High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems

Home , Aipysurus eydouxii, Carrion, Fordonia, Gerarda prevostiana, Hydrophis elegans, Isopoda

Provided for non-commercial research and educational use. Not for reproduction, distribution or commercial use.

This chapter was originally published in Treatise on Estuarine and Coastal Science, published by Elsevier, and the attached copy is provided by Elsevier for the author's benefit and for the benefit of the author's institution, for non- commercial research and educational use including without limitation use in instruction at your institution, sending it to specific colleagues who you know, and providing a copy to your institution's administrator.

All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution's website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier's permissions site at:

http://www.elsevier.com/locate/permissionusematerial

Davenport J (2011) High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems. In: Wolanski E and McLusky DS (eds.) Treatise on Estuarine and Coastal Science, Vol 6, pp. 227–249. Waltham: Academic Press.

6.09 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems J Davenport, University College Cork, Cork, Republic of Ireland

6.09.1 Introduction/Paleoecology 227 6.09.2 Amphibia 229 6.09.3 Reptiles 230 6.09.3.1 Lizards 230 6.09.3.2 Snakes 233 6.09.3.2.1 Sea snakes 233 6.09.3.2.2 Other snakes 236 6.09.3.3 Crocodilians 236 6.09.3.4 Turtles 237 6.09.3.4.1 Sea turtles 237 6.09.3.4.2 Other turtles 244 6.09.4 General Conclusions 246 References 246

Abstract

This chapter briefly considers the trophic relationships of Mesozoic coastal reptiles (ichthyosaurs, plesiosaurs, mosasaurs, pterosaurs, and placodonts). The few amphibian species that exploit coastal/estuarine food resources are described. Coastal lizards are dealt with in detail, while there is a large section devoted to the trophic specialization and niche separation of sea snakes. The trophic biology of estuarine crocodilians is reviewed. The rest of the chapter is devoted to sea turtles plus a few other chelonian species that exploit estuaries and coastal waters. The consequences of anthropogenic depletion of reptile populations form a major theme, as do inter-ecosystem energetic subsidies.

6.09.1 Introduction/Paleoecology artifact of the bias created by the presence of indigestible belemnite hooklets in the stomachs; cf. squid diets of ele Amphibian ancestors were basal tetrapods, derived from fresh phant seals and sperm whales), coastal forms appear to have water fish, dating to the Devonian (~365 million years ago had a broader diet, some having jaws suitable for crushing (MYA)). Modern amphibians (mainly frogs, toads, newts, and shellfish and others apparently taking fish. Recently, evidence salamanders) are mostly tied to freshwater for reproduction of a truly catholic diet in a widespread coastal ichthyosaur was and there is no fossil evidence of ancestral forms from marine found in the fossilized stomach contents of a specimen of the sediments, indicating an absence of an amphibian role in Cretaceous Platypterigius, which had been eating fish, hatchl Devonian coastal or estuarine ecosystems. ing sea turtles, and a bird (Kear et al., 2003). Given their size Basal tetrapods (formerly known as labrynthodont amphi diversity (and relatively small size at birth), it is virtually bians) gave rise to reptiles in the Carboniferous (320–310 certain that members of the ichthyosaur feeding guild preyed MYA). Reptiles were truly terrestrial, laying shelled eggs and upon each other and there is some fossil evidence to support having relatively impermeable skins. During the Mesozoic this view (e.g., McGowan, 1974). (230–63 MYA), reptiles became extremely important in marine Plesiosaurs (190–65 MYA) were highly successful Mesozoic trophic biology and occupied coastal and estuarine niches that, reptiles, many of which were large and oceanic predators on since major extinctions at the end of the Mesozoic, have sub fish and cephalopods (e.g., belemnites). Although the best- sequently been taken over by a variety of fish and marine known species were long necked with a small head and four mammals. Some groups (notably the ichthyosaurs) were vivi roughly equal-sized flippers, morphological diversity was high, parous and independent of land. with short-necked, large-headed pliosaurs being apex predators Ichthyosaurs (230–90 MYA) formed a highly diverse that competed with ichthyosaurs. Many plesiosaurs were cer group, ranging from 2 to 20 m in overall length and occupied tainly coastal and omnivorous; McHenry et al. (2005) recently both oceanic and coastal ecological niches. Coastal species described the gut contents of elasmosaurs (extremely long- appear to have been smaller and rather less streamlined than necked plesiosaurs) from early Cretaceous deposits of the well-known highly streamlined tuna-like forms with large Australia. They found remains of fish and belemnites – and eyes (e.g., Opthalmosaurus) that probably foraged offshore in also remains of gastropod and bivalve mollusks as well as deep water. Whereas oceanic forms seem to have specialized crinoid echinoderms and crustacean exoskeletons. They noted in feeding on belemnite cephalopods (though this may be an that the elasmosaurs had conical teeth like modern fish and

227 Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 228 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems

Placodus Cyamodus Henodus Figure 1 Placodont diversity. Reproduced with permission from Naish, D., 2004. Fossils explained 48. Placodonts. Geology Today 20, 153–158.

squid eaters, but that the stomachs contained gastroliths (as do there is plentiful evidence that they also ate ammonites and modern crocodilians) which would have allowed hard-shelled nautiloids, besides preying on other mosasaurs, small plesio prey to be triturated, even though the animals had first to be saurs, sea turtles, and seabirds. Mosasaur anatomy was not swallowed whole. compatible with high swimming speeds, so it seems likely Triassic (230–63 MYA) placodonts (see Naish (2004) for that they were ambush, rather than pursuit predators (except review; Figure 1) were modest-sized (1–3 m overall length), when preying on slow-moving groups such as nautiloids). As shallow water durophagous benthic predators that appear to early mosasaurs had highly kinetic skulls (like snakes), they have subsisted predominantly upon bivalve mollusks, crushed were able to ingest relatively large prey items (e.g., large teleost by palatine teeth, although crustaceans and even brachiopods fish). With increasing study, it has become clear that mosasaurs have been suggested as possible food items. Many placodonts were extremely diverse in form, and occupied a variety of were heavily armored and superficially resembled turtles, being trophic niches, exploiting pelagic and benthic food resources. short-legged and often with carapaces (Figure 1). They prob For example, Martin (2007) confirmed that a mosasaur ably foraged predominantly on sandy and muddy bottoms, (Globidens) with rounded crusher teeth had bivalve remains in digging into the substratum with limbs and/or snouts to extract its stomach. Whereas juvenile mosasaurs probably fell prey to a bivalves. However, most placodonts had robust, spatulate teeth range of pterosaurs, birds, and fish, it seems likely that adult at the front of the jaws, so appear to have been capable of mosasaurs were only preyed upon by other mosasaurs, by plucking prey from rocky substrata; again, bivalve mollusks marine crocodiles, or by sharks. Healed bite marks on mosa may have been the most important prey, though gastropods saur skeletons are evidence of shark predation (rather than and even large barnacles are other likely candidates. Placodonts scavenging) (Rothschild et al., 2005). were slow moving and it has been suggested that adults were In recent years, paleontological studies have revealed that preyed upon by sharks or larger coastal marine reptiles such as flying pterosaurs (which arose in the Triassic 225 MYA and crocodilian phytosaurs (Mazin and Pinna, 1993). Certainly disappeared about 65 MYA, but coexisted with birds for 100 juveniles are known to have fallen prey to the small (1 m) million years) predominantly occupied coastal niches, nothosaur (plesiosaur precursor) Lariosaurus (Tschanz, 1989) whereas birds dominated in terrestrial and inland habitats and, by analogy with bird predation, it has been suggested that (Wang et al., 2005). Pterosaurs formed a highly diverse pterosaurs may have preyed upon them (Mazin and Pinna, group anatomically and ecologically; old ideas of weak, lim 1993). It seems probable that placodonts were egg layers like ited fliers that were nearly helpless on land have been replaced sea turtles, and that eggs and hatchlings would therefore have by concepts of powerful, agile flying and running animals provided a marine energetic subsidy for coastal terrestrial (probably endothermic) that filled ecological niches currently predators. occupied by albatrosses, pelicans, gulls, waders, etc. It is evi Mosasaurs made up the last great group of marine reptiles of dent that many were capable of landing on, and taking off the Mesozoic, largely replacing the extinct ichthyosaurs and the from water, just like modern seabirds. It can be speculated much-reduced plesiosaurs in the late Cretaceous (99–65 MYA). that smaller species were predominantly near coastal in their They were marine lizards and many genera were giants foraging, while gigantic forms may have foraged over ranges (<14–18 m length). Although early mosasaurs possibly laid comparable with albatrosses. There is plentiful fossil evidence their eggs on land, it is known that later, highly adapted, of fish capture, but some jaw/tooth structures indicate the forms were viviparous. Similar to most Mesozoic marine likelihood of foraging on a wider variety of marine organisms reptiles (except turtles), they primarily ate fish and squid, but including mollusks, crabs, and zooplankton (see Chatterjee

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems 229 and Templin (2004) for review). It has recently been con brackish conditions. Coastal toads include the Natterjack and firmed that pterosaurs laid eggs, probably with thin, leathery Green toads (Bufo calamita and Bufo viridis) in Europe and the shells (Ji et al., 2004); hence, it is likely that these provided an Western toad (Bufo boreas) in North America. All can breed in energetic subsidy for any terrestrial animals that preyed upon slightly brackish water and live close to the sea (the Western them. Many pterosaurs were probably preyed upon by large toad can swim in seawater for hours and inhabits many islands fish and a variety of marine reptiles (just as modern coastal off western USA and Canada); however, little evidence is avail birds are taken by marine mammals, sharks, and crocodiles), able to confirm that they include significant amounts of but direct evidence of this appears to be lacking. marine-related food items in their invertebrate diets. Some Chelonians (turtles) date from around 200 MYA and it is frogs live in similar habitats (e.g., the marsh frog Rana ridibunda generally believed that ancestral forms were freshwater/marsh that lives in coastal marshlands in much of Europe). There are dwellers. Sea turtles arose repeatedly from such groups from also occasional reports of frogs foraging on invertebrates of 150 MYA onward, apparently responding to the richer food beach strandlines. For example, Davenport et al. (1995) supply available in coastal waters (see Bels et al. (2008) for found frogs (species unknown) feeding on strandline organ review). They thrived until the last few centuries during which isms at the brackish Laguna San Rafael in Chile, albeit under their numbers have been severely reduced by human exploita conditions of heavy rainfall. tion. Their breeding strategy and trophic biology appears to In only two species have the diets of coastal/estuarine have been similar to living species, so will not be discussed amphibians been studied in any detail. The first of these is further in this section. undoubtedly a fully estuarine anuran, the crab-eating frog, Mesozoic crocodilians were far more widespread and much Fejervarya (=Rana) cancrivora (Figure 2) that inhabits SE more marine than their living relatives. Many were large oceanic Asian coastal/estuarine mangrove communities of Vietnam competitors for plesiosaurs, ichthyosaurs, and mosasaurs. and Thailand (as well as neighboring freshwater systems). However, unlike those groups, they were tied to land for repro F. cancrivora adults tolerate salinities <18 for long periods, duction and so certainly affected coastal systems, even at the level while tadpoles are found in pools and ditches containing of simply providing food (eggs) for shoreline predators. There water <35. Although F. cancrivora is insectivorous in freshwater were also definitely coastal specialists. Among marine crocodi systems, the brackish mangrove populations subsist extensively lians were relatives of the modern gharial (Gavialis gangeticus), a on sesarmid crabs (Elliott and Karunakaran, 1974). There highly specialized fish eater with long slender jaws. Whereas the appear to be no data concerning the diet of tadpoles. Equally, sole living species is entirely freshwater in distribution, many of there are no reports of the frogs’ natural predators, though these the earlier species were coastal, notably the giant Pliocene probably include birds, monitors, and mangrove snakes. (5.5–1.8 MYA) Ramphosuchus crassidens that was 15–18 m long Crab-eating frogs, like many other ranids, are harvested in and presumably fed on shoaling marine teleost fish. substantial numbers for human food; they are extensively Snakes fossilize poorly, so little is known about their role in traded in SE Asia. ancient oceans; all living marine species are of relatively recent The second example is intriguing: the highly aquatic African origin (<25 MYA). A hypothesis that all snakes arose in the sea clawed frog Xenopus laevis is native to sub-Saharan Africa and from a common snake/mosasaur varanoid ancestor has was distributed originally only in freshwater habitats at alti recently been discredited to some extent on molecular grounds, tude. However, due to its use in human pregnancy testing in the because of a lack of a close relationship between living vara 1940s and 1950s, it has been introduced to habitats (some of noids (varanid monitor lizards) and snakes (Vidal and Hedges, them estuarine) in many countries and is euryhaline, tolerating 2004). However, there were Upper Cretaceous (146–65 MYA) sustained salinities of 12–15 (probably because it can cope snakes that fossilized in coastal, shallow marine sediments with dehydration during estivation in its natural habitat). (e.g., Tchernov et al., 2000). Some of these (e.g., Haasiophis) Lafferty and Page (1997) studied the gut contents of clawed had well-developed hind limbs, and cladistic analysis suggests frogs living in the Santa Clara estuary, southern California affinities with living boas and pythons. Nothing definite is (salinities ranging between 3 and 37). They confirmed that known of their diet, but it is probable that they preyed upon the frogs ate several fish species (including the endangered the contemporary diverse teleost fish community, as do their tidewater goby, Eucyclogobius newberryi), amphipods, beetles, modern relatives (see below). and hemipterans. They also observed predation on clawed frogs and their tadpoles by typical estuarine avian predators (herons, egrets, and gulls). 6.09.2 Amphibia

Few amphibians are capable of tolerating brackish water; because of osmotic/ionic problems none are fully marine. Living amphibians are made up almost entirely of the Urodela (newts and salamanders) and the Anura (frogs and toads). Virtually all of these animals have some ability to exchange gases across the skin, which consequently has to be very permeable to water, making osmotic dehydration a severe risk. Eggs and larvae are also rarely capable of development under saline conditions. However, there are a number of amphibian species that live Figure 2 Fejervarya (=Rana) cancrivora (crab-eating frog) from in coastal/estuarine areas where they encounter tolerable Singapore. Photograph: Colleen Goh.

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 230 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems

6.09.3 Reptiles 6.09.3.1 Lizards Many lizards, particularly island species, exploit the high pro ductivity of neighboring coastal marine ecosystems to supply their energy needs either wholly or partially. Coastal terrestrial habitats (especially those of subtropical or tropical islands) are often energy poor; therefore, this type of exploitation repre sents a form of marine energetic subsidy (see Barrett et al. (2005) for discussion). Most of such lizards forage intertidally and do not enter the water to catch prey. A high proportion are primarily omnivores or insectivores, taking prey such as strand line dipteran flies or talitrid amphipods that differ little in characteristics and behavior from terrestrial prey items (e.g., the phrynosomatid lizard Uta stansburiana from Mexico Figure 4 Bermudian skink (Eumeces longirostris). Photo: Mike and California; Galina-Tessaro et al., 1997; Barrett et al., 2005). Pienkowski. Stable isotope studies on U. stansburiana demonstrated a sea weed/algal detritus–insect–lizard food chain (Barrett et al., 2005); similar food chains have been demonstrated in Peru scavenge marine material (e.g., fragments of squid and fish) (Catenazzi and Donnelly, 2007). from the burrows of nesting marine tropic birds (Phaethon It is probable that many more lizard species forage on lepturus) and cahows (Pterodroma cahow), as well as feeding intertidal material than is currently recognized. The volcanic upon broken eggs and dead chicks (see Davenport et al. Atlantic island of Madeira is inhabited by an endemic poly (2001) for review). The birds nest in burrows, usually a few morphic, omnivorous lacertid lizard species (Podarces dugesii) meters above high water; they effectively subsidize the lizards that is distributed throughout the island, feeding on insects, during the breeding seasons of each party to the trophic rela nectar, grapes, and other fruit. Davenport and Dellinger (1995) tionships. Historically (i.e., before human settlement on discovered a melanic morph (Figure 3) that forages on the dark Bermuda), the major predator on skinks was probably the gray pebble beaches of the island, mainly eating the large extinct night heron (Nyctanassa carcinocatactes), a heavily built intertidal isopod Ligia italica, which itself feeds on intertidal night heron that fed primarily on land crabs (Olson and algae and detritus. More recently, demonstration of predation Wingate, 2006). Nowadays, the lizards are preyed upon by a on Ligia occidentalis by U. tumidarostra in Baja, California suite of introduced lizards, mammals, and birds (Davenport (Hazard et al., 1998) and upon L. italica by Podarces atrata on et al., 2001). the Columbretes archipelago of the Mediterranean (Castilla The best-studied island lizard does enter the sea to forage. et al., 2008) reinforces the likely geographically widespread The endemic Galápagos marine iguana Amblyrhynchus cristatus nature of such foraging. It should be noted that all of these (Figure 5) is a far larger animal than the lizards considered so lizards are opportunistic foragers that also forage in the nearby terrestrial habitat (e.g., P. dugesii also eats insects, pollen, and fruit debris discarded by tourists; Davenport, personal observation). Island lizards may not even need to enter the intertidal zone to gain access to marine food. Juvenile, subadult and adult Bermudian skinks (Eumeces longirostris)(Figure 4) forage pri marily on insects (hatchlings are strict insectivores), but also

Figure 3 Madeiran lizard (melanic intertidal specimen)Podarces dugesii. Figure 5 Large male marine iguana (Amblyrhynchus cristatus). Photo: Photograph: Thomas Dellinger. Martin Wikelski.

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems 231 far, and lives extensively within the rocky intertidal zone of all by maintaining higher body temperatures than adults and islands of the Galápagos archipelago, though iguanas have to passing relatively greater quantities of food through the gut at move up to 300 m inland to find sand or volcanic ash in which relatively higher speeds; yearlings take about 6 days for food to to lay their eggs, and they commonly leave the shore entirely at pass through the gut, whereas adults take around 12 days night or when the tide is high. Unlike all other living lizards, (Wikelski et al., 1993). Amblyrhynchus has a wholly marine diet. The iguanas are almost The influence of El Niño on iguana nutrition has attracted completely herbivorous, although Carpenter (1966) reported much attention. Briefly, El Niño events (roughly every 3–4 that they would sometimes take crustaceans (they share years, but of variable intensity) occur when the normal trade foraging areas with the omnivorous sally light foot crab, winds that blow water westward across the Pacific slacken Grapsus grapsus)ors ealp lacentas,w hileWi kelski et al. (1993) (Figure 6). This results in the sea around the Galápagos found that they occasionally ate feces of other marine iguanas, or Islands becoming warmer and containing far less nutrients of sea lions. Iguanas eat intertidal and sub-tidal algae (mainly a (because upwelling is suppressed). In addition, the sea surface wide variety of red and green algal species) that form turfs on the rises by about 0.5 m and wave action increases. In consequence, lower portions of shores of the Galápagos Islands. Detailed less intertidal zone is available (spatially and temporally) for study has revealed that marine iguanas have a remarkably wide foraging and the lack of nutrients in the seawater results in a range of maximum body sizes (1–12 kg body mass; Laurie, much-reduced macroalgal growth. Vinueza et al. (2006) stu 1989) that impacts on foraging strategies. The wide size range died marine iguanas, sally lightfoot crabs, and fish (all of iguanas is related to markedly different food resources on macroalgal grazers) at Santa Cruz island for several years during different islands; those islands to the west of the archipelago and after the strong 1997–98 El Niño event. They found that are supplied with nutrient-rich water from the cold southern grazing pressure substantially cropped back the reduced Humboldt Current and Cromwell upwelling; they feature lush amount of macroalgal turf available during the event itself; algal turfs. Other islands (e.g., Genovesa) are more influenced by iguana mortality was high (Steinfartz et al. (2007) reported the warmer waters of the intertropical convergence zone that are values as high as 90% for some islands) and crabs’ densities poorer in nutrients, with correspondingly sparse algal resources reduced. There were associated changes in macroalgal commu (Wikelski et al., 1997). Marine iguanas show no territoriality nity composition and diversity. Wikelski et al. (1997) had over food resources and are ‘scrambling’ (equal access) rather earlier studied the consequences of El Niño events for marine than ‘contest’ (unequal access) foragers. iguanas and found that, on a given island (each island having Although broadcast natural history films have highlighted different macroalgal resources and thermal conditions), igua the remarkable ability of marine iguanas to dive and swim to nas larger than a threshold size (which varied between islands feed at depths of 3–5 m, only relatively large animals can afford and between strengths of El Niño events) lost body mass, such long-duration dives as the waters around the Galápagos whereas smaller iguanas actually increased in body mass. are cold and small animals inevitably cool more quickly than Larger iguanas therefore suffer higher mortality than smaller large ones. In consequence, young and/or small iguanas are ones when food is short. Marine iguanas are remarkably plastic limited mainly to intertidal foraging, and rarely for more than animals; apart from the phenotypic variability of body mass, − 2hd 1, as Pacific tides are semidiurnal and algal turfs are only Amblyrhynchus is even capable of shrinking to reduce energy found at shore levels where they are covered on every tide demand during El Niño years (Figure 7). Individuals studied (i.e., below low-water neap-tide level). It is also noteworthy over a number of years showed length reductions <20% during that even hatchling animals (50–70 g) are almost completely 2-year periods of poor algal supply, apparently because of bone herbivorous, only gaining a little crustacean material inciden reabsorption as well as reduced quantities of cartilage and tally with their macroalgal diet (Wikelski et al., 1993), and connective tissue. It has been speculated that this may result there is no sign of the ontogenetic dietary shifts characteristic from the prolonged inactivity of marine iguanas during severe of many reptile species in which early life history stages eat El Niño years, leading to enhanced cortisol levels (Wikelski and higher protein diets (e.g., green turtle (Chelonia mydas) hatchl Thom, 2000). The net result is that the shrunken iguana has a ings are carnivorous, while adults are predominantly lower metabolic rate, and hence requires less food. Large igua herbivores). Feeding behavior in marine iguanas is complex nas that shrink most survive longest (Figure 7). The shrinkage as it is only diurnal, but obviously affected by the tide and reverses in years when food is plentiful. weather – the iguanas have to bask to raise their body tempera It is usually expected that shortage of food will result in a tures in between bouts of foraging on the shore or sub-tidally. broadening of dietary niche width; however, in general, this Large iguanas that forage sub-tidally gain two advantages: first, appears not to apply to marine iguanas, which can starve to the sub-tidal algae are more abundant and second sub-tidal death when sufficient macroalgal resources are unavailable, foraging is possible independent of tidal state. Some islands are even though an abundance of terrestrial vegetation is present characterized by relatively small iguanas (e.g., Santa Fe; inland. However, one subpopulation of Amblyrhynchus (on the maximum size is ~2.5 kg) and for these lizards, sub-tidal fora west coast of Seymour Norte island) does use an alternative ging is limited and only supplements an essentially intertidal food source – the succulent saltwort Batis maritima, which diet. On islands where much larger animals are found grows on beaches and around lagoons in the Galápagos, and (e.g., Fernandina), big iguanas feed almost entirely sub-tidally. so is very close to the sea. These Seymour Norte iguanas routi Amblyrhynchus shows no ontogenetic change in diet or nely eat saltwort, especially large animals, which can reach assimilation efficiency (at 70% a high value for a herbivore), double the size of other iguanas on the same island (Wikelski yet young iguanas require proportionally more food because and Wrege, 2000), despite the relatively poor food quality of they have twice the mass-specific energy consumption of adult the succulent (roughly half that of macroalgae); they forage animals (and also have to achieve growth). They manage this over long periods including those when macroalgae are

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 232 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems

(a)

Strong trade winds Inclined sea surface Entrained water flow

South America Warm water Thermocline accumulation Strong upwelling Galapagos

(b)

Trade winds fail Warm water flow

Southeast Asia

Weak upwelling Galapagos

Figure 6 Diagrammatic representation of El Niño phenomenon in relation to Galápagos: (a) Normal conditions and (b) El Niño event.

Male Santa Fe 75 Female Male Genovesa 50 Female 25

−25 Change in SVL (mm) −50 150 200 250 300 350 400 SVL (before 98/99 ENSO; mm)

97 n = 258 305 47

97 Survived until year (N) Survived until year

94 >0 0 to −20 −21 to −60 Change in body length (mm) Figure 7 Changes in body length and survival of marine iguanas (Amblyrhynchus cristatus) in response to the 1998/99 ENSO (El Niño) event. (Above) Reduction in snout-vent length (SVL) of male and female iguanas from two Galápagos Islands. Males shrank less at a given body length than females; Genovesa iguanas were smaller and shrank less than Santa Fe animals. (Below) Relationship between the amount of shrinkage and survival in Santa Fe adult iguanas (both sexes). Large (>300 mm SVL) adult animals that shrank most, lived longest. Reproduced with permission from Wikelski, M., Thom, C., 2000. Marine iguanas shrink to survive El Niño. Nature 403, 37–38.

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems 233 inaccessible at high tide. This food supplementation has appar Both species eat crabs (their most common marine prey), ently allowed large iguanas to survive several El Niño episodes; mollusks, fish, sea turtle eggs, and hatchlings. De Lisle (2007) however, ironically, during the 1997–98 event, high tides sub studied the behavior of V. salvator in a brackish lagoon in North merged the saltwort and killed it – subsequently, all of the large Sulawesi. He noted the controversy over whether monitors iguanas of the subpopulation died. could or could not catch live fish, reporting that the monitors It is currently predicted that global climate change will he observed certainly caught large fish (probably mullet), and result in more frequent and stronger El Niño years and that possibly caught small fish in dense aggregations. V. indicus is this might adversely affect marine iguana populations. known to scavenge the nests of turtles on coastal beaches, However, it should perhaps be borne in mind that the including the endangered leatherback sea turtle (Read and Galápagos climate and hydrography have changed consider Moseby, 2006). ably in the past, even within historical times; Sachs et al. (2009) There is little published material about predators of the two have recently presented evidence that, during the Little Ice Age monitor species. It is likely that most predatory pressure is on (AD 1400–1850), the intertropical convergence zone (featur the egg and hatchling stages, which will be vulnerable to large ing warm and wet conditions) coincided with the Galápagos fish, birds, crocodiles, and other monitors. Both monitor spe archipelago (it is now around 500 km further north). These cies are exploited by humans, for food and their skins (about 1 environmental conditions would likely have reduced available million V. salvator skins were traded annually in the mid-1990s; crops of intertidal macroalgae. Marine iguanas have high phe Shine et al., 1996). notypic plasticity and appear to be able to survive repeated population crashes, usually with minimal genetic restriction 6.09.3.2 Snakes (Steinfartz et al., 2007). Predation pressure on marine iguanas comes from two 6.09.3.2.1 Sea snakes sources, the sea and the land. There are no natural terrestrial Sea snakes are by far the most speciose of living coastal and predators other than birds, but hatchlings in particular are estuarine reptiles (see Rasmussen (2001) for review). More taken by the Galápagos hawk (Buteo galapagoensis) and (less than 60 species have been described, but their taxonomy is commonly) the Galápagos subspecies of the short-eared owl unstable and controversial. Although sometimes placed within (Asio flammeus). However, predation pressure is believed gen a single family (Hydrophiidae), it is believed that they have erally to be low and populations are accepted to be primarily separately evolved from the terrestrial snake family Elaphidae regulated by food availability, with little mortality during non- (which includes cobras, kraits, and mambas) on at least three El Niño years. Less information is available for marine preda occasions. Commonly, the amphibious sea kraits (Laticauda tors, although the Galápagos shark (Carcharhinus galapagensis) spp.) are placed in a subfamily (Laticaudinae) separate from certainly takes marine iguanas as well as sea lions, while crabs the ‘true sea snakes’ (subfamily Hydrophiinae), which are are known to take hatchlings (Roy, 2000). Unusually for entirely aquatic, all viviparous, and therefore form the only lizards, marine iguanas also provide food for predators by living group of reptiles that has no terrestrial life history stage. association. Galápagos mocking birds (Nesomimus parvulus) Sea snakes occur throughout the tropical and subtropical and Darwin’s finches (Geospiza fuliginosa) have both been Pacific and Indian oceans including many islands, and appear reported to have cleaning/feeding relationships with marine to be constrained by the 18 °C isotherm; this thermal limita iguanas, in which iguanas adopt specific postures to initiate tion also limits them to the top 20–50 m of the water column. and facilitate collection of ticks by the birds (Carpenter, 1966). Voris (1972) showed that the center of biodiversity for the Monitor lizards are a group of varanoid lizards with ancient group is in the Straits of Malacca, where 27 species coexist; origins that appear to have evolutionary links with both snakes biodiversity declines westward and eastward. Sea snakes are and mosasaurs (Molnar, 2004). Monitors are carnivorous top absent from the Atlantic apparently because they evolved in predators and carrion eaters in many ecosystems and include a the Indo-Pacific relatively recently (<25 MYA) after the Panama number of very large lizards, including the Komodo dragon isthmus had closed; the seas off the tip of South Africa are too (Varanus komodoensis). Living monitor lizards are widespread cold for them to move around the Cape of Good Hope. They from Africa to the Pacific Islands. Fossil monitors commonly are highly venomous predators; the venom of P. platurus is 5 had coastal distributions (Molnar, 2004). There are two living times as toxic as cobra venom (Priede, 1990). Almost all species ‘species’ (actually species complexes) that are common in are coastal foragers including around islands, in estuaries, and estuarine and coastal habitats and eat marine prey; they are even into freshwater. However, Pelamis is fully oceanic across able to do so because they have nasal salt glands that allow the entire group’s range from Africa to Panama; it specializes in them to excrete salt loads (Dunson, 1974). These are the water feeding at frontal drift lines, whether these are coastal or far out monitor Varanus salvator, distributed from Sri Lanka through SE at sea. Sea snakes are almost helpless on land (with the excep Asia to Indonesia, and the mangrove monitor Varanus indicus, tion of the sea krait genus Laticauda which has ventral scutes which lives in Australia, New Guinea, and many Pacific islands that permit terrestrial locomotion and which lay eggs in shal including the Marianas. The former species is very large, reach low (fresh) muddy water within tidal caves; Tu et al., 1990). ing 3 m and 90 kg; the latter is much smaller (maximum length Sea snakes are relatively well studied in terms of trophic 1.2 m). Both ‘species’ are complexes of many species and sub biology, particularly in the case of sea kraits that return to land species and they occur in damp forests and freshwater habitats to digest meals as well as to slough their skins and reproduce, as well as being common in mangrove-fringed estuaries and because they are not subject to conservation legislation, so lagoons. They are powerful swimmers and can swim for long capture for stomach analysis has been simple (some species periods; they may well have been distributed so widely by readily regurgitate food, allowing nonlethal study too). virtue of being swept out to sea by tsunami backwashes. However, because they spend about 90% of their life

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 234 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems submerged and are inconspicuous despite relative large adult that sea kraits (and other sea snakes) are sometimes killed by size (1–3 m), it is difficult to assess their importance in coastal/ their potential prey. estuarine food webs. They are common nearshore predators in Most sea snakes are piscivorous ambush predators, using Indo-Pacific regions (some estimates suggest that they are the highly neurotoxic venom to subdue their prey, but one species, most abundant reptiles on the Earth; a shoal of Astrotia stokesii the olive sea snake Aipysurus laevis (common on reefs of western, 100 km long and 3.5 m wide was seen at the surface in the northern, and eastern Australia) is an opportunistic generalist Malacca Strait in 1932 (Priede, 1990) and must have contained benthic carnivore that takes mollusks and crustaceans as well as millions of snakes). Ineich et al. (2007) studied sea kraits fish, whereas sea kraits actively seek out their eel prey, pursuing (Laticauda spp.)and they found a population of at least them into burrows and crevices. It is likely that extremely toxic 1400 sea kraits with a biomass of about 350 kg around Signal venom is needed to kill prey items virtually instantaneously so Island (New Caledonia) – a very small, elliptical, 6-ha flat that they do not escape; sea snakes cannot afford to wait for island only 950 m � 1450 m in size, and on which they their prey to die and track it while it does, as is the case for most sampled only 450 m of coastline. They provided (limited) terrestrial venomous snakes. Interestingly, the marbled sea evidence that the Signal Island sea kraits foraged over no snake Aipysurus eydouxi, which only eats fish eggs (Voris and more than 130 ha and swam no more than 1.6 km (sea kraits Voris, 1983), has a venom that is 50–100 times less toxic than are highly philopatric; Shetty and Shine, 2002a). They showed other species of the genus; the species also has reduced fangs, that the sea kraits took around 37 000 fish (biomass 972 kg) much smaller (but still functional) venom glands, and has per year. These observations, if reflecting generalities of limited mouth gape as well as being a much smaller species distribution and trophic impact, imply that sea snakes are overall than its congeneric relatives (see Li et al. (2005) for important top predators in coastal tropical systems, especially review). coral ecosystems, and are therefore likely to have significant There is a great deal of trophic niche separation among sea top-down regulatory effects on lower trophic levels (cf. Finke snake communities as first revealed by seminal studies more and Denno, 2004) as they prey on other strict predators as well than 30 years ago (e.g., Voris, 1972; McCosker, 1975). A parti as herbivores (see below). Brischoux et al. (2007) have subse cularly important paper is that of Voris and Voris (1983), who quently published distinctly different foraging ranges for the investigated the stomach contents of more than 1000 stomachs Signal Island sea kraits. They relied on the fact that sea kraits of 39 species. They found that teleost fish overwhelmingly appear to swim back to shore as soon as they have eaten a large dominate sea snake diet, but that only about a third of the meal (because swimming capability is compromised (Shine teleost families of the Indo-Pacific region are represented in and Shetty, 2001) and sea kraits prefer to digest prey on snake stomach contents, indicating a high degree of selectivity. shore). By assessing the rates of digestion, establishing the Snakes usually have to swallow prey whole. Although snakes state of digestion of regurgitated prey, and from known swim are able to swallow relatively large prey because of their mobile ming speeds, they calculated that about one-third of foraging skulls, unfused mandible tips, and elastic buccal connective trips were within the small range postulated by Ineich et al. tissues, the size and morphology of the snake head and neck (2007), the rest of foraging trips had a mean radius of 14 km are still linked to the characteristics of prey morphology. (Laticauda laticauda) and 21 km (L. saintgironsi), implying very Figure 8 (from Voris and Voris (1983)) shows the relative large foraging areas indeed, not only on coral reef habitats, but importance of prey of different morphologies to overall sea also on lagoon soft bottoms. Unfortunately, it is currently snake diet. Elongate (eel-like) fish are the most common impossible to assess eel densities, so the relative impact of the prey, to the extent that Ineich et al. (2007) recently used sto sea snakes upon them cannot be assessed. Sea kraits take very mach contents’ analysis of two sea krait species (L. laticauda and large meals that take several days to digest, so it is presumably L. saintgironsi) that specialize almost entirely (99.6%) on eels, worthwhile to swim the longer distances. To further illustrate to demonstrate that coral reef eel species diversity was much the difficulties of quantification of trophic interactions, the greater in the study area than previously realized – snakes are Signal Island research group has recently readdressed the size much better at collecting eels than humans! However, it is not of the local sea krait population in the light of enhanced mark– the case that all sea snakes specialize on eels; Lobo et al. (2005) recapture rates (Brischoux and Bonnet, 2008); their latest demonstrated that the hydrophiinid Lapemis curtus was a gen estimate is of 4087 snakes with a total mass of >660 kg. They eralist feeder that concentrated upon clupeids (herring group) also note that there are substantial populations of sea snakes and cynoglossids (flatfish), which occupy different positions in associated with other islands within the New Caledonia the water column, being pelagic and benthic, respectively. lagoon, so that the entire lagoon is effectively an important In the Indo-Pacific area, there is much overlap in the dis sea krait foraging area where tens of thousands of large eels tribution of sea snake species, but Voris and Voris (1983) must be taken every year. Bonnet et al. (2010) have recently demonstrated great feeding niche specialization among an shown that the relationship between sea kraits and their eight species’ matrix, with diet overlap values (Schoener anguilliform prey is more complex than previously realized. index) being low (0–31%). At a very local level, between They found that many sea kraits had scars that reflect serious pairs of closely related Hydrophis species, diet overlap values bites by prey such as moray eels. They present photographic were even lower (0–20%). Such local niche separation is also evidence to show that prey can bite the snake predators, even characteristic of the sea krait lineage; Brischoux et al. (2007) when being swallowed (as kraits sometimes ingest prey demonstrated low levels of diet overlap in two Laticauda species tail-first). They point out that this has energetic implications in New Caledonia. for the snakes (in that there are tissue repair costs), but also that Voris (1972) demonstrated links between sea snake size and the trophic relationship has considerable inherent risk, morphology on food selection; so, it is not surprising that there despite the powerful venom of sea kraits. It seems probable are commonly ontogenetic changes in diet, with young sea

Number of snake species Number of snake 5

0 5 4 1 8 11 16 6 2 1 2 1 2 3 4 5 6 78910 11 12 Fish shape categories Figure 8 Trophic niche separation in sea snakes. Shape categories 1–10 represent fish morphologies; shape category 11 represents fish eggs; and category 12 represents invertebrates. Shape categories 1–4 show a tendency toward increasing diameter and decreasing length; categories 5–8 show tendencies toward lateral or dorso-ventral flattening; categories 9 and 10 represent fish that can inflate themselves or have spines. Numbers within histogram bars indicate numbers of fish families eaten (56 in total). Reproduced with permission from Voris, H.K., Voris, H.H., 1983. Feeding strategies in marine snakes: an analysis of evolutionary, morphological, behavioral and ecological relationships. American Zoologist 23, 411–425. snakes taking different fish species from adults. However, there whereas lemon sharks (Negaprion acutidens) and milk sharks can even be sexual divergence in diet; Shetty and Shine (2002b) (Rhizoprionodon acutus) apparently take none at all (White found that there was significant sexual dimorphism between et al., 2004). However, Wirsing et al. (2007) have convin male and female yellow-lipped sea kraits (Laticauda colubrina), cingly argued that tiger sharks are primarily drawn to with the females being much larger than males. Not only were western Australian waters by (1) elevated temperatures and females longer, but their heads were proportionately larger as (2) the availability of dugongs (Dugong dugon)asp rey,s o the well. Whereas adult female snakes preyed predominantly on sea snake specialization is probably secondary. Tiger shark large conger eels, male snakes took mainly smaller moray eels. predation on sea snakes is not unique to Australian waters. Interestingly, juveniles of both sexes took small moray eels, but Masunaga et al. (2007) found that tiger sharks ate both hydro the females switched to conger eels as they grew larger. Because phiid sea snakes and sea kraits in southern Japanese waters conger eels live in rather deeper water than the shallow, reef- (Yaeyama Islands); again, sea snake predation by other shark dwelling morays, this results in progressive spatial/ecological species was unusual. separation of the sexes. Simpfendorfer et al. (2001) speculated that sea snakes (as Rather less is known about the predators of sea snakes. well as turtles and dugongs) might be vulnerable to tiger sharks These include sea eagles (Haliastur indus and Haliastur leucoga because of their requirement for breathing air, and hence need ster) from above (Heatwole, 1999), and sharks, large predatory ing to swim up in the water column. Kerford et al. (2008) have teleost fish, eels, and perhaps even crocodiles from below. Only recently investigated the predatory relationship between tiger the surface-dwelling yellow-bellied sea snake (P. platurus)is sharks and the bar-bellied sea snake (Hydrophis elegans)ina believed to be unpalatable, as there are no reports of sharks, pristine habitat (Shark Bay) with high shark and snake popula fish, or predatory birds eating them. White-bellied sea eagles tions. The snakes had access to two habitats, seagrass meadows (H. leucogaster) are common avian predators, as their coastal (refuge area with little available food) and exposed sand areas distribution overlaps sea snake distribution to a considerable (where snake eel prey (Ophichthidae) was abundant). They extent. They catch sea snakes at the surface with their talons, demonstrated that, at low tide, the sharks could not access usually within 1 km of the coast. these habitats readily (water depth ~1.3 m) and the snakes Shark predation appears to be particularly important; 21% foraged widely upon the sandy areas, but at high tide, when of tiger sharks (Galeocerdo cuvier, a large and widespread spe both habitats could be reached by sharks, the snakes retreated cies) in western Australian waters (Shark Bay) had sea snakes to the seagrass refuges where they are cryptic, trading off safety in their stomachs, while sea snakes were the third most com against food availability. This illustrates the potential complex mon prey (after teleosts and turtles) overall (Simpfendorfer ity of interaction of abiotic and biotic influences on sea snake et al., 2001). In the case of small tiger sharks, sea snakes were trophic biology. actually the most preferred prey (occurring in 50% of sto Sea snakes are heavily exploited by humans throughout the machs). Tiger sharks may well specialize on sea snakes; the Indo-Pacific (including Australia and Japan) despite their veno nervous shark (Carcharhinus cautus) found in the same waters mous nature. They have been deliberately taken for their meat only occasionally eats sea snakes (1.7% of stomach contents), and skin on an industrial scale in the Philippines for at least 80

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 236 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems years and their organs are used in Chinese medicine. The skin is estuarine crocodile Crocodylus porosus appears to have a substan particularly valued as it is small scaled and does not (apart tial estuarine trophic role at present. Although adult estuarine from in Laticauda spp.) possess the large ventral scutes of crocodiles are common in estuaries and mangrove swamps, and terrestrial snakes. Exploitation is largely uncontrolled (save in are quite often seen in coastal water (probably on passage Queensland, Australia), and it is thought that overfishing is between estuaries: there is no indication that they feed at sea), widespread (Rasmussen, 2001). Sea snakes are also caught as young crocodiles probably have a greater marine component in nontarget by catch. Fry et al. (2001) reported that prawn traw their diets. Hatchlings tend to be primarily insectivorous, but lers operating in northern Australian waters took 17 species of crabs are a dominant item of diet afterward, usually caught in sea snakes as by catch and that the proportion of mature snakes shallow water or at the water’s edge. Saltwater crocodiles are in catches was high (67% males, 89% females). As sea snakes relatively inefficient foragers on fish or other immersed prey have a relatively low fecundity reproductive strategy (annual (Davenport et al., 1990a) and snap at schools rather than attack breeding, 1–20 young), the heavy capture of mature adults ing individuals. On the other hand, they are highly successful indicates a high trawling impact. It is evident that sea snakes water’s edge specialists that (as adults) mainly take terrestrial merit coordinated international study to assess human impacts birds and mammals. Adult crocodiles will also scavenge from upon these important predators as well as to inform manage large carcasses. Another crocodile, Crocodylus acutus (American ment and conservation initiatives. crocodile), lives in similar habitats to C. porosus (tidal estuaries, coastal lagoons, and mangrove swamps) and is known to eat 6.09.3.2.2 Other snakes fish and turtles. Its ecology has been reported by Thorbjarnarson Besides sea kraits and true sea snakes, a few other snakes enter (1989) and it appears to be an American equivalent of the latter brackish and coastal waters and feed on marine species. species. However, C. acutus has been heavily depleted (by shoot Prominent groups include the file snakes (family ing, habitat loss, and tourist development) over its extensive Acrochordidae) and mud snakes (family Colubridae, subfamily range (Florida, Caribbean, and Central and South America), Homalopsinae). File snakes (three species) are nonvenomous which includes freshwater habitats, so its present trophic impor snakes, widely distributed from western India to the Solomon tance is limited. In Belize, the demise of C. acutus has allowed yet Islands and are entirely aquatic ambush predators that have another crocodile species, Crocodylus moretti,toe xpandi nto convergent morphological and physiological similarities with mangrove habitats with salinities as high as 22 (Platt and sea snakes (e.g., laterally flattened body, ovoviviparous repro Thorbjarnson, 2000), where it presumably feeds on a variety of duction, and prolonged diving capacity), but are remote marine invertebrates and fish. taxonomically. One species is estuarine and coastal Although species of the family Alligatoridae (alligators and (Acrochordus granulatus; it has a small salt gland and has to return caimans) do not possess the lingual salt glands found in mem to freshwater at intervals) and takes fish in estuarine systems, bers of the family Crocodylidae (true crocodiles) (Taplin et al., particularly mangroves as well as the sea itself (Lillywhite, 1991). 1982), hence cannot remain in brackish water for too long, there It selects relatively small and elongate gobioid fish, including is good evidence that American alligators (Alligator mississippien Trypauchenidae (Voris and Glodek, 1980). sis) living in coastal regions spend time in brackish water and eat Mud snakes are viviparous back-fanged venomous snakes marine organisms. Tamarack (1988) studied coastal island alli occurring from India to Australia. Some are living in freshwater, gators of the US state of Georgia. He found that subadult and but most are estuarine, especially in mangrove areas. Bitia adult alligators fed extensively on striped mullet (Mugil cephalus), hydroides is coastal in distribution and eats a variety of marine although they used the inefficient sideways snapping also gobies, of sizes up to 38% of the snake’s body mass (Jayne reported for crocodiles feeding on submerged prey (Davenport et al., 1995). Most estuarine mud snakes are fish eaters, but two et al., 1990a); they caught a very small proportion, but the fish crustacean eaters have attracted attention recently after studies were superabundant and crowded into narrow waterways by the in Singapore mangroves. Fordonia leucobalia, the crab-eating falling tide. Tamarack (1988) also reported that juvenile and snake, has a large skull and massively enlarged fangs that can subadult alligators ate fingerling mullet, shrimps, and blue penetrate the crab exoskeleton (Murphy and Voris, 2002), crabs (Callinectes sapidus) when in brackish ponds; such ponds allowing the snake to subdue hard-shelled crabs quickly, were close to freshwater swamps, so access to freshwater for which it then swallows whole. Another species, Gerard’s water drinking was presumably not a problem. Tamarack (op. cit.) snake (Gerarda prevostiana), has an apparently unique beha even found that alligators foraged on open coast beaches, taking vioral feeding mechanism. It selects soft, newly molted crabs crabs in the surf and attacking schools of fish; they also ate dead that can be far larger than could be swallowed whole. It bites fisheries’ discards (benthic fish), washed in by the tide. the crab, and then pulls the crab through a tight loop of the The widespread common caiman of the Caribbean, and body, tearing it to pieces in the process (Jayne et al., 2002). As Central and South America, Caiman crocodilus, is predomi with file snakes, there appear to be no specific records of mud nantly a freshwater crocodilian, but has been reported from snake predators. Some mud snakes are collected commercially estuaries and coastal lagoons as well. Like C. porosus, they are for the snakeskin trade. capable of assimilating unsaturated long-chain fatty acids from a marine diet (Davenport et al., 1990a, 1992a) and as juveniles and subadults take crustaceans, mollusks, fish, and turtles (e.g., 6.09.3.3 Crocodilians Velasco et al., 1994). Crocodiles have an extensive marine fossil history (they were Overall, it has been suggested (e.g., Brochu, 2001) that, among Jurassic marine top predators), but among 23 living prior to overexploitation by humans in the last 150 years, crocodilian species, only the widely distributed (Sri Lanka in coastal and estuarine crocodilian populations were dominated the west, to the Solomon Islands in the east) saltwater or by the two cosmopolitan species C. porosus and C. acutus that

Table 1 Distribution and foraging habitats of living species of sea turtles

Hatchling/ Adult foraging juvenile foraging Family Species Common name Distribution habitat habitat

Dermochelyidae Dermochelys coriacea Leatherback turtle Pan tropical to cold temperatea Coastal and Oceanic surfacec oceanic pelagicb Cheloniidae Chelonia mydas Green turtle Pan tropical to warm temperatea Coastal benthic Oceanic surface Eretmochelys Hawksbill turtle Pan tropicala Coastal benthic Oceanic surface imbricata Caretta caretta Loggerhead turtle Pan tropical to cool temperatea Coastal benthic, Oceanic surface oceanic pelagicb Lepidochelys kempii Kemp’s Ridley Limited; Mexico, Texas and Coastal benthic Coastal/oceanic turtle eastern USA. surfacec Tropical to warm temperate. Lepidochelys olivacea Olive Ridley turtle Pan tropical to warm temperatea Coastal benthic Oceanic surfacec Natator depressus Flatback turtle Limited to northern Australia and Coastal benthic Coastal surfacec New Guinea. Tropical/ subtropical a Indicates species that are found in Indo-Pacific and Atlantic Oceans: b Indicates species that will also forage on fishery long line baits: c Indicates limited knowledge, but assumed.

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 238 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems

Table 2 Reproductive output of sea turtles

Mean Total egg mass Typical adult Mean no. Mean no. Mean no. individual per turtle per Mean individual body mass eggs per clutches per eggs per egg mass season hatchling mass Species (kg)a clutch season season (g) (g) (g)

Dermochelys 400 81.5 6.17 503 75.9 38 178 44.4 coriacea Chelonia 180 112.8 2.93 331 46.1 15 259 24.6 mydas Eretmochelys 80 130.0 2.74 356 26.6 9 470 14.8 imbricata Caretta caretta 140 112.4 3.49 393 32.7 12 851 19.9 Lepidochelys 40 110.0 1.80 198 30.0 5 940 17.3 kempii Lepidochelys 45 109.9 2.21 243 35.7 8 675 17.0 olivacea Natator 84 52.8 2.84 150 51.4 7 710 39.3 depressus

Modified from Miller, J.D., 1997. Reproduction in sea turtles. In: Lutz, P.L., Musick, J.A. (Eds.), The Biology of Sea Turtles. CRC Press, Boca Raton, FL, pp. 51–81. turtle eggs have a high hatching success rate (90% is common), centuries ago. Originally no different in principle from predation but this is not true of leatherback turtles which often suffer high by other large mammals, it became increasingly damaging with levels of embryonic mortality (Bell et al., 2003), so that around exponential increases in coastal human populations, the intro half of embryos die in the egg and decompose (providing duction of cash economies, and the development of transport nutrient enrichment to nutrient-poor sandy beach substrata). mechanisms that permitted inland sale of turtle eggs. Further Egg predators are numerous. Mammals (e.g., raccoons, arma factors have been the widespread use of turtle eggs in tropical dillos, foxes, pigs, and rats) are especially effective as they have cuisines and medicine, plus a common belief in their aphrodisia a good sense of smell and digging capability; all are widely cal value. Human poaching of turtle eggs has been, and remains, a distributed naturally and have often been introduced outside potent cause of population extinction (e.g., Chan, 2006). their native range. Raccoons (Procyon spp.) have been exten Relatively large mammals are not the only egg predators. sively studied in the United States and the Caribbean, where Monitor lizards (Varanidae) are important turtle egg predators they prey on all species of turtle eggs. Depredation rates can be in Asia and Australia; similar to foxes and raccoons, they are highly variable, ranging from natural levels of perhaps 5–15% opportunistic generalist predators and scavengers that enter of nests to 92% where there are local concentrations of rac beach ecosystems as well as foraging in terrestrial and estuarine coons as a result of human food waste availability. Similarly, habitats. Limpus (1971) reported that almost all eggs of the red foxes (Vulpes vulpes) were recorded as taking 88% of logger flatback turtle N. depressus were taken on one nesting beach in head turtle eggs from a single beach throughout the loggerhead southeast Queensland, Australia. Ghost crabs (Ocypode spp.) turtle (Caretta caretta) breeding season in Turkey (MacDonald are also common inhabitants of turtle breeding beaches, and et al., 1994); in this case, the foxes cached eggs inland before are often the next most important egg predators after raccoons. feeding them to their cubs. Such high predation rates may, to It has been found that there are intra-guild effects where rac some extent, be an artifact of reduced turtle populations, as coons are controlled, as they are also predators of ghost crabs. some species at least appear to have natural reproductive stra Barton and Roth (2008) recently found that reducing raccoon tegies designed to saturate predators. The two Ridley turtles populations in Florida caused increased populations of ghost show synchronized ‘arribada’ breeding, during which thou crabs, and allowed the crabs to reach larger size. In this case, the sands of females come ashore during short periods to lay highest rates of loggerhead turtle egg predation (31%) occurred their eggs. Famously, the Kemp’s Ridley (Lepidochelys kempii) where raccoon densities were lowest, and crab populations lays eggs (during the day) almost entirely on a single beach were highest. Understanding food-web connectivity is clearly system (at Rancho Nuevo) in eastern Mexico. In the late 1940s, crucial to turtle conservation. around 40 000 females came ashore during the annual arri It has also become clear in recent years that some turtle egg bada. By the mid-1980s, this population had been reduced to predators are small and cannot be excluded by fences. Insect 200 by egg poaching by humans and by introduced egg pre predation is common, but inconspicuous. Donlan et al. (2004) dators (e.g., pigs), but is showing strong signs of recovery after reported that coleopteran larvae (of the click beetle Lanelater vigorous conservation initiatives. Unsurprisingly, a standard sallei) were major predators of loggerhead turtle nests in Cape and effective breeding beach turtle conservation management Florida State Park (CFSP). From 2001 to 2003, they found that tool involves fencing nests so that they cannot be taken by such the beetle larvae were the most important loggerhead turtle egg predators. It is also common to control predators by trapping predators in the CFSP, infesting up to 47.5% of nests and and removing/culling them. killing substantial numbers of eggs within each of those nests Human predation on sea turtle eggs has been unremitting for (Figure 9). They also reviewed the literature devoted to insect millennia, although ecologically unimportant until a few predation on turtle eggs and found it to be worldwide and

(a) (b)

Figure 9 Insect predation on sea turtle eggs: (a) Larva of click beetle Lanelater sallei and (b) Loggerhead turtle eggs destroyed by click beetle larvae. Photo: Josiah Townsend and Ellen Donlan. applying to all sea turtle species. Illustrating that this is a by Gyuris (1994). The extent of such predation is difficult to geographically widespread problem, Maros et al. (2003) quantify and evidence is conflicting, probably because it found high levels of predation (18%) of leatherback turtle reflects a highly variable situation (geographically and tempo eggs in French Guiana by mole crickets (Scapteriscus didactylus). rally). Gyuris (1994) found that 93.4% of 47 observed free- Even after two centuries of serious depletion, there are still swimming green turtle hatchlings were eaten between entering some turtle breeding areas that illustrate the great subsidy the water and reaching the reef edge at Heron Island, Australia. provided to terrestrial and nearshore ecosystems by egg-laying However, predation rates on much larger numbers (>1700) of sea turtles. Witt et al. (2009) recently surveyed the world’s semitethered hatchlings were far lower (mean 31%) and clearly largest remaining leatherback rookery in Gabon, West Africa. affected by state of tide and phase of moon. In contrast, Stewart They calculated from aerial survey that 36 185–126 480 and Wyneken (2004) studied mortality of 217 loggerhead clutches were laid by 15 730–41 373 female D. coriacea each hatchlings swimming offshore from beaches in Florida to year. Using the data of Table 2, this suggests that 224–782 deep water; only 11 (5.1%) were taken by aquatic predators. metric tons of leatherback eggs are laid per annum in Gabon, Thus, Gyuris (1994) thought fish predation on green turtles or 0.37–1.3 metric tons per kilometer of coastline. It is evident during the swimming frenzy was the most important source of that the eggs will provide nutrition for an extensive terrestrial mortality during the turtles’ life history, while Stewart and food web, either directly or indirectly. Wyneken found it to be a relatively minor element. However, Sea turtle eggs hatch asynchronously, but commonly the the latter authors stress that the Floridian loggerhead hatchlings hatchlings of a single nest emerge nearly simultaneously, entered the water unpredictably over a long coastline, making it usually at night, often near dawn. The simultaneous emergence relatively difficult for predators to locate them. On small of hatchlings that have emerged from eggs over several days has islands, or at concentrated arribada beaches, it may be easier been interpreted as a predator-swamping mechanism (Tucker for concentrations of fish predators to build up and hence take et al., 2008). Widely introduced fire ants (Solenopsis invicata) a greater proportion of hatchlings. take advantage of this and are known to attack, kill, and skele Once sea turtle hatchlings complete their swimming frenzy, tonize turtle hatchlings before they leave the nest (e.g., Parris they enter a prolonged oceanic trophic phase (outside the et al., 2002), although their impact in terms of hatchling mor scope of this chapter) lasting several years (6.5–11.5 years in tality is currently unclear. Fire ants are opportunistic predators the case of loggerhead turtles; Bjorndal et al., 2000), during that also take chicks of ground-nesting birds and hatchlings of a which they subsist predominantly on neustonic food items. In wide variety of reptiles including lizards and snakes. Turtle egg the case of the leatherback turtle D. coriacea (Figure 10), all incubation and hatching coincide with periods when ant’spro tein demand is particularly high because of brood production. After leaving the nest, turtle hatchlings head directly for the sea and then swim continuously away from the coast (except in the case of flatback turtle hatchlings; Walker and Parmenter, 1990) in the ‘swimming frenzy’ until they are entrained in offshore current systems. While crossing the beach and shallow coastal waters they fall prey to a wide range of terrestrial and marine predators. Almost all of the egg predators already described also take hatchlings, drawn to the nest by scent and sound. Avian predation of hatchlings is very common from a suite of bird species including corvids (e.g., fish crows (Corvus ossifragus) in Florida), seagulls, herons, storks, eagles, and even owls (e.g., great horned owls Bubo virginianus; Toland, 1991). Semiterrestrial crabs, particularly ghost crabs, take large num bers of hatchlings too. Hatchlings are also preyed upon in the water by inshore predators, particularly fish. These include sharks, although tele ost Serranidae and wrasses (Labridae) were dominant fish Figure 10 Female leatherback turtle emerging from the sea to lay eggs predators in a study of predation on green turtle hatchlings on beach in French Guiana. Photo: Jean-Yves Georges.

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 240 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems juvenile and subadult development takes place in the open the Gulf of St. Lawrence in late summer. She utilized most ocean and much adult foraging too. However, adult leather of the water column from near the surface to almost 100 m backs do forage at high latitude in coastal waters and have and repeatedly dived (1312 dives in a 44-day period in the a specialized trophic niche; they forage exclusively on gela Gulf of St. Lawrence) into near-freezing waters when fora tinous plankton throughout the water column, mainly ging near the bottom. Jellyfish form an extremely low- upon large cnidarian medusae (e.g., Rhizostoma octopus and quality diet (see Doyle et al. (2007) for proximate analysis), Cyanea capillata). Such medusae commonly occur in coastal being about 96% water, with around 67% of the dry mass aggregations, and sightings and strandings of leatherback being inorganic. This means that leatherbacks have to eat turtles have been shown to correspond with such aggrega exceptionally large quantities of food (Duron, 1978), from tions (Houghton et al., 2006;seeFigu res 11 and 12). James more than 100% body mass per day in hatchlings et al. (2006) used satellite tagging to follow a female (Lutcavage and Lutz, 1986), to at least 50% body mass leatherback turtle around the coast of Nova Scotia and in per day in adults (Davenport, 1998). Despite this situation,

(a) (b) 55 D

54 100 km Ireland N)

� 53 C B Wales 52 Latitude ( A

51 (c) –8 –6 –4 –2(d) –8 –6 –4 –2 55

54 Relative scale: 100 km >1000 N)

� Ireland 53 300–800

Wales 10–50

Latitude ( 52

51 –8 –6 –4 –2 –8 –6 –4 –2 (e) (f) 55

54 N)

� 100 km Ireland Ireland 53

Latitude ( Wales Wales 52

51 –8 –6 –4 –2 –8 –6 –4 –2 Longitude (� W) Longitude (� W) Figure 11 Coastal foraging by leatherback turtles Dermochelys coriacea around the British Isles I. Areas covered during aerial surveys are shown for (a) 2003, (c) 2004, and (e) 2005. Each square represents the midpoint of a 5-min survey unit (7710 m2). Distribution of Rhizostoma aggregations are also shown for (b) 2003, (d) 2004, and (f) 2005. Data are total abundances for the period between July and September (leatherback peak season) in each year. Each circle represents a measure of abundance during a single 5-min observation period. Relative scale of aggregations is shown in (d) ranging from >1000 to 10–50 jellyfish per 5 min. Locations of hotspots are shown in (b): A, Carmarthen Bay; B, Rosslare Harbor; C, Tremadoc Bay. A fourth possible hotspot (D, Solway Firth) is also shown, although this site was only surveyed once under good conditions, thus preventing a full assessment of its temporal and spatial constancy. From Houghton, J.D.R., Doyle, T.K., Wilson, M.W., Davenport, J., Hays, G.C., 2006. Jellyfish aggregations and leatherback turtle foraging patterns in a temperate coastal environment. Ecology 87, 1967–1972.

(a) 55 good information about the foraging and nutrition of two species, the loggerhead turtle C. caretta and the green turtle C. mydas. Turtle foraging and nutrition have been extensively 54 reviewed, particularly by Mortimer (1982) and Bjorndal (1985, Ireland 1997). The loggerhead turtle is predominantly carnivorous, 53 ingesting limited amounts of macroalgae and seagrasses, and exploits coastal benthic habitats throughout its juvenile and Wales adult life, taking more benthic prey as it grows (e.g., Casale Longitude (° N) Longitude (° 52 et al., 2008), although it is becoming clear that some popula

100 Km tions also exploit the pelagic and offshore realms as well as 51 scavenging at the sea surface and taking fishery bycatch. –8 –6 –4 –2 The most comprehensive studies of diet have been conducted (b) 55 in the USA, notably by Seney and Musick (2007) who studied gut contents of loggerheads collected in the coastal waters of 54 Virginia between 1983 and 2002. During the early part of this period, the loggerheads mainly ate horseshoe crabs (Limulus Ireland polyphemus), then switched to blue crabs (C. sapidus). From the 53 mid-1990s onward, they predominantly ate finfish, presumed Wales to be largely discards or scavenged from nets as loggerheads are

Longitude (° N) Longitude (° 52 too slow moving to catch live fish. These findings were linked to a sequence of overfishing, first of horseshoe crabs (used as 100 Km 51 whelk bait), then of blue crabs. This study also indicated a –8 –6 –4 –2 plastic, opportunistic diet, especially as other regularly Longitude (° W) recorded food item species included 11 crustaceans, 20 mollusks, 9 teleost fish, and 7 macroalgae/other plants. Figure 12 Coastal foraging by leatherback turtles Dermochelys coriacea Loggerheads have especially powerful jaws and can crush dec around the British Isles II. (a) All leatherback sightings (for a given section of the Irish Sea) from 1950 to 2005 (n = 143). Data are plotted according apod crustaceans and shelled mollusks with increasing ease as to decade: 2000–05 (open circles); 1990s (solid circles); 1980s (open they grow. Their ability to exploit bycatch of the fishing indus squares); 1970s (solid squares); 1960s (open diamonds); and 1950s tries may have helped them to cope with diminishing (solid diamonds). (b) Sightings where turtles were associated with jelly populations of ‘normal’ prey, but, as with other turtles, they fish (solid triangles) and when foraging activity was confirmed (open are vulnerable to mortality caused by active and passive fishing triangles). Three live turtles sighted during 2003 and 2004 aerial surveys gears. Less comprehensive studies have confirmed the oppor are marked with stars. The three sightings confirming predation on tunistic nature of loggerhead feeding; Plotkin (1989) found Rhizostoma octopus were in Carmarthen Bay (51.658° N, 4.538° W and that sea pens (Virgularia presbytes) (octocorals that occur from 51.618° N, 4.738° W) and Tremadoc Bay (52.808° N, 4.368° W). 9 to 90 m depth in nearshore waters) were a major part of the Predation of Chrysaora hysoscella was observed on a single occasion in diet of Texan loggerheads, whereas Burke et al. (1993), who Tremadoc Bay (54.678° N, 3.738° W). From this figure and the preceding one it can be seen that leatherback sightings coincide with coastal jellyfish studied fecal samples from juvenile loggerheads caught at Long hotspots. From Houghton, J.D.R., Doyle, T.K., Wilson, M.W., Davenport, J., Island, New York, found a dominance of crabs, particularly Hays, G.C., 2006. Jellyfish aggregations and leatherback turtle foraging spider crabs (Libinia emarginata). Generally, plant material patterns in a temperate coastal environment. Ecology 87, 1967–1972. appears in most loggerhead guts, but it is not clear whether loggerheads can digest plant material, or simply take it in incidentally when preying upon animals. the leatherback is probably the fastest-growing reptile, Although hatchling and small juvenile green turtles are carni reaching maturity in 6–16 years (Rhodin, 1985; Snover vorous or omnivorous, large juveniles, subadults and adults are and Rhodin, 2008; Jones et al. 2011). primarily herbivores. This scenario has been postulated from Natural predation on adult D. coriacea in coastal waters traditional diet and behavioral studies for many years (e.g., appears to be low and probably the most important predators Bjorndal, 1997), but has recently been confirmed by extensive are large sharks of various species (see Heithaus et al. (2008) stable isotope studies performed on green turtles in the south for review) that take adult turtles when they are close to nesting western Pacific (Arthur et al., 2008), which recruit to inshore beaches. There is at least one record of a herd of killer whales foraging habitats at about 44 cm curved carapace length. There (Orcinus orca) killing and eating a leatherback turtle foraging in are regional differences in the size at such recruitment; Bjorndal temperate waters off the Californian coast (Pitman and Dutton, and Bolten (1988) reported a much smaller size for the Bahamas 2004), while there is evidence of jaguars (Panthera onca) killing (20–25 cm) in the western Atlantic, whereas the author (personal nesting leatherbacks in Central and South America. However, observations) has seen green turtles as small as this on seagrass natural predation from these sources is miniscule by compar beds rather further north in Bermuda. Balazs (1980) reported ison with the mortality caused by fishing gear (e.g., longlines) recruitment at around 35 cm in Hawaii. in oceanic waters, which has nearly extirpated leatherbacks in Subadult and adult green turtles mainly eat plant material, the Pacific (Spotila et al., 2000). although they do take some sponges and eat jellyfish and other All of the other sea turtle species forage extensively in invertebrates if available (cf., Bjorndal, 1997); the author has inshore waters after completing their pelagic oceanic phase. observed an adult green turtle living in an anchialine pool in Principally they are benthic feeders and there is particularly Bermuda (apparently marooned there many years ago after a

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 242 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems

populations has been underrated; they found that, at Shoalwater Bay in central Queensland, there was a population of immature and adult C. mydas that entered mangrove forests at high tide to feed on Avicennia marina leaves and fruit. The turtles retreated to seagrass meadows on falling tides, so often consumed a mixture of the two plant types. Limpus and Limpus also found that fruit eating was much more important than eating of mangrove leaves. In several locations, green turtles are predominantly consu mers of macroalgae. Numerous studies have been conducted over nearly three decades in the Hawaiian archipelago, where some populations feed mostly on macroalgae. Russell and Balazs (2009) have recently summarized findings, noting that green turtles have been recorded eating 130 species of marine vegetation in that period, and also that the three most common current seaweeds consumed are actually particularly nutritious non-native soft red seaweed (Rhodophyta) species: Figure 13 Green turtle, Chelonia mydas, browsing on marine algae. Acanthophora spicifera, Hypnea musciformis, and Gracilaria salicor Photo: copyright Ron Lucas, Bermuda. nia. The green turtles progressively switched from native to non-native species over a 10-year period. Other areas where macroalgae are the most important dietary items include the hurricane) where it subsists on mangrove vegetation and fruit, Ogasawara areas of Japan and a number of central Pacific plus quantities of the ‘upside-down’ jelly fish (Cassiopea xama Islands as well as Brazilian waters from which the Ascension chana) that carpet the floor of the pool. Clearly, green turtles Island nesting green turtles (Figure 14) migrate (see review of have the ability to consume a wide variety of food items, but in Bjorndal (1997)). many locations do not do so. Generally, green turtles either eat Green turtles from the eastern Pacific (sometimes, but con a great variety of seagrasses, or eat a variety of macroalgae troversially, classified as black turtles, C. mydas agassizzi) are (e.g., Figure 13). It is unusual for them to eat both, although notably more omnivorous, eating animal material as well as Ross (1985) found both food types in the stomachs of three plants. Several observations were summarized by Bjorndal, out of nine green turtles sampled from the northern Indian (1997); more recently, Seminoff et al. (2002) studied green Ocean (the rest had eaten one or other category). Bjorndal turtles from the Gulf of California over a 4-year period, using (1985) attributed this dietary separation to the great differences a variety of techniques (gastric lavage, fecal analysis, and dis in complex carbohydrates between these major plant groups – section of carcasses). They reported ingestion of a variety of red, which require different gut microflora to break them down. brown, and green macroalgae, as well as a large number of Seagrass carbohydrates are dominated by cellulose, whereas animal taxa, notably sabellid polychaetes, sea hares (Aplysia), algal structural carbohydrate is complex, including glucan, mollusks, and sponges. Clearly, the common assumption that mannan, xylan, carrageenan, alginic acid, and uronic acid. green turtles are obligate herbivores, mainly on seagrasses, is Bjorndal also pointed out that dietary shifts from one plant fallacious. group to the other would require an inefficient alteration in gut Hawksbill turtles, Eretmochelys imbricata, are specialized for microflora, unlikely in the short term. agers as adults. Like other turtle species, the hatchlings and Initial study of foraging and nutrition was mainly in the Caribbean, where the major food plant is the seagrass Thalassia testudinum, making up 79% of the dry mass of stomach con tents, with other seagrass species making up a further 10% (Mortimer, 1981). In these extensive shallow-water seagrass meadows, green turtles are almost the only browsing animals and apparently maintain grazing plots, preferentially consum ing new growth with lower lignin content and higher protein levels (Bjorndal, 1985). Seagrass browsing is associated with relatively slow growth to sexual maturity. For example, imma ture green turtles of the large sea population foraging on seagrass on the reef of Bermuda appear to spend some 30 years there before leaving to breed at a variety of locations in the Caribbean (Jennifer Gray, personal communications). Such seagrass browsing is also characteristic of populations of sea turtles living in other parts of the world, including the coasts of India, Yemen, and Queensland, Australia (see Bjorndal (1997) for review). However, several studies have revealed different or more complex situations. Figure 14 Adult female green turtle, Chelonia mydas, crawling down For example, Limpus and Limpus (2000) noted that the Ascension Island beach after laying eggs. Note the large number of importance of mangrove vegetation to some green turtle nesting tracks in sand. Photo: Graeme Hays.

changed the turtles’ diet. Centuries ago, hawksbill meat was undoubtedly toxic to humans because of their diet of sessile invertebrates with elaborate chemical defense systems. Now hawksbills do not show such toxicity. McClenachan et al. argue that this change reflects a broadening of hawksbill diet as population densities have fallen and competition has weakened. There is some debate about whether hawksbills are predo minantly spongivorous in other parts of their extensive and diffuse tropical range. Data are summarized by Bjorndal (1997), and it would seem that there is a tendency for small juveniles that have recently recruited to coastal waters to be much more omnivorous (feeding on red algae, ascidians, sea urchins, and fish). However, limited data for the Pacific and Indian oceans appear to confirm that large hawksbills are fairly strict spongivores. The feeding ecology of the two Ridley sea turtle species in coastal waters has been studied in rather less detail. Adult and subadult Kemp’s Ridleys (L. kempii), the smallest sea turtle species, have been studied by Shaver (1991) in Texan waters. Figure 15 Adult hawksbill turtle foraging on Caribbean coral reef. Photo: She reported that Kemp’s Ridleys were specialist crab eaters in copyright Ron Lucas, Bermuda. shallow nearshore waters, crabs making up 78% of the food items present in the digestive tract, and 94% of ingested dry early juveniles have an oceanic and neustonic trophic niche; they mass. Mollusks made up most of the rest of the gut contents. have been reported from rafts of Sargassum.Theyr ecruitt o A recent study of immature L. kempii caught in Gullivan Bay, coastal areas, typically coral reefs or seagrass meadows, at around southwest Florida (Witzell and Schmidt, 2005), indicates a 20–25 cm carapace length in the Atlantic and about 38 cm more catholic diet, dominated by benthic tunicates and crabs. carapace length in the Pacific. In a seminal paper, Meylan However, Witzell and Schmidt also reviewed the literature (1988) clearly demonstrated that Caribbean adult hawksbills devoted to Kemp’s Ridley diet and it is apparent that this (Figure 15) are specialist spongivores, browsing on sponges of species is an opportunistic carnivore, similar in many respects coral reefs and competing with a diverse community of spongi to the loggerhead turtle. Although the olive Ridley (Lepidochelys vorous fish. She found that 95.3% of the dry mass of the gut olivacea) occurs in far larger numbers and throughout the contents of 61 hawksbills consisted of sponges. She also demon Atlantic, Indian, and Pacific oceans, there has been remarkably strated that Eretmochelys was highly selective in the sponges that little systematic study of its trophic relationships, albeit it it eats, 99% of all sponges consumed being derived from only 3 seems to be a similarly opportunistic species, consuming of 12 demosponges. Chondrilla nucula was the most preferred crabs, shrimp, rock lobsters, jellyfish, and tunicates in near- sponge. Hill (1998) pointed out that this sponge is highly shore waters up to 80 m deep (e.g., Bjorndal, 1997). successful in interspecific sponge: coral competition for space, Probably least well documented is the diet of the flatback and that selective spongivores such as the hawksbill turtle, con turtle (N. depressus). This species feeds in coastal turbid waters sequently control the diversity and relative abundance of of tropical Australia (Limpus et al., 1983). Flatbacks appear to sponges – they also liberate more space for corals. feed in shallow waters on a mixed pelagic and benthic carni Unfortunately, hawksbill populations are greatly dimin vorous diet of soft corals, sea cucumbers, mollusks, and ished (by far more than 90%) as a result of long-term jellyfish (Bjorndal, 1997). sustained human exploitation over many centuries (mainly Natural predation levels on adult and subadult cheloniid for decorative tortoiseshell, but more recently for highly sea turtles, as with leatherback turtles, appear to be low. polished whole-animal tourist curios as well). It is highly prob Nesting female sea turtles are sometimes taken by large terres able that the community composition on many coral reefs has trial carnivores such as jaguars or crocodiles; Whiting et al. been altered by the loss of large populations of hawksbills. (2008) reported regular beach mortalities of flatback females However, lest it be thought that hawksbills are entirely positive due to large estuarine crocodiles ((C. porosus) during the nest for coral species, the study of Leon and Bjorndal (2002) should ing season). At sea, all sea turtles are subject to shark attack be born in mind. They studied selective feeding by hawksbills (Heithaus et al., 2008), and this is likely to occur in coastal on Dominican Republic reefs. They found extensive spongiv waters close to breeding beaches. Commonly, sharks bite flip ory, but 59% of their stomach lavage samples contained pers and not all attacks are immediately fatal. Loggerhead quantities of the coral Ricordea florida. Interestingly, Anderes turtles found in northern European waters sometimes feature Alvarez and Uchida (1994) reported that gravid female hawks healed flipper wounds (personal observations). However, bills around Cuba eat quantities of calcareous coralline algae – adult sea turtle mortality in coastal waters because of human perhaps to provide adequate calcium for eggshells. Overall, it is activities is massively greater than natural mortality. Mostly this evident that the adult diet of hawksbill turtles in the Caribbean is because of inshore finfish/shellfish fishing activity and is not is very different from that of all other sea turtles. McClenachan targeted at sea turtles per se. However, extensive direct poaching et al. (2006) have recently argued convincingly from historical of sea turtles still takes place (e.g., in Seychelles, southeast Asia, evidence that great decreases in hawksbill populations (they Central America, and Micronesia), mostly for green turtles that claim >99.7%) over the last few centuries have actually are highly valued for their meat by local people and traders,

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 244 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems and also for hawksbills that are taken for their shells. All that diamondbacks ate a certain amount of vegetation too; there sea turtles are theoretically protected by Convention on fore, it would seem that M. terrapin exploits a large fraction of the International Trade in Endangered species (CITES) legislation available resources. However, in the most detailed available that prohibits international trade, but enforcement is extremely account of diamondback trophic ecology, Tucker et al. (1995) lax. Indonesia, particularly the island of Bali, has been repeat found that terrapin diet in South Carolina salt marshes was edly demonstrated to flout CITES rules both in direct fishing dominated (76–79%) by periwinkles (Littorina irrorata)which − and in trade. In addition to directly consuming large quantities reach enormous densities (<700 m 2), and which the turtles fed of green turtles collected from large areas of the Indo-Pacific on at high tide. They also recorded a similar range of invertebrate region (e.g., Davenport, 1988), Indonesia has also exported prey as reported by Coker (1906): snails, small crabs (including products (tortoiseshells and curios) representing several Uca and Callinectes), polychaetes, and fish (captured or sca million turtles in the past 30 years (e.g., Barr, 2001), especially venged). In addition, they showed that females (much larger to Japan. Japan has been the center of the worldwide hawksbill than males) ate larger prey. Davenport et al. (1992b) found that tortoiseshell trade for centuries and built up extensive stocks adult male diamondbacks (101–117 mm carapace length) were in an anticipation of a trading ban by the mid-1990s limited to the consumption of relatively small littorinids (188.4 metric tonnes of tortoiseshell, equivalent to about (<8 mm shell height) and mussels (<30 mm) by virtue of gape 1.8 million hawksbill turtles). The tortoiseshell carving indus size and force of the jaws. try is still healthy in Japan, which is still attempting to reopen Davenport and Ward (1993) found that diamondbacks had international trade, while clandestine trading is reported from an unusually large appetite, eating satiation meals of 7.2% body many parts of the world. weight (when fed on mussel flesh) at 25 °C, and averaging 3.7% body weight per day. These values are about 10 times as much 6.09.3.4.2 Other turtles food as is eaten (size for size and at the same temperature) by Among other turtles that exploit coastal/estuarine resources, related emydid freshwater species. To some extent, this is a the emydid diamondback terrapin (Malaclemys terrapin) general characteristic of estuarine carnivores, which tend to eat (Figure 16) has been the most studied. It lives in estuaries large quantities of food, processed inefficiently, but another and salt marshes along the east coast of the USA from New factor may be at work in diamondbacks. Although diamond York State to Texas, and is apparently the most salt-tolerant backs do eat while in full seawater, Davenport and Ward (1993) nonsea turtle species–recently it has been found that a popula found that they progressively lose appetite under such condi tion of diamondbacks lives on Bermuda, presumably traveling tions, to the extent that appetite after 18 days’ exposure to at least 960 km of sea from the US coast to naturally colonize seawater is only 22–54% of that shown when freshwater is this isolated archipelago (Davenport et al., 2005; Parham et al., readily available. Diamondbacks in their coastal habitats may 2008). However, diamondbacks cannot survive indefinitely well face periods of 2–4 weeks when no rain falls. All of their in seawater; they need intermittent access to freshwater prey items are invertebrates that will be isosmotic with seawater, (Davenport and Macedo, 1990). so will have body fluids more than twice as concentrated as By exploiting estuarine and salt marsh ecosystems, diamond terrapin blood. It is also probable that terrapins will take in backs gain access to far richer food resources than are some seawater with their food, although they avoid deliberate characteristic of freshwater ecosystems. Coker (1906) long ago drinking of high salinity media (Davenport and Macedo, 1990). showed that diamondbacks are predominantly carnivores which In consequence, it is clear that diamondbacks restrict food intake eat a range of salt marsh invertebrates (crabs, littorinid snails, when in seawater for long periods. Their great appetite when and nereid worms). Captive adult animals readily take small recently hydrated presumably helps them to compensate. intact mussels, crabs, and even snails as small as Hydrobia ulvae Salt marshes on the east coast of the USA have a very diverse (<4 mm shell height) (Davenport, 1992). Several studies have crab fauna, which includes several species of small fiddler crabs demonstrated that newly caught wild diamondbacks defecate (Uca), the (introduced) medium-sized portunid crab Carcinus fragments of mussel shells, while Pritchard (1979) reported maenas and the large (<300-mm carapace width) portunid blue crab C. sapidus. Crabs are therefore a source of both food and danger as far as M. terrapin is concerned. Using C. maenas as a test prey species, it has been found that diamondbacks have a complex response to crabs (Davenport et al., 1992b). First, as with molluskan prey, they exhibit size selection of crabs (see Figure 17), mainly eating whole crabs in the small size range, taking a few medium-sized crabs, and no whole large crabs. However, further research showed that diamondbacks did exploit a wider size range of crabs, not by eating them whole, but by biting off legs (predominantly of medium-sized crabs, leaving the animal alive (see Figure 17). Small crabs were rarely ‘cropped’ in this fashion because they were eaten whole, while few large (and dangerous) crabs were cropped either. Detailed investigation showed that diamondbacks make complex deci sions when cropping medium-size and large crabs. Normally they attack from behind and bite off the rearmost walking Figure 16 Adult female diamondback terrapin, Malaclemys terrapin limbs first, as these are the furthest from the dangerous claws (Bermudian specimen). Photo: John Davenport. (chelipeds). A significant number of crabs in the natural habitat

100 67% of radio-tagged hatchlings were killed by rats in 30 days. The hatchlings and juveniles were taken in vegetation mats at high water level – a habitat which is normally a relatively safe 80 hiding place for young diamondbacks. Avian predators of adult diamondbacks include the bald eagle (Halliaetus leucocephalus) (Clark, 1982), night herons (Nycticorax violacea), and fish crows 60 (C. ossifragus)(Draud et al., 2004). The only other major nat ural predators reported are ghost crabs Ocypode quadrata (Arndt, 40 1991, 1994). However, many diamondback terrapin popula tions have been in long-term decline and a major cause of this % Crabs ‘eaten’ % Crabs is mortality due to incidental capture in crab pots (especially 20 abandoned ‘ghost’ pots), which, because of the species’ strong sexual dimorphism and consequent selective capture, has caused skewed sex ratios as well as a reduced and aging popula 0 tion (Dorcas et al., 2007). 10−25 mm 30−50 mm 52−75 mm Several large batagurine river turtles of SE Asia (including Crab size offered Callagur borneoensis and Batagur baska) forage to some extent in 100 brackish waters at the mouth of the river systems they live in (predominantly on vegetation and fruit, including mangrove fruits). C. borneoensis nests on coastal beaches, but has limited 80 ability to control its blood salt levels, so retreats to brackish and freshwater to forage (Moll and Moll, 2004). One group of nonsea turtles seems to be coastal inhabi 60 tants to an unappreciated extent: giant soft-shell turtles (Trionychidae) of Africa and Asia (Moll and Moll, 2004). 40 Several species formerly thought of as purely freshwater have been found repeatedly in coastal waters. The large % Crabs ‘cropped’ % Crabs Nile Soft-shell Trionyx triunguis overlaps in size with the 20 smaller sea turtles, large specimens exceeding 1 m in cara pace length. Although widely distributed in rivers and lakes of Africa and countries of the eastern Mediterranean, it is 0 increasingly being recognized as an estuarine and coastal 10−25 mm 30−50 mm 52−75 mm species as well. There are several records from estuaries and Crab size offered nearshore habitats of West Africa (e.g., Loveridge and Figure 17 Crab (Carcinus maenas) eating by adult male diamondback Williams, 1957), while Taskavak and Akcinar (2009) have terrapins (Malaclemys terrapin). Above (blue histograms): size selection recently summarized Turkish marine records of the species by 11 terrapins (101–117 carapace length) offered 30 crabs of each size from the early twentieth century onward. Not only are the class. Terrapins predominantly ate small crabs and did not attack large soft shells common in Mediterranean deltas, estuaries, and crabs. Below (red histograms): ‘cropping’ (i.e., removing and consuming salt marshes, they also forage in coastal waters to depths as crab legs) of crabs by 11 terrapins offered 20 crabs of each size class after much as 55 m and a distance as much as 200 km from ’ ‘ ’ 3 days fasting. Terrapins predominantly cropped medium-sized crabs; known breeding areas. Although they have been caught they ate small crabs whole and rarely attempted to crop large crabs. From (as bycatch) in large numbers on Turkish fish trawling Davenport, J., Spikes, M., Thornton, S.M., Kelly, B.O., 1992. Crab-eating in the diamondback terrapin Malaclemys terrapin (Latreille); dealing with grounds, no study has yet been made of their gut contents dangerous prey. Journal of the Marine Biological Association of the United (Taskavak, personal communications). However, Atatür, Kingdom 72, 835–848. Meylan and Tashkavak (2000) reported that soft shells are omnivorous, taking mainly fish and gastropods, and also crustaceans; they are also scavengers, so may take fisheries’ are without one or both chelipeds (because of intraspecific discards. Soft shells in general are ambush predators that aggression or because they have been preyed upon and are not adapted for rapid swimming. They lay eggs on escaped). Diamondbacks faced with a clawless crab are much coastal and island beaches (Atatür, 1979), so clutches are more aggressive and attack from the front, even if the crab is likely to provide an energetic subsidy for terrestrial preda large. If the crab has a single claw, they invariably attack from tors, whereas hatchlings are probably eaten by the same the less dangerous side. predators as those of the sea turtle species (e.g., seabirds, There are numerous predators that take diamondback terra crabs, and fish). Although not specifically targeted by fish pins. Nests (eggs) are predated by a number of terrestrial ers, they are commonly caught as by catch by trawls and predators, but especially raccoons (Procyon lotor), which also longlines (Taskavak and Akcinar, 2009), to the extent that take adult terrapins. Hatchlings and juveniles of the New York this is a conservation concern. State terrapin population are preyed on extensively by Norway The Asian giant soft-shell turtle (Pelochelys bibroni) overlaps rats (Rattus norvegicus)(Draud et al., 2004), which essentially in size with olive Ridley sea turtles, and often nests on the same achieve high population levels by virtue of human energetic beaches (Hussain, 2003). No specific details appear to be avail subsidy (in the form of stored/waste food). They found that able concerning its marine diet, although it is known to eat fish,

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 246 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems shrimps, and mollusks which it takes as an ambush feeder predator guilds), but also the shorter leaves would have per (Nutaphand, 1979) – clearly, soft-shells merit further study. mitted physically faster current flow rates over the beds, and As with all beach-nesting turtles, eggs provide a subsidy for promoted different regimes of gaseous and nutrient exchange. terrestrial predators, while hatchlings will presumably be Pristine herds of green turtles were keystone controllers of global taken by a suite of avian and marine predators. seagrass bed function; their descendants have only marginal effects. Similar considerations, no doubt, apply to control of reef ecosystems by hawksbill turtles, to foraging by leatherback 6.09.4 General Conclusions turtles on coastal concentrations of jellyfish, and to the top- down predatory influences of estuarine crocodile species. Two pervasive themes are evident in any consideration of the trophic relationships of reptiles that inhabit estuarine and coastal areas. First, they are characterized by substantial exchanges of Reference energy between ecosystems, between oceanic and coastal sys tems, and among coastal, estuarine, and terrestrial systems. Anderes Alvarez, B.L., Uchida, I., 1994. Study of hawksbill turtle (Eretmochelys Marine subsidy of coastal terrestrial systems is especially marked. imbricata) stomach contents in Cuban waters. In: Study of the Hawksbill Turtle in Second, throughout this chapter, there has been repeated Cuba (I). Ministry of Fishing Industry, Cuba. reference to the depletion of populations that has resulted from Arndt, R.G., 1991. Predation on hatchling diamondback terrapin, Malaclemys terrapin (Schoepff), by the ghost crab, Ocypode quadrata (Fabricius). Florida Scientist 54, human activities. There is no doubt that existing populations of 215–217. coastal/estuarine reptiles are of marginal ecological importance Arndt, R.G., 1994. Predation on hatchling diamondback terrapin, Malaclemys terrapin in most cases. Virtually all species are still exploited for their (Schoepff), by the ghost crab, Ocypode quadrata (Fabricius) II. Florida Scientist 57, eggs and meat, for their skins, and for their shells; they are also 1–5. subject to serious damage by fishing gear impact, coastal devel Arthur, K.E., Boyle, M.C., Limpus, C., 2008. Ontogenetic changes in diet and habitat use in green turtle (Chelonia mydas) life history. Marine Ecology Progress Series 362, opment, and general anthropogenic disturbance. Conservation 303–311. measures are often in place, and there are success stories result Atatür, M.K., 1979. Investigations on the Morphology and Osteology, Biotope and ing from them (e.g., the recovery from near-extinction of the Distribution in Anatolia of Trionyx triunguis (Reptilia, Testudines) with Some Kemp’s Ridley turtle in Mexico and increasing populations of Observations on Its Biology. Ege University Faculty of Science Monographs, Izmir, Turkey, vol.18, pp. 1–75. saltwater crocodiles in Australia), but only against baselines of Atatür, M.K., Meylan, P., Taskavak, E., 2000. Trionyx triunguis (Forskal 1775), Nile virtual ecological extinction a few decades ago. In many parts of soft-shelled turtle. In: Pritchard, P.C.H., Rhodin, A.G.J. (Eds.), The the world, population trends still involve consistent declines, Conservation Biology of Freshwater Turtles. Chelonian Research Foundation, especially in the Indo-Pacific region, though there are local Lunenburg, MA. (in press). exceptions (e.g., the rising Hawaiian green turtle population). Balazs, G.H., 1980. Synopsis of biological data on the green turtle in the Hawaiian Islands. US Department of Commerce, NOAA Technical Memorandum NOAA-TM The seminal review of Jackson et al. (2001) convincingly NMFS-SWFC-7, 141 pp. demonstrated the enormous impacts that millennia of overex Barr, C., 2001. Current status of trade and legal protection for sea turtles in Indonesia. ploitation have had on populations of large coastal consumers Marine Turtle Newsletter 54, 4–7. in general. These impacts changed coastal ecosystems well before Barrett, K., Anderson, W.B., Wait, D.A., Grismer, L., Polis, G.A., Rose, M.D., 2005. Marine subsidies alter the diet and abundance of insular and coastal lizard the modern period of scientific study, in many cases irreversibly populations. Oikos 109, 145–153. as far as timescales of human significance are concerned. Barton, B.T., Roth, J.D., 2008. Implications of intraguild predation for sea turtle nest Millions of years ago, coastal ecosystems were dominated by protection. Biological Conservation 141, 2139–2145. reptilian predators (Section 6.09.1), but even a few centuries Bell, B.A., Spotila, J.R., Paladino, F.V., Reina, R.D., 2003. Low reproductive success of ago, reptiles remained important top-down controllers of coastal leatherback turtles, Dermochelys coriacea, is due to high embryonic mortality. Biological Conservation 115, 131–138. ecosystems. To take a single example based on the data given by Bels, V., Baussart, S., Davenport, J., Shorten, M., O’Riordan, R.M., Renous, S., Jackson et al. (2001): the green turtle C. mydas is a large herbi Davenport, J.L., 2008. Functional evolution of feeding behavior in turtles. In: vore that browses on seagrass meadows in shallow tropical and Wyneken, J., Godfrey, M., Bels, V. (Eds.), Biology of Turtles: From Structure to – subtropical coastal waters throughout the world’s oceans. In the Strategy of Life. CRC Press, Boca Raton, FL, pp. 164 187. Bjorndal, K.A., 1985. Nutritional ecology of sea turtles. Copeia 1985, 736–751.

Caribbean, there are currently about 1 million adult green tur Bjorndal, K.A., 1997. Foraging ecology and nutrition of sea turtles. In: Lutz, P.L., Musick, tles. Estimates of adult populations before the arrival of J.A. (Eds.), The Biology of Sea Turtles. CRC Press, Boca Raton, FL, pp. 199–231. Europeans (already depleted by aboriginal exploitation) are Bjorndal, K.A., Bolten, A.A., 1988. Growth rates of immature green turtles, Chelonia varied, but are of the order of 16–33 million, indicating a mydas, on feeding grounds in the southern Bahamas. Copeia 1988, 555–564. 94–97% reduction over the last three to four centuries. A more Bjorndal, K.A., Bolten, A.A., Martins, H.R., 2000. Somatic growth model of juvenile loggerhead sea turtles Caretta caretta: duration of the pelagic phase. Marine Ecology recent estimate by McClenachan et al. (2006) has even claimed Progress Series 202, 265–272. reductions in populations of 99.67 % for Caribbean green turtles Bonnet, X., Brischoux, F., Lang, R., 2010. Highly venomous sea kraits must fight to get (and 99.73% for hawksbill turtles) since historical times their prey. Coral Reefs 29, 379–379. (roughly 15th century). In ecosystem terms, the species is func Brischoux, F., Bonnet, X., 2008. Estimating the impact of sea kraits on the anguilliform fish community (Congridae, Muraenidae, Ophichthidae) of New Caledonia. Aquatic – tionally extinct and the Caribbean situation is actually better Living Resources 21, 395–399. than in many other parts of the world. C. mydas is a selective Brischoux, F., Bonnet, X., Shine, R., 2007. Foraging ecology of sea kraits Laticauda spp. cropper of seagrasses, and its past populations would have kept in the Neo-Caledonian Lagoon. Marine Ecology Progress Series 350, 145–151. seagrass turfs much shorter than is the case today and consider Brochu, C.A., 2001. Congruence between physiology, phylogenetics and the fossil ably reduced the proportion of older, lignin-rich leaves, as well record on crocodylian historical biogeography. In: Grigg, G.C., Seebacher, F., Franklin, C.E. (Eds.), Crocodilian Biology and Evolution. Surrey, Beatty and Sons, as recycling a generally refractory food resource. Not only would Chipping Norton, NSW, pp. 9–28. this have resulted in dramatically different epiphytic assem Burke, V.J., Standora, E.A., Morreale, S.J., 1993. Diet of juvenile Kemp’s ridley and blages and detrivorous communities (with knock-on effects on loggerhead sea turtles from Long Island, New York. Copeia 1993, 1176–1180.

Carpenter, C.C., 1966. The marine iguana of the Galapagos Islands, its behavior and Fry, G.C., Milton, D.A., Wassenberg, T.J., 2001. The reproductive biology and diet of physiology. Proceedings of the California Academy of Sciences 34, 329–376. sea snake bycatch of prawn trawling in Northern Australia: characteristics Casale, P., Abbate, G., Freggi, D., Conte, N., Oliverio, M., Argano, R., 2008. Foraging important for assessing the impacts on populations. Pacific Conservation Biology ecology of loggerhead turtles Caretta caretta in the central Mediterranean Sea: evidence 7, 55–73. for a relaxed life history model. Marine Ecology Progress Series 372, 265–276. Galina-Tessaro, P., Ortega-Rubio, A., Romero-Schmidt, H., Blázquez, C., 1997. Castilla, A.M., Vanhooydonck, B., Catenazzi, A., 2008. Feeding behaviour of the September diet and reproductive state of Uta stansburiana (Phrynosomatidae) at Isla Columbretes lizard Podarcis atrata, in relation to Isopoda (Crustaceae) species: Ligia San Roque, Baja California Sur, México. Journal of Arid Environments 37, 65–70. italica and Armadillo officinalis. Belgian Journal of Zoology 138, 146–148. Gyuris, E., 1994. The rate of predation by fishes on hatchlings of the green turtle Catenazzi, A., Donnelly, M.A., 2007. The Ulva connection. Marine green algae subsidize (Chelonia mydas). Coral Reefs 13, 137–144. terrestrial consumers in coastal Peru. Oikos 116, 75–86. Hays, G.C., 2001. The implications of adult morphology for clutch size in the flatback Chan, E.H., 2006. Marine turtles in Malaysia: on the verge of extinction? Aquatic turtle (Natator depressus). Journal of the Marine Biological Association of the United Ecosystems Health and Management 9, 175–184. Kingdom 81, 1063–1064. Chatterjee, S., Templin, R.J., 2004. Posture, Locomotion, and Paleoecology of Hazard, L.C., Shoemaker, V.H., Grismer, L.L., 1998. Salt gland secretion by an intertidal Pterosaurs. Geological Society of America Special Paper. Geological Society of lizard, Uta tumidarostra. Copeia 1998, 231–234. America, Boulder, CO, 64 pp. Heatwole, H., 1999. Sea Snakes. University of New South Wales Press, Sydney, NSW. Clark, W.S., 1982. Turtles as a food source of nesting bald eagles in the Chesapeake Bay Heithaus, M.R., Wirsing, A.J., Thomson, J.A., Burkholder, D.A., 2008. A review of lethal region. Journal of Field Ornithology 53, 49–51. and non-lethal effects of predators on adult marine turtles. Journal of Experimental Coker, R.E., 1906. The natural history and cultivation of the diamond-back terrapin with Marine Biology and Ecology 356, 43–51. notes on other forms of turtles. North Carolina Geological Survey Bulletin 14, 1–69. Hill, M.S., 1998. Spongivory on Caribbean reefs releases corals from competition with Davenport, J., 1988. The turtle industry of Bali. British Herpetological Society Bulletin sponges. Oecologia 117, 143–150. 25, 16–24. Houghton, J.D.R., Doyle, T.K., Wilson, M.W., Davenport, J., Hays, G.C., 2006. Jellyfish Davenport, J., 1992. The biology of the diamondback terrapin Malaclemys terrapin aggregations and leatherback turtle foraging patterns in a temperate coastal (Latreille). Testudo 3, 21–32. environment. Ecology 87, 1967–1972. Davenport, J., 1998. Sustaining endothermy on a diet of cold jelly: energetics of the Hussain, I.S., 2003. Olive ridley sea turtles in Porto-Novo, Tamil Nadu, India, with an leatherback turtle Dermochelys coriacea. British Herpetological Society Bulletin 62, observation of an Asian giant softshell turtle. Marine Turtle Newsletter 101, 25. 4–8. Ineich, I., Bonnet, X., Brischoux, F., Kulbicki, M., Shine, R., Séret, B., 2007. Anguilliform Davenport, J., Andrews, T.J., Hudson, G., 1992a. Assimilation of energy, protein and fishes and sea Kraits: neglected predators in coral-reef ecosystems. Marine Biology fatty acids by the spectacled caiman Caiman crocodilus crocodilus L. Herpetological 151, 793–802. Journal 2, 72–76. Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque, B.J., Davenport, J., Dellinger, T., 1995. Melanism and foraging behaviour in an intertidal Bradbury, R.H., Cooke, R., Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S., population of the Madeiran lizard Podarcis (=Lacerta) dugesii (Milne-Edwards, Lange, C.B., Lnihan, H.S., Pandolfi, J.M., Peterson, C.H., Steneck, R.S., Tegner, 1829). Herpetological Journal 5, 200–203. M.J., Warner, R.R., 2001. Historical overfishing and the recent collapse of coastal Davenport, J., Glasspool, A.F., Kitson, L., 2005. Occurrence of diamondback terrapins, ecosystems. Science 293, 629–638. Malaclemys terrapin, on Bermuda: native or introduced? Chelonian Conservation James, M.C., Davenport, J., Hays, G.C., 2006. Expanded thermal niche for a diving and Biology 4, 956–959. vertebrate: a leatherback turtle diving into near-freezing water. Journal of Davenport, J., Grove, D.J., Cannon, J., Ellis, T.R., Stables, R., 1990. Food capture, Experimental Marine Biology and Ecology 335, 221–226. appetite, digestion rate and efficiency in hatchling and juvenile Crocodylus porosus Jayne, B.C., Voris, H.K., Ng, P.K.L., 2002. Snake circumvents constraints on prey size. Schneider. Journal of Zoology 220, 569–592. Nature 418, 143. Davenport, J., Hills, J., Glasspool, A.F., Ward, J., 2001. Threats to the endangered Jayne, B.C., Ward, T.J., Voris, H.K., 1995. Morphology, reproduction, and diet of the marine endemic Bermudian skink, Eumeces longirostris. Oryx 35, 332–339. homalopsine snake Bitia hydroides in peninsular Malaysia. Copeia 1995, 800–808. Davenport, J., Macedo, E., 1990. Behavioural osmotic control in the euryhaline Ji, Q., Ji, S.-A., Cheng, Y.-N., You, H.-L., Lü, J.-C., Liu, Y.-Q., Yuan, C.-X., 2004. diamondback terrapin Malaclemys terrapin Latreille: responses to low salinity and Pterosaur egg with leathery shell. Nature 432, 572. rainfall. Journal of Zoology 220, 487–496. Jones, T.T., Hastings, M.D., Bostrom, B.L., Pauly, D., Jones, D.R., 2011. Growth of Davenport, J., Other, 1995. The marine ecology of the Laguna San Rafael (Southern captive leatherback turtles, Dermochelys coriacea, with inferences on growth in the Chile): ice scour and opportunism. Estuarine, Coastal and Shelf Science 41, 21–37. wild: implications for population decline and recovery. Journal of Experimental Davenport, J., Spikes, M., Thornton, S.M., Kelly, B.O., 1992b. Crab-eating in the Marine Biology and Ecology 399, 84–92. diamondback terrapin Malaclemys terrapin (Latreille); dealing with dangerous prey. Kear, B.P., Boles, W.E., Smith, E.T., 2003. Unusual gut contents in a Cretaceous Journal of the Marine Biological Association of the United Kingdom 72, 835–848. ichthyosaur. Proceedings of the Royal Society of London Series B270 (supplement), Davenport, J., Ward, J.F., 1993. The effects of salinity and temperature on appetite in the S206–S208. diamondback terrapin Malaclemys terrapin (Latreille). Herpetological Journal 3, Kerford, M.R., Wirsing, A.J., Heithaus, M.R., Dill, L.M., 2008. Danger on the rise: diurnal 95–98. tidal state mediates an exchange of food for safety by the bar-bellied sea snake De Lisle, H.F., 2007. Observations on Varanus s. salvator in North Sulawesi. Biawak 1, Hydrophis elegans. Marine Ecology Progress Series 358, 289–294. 59–66. Lafferty, K.D., Page, C.J., 1997. Predation on the endangered tidewater goby, Donlan, E.M., Townsend, J.H., Golden, E.A., 2004. Predation of Caretta caretta Eucyclogobius newberryi, by the introduced African clawed frog, Xenopus laevis, (Testudines: Cheloniidae) eggs by larvae of Lanelater sallei (Coleoptera: Elateridae) with notes on the frog’s parasites. Copeia 1997, 589–592. on Key Biscayne, Florida. Caribbean Journal of Science 40, 415–420. Laurie, W.A., 1989. Effects of the 1982–1983 El Niño-Southern oscillation event on Dorcas, M.E., Wilson, J.D., Gibbons, J.W., 2007. Crab trapping causes population marine iguanas (Amblyrhynchus cristatus, Bell, 1825) populations in the Galápagos decline and demographic changes in diamondback terrapin over two decades. islands. In: Glynn, P. (Ed.), Global Ecological Consequences of the 1982–1983 El Biological Conservation 137, 334–340. Niño-Southern Oscillation. Elsevier, New York, NY, pp. 121–141. Doyle, T.K., Houghton, J.D.R., McDevitt, R., Davenport, J., Hays, G.C., 2007. The energy Leon, Y.M., Bjorndal, K.A., 2002. Selective feeding in the hawksbill turtle, an density of jellyfish: estimates from bomb-calorimetry and proximate-composition. important predator in coral reef ecosystems. Marine Ecology Progress Series Journal of Experimental Marine Biology and Ecology 34, 239–252. 245, 249–258. Draud, M.M., Bossert, M., Zimnavoda, S., 2004. Predation on hatchling and juvenile Li, M., Fry, B.G., Kini, R.M., 2005. Eggs-only diet: its implications for the toxin profile diamondback terrapins (Malaclemys terrapin) by the Norway rat (Rattus norvegicus). changes and ecology of the marbled sea snake (Aipysurus eydouxii). Journal of Journal of Herpetology 38, 467–470. Molecular Evolution 60, 81–89. Dunson, W.A., 1974. Salt gland secretion in a mangrove monitor lizard. Comparative Lillywhite, H.B., 1991. The biology and conservation of acrochordid snakes. Hamadryad Biochemistry and Physiology 47A, 1245–1255. 16, 1–9. Duron, M., 1978. Contribution à l’étude de la biologie de Dermochelys coriacea Limpus, C.J., 1971. The flatback turtle, Chelonia depressa Garman in southeast (Linné) dans les Pertuis Charentais. Ph.D. Thesis, University of Bordeaux, Queensland, Australia. Herpetologica 27, 431–446. Talence, France. Limpus, C.J., Limpus, D.J., 2000. Mangroves in the diet of Chelonia mydas in Elliott, A.B., Karunakaran, L., 1974. Diet of Rana cancrivora in fresh water and brackish Queensland, Australia. Marine Turtle Newsletter 89, 13–15. water environments. Journal of Zoology (London) 174, 203–215. Limpus, C.L., Parmenter, C.J., Baker, V., Fleay, A., 1983. The flatback turtle, Chelonia Finke, D.L., Denno, R.F., 2004. Predator diversity dampens trophic cascades. Nature depressa, in Queensland: post-nesting migration and feeding ground distribution. 429, 407–410. Australian Wildlife Research 10, 557–561.

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5 Author's personal copy 248 High-Trophic-Level Consumers: Trophic Relationships of Reptiles and Amphibians of Coastal and Estuarine Ecosystems

Lobo, A.S., Vasudevan, K., Pandav, B., 2005. Trophic ecology of Lapemis curtus Priede, I.G., 1990. The sea snakes are coming. New Scientist 1742, 29–33. (Hydropiinae) along the west coast of India. Copeia 2005, 637–641. Pritchard, P.C.H., 1979. Encyclopaedia of Turtles. TFH Publications, Hong Kong. Loveridge, A., Williams, E.E., 1957. Revision of the African tortoises and turtles of the Rasmussen, A.R., 2001. Sea snakes. In: Carpenter, K.E., Niem, V.H. (Eds.), The Living suborder Cryptodira. Bulletin of the Museum of Comparative Zoology Harvard 115, Marine Resources of the Western Central Pacific, Volume 6, FAO Species 163–557. Identification Guide for Fishery Purposes. Food and Agriculture Organization of the Lutcavage, M., Lutz, P.L., 1986. Metabolism and feeding energetics of the leatherback United Nations, Rome, pp. 3987–4008. sea turtle. Copeia 1986, 796–798. Read, J., Moseby, K., 2006. Vertebrates of Tetepare Island, Solomon Islands. Pacific MacDonald, D.W., Brown, L., Yerli, S., Canbolat, A.-F., 1994. Behaviour of red foxes, Science 60, 69–79. Vulpes vulpes, caching eggs of loggerhead turtles, Caretta caretta. Journal of Rhodin, R.G.J., 1985. Comparative chondro-osseous development and growth of marine Mammalogy 75, 985–988. turtles. Copeia 1985, 752–771. Martin, J.E., 2007. A new species of durophagous mosasaur, Globidens (Squamata: Ross, J.P., 1985. Biology of the green turtle, Chelonia mydas, on an Arabian feeding Mosasauridae) from the Late Cretaceous Pierre Shale Group of central South Dakota, ground. Journal of Herpetology 19, 459–468. U.S.A. In: Martin, J.E., Parris, D.E., (Eds.), The Geology and Palaeontology of the Rothschild, B.M., Martin, L.D., Schulp, A.S., 2005. Sharks eating mosasaurs, dead or Late Cretaceous Marine Deposits of the Dakotas. Geological Society of America alive? Netherlands Journal of Sea Research 84, 335–340. Special Paper. Geological Society of America, Boulder, CO, pp. 167–176. Roy, K., 2000. “Amblyrhynchus cristatus”, Animal Diversity Web. http://animaldiversity. Maros, A., Louveaux, A., Godfrey, M.H., Girondot, M., 2003. Scapteriscus didactylus ummz.umich.edu/site/accounts/information/Amblyrhynchus_cristatus.html (Orthoptera, Gryllotalpidae), predator of leatherback turtle eggs in French Guiana. (accessed August 2010). Marine Ecology Progress Series 249, 289–296. Russell, D.J., Balazs, G.H., 2009. Dietary shifts by green turtles (Chelonia mydas) in the Masunaga, G., Kosuge, T., Asai, N., Ota, H., 2007. Shark predation of sea snakes K¯ane‘ohe Bay region of the Hawaiian Islands: a 28-year study. Pacific Science 63, (Reptilia: Elaphidae) in the shallow waters around the Yaeyama Islands of the 181–192. southern Ryukyus, Japan. JMBA2- Biodiversity Records. http://www.mba.ac.uk/ Sachs, J.P., Sachse, D., Smittenberg, R.H., Zhang, Z., Battisti, D.S., Golubic, S., 2009. jmba/pdf/5970.pdf (accessed August 2010). Southward movement of the Pacific intertropical convergence zone AD1400–1850. Mazin, J.-M., Pinna, G., 1993. Palaeoecology of the armoured placodonts. Paleontologia Nature Geoscience 2, 519–525. Lombarda, N.S. 2, 83–91. Seminoff, J.A., Resendiz, A., Nichols, W.J., 2002. Diet of east Pacific green turtles McClenachan, L., Jackson, J.B.C., Newman, M.J.H., , 2006. Conservation implications (Chelonia mydas) in the central Gulf of California, México. Journal of Herpetology of historic sea turtle nesting beach loss. Frontiers in Ecology and the Environment 4, 36, 447–453. 290–296. Seney, E.E., Musick, J.A., 2007. Historical diet analysis of loggerhead sea turtles (Caretta McCosker, J.E., 1975. Feeding behaviour of Indo-Australian Hydrophiidae. In: Dunson, caretta) in Virginia. Copeia 2007, 478–489. W.A. (Ed.), The Biology of Sea Snakes. University Park Press, Baltimore, MD, Shaver, D.J., 1991. Feeding ecology of wild and head-started Kemp’s ridley sea turtles in pp. 217–232. south Texas waters. Journal of Herpetology 25, 327–334. McGowan, C., 1974. A revision of the longipinnate ichthyosaurs of the Lower Jurassic of Shetty, S., Shine, R., 2002a. Philopatry and homing behavior of sea snakes England, with descriptions of two new species (Reptilia, Ichthyosauria). Life Science (Laticauda colubrina) from two adjacent islands in Fiji. Conservation Biology Contributions of the Royal Museum of Ontario 97, 1–37. 16, 1422–1426. McHenry, C.R., Cook, A.G., Wroe, S., 2005. Bottom-feeding plesiosaurs. Science 310, Shetty, S., Shine, R., 2002b. Sexual divergence in diets and morphology in Fijian sea 75. snakes Laticauda colubrina (Laticaudinae). Austral Ecology 27, 77–84. Meylan, A., 1988. Spongivory in hawksbill turtles: a diet of glass. Science 239, Shine, R., Harlow, P.S., Keogh, J., 1996. Commercial harvesting of giant lizards: the 393–395. biology of water monitors (Varanus salvator) in southern Sumatra. Biological Miller, J.D., 1997. Reproduction in sea turtles. In: Lutz, P.L., Musick, J.A. (Eds.), The Conservation 77, 125–134. Biology of Sea Turtles. CRC Press, Boca Raton, FL, pp. 51–81. Shine, R., Shetty, S., 2001. Moving in two worlds: aquatic and terrestrial locomotion in Moll, D., Moll, E.O., 2004. The ecology, exploitation and conservation of river turtles. sea snakes (Laticauda colubrina, Laticaudidae). Journal of Evolutionary Biology 14, Oxford University Press, New York, NY. 338–346. Molnar, R.E., 2004. The long and honorable history of monitors and their kin. In: Simpfendorfer, C.A., Goodreid, A.B., McAuley, R.B., 2001. Size, sex and geographic Pianka, E.R., King, D.R., King, R.A. (Eds.), Varanoid Lizards of the World. variation in the diet of the tiger shark, Galaeocerdo cuvier, from Western Australian Indiana University Press, Bloomington and Indianapolis, pp. 10–67. waters. Environmental Biology of Fishes 61, 37–46. Mortimer, J.A., 1981. The feeding ecology of the west Caribbean green turtle (Chelonia Snover, M.L., Rhodin, R.G.J., 2008. Comparative ontogenetic and phylogenetic aspects mydas) in Nicaragua. Biotropika 131, 49–58. of chelonian chondro-osseous growth and skeletochronology. In: Wyneken, J., Mortimer, J.A., 1982. Feeding ecology of sea turtles. In: Bjorndal, K.A. (Ed.), Biology and Godfrey, M., Bels, V. (Eds.), Biology of Turtles: From Structure to Strategy of Life. Conservation of Sea Turtles. Smithsonian Institution Press, Washington, DC, CRC Press, Boca Raton, FL, pp. 17–43. pp. 103–109. Spotila, J.R., 2004. Sea Turtles: A Complete Guide to their Biology, Behavior and Murphy, J.C., Voris, H.K., 2002. Aquatic snakes with crustacean-eating habits elude Conservation. The Johns Hopkins University Press, Baltimore, MD. herpetologists for two centuries. Litteratura Serpentium 22, 107–114. Spotila, J.R., Reina, R.D., Steyermark, A.C., Plotkin, P.T., Paladino, F.V., 2000. Pacific Naish, D., 2004. Fossils explained 48: Placodonts. Geology Today 20, 153–158. leatherback turtles face extinction. Nature 405, 288–290. Nutaphand, W., 1979. The Turtles of Thailand. Siam Farm Zoological Gardens, Steinfartz, S., Glaberman, S., Lanterbeq, D., Marquez, C., Rassman, K., Caccone, A., Bangkok. 2007. Genetic impact of a severe El Niño event on Galápagos marine iguanas Olson, S.L., Wingate, D.B., 2006. A new species of night-heron (Ardeidae: Nyctanassa) (Amblyrhynchus cristatus). PLoS ONE 2 (12), e1285. from Quaternary deposits on Bermuda. Proceedings of the Biological Society of Stewart, K.R., Wyneken, J., 2004. Predation risk to loggerhead hatchlings at a Washington 119, 326–337. high-density nesting beach in southeast Florida. Bulletin of Marine Science 74, Parham, J.F., Outerbridge, M.E., Stuart, B.L., Wingate, D.B., Erlenkeuser, H., 325–335. Pappenfus, T.J., 2008. Introduced delicacy or native species? A natural origin Tamarack, J.L.,1988. Georgia’s coastal island alligators, variations and habitat and prey of Bermudian terrapins supported by fossil and genetic data. Biology Letters 4, availability. Proceedings of the Eighth Working Meeting of the Crocodile Specialist 216–219. Group. IUCN, Gland, Switzerland, pp. 105–118. Parris, L.B., Lamont, M.M., Carthy, R.R., 2002. Increased incidence of red imported fire Taplin, L.E., Grigg, G.C., Harlow, P., Ellis, T.M., Dunson, W.A., 1982. Lingual salt glands ant (Hymenoptera: Formicidae) presence in loggerhead sea turtle (Testudines: in Crocodylus acutus and C. johnstoni and their absence from Alligator Cheloniidae) nests and observations of hatchling mortality. Florida Entomologist 85, mississipiensis and Caiman crocodilus. Journal of Comparative Physiology B 149, 514–517. 43–47. Pitman, R.L., Dutton, P.H., 2004. Killer whale predation on a leatherback turtle in the Taskavak, E., Akcinar, S.C., 2009. Marine records of the Nile soft-shelled turtle, Trionyx northeast Pacific. Pacific Science 58, 497–498. triunguis. JMBA2-Biodiversity Records. e9 doi:10.107/S1755267208000092. Platt, S.G., Thorbjarnson J.B., 2000. Population status and conservation of Morelet’s Tchernov, E., Rieppel, O., Zaher, H., Polcyn, M.J., Jacobs, L.L., 2000. A fossil snake with crocodile. Biological Conservation 96, 21–29. limbs. Science 287, 2010–2012. Plotkin, P., 1989. Feeding ecology of the loggerhead turtle in the northwestern Gulf of Thorbjarnarson, J., 1989. Ecology of the American crocodile, Crocodylus acutus. In: Mexico. In: Eckert, S.A., Eckert, K.L., Richardson, T.H. (Eds.), Proceedings of the Crocodiles. Their Ecology, Management and Conservation. A Special Publication of Ninth Annual Workshop on Sea Turtle Conservation and Biology. US Department of the Crocodile Specialist Group. IUCN, Gland, Switzerland, pp. 228–258. Commerce, National Oceanic and Atmospheric Administration, Washington, DC, Toland, B., 1991. Great horned owl predation of Atlantic loggerhead turtle hatchlings. pp. 139–141. Florida Field Naturalist 19, 117–119.

Tschanz, K., 1989. Lariosaurus buzzii n.sp. from the Middle Triassic of Monte San Wang, X., Kellner, A.W.A., Zhou, Z., Campos, D.A., 2005. Pterosaur diversity and Giorgio (Switzerland) with comments on the classification of nothosaurs. faunal turnover in Cretaceous terrestrial ecosystems in China. Nature 437, Palaeontographica 208, 153–179. 875–879. Tu, M.C., Fong, S.C., Lue, K.Y., 1990. Reproductive biology of the sea snake, Laticauda White, W.T., Platell, M.E., Potter, I.C., 2004. Comparisons between the diets of four semifasciata, in Taiwan. Journal of Herpetology 24, 119–126. abundant species of elasmobranchs in a subtropical embayment: implications for Tucker, A.D., Fitzimmons, N., Gibbons, J.W., 1995. Resource partitioning by the resource partitioning. Marine Biology 144, 439–448. estuarine turtle Malaclemys terrapin; trophic, spatial, and temporal foraging Whiting, A.U., Thomson, A., Chaloupa, M., Limpus, C.J., 2008. Seasonality, abundance constraints. Herpetologica 51, 167–181. and breeding biology of one of the largest populations of nesting flatback turtles, Tucker, J.K., Paukstis, G.L., Janzen, F.J., 2008. Does predator swamping promote Natator depressus: Cape Domett, Western Australia. Australian Journal of Zoology synchronous emergence of turtle hatchlings among nests? Behavioral Ecology 19 56, 297–303. (1), 35–40. Wikelski, M., Carrillow, V., Trillmich, F., 1997. Energy limits to body size in a grazing Velasco, A., Trejo, V., Zapata, I., 1994. Stomach contents of Caiman crocodilus from the reptile, the Galapagos marine iguana. Ecology 78, 2204–2217. Orinoco delta of Venezuela. Crocodile Specialist Group Newsletter 13 (3), 20–21. Wikelski, M., Gall, B., Trillmich, F., 1993. Ontogenetic changes in food intake and Vidal, N., Hedges, S.B., 2004. Molecular evidence for a terrestrial origin of snakes. digestion rate of the herbivorous marine iguana (Amblyrhynchus cristatus, Bell). Proceedings of the Royal Society of London B271 (supplement), S226–S229. Oecologia 94, 373–379. Vinueza, L.R., Branch, G.M., Branch, M.L., Bustamante, R.H., 2006. Top-down herbivory Wikelski, M., Thom, C., 2000. Marine iguanas shrink to survive El Niño. Nature 403, and bottom-up El Niño effects on Galápagos rocky-shore communities. Ecological 37–38. Monographs 76, 111–131. Wikelski, M., Wrege, P.H., 2000. Niche expansion, body size, and survival in Galápagos Voris, H.K., 1972. The role of sea snakes (Hydrophiidae) in the trophic structure of marine iguanas. Oecologia 124, 107–115. coastal ocean communities. Journal of the Marine Biological Association of India 14, Wirsing, A.J., Heithaus, M.R., Dill, L.M., 2007. Can measures of prey availability 1–14. improve our ability to predict the abundance of large marine predators? Oecologia Voris, H.K., Glodek, G.S., 1980. Habitat, diet and reproduction of the file snake, 153, 563–568. Acrochordus granulatus, in the Straits of Malacca. Journal of Herpetology 14, Witt, M.J., Baert, B., Broderick, A.C., Formia, A., Fretey, J., Gibudi, A., Avery, G., 108–111. Mounguengui, M., Moussounda, C., Ngouessono, S., Parnell, R.J., Roumet, D., Voris, H.K., Voris, H.H., 1983. Feeding strategies in marine snakes: an analysis of Sounguet, G.-P., Verhage, B., Zogo, A., Godley, B.J., 2009. Aerial surveying of the evolutionary, morphological, behavioral and ecological relationships. American world’s largest leatherback turtle rookery: a more effective methodology for large- Zoologist 23, 411–425. scale monitoring. Biological Conservation 142, 1719–1727. Walker. T.A., Parmenter, C.J., 1990. Absence of a pelagic phase in the life cycle of Witzell, W.N., Schmidt, J.R., 2005. Diet of immature Kemp’s ridley turtles (Lepidochelys the flatback turtle, Natator depressa (Garman). Journal of Biogeography 17, kempi) from Gullivan Bay, Ten Thousand Islands, Southwest Florida. Bulletin of 275–278. Marine Science 77, 191–199.

Treatise on Estuarine and Coastal Science, 2011, Vol.6, 227-249, DOI: 10.1016/B978-0-12-374711-2.00618-5