; Weihe and Gray 1968), important to an individuals ability to survive and avoid predation when prone. Populations of E. aberrans and E. michelini have been found to be almost always distinct off the coast of North Carolina (Phelan 1972). These observations suggest the three sp ecies have preferred habitats that are re lated to depth and substratum type. Sand dollars exhibit diel patterns in activity. Salsman and Tolbert (1965) documented nocturnal behavior of Mellita tenuis The sand dollars began moving through the substratum around early evening and eventually reached a maximum speed approximately one hour after activity began. Ac tivity declined in the hours just before dawn. Diel rhythms can result from rhythm s of physical conditions, such as light, predators or food. As sand dollars do not ha ve vision, it is unlikel y light affects diel activity other than as a cue. Mellita tenuis may intensify feeding in the hours just before or after sunset associated with the nocturnal rhythm noted by Salsman and Tolbert (1965). Lane and Lawrence (1982) reported significantly more food in the gut in the night than in the day. Predation could be one possible reason for a diel activity pattern. Crozier (1919) suggested fish were predators of the sand dollar Leodia sexiesperforata and MacGinitie
7 and MacGinitie (1968) suggest ed the spiny lobster, Panulirus interruptus, prey on Dendraster excentricus. Merrill and Hobson (1970) saw the crabs Loxorhynchus grandis and Cancer sp. feeding on the test ambitus of D. excentricus The gray triggerfish, Balistes capriscus, feeds on the test ambitus of Mellita tenuis (Frazer et al. 1991, Kurz 1995). As fish are primarily diurnal and crabs nocturnal, fish predation may be the cause of nocturnal behavior in sand dollars. Seasonally, feeding behavior may change due to food availability, light or temperature. Cool winter temperatures and light may alter plankt on and benthic biomass (Findlay and Watling 1998, Mitbarkar and An il 2002), and reproduction, i.e. allocating resources to gonad growth versus somatic growth (Lane and Lawrence 1979). The caloric content from lipid of the silt-clay fr action of the substratum at Mullet Key with a population of Mellita tenuis is highest in the summer (Lane and Lawrence 1982). Lipid is a major component of diatoms (Dawes 1998). Lane and Lawrence (1982) reported feeding rate by M. tenuis is higher in summer than winter. The microbial community of sand habitats exhibits patterns in behavior, also. Pennate diatoms migrate between the sediment and water surface level to optimize light conditions and avoid predation by surface pr edators (Mitbarkar and Anil 2002). Centric diatoms rely on tidal currents and wave acti on for resuspension. Furthermore, pennate and centric diatoms enter resting stages and lie dormant in the substrate while retaining their photosynthetic capacity un til either resuspended or more optimal environmental conditions return (Mitbarkar and Anil 2002). Though phytoplankton biomass may be small, they can contribute 30-70% of the total primary productivity at the sediment-water interface in shallow waters (Findlay and Watling 1998). In addition, diatoms are a
8 valuable source of proteins and lipids for grazers (Romer and MacLachlan 1986). Therefore, migration and resuspension coul d have cascading effects on the benthic food web. The copepods Acartia grani and Centropages typicus emerge from the substrate at dusk and migrate to the surface to feed (C albert 1999). Ostracods exhibit a similar behavior, but some ostracods only migrate be tween 0 and 50 cm from the substrate to feed, providing a food source for mid-colu mn suspension feeders (Macquart-Moulin 1999). Findlay and Watling (1998) found seasonal variations in the microbial conditions in shallow water communities depending on tu rnover and nutrient avai lability. Possibly, the rhythms of consumers mirror the rhythms of their prey. Consequently the diel rhythms observed in sand dollars might be explained by the diel rhythms of their food. Sand dollars have a profound effect on th eir environment (Bell and Frey 1969). Mellita tenuis may be packed on a sand plain to the point of overlapping individuals (Weihe and Gray 1968, Hyman 1955, pers. obser vation). The layer in which sand dollars can live and respire is shallow, yet still can be incredibly productive and made even more so by their presence, due to their bioturba tion. Findlay and White (1983) found that the passage of a sand dollar effectively increased the depth of redox discontinuity from 0.2 cm to 0.8 cm, however, the chlorophyll a concentration in the sediment was unaffected by M. tenuis feeding. Bioturbation, such as that from sand dollars, enhances the oxygen content of the benthos, loosens sediments, and stimulates sediment transport, all important factors in determining species dive rsity and community structure (Dernie et al. 2003). Lane and Lawrence (1982) suggested in months when M. tenuis had negative absorption efficiencies, i.e. excreting more car bohydrates and proteins than taken in, they effectively enrich the sediment. This would especially be true when sand dollars travel
9 from high to low nutrient substrates. Furt hermore, during summer 2005, a severe red tide on the west coast of Florida caused mass mort alities of fish and invertebrates offshore and caused an anoxic zone of approximately 200 km 2 (FWRI 2005). In September 2005, a site 8.5 km west of Egmont Key was visi ted. No living sand dollars were found, only Mellita tenuis and Encope aberrans tests were found along with a visible detrital layer on top of the substrate. However, in shore off Fort De Soto North Beach, M. tenuis appeared closer to the shore than normally, in only a few centimeters of water (pers. observation). Possibly, M. tenuis was migrating inshore to avoid th e anoxia, thus moving between high and low nutrient areas. Sand dollars, by aerati ng the sediment and grazing on the benthic community, have a distinct impact on their habitat. A better unders tanding of sand dollar feeding may further highlight th eir importance to the ecosystem. Information about the food and feeding of sand dollars is extremely limited. Except for Timko (1976) and ONeill (1978) on Dendraster excentricus and Lane and Lawrence (1982) on Mellita tenuis, information is restricted to general statements about food composition. Even their studies are li mited to a single site and laboratory observations. Additional information on population differences of a species and of other species of sand dollars would contribute to understanding better their biology and ecology. Mellita tenuis Encope aberrans and Encope michelini of the central Florida Gulf Shelf provide the opportunity to make such comparisons. They are scutellid sand dollars in the family Mellitidae (Mooi 1989). They have interesting differences in their distribution. Serafy (1979) repor ted the distribution of the th ree species on the Florida Gulf Shelf. Mellita tenuis was most abundant at a stati on with well sorted, fine quartz
10 sand with a modal grain size of 0.18 mm at 6 m depth. It was much less abundant at a station with crushed shell and poorly sorted quartz sand with a model grain size of 0.6 mm. A few Encope michelini also occurred at this site. Encope michelini and Encope aberrans occurred in fewer numbers at greater de pths of 18 and 37 m on crushed shell and quartz sand. Encope michelini was much more common than E. aberrans. We have found E. michelini and occasional E. aberrans at one of Serafys 18 m sites and E. aberrans along with abundant M. tenuis at one of his 6 m site s. Neither species of Encope were found at sites less than 4 m depths fr om Panama City to Naples, Florida where M. tenuis was found (Tan and Lawrence 2001). However, Telford and Mooi (1986) and Kurz (1995) found M. tenuis and E. michelini together near Cedar Key, Florida and in the Florida Keys. Leodia sexiesperforata is also found at Cedar Key (J. Herrera, pers. communication). The question addressed here is whether f ood and feeding of three species of sand dollars in the family Mellitidae ( Mellita tenuis Encope aberrans and Encope aberrans ) vary with respect to species and over space and time. Organic content of the gut contents and sediment are expected to differ on both a diel and seasonal scale. In addition, the organic content of the sediment across si tes should be different; thereby supporting different sand dollar species. Particle sizes in the guts should be smaller since smaller particles, due to their surf ace area to volume ratio, have a higher nutritive value than larger particles and sand dollars possess powerful la nterns to further crush those particles.
11 Materials and Methods Collection Sites For the seasonal study, specimens and sediment were collected from March through September 2004. Encope michelini were collected approximately 22.4 km off Sanibel Island, Florida (26 32.43 N, 82 29.163 W, Hourglass Station J, Serafy 1979) at 18 m depth (Fig. 1). Encope aberrans and Mellita tenuis were collected approximately 8.5 km off Egmont Key, Florida (27 35.00N, 082 50.00 W, Hourglass Station A, Serafy 1979) at 7 m depth in March. Mellita tenuis were also collected in June and September 2004 off of Bean Point, Anna Maria Island (27 31.54 N, 82 45.26 W) at 1-2 m depth. Additional sediment collecti ons and sand dollar densities measurements were taken from sites in Tampa Bay (27 49.42 N, 82 24.10 W) at 1-2 m depth. Sediment only was collected from sand dolla r sites at the North (27 38.41 N, 82 45.49 W) and East (27 35.48 N, 82 42.10 W) beaches of Fort De Soto Park, and Lido Key (27 9.40 N, 82 35. 12 W) in 1-2 m depth (Table 1). At each site, when weather conditions pe rmitted, a 30 m transect line was run in triplicate. A 1 m 2 quadrat was placed every 5 m and the number of sand dollars was counted to determine the density. In poor weather conditions, a 1 m 2 quadrat was thrown randomly (n=3) and sand dollars were counted. The width of each was measured at the
Figure 1. Sand dollar collection sites. Encope michelini was collected from the Offshore Charlotte Harbor site. Encope aberrans was collected from Of fshore Egmont Key. Mellita tenuis was collected at all other sites (map: Google Earth v3.0 2005). widest point (just above the II and IV ambulacra, after Love n 1874) with Vernier calipers to determine a size-frequency distribution. For the diel study, 20 Mellita tenuis were collected ever y 34 hours at Anna Maria Island, June 20, 2004 and September 25, 2004. Encope michelini were collected in a similar manner at the offshore Charlotte Harbor site in March 20, 2004 and September 17, 2004. 12
13 Table 1. Collection sites. Collection Site Coordinates Depth (m) Date Species present 22.4 km W Charlotte Harbor 26.43' N, 82.163' W 18 9/27/2003 Encope michelini, Encope aberrans 3/20/2004 Encope michelini 9/28/2004 Encope michelini 8.5 km W Egmont Key 27.00' N, 82.00'W 7 9/28/2003 Mellita tenuis, Encope aberrans 3/21/2004 Mellita tenuis, Encope aberrans Anna Maria Island 27.09' N, 82.49' W 2-3 6/20/2004 Mellita tenuis 9/25/2004 Mellita tenuis Tampa Bay, W Apollo Beach 27.07' N, 82.23' W 2 10/24/2004 Mellita tenuis Lido Key 27.04' N, 82.12' W 2-3 10/28/2004 Mellita tenuis Fort De Soto North Beach 27.11' N, 82.24' W 2-3 10/28/2004 Mellita tenuis Fort De Soto East Beach 27.51' N, 82.36' W 3-4 10/28/2004 Mellita tenuis Preservation of Samples All specimens collected were preserved by the method of Telford et al. (1987). Immediately after collection, specimens were injected with 20% formalin in seawater into the coelomic cavity. They were then immersed in the formalin solution and stored until analysis. Sediment collected was also pres erved with 20% formalin and refrigerated. Sediment from the offshore Gulf site with and without formalin was compared because addition of formalin may sli ghtly increase the organic co ntent of the samples (Crisp 1971). Sediment samples with formalin were, on average, 1.5% higher than the sediment without formalin. However, the necessity of specimen preservation, i.e. eliminating decomposition before samples could be processed, outweighed the small increase in values.
14 Analysis of Gut Contents and Substratum The aboral test and gonads were removed from sand dollars with forceps and the guts were extracted from 20 specimens per coll ection. Guts and their contents were stored in vials and bathed in 20% formalin. The guts were divided into four segments: stomach, upper intestine, mid-intestine, and lo wer intestine. The lo cation of food in the gut tract was noted (Fig. 2). Extracted guts were cleaned of any test or gonad material. Samples of gut contents were observed under a microscope and the sizes of 150 ingested sediment particles were measured using a ca librated micrometer (following the method of Telford 1986). The optimum sample size was determined by comparing variances for samples sizes of 50, 100, 150, 200 and 250. Similar variances existed between sample sizes of 150, 200 and 250; consequently a samp le size of 150 was used. A wet smear of gut contents was placed on a slide and examined under a light microscope. Several different locations on the slides were exam ined for particle size measurements. Resolution decreased at magnifications higher than 400X so that accurate measurement of particles was difficult. Thus, the maximum magnification used was 400X. The smallest particles measured were 0.005 mm though the actual size may have been smaller. Afterwards, the entire gut sample was dried for at least 24 hours in a fume hood under loose cover. The dried gut contents were weighed, ashed in a muffle furnace at 500 C for 4 hours and weighed again to calcul ate the percent organic material (Paine 1971). Possibly, some organic material from the gut was present in the samples used to measure percent organic material.
Figure 2. Diagram of internal anatomy of sand dollar. The gonads are aboral to guts and have been removed to show the intestinal tract. The basic layout for the guts of Encope aberrans Encope michelini and Mellita tenuis are similar. (Key: A= Pharynx going through Aristotles lantern, B= Stomach, C= Last part of intestine, D= Mid intestine, E= Anus, F= First part of intestine.) (Drawing by author, modified from Hyman (1955).) Sediment was sampled at the collection s ites. Three samples of approximately 60 ml of sediment were collected from the surface layer at each collection. Sediment samples were fixed in 20% buffered forma lin immediately after collection. Before analysis, samples were dried in an oven at 60 C for at least 48 hours. Sediment samples were dry-sieved using the US standard siev e series and sieve segments were weighed (adapted from Mitbarker and Anil 2002). A small portion of each sediment sample was taken to measure the percent organic content in the same manner as the gut contents. The 15
16 ashing treatment at 500 C does not degrade carbonate sediment, found at the offshore site (Lawrence, pers communication). Food Groove Analysis Ten Encope michelini collected March 2005, from O ffshore Charlotte Harbor, and ten Mellita tenuis collected October 2005, from Anna Maria Island were preserved in 20% buffered formalin in individual plastic bags. Sand pa rticles from the food grooves were removed using a spatula and smeared on a slide, along with a small amount of deionized water. Particle sizes were meas ured following the same method for particle analysis of the gut contents. Statistical Analysis The size-frequency distribution of test diameters between groups was analyzed using the Kolmogorov-Smirnoff Test (Zar 1999). The particle size distribution in the guts, food grooves and sediment was analy zed using a percent cumulative frequency distribution (Folk 1965). Sedime nt grain size distribution wa s also analyzed using the program GRADISTAT version 4.0 (Blott 2000). Particle distributions were compared using the Kolmogorov-Smirnoff Test (Zar 1999) and mean particle sizes were compared using non-parametric tests (Kruskal-Wall ace and Mann-Whitney U-test) (SPSS student, version 13.0 2004). The percent organic material of gut samples and sediment was compared using ANOVA techniques and, when n ecessary, non-parametr ic tests (KruskalWallace and Mann-Whitney U-test) (SPSS student, version 13.0 2004). The units, a geological measurement ( = -log 2 mm), were used when dealing with sediment because the sediment was sieved giving a mass for each size class. However, SI units were used for analyzing the gut particles, since the inte rest there was in the frequency and sizes of
17 the sand grains (after Telford and Mooi 1986). SI units were used to compare sediment, gut and food groove particles. Descriptive st atistics were used to evaluate populations and diel variation in particle sizes and organic content of sediment and gut contents (SPSS student, version 13.0 2004).
18 Results Description of Populations The average size of E. michelini off Sanibel Island was 8.12 0.47 cm in Fall 2003, 8.46 0.53 cm in Spring 2004 and 9.03 0.59 cm in Fall 2004 (Fig. 3). The sizefrequency distributions increased over time a nd were all significantly different from each other (Kolmogorov-Smirnoff Test, D 0.05, 25 = 0.26404). The average size of M. tenuis off Egmont Key (Fall 2003) was 6.55 0.50cm a nd Anna Maria Island (Summer 2004) 7.15 0.85cm, (Fall 2004) 7.70 0.77cm and Tampa Bay (Fall 2004), 10.85cm 0.85 (Fig. 4). The September 2004 Anna Maria Island co llection was significantly greater than the June 2004 collection. The size-frequency distributions for the Tampa Bay collections were significantly greater than the other collections of M. tenuis which differed significantly among themselves as well. De nsities of sand dollars ranged from 0.43 0.60 E. michelini m -2 (September 2004) to 49 21 M. tenuis m -2 (Fort De Soto, Apr. 2004). At Anna Maria Island (June 2004), mean density during the day was 21.4 7.9 M. tenuis m -2 but after sunset th e density was 7.6 3.6 M. tenuis m -2
0 5 10 15 20 25 30 35 40 45 50 6.50-6.997.00-7.497.50-7.998.00-8.498.50-8.999.00-9.499.50-9.9910.00-10.49 Size frequency categories (cm)% Frequency Fall '03 (N=96) Spring '04 (N=100) Fall '04 (N=79) Figure 3. Size-frequency distribution of Encope michelini (test width = diameter taken above anterior notches). 0 5 10 15 20 25 30 35 40 45 504 .50-4.99 5 005. 49 5 .50-5.9 9 6 006. 49 6 .50-6.99 7 007. 49 7 .50-7.99 8 008. 49 8 .50-8.99 9 .00 9. 49 9 509 .99 10. 0 0 10 49 10.5 0 -10.99 11. 0 0 11.49 11.5 0 -11. 9 9 12. 0 0 12.49 12. 5 -12.99Size-frequency categories (cm)% Frequency Tampa Bay Fall '04 (N=28) Anna Maria Island Sum '04 (N=80) Egmont Key Fall '03 (N=74) Egmont Key Spr '04 (N=20) Anna Maria Island Fall '04 (n=79) Figure 4. Size-frequency distribution of Mellita tenuis (test width = diameter from above anterior lunules). 19
20 Particle Analysis Sediment The sediment of the eastern Gulf coast is generally quartz sand inshore and quartz mixed with shell hash offshore (FIO 1994). The sediment particle-size distribution varied with site (Fig. 5). The sites with M. tenuis had smaller particles and differed among themselves (Table 2). The sediment off Sanibel Island ( E. michelini site) had a graphic mean particle size of 0.352 0.937 which corresponds to the coarsest sediment of all the collections. Sediment ra nged from very fine gravel to fine sand. The sediment off Anna Maria Island ( M. tenuis site) had a mean particle size of 1.645 0.628 The site with the finest sediment wa s Fort De Soto North Beach (2.260 0.461 ). The inshore sites ranged from course sand to very fine sand. Moreover, sediment grains from offshore and inshore sites are different (F ig. 6). Sediment from the inshore sites is smaller and mostly quartz with very little sh ell fragments. Offshore, the particles are larger and carbonate particles are more fre quent. Both inshore a nd offshore sediments are composed of smooth, rounded particles. No significant differences were found among sediment particle distributions (Kolmogorov-Smirnoff Test). However, this seemed doubtful since the particles from inshore and offshore appeared so different in composition and size. Consequently, mean particle sizes from different sites were comp ared and some differences were observed. Mean particle size of the sand dollar site sa mples were significantly different (KruskalWallis Test, 2 = 23.249, p<0.003) (Fig. 7). The inshore sites in general have smaller mean particle sizes than the two offshore sa mpling dates. Most pair-wise comparisons
-20 0 20 40 60 80 100 120 -1.5-1-0.500.511.522.533.5 Sediment particle size (phi = -log2mm)Mean cumulative percent frequency (g ) 3/04 Offshore 9/04 Offshore 3/04 Egmont Key 6/04 Anna Maria Island 9/04 Anna Maria Island 10/04 Tampa Bay Lido Key 10/04 Ft. DeSoto N Beach 10/04 Ft. DeSoto E Beach 10/04 Figure 5. Percent cumulative frequency distributi ons of sediment particles from habitats with E. michelini and M. tenuis Each curve represents a cumulative percent frequency distribution of the mean values in each size category. Note the phi units are on a log scale. Figure 6. Photographs of offshore and inshore sediment from a light microscope (40X). Left: Sediment from Anna Maria Island. Right: Sediment from Offshore Charlotte Harbor. 21
22 Table 2. Mean particle sizes of sediment, food grooves and gut contents. Collection Mean particle size (mm) 2 Sediment Offshore 9/2003 0.78 0.0019 Offshore 3/2004 0.61 0.0021 Offshore 9/2004 0.60 0.0020 Offshore 3/2005 0.29 0.0018 Egmont Key 3/2004 0.27 0.0017 Egmont Key 3/2005 0.22 0.0016 Anna Maria Island 6/2004 0.34 0.0016 Anna Maria Island 9/2004 0.32 0.0015 Ft. De Soto N 5/2004 0.24 0.0016 Ft. De Soto N 10/2004 0.21 0.0014 Ft. De Soto E 10/2004 0.21 0.0014 Lido Key 10/2004 0.22 0.0014 Tampa Bay 10/2004 0.26 0.0016 Gut Contents M. tenuis 3/04, Egmont Key 0.010 0.018 M. tenuis 6/04, Anna Maria Island 0.020 0.038 M. tenuis 9/04, Anna Maria Island 0.0086 0.013 E. aberrans 3/04, Egmont Key 0.076 0.12 E. michelini 3/04, Offshore 0.010 0.026 E. michelini 9/04, Offshore 0.0093 0.013 Food Grooves M. tenuis 10/05, Anna Maria Island 0.066 0.14 E. michelini 3/05, Offshore 0.015 0.038 were significantly different, except seasonal collections from Anna Maria Island (June and September 2004) and Offshore Charlotte Harbor (March and September 2004) were not significantly different (Mann-Whitney U-test, p<0.05). Sediment from Egmont Key (March 2004) was not significan tly different from any other inshore sites, except Anna Maria Island (June 2004). Gut Contents The gut particle size-frequency distributi ons among collections (Table 2) were not significantly different between any pairing (K olmogorov-Smirnoff Test). However, this
Offshore 9/04 Offshore 3/04 Egmont Key 3/04 Tampa Bay 1 10/04 Lido Key 10/04 Ft DeSoto N 10/04 Ft DeSoto E 10/04 AMI 6/04 AMI 9/04 1:30pm Sand dollar sites 700.00 600.00 500.00 400.00 300.00 200.00 Mean particle size (micrometers) Figure 7. Box and whiskers plot of mean sediment particle sizes at sites. 0 10 20 30 40 50 60 70 80 90 0-9.910-99.9100-199.9200-299.9300-399.9400-499.9500-999.91000+ Particle Size Range (micrometers)Percent of Sample M. tenuis 9/04 (A.M.I.) M. tenuis 6/04 (A.M.I.) M. tenuis 3/04 (E.K.) E. aberrans 3/04 (E.K.) E. michelini 9/04 E. michelini 3/04 Figure 8. Size-frequency distribution of particles in the guts of Encope aberrans, Encope michelini and Mellita tenuis (A.M.I. = Anna Maria Island, E.K. = Egmont Key.) 23
24 may be suspect because approximate ly 75% of the gut particles for Encope michelini and Mellita tenuis were less than 9.9 m, which may unduly skew the results of the distribution test. Moreover, the shapes of the size-freque ncy distributions indicate differences (Fig. 8). Encope aberrans has larger gut particles. Approximately 75% are between 10-99.9 m. Larger-sized particles were also present. The cumulative frequency distributions for E. michelini and M. tenuis were similar, but, the shape of the cumulative frequency for E. aberrans is appears different fr om the other two species (Fig. 9). The gut contents of E. michelini have smaller particles than M. tenuis (Figs. 10 and 11). The guts are characterized by some large particles (10-200 m) and many very small particles (< 9.9 m). The guts of M. tenuis have more loose organic material, probably filamentous algae or detritus. The guts of Encope aberrans had the largest average particle size, 0.0757 0.1153 mm, as well as the largest maximum size (1.6925mm). In addition, the distribution wa s different from the other two sand dollar species (Fig. 8). The mean sizes of gut particles were compared across seasons for M. tenuis and E. michelini and species co-occurring at Egmont Key, M. tenuis and E. aberrans (Fig. 12). Nonparametric statistics were employed to compare the means as variances were not homogeneous. Gut particles from M. tenuis collected June 2004 were significantly larger than September 2004 (Mann-Whitney U-te st, p<0.000). The gut particles of E. michelini, collected in March 2004 were significantly gr eater than the September 2004 collection (p<0.000). Gut particles from E. aberrans were significantly greater than M. tenuis, both collected March 2004 from Egmont Key (p<0.000). Notably, E aberrans had the largest range of values, whilst the ranges of the othe r two species were much less. Centric and
-20 0 20 40 60 80 100 120 -2-1012345678 Phi (-log2mm)Percent cumulative frequenc y M. tenuis 9/2004 M. tenuis 6/2004 M. tenuis 3/2004 (E.K.) E. aberrans 3/2004 E. michelini 3/2004 E. michelini 9/2004 Figure 9. Percent cumulative frequency distributions of particles in the guts of Encope aberrans, Encope michelini and Mellita tenuis. pennate diatoms, filamentous green algae, flat worms (possibly parasites), and foraminifera were observed in the gut contents of all species. Comparison of Sediment, Food Grooves and Gut Particles The sediment particle, food grooves, and gut particle distributions were compared. The average particle size in the food grooves was 0.0147 0.0383 mm for E. michelini and 0.0659 0.136 mm for M. tenuis. This is not signifi cantly different from all their gut contents. However, the particle distribution of the food grooves is larger than the gut contents (Fig. 13). The average gut particle sizes ranged from 0.0086 0.013 for Mellita tenuis to 0.076 0.12 mm for Encope aberrans (Table 2). The average sediment particle sizes ranged from 0.7834 0.001915 mm (offshore September 2004) to 0.2088 25
Figure 10. Photographs of gut contents of E. michelini from a light microscope (100x). Left: (A) a foraminiferan test is left center. Right: (B) long pennate forms. Figure 11. Photographs of gut contents of M. tenuis from a light microscope (100X). Left: (A) several large sediment particles with jagged edges and (B) some gut tissue. Right: (C) tufts and filamentous strands of green algae. 26
Emich 9/04 Emich 3/04 Eaber 3/04 Mtenu 9/04 Mtenu 6/04 Mtenu 3/04 Collection 0.1200 0.1000 0.0800 0.0600 0.0400 0.0200 0.0000 Mean gut particle Size 76 28 21 Figure 12. Box and whiskers plot of mean particle sizes of gut contents of Encope aberrans, Encope michelini and Mellita tenuis 0 10 20 30 40 50 60 70 80 0-9.9 10-99.9100-199.9200-299.9300-399.9400-499.9 Particle size range (micrometers)Percent of sample Gut 9/04 Gut 3/04 Food Grooves 3/05 0 10 20 30 40 50 60 70 80 0-9.910-99.9100-199.9200-299.9300-399.9400-499.9500-999.91000+ Particle size range (micrometers)Percent of sample Gut 9/04 (A.M.I.) Gut 6/04 (A.M.I.) Gut 3/04 (E.K.) Food Grooves 10/05 Figure 13. Comparison of particle size distribution in food grooves and gut contents. Left: Encope michelini collections. Right: Mellita tenuis collections. 27
28 0.00138 mm (Fort De Soto North Beach April 20 04). The mean particle size for the sediment is larger than the gut particles by one order of magnitude. It is important to note that the sample sizes of sand dolla rs and sediment were disparate (n gutparticle = 20, n sediment = 3); which prevents valid st atistical comparison. Sedime nt particles are large and rounded (Fig. 6), while the gut particles are much smaller and often have jagged edges (Figs. 10 and 11). Comparison of sediment and gut particles proved difficult because the size fractions of the gut particles were so small and there was very little overlap between them and sediment particles. Organic Content Analysis Sediment The percent organic content of th e sediment was examined. Across Mellita tenuis sites, the percent organic content was significantly different (Kruskal-Wallis Test 2 = 29.163, p=0.001). In addition, sites were compared pair-wise (Mann-Whitney U-test, p<0.05) to analyze specific differences. Se diment from Fort De Soto East Beach (October 2004) had significantly greater orga nic content than other inshore sites (Fig. 14). Sediment from Anna Maria Island 9:30 am, September 2004 was not significantly different from any other sites. Yet, other than the collections from the same day, the pvalues were very close to be ing significantly different (Tab le 3). The Tampa Bay sites, although not significantly different from each other, had significantly lower organic content compared to all other sites, excep t Anna Maria Island 9:30 am, September 2004. Lido Key shared the most with the other si tes and was only signifi cantly different from four other sites (Anna Maria Island, Se ptember 2004 1:30 pm, 5:00 pm, and 8:00 pm collections and Ft. De Soto E. Beach Oct ober 2004). Seasonally, the sediment from
Offshore 9/04 Tampa Bay 3 10/04 Tampa Bay 2 10/04 Tampa Bay 1 10/04 Lido Key 10/04 Ft DeSoto N 10/04 Ft DeSoto E 10/04 AMI 6/04 AMI 9/04 8pm AMI 9/04 5pm AMI 9/04 1:30pm AMI 9/04 9:30am Sand dollar sites 6.0 5.0 4.0 3.0 2.0 1.0 0.0 % Organic Content Figure 14. Box and whiskers plot of percent organic content of sediment from sites. Percent organic content is calculated by dividing the weight (g) of the sample post-ashing by weight (g) pre-ashing. Table 3. Pair wise comparisons of percent organic content of sediment from sites with Mellita tenuis The nonparametric Mann-Whitney U-test was u tilized because variances we re not homogeneous. The upper right triangle gives information on significant differences (sig. = significantly different at p<0.05, n.s. = not significantly differe nt); the lower left triangle gives the p-values for the pair wise test. Collection AMI 9/04 11am AMI 9/04 3pm AMI 9/04 6pm AMI 9/04 9pm AMI 6/04 Ft. De Soto E 10/04 Ft. De Soto N 10/04 Lido Key 10/04 Tampa Bay 1 10/04 Tampa Bay 2 10/04 Tampa Bay 3 10/04 AMI 9/04 11am X n.s n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. AMI 9/04 3pm 1 x n.s. n.s. sig. sig. sig. sig. sig. sig. sig. AMI 9/04 6pm 0.8 0.184 x n.s. sig. sig. sig. sig. sig. sig. sig. AMI 9/04 9pm 0.564 0.275 0.827 x sig. sig. sig. sig. sig. sig. sig. AMI 6/04 0.64 0.034 0.034 0.034 x sig. n.s. n.s. sig. sig. sig. Ft. De Soto E 10/04 0.083 0.05 0.05 0.05 0.034 x sig. sig. sig. sig. sig. Ft. De Soto N 10/04 0.083 0.05 0.5 0.05 0.48 0.05 x n.s. sig. sig. sig. Lido Key 10/04 0.083 0.05 0.05 0.05 0.077 0.05 0.127 X n.s. n.s. n.s. Tampa Bay 1 10/04 0.076 0.046 0.046 0.046 0.032 0.046 0.046 0.825 x n.s. n.s. Tampa Bay 2 10/04 0.083 0.05 0.05 0.05 0.034 0.05 0.05 0.513 0.268 x n.s. Tampa Bay 3 10/04 0.083 0.05 0.05 0.05 0.034 0.05 0.05 0.513 0.487 0.275 x 29
30 Anna Maria Island had significantly lower or ganic content in June 2004 than September 2004 (Kruskal-Wallis 2 = 8.265, p=0.004). Geographically, the sites vary across space (Fig. 1). The sediment organic content did not vary over a diel period (Anna Maria Island, September 2004, Table 3). Inshore, the average organic content ranged from 0.355 0.046% (Tampa Bay 2 October 2004) to 1.759 0.294% (Ft. De Soto E October 2004). At the offshore Encope michelini site, sediment organic content was 4.24 1.26% (September 2004), which is on average 2-4 ti mes higher than the inshore sediment. The mean particle size and mean organic content of sites were compared. Smaller particles have slightly lower organic content, however, th ere was some variation (Fig. 15). The sediment from Fort De Soto East Beach, at the mouth of Tampa Bay, has a greater organic content than any other inshore sites although its mean particle size is quite small. In contrast, the offshore site has la rger particles and great er organic content. Larger particles correlated w ith higher organic content (R 2 linear = 0.699). Gut Contents The organic content of the gut samples varied over time, both seasonally and hourly. Gut mean percent organic content for Mellita tenuis ranged from 26.96 6.28% (3 pm June 2004) to 44.48 4.93% (1:30 pm September 2004). The sediment organic content at Anna Maria Island is, on average, 0.579 0. 185%; this corresponds to a 47-77 fold difference in organic content between th e gut contents and the ambient sediment. The gut contents from Mellita tenuis collected in Septembe r 2004 had a significantly greater overall organic content than th e June 2004 collection (ANOVA, F = 68.510,
700.00 600.00 500.00 400.00 300.00 200.00 Mean particle size (micrometers) 6.0 5.0 4.0 3.0 2.0 1.0 0.0 % Organic content Fit line for Total Offshore 9/04 Ft DeSoto E 10/04 Tampa Bay 1 10/04 Lido Key 10/04 Ft DeSoto N 10/04 AMI 6/04 AMI 9/04 1:30pmS.D. Site R Sq Linear = 0.699 Figure 15. Scatter plot of mean particle size and percent organic content at sites. p<0.000) (Fig. 16). Furthermore, the sampli ng times were significan tly different from each other (ANOVA, F= 18.292, p<0.000). In June 2004, the 11 am collection was significantly higher than the 3 pm collection, and the 3 pm collection was significantly lower than the 9 pm collection (Tukey HS D, p<0.05). For September 2004, the diel collections were not significantly differe nt from each other. Most September 2004 collections were significantly greater than the June 2004 collections, except 11 am June 2004 and 9:30 am September 2004 (Table 4). For the June 20, 2004 collections, the 11 am collection had the greatest organic content and the 3 pm collection had the lowest. In September 25, 2004, the 8 pm collection had the greatest organic content and the 9:30 am collection had the lowest, albeit this was great er than any of the June 2004 collections. The gut contents had 30-55 times more organic content than the sediment. 31
Table 4. Pair wise comparisons of percent organic content of gut contents of Mellita tenuis Upper right triangle relates significant differences (=sig.) or not significant (n.s.) using Tukeys HSD; lower right triangle, are pair wise comparisons using the Bonferroni post-hoc test. Two different measures were used to test whether the statistics were equiva lent in sensitivity, however there was only 1/28 cases where the two tests disagreed. Collection 6/04 11am 6/04 3pm 6/04 6pm 6/04 9pm 9/04 9:30am 9/04 1:30pm 9/04 5pm 9/04 8pm 6/04 11am x sig. n.s. n.s. n.s sig. sig. sig. 6/04 3pm sig. x n.s. sig. sig. sig. sig. sig. 6/04 6pm n.s. n.s. x n.s. sig. sig. sig. sig. 6/04 9pm n.s. sig. n.s. x n.s sig. sig. sig. 9/04 9:30am n.s. sig. sig. n.s. x n.s. n.s. n.s. 9/04 1:30pm sig. sig. sig. sig. n.s x n.s. n.s. 9/04 5pm n.s. sig. sig. sig. n.s n.s. x n.s. 9/04 8pm sig. sig. sig. sig. n.s n.s. n.s. x 4 3 2 1 Time of Day 60.0 50.0 40.0 30.0 20.0 10.0 Gut % Organic September JuneSeason Figure 16. Scatter plot of seasonal mean percent organic content of gut contents of M. tenuis at Anna Maria Island. (Time of day codes are 1 = 9:30 -11:30am, 2 = 1:30-3:30 pm, 3 = 5-6pm, and 4 = 8-9 pm. Percent organic content is calculated by dividing the weight (g) of the sample post-ashing by weight (g) pre-ashing. 32
33 The organic content of gut samples of Encope michelini across seasons was significantly different (ANOVA, F = 110.713, p< 0.000) (Fig. 17). Sampling times were also significantly different from each other (Kruskal-Wallis, 2 = 79.706, p<0.000). The 5:55 pm collection for March 20, 2004 had significantly lower organic content, 23.33 1.91%, than collections at 3:10 pm, 8:00 pm and 10:30 pm (Mann-Whitney U-test, p<0.05) (Table 5). The 10:30 pm collection had the highest mean organic content for that day, 27.96 4.80%. The September 17, 2004 colle ctions overall had a significantly greater mean organic content than the March 2004 collections. The co llections from that date were not significantly different from each other; but they were all significantly higher than the March 2004 co llections (Table 4). The 4:20 pm collection had the greatest mean percent organic content, 36.22 6.77%, of all co llections, albeit nonsignificant. The gut contents had 5.5-8.5 times more organic content than the sediment. The organic content of the sediment and the gut contents could not be compared statistically because of the differe nce in sample sizes. The organic content of the gut contents of Encope aberrans and Mellita tenuis at the same site were compared. Encope aberrans collected from 8.5 km west of Egmont Key had a significantly smaller mean organic content than M. tenuis (Kruskal-Wallis, 2 = 29.268, p<0.000) (Fig. 18). The mean gut particle size and mean pe rcent organic content for each collection were compared (Fig. 19). The gut contents of Mellita tenuis and Encope michelini clumped closely to each other across seasons and sites. Encope aberrans clumped separately from the other two species. Notably, M. tenuis, collected in March 2004 at Egmont Key along with E. aberrans clumped together with other M. tenuis collected at
5 4 3 2 1 Time of Day 50.0 40.0 30.0 20.0 Gut % Organic Content September MarchSeason Figure 17. Scatter plot of seasonal mean percent organic content of gut contents of E. michelini at offshore Charlotte Harbor. (Time of day codes are 1 = 8:30-9:30 am, 2 = 1:30-3:30 pm, 3 = 4-6 pm, 4 = 89pm, and 5 = 10:3011:30 pm.) Table 5. Pair wise comparisons of percent organic content of gut contents of Encope michelini The nonparametric Mann-Whitney U-test was utilized because the variances were not homogeneous (sig. = significantly different, n.s. = no t significantly different). Collections 3/04 9:25am 3/04 3:10pm 3/04 5:55pm 3/04 8pm 3/04 10:30pm 9/04 8:30am 9/04 1:30pm 9/04 4:20pm 9/04 8:25pm 3/04 9:25am x sig. n.s n.s sig. sig. sig. sig. sig. 3/04 3:10pm x sig. n.s n.s sig. sig. sig. sig. 3/04 5:55pm x sig. sig. sig. sig. sig. sig. 3/04 8pm x n.s. sig. sig. sig. sig. 3/04 10:30pm x n.s. sig. sig. sig. 9/04 8:30am x n.s. n.s n.s. 9/04 1:30pm x n.s. n.s. Dive #3 9/04 x n.s. 9/04 8:25pm x 34
Mtenu 3/04 Eaber 3/04 Egmont Key 40.0 30.0 20.0 10.0 % Organic Content Figure 18. Box and whiskers plot of mean percent organic content from the gut contents of Encope aberrans and Mellita tenuis, two co-existing species off Egmont Key. 0.120 0.100 0.080 0.060 0.040 0.020 0.000 Mean Gut Particle Size (mm) 60.0 50.0 40.0 30.0 20.0 10.0 0.0 % Organic Content Fit line for Total Emich 9/04 Emich 3/04 Eaber 3/04 Mtenu 9/04 Menu 6/04 Mtenu 3/04Collection R Sq Linear = 0.617 Figure 19. Scatter plot of mean gut particle size and mean percent organic content of gut contents of Encope aberrans, Encope michelini and Mellita tenuis 35
36 Anna Maria Island and E. michelini, collected offshore. The sm allest mean particle size correlated with the grea test organic content (R 2 linear = 0.617).
37 Discussion Description of Populations The change in size distributions for Encope michelini may indicate a single aging cohort. Recruitment has not occurre d at the offshore site (Lawrence, pers. communication). Recruitment of Mellita tenuis is higher inshore with typically biannual events (Lawrence, pers. communication). The size-frequency distributions of M. tenuis from date to date generall y cluster around the sa me mode, and are probably a single cohort. The similarity in modes among co llections may indicate that recruitment is widespread in the Gulf populations of M. tenuis. The size-frequency distributions of M. tenuis at Anna Maria Island indi cate cohort aging as the mean increased from June to September. The Tampa Bay collection of M. tenuis is much larger than those from other sites. The difference in their size could i ndicate a difference in age or growth rate. However, size and age are uncoupled in M. tenuis (Tan and Lawrence 2001). Moreover, M. tenuis is a stress-tolerant ruderal species, char acterized by high fecundity and short life span (Lawrence 1990). Less information on the life-history of Encope michelini is available to determine its life-hi story strategy. However, since E. michelini exists in deeper water, it probably experiences fewer di sturbances and more st ress (due to lower primary productivity) than M. tenuis. More research on fecund ity and recruitment for this species is needed. The difference in distribution of the two species might indicate
38 specific habitat requirements, possibly a ssociated with particle size-frequency distribution of the sediment. Particle Analysis The particle sizes of sediments from the habitats of E. michelini and M. tenuis are different. The offshore sites have much co arser sediment than the inshore sites. Encope aberrans the largest sand dollar, is found in both inshore and offshore sites intermittently. The range of sediments that M. tenuis inhabits is quite wide; from the very fine sand of the East Beach of Fort De Soto and the Tampa Bay sites to the coarser sandy sites west of Egmont Ke y. The results here, a mixture of quartz and carbonate, are consistent with past studies (Brooks et al 2003). Quartz part icles are predominant inshore and quartz and carbonate particles are mixed offshore. Carbonate sands are porous while quartz is solid (Folk 1965). The types of benthic microalgae could be intrinsically affected by this at the offshore and inshore site s, and may partially explain differences in species distributions. The gut particles are small. For both E. michelini and M. tenuis 60-75% of the particles were less than 10 m. Encope aberrans had larger particles; the mean particle size was 75 m. However, all species feed on a much smaller subset of the ambient sediment. Lane and Lawrence (1982) found that the majority of the particles in the guts were less than 62 m. Telford et al. (1985) found th at 79% of particles in the gut M. tenuis were less than 100 m. Both of these past studies are close to what was observed in the current study. In addition, mean particle sizes in the food grooves were more similar to those in the guts than the se diment. Sand dollars probably select small particles, and then further crus h particles with their lanterns There is evidence of sand
39 grain fracture as particles in the guts had sharp edges while those in the sediment were rounded and polished. However, it is unlikel y that the calcium carbonate lanterns are able to crush the harder quartz particles. The results here corr espond to the findings of Telford et al. (1985): podia select manageable particles and the lantern crushes softer particles to aid digestion. The difference in gut particle sizes between Encope aberrans, Encope michelini and Mellita tenuis may explain, in part, why E. aberrans can coexist with the two other species. It competes with neither for particles. Whitl ach (1980) found advantages to feeding on both large and small particles, while studying the feeding niches of polychaetes. Specializing on small particles pr ovides higher quanti ties of nitrogen and bacteria, because of the inverse relati onship between surface area and volume. Specializing on larger particles, though th ey are less abundant, greatly increases the probability of encountering a food particle. Statistical comparison of sediment and specimen particles proved difficult. When analyzing sediment and gut particles, differe nt techniques were necessarily employed. Mean particle size of sediment samples was determined by sieving and weighing the mass for each size fraction. The volume of gut contents retrie ved was so small, using a sieve was impossible. Mean particle size of gut and food groove samples was determined by measuring the first 15 0 particles encountered. There is an inherent bias between these two analytical techniques. However, the goa l for each technique was to estimate the true particle size of the sample and each does so successfully to a degree. Yet, the objective here was to determine whether the guts c ontained different sized particles from the ambient sediment. Upon visual examination of the different samples, it was evident that
40 the gut particles are much smaller than the sediment. The measurements provided by the two techniques substantiate this; there was very little overlap in par ticle sizes for gut and sediment particles. This alone may be evid ence of differences in particle size between gut contents and sediment. Moreover, the reso lution of the very small particle sizes at high magnification decreased making precise measurement difficult. However, the purpose of measuring gut particles was for co mparison to the sediment. In general, particles that were measured as 0.005 mm could be that size or smaller. Particles smaller than 0.005 mm would only further validate the conclusion there are extreme differences between the gut and sediment particle sizes. Organic Content Analysis The organic content of the sediment at both Encope and Mellita sites varied across seasons. This, of course, is expected, given the annual increase of algae during the summer and decrease with the cooler temperatur es and reduced daylight of winter in the subtropics (Dawes 1998). The sediment or ganic content did not change significantly over the course of the day. The Mellita Gulf sites are fairly similar in terms of organic content, however elsewhere where Mellita exist the range of organic content is great. The site located on the mouth of Tampa Bay (Fort De Soto East beach) has the highest organic content, most likely because daily tidal cycles bring in and draw out large amounts of water to the bay through the mout h which would probabl y yield organically rich sediment. Moreover, the Tampa Bay site s had the lowest organic content of all the sites. These sites are inside an estuary, subj ect to increased fluctuations in salinity, water temperature and sedimentation rates (Dawes 1998). Given the ecological, hydrographical and geographical differences fo r The Tampa Bay sites, it is not unexpected those sites are
41 significantly different from the Gulf sites. Possibly, because of these habitat differences, the biology of the sand dollars in Tampa Bay is quite different from populations in the Gulf waters and deserves further investigation. Similar to the particle analysis, the diffe rence in the organic content of sediment and the gut contents could not be statistically compared because of the difference in sample sizes. However, the great differences in values, less than 1% sediment organic content and 20-40% gut organic cont ent, are probably different. The organic content at the Encope site was much higher than the inshore sites. One may expect the inshore sites to have a higher organic conten t, given terrestrial nutrient inputs. However, areas of low and high organic content are patchy in the West Florida Shelf (Brooks et al. 2003). Detrital precipitation, cu rrents carrying allochthonous nutrient sources, lower herbivore densities or in situ production all may affect levels of organic content offshore. These factors pr obably play a role in determining species distributions. Although, the organic content of the sediment does not fluctuate over a diel period, the organic content of the guts does. Higher organic content in the guts may indicate time periods when the sand dollars ar e feeding. Differences in behavior over a diel period have been observed for de posit feeding echinoderms. Hammond (1982) found diel patterns of activity for severa l species of deposit feeding echinoderms ( Holothuria mexicana, Isostichopus badionotus, Meoma ventricosa, and Clypeaster rosaceus) in Jamaica. All four species increased activity in the afternoon leading up to midnight and decreased activity in the early morning, a nd by midday were virtually immobile. Moreover, the behavior that Salsman and Tolbert (1965) documented for
42 Mellita quinquiesperforata off Panama City, Florida is similar to Hammonds study. Lane and Lawrence (1982) found signi ficantly more food in the gut of Mellita quinquiesperforata at night than in the day. Interestingly, the peak times of organi c content of the gut contents were not consistent across collection dates. For Mellita tenuis the peak organic content was 11 am June 2004 and 8 pm September 2004. Encope michelini had peak organic content at 10:30 pm March 2004 and 4:20 pm September 2004. An animal may have a preferred feeding time, for example, around dusk or daw n. A change in peak feeding time could be due to conditions changing with season. Daylight hours, fish activity and predation, and abundance of microalgae all change throughout the year. For example, Buglossidium luteum experiences increased predation seasona lly (Nottage and Perkins 1983). Bay scallops in North Carolina experience greater predation when blue crab populations are highest in the summer (Bishop et al. 2005). Reiss and Kroncke (2005) found predation is a factor in seasonal variabil ity of mean abundances of infaunal organisms in the North Sea. During summertime in the subtropics, the abundance of diatoms, dinoflagellates and foraminiferans is highest (Dawes 1998). Deposit feeding polychaetes increase their ingestion rates when high quality food is available and decrease when low quality food is available (Taghon 1982). An optimal feeding time in the summer may not be the same as in the winter. Sand dollars may take adva ntage of the higher benthic algae populations and conduct their feeding during the shorter summer nighttime periods, when the risk of piscine predation is lower. In addition, during the summer, they can allocate more time to other activities, such as gonad growth (Lane and Lawrence 1979). Conversely, in the winter, when food abundances are lower, sand dollars may allocate more time to foraging
43 in order to meet minimum nutrient requirements. Sand dollars store lipids in their gonads in the limited space between the aboral surf ace and the guts (Moss and Lawrence 1972). Thus, sand dollars cannot tolerant long periods of low food supply, i. e. the winter (Moss and Lawrence (1972). Lane and Lawrence ( 1982) found negative abso rption efficiencies of food during the winter months. They al so noted constant feeding. However, their observations were made in the laboratory. Ma ny marine species are sensitive to chemical cues from predators and may alter behavi or in response (Brown 2003; Lawrence 1991; Rosenberg and Selander 2000; and Solan and Ba ttle 2003). Additionally, differences in photoperiod or light intensity in the labor atory may affect behavior. Since food availability does not vary over the course of the day and peak gut organic content does, the periodicity of feeding in sand dollars most likely is driven by the risk of predation. The organic content of the gut contents of the three sp ecies, by and large, were different from each other. The bioenergetic requirements probably vary with species. In general, metabolic rate scales with power of biomass; thus larger organisms have a lower metabolic rate (Banavar et al. 2002; West et al. 1997). Mellita tenuis the smallest, would most likely have the highest energy re quirements. Yet, the inshore sediment inhabited by M. tenuis has low organic content. Indeed, the gut organic content for Mellita tenuis is the highest of the three species. Encope michelini is larger and would have smaller energy requirements and the organi c content of its gut contents is lower than M. tenuis. Encope aberrans is the largest sand dollar and has the lowest organic content of its gut contents. Ironica lly, the organic content of th e sediment does not correspond with their presumed metabolic rate s. Possibly, the high abundance of Mellita tenuis inshore locally depletes the microalgae populatio n, which otherwise would be quite high.
44 However, populations of M. tenuis persist despite low organic content. Possibly, incoming tides and currents regularly replenish mi croalgae at the inshore sites. Encope aberrans appears to have an adaptive st rategy quite different from the other two species and may furt her explain why it can coexis t with other species of sand dollar. It ingests larger pa rticles and inhabits both inshor e and offshore habitats. Where E. aberrans and Mellita tenuis were found together, the organic content of their gut contents and the particle sizes were significantly different. Thus their niches may not overlap enough for species displacement to occur. Encope michelini and M. tenuis appear to have distin ct habitat preferences; E. aberrans appears more flexible. Whitlach (1980) found that deposit-feedi ng polychaetes living in soft bottom habitats were either habitat generalistsresource sp ecialists or habitat specialis tsresource generalists. Possibly, E. michelini and M. tenuis fall into the latter category while E. aberrans is in the former. Though much work has been done on deposit feeders generally, and polychaetes specifically, it remain s to be tested whether these rules hold true across taxa. Arguably, sand dollars and polychaetes are subj ect to similar ecological and evolutionary pressures of deposit feeding in soft bottom habitats, and may follow similar rules for niche occupation. Furthermore, differences in the organic c ontent of the gut contents in all three species could be related to th e types of food each ingests. Mellita tenuis had visibly more loose organic material present in the guts than Encope michelini. Even though particle sizes were approximately the same for the two species, E. michelini may feed more on epipelic algae and M. tenuis may feed more on interstitial free microalgae, as evidenced by gut contents. This was not tested for in this study, however, it de serves investigation.
45 The porous carbonate particles offshore may provide more sites of attachment for epipelic algae as opposed to th e quartz particles inshore, wh ich are more solid and would resist algae attachment. The quartz particle s examined were mostly translucent or pure white. The opaque carbonate particles are of ten stained green. A lthough particle sizes were similar for the two species, the food material may be quite different. An interesting relationship between organi c content and particle size for sediment and the gut contents was found. Generally, a negative relationship between particle size and organic content exists (Levinton 1989, Whitlach 1981). For the gut contents, Mellita tenuis and Encope michelini had small particles and high organic content and E. aberrans had larger particles and low organic conten t. However, for the sediment, the exact opposite was found. The inshore sediment partic les are smaller, but have a lower organic content and vice versa for the offshore sedime nts. Inshore sediments should have higher primary productivity given the shallow depth and less light attenuati on than in offshore deeper water. Moreover, M. tenuis, with its small size and hi gh organic content of the gut contents, would presumably require an abunda nt food source. The sediments it inhabits have low organic content (<2%). There are several explanations fo r this trend. Though the organic content of the sediment offshore wa s higher, not all of that may be edible or palatable to sand dolla rs, which may preclude M. tenuis from inhabiting offshore sites. Secondly, it is possible that M. tenuis do not feed at the place they were collected from. During the June 2004 collections at Anna Ma ria Island, the density of sand dollars decreased after dusk. A lower density at nigh t may indicate a periodic migration, deep burrowing, or increased difficulty in finding sand dollars at ni ght. However, we typically searched 5-10 cm into the substrate, thus making burrowing behavior unlikely. In June
46 2004, the gut organic content was significantl y higher in the morning and night than midday. Thus, at least in the summer, M. tenuis appears to increase its feeding during the night. Alternatively, the low organic content of the sediment could be the result of sand dollar feeding decreasing the overall organic content in th e sediment at least where they occur. The density of M. tenuis inshore is high rela tive to the offshore E. michelini populations. The guts of M. tenuis had relatively more strands of filamentous algae than E. michelini and E. aberrans indicating that inshore sand dollars can sequester food despite low availability. One way to test if sand dollars alter the sediment organic content would be to compare sediment from adjacent sand dollar and non-sand dollar sites, and to exclude sand dollars from know n habitats and test for increased sediment organic content. In addition to differences in substrate, food types and energetic requirements of Mellita tenuis and Encope michelini differential predation pressure may also be a component in the species distribution of sand dollars. In artificial reefs off Cedar Key, Fl, M. tenuis experiences almost twice the amount of predation by triggerfish than E. michelini (Kurz 1995). Mellita tenuis has a thinner test with relatively looser internal calcareous pillar structures (Seilacher 1979) that is easier to break open (pers. observation). This would make it mo re vulnerable to predation than E. michelini Mellita tenuis may not be able to inhabit offshore sand bottoms where fish predation may be higher than in the subtidal beach zone. Encope aberrans occurring both inshore and offshore, has the largest and thickest test of the three species (Hendler 1995; pers. observation), and may even reach a refuge in size, where few adults are preyed upon because of the thick test. Predation pressure may be important in determining the species
47 distribution. However, more research needs to be done on predation of sand dollars in order to test this. Deposit feeders differ from other herbivores in that they ingest large quantities of substrate, liberate and digest organic material, and egest the clean substrate. The substrate itself presents very little nutriti on other than possible minerals. Rather, the organic material on and amongst the substrat e is sought by the feeder. Polychaetes, bivalves, sea cucumbers and sand dollars all ob tain food this way in the marine benthos. Levinton (1989) states Deposit feeders satisfy their nutritional requirements from the organic fraction of ingested sediment. Th is noncommittal statemen t masks a number of problems and controversies that have occupi ed the efforts of nut ritional biologists, biological oceanographers, and sedimentologist s. At the heart of these problems and controversies are the mechanisms of processing sediment that can have a profound impact on the community. Specifically, where densities of sand dollars are high, such as inshore sand beds with populations of Mellita tenuis the animals may indeed alter the conditions of their habitat. Sediment is c onstantly disturbed, part icles abraded, epipelic and interstitial growth removed, and cleaned substrate made available for colonization (Findlay and White 1983; Reidenauer 1989) Moreover, wave action makes the environment constantly unstable, carrying away and bringing in nutrients. Even at the offshore site, in 18 m depth, deep troughs were observed, probably formed by waves and currents. Sand dollars must contend with th is to survive and reproduce. The sand dollar ekes out an existence in the thin veneer of oxygenated substrate, while being subject to fish predation. Sand dollars plow through the substrate daily and thus allow other organisms to exist in their wake.
48 Conclusions Sediment particles are larger offshore and have a higher organic content than inshore sediment. The offshore and inshore se diment have distinct cumulative frequency curves. Despite geographic differences, ins hore Gulf sediments are similar. Particle sizes in the sand dollar guts are a smaller subset of the ambient sediment. Encope aberrans has larger particles in their guts than Mellita tenuis at offshore Egmont Key were they co-occur. Based on data from food grooves, sand dollars select small particles that are most likely crushed by the lantern. The organic content of the sediment differed from inshore sites 0.05-1.8% to 4.2% offshore. Though particle sizes and or ganic content of the gut contents for Mellita tenuis and Encope michelini are similar, the types of food th ey ingest may be different. Food types available may differ between inshore and offshore sites, and thus may affect food type ingested. The guts of M. tenuis had visibly more filamentous algae than E. michelini Encope aberrans had significantly lower organic content in the gut contents than M. tenuis, both from Egmont Key. Encope aberrans may select different particles, allowing it to co-exist with both E. michelini and M. tenuis at different sites. Differences in peak feeding times change over the course of the year and are not consistent across species. Inverse relationships between particle si ze and organic content exist for sediment and sand dollar gut contents. Sand dollar dens ities offshore are low and may not affect
49 the overall organic content of that habitat. Mellita tenuis may deplete the sediment organic content, due to their much higher density inshore. Thus, M. tenuis may reverse the negative relationship be tween particle size and organic content. In general, more research on deposit -feeding echinoderms needs to be done. Basic information on spawning, recruitment, and adaptive strategies for stress and disturbance is needed for Encope aberrans and Encope michelini Future studies of sand dollar feeding should include exclusion experi ments, long-term mon itoring, and further sediment studies. To determine whether sand dollars deplete resources in their habitats, exclusion experiments could be conducted to test for increased sediment organic content. Sand dollars appear to ingest pa rticles smaller than the average particle size available. Study of the size fraction ingested may pr ovide further insight to their dietary preferences. Moreover, to further estab lish a periodicity of feeding, sampling over 24 hour periods and over the course of several years would be necessary to document full patterns of their behavior. In addition, it appe ars that predation may be the driving force in determining peak feeding times for sand dol lars. Further inves tigation of predatorprey relationships may illuminate more on th e feeding ecology of sand dollars. Finally, echinoids are generally regarded as being stenohaline, however, Mellita tenuis is found in Tampa Bay and grows to a large size. More research on bay sand dollars needs to be conducted in order to understa nd their adaptive strategy to an estuarine habitat.
50 References Banavar, J. R., J. Damuth, A. Maritan, a nd A. Rinaldo. 2002. Supply-demand balance and metabolic rate. Proceedings of the National Academy of Science 99, 1050610509. Bell, B.M. and R. W. Frey. 1969. Observations on ecology and the feeding and burrowing mechanisms of Mellita quinquiesperforata (Leske). Journal of Paleontology 43, 553-560. Bishop, M. J., J. A. Rivera, E. A. Irlandi, W. G. Ambrose, Jr., and C. H. Peterson. 2005. Spatiotemporal patterns in the mortality of bay scallop recruits in North Carolina: Investigation of a lif e history anomaly. Journal of Experiment al Marine Biology and Ecology 315, 127-146. Blott, S. 2000. GRADISTAT v5.0 software. Department of Geology, Royal Holloway University of London, UK. Borzone, C. A., Y. A. G. Tavares, and C. R. Soares. 1997. Morphological adaptation of Mellita quinquiesperforata (Clypeasteroida, Mellitidae) to exploit high hydrodynamics environments. Iberingia 82, 33-42. Brown, G. E. 2003. Learning about danger: chemical alarm cues and local risk assessment in prey fish. Fish and Fisheries 4, 227-234. Calbert, A., E. Saiz, X. Irigoien, M. Alcaraz, and I. Trepat. 1999. Food availability and diel feeding rhythms in the marine copepods Acartia grani and Centropages typicans Journal of Plankton Research 21, 1009-1015. Cowen, R. 1981. Crinoid arms and banana pl antations: an economic harvesting analogy. Paleobiology 7, 332-343. Crisp, D. J. 1971. Holme, N. A. and A. D. McIntyre, eds. Energy Flow Measurements. Methods for the Study of Marine Benthos Blackwell Scientific Publications: Oxford, U. K. Crozier, W. J. 1920. Notes on the bionomics of Mellita American Naturalist 54, 435442.
51 Culver, S. 1961. Observations on the biology of the sand dollar, Mellita quinquiesperforata M. A. Thesis, Duke University, Durham, SC. Dawes, C. J. 1998. Marine Botany, Second Edition John Wiley and Sons, Incorporated: New York, NY. Dernie, K. M., M. J. Kaiser, E. A. Richardson, R. M. Warwick. 2003. Recovery of soft sediment communities and habitats following physical disturbance. Journal of Experimental Marine Biology and Ecology 285-286, 415-434. Findlay, R. H. and D. C. White. 1983. The effects of feeding by the sand dollar Mellita quinquiesperforata (Leske) on the benthic microbial community. Journal of Experimental Marine Biology and Ecology 72, 25-41. Findlay, R.H. and L. Watling. 1998. Seasona l variation in the st ructure of a marine benthic microbial community. Microbial Ecology 36, 23-30. Florida Fish and Wildlife Research Institute (FWRI). 2005. Offshore red tide-associated mortalities and FWRI even t response. Online public ation. Accessed 11/7/2005. www.floridamarine.org/featur es/view_article.asp?id=25276. Florida Institute of Oceanography (FIO). 1994. The West Florida Shelf: Past, Present and Future. Florida Institute of Ocea nography: St. Petersburg, FL. Frazer, T. K., W. J. Lindberg, and G. R. Stanton. 1991. Predation on sand dollars by gray triggerfish, Balistes capriscus in the northeastern Gulf of Mexico. Bulletin of Marine Science 48, 159-164. Ghiold, J. 1979. Spine morphology and its si gnificance in feeding and burrowing in the sand dollar, Mellita quinquiesperforata (Echinodermata: Echinoidea). Bulletin of Marine Science 29, 481-490. Ghiold, J. 1984. Adaptive shifts in clypeas teroid evolution feed ing strategies in the soft-bottom realm. Journal of Geological Paleontology 169, 41-73. Goodbody, I. 1960. The feeding mechanism in the sand dollar Mellita sexiesperforata (Leske). Biological Bulletin 119, 80-86. Google. 2005. Google Earth v3.0. Hammond, L. S. 1982. Patterns of feeding a nd activity in deposit -feeding holothurians and echinoids (Echinodermata) from a shallow back-reef lagoon, Discovery Bay, Jamaica. Bulletin of Marine Science 32, 549-571.
52 Hendler, G., J. E. Miller, D. L. Pawson, and P. M Kier. Sea Stars, Sea Urchins and Allies: Echinoderms of Florida and the Caribbean Smithsonian Institution Press: Washington, DC. Hyman, L. H. 1955. The Invertebrates: Echinodermata McGraw-Hill Book Company, Incorporated: New York, NY. Kaplan, E. H. 1988. A Field Guide to Southeastern and Caribbean Seashores Houghton Mifflin Company: Boston, MA. Kenk, R. 1944. Logical observations on two Puerto-Rican echinoderms, Mellita lata and Astropecten marginatus Biological Bulletin 81, 177-187. Kurz, R. C. 1995. Predator-prey inte ractions between gray triggerfish ( Balistes capriscus Gmelin) and a guild of sand dollars around artificial reefs in the northeastern Gulf of Mexico. Bulletin of Marine Science 56, 150-160. Lane, J. M. 1977. Bioenergetics of the sand dollar, Mellita quinquiesperforata (Leske, 1778). Ph. D. Dissertation, University of South Florida, Tampa, FL. Lane, J. M. and J. Lawrence. 1979. Gona dal growth and gametogenesis in the sand dollar Mellita quinquiesperforata (Leske, 1778). Journal of Experi mental Marine Biology and Ecology 38, 271-285. Lane, J. M., and J. Lawrence. 1980. Seasonal variation in body growth, density and distribution of a populat ion of sand dollars, Mellita quinquiesperforata (Leske). Bulletin of Marine Science 30, 871-882. Lane, J. M. and J. Lawrence. 1982. Food, feeding and absorption efficiencies of the sand dollar, Mellita quinquiesperforata (Leske). Estuarine, Coastal and Shelf Science 14, 421-431. Lawrence, J.M., A.L. Lawrence, and A.C. Giese. 1966. Role of the gut as a nutrientstorage organ in the purple sea urchin ( Strongylocentrotus purpuratus ). Physiol. Zool. 39, 281-290. Lawrence, J. M. 1990. The effect of st ress and disturbance on Echinoderms. Zoological Science 7, 17-28. Lawrence, J. M. 1991. Chemical alarm response in Pycnopodia helianthoides (Echinodermata: Asteroidea). Marine Behavioral Physiology 19, 39-44. Lawrence, J. M., J. Herrera, and J. Cobb. 2004. Vertical posture of the Clypeasteroid sand dollar Encope michelini Journal of Marine Biologi cal Association of the United Kingdom 84, 407-408.
53 Lawrence, J.M., J. Cobb, S. Hilber, and J. Sw igart. Characteristics of populations of the sand dollars Encope michelini and Encope aberrans off the central Florida Gulf coast. Society for Integrative and Comparative Biology Annual Meeting, Orlando, FL. 4-8 January 2006. Levinton, J. S. 1989. Deposit Feeding and Coastal Oceanography. G. Lopez, G. Taghon, and J. Levinton, eds. Ecology of Marine Deposit Feeders Springer-Verlag: New York NY. Loven, S. 1874. tudes sur les chnoides P. A. Norstedt and Sner: Stockholm, Sweden. MacGinitie, G. E. and N. MacGinitie. 1949. Natural History of Marine Animals McGraw-Hill Book Company: New York, NY. Macquart-Moulin, C. 1999. Diel vertic al migration and endogenous swimming rhythms in Asterope mariae (Baird) and Philomedes interpuncta (Baird) (Crustacea Ostracoda Cyprinidae). Journal of Plankton Research 21, 1891-1910. McNulty, J.K., R.C. Work, H. B. Moore. 1962. Level sea bottom communities in Biscayne Bay and neighboring areas. Bulletin of Marine Science of the Gulf and Caribbean 12, 204-233. Merrill, R. J. and E. S. Hobson. 1970. Field observations of Dendraster excentricus, a sand dollar of western North America. The American Midland Naturalist 83, 595624. Mitbarkar, S. and A. L. Anil. 2002. Diatoms of the microphytobenthic community: population structure in a tropi cal intertidal sand flat. Marine Biology 140, 41-57. Mooi, R. 1989. Living and fossil genera of the Clypeasteroida (Echinoidea: Echinodermata): an illustrated key and annotated checklist. Smithsonian Contributions to Zoology 488. Nottage, A.S. and E. J. Perkins. 1983. The biology of solenette, Buglossidium luteum (Risso), in the Solway Firth. Journal of Fish Biology 22, 21-27. ONeill, P. L. 1978. Hydrodynamic anal ysis of feeding in sand dollars. Oecologia 34, 157-174. Paine, R. T. 1971. The measurement and application of the calorie to ecological problems. Annual Review of Ecology and Systematics 2, 145-164. Phelan, T. F. 1972. Comments on the echinoid genus Encope and a new subgenus. Proceedings of the Biological Society of Washington 85, 109-130.
54 Pomory, C. M., B. D. Robbins, and M. T. Lares. 1995. Sediment grain size preference by the sand dollar Mellita tenuis Clark, 1940 (Echinodermata; Echinoidea): a laboratory study. Bulletin of Marine Science 56, 778-783. Reidenauer, J. A. 1989. Sand-dollar Mellita quinquiesperforata (Leske) burrow trails: sites of harpacticoid disturba nce and nematode attraction. Journal of Experimental Marine Biology and Ecology 130, 223-235. Reiss, H. and I. Kroncke. 2005. Seasonal variability of infaunal communities in three areas of the North Sea under diffe rent environmental conditions. Estuarine, Coastal and Shelf Science 65, 253-274. Romer, G. S., and A. MacLachlan. 1986. Mu llet grazing on surf diatom accumulations. Journal of Fish Biology 28, 93-104. Rosenberg, R. and E. Selander. 2000. Al arm signal response in the brittle star Amphiura filiformis Marine Biology 136, 43-48. Salsman, G. G. and W. H. Tolbert. 1965. Observations of the sand dollar, Mellita quinquiesperforata Limnology and Oceanography 10, 152-155. Seilacher, A. 1979. Constructional morphology of sand dollars. Paleobiology 5, 191221. Serafy, D. K. 1979. Memoirs of the Hourglass Cruises, Volume V, part III: Echinoids (Echinodermata: Echinoidea) Florida Department of Natural Resources: St Petersburg, FL. Simard, Y., G. Lacroix, and L. Legendre. 1985. In situ twili ght grazing rhythms of diel vertical migration of a scattering layer of Calanus finmarchicus Limnology and Oceanography 30, 598-606. Smith, A. 1984. Echinoid Palaeobiology George Allen and Unwin: London, Great Britain. Solan, M. and E. J. V. Battle. 2003. Does the ophiuroid Amphiura filiformis alert conspecifics to the danger of predator s through generation of an alarm signal? Journal of the Marine Biologica l Association of the United Kingdom 83, 1117-1118. SPSS Inc. 1989-2004. SPSS v13.0 for window s student version software. Taghon, G. L and P. A. Jumars. 1984. Variable ingestions rate and its role in optimal foraging behavior of marine deposit feeders. Ecology 65, 549-558.
55 Tan, C. and J. M. Lawrence. 2001. Age determination in the sand dollar Mellita tenuis Gulf of Mexico Science 19 43-49. Telford, M. 1981. A hydrodynamic interpretation of sand dollar morphology. Bulletin of Marine Science 31, 605-622. Telford, M., R. Mooi, O. Ellers. 1985. A new model of podial de posit feeding in the sand dollar, Mellita quinquiesperforata (Leske): the sieve hypothesis challenged. Biological Bulletin 169, 431-448. Telford, M. and R. Mooi. 1986. Resource part itioning by sand dollars in carbonate and siliceous sediments: evidence from podial and particle dimensions. Biological Bulletin 171 197-207. Telford, M., R. Mooi, and A. S. Harrold. 1987. Feeding activities of two species of Clypeaster (Echinoides, Clypeasteroida): furthe r evidence of clypeasteroid resource partitioning. Biology Bulletin 172, 324-336. Timko, P. L. 1976. Sand dollar as suspension feeders: a new descri ption of feeding in Dendraster excentricus. Biological Bulletin 151 247-259. Weihe, S. C. and I. E. Gray. 1968. Observations on the biology of the sand dollar Mellita quinquiesperforata (Leske). The Journal of the Elis ha Mitchell Society 315327. West, J. B., J. H. Brown, and B. J. Enquist 1997. A general model for the origin of allometric scaling laws in biology. Science 276, 122-126. Whitlach, R. B. 1980. Patterns of resource utilization and coex istence in marine intertidal deposit-feeding communities. Journal of Marine Research 38, 743-765.