Quick viewing(Text Mode)

Synopsis of the Biological Data on the Green Turtle

Synopsis of the Biological Data on the Green Turtle

BIOLOGICAL REPORT 97(1) AUGUST 1997

Synopsis of the Biological Data on the Green

Harold F. Hirth

Department of Biology University of Utah Salt Lake City, Utah 841 12 USA

Fish and Wildlife Service U. S. Department of the Interior Washington, D. C. 20240 Preparation of this Synopsis

This review of the green turtle, Chelonia mydas. has mitt= aided travel to libraries. Ms. Jeanette Stubbe gra- been prepared following the FA0 fisheries synopsis out- ciously and conscientiously typed several versions of the line of Rosa (1965) and as applied to marine by manuscript. Mr. Keny Matz prepared the figures. Dr. ffirth (197Ib). John Roth; Chairman of the Biology Department, sup- The main purposes of this synopsis am to bring together parted the author with. some sabbatical leave time. the current and salient information on the biology of the Itis a pleasure to acknowledge Dr. David W. Ehrenfeld, green turtle and to draw attention to some of the major Dr. Nicholas Mrosovsky, Dr. Mark Nielsen and Dr. Peter gaps in our knowledge of the . Because of the C. H. Pritchaid for their helpful comments after review- nature of a synopsis, i.e., that of providing an entry into ing an earlier version of the manuscript the literature, researchers should peruse die original pa- The author'thanks Drs. Leslie Dieraof and Richard pers for details of methodologies and conclusions. Byles, Ms. Susan MacMullin and Mr. Art Needleanan of The author is indebted to Ms. Linda Bums, Imerlibrary the Endangered Species Office,U. S. Fish and Wildlife Loan Supervisor, University of Utah, for helpful and Service, Albuquerque, New Mexico for help and for sup timely assistance in obtaining some rare publications. porting the publication of this synopsis. Assistance from the University of Utah Research Corn- I

ii

--- Abstract

This document reviews the salient and current litera- studies on a few nesting beaches have disclosed signifi- ture on the biology of the green turtle, Chelonia my& cant annual fluctuations in numbers of nesters. Under (Linnaeus, 1758) including taxonmy, distribution,phy &- natural conditions there is high predation on eggs and ~logy,morphology, ecology, demography, exploitation hatchlings and low predarion on adults. Long life was and conservation. Fifteen figures and 17 tables supple- probably the norm for those that survived neonatal mor- ment the text. tality. before the advent of humans. A green turtle survi- In general, green turtles arelaige sea turtles well adapted vorship curve is roughly concave, under natural condi- to marine life. They are cireumglobal, conanonly occur- tions. Green turtles are host to a large variety of para- ring in warm, tropical seas. They occur in offshore wa- sites, and fibropapillomatosis is a significant disease in ters or on the nesting beaches of at least 139 countries ~inewidely scattered areas. Major food competitors on and terriiories. Most nesting sites are located between the seagrass pastures are dugongs, fishes and sea urchins. 30ÂN and 30Â S latitudes. The green turtle is a morpho- As expected, studies have shown how many aspects of species, made-up of several distinct populations and the turtles' physiology are related to their feeding habits, meiapopulations. The total range of a population~en- reproductive cycles, prolonged swimming, diving and compassing the nesting beach, epipelagic , feed- migrations. References are provided on morphological ing grounds and migrations~canbe very extensive. descriptions of the embryo, egg shell, skull, lung, kidney. Sex is determined by substrate temperatures during in- ovary and oviduct. cubation with wanner temperatures producing females, Some major gaps in our knowledge of green turtles are The diploid chroniosome number is 56 and there are no speciation rates, natural sex ratios, ecologies of hatchlings heteromorphic sex chromosomes. Hatchlings use visual and juveniles during the "lost years", biology of males, cues in crawling to the sea and then, in shallow water, survival rates of different size classes, and navigation they orient by swimming into the waves. Magnetic cues mechanisms. Obtaining information on some of these may be used for orientation in deep water. The cues used parameters can be aided by the development of reliable by navigating adult turtles in their long-distance gametic marking and tracking systems. migrations are unknown. Satellite telemetry may prove Because of many decades of overexploitation by hu- useful in this regard. Recent mWNA research supports-a roans, most green turtle populations are endangered or natal homing hypothesis. Hatchlings and small juveniles threatened today. Degradation of nesting beaches and are chiefly carnivorous (or omnivorous) while subadults oceanic pollution are additional threats to green turtle and adults are chiefly herbivorous. Trophic level changes survival almost everywhere now. Conservation of any are associated with ontogenetic habitat shifts. one population will almost certainly involve regional co* Much more is known about females than males because operation. All populations are important because they the former are easily studied on the nesting beaches. Green are the evolving units in nature and because they repre- turtles are characterized by slow growth, delayed sexual sent genetic diversity. For conservation purposes, each maturity, high fecundity, iteroparity, and a relatively long green turtle nesting population should be viewed as an reproductive life (under natural conditions). Reproduc- autonomous demographic entity. Preservation of the tive data, from many nesting sites, are provided in tabular turtles' critical , education, and enforcement of form: sizes ofnesters, clutch sizes and number of clutches existing protective regulations are among (he management per season, egg and hatchling dimensions, remigration strategies discussed. intervals and hatching success. Long-range demographic

ill Contents

Preparation of &is Synopsis ...... ii Abstract ...... ,...... *.,...... 111

1.1 Nomenclature ...... ; ...... *...... ‘...... 1 1.1.1 Valid name ...... 1 1.1.2 Synonymy ...... ã...... ã...... 1 1.2 ...... 2 1.2.1 Afiimties ...... ; ...... 2 1.2.2 Taxonomic status ...... *....2 1.2.3 Subspecies .....-...... 2 1.2.4 Standard common names ...... -...... 5 1.2.5 Definition of size categories ...... ; ...... 6 1.3 Morphology ...... ; ...... 6 1.3.1 External/internal morphology and coloration ...... 6 1.3.2 Cytomorphology ...... : ...... 11 1.3.3 Protein composition and specificity and general physiology ...... 12 DIS.TMBTJTION ...... 13 2.1 TotalArea ...... 13 2.2 Differential Distribution ...... 14 2.2.1 Hatchlings ...... 14 2.2.2 Juveniles. subadults. and adults ...... 14 23 Determinants of Distributional Changes ...... 25 2.4 Hybridization ...... :...... :...... ,...... 25 BIONOMICS AND LIFE HISTORY ...... 25 Reproduction ...... 25 3.1.1 . Sexuality ...... 25 3.1.2 Maturity ...... 26 3.1.3 Maring ...... 26 3.1.4 Fertilization...... 29 3.1.5 Gonads ...... 29 3.1.6 Nesting process ...... 30 3.1.7 Eggs ...... 30 Embryonic and Hatchling Phase ...... ã...... 39 3.2.1 Embryonic phase ...... 39 3.2.2 Hatchling phase ...... 39 Juvenile. Subadult, and Adult Phase ...... -....-...... 43 3.3.1 Longevity ...... 43 3.3.2 Hardiness ...... 44 3.3.3 Competitors ...... 45 3.3.4 Predators ...... 46 3.3.5 Parasites, commensals and diseases ...... 46 Nutrition. . and Growth ...... 55 3.4.1 Feeding ...... 55 3.4.2 Food ...... 56 3.4.3 Growth rate ...... 62 3.4.4 Metabolism ...... 62 3.5 Behavior ...... 64 3.5.1 Migrations and local movements ...... 64 3.52 Schooling ...... 71 3.5.3 Responses to stimuli ...... 71. 4 . POPULATION ...... 72 4.1 Structure ...... 72 4.1.1 Sex ratio ...... 72 4.1.2 Age composition...... 72 4.1.3 Size composition...... -...... 72 4.2 Abundance and Density ...... + ...... 72 4.2.1 Average abundance and density ...... *...... 72 4.22 Changes. in abundance and density ...... 72 4.3 Natality and Recruitment ...... 73 4.3.1 Reproduction rates ...... 73 4.3.2 Factors affectingreproduction ...... 73 4.3.3 Recruitment ...... 75 4.4 Mortality ...... '...... 75 4.4.1 Mortality rates ...... 75 4.4.2 Factors causing or affectingmortality ...... 75 4.5 Dynamics of Populations...... 77 4.6 The Population in the Community and the Ecosystem ...... 79 5 . EXI^OITAIION ...... 79 5.1 Fishing Equipment and Methods ...... 79 5.2 Fishing Areas ...... 80 5.3 Fishing Seasons ...... 80 5.4 Fishing Operations and Results ...... w...+...... $0 6 . PROTECTION AND MANAGEMENT ...... 82 6.1 Regulatory Measures ...... 82 6.2 Management Strategies ...... 84 7 . MARICULTURE ...... : ...... 87 8. REFERENCES .....: ...... 88 1. IDENTITY ChelorUa lachrymauiCu4ac, 1829: 13. Type-locality, not 1.1 Nomenclature stated. Holotype, possibly in Mus. Hat. Hist. Natur., Paris I. I. 1 Valid name (Roux, pers. conun.1. Chelonia mydas (Linnaeus), 1758 1.I.2 Synonymy Chelonh maculosa Cuvier, 1829: 13. Type-locality, riot stated, restricted to "Ascension Island" by Smith and Tay- lor (1950: 17). Holotype, possibly in Mus, Nat Hist. mydas Linnaeus, 1758: 197. Type-locality Natur. Paris (Roux, pers. corn.). "insulam Adscensionis". Chelonia bicarinata Lesson, 1834: 301. Type-locality. "1 Testudo macropus Walbaum. 1782: 112. Type-locality 'Oc6an atlantique." Holotype, possibly in Mus. Nat. Hist. not stated. Holotype not designated. (illegitimate name). Natur.. Paris (Roux. pers. conun.).

Testudo viridis Schneider, 1783: 299. Type-locality un- Chelonia marmorata Dum6ril and Bibron, 1835: 546. known,restricted to Charleston, South Carolina by Smith Type-locality, Ye de 1'Ascension." Holotype, Mus. Nat. and Taylor (1 950: 17). Holotype not designated. Hist. Natur. Paris 7878.

Tesrudo japonica Thunberg, 1787: 178. Type-locali ty Euchdys macmpus Girard, 1858: 448 (=Te-stud0 mydas "Japan". Holotype not designated. Linnaeus 1758) by monotypy.

Testudo marina vaaris Lac6pfede, 1788: 54 and Table. Chelonia formosa Girard, 1858: 456. Type-locality, (illegitimate name; substitute name for Testudo mydas "Feejee Islands". Holotype, U. S. Nat. Mus. 12386, Linnaeus, 1758). adult carapace, Fiji Islands. U. S. Exploring Expedi- tion, 1840. Testudo viridi-squamosa Lac6pMe. 1788: 92 and Table. (illegitimate name). Chelonia tennis Girard, 1858: 459. Type-locality, "Honden Island, Paumotu Group; Tahiti and Eimo; Rosa Chdonia mydas Brongniart, 1800: 89 (see generic syn- Island." Holotype, U. S. Nat. Mus. 12390, male cara- onymy 1.2.1) pace, Rosa Island (=Rose Atoll).

Testudo rugosa Daudin, 1801: 37. Type-localily, "lamer Chelonia albtventer Nardo, 1864: 1420. Type-locality, des Indes ...environs trois d6grts des iles Maldives." *Adriatico.. .prossimat& del porto di Malamocco," Holo- Holotype not designated. type, Mus. Civico Storia Natur. Venezia, Italy, unnum- bered dry specimen. Testudo cepediana ~audin,1801 : 50. Type-locality, not stated. Holotype not designated. Chelonia agassizii Bocouri, 1868: 122. Type-locality. "embouchure du Nagualate.. .Pacifique (Guat6mala). Chelonia mydas Schweigger, 1 8 12: 29 1. Holoiype, Mus. Nat. Hist. Natur., Paris.9537.

Chelonia virgata Schweigger, 1812: 29 1. Type-local- Chelonia lata Pbilippi, 1887: 84. Type-locality. "Insel ity "man sub zona torrida," restricted to "Bermuda Chiloe7'Chile. Holotype, Mus. Nat. Hist Natur., Santiago Islands" by Smith and Taylor (1 950; 17). Holotype 100201. not designated. CMonia mydas carriliegru Caldwell, 1962a: 4. Type- Carem cepedii Menem, 1820: 18. (substitute name for locality, "Bahia de Los Angeles, Baja California None, Testudo cepediana Daudin, 1801). Mexico." Holotype, Los Angeles Co. Mus. 1696.

Caretta esculenta Merrem, 1820: 18. ~y~e-locality, The preceding, abbreviated synonymy is adapted from "Oceano AtIantico," Holotype not designated. Hirth (1980b) and Pritchard and Trebbau (198.4). More detailed synonymies are in Bound (1941)' Wermuth and Caretta thwibergii Merrem, 1820: 19. Type-locality, Ja- Mertens (1977). Smith and Smith (1979) and .Pritchard pan (substitute name for Testudo japonica Thunberg and Trebbau (1984). Wallin (1985) recommends a nee 1787)., type in the Stockholm Museum. 1.2 Taxonomy . to marine life (Rg. 1). The elevated carapace has juxt 1.2.1 Affinities posed scutes, is oval to heart-shaped, with four pails -Suprageneric cosrals (the first separated from the nuchal). The car Phylum Chordata pace is constrictedsharply above the hind flippers in so; Subphylum Vertebrata eastern Pacific populations. Carapace ground color w Superclass Tetrapoda ies from predominantly green to olive, or brown, or gr Class Reptilia to blade; and, wirh a varying number of blotches or strea Subclass Anapsida of yellow, green, brown, copper and black. The adult pk Testudines con varies from white to cream-yellow but in some pop Suborder ladons has various sized infusions of gray or black. T Superfamily Chelonioidea bridge has four enlarged inframarglnal scutes which la pores. The head has a pair ofprefrontal scales,and us Generic ally four postoculars. The tomiurn of the lower jaw senate while that of the upper jaw has strong vertic Chelonia Brongniart, 1800: 89. Type-species designated ridges on its inner surface. The single nail on t as Chelonia mydas Cuvier, 1832 (=Ted0 mydas foreflippers is mare elongate and curved in the male. T Linnaeus, 1758) by Fitringer, 1843: 30. strongly prehensile tail of the male is much longer th that of the female, extending well beyond the posten' Chelae Sonnini and Laireille, 1802: 22. Type species margin of the carapace (Kg. 2). The carapace and pit designated as Testudo my& Linnaeus, 1758by Fitzinger, troa of hatchlings are slale black (darker when wet) a 1843. white, respectively. Detailed descriptions are Demiyagala (1939), Can- (19521, Smith and Smi Chelonius Rafinesque, 1814 66. Emendation. (1979), Pritchard and Trebbau (1984), Ernst and Barbo (1989) and Mfcquez (1990). MyhCocteau, 1838: 22. Type-species, Testudo mydas Linnaeus, 1758 by tautonomy. 1.2.2 Taxonomic status It is the opinion of the author that the green turtle fC?. Mydasea Gervais, 1843: 457. Type-species. Tesmdo lo& mydas} is a dremnglobal, morpho-species. The sg my& Linnaeus. 1758 by monotypy. cies is made-up of several distinct populations sc- metapopulanouSmThe populations can be identified ' Euchelonia Tschudi. 1846: 22. Type-species, Test& the name of their nesting beach or beaches used in as mydas Linnaeus. 1758 by monotypy. ciation with Chelonia mydas (see following section). / the populations are important because they are the evol Megemys Gistl, 1848: 4%(substitute name for Chelm ing units in nature and because they represent genetic i Sonnini and Latrdlle 1802). versity.

Euchelys Girard, 1858: 447. Type-species, Euchelys 1.2.3 Subspecies macropus Girard, 1858: 448 (= Tedmydas Linnaeus, Several populations of green turtles have teen describe 1758) by monotypy. in morphological and biochemical tern, and some ha been given specific or subspecific names. Some of I The preceding abbreviated synonymy is adapted from more recent descriptions follow. Wermuth and Mertens (19771, Smith and Smith (19791, M&quez (1 990) recognizes the dark Cheloniu ocxs Hi& (1980a). and Cogger et al. (1983). and these ac- ring in the eastern Pacific Ocean principally from Bi counts can be consulted for details. California to Peru, and with major nesting beaches Smith and Smith (1979) recognize Brongniait as the Mexico and the Galfoagos Islands, as a species, Che, source for Chel& and they discuss uses of Sonnini and nia agassizii. la addition to the species bring smal Latreille. 1802, and Schweigger, 1812. than the typical Chelonk mydas, Mkquez (1990) li some of the diagnostic features: adult carapace oft Generic strongly vaulted; the carapace, in dorsal view, The Chelonia is monotypic, in the opinion of tile subcardiform and deeply ernarginate over the rear fl author (see following two sections). pers; the carapace width becomes relatively narrower w age; adult carapace is slate gray to black with a blotch Specific brown and olive pattern; upper surfaces of the head a Diagnosis: Medium to large size turtles well-adapted flippers dark; plastron varying from whitish-gray Fig. L A typical adult female green turtle on Ascension Island, the type locality. Standard straight line carapace length of this individual is 112 on. Females here are among the largest in the world and they migrate to the Island from Brazil, a round-trip distance of about 4,600 km.

Fig. 2. A typical adult male green turtle on Aldabra Atoll. Standard straight line carapace length of this individual is 95 cm. The size of the tail easily identifies adult males, anywhere. bluish or olive gray. Younger individuals are usually more mydas mydas be used for the Ascension Island pop1 colorful and similar to those in the Atlantic populations. lion; and, that Chelonia mydas viridis be used for Further, Mihquez (1 990) is of the opinion that Chelonia Tortuguero colony. mydas is comprised of two subspecies: C. m. mydas in The use of color as a taxonomic index must be u the Atlantic Ocean and C. rn. japonica in theIndian Ocean with caution inasmuch as Prazier (1971) illustrated m and in" the western and central Pacific Ocean. color variation within a single population at AldabraAt Alvarado and Figueroa (1 990) also consider the East- More recently, molecular techniques have been us& turtle or black turtle a species, Chelonia detennine gene flow between green untie populatk agassizi. Eight rnorphometric characters in the Mitochondrial DNA data do not support the evolution Michoacan-nesting C. agassizi population and in the distinctness of C. agassizi. The mtDNA data suggest 1 'Ibrtuguero-nesting mydas population were compared. the Chelonia complex should probably be divided i A principal component analysis of the data indicated a Atlanp-Medimean and Indian-Pacific subspec clear distinction between the two populations (Alvarado with additional population-level distinctness recogpi and Figueroa 1990). within each ocean basin (Bowen et d. 1992). Under Kamezaki and Matsui (1995) analyzed twenty cranial division, the Indian-Pacific subspecies would be nai traits of 145 green turtle skulls and four mdibular traits japonica. Based on nucleotide sequences from the q of 103 mandibles from six localities in three oceans chrome b gene of mtDNA, Bowen et al. (1993) detenni (Comoros Is., Seychelles Is., Ogasawara Is.. GaMpagos hit C my& is paraphyletic with respect to C. ap.~ Is., Guyana and Caribbean Costa Rica). The Galapagos in terms of matriarchal phylogeny. An analysis of nuc sample was completely separated from the other samples DNA indicated some genetic similarity between pop by a canonical discriminant analysis but was not differ- tions in Michoacan, Mexico, and the Gal6pagos (Ka entiated from the others by any single character dimen- al. 1992). Although nesting populations appear to be sion relative to skull length. The authors, therefore, sup- lated with respect to female inuDNA) lineages, the w port the recognition of the eastern Pacific population, in- of Kari et al. (1992) with nDNA indicates a moderase 1f cluding the Galapagos Is. population, as a distinct sub- of male-mediated gene flow. Moritz (1994a) reconnnc species, C. m. agassizi, but not as a distinct species. that the circumglobal green turtle should be manage Dutton and McDonald (1 992) found it difficult to dis- two ESUs (Evolutionary Significant Units), is. the tinguish between Chelonla agassizi and CheIoiid mydas. ]antic-Medimean and the Indo-Pacific. with each E based on carapace color and shape and plastron color, in consisting of multiple MUs (Management Units). . a small population in San Diego Bay. The D-loop nucle- cording to Moritz(1994a) "ESUs should be redproc, aide sequences of mtDNA from four San Diego Bay monophyletic for rntDNA alleles and show signific turtles were compared with those of five Hawaiian turtles divergence of allele frequencies at nuclear loci" and '% from the French Frigate Shoals colony and with three are therefore recognized as populations with signific Mexican black turtles from the Michoacan nesting colony divergence of allele frequencies at nuclear or mitoch (Dutton et al. 1994). The San Diego Bay turtles appeared drial loci, regardless of the phylogenetic distinctiver to be more like the Mexican turtles, although some dif- ."of the alleles." ferences in the mtDNA suggested that they may not have Green turtles (and other Testudines) have slower originated from the Michoacan colony. croevolutionaiy rates for mtDNA than other venebn . Pritchari and Trebbau (1984) consider the East Pacific (Avise et al. 1992). Karl and Avise (1993) discuss Cheloniu populations from the Galfipagos Islands and the advantages of using polymerase chain reaction (P( mainland shores of the Americas a distinct species, Cheto- techniques to generate population genetic data and t tzb agassat. Pritchard in Pritchard and Trebbau (1984) also use green turtle data to show how this approach is use noted occasional syrnpatty between C rnydas and C agassizi Nonnan et al. ( 1994)used PCR to evaluate sequence va in Pacific Mexico, the Gal6paaos and Papa New Guinea. don in the hypervariable control region of green tu Hendriclcson (1 980) recommended that Chelonia nydas mtDNA and concluded that the Indo-Pacific green tur caminegra be elevated to species status. The position of include a number of genetically differentiatedpopulati carrinegra in green turtle taxonomy was reviewed by with minimal female-mediated gene flow among (hi Groombridge and Luxmoore (1989). In a mini-review, Moritz (1994b) draws a distinction Earlier, Can (1975) suggested that the eastern Pacific tween the use of mtDNA analysis to identify and man form of the green turtle, extending from Baja California genetic diversity and its use as a tool for demogral to the Gatepagos Islands and Peru, and westward to the studies of populations. Dutton and Balazs (1995) brii Hawaiiankchipelagoand the Marshall Islands, be called describe how DNA can be obtained for PCR analysis ft agassizi; that juponica be used for Cheloniu in the Indian small skin tissue biopsies using a relatively non-inva; Ocean and the western tropical Pacific; that Chelonia sampling procedure. In this procedure, a biopsy punc used to obtain a disc of tissue from the skin in the dorsal al. 1993);Andaman and Nicobar Islands - dudh kacchua, axial region of the rear flipper. No suture is necessary. kap-ka, kap-troeje, yadi-da (Bhaskar 1979); Angola - A distinct division was revealed between western Car- yofirie (Can-and Can 1991); Anguilla - greenback (Meylan ibbean nesting populations (Florida and Costa Rica) and 1983); Argentina - cay, tortuga Carey. tortuga franca, eastern Caribbean nesting colonies (Aves Island and tortuga de mar, mgaveide (Mittenneier et al. 1980; Suriname) and an inverse relationship was found between Freiberg 1981); - green turtle, malunba, wani nesting colony size and mtDNA diversity when analyz- (Nietschmann 1989; Bradley 1991; Cogger 1992); Bar- ing mtDNA control region sequences (Lahanas et al. bados - greenback, green turtle (Horroda 1992);Belize - 1994). Encalada (1994) also found greater mtDNA di- white turtle (M&quez 1990); Brazil - suman&,taruauga, versity among smaller nesting colonies in the Atlantic tartaniga do mar. ma(Mittenneier ei aL 1980; Freibeig Ocean when sequencing the narDNA control region of 1981); Chile - tortuga comestible, tortuga verde nesters from nine colonies. (Mittenneier et al. 1980); China - lu gui (Frazier et aL Rtzsimns et al. (1993-94) and Rtzsimmons et aL 1988); Cocos-Keeling Is. - penyu (Gibson-Hill 1950); (19941, respectively, very briefly describe the applicabil- Colombia - caguama, moro (female) and yauc (male), ity of microsatellite analysis in population structure of tortuga, tortuga blanca, tortuga de mar, tortoga verde marine turtles and the use of microsatellite techniques KÃ (MittennEsier et al. 1980; Green and Qrtiz-Crespo 1982); study male-mediated gene flow among populations in- Costa Rica (Caribbean) - tonuga. tortuga blanca (Can- cluding paternity of clutches. Details of how 1983);Costa Rica (Pacific) - tore, tonuga negra (Cornelius microsatellite analysis can be a valuable tool to comple- 1976; MA-quez 1990); Cuba - tortuga verde (Mibquez ment assays of sequence variation in nDNA and 1990); Ecuador - tortuga prieta (MA-quez 1990);Egypt - mfDNA is given by Ftasimrnons et al. (1995a). biswa, p'saya, tersa (Frazier et al. 1987); El Salvador - The genetic information on green turtle populations is tortuga verde (Mfirquez 1990);Fiji - ika darnu, mato 10% accumulatingrapidly but as stated by Lahanas et al. (1994) vonu damn, vonu loa (Hirth 197la); French Guiana - "It should be cautioned that finer-scale genetic data do kaouane, ouyainoury (Mittenneier et al. 1980); French not necessarily translate into greater geographic resolu- Polynesia - tortue, honu (Hirth 1971a); Gabon - nkudu, tion of population structure." A short, popular account of nkunu, ikes. tchiches, ehu, tortue verte (Fretey and the mtDNA work with green turtles was prepared by Girardin 1989); GalQagos Is. - tortuga negra, tortuga Bowen and Avise (1994). amarilla (Green and Ortiz-Crespo 1982); Gold Coast - After briefly reviewing the state of affairs, Paifiam and anwa, apuhulu. apuhuru. hala, klo (Irvine 1947); Zug (1996) recommended that the name Chelonia mydas Guadeloupe - tortue, tortue blanche, tortue verte (Meylan be used, with no formal subspecific recognition, for green 1983); Guatemala (Caribbean) - tortuga verde (Marquez turtle populations throughout the world, 1990); Guatemala (Pacific) - pariama, tortuga negra, The size, sample size, and history of the green turtle breed- tortuga verde (MArquez 1990); Guyana - bettia ing population bang analyzed can have significant bearing (Mittennder et d. 1980); Hawaii - green turtle, green sea on taxonomic conclusiork Small populations are subject turtle, honu (Bh1980); Honduras (Pacific) - guiltora to genetic drift and inbreeding. Populations emanating (Mfojuez 1990); Indochina - lemech, vich (Bourret 1941); (Founder effect) or being rebuilt (population bottleneck Indonesia - penyu biasa, penyu daging, penyu nijan, penyu followed by population flush) from a few individuals sda, wau kaku (Rhodin et al. 1980; Suwelo et al. 1982); usually have less genetic diversity than large populations. Israel (Red Sea) - turas al abiad (Prazier et al. 1987); Mad - fanojoaty (Hughes 1975);Maylayria - penyu 1.2.4 Standard common names agar. penyu empegit, penyu pulau, penyu pulo (Chin 1971; Standard common names are: English - green turtle, Tow and Moll 1982); Maldive Is. - vela (Deraniyagala , edible turtle, greenback turtle; German - 1956); Maisirah Is. - humsa asfah (PA0 1973); Mayotte - SuppenschildkrSte; Dutch - Soepschildpad; French - fanu, kasa, nyamba, tortue de mer, tortue franche, tortue tonue franche, tome de mer, tome verte; Portuguese - mangeable, tortue verte (Frazier 1985); Mexico (Carib- tanaruga, tartaruga do mar, tanaruga verde; Japanese - bean) - turtuga blanca (Mdrquez 1990); Mexico (Pacific) ao umigame; Caribbean Spanish - tortoga, tortuga blanca, - caguama negra, caguama prieta, moosni, parlama, tortuga verde. sacacfflo, tortoga negra, tonugaprieta (Cliffton et al. 1982; Common and local names of sea turtles are very im- Felger and Moser 1985; Mfirquez 1990); Micronesia - ponant tooh for field biologists and conservation agents. calap, melop, mwon, wel, winimon, won (Priichard 1977); Common names of Chelonia mydas for some countries Mobeli - dusi, kasa, nyamba (Frazier 1985);Mozambique and regions are: Aldabra Atoll - tortue de mer, tortie de - asa, casa, hassa, icaha, paten, sinernbo, tmga(Hughes mer (Hirth and Can- 1970); American Samoa - fonu, 1971); Netherlands Antilles - greenback, tortuga blanku laumei ma ena, laumei leai se uga (Tuato'o - Bartley et (Sybesma 1992); Nevis -greenback (Meylan 1983); Nica- (Caribbean) - torniga venk, tunel, wli (Nietschmann green turtle with moomi being commonly used. Otfa 1973); Nicaragua (Pacific) - torita (Marque2 1990); names are based on coloration, size and condition (Felg Ogasawara Is. - ao &game (Fukada 1965); Panama - and Moser 1985). tomga verde (Maiquez 1990); Papua New Guinea - 50+ names (Rhodinet al, 1980); Peru - tortug& tortuga blanca, 1.2-5 Definition of size categories , tmuga comestible, tortuga verde (Mittenneier et al. 1980; Size categories for green turtles are defined as follow Brown and Brown 1982; M@uez 1990); Philippines - (carapace lengths are standard straight-line tneasur bildog, katuan, pawikan, payukan, pudno, tortuga ments): (Pawikan Conservation Project Staff 1993); Portugal - tartaniga do mar, tartaruga verde (Osorio de Castro 1954); hatchimg-from hatching (still bearing conspicuot Ryukyu Is. - ao umigame (Fukada 1965); Sfc Tom6 - umbilical scar) to the first few weeks of life. ambo, m5o branco (Graff 1995); Senegal - dud,mawa, ndumar, tortue franche, tme verte (Maigret 1977); juvenile~posthatchlingto 40 cm carapace length. Tb Solomon Is. - 29 names (Vaughan 1981); Southeast Af- stage iidssent.ally the carnivorous (or omnivorous) p~ rica - asa, casa, fano, fanohara, fanojoaty, fanovua, green lagic stage. By about 40 cm carapace length most gree turtle, groenseeskilpad, hassa, icaha, ifudu. pateri, turtles have entered their nearshore feeding habitat aa sinernbo, tartaruga, tortie de mer, tortue de mer, tome are chiefly herbivorous (see section 2.2.2) franche, tortue verte, tsakoy (Hughes 1974);Sri Lanka - gal kfisbSva, mas kfisbBva, pal amai, pen amai, -vSli subadult-from 41 cm to the onset of sexual mafurit kSsbSva (Deraniyagala 1939); St. Barthelemy - tome about 70 to 100 cm carapace length, depending upon th (Meylan 1983); St. Kitts - greenback (Meylan 1983); St. population. Martin - greenback (Meylan 1983); Suriname - kadaloe, krap6, ouyamouri, pefiiing, portoka, soepschildpad adult-sexual maturity, >70-100ern carapace length de (Schuiz 1975; Mittenneier et al. 1980); Thailand - tao pending upon the population. The size at sexual maturit saeng-atit, tao ta-nu (Nutaphand 1979);Tonga - fonu, fonu for males is presumed to be similar to that of females (bt til ' akula, fonu tu ' apolata, fonu tu ' a ' uli, tuai fonu see section 3.1.1). (Hirth 1971a); Turkey - tirros (Hathaway 1972); Uruguay - tortuga verde (Mittermeier et al. 1980); Venezuela - 1.3 Morphology caguamo, tortuga blanca, tortuga comestible, tortuga de sopa, tortuga franca, tortuga verde (Mittenneier et al. 1.3.1 E~tern~ntemalmorphology and col 1980; Pritchard and Trebbau 1984); Western Samoa - oration laumei (Hirth 1971 a); Yemen - bissa, harnas, humea (FA0 General external morphology is described ii 1973). Deraniyagala (1939), Can- (1952). Pritchard (1979b) Western Islanders in the Torres Strait distinguish 13 ftitchard and Trebbau (1984). Ernst and Barbour. (1989) kinds of green turtles (waru)based on size, sex, age, color, Mfirquez (1990), Cogger (1992) and Ernst et al. (1994 habitat, agility, appearance and taste of the 's fat (see Rg.3). (Nietschmann 1989). The least desirable, garau warn A few investigators have given some populations spe ("drying reef turtle") are old, move slowly, are sedentary ciKc or subspecific names based mainly on morphologi- residents of drying reefs and graze on several algas which cal and biochemical traits and color. These population; the Islanders say produce poor-tasting, black calipash fat. and the traits have been discussed in section 1.2.3. The kapu waru ("good turtle*) are younger, bigger, faster, Morphometric measurements of adults and hatchlings eat mostly seagrasses, migrate to nesting beaches on the from a wide range of locales are described in sections Barrier Reef, and the Islanders say have good-tasting green 3.1.2 and 3.2.2, respectively. calipash fat. Unfortunately, turtle workers have used different names The Yanyuwa hunters in the southwestern Gulf of for some shell structures. The scutes (or laminae) of the Carpentaria have a long historical and spiritual associa- carapace are the vertebrals (=centrals), cost& (=laterals, tion with green turtles and dugongs. The green turtle is pleurals), marginals, nuchal (=precentral, cervical), and known as malurrba but there are additional names based supracaudals (=I 2th marginals, postcentrals) (see Fig, 4). on sex. size, coloration, condition and combinations of There is more uniformity in the terminology of the plas- these traits (Bradley 1991). tral scutes: intergular, gulars, humerals, pectorals, In Tonga some fishermen have names for color phases, abdominals, femorals. and anals. The bones of the cara- sizes and sex (ffirth 1971a). and the Raroians have names pace are the nuchal (=proneural), neurals, pleurals for differentsizes of sea turtles (Danielsson 1956). (scostals), peripherals (=marginals). pygal, and The Seri of Pacific Mexico have eight names for the suprapygal. The paired bones of the .plastron are the Fig. 3. Chelonia mydas, ventral view of skeleton, with plastron removed. 1-nuchal plate; 2-scapula; 3-minion process of scapula; 4-coracoid; 5-humerus; 6-radius; 7-ulna; 8-pubis; 9-ischiuni; 10-ilium; 11-femur; 12-tibia; 13- fibula; 14pisifom; 15-carpals; 1&metacarpals; 1?-phalanges; 18-fontanelle: 19-pleural plate; 20-peripheral plate; 21- tarsals; 22-metatarsals; 23-phalanges; I-V-digits. From DeWitte in Vielliers, A. 1958 (with some nomenclatorial addi- tions). Tortues et crocodiles & 1' Afrique NoireFrangaise. Initiations Afncaines. Institut Frangais D'Afrique Noire 15: 1-354.

Fig. 4. Sketches of Chelonia mydas illustrating the epidermal laminae (right) and bony elements (left) of the carapace (A) and the epidermal laminae (right) and bony elements (left) of the plastron (B). abd-abdoininal scute; asc-and scute; cpl-anteriormost of eight costal plates; csc-one of five central scutes; enp-entoplastron; epp-epiplastron; fsc- femoral scute; gsc-gular scute; hsc-humeral scute; hpp-hypoplasmon; hyo-hyoplastron; isc-intergular; lsc-anteriormost of four lateral scutes; msc-marginal scutes; nup-neural plate; pec-pectoral; pnp-preneural plate; prb-posterionnost of eleven peripheral bones; psc-precentral scute; pyb-pygal bone; spb-suprapygal; xppxiphyplastron. Interlaminal seams are shown by solid lines and sutures by irregular lines. From Legler, J. M. 1993. Morphology and physiology of the Chelonia. Pages 108-1 19 in C. J. Glasby, G. J. B. Ross and P. L. Beesley (eds.). , Vol. 2A. Amphibia and Reptilia. Australian Government Publishing Service, Canberra, Australia. Commonwealth of Australia copyright reproduced by permission. epiplastra. hyoplastra, hypoplastra and xiphiplastra. The brane is thick and contains a large amount of fatty ti% single bone anteriorly is the entpplastron. Zangeri (1969). This material serves to link the surface layer to Pritchaid and Trebbau (1 984) and Ernst and Barbour extracolumellar kuob. The cochlea attains a length of 1' (1989) discuss the shell terminology used by different p. Ridgway et al. (1969) measured cochlear potent investigators. and found maximum sensitivities in the regions of 3(X The streamlined green turtle carapace has five verte- 400Hz.. bral scutes, four pairs of costals and twelve pairs of The morpholo& of the pineal-paraphyseal comple> marginals. Carapacial scutes are juxtaposed. The rela- large structure projecting dorsally and anteriorly abc tively small head has a pair of elongate prefontal scales. the prosencephalon, was described by Owens and Ral The minimally retractile neck is thickand relatively short. (1978). They describe two pineal cell types which i The strong forelimbs are elongate and paddle-shaped. The pew to correspond to the neuroglial supportive cells a hindlimbs are smaller than the forelimbs. A median the secretory rudimentary photoreceptor cells of other a present on the hatchling carapace, is weakly present in niotes. The production and role of melatonin was d some juveniles and absent in adults. cussed^y Owens et al. (1980) and Owens and Ge Wyneken (1994) speculatesthat the changes in the shell (1985). shape of green turtles represent strategies for predator Winokur and Legler (19%) found that green turtles la avoidance. In the epipelagic stage the carapace grows typical rosnal pores (= epidermal invaginations in t from an elliptical shape to a nearly circular shape. Later, internarid region) but they have a deep sagittal fissia when green turtles have entered coastal habitats, the cara- Vath an expanded basal portion, between the prefroni

pace has regained its elliptical appearance. scales, and this may be a pore hornologue. In his study 1 According to Pritchard and Trebbau (1984) the bones buccopharyngeal mucosa of turtles. Winokur (1988) di of the carapace are relatively thick in adults; the neural covered that green turtles were the only species studit bones are narrow; all peripheral bones except I, If and X with what he termed "pharyngeal tonsils." These were bear a pit which receives the end of the rib; there are nine series of five to seven pits. lined with cells, in the pha plastral bones with persistent fontanelles along the mid- ynx, posterior to the glottis. line and at the center of each bridge; and, the skull is dor- Quesada and Madriz (1986) give an account of tk sally flat and extensively roofed. anatomy of the adult heart and Jaffee (1969) very briefl Zangerl(1980) described the phylogenetic relationships describes some aspects of the ontogeny of the embryoni among Chelonia rnydas, Chelonia asmondai-and other heart. Barragfin (1994) very briefly described the cardic cheloniids based on morphology of the carapace, plas- vascular anatomy of a juvenile Mexican black turtle. Ill tron and limb skeleton. The shell of turtles is described aortas of several, juvenile green turtles from the Cap and the family Cheloniidae is placed in the metachelydian Turtle Farm were found to have gross aneurysmal dila level of organization by ZangerI(1969). tions and multiple raised plaques which resembled bot Solomon et al. (1 986) described the heavily keraiinized the aortic lesions in Marian's syndrome in humans am plastron and carapace of the green turtle. The epidermis those induced by chemical treatments in (Toda e is generally 2-4 cells thick but at growing points it can al. 1984). Sapsford ( 1978) reported on a muscular sphinc attain 6 cell layers. Examination of the bones of C. my& tea" in the pulmonary arteries of four species of sea turtles revealed typical chelonian articular surfaces without including the green turtle, and suggested that its piesena transphyseal vascularization (Rhodin 1985). Gaffhey provides a mechanism for the control of blood flov (1979) reviewed the earlier literature, then updated, the through the heart. cranial morphology of sea turtles. He gives anatomical The anatomy of the lung was described by Solomor descriptions of horizontally (Rg. 5) and medially sec- and Pnrton (1984) and among their findings were: the res- tioned skulls of C. mydas. Albrecht (1976) described in piratory epithelium is typically vertebrate, being detail the cranial arteries (Fig. 6). The large stapedial and pseudostratified columnar with cilia; thegaseous exchange palatine arteries of the green turtle, and other sea turtles, areas appear at all levels from the respiratory bronchi to may be similar to the primitive cranial arterial pattern for the alveoli; and the epithelial lining of the alveoli is em- turtles. posed of type I and type Il pneumocytes which are mor- The papillae along the lateral choanal margin and the phologically similar to those of and mammals. nasal cavities are described by Parsons (1968, 1970). Patterson (1973) reported a lung volume (in 3) to body Saint-Girons (1991) studied the nasal cavity, histologi- mass (02) ratio of 0.049 in a green turtle. cally. Liebman and Granda (1971, 1975) described the Solomon and Tippea (1991) determined that the livers rods, cones and oil droplets in the eyes. of male and female, fann-reared turtles are fat laden and According to Wever (1 978), who descri that liver weight and fat accumulation increase with ani- turtles, in C mydas the middle layer of the tympanic mem- mal weight. Fig. 5. Cranial morphology of a green turtle. Dorsal view of a horizontally sectioned skull. Hatched areas indicate cut surface. Anatomical abbreviations: bo, basioccipital; bs, basisphenoid; ex, exoccipital; ju, jugal; nix, maxilla; op, opisthotic; pal. palatine; pf. prefrontd; pm. premaxilla; PO, postorbital; pr, prootic; pt, pterygoid; qj, quadratojugal; qu. quadrate; vo, vomer. From Gaffney, E. S. 1979. Comparative cranial morphology of recent and fossil turtles. Bull. Amer. Mus. Nat. Hist 164(2): 65-376. Copyright American Museum of Natural History 1979. Courtesy American Museum of Natural History Library.

Fig. 6. Semidiagrammatic dorsal view of the cranial arteries (right side) and cranial arterial foramina and canals (left side) of Chelonia mydas. 1-F. posterior canalis carotici interni; 2-Canalis caroticus internus; 4-Auditus canalis stapedie temporalis; 5-Canalis stapedio-temporalis; 6-F. stapedio-temporale; 7-Canalis cavernosus; 8-F. cavernosum; 10-F. caroticum laterale; 11-F. anterior canalis carotici interni; 17-F. orbito-nasale; 18-F. alveolare superius; 19-Canalis alveolaris superior. 20-F. supraorbitale; 21-Fissura ethmoidalis; 23-F. arteriomandibulare; 24-Sulcus caroticus; 28- Internal carotid; 29-Stapedial; 30-Cervical; 3 1-Palatine; 32-Vestigial mandibular; 33-Mandibular; 340bital; 37-In- fraorbital, 39-Supraorbital; 40-Alveolar-nasal; 41 -Posterior nasal; 42-Superior alveolar. From Albrecht, P. W. 1976. The cranial arteries of turtles and their evolutionary significance. J. Morphology 149(2}: 159-182. Copyright. The Wistar Institute Press 1976. mekidneys of green turtles are flattened, lobed and carapacial shields. closely applied to the posterior wall of the pleuroperitoneal Qreen turtle hatchlings are blackish above and I cavity (Solomon 1985). Using light, scanning and trans- below. The plastron usually remains light-colored a mission mimpy, Solomon (1985) descn¥bethe func- turtle grows. However, Balazs (1986) noted that Hz tional nephron as being comprised of a glomerulus, proxi- iau hatchkip passed through a pronounced color p mal tubule, intermediate, segment which can be subdivided in early life. The plastron of Hawaiian neonates (50 into a proximal non-secretory segment and a distal mu- carapace length) are white but soon become diffused cus secreting segment, distal convoluted tubule, and col- gray and black pigment t intensity between lecting tubule. 80 mm carapace length) en the dark color f The ovary is a membranous structure with a relatively away and usually disappears completely (by 1SOmm c short attached border resulting in significant folding. In pace length) leaving the plastron white again. All reproductively inactive animals a narrow, compact cortex hatchlings have been reported in Sarawak (Harry and spongy medulla are easily recognized (Aitken et al. 1963) 'Ibrmmero (Can-1 %?a; Fig. 7) Florida (Fletem 1976). The roles of the several subdivisions of the ovi- 1977) nd North Carolina (Schwartz and Peterson 19 duct in egg formation are elucidated by Aitken and Carr 667a) staked that four albinos had been foum Solomon (1976) and Solomon and Baird (1979). some one hundred and fifty thousand hatchling: Using light microscopy, Ehrenfeldand Ehrenfeld (1973) Tbrtugum. described the milky and inguinal glands. They pmtu- 39) described eight color stages in lated that the secretion of the glands may serve as a de- . Hatchlings' carapaces are dark gre fense substance and/or the secretion may play a role in fch bronze and after three or four months they beco innaspecies cornmication. Later, Solomon (1984) us- brownish-red. This background color then is variega ing scanning electron microscopy and transmission elec- by black, brown and yellow streaks. Ultimately the ca iron microscopy, as well as light microscopy, elaborated pace color is suffused with olive-green and the black a on the structure of the axillary gland. brown streaks are broken-up into small spots. Ventral The phallus of the green turtle is similar in structure to the color is white in the young and light yellow in t those of the hawksbill, loggerhead and leatherback turtles, adult. in that a single U-shaped fold forms the glans and the Juvenile carapace coloring is highly variable. Ma seminal groove is single and terminates medially on the regional handbooks and field guides give word descri inner surface of the fold (Zug 1966). tions of color patterns of immature and/or mature turd The major visceral organs of the green turtle are illus- in their respective areas. In an attempt to standarize mi (rated in the booklet by Rainey (1981), and Wolke and descriptions of immature green turtles, Hiet aL (1 99 George (1981) have written a guide for conducting necrqp- used Munsell soil color chips to describe pigmentation sies under laboratory and field conditions. Some of the Wuvulu Island turtles. Here, the coloration of subadol older references to the morphology and physiology of the with carapace lengths of between 45 and 72 cm were b. green turtle are cited in Hi& (1971b). sically similar. For example, the large vertebral and ca Miller (1985) reviewed the literature and concluded that tal scutes were, when wet, brown (7.5 YR 4/41 to dai the frequency of occurrence of abnormal embryos and brown (7.5YR 3/2) at the basal seam with emanating dar hatchlings is low among marine turtles. The most com- brown and olive gray (5Y 5/2) rays of varying lengtt mon malformation is variation in scale patterns. From Hie plasm varied from white (5Y 8/1) to pale yellow their personal observations and a review of the literature, (5Y 8/3). Rhodin et al. (1984) concluded that the incidence of spi- The coloration of adult males and females is higw nal deformities and kyphosis among a total of 4.207 green variable. Frazier (1971) illustrated a wide range of adul turtles, fromfour localities, was, respectively, 0.14% and color patterns within the Aldabra population. Pritchan 0.10%. (1971) described several morphotypes in the Galeago; Deviations from normal central and lateral carapace ranging from a "yellow" type to a few with mydas-lik scutes were statistically different in hatchlings from a carapaces to those (majority) with blackish dorsa. hatchery (mean 12.8%) and in hatchlings from natural Photographs illustrating coloration and general exter- beaches (mean 4.9%) in the Ogasawara Islands nal morphology are provided by: Can- (1967b), Costa (Suganuma et al. 1994). Although not statistically differ- Rica (Caribbean), hatchlings, adult female; Frazier (1971). ent, 5% of adult females (N=1,252) and 3.3% of adult Aldabra Atoll, adult males and females; Can- (1972b), males (N=661) in the Ogasawara population exhibited Australia, sleeping underwater, Costa Rica (Caribbean), similar scute abnormalities. nester; Bustard (1973). Australia, large females, male; Demetropoulos and Hadjichristophorou (1995) pro- Ehrenfeld (1974),Australia, copulating. hatchlings, nestfa, vide a color photo of an adult green turtle lacking all swimming male (cover); Hallowell (1979), Austtalia, large Fig. 7. A normal green turtle hatchling and a partial albino from Tortuguero, Costa Rica. Each weighs about 26 g. Hatchlings from this population will increase in weight approximately 4.700X by the time they are reproductively mature. Note that unlike adults, hatchlings crawl using diagonal flippers (i. e. front limb moved in conjunction with hind limb on opposite side). male; Pritchard (1 979b). Australia, large male, Galfoagos, able after hybridization with Bkm 2(8) (Demas and nester, Guyana, adult female, captivity, swimming male;" Wachml 1991). Spring (1 980). Papua New Guinea, juvenile and large fe- Frair (1 977a) made a comprehensive survey of the turtle males; Pritchard et al. (1983). locations not given, literature concerning packed red blood cell volumes posthatchling,juvenile, adult male and female; Sheppard (PCV). red blood cell counts (RCC) and red blood cell (1 983). Australia, copulating and stack of four; Pritchard sizes. The mean PCV of green turtles from several stud- and Trebbau (1984). Suriname, nester, captivity, ies ranged from 24.8 to 31.6 cm^/l00 cm3 and the mean posthatchling; Bonnert et al. (1985a). Southwest Indian RCC in two studies was 523 and 530/nun3x 10-3 . Larger Ocean, hatchlings and nester; Cornelius (1986), Costa green turtles tend to have higher PCVs, larger red cells Rica (Pacific),large female; Miller (1989), Saudi Arabia, and lower RCC (Frair 197%). There is also a significant nesters, copulating; Cogger (1992). Australia, copulating positive correlation between carapace length and total stack (2 males, 1 female); Rudloe and Rudloe (1994), serum protein (Frair and Shah 1982). Grumbles et al. Mexico (Pacific) copulating; Demetropoulos and (1990) determined that, in wild turtles off the Pacific coast Hadjichristophomu (1995), Mediterranean, adult male and of Mexico, packed cell volumes and red blood cell counts female, juvenile and hatchling; Lindsay (1995). Indone- were not significantly different between nesting females, sia, adult male swimming; Wuethrich (1996), Hawaii, females at sea, and males at sea, bat white blood cell counts adult swimming. were significantly lower for males at sea than nesting fe- The common name, green turtle, does not refer to its males or females at sea. Bolten and Bjorndal(1992) found external color, but to the color of its fat. that PCV was not significantly related to differences in body size or sex in a population of juvenile green turtles 1.3.2. Cytomorphology in the Bahamas. They did find that plasma uric acid and Chelonia mydas has a diploid number of 56 chromo- cholesterol were significantly different between females somes. There are no heteromorphic sex chromosomes and males. Wood and Ebanks (1 984) identified six cell (Bickham et al. 1980, Bachftre 1981). types in the blood: red cells, lymphocytes, eosinophils, Using Bkm 2(8) probes, Demas et al. (1 990) found male basophils, neutrophils and thrombocytes. The infrastruc- and female-specific DNA fragments in the green turtle. ture of thrombocytes are described by Bonnet et al. Individual-specificDNA fingerprints are readily identifi- (1985b). McKinney and Bentley (1985) found that blas- togenic and cytotoxic responses of leukocytes to some *to-n)acroglobulb in green turtle plasm was about 4 mitogens and antibodydependent cell-mediated cytotox- ml and that the concentration of ovomacroglobulin in g icity were of significant magnitude. Based upon the ex- Meegg white was about 0.4 nag/mL They postulated Unction coefficient, the total carobnoid content in the se- the difference was due to divergent evolution. Elec rum of two Pacific green turtles was determined to be micrographs ofalpha-inacroglobuiin and o~~naacroglot 1.27 &ml (Nakarnura 1980). revealed similarities in their fandmental architecture Owens and Ruiz (1980) developed methods for obtain- differences in some details (Sksd et al. 1988). ing blood samples from the paired dorsal cervical sinuses Egg lipids from green, loggerhead, leatherback and cerebrospinal.fluids from the foramen magnum. hawksbill turtles have distinct profiles and this IOK Kornent and Haines (1982) describe the establishment edge has been used as a forensic tool to support enfo and characterization of a skin cell line from a young green rnent of protective regulations (Seaborn sad Mows 19 turtle. Sage and Gray (1 979) determined the amino acid o positWn of elastin in the aorta. 1.3.3 Protein composition and specificity and Myoglobins from an Atlantic and a Pacific green ti general physiology exhibited similar amino acid compositions but with 1 Serum electrophoresis studies indicated that proteins sible differences in lysine, histidine, serine, glutamic a from Chelonia are more like those of Carem and proline and glycine residues (Williams and Brown 19' Lepidochelys than like Eretmochelys (Frair 1982). Sys- 'Jhe amino acid sequence of the main component in tematic information derived from imunoelectrophoretic globin from skeletal muscle of the Pacific green tu work (Mao and Chen 1982) are in general agreement with was analyzed by Watts el al. (1983). taxonomic-relationships established by morphological The complete amino acid sequences of green tu criteria. growth hormone and prolactin consist of, respectively, ars that organic phosphate modulators regulate and 198 amino acid residues (Yasuda et al. 1989 i whole blood oxygen affinity during embryonic develop- Yasuda et al. 1990). ment but not in the adult (Isaacks and Hartaiess 1980). The lysine: histidine ratio in the shell keratin of v Wells and Baldwin (1994) described how hatchling eryth- green lurries and farm-reared green turtles are significal rocyte (red blood cell) mean cell volume is approximately different (Hendrickson et al. 1977). These results in half of the adult value, but hematocrit, blood hemoglobin care a dietary influence on shell composition at least concentration and blood viscosity of hatchlings and adults to some point. are similar. Friedman et al. (1985) found that sea turtle Depot fatty acid composition in Caribbean turtles w hemoglobins are designed for efficient oxygen transport studied by Joseph et al. (1985) and fatty acids in and release to tissues rather than storage. They also de- fats ofgreen turtles from Hawaii and Johnston Atoll, wh scribe how the temperature response of the oxygenated feed principally on marine algae, were analyzed hemoglobin may be related to its ability tomaintain meta- Ackman et al. (1992). bolically active tissues at several degrees higher than aria- Measurements made of the water content and fat ofsti bient temperatures. This regional endothermy may assist dard cores of fat lining the inner carapaceof turtles caul the turtle in long migrations. off Dam, Papua New Guinea, revealed that the amount Working with Australian green turtles, Reina (1994) depot fat, total lipid and neutral lipid per core varied w found that hatchlings have significantly higher levels of the sex, reproductive status and maturity of the individi sodium and potassium in their plasma than do adults and (Kwan 1994). Cores from adult females had a signi the differences may be associated with diet (hatchlings candy greater fat content than those from adult ma11 feed on macroplankton: adults on seagrasses). Plasma Cores from pubescent and vitellogenic females bad t zinc. analyzed by atomic absorption spectrophotometry, highest fat content. Results.of this study suggested tf in five green turtles from Costa Rica, averaged 1.00 M/ sub-carapace depot fat supplies the energy formigratio ml (range 0.67- 1.29 pgfml) (Lance et al. 1995). This av- and egg production. erage was slightly higher than the means of 0.64 and 0.84 Penick et al. (1996) determined that Qlo values of p pglrnl reported for five olive ridleys and one Kemp's rid- fat, small intestine, nonswiinming skeletal muscle, pe ley. respectively. toralis muscle, liver, heart and kidney tissues ranged fro Using starch gel electrophoresis, Smith et al. (1978) 0.65 to 3.38. Tissue metabolic rates were highest in ti analyzed thirteen biochemical loci in green turtles from kidney and heart tissues and lowest in the green fat ai Florida and the Caribbean. They found that 46.2 to 69.2% small intestine tissues. Muscle tissue had a high oxygf of the loci were polymorphic and that heterozygosity av- consumption and this elevated metabolism may be . eraged 11 '9%. tive for long migrations. Osada et al. (1988) determined that the concentration of Owens and Morris (1985) reviewed the literature c the comparative endocrinology of Cheloniu, Carem and Zealand, Nicaragua, ~o~~arianas,Oman, Pakistan, Lepidochelys. This review included references to research Palau Republic, Panama,Papua New Guinea, Peru, Phil- on pituitary homogenate, follicle stimulating hormone, ippines, Pitcairn. Puerto Rico, Qatar. Reunion (Europa, luteinizing hormone, growth hormone, thyroid stimulat- Tmmdin. Bes Glorieuses, Juan de Nova), Silo Tom6 and ing hormone, gonadotropin releasing hormone, arginine Prindpe, Saudi Arabia, Senegal, Seyqhelles, Siena Leone, vasotocin, melaionin, restostenme, cstradid, progesterone Solomon Islands, Somalia, South Aftica, Sri Lanka, St. and cordcosterone. Subsequent to this review, some other Kim-Nevis, St. Lucia, St. Vincent and the St. Vincent accounts of green turtleendocrinology and physiology are Grenadines. Sudan. Suriname, Taiwan, Tanzania, Thai- provided by Ijcht et aL (1984) and Licht andPapkoff (1985) land, Togo, Tokelau, Tonga, Trinidad and Tobago, Tur- on glycoproteins; Licht et al. (1 985) on thyroxine and test- key, Turks andCaicos Islands, Tuvalu, United Arab Emir- osterone; Licht et al. (1991) on thyroxine; and. wbbels et ates. United States ofAmerica, U.S. Pacific Islands (Jarvis al. (1992) on follicle stimulating hormone, luteinizing Island, Johnston Atoll, Howland and Baker Reefs, Palmyra hormone, progesterone and testosterone. Island, Wake Island), U.S. Virgin Islands, Vanuatu, Ven- Herbst and Klein (1995b) showed how monoclonal an- ezuela. Viet Nam, Western Samoa, Yemen, Zaire. tibodies may be useful for inununodiagnostic applications To the aforementioned list of areas can be added: Ar- in die green turtle. Shaw et al. (1995a) tested isoflurane gentina (F'er198% Richard and Moulin 1990, Scolaro on juvenile and subadult green turtles and found it to be a 19901, Canada (Carl 1955). Prance (Knoepffler 1961). safe and effective anesthetic. Greece (Margaritouliset al. 19861, Italy (Gramentz 1989; Basso 1992), Korea (Shannon 1956), Malta (Despott 1930; Brongersma and Can" 1983), Netherlands (Brongersma 1982); Portugal (Brongersm 1982). Spain (Bmngersma 2. DISTRIBUTION 1982; Pascual1985). Tunisia (Laurent et al. 1990; Laurent and Lescure 1992),United Kingdom (Penhallurick 1990), 2.1 Total Area and Uruguay (Gambarotta and Gudynas 1979; Gudynas Green turtles are circumglobal, commonly found 1980; Frazier l984a). throughout the tropical seas and as stragglers in a far Some recent, detailed accounts of their distribution in more extensive area. In general, green turtles are seen some areas of the Pacific Ocean can be found in Hih between 40° and 40"s latitudes, but there are a dearth (1 971&,1993),Polunin (1975). Pritchard (1 977,1979a). of sightings ia the east-central Pacific Ocean and the Balazs (1980). UNEP/IUCN (1 988~1,Lockhart (1 989), northeast Atlantic Ocean. They occur on the nesting and Smith and Smith (1979,1993). On the Pacific coast beaches or in offshore waters of at least 139 countries of the USA, green turtles have been documented from and territories. Groombridge and Luxmoore (1989) Alaska (Hodge 1981), Washington (Slater 1963) and Cali- briefly review their occurrence in 126 areas: Ameri- fornia (Carr 1952). The following references provide can Samoa, Angola, Anguilla, Antigua and Barbuda, detailed distribution accounts of green turtles in and Ascension and St. Helena, Australia, Bahamas, around the Atlantic Ocean: Villiers (1958). Brongersma Bahrain, Bangladesh, Barbados, Belize, Bermuda, (1972), Carr et al. (1982). Meylan (1983). Pritchard and Brazil, British Indian Ocean Territories, British Vir- Trebbau (1984), Bacon et al. (1 984), Delaugen-e (1987), gin Islands, Burma, Canary Islands, Cape Verde Is- UNEP/IUCN (1 988a), Ogren et al. (1989), Groombridge lands. Cayman Islands, Chile, China, Colombia, (1 990) and Smith and Smith (1979, 1993). Green turtles Comoro +Islands,Congo, Cook Islands, Costa Rica, occasionally stray into the Black Sea (Vdkanov 1949; Cuba, Cyprus, Djibouti, Dominica, Dominican Repub- Fuhn and Vancea 1961; Geldiay et al. 1982). In the east- lic, Ecuador, Egypt, El Salvador, Equatorial Guinea, ern USA, green turtles have been documented off the At- Eritrea, Federated States of Micronesia, Fiji, French lantic and Gulf coasts from Massachusetts to Texas Guiana, French Polynesia, Gabon, Ghana, Grenada and (Magnuson et al. 1990)although they are much more com- the Grenadian Grenadines, Guadeloupe, Guam, Gua- mon in the warmer waters. In continental USA the main temala, Guinea, Guinea Bissau, Guyana, Haiti, Hawaii, nesting grounds are in Florida. The northernmost nest- Honduras, Hong Kong, India, Indonesia, Iran, Israel, ing record on the U.S. Atlantic coast is North Carolina Jamaica, Japan, Kampuchea, Kenya, Kiribati, Kuwait, (Peterson et al, 1985). Some detailed accounts of their Liberia, Madagascar, Madeira and Azores, Malaysia, distribution in the Indian Ocean are in Peters and Lionnet Maldives, Marshall Islands, Martinique, Mauritania. (1973). Bonnet (1 986), UNEP/IUCN (1 988b), Miller - Mauritius and Dependencies (Rodrigues, St. Brandon (1989) and Frazier (1990). Shoals, Mayotte), Mexico, Montserrat, Mozambique, Some worldwide distribution accounts are in Parsons Namibia, Netherlands Antilles (Curacao, Bonaire, (1962), Hirth (l971b. 1980b), Sternberg (1981) and Saba, St. Eustatius, St. Maarten), New Caledonia, New Bjorndal(1982a). 2.2 Differential Distribution ' that both the northern and southern contingents of Tortuguero breeding population pass their entire life cy 2.2.1 Hatchlings within the Southwest Carribbean Gyre and the reg the sea, it is postulated that the hatchlings around its perimeter." Coston-Oements et al. (1991) actively swim (the so-called "swim frenzyw)directly away view the literature and report how in addition to fours from land until they encounter zones of conveqence and/ cies of sea turtle^, pelagic sargassm supports a dive or, where present, sargassum rafts (see Fig. 8). These community of epiphytes, fimgi and more than one b convergence zones, or rafts, are rich in prey for the dred species of invertebrates and fishes. hatchlings and provide shelter. At this stage thehatchlings Pitman (1992) observed two green turtles, as well and young juveniles are thought to be chiefly carnivorous three other species of turtles and a number of unida but some omnivory may prevail. This tune in the lives of fied turtles, associated with flotsam in the eastern rrc green turtles, as well as other marine turtles, was cm- cal Pacific Ocean. Witham (1991), however, pointed I monly referred to as the "lost year". It is now thought that Ha/See may be significant predation on marine tun that this epipelagic phase of the green turtle, when it is associated with sargassum and flotsam and he sugges presumed to be drifting in ocean currents and gyres, may hatchlings may have a higher survival at sea away fn be somewhat more protracted than a year and the plural sargassum and biomass accumulations. Collard (19! "lost yearsn is more accurate. It should be emphasized found that in some places in the Gulf of Mexico and No that the turtles are not "lost"; humans have just not been Atlantic food would be available for pelagic sea turtle able to track them. Some post-hatehlkg recovery sites, non-frontal zones. The Sea Turtle Research Unit (1% although few are recorded, may be quite distant from of the Univmiti Pertanian Malaysia has recently dev known nesting beaches. Strong El-Nifio years must also oped a technique to monitor movements of hatchlings have an effect on hatchling distribution. In a couple of sea by miniaturization of radiotelemetry. seminal papers Carr (1986, 1987a) developed a model The epipelagic phase in green turtles has been estima showing how hatchling and juvenile loggerheads may at little more than a year (Hughes 1974b), from 7 to drift, perhaps for several years, with major currents and months (Can- et al. 1978). at least two years (Balazs et eyres between North America and Europe and Has model, 1987), from 1 to 3 years (Ehrhart and Witherington 195 where applicable, may be relevant to green turtles. Eckert and Honebrink 1992). and from 2 to 5 ye< In laboratory experiments. Mellgren et al. (19941 found (Aqhet al. l994b). that hatchling green turtles, unlike hawksbill and logger- head hatchlings, did not orient to or congregate in artifi- 2.2.2 Juveniles, subadults, and adul cial weed beds or in real seaweeds. They concluded that As mentioned in the preceding paragraphs, the epir the lost years habitat of the green turtle has yet to be de- lagic stage of hatchlings and juveniles may persist fa termined. year or more. The meroplanktonic green turtles are cca While not dealing specifically with green turtles, monly seen again when they enter shallow water near Brongersma (1972) discussed the possible roleof currents lands or the neritic habitat (see Pig. 8). At this stage th in carrying sea turtles across the North Atlantic Ocean. are chiefly herbivores, feeding on seagrasses and algt Hughes (1989) stated "green turtles nesting on Europa The approximate sizes (carapace lengths) at which juv are carried away from the shark-infested inshore waters nile green turtles leave die epipelagic habitat and en1 by an eddy of the Mozambique Cmrent, while the South their shallow water feeding habitat, in the Pacific Oce, Equatorial Current carries green turtle hatchlings away area, are: 30.5 cm, in Western Samoa (Witzell 1982); : from St. Brandon and Tromelin." Witham (1980) reported cm, in the Hawaiian Islands (Balazs 1982a); 35 cm, that tag returns from pen-reared yearling green turtles Johnston Atoll (Balazs 1985a); 36.8 cm, at Wuvulu I suggested ocean current dispersal and he concluded "Our land in Papua New Guinea (Hirth et al. 1992); 36 cm, data strongly suggest that the initial posthatching period, Heron Reef, Australia (Limpus and Reed I985a); 36 a 'the lost year', is a period of oceanic existence, when at Crown Island in Papua New Guinea (Spring 1983); turtles oppo&nistically use ocean currents and food re- cm. in , Australia (Limpus 1982a); and, 43 sources for dispersal and survival." The fiatback turtle, cm, in the Solomon Islands (Vaughan 1981). In the A Natator depressus, may be unique among sea turtles in lantic Ocean and Mediterranean Sea area the smalle not having a pelagic phase in its life cycle (Walker and green turtles found in some benthic feeding habitats ar Pannenter 1990). 21 cm,in Florida (Ehrhart 1983); 22-24 cm. in Turk Can- and Meylan (1980) postulated that Tortuguem (Groombridge 1990); 22.2 cm in Texas (Coyne and Landj hatchlings associate themselves with sargassumrafts,and Jr. 1994);23.6 cm,in Florida (Henwood and Ogren 1987 drift in these rafts with currents. Developing this idea 24.6 em, in Puerto Rico (Collazo et al. 1992); 25 cm, I further, Can- (1980) posited that "it is therefore possible the Bahama Islands (Bjarndd and Bolten 1988); 25 cr Immature and adult -shallow - water feeding habitats

; indicates Fig.,8. Generalized life cycle of green turtles. The dashed line from the nesting beach to the feeding habitats that the routes taken (i. e. direct or indirect) are generally unknown. (Figure modified from FAO, 1973).

15 in the U.S. Virgin Islands (Boulon and Frazer 19%); 26.6 l994b). cm, in Texas (Shaver 1994); 29.5 cm. in Florida Tagging projects now underway in some localities m (Mendonqa 1981); 30 cm in Bermuda (Meylan et al. eventually link developmental habitats with specific ad 1992a); and, about 31 cm in Brazil (da Costa 1969). Al- foraging habitats and with specific nesting beacht -though the methodologies, objectives, and sample sizes Norman et al. (1993-94) briefly describe how gene in these studies were different, it appears that Pacific ju- markers are being used to identify green turtle stocks venile green turtles enter the nearshore feeding habitat at feeding grounds off eastern Australia. In a popular, slightly larger sizes than their Atlantic counterparts. tide. Bowen (1995) discusses how natural tags, e. Some green turtles may move through a series of "de- mtDNA polymorphisrns, can now be assayed to link velopmental" feeding habitats as they grow. For example, population's widely distant breeding and feeding sites subadults are found in a feeding pasture off the west coast On the other hand, some shallow water feeding gmun of Florida (Can- and Caldwell 1956) and juveniles and harbor aggregations of both immature and mature turtli subadults are seen in the developmental habitats off the ~razrf(da Costa 1969), Yemen (Hirth and Can 197t east coast of Florida (Mendonqa and Ehrhart 1982; Ehrhart Nicaragua (Mortirner 1981)' Oman (Ross 1985) and Ti 1983; Wershoven and Wershoven 1992; Schmid 1995). key (Groombridge 1990). A resident population Speaking about Florida turtles, Ehrhart and Witherington immatures and adults may reside in Mussulo Bay, Angt (1992) stated that "Juveniles 2-60 kg (4-130 Ib) forage as (Can" and Can 1991). On a map in Can et al. (198 herbivores in shallow coastal waters before abandoning Well-known adult benthic foraging habitats of the Can this developmental habitat as sub-adults." The waters off bean green turtles are depicted in Mexico, Nicaragi the North Carolina coast may provide important develop- Colombia, Venezuela and in some places in the Less mental habitats for loggerhead, green and Kemp's ridley Antilles. sea turtles (Epperly et al, 1995a. b). Bermuda, now. is The distribution of adults is determined to a large e strictly a developmental habitat (Meylan et al. 1992a). tent by the locations of their nesting beaches and feedi The evidence for a regular summering population of grounds. Nesting sites are shown in Figs. 9. 10, and 1 immatures in Nantucket Sound is problematical (Laze11 and nesting seasons are given in Table 1. All the maj 1980). Collazo et al. (1992) reported a juvenile and sub- and minor nesting sites (except those in Turkey) are 1 adult feeding population in Puerto Rico. Evidence is ac- cated between 30° and 30° latitudes. All nesting pop cumulating that there is a green turtle developmental habi- lations are important and need to be conserved. tat off the Texas coast (Coyne and Landry Jr. 1994). Forty- A few pertinent comments on some well-known ne, nine juvenile and subadult turtles have been captured there ing beaches and adult distributions in the Pacific, Indi with straight line carapace lengths ranging from 22.2 to and Atlantic Oceans are given here. In addition to t 81.5 cm. Possibly, there is an immature feeding popula- Xisha Islands, Zhao and Adler (1993) list the occurren don in Greece (Margaritoulis et al. 1992). Williams (1988) of green turtles off the Chinese mainland provinces described the feeding behavior of an immature popula- Shandong, Jiangsu, Zhejiang, Fujian. Guangdong a tion in the U.S. Virgin Islands. Kaneohe Bay on Oahu, Guangxi and off the islands of Taiwan and Hainan. T Hawaii, harbors a feeding population of about 500, mostly KO Adang Group in Thailand is now part of the Tarut immature green turtles (Balazs et al. 1993). There is a National Park. A brief history of the Park and the s foraging ground off southern Peru where 89% of 416 turtle work is provided by Howlett (1982). The Saraw turtles in a sample were immatures (no hatchlings ) (Brown Turtle Islands are Satang Besar, Talang Talang Besar sa and Brown 1982). On a map in Can- et al. (1982). well- Talang Talang Kecil and the three principal nesting sit known Canibbean green turtle developmental foraging in the Sabah Turtle Islands (now incorporated into t habitats are plotted in Mexico, Nicaragua, Costa Rica, Turtle Islands National Park) are Pulau Selingaan, Pul Panama, southern Bahamas, and in some places in the Gulisaan and Pulau Bakkungan Kecil. According Lesser Antilles. Mortimer et al. ( 1993) important foraging areas for gre! Lanyon et al. (1989) reviewed the Australian literature turtles and hawksbills are found along the coasts oft and stated that on the seagrass habitats of the Macarthur States of Melaka and Negeri Sembilan in Malaysia. Lie

River Delta, Cleveland Bay, Shoalwater Bay, Repulse Bay and Chan (1 993) state that the island of Pulau Redang, 1 and Moreton Bay, large immatures and adults predomi- km off the coast of Terengganu, provides nesting habii nate; but small-to-medium-sized immature turtles domi- for the largest concentration of green turtles in Penins nate the population structure in coral reef habitats off lar Malaysia. Heron Island, Cairns and eastern Torres Strait. The turtles The major nesting sites in the Philippine Turtle Islan feeding on Moreton Banks, Queensland, are mostly im- are Baguan Is.. Taganak Is., Langaan Is.. Ore . mature individuals, but this population may be in a state Bakkungaan Is.. LihiiIs., and Boaan Is. A small pop of recovery from past overharvesting (Limpus et al. lation of green turtles nests on several beaches on Wa Fie. 9. Major nesting sites (solid circles; more than 500 nest annually, or an average of 500 if nesting is cyclic) and minor nesting sites (open circles; from 100 to 500 nest annually) of the green turtle. l=Xisha Is. (Groombridge and Luxmoore 1989) 2=Ko Khram (Groombridge and Luxmoore 1989) 3=Tharnihla Kyun (Groombridge and Luxmoore 1989) 4=Terengganu State (Groombridge and Luxmoore 1989) 5=Sarawak Turtle Islands (Hendrickson 1958; Groombridge and Luxmoore 1989) 6=Sabah Turtle Islands (de Silva 1982; Groombridge and Luxmoore 1989) 7=Phil- ippine Turtle Islands (Trono 1991; Pawikan Conservation Project Staff 1993) 8=Aceh and North Sumatra Province (Salm 1984 in Groombridge and Luxmoore 1989)'9=Riau Province (Schuiz 1987 in Groombridge and Luxmoore 1989) 10=West Sumatra Province (Salm 1984 in Groombridge and Luxmoore 1989) 1l=South Sumatra Province (Schulz 1987 in Groombridge and Luxmoore 1989) 12=West JavaProvince (Salm 1984 and Schulz 1987 in Groombridge and Luxmoore 1989) 13=East Java Province (Schulz 1984 and 1987 in Groombridge and Luxmoore 1989) 14=West Nusa Tenggara Province (Schulz 1989 in Groombridge and Luxmoore 1989) 15=West Kalimantan Province (Schulz 1987in Groombridge and Luxmoore 1989) 16=Central Kalimantan Province (Salm 1984 in Groombridge and Luxrnoore 1989) 17=South Kalimantan Province (Schuiz 1987 in Groombridge and Luxmoore 1989) 18=East Kalimantan Prov- ince (Schulz 1984 in Groombridge and Luxmoore 1989) 19=North Sulawesi Province (Salm 1984 in Oroombridge and Luxmoore 1989) 20=Central Sulawesi Province (Salm 1984 in Groombridge and Luxmoore 1989) 21=South Sulawesi Province (Schulz 1989 in Groombridge and Luxmoore 1989) 22=Southeast Sulawesi Province (Salm 1984 in Groombridge and Luxmoore 1989) 23=Maluku Province (Schuiz 1989 in Groombridge and Luxmoore 1989) 24=Irian Jaya Province (Salm 1984 in Groombridge and Luxmoore 1989) 25=Northwest Cape-Barrow Is. complex (Prince 1993; Limpus, pers. comm.) 26=Lacepede Is. complex (Prince 1993; Limpus, pers. comm.) 27=Wellesley group (Limpus 1982a) 28=Raine Is.-Moulter Cay complex (Limpus et al. 1993) 29=Capricorn-Bunker Group (Limpus 1980) 3Mgasawara Is. (Suganuma 1985) 31=Merir Is. (Pritchard 1977; Johannes 1986) 32=Helen's Reef (Pritchard 1977; Johannes 1986) 33=Manus Province Is. (Spring 1982a) 34=Long Is. (Spring 1983) 35=Bikar Atoll (Fosberg 1969, 1990; Hendrickson 1972) 36:=d'Entrecasteaux Reef system (Pritchard 1982% 1987, Anon 1989) 37=Palmerston Atoll (Powell 1957; Groombridge and Luxmoore 1989) 38=Scilly Atoll (Lebeau 1985) 39=French Frigate Shoals (Anon. 199la). Fig.10. Major nesting sites (solid circles; more than 500 nest annually, or an average of 500 if nesting is cyclic) anc minor nesting sites (open circles; from 100to 500 nest annually) of the green turtle. l=Islas Revillagigedo(Brattstron 1982;Awbrey et al. 1984) 2=Galgpagos Is. (Green 1983) 3=Colola and Maruata(A1varado and Figueroa 1990)4=Play~ Naranjo (Cornelius 1976) 5=Southeast Florida (Ehrhart and Witheringon 1992) 6=Yucatan Peninsula (Marquez in Ogren 1989) 7=Tortuguero (Can- et a1 1982) 8=Shell Beach (Pritchard 1969) 9=Suriname (Schulz 1982) 10=Frencb Guiana (Fretey 1984) 1 l=Dominican Republic (Ottenwalder 1981 in Groombridge and Luxmoore 1989) 12=Aves Is. (Medina and Sole in Ogren 1989) 13=Atol dm Rocas (Bellini et al. 19%) 14=Ilha de Trindade (Moreira et al. 1995).

An Island, Peng-Hu Archipelago, Taiwan (Chen and for the past 1,130 years. In addition to these well-known Cheng 1995). Green turtles are known to nest on islands nesting sites. Miller and Limpus (1991) list several is- in the Ryukyu Archipelago (Kikukawa et al. 1996). The lands off northeast Australia where several dozen turtles Indonesian nesting sites include several beaches. It is were recorded in the breeding season. About 5,000 nest estimated that between 25,000 and 35.000 females nest annually in the Capricorn - Bunker group (Limpus and annually in Indonesia (Groombridge and Luxmoore 1989). Fleay 1983). Lithou Cays, Magdeline Cays, Diamond Thereis some nesting by green turtles on Inggresau Beach, Islets and Wilis Islets are green turtle nesting sites of Yapen Island, Irian Jaya Province in Indonesia undetermined density, off Queensland, and should be in- (Maturbongs et al. 1993). Villagers here say the nesting vestigated (Lipus 1980). season is April to July. Pritchard (1977) estimated that several dozen nested There are five major nesting concentrations in Austra- on a good night on Merir Is. and Helen's Reef. The peak lia. Limpus (1982a) states that the annual nesting popu- nesting season appears to be in the northern hemisphere lation in the Wellesley group (including Bountiful and summer. (1976) reported that a foreign fishing Pisonia Islands) is usually thousands. Limpus et al. (1993) vessel had illegally caught over 214 green turtles in ten declare that the Raine Is. - Moulter Cay complex (includ- days at Helen's Reef. Rodda et al. (1991) list the green ing Bramble Cay, and No. 7 and No. 8 Sandbanks) is the turtle as occurring on Cocos, Guam, Rota, Tinian and largest green turtle nesting aggregation in the world. The Saipan in the Mariana Islands. The annual nesting popu- annual nesting population varies between a few hundred lation in the Ogasawara Islands was estimated at between to tens of thousands. Low (1985) reported that 11,467 43 and 162 between 1985 and 1993 (Horikoshi et al. 1994). turtles were on Raine Is. at one time. Limpus (1987) de- In a September, 1988, survey of Pitoar Atoll (=Bikar), scribes how Raine Is. has probably been a nesting ground Jemo Island andAdkupAtol1, 176.53 and 49 sets of turtle Fig. 11. Major nesting sites (solid circles; more than-500 nest annually, or an average of 500 if nesting is cyclic) and minor nesting sites (open circles; from 100 to 500 nest annually) of the green turtle. l=Ascension Is. (Mortimer and Can- 1987) 2=Bijagos Archipelago (Limoges 1991 in Agardy 1991) 3=Bioko Is. (Adada, pers. corn; Butynski, pen. comm.) 4=Angola (Can- and Carr 1991) 5=Southeast Turkey (Groombridge 1990; Baran et al 1991; Coley and Smart 1992; Society for Protection of Nature 1992) 6=Southern Somalia (Goodwin 1971) 7=Maziwi Is. (Frazier 1982a) 8=Aldabra Atoll (Seabrook 1991) 9=Assumption. Astove and Cosmoledo Islands (Mortimer 1984) lO=Moheli Is. (Frazier 1985) 1l=Mayotte Is. (Frazier 1985) 12=Primeiras Is. (Hughes 1974a) 13=Europa Is. (LeGall et al. 1986) 14=Tromelin Is, Wallet al. 1986) 15=St. Brandon (Hughes 1974a, 1976) 16=Karan Is. (Miller 1989) 17=Jana Is. (Miller 1989) 18=Shihr and Shuhair (Hirth and Can 1970) 19=Shanna and Ithrnun (Hirth and Carr 1970) 20=Ras al Madrakah and Salalah (Ross and Barwani 1982) 21=Masirah Is. (Ross 1985) 22=A1 Ashkara and Ras Jibsh (Ross and Barwani 1982) 23=Ras al Hadd (Ross and Barwani 1982) 24=Damanyat Is. (Ross and Barwani 1982) 25=Makran Coast (Groombridge et al. 1988) 26=Hawkes Bay and Sandspit (Kabraji and Firdous 1984 in Groombridge and Luxmoorc 1989) 27=Gujarat State (Bhaskar 1984 in Groombridge and Luxmoore 1989) 28=Lakshadweep (Kar and Bhaskar 1982) 29=MaIdives (Frazier 1990) 30=Chagos Archipelago (Frazier 1990) 31=Andaman and Nicobar Is. (Bhaskar 1979; Bhaskar 1984 in Groombridge and Luxmoore 1989). tracks respectively were counted (Maragos 1994). A few turtles on Huon on 11 December 1991 and counted 1,800 green turtles have been seen in the lagoon on Caroline tracks there. They also counted 572 tracks on He Fabre Atoll, Southern Line Islands, and a few may nest on the and 310 crawls on lie Surprise. The nesting population Atoll (Kepler et al. 1994). on Scilly Atoll in the mid-1980's was estimated at be- There is scattered nesting in many areas of Papua New tween 300 and 400 (Lebeau 1985). Balazs et al. (1995) Guinea. Spring (1982a) mentions Mussau Is. in New Ire- estimated a similar number of turtles nested there in 1991. land Province and several islands in Milne Bay Province Up to fifty adult turtles of both sexes are consumed annu- where an undetermined number of green turtles nest. In ally under special governmental permission. Consider- New Caledonia, the d'Entrecasteaux Reef system includes able incentive exists for poaching because an adult green the Islands of Surprise, Leizour, Pabre and Huon. turtle can be illegally sold in Tahiti for about US $1000 Pritchard (1994) and his associates tagged 149 green (Balazs et al. 1995). The nesting population in nearby Table 1. Nesting locations and nesting seasons of green turtles. Parentheses indicate peak nesting months. Month Location J FMAMJ J ASOND Reference Western Pacific Ocean China

Nine Dragon's Beach JJASO Morion (1992)

Xisha Is Thailand

Ko Khram J F (MA MJ J &S)O N D Penyapol(1958) KO Adang J FMA N D Polunin (1975) Malaysia Peninsula, east coast Leong and Siow 1984 in Groombridge and Luxmom (1989) Sarawak J FMAMJ &A)SOND Hendrickson (1958). Chin (197 Sabah J FMAMJ (JAS0)ND deSilva (1970) Indonesia South Natuna Is. Schulz 1987 in Groombridge ai Luxmoore (1989) Tambelan Arch. Schulz 1987 in Groombridge a Luxmoore (1989) J FMAMJ (JAS0N)D Schub 1984 in Groombridge ar Luxm(1989) Aru Is. J FMAMJ JASOND Compost 1980 in Groombridge and Luxmoore (1989) Sambas-Paloh JFMAMJ{JA)SOND Schulz 1987 in Groombridge ar Liixmom (1989) J FMAMJ JASOND Schuiz 1984,1987 in Groombridge and Luxmoore (1989) Sukamade J FM)AMJ JASON(D Suwelo (1975) Al-Ketapang AMJ J Nuitja and Laze11 (1982) Mubrani-Jonsoribo A M Adipati and Patay (1 983) Philippine Turtle Is. J FMA(MJ J AS)OND Dommtay (1952-53) Taiwan, Wan-An Is. J (J A) S 0 Chen and Cheng ( 1995) Ogasawara Is. M (J J) A Suganuma (1985) Merir Is. and Helen's Reef J FMAM(JJ)ASOND Pritchard (1977) Papua New Guinea Manus Province MI 1 AS Spring (1982a) Long 1s. J F MA(MJ J A)S OND Spring (1983) Milne Bay M A Spring (1982a) Eastern Australia J) f M 0 N (D Bustard (1 974). Kowarsky (1978). Stoddart a al. (1981) Caroline Is. MAMJ 3 AS McCoy (1982)

Solomon is. J F M) A M J J A (S 0 N D Vaughan(1981) Vanuatu J S 0 N D Pritchard(1982a) d'Entrecasieaux Reef J F N D Priiehard (19871, Anon (1989) Central And Eastern Pacific Ocean Bikar Atoll J J A Posherg (1969). Hendrickson (1972). Pritchan! (1982b) Phoenix Is. J F M A M J J A S (0 N) D Balazs (1975) Fiji J F N D Pritchard (1982a) Tokeiau Is. SON Balazs (1983b) Table 1. Continued. Month

Location M3 3 Rose Atoll

Bnga Reach Frigate Shoals Scilly Atoll Society Is.

LÂ¥.la Revillagigedos GaUpagos Is. Mexico. Colola and Maruata Guatemala El Salvador

Costa Rica, Playa Namjo Panama Ecuador Western Atlantic Ocean USA, Southeast Florida

Eastern Mexico Mexico, Contoy Is. Mexico, El Cuyo Belize Honduras

Costa Rica, Tonuguem Panama Colombia

Venezuela

Guyana, Shell Beach Suriname Bench Guiana Timil Mainland

Praia do Forte At01 das Rocas Trindnde and Fernando de Noronha Is, FMA

Bahamas

. . Turks and Caicos Is. Cuba Haiti Dominican Republic ".

Puerto Rico British Virgin 1s. - U. S. Virgin Is. Table 1. Continued. Month

Location J FMAMJ J AS OND Reference 1 ! St. KittsÑNevi MAMJ J AS0 Wilkins and Meylan (1 984) Barbuda MJJASON' Joseph et al. (1 984) Antigua J J Joseph et al. (1984) Guadeloupe AMJ J AS Can- et al. (1982) ' Aves Is. (J J A) Rainey (1971) Dominim JJASO Can- et al. (1982) Martinique MJ J (AS)ON Dropsy 1987 in Groombridge and Luxmoore (1989) St. Lucia MJ J ASc Munay (1984) St. VincentGrenadines AMJ JA Moms (1984) Grenada AMJJ AS Finley (1984) -Trinidad AMJ J A Cheong (1984) Eastern Atlantic Ocean and Mediterranean Sea Ascension Is. J (F M A) M t D Mortimer and Carr ( 1987) Mauritania J Maigret (1983) Senegal JFM JASO Dupuy (1 986) Equatorial Guinea, Bioko Is. J FjMA S 0 N (D Adada, p.comrn.,Butynski, pers. comm., Eisentraut (1964) Angola N D Can-andCan-(1991) Turkey, southeast coast JJA Baran et al. (1991) Israel AMJ J Sella (1 982)

JJAS Demetropoulos and Hadjichristophamu (1982). Godley and Broderick (1993) Western Indian Ocean and Red Sea Dahlak Arch. M Urban (1970) Somalia SON Karaani, pers. comm. Maziwi Is. JFMAMJ JASO Frazier (1 984b) Primeiras Is. J F D Hughes (1971) Aldabra Atoll J FMA)MJ J ASO(ND Seabrook (1989a) Assumption Is. FMAMJ J ASOND Frazier (1984b) Cosmoledo Atoll FMAM Frazier (19B4b) Astove Atoll J FMAMJ J Frazier (1 984b) Moheli Is. J F M A M (J)J Frazier (I 985) Mayotte Is. FMAMJ J Frazier (I 985) lies Glorieuses (M J J) Vergonzanne in Groombridge and Luxmoore ( 1989) Eumpa Is. Servm (1 976) Tromelin Is. Hughes (1974~).LeGall et al. (1986) St. Brandon Is. Hughes (1 974a, 1976) Saudi Arabia Ras Baiidi MJ J AS Miller (1 989) Karan Is. J (J) A S Miller (1989) Yemen Abdul Wadi J FMAMJJAS(0ND) Hinh and Cam (I 970) Shuhoir (0 N) Hirth and Cam (1970) Shihr (0 N) Hinh and Can- (1 970) Shtnna J F MAMJ J AS (0N)D Hirth and Can (1970). FA0 ( 1973) J FMAMJ J ASOND Hirth and Can- (1970). FA0 (1973) Table 1. Continued. Month

Location J FMAMJ J ASOND Reference Oman Masirah 1s. J J J A (S 0 N) D Ross and Barwani (1982) Ras a1 Hadd J F M A M J J (A S 0 N D) RossandBwwa~(1982) Northern and Eastern Indian Ocean Pakistan Makran coast JFM S 0 N D Groombridgeei al. (1988) Hawkes Bay and Sandspit J F M A M J J A (S 0 N) D Kabmji andKidous 1984 in Groombridge and LBXITIOOIC (1989) India Gujmt J J A S 0 N D Bhaskar I984 in Groombridge and Luxmoore (1 989) Lakshadweep J J AS Kar and Bhaskar (1982) Andanurn and Nicobar Is. J F M A (M J J A S) 0 N D BhÈskar(1979 Maldives J F M A M J J A S 0 N D FrazierandMer19871n Groombridge and Liixmoore (1 989) dagos Arch. J JAS Fmzier (1990) Sn Lanka, Kosgoda J F M (A M) J J A S 0 N D Dattatri and Saniarajiva 1983 in Groombridge and Luxmoore (1989) Bangladesh J F 0 N D Khan 1985 in Groombndge and Luxmoore (1 989) Burma, Thamihln Kyun J F M A M J (J A S 0 N) D KarandBhskar(1982) J F N D Prince (1993)

Mopelia Atoll should be investigated since Sachet (1 983) ary 1993, 848 green turtle tracks were counted (Arauz- was informed by plantation workers that they take about Almengor and Morera-Avila 1994). There is low-level, 200 green turtles annually on Mopelia. Green turtles are sporadic nesting year-around on the OsaPeninsula, Costa reported to nest on Nukutipipi Atoll, in the Tuamotu AT- Rica (Drake 1996). chipelago (Salvat and Salvat 1992). Apparently, green In southeast Florida, the greatest nesting concentrations turtles lay eggs on the few beaches on Henderson Island are on Melbourne Beach, Hutchinson Island and Jupiter (Quayle 1922 in Fosberg et a1 1983). Based upon obser- Island. Whether or not the numbers of green turtles in vations made in 1991-1992. a little nesting (about ten fe- Florida are slowly increasing is a moot issue (Dodd 1982a. males per season with peak nesting January-March) oc- 1995;Thompson 1988,1991; Magnuson et al. 1990; Dodd curs on Henderson bland (Brooke 1995). Weisler (1995) and Byles 1991; Ehrhart and Witherington 1992). When documents that nesting green turtles and their eggs were finalized, the Archie Carr National Wildlife Refuge in taken by prehistoric humans on Henderson Is. south Brevard and north Indian River counties will en- Between 1,200 and 3,500 females nest annually in the compass about 40% of the green turtle nesting area Galapagos Is. (Green 1983). Most nesting, as far as is (Ehrhart and Witherington 1992). hi 1992, 12,754 log- known, takes place on Isabela, Baltra, Santa Cruz and gerhead and a record high of 686 green turtle nests were Santiago Islands. In 1989-90 an estimated 1,280 turtles laid in the Refuge (Anon 1993b). Johnson and Ehrhart nested in Michoacan State, Mexico, with about 967 of (1 995) stated that the proposed Archie Carr National Wild- these at Colola and Maruata beaches (Alvarado and life Refuge "may very well produce more Florida green Figueroa 1990). Nesting activity continues on Playa turtle hatchlings annually than any other beach in the Naranjo, Costa Rica. Between October 1989 and Janu- state."

23 Ogren (1989) calculated about 283 to 420 females nest Fernando Pw, SbTho&, Principe, Rolas, Congo, Zaire annually on Mexico's Gulf of Mexico and Caribbean Angola, and on rare occasions on St. Helena." Needless beaches and most of the nesting is probably on theYucatan to say these nesting locales, if still extant, need to be ac- Peninsula and nearby islands. curately censused. Accwding to Graff (1995) green turtles The Tortuguero, Costa Rica, nesting beach is 35 km nest on several beaches in Sa6 Tome with peak nesting long and extends between the mouths of the Tortuguero probably between November and January. Green turtles and Parismina Rivers. The most concentrated nesting is have been reported nesting in Gabon (Fretey and Girardin about midway between these two river mouths. The esti- 1989). Green turtles are still somewhat common in the mated number of females nesting annually between 1971 foraging pastures off Mauritania and Senegal (Maigret and 1981 ranged between 5,178 and 52,046 (Can- et al. 1983 and Dupuy 1986, respectively) and some nesting is 1982). The nesting population at Tortuguero is now in its reported there. Green turtles have been reported as breed- fortieth consecutive year of study. The major Suriname ing in the Cape Verde Islands (Parsons 1962) but beaches are near Galibi, Matapica and Krofajapasi. It is Brongersp (1982) suggested that the species may have estimated that between about 1,000 to 3,000 green turtles been misidentified and he recommended verification. nested in Suriname annually from 1968 to 1979 (Schulz Sandy s-Winsch and Harris ( 1994) mention only logger- 1982) and between 1,464 and 2,160 annually between heads and hawksbills as breeding on the sandy beaches in 1983 and 1987 (Mohadin in Ogren 1989). Some of the Cape Verde Islands, and these two species are hunted for better known nesting sites in French Guiana are: Awara, their eggs and meat. Farez Kawana, Les Hattes and Pointe Isere (Fretey 1984). The southeast Turkey beaches, especially the beaches In the Dominican Republic there is scattered nesting on at Kazanli, Akyatan and Samandagi, harbor the largest many beaches along the coast, with several dozen nesting green turtle nesting aggregation in the Mediterranean Sea. in the Pedernales and La Altagracia Provinces. The 1980 It is estimated that several hundred nest here annually. nesting population in the country was estimated at be- Groombridge (1990) estimated that about 75 females nest tween 160 and 360 (Ottenwalder 1981 in Groombridge annually in Cyprus, about 25 and 50 on the south and and Luxmoore 1989). The annual average number of nest- north coasts, respectively. Godley and Broderick (1994) ing females in the mid-1980's on Aves Is. was about 376 estimated that 154 green turtles nested on the beaches of (Medina and Sole in Ogren 1989). Northern Cyprus in 1994 (compared to about 29 and 107 At01 das Rocas, in Brazil, is a minor nesting beach. It in 1992 and 1993 respectively). Some important nesting is estimated that about 20 green turtles nest annually on beaches in Cyprus are near Alagadi, Chelones Bay, nearby Fernando de Noronha (Wells 1987) but more de- Akdeniz, Esentepe and the Karpaz Peninsula. According tailed nesting surveys are currently being conducted in to Demetropoulos and Hadjichristophorou (1995) "It is the Archipelago (T. Sanches, pers. comm.). According to assumed that the green turtle population nesting in Cyprus the literature references cited in Groombridge and does not exceed 200 turtles, the north and east coast Luxmoore (1989) the degree of nesting on the mainland beaches combined." Recent observations in Eritrea re- of Brazil is debatable and needs reevaluation. D'Amato ported by Hillman and Gebremariarn (1995) indicate that and Marczwski (1993) discovered that a few green turtles green turtles may nest there in April, May and June, in nested at Praia do Forte on the coast in the State of Bahia addition to the nesting noted there in March by Urban but the vast majority of nesters here were loggerheads. A (1970). Aldabra Atoll is a major nesting site and in 1982 total of 23 green turtle neste were made at Praia do Forte it was designated a.World Heritage Site. Accounts of the between 1987 and 1993 and the overall nesting season ecology and geology of the Atoll are in Stoddart (1967, was from August to April (Marcovaldi and Laurent 1996). 1971). The estimated annual nesting population on Moheli Based upon current information, Dhe de Trindade is the is 1.800 (Frazier 1985). Maziwi Island, Tanzania, is re- main nesting site for green turtles in Brazil (Moreira et ported to have been washed over by the sea (Anon. 1982). al. 1995). About 1,800 nests are made annually on the A legend about Maziwi told by the Arabs along the Tan- Island and the peak nesting season is from January through zania coast is that if Maziwi Island should ever disappear March. beneath the sea the world will come to an end (Anon. Ascension Island is a major nesting site. About 1,650- 1969). Green turtles nest at some sites along the north- 3,000 nested annually in the late 1970's (Mortimer and east and west coasts of Madagascar, between October and Carr 1987). There is a dearth of confirmed major and January, but the current sizes of the nesting populations minor nesting beaches on the west coast of Africa. How- need to be determined (Rakotonirina and Cooke 1994). ever, Brongersma (1982) was of the opinion that "from Miller (1989) estimated that 1,350turtles nested on Karan the list of records it is clear that C. mydas breeds in many Is., Arabian Gulf, in 1986 and 450 in 1987. Jana Is.. places: perhaps in southernmost Morocco, but definitely nearby, is a minor nesting site. Other islands in the Saudi in Mauritania, Senegal, Sierra Leone, Liberia. Ghana. Arabian sector of the Gulf of Arabia where nesting oc- curs are Kurayn Is., Harqis Is., and &ayd Is. On the their intemesting period of 10-1 1 days. They state that Saudi Arabian side of the Red Sea some nesting occurs in "they hardly ventured away from the vicinity of their nest- the vicinity of Tiran Is., the Wejh Bank, along the, coast ing beach except turtle T2 which moved longshore 6 km north ofYanbu, on the offshore islands south of Qunfadah, south from her nesting site. She later returned to the vi- and on the islands off the Farasan Bank. The largest single cinity of her nesting site a few days prior to renesting." nesting site is at Ras Bairdi, just north of Yanbu where Liew and Chan (1993) also found that the turtles spent I between 50 and 100 green turtles nest from May through the majority of their time resting on the seafloor at a mean September (Miller 1989). Ross and Barwani (1982) esti- depth of 10 m; that none of the turtles were observed for- mate that 6,000 green turtles nest annually at Ras al Hadd aging; and that one was seen mating during the intemesting in Oman. Salm and Salm (1991) estimate that up to20,000 period. green turtles nest annually in Oman, with most nesting along a 45 km stretch of coast from Ras al Hadd south to 2.3 Determinants of Distributional Changes Ras al Khabbah. Didi (1 993) summarizes some of the Green turtles are commonly found in warm tropical seas current information about sea turtles in the Maldives. with major concentrations in their feeding pastures and at During a February-May, 1991, nesting beach survey of their nesting beaches. six islands in the southern Nicobars, Tiwari (1 994) saw Distribution of hatchlings and posthatchlings may be twelve green turtle crawls and six body pits but the vast affected by changes in ocean currents, gyres and strong majority of nesting was by leatherbacks (433 nests). El-Niiio years (section 2.2.1). Factors affecting repro- Current accurate surveys should be conducted at some duction and mortality can affect distribution (sections 4.3.2 sites that were censused decades ago. Nesting is reported and 4.4.2). The distribution of green turtles was certainly at other sites around the tropics but quantitative data are more extensive before human interference and exploita- lacking. tion. The intemesting habitat (see Fig. 8) needs more study since some individuals may spend up to three months here 2.4 Hybridization as they renest at about thirteen day intervals. Most mat- Turtles hatched from a clutch of Suriname eggs appeared ing takes place in, or near, this habitat (but there are ex- to be hybrids of Chelonia mydas and Eretmochelys ceptions). Little is known about the duration of males in imbricafa and all the hybrids were males (Wood et al. the intemesting habitat. Ecologically, this habitat may be 1983a). These hybrids, raised in the Cayman "RirtleFarm, quite different from the shallow water feeding habitat. had high mortality and were more susceptible to lung in- Characteristics of the internesting habitat may provide fection than captive green turtles but at least one male cues to homing turtles. The internesting habitat, along hybrid survived to maturity, had motile sperm, and was with the nesting beach, can be an area of intense human observed mating with resident female green turtles (Wood contact, directly or indirectly (e.g., recreational use, fish- in Karl et al 1995). ing, run-off pollution). Karl et al. (1995) employed molecular genetic assays Dizon and Balazs (1982) found that the intemesting to document the natural occurrences of interspecific hy- habitat of females (and males) at French Frigate Shoals, brids between a male green turtle and a female logger- Hawaii, was in close proximity to the nesting and bask- head (N=4 hatchling clutch-mates collected in Brazil) and ing islands where the turtles were captured. At Tortuguero, between a female green turtle and a male hawksbill (N=1 Meylan (1982a) found that the intemesting turtles trav- individual collected in Suriname). The species involved elled parallel to shore, within the 24 meter contour line, in these hybridizations represent evolutionary lineages and that the maximum longshore distance moved from thought to have separated about 50 million years ago and site of nesting was 10 km. At Ascension Is., Mortimer thus may be among the oldest vertebrate lineages capable and Portier (1989) tracked a few individuals and found of producing viable hybrids in nature (Karl et al. 1995). that after oviposition most of them moved to a shallow Some potential problems of intraspecific hybridization, area off the northwest coast of the island. Alvarado and as a consequence of manipulative management practices, Rgueroa (1990) conducted preliminary studies on turtles are discussed by Can- and Dodd (1983). The F2 green in the internesting habitat off the Pacific coast of Mexico. turtle hatchlings produced on the Cayman Turtle Farm A nesting female at Maruata Beach was fined with a ra- are most likely intraspecific hybrids (see section 7). dio transmitter and over the next seven days she remained in Maruata Bay. Another nester from Colola Beach was 3. BIONOMICS AND LIFE HISTORY radio-tracked as she moved 10 km east of the nesting beach 3.1 Reproduction over two days. Liew and Chan (1993) found that three Malaysian females tracked by means of radio and ultra- 3.1.1 Sexuality sonic telemetry remained within 1 km of the coast during Green turtles are bisexual but sexual dimorphism is ex- teraally apparent only in large subadults and adults. Adult turtles, males have a much thicker and longer tail than females. The male's tail extends well beyond the posterior marein 3.1.2 Maturity of the carapace and it is somewhat prehensile. Mature The age at sexual maturity may vary among individu- males have longer claws on the foreflippers than do fe- als of the same population and among individuals of dif- males. Adult males are sometimes viewed as having a ferent populations. The age at maturity has been esti- more elongate carapace than females, especially in east mated at more than 30 years in Australia (Limpus and Pacific populations (Can-1952; Stebbins 1985). In a small Walter 1980); 25 to 30 years in Florida (Mendonga 1981); sample (llm8f') of adult green turtles caught off about 27 years in Bermuda (Burnett-Herkes et al. 1984); Michowan, Mexico, males had soft plastra while females 18 to 27 years in Florida (Frazer and Ehrhart 1985); 40 to exhibited well-cornified plasm (Wibbels et al. 1991). 50 years in Hawaii (Zug and Balazs 1985); 12 to 26 years Mature females sometimes possess a "mating notch" in in Costarica, 17 to 35 years at Ascension Island, 27 to the shoulderregions of the carapace, produced by the grip- 33 years in the Virgin Islands, and 24 to 36 years in ping of the male's claws during copulation. Suriname (Frazer and Ladner 1986); about average of 25 Evidence is slowly accumulating that mature males are years in Hawaii (Anon 199 la); 19 to 24 years in the west- smaller (i. e. shorter carapace length) than mature females em central Atlantic Ocean (Ehrhardt and Witham 1992); in the same population: Witzell (1982) measured a few and, at least 36 years in the Galapagos Islands (Green foraging adult turtles off Western Samoa and found males 1993). These estimates are much longer than some of the (mean straight carapace length=92.2 cm, range=86.5- 102 earlier estimates cited in Hi& (1971b). The mean age of cm, N=5) were smaller than adult females (mean straight nesting on the Cayman Turtle Farm is 16 years (Wood carapace length = 96.9 cm, range=91.5-109 cm, N=9); and Wood 1993a). Frazier (1984b) found breeding males (MCCL [mean Carr and Cm(I 970b) discussed the variation in matu- curved carapace length] =a. 102 cm, range ca. 94-1 16 ration size in the Tortuguero nesting colony. There is much cm, N=84) were smaller than breeding females information available on the sizes (carapace lengths) of (MCCkca. 108 cm, range=ca. 97-120 cm, N=54) in the nesting turtles (Table 2) and some information on the Aldabra Atoll population; Limpus and Reed (1985a) stated weights of nesters (Table 3). As mentioned in the preced- that adult males (MCCL98 cm, range=90.5- 105.5 cm, ing section, evidence is accumulating that mature males N=24) in a feeding population off Heron Island, Austra- are smaller than mature females in some populations. lia, were significantly smaller than adult females (MCCL103 cm. range=91.5-109.5 cm, N=16); Limpus 3.1.3 Mating and Reed (198%) recorded that adult males (MCCL-98.4 Most mating takes place in the vicinity of the nesting cm, range 92.5-104.5 cm, N=9) in a feeding population beach, but there are exceptions. Ross (1984) noted occa- in the Gulf of Carpentaria, Australia, were significantly sional mating on a feeding pasture near Masirah bland. shorter than adult females (MCCL=107.6 cm, range 98- Meylan et al. (1992b) reported that some Tortuguero 123.5 cm. N=18); Miller (1989) said adult males Beach-bound turtles mate approximately 240 km distant (MCCL=94.7 cm, range=87-103 cm, N=14) were in Panama. In the southern system, signficantly smaller than adult females (MCCL=100.6 cm, Limpus (1 993b) observed that mated females dispersed range=%%-113 ern, N=21) in a feeding pasture at Dawhat from the courtship area to beaches within 92 km, without Abu Ali, Saudi Arabia; Miller (1989) found breeding necessarily nesting on the closest nesting beach to the males (MCCk91.3 cm, range=84-96 cm, N=21) to be courtship area. A female tagged while mating off one significantly smaller than mature females (MCCL=98.8 atoll in Yap State subsequently nested at another site 101 cm, range=81.5-108.5 cm, N=43) in a nesting population krn distant from the mating event (Kolinski 1995). at Karan Island, Saudi Arabia; and Limpus (1993b) stated It is unknown whether the males accompany the females that breeding males (MCCL=100.6 cm, range=89.5-114.5 from the feeding pastures to the nesting beach or whether cm, N=361) in the southern Great Barrier Reef popula- they make synchronized rendezvous with females off the tion were smaller than mature females (MCCL=107 cm, beach. range=91-124 cm, N=1942) in the same population. Evidence is accumulating to indicate that sperm pro- The sex of hatchlings, juveniles and small subadults duced during the copulations, usually at the beginning of can be determined by dissection, histological examina- the nesting season, is used only for the current season's tion, radioimmunological assays or by laparoscopy ovulations (Booth and Peters 1972; Owens 1980; Wood (Owens 1982; Wood et al. 1983b; Van der Heiden et a]. and Wood 1980; Owens and Moms 1985). Green turtles 1985; Jackson et al. 1987; Wibbels et al. 1993). Wellins in the Caribbean appear to follow a pre-nuptial pattern of (1987) suggested that an H-Y antigen serological assay spermatogenesis: i. e., tesricular recrudescence and ac- may be useful in determining the sex of immature sea tive spermiation precede mating (Engstrom 1994). Mul- Table 2 Carapace lengths (straight line, cm) of nesting green turtles. *curved carapace length, not given. ss. (drca/appmximiely). Location Mean Ranee Reference Pacific Ocean KoKhnun-Thailand Penyapol(1958) Sarawak @ 97.5" Hackson (1 958) Sukumade, Indonesia 99.7 Nuitja (1993-4); pets. comm. Baguan Is, Philippines 99.5* Trono (1991) Wan-An I#., Taiwan 96.6 Chen and Cheng (1995) Long Is, Papua New Guinea 94.7 Spring (1983) Long Is. Papua New Guinea 97.5 Pritchard (1979a) Solomon Is. 85 McKeown (1977) Solomon Is. no* Vaughan(1981) Bramble Cay, Australia 99.7 Kowarsky (1978) Raine Is., Australia 100.2 Kowarsky (1978) Raine Is., Australia 1W Stoddart et al. (1981) Heron Is., Australia Q l06* Bustard (1972) Heron Is., Australia 107* Limpus (1980) lie Huon, New Caledonia 107.6* Anon (1989) He Surprise, New Caledonia 105.8* Anon (1989) Western Samoa 96.9' Witzell (I 982) Scilly Atoll. French Polynesia 95.6** Doumeage (1973) Scilly Atoll, French Polynesia 96.3 Lebeau (1985) French Frigate Shoals, Hawaii 92.2 Balazs (1980) Colola and Maruata, Mexico 82* Alvarado and Figuema (1990) Playa Naranjo, Cosm Rica 82.9 Cornelius (1976) Gal6pagos Is. m81.3 Pritchard (1971) GaKpagos Is. 81.4 Green (1994) Atlantic Ocean and Mediterranean Sea Kennedy Space Center, Florida 100.5 Ehrhan (1979b) Cape Canaveral and Witherington and Ehrhart Melbourne, Florida 101.5 (1989a) Melbourne, Florida 110 Bjomdal et al. (1983) Hutchinson Is., Florida 101.1 Gallagher et al., (1972) Contoy Is., Mexico 99 Nfiera (1991) El Cuyo, Mexico log* Rodriguez and Zambrano (1991) Tortuguero, Costa Rica cal00.31' Can- and Giovannoli (1957) Tortuguero, Costa Rica 100.11' Can- and Ogren (1960) Tonuguero, Costa Rica 100.3~ Can and Hirth (1962) Aves Is. 107.7 Rainey (1971) Shell Beach, Guyana 103.9 Pritchard (1969) Bigi Santi, Suriname 111.8 Pritchari (1969) Suriname 109~ Schulz (1975) Trindade, Brazil 116.8- MordiÃet al. (1995) Praia do Forte, Brazil 123.3* Marcovaldi and Laurent (1996) At01 das Rocas, Brazil 11 8.6* Bellini et at (19%) Ascension Is. 108.1 Can-&Kith (1962) Ascension Is. ll6.8* Simon and Parkes (1 976) Alagadi, Cyprus 92* Godley and Broderick (1993) Kazanli. Turkey ca 96 Baran et al. (1991) Kazanii, Turkey 96* Coley and Sman (1992) Indian Ocean Maziwi Is. 113* Frarier (1984b) Europa Is. 106.5 Hughes (1974a) Europa Is. 108.9 Swan (1976) Mayone Is. 1 10.8* Frazier (1 985) Moheli Is. 1 12.3* ITazier (1985) Table 2. Continued. Location Mean Range N Reference Aldabra Atoll Hirth and Cair (1970) Aldabra Atoll Prazier (1971) Assumption Is. =rth and Can- (1 970) Assumption Is. ftazfcr (1984b) Tmmelin Is. Hugbes (1974a) Ras Bairdi, Saudi Arabia Milter (1989) Koran Is. Saudi Arabia Miter (1 989) Shaima, Yemen Hinhand Cair(1970) Sharma, Yemen FA0 (1973) Masirah Is., Oman Ross and Barwani (1982) RasdHadd,Omari Ross and Barwani (1982) a Non-nesting adult females b Total straight line carapace length

Table 3. Weights (kg) of nesting green turtles (after laying or presumably after laying). Location Mean Range N Reference Pacific Ocean Sarawak Ill 89-126 10 Heridrickson (1958) Long Is., Papa New Guinea 112.7 81-169 100 Spring (1983) Scilly Atoll, hchPolynesia 126.7 75-205 255 Doumenge (1973) French Frigate Shoals, Hawaii 110 68-148 69 Kridler and Olsen in Balazs (1 980) Galfipagos Is. 81.9 45.5-1 72.7 Oreen (1994)

Atlantic Ocean Kennedy Space Center, Florida 139.7 121.5-176.9 10 Ehrhart (1979b) Cape Canaveral and Witherington and Ehrhan Melbourne, Florida (1 989a) Tonuguero, Costa Rica Can and Hilth (1962) Aves Is. Rainey (1971) Bigi Santi, Suriname Priichart (3 969) Suriname Schulz (1975) Ascension Is. Can and Hi (1 962) Indian Ocean Europa k Hughes (1974a) Europa Is. Swan (1976) Tmmelin Is. Hughes (19%) Ras Bairdi. Saudi Arabia Miller (1989) Karan Is., Saudi Arabia Miller (1989)

Hawks Bay, Pakistan ' Minton (1 966)

tiple marines of females during a nesting season have been lished that some females were using sperm from two to reported (Alvarado and Figueroa 1990, 1991; Postal et four males to fertilize their eggs. The authors also very al. 1990) and this behavior, and its significance, deserves briefly discuss how mating systems (e. g., promiscuous more attention. Peare et al. (1994) stated that the preva- vs. monogamous) can influence both the effective popu- lent mating system in the Tortuguero population is pro- lation size and the level of genetic variability. miscuity-i. e., females mate with multiple males to fer- The diminution of observed mating activity offshore of tilize each clutch. Evidence for this came from using Tortuguero nesting beach, after about the middle of the multilocus minisattelite DNA fingerprinting which estab- nesting season, may indicate that males leave the nesting area before females, or that they are just less visible in (1990) reported that escort males in Hawaii, Saudi Arabia some other area of the internesting habitat (Carr et al. and Mexico, respectively, also attempt to dislodge a mat- 1978). Speaking of sea turtles, Owens and Morris (1 985) ing male by biting the flippers and tail. Miller (1989) staled "It appears that mating aggregationspeak just prior noted that the time interval between the sighting of a to,the beginning of the nesting season, and that males sub- mounted pair and the first nesting attempt by the female sequently leave the nesting area." In the southern Great averaged 15 days (range 1-38 days, N=10). Green (1983) Barrier Reef region, males are sexually active for about a noted that coupling lasts up to six hours in the GalApagos. month

3.1.6 Nesting process ' 3.1.7Eggs Nesting behavior is an important aspect of a green Freshly laid green turtle eggs are spherical and white turtle's life history because it relates directly to fitness. with flexible shells. Using scanning electron microscopy, There is a large amount of information available on the Baird and Solomon (1979) showed that the structure of nesting behavior of green turtles. Much of the informa- the calcified layer of egg shells from farm-reared and wild tion is covered in several reviews (Hirth 1971b; Ehrenfeld turtles w somewhat different. Egg shells from fanned 1979; Bu-hart 1982; Hendrickson 1982). It is believed, turtles 2tained distinct regions of blocks of calcite and but not proven, that females nest on the beach where they spherulites of aragonite. The egg shells from feral popu- hatched. Renestings during a nesting season are usually lations consisted only of the spherulites. Shells with the on the same sector of beach. Most females tend to return more open calcitic framework are more susceptible to to the same nesting area on their reproductive migrations. fungal invasion, Aspergillus sp. being one kind (Solomon Most nesting takes place at night (reflecting adaptations and Baird 1980). The hyphae may impair gas exchange to diurnal predators and heat stress) on an elevated beach and also may create a calcium deficiency. Gas exchange platform (berm). The turtles nest on beaches that vary between eggs and their ambient environment is discussed markedly in tenns of sand texture, mineral composition in section 3.4.4. and color (Hirth and Can- 1970; Mortimer 1990). Green The average clutch size varies widely (Table 4) but there turtles have deposited eggs on man-made beaches (e-g., is a relationship between clutch size and carapace length Cayman Turtle Farm) and on nourished beaches in Florida (Fig. 12). The females nesting at Colola and Maroata (Witham 1990). But this does not mean that bargeloads (Pacific Mexico), Play a Naranjo (Pacific Costa Rica) and of sand can be randomly imported to restore green turtle in the Galapagos Islands are among the smallest and they nesting beaches. The sand should be of the same type as deposit the fewest eggs per clutch. The nesters in that on the natural nesting beaches in the area: Suriname, Brazil, Europa Island and Maziwi Island are Can- and Ogren (1 960) divided the nesting behavior into among the largest green turtles and they lay large clutches. eleven stages (several of these stages are combined by The diameter of green turtle eggs, from 18 localities, other investigators): 1. stranding, testing of stranding site, ranged from 33.8 to 58.7 nun, with an overall average of and emergence from wave wash 2. selecting of course about 45 nun (N=28 samples) (Table 5). The weight of and crawling from surf to nest site 3. selecting of nest eggs, from 16 localities, ranged from about 21 to 66 g, site 4. clearing of nest premises 5. excavating of body with an overall mean of about 47 g (N=20 samples) pit 6. excavating of nest hole 7. oviposition 8. filling, (Table 6). covering, and packing of nest hole 9. filling of body pit The number of clutches laid in one nesting season can and concealing of site of nesting 10. selecting of course, range from 1 to 9 in some nesting colonies but the overall and locomotion back to the sea 11. re-entering of wave average is about 3.3 (N=21 samples) (Table 7). Green wash and traversal of the surf. turtles renest at about 13 day intervals (Table 8) and they Data in Table 1 indicate that there is year-around nest- occupy an internesting habitat (see section 2.2.2) between ing (but with seasonal peaks) at some sites while at other nesting episodes. Remigration intervals at different nest- locales shorter seasonal nesting prevails. ing colonies are provided in Table 9. Green turtles spend about two and one-half hours on Several researchers have reported a positive relation- the beach for nesting (ffirth 1980~);about two hours in ship between size of nesting female and clutch within a the actual construction and camouflaging of the nest (Hirth nesting population: Homell (1927) on Aldabra Atoll; and Samson 1987). At Tortuguero, green turtles spend Bustard (1 972) on Heron Island; Simon and Parkes (1 976) about 23 minutes in digging the body pit, about 23 minutes and Hays et al. (1993) on Ascension Island; Balms (1980) in digging the egg chamber, approximately 15 minutes in on French Frigate Shoals; Hirth (1988) and Bjorndal and laying eggs, about 12 minutes in filling the egg chamber, Can- (1989) at Tortuguero; Witherington and Ehrhart and, approximately 43 minutes in filling the body pit and (1989a) in Florida; and, Chen and Cheng inTaiwan (1995). camouflaging the nest site. Inexperienced and experienced On the other hand, Hughes (1974a) found no correlation nesters at Tortuguero exhibit similar patterns of nesting and between size of female and clutch size on Europa Island, data indicate strong natural selection far a fixed nesting nor did Miller (1989) find a correlation on Khan Island. behavior in green turtles (Hirth and Samson 1987). Alvarado and Figueroa (1990) found no correlation be- (1976) include an ovary with compact cortical layer and On Ascension Island, nest site selection seems to in- stromal space. Rainey (1981) illustrates the gonads of two volve cues provided by an uneven beach topography- immature males (with carapace lengths of 49 and 64 cm) i. e., turtles usually attempted to nest only after they had two immature females (with carapace lengths of 61.5 and crawled into the uneven beach zone above the spring high 68.5 cm) and one adult female (carapace length 86 cm). water line (Hays et al. 1995a).

3.1.6 Nesting process ' 3.1.7 Eggs Nesting behavior is an important aspect of a green Freshly laid green turtle eggs are spherical and white turtle's life history because it relates directly to fitness, with flexible shells. Using scanning electron microscopy, There is a large amount of information available on the Baird and Solomon (1979) showed that the structure of nesting behavior of green turtles. Much of the inform- the calcified layer of egg shells from farm-reared and wild tion is covered in several reviews (Hirth 1971b; Ehrenfeld turtles what different. Egg shells from fanned 1979; Ehrhart 1982; Hendrickson 1982). It is believed, turtles distinct regions of blocks of calcite and but not proven, that females nest on the beach where they sphemlites of aragonite. The egg shells from feral popu- hatched. Renestings during a nesting season are usually lations consisted only of the spherulites. Shells with the on the same sector of beach. Most females tend to return more open calcitic framework are more susceptible to to the same nesting area on their reproductive migrations. fungal invasion, AspergiUus sp. being one kind (Solomon Most nesting takes place at night (reflecting adaptations and Baird 1980). The hyphae may impair gas exchange to diurnal predators and heat stress) on an elevated beach and also may create a calcium deficiency. Gas exchange platform (berm). The turtles nest on beaches that vary between eggs and their ambient environment is discussed markedly in terms of sand texture, mineral composition in section 3.4.4. and color (Hid and Can- 1970; Mortimer 1990). Green The average clutch size varies widely (Table 4) but there turtles have deposited eggs on man-made beaches (e.g., is a relationship between clutch size and carapace length Cayman Turtle Farm) and on nourished beaches in Florida (Fig. 12). The females nesting at Colola and Maruata (Witham 1990). But this does not mean that bargeloads (Pacific Mexico), Playa Naranjo (Pacific Costa Rica) and of sand can be randomly imported to restore green turtle in the Galapagos Islands are among the smallest and they nesting beaches. The sand should be of the same type as deposit the fewest eggs per clutch. The nesters in that on the natural nesting beaches in the area.- Suriname, Brazil, Europa Island and Maziwi Island are Can- and Ogren (1 960) divided the nesting behavior into among the largest green turtles and they lay large clutches. eleven stages (several of these stages are combined by The diameter of green turtle eggs, from 18 localities, other investigators): 1. stranding, testing of stranding site, ranged from 33.8 to 58.7 nun, with an overall average of and emergence from wave wash 2. selecting of course about 45 rnm (N=28 samples) (Table 5). The weight of and crawling from surf to nest site 3. selecting of nest eggs, from 16 localities, ranged from about 21 to 66 g, site 4. clearing of nest premises 5. excavating of body with an overall mean of about 47 g (N=20 samples) pit 6. excavating of nest hole 7. oviposition 8. filling. (Table 6). covering, and packing of nest hole 9. filling of body pit The number of clutches laid in one nesting season can and concealing of site of nesting 10. selecting of course, range from 1 to 9 in some nesting colonies but the overall and locomotion back to the sea 11. re-entering of wave average is about 3.3 (N=21 samples) (Table 7). Green wash and traversal of the surf. turtles renest at about 13 day intervals (Table 8) and they Data in Table 1 indicate that there is year-around nest- occupy an internesting habitat (see section 2.2.2) between ing (but with seasonal peaks) at some sites while at other nesting episodes. Remigration intervals at different nest- locales shorter seasonal nesting prevails. ing colonies are provided in Table 9. Green turtles spend about two and one-half hours on Several researchers have reported a positive relation- the beach for nesting (Hirth 1980~);about two hours in ship between size of nesting female and clutch within a the actual construction and camouflaging of the nest (ttd nesting population: Hornell (1927) on Aldabra Atoll; and Samson 1987). At Tortuguero, green turtles spend Bustard (1 972) on Heron Island; Simon and Parkes (1976) about 23 minutes in digging the body pit, about 23 minutes and Hays et al. (1993) on Ascension Island; Balazs (1980) in digging the egg chamber, approximately 15 minutes in on French Frigate Shoals; Hirth (1988) and Bjorndal and laying eggs. about 12 minutes in filling the egg chamber, Can- (1989) at Tortuguero; Witherington and Ehrhart and, approximately 43 minutes in filling the body pit and (1989a) in Florida; and, Chen and Cheng in Taiwan (1 995). camouflaging the nest site. Inexperienced and experienced On the other hand, Hughes (1974a) found no correlation nesters at Tortuguero exhibit similar panerns of nesting and between size of female and clutch size on Europa Island, data indicate strong natural selection for a fixed nesting nor did Miller (1989) find a correlation on %an Island. behavior in green turtles (Hirth and Samson 1987). Alvarado and Figueroa (1990) found no correlation be- Table 4. Clutch sizes of grew) turtles. Location Mean Padfie Ocean XiIs., China Huang Chu-Chien (1982) Ko Khram, Thailand 70-130 Penyçpol1958) Sarawak 3-184 8,147 Hendrickson (1958) Sabah 3-190 de Silva (1970) Pangumbahan, Indonesia Suwdo and Kuntjoro (1969) Sukumade, Indonesia Nuitjs ( 1993-4) Pulau Berhala, Indonesia 48- 175 22 Mohr (1927) Baguan Is., Philippines 146 Truno(1991) Wan-& Is., Taiwan 64-172 63 Chen and Cheng (1995) Ogasawara Is. 70-150 FBkada (1965) Long Is., Papua New Guinea 71-165 126 Spring (1983) Solomon Is. 45-156 5 McKeown (1977) Solomon Is. 37-143 8 Vaughan(1981) Bramble Cay, Australia 8 Kowarsky (1978) Raine Is., Australia 6 Kowarsky (1978) Heron Is., Australia 50-200 Bustanl(1972) Heron Is., Australia 61-153 35 Unip~(1980) Scilly Atoll, French Polynesia 37-152 12 Lebeau (1985) French Frigate Shoals. Hawaii 38-145 50 Balm (1980) Michoacan, Mexico Matquez et al(1982) Colola and Marunra. Mexico 11-146 397 Alvarado et al(1985) Colola and Maruata, Mexico 5-136 636 Alvarado and Figueroa(1986) Colola and Maruaia, Mexico 1-130 916 Alvarado and Rgueroa (1990) Playa Naranjo, Costa Rica 65-107 10 Cornelius (1 976) GalSpagos Is., James Bay 48-131 15 Pritchard (1971) Galipagos Is., Indefatigable 19-116 27 Pritchari (1971) Galipugos Is. 26-144 Green (1994) Atlantic Ocean and Mediterranean Sea Canaveral National Seashore, Florida Bryant (1986) Cape Canaveral and Witherington and Melbourne, Florida Ehrhan (1989a) Hutchinson Is., Florida Gallagher et al. (1972) Browad County, Florida Broward County Erosion Prevention Dist. (1987) Contoy Is., Mexico Nijm (1991) El Cuyo, Mexico Rodriguez and Zarnbrano (1991) Tortuguero, Costa Rica OUT and Kinh (1962) Tomguero, Costa Rica Fowler (1979) Tortuguero, Costa Rica Bjomdal and Carr (1989) Shell Beach. Guyana Pritchard (1969) Eilanti, Suriname Pritchard (1969) Bigi Sand, Suriname Pritchard (1969) Suriname Schuk (1975) Praia do Pone, Brazil R' Amato and Marczwski (1993) Praia do Forte. Brazil Mareovaldi and Lourent (19%) Ascension Is. Carr and Hi& (1962) Ascension Is. (1 973) Simon and PaAes (1974) (1976) Ascension Is. Hays et al(1993) Ascension Is. Mortimer and Can- (1987) Alagadi, Cyprns Godley and Broderick (1993) Table 4. Continued. Location Mean Range N Reference

KazanIi, 7\iAey 1126 43 Barno et a1 (1991) Kazanli, nrkey 122 7 Coley and Smart (1992)

Indian Ocean Maziwi Is. Prazier (1984b) Primeim Is. Hughes (197411) Europa Is. Hughes (1974a) Europa Is. Hughes (1974b) Europe 1s. Servan (1976) May* Is. Brazier (1985) ' Moheli Is. Razier (1985) ~mraAtoll Hornell (1927) Aldabra AmU Frazier(1971) Tromelin Is. Hnghes (19741) Tromelin I& Hugheg (1974b) Koran Is., Saudi Arabia Miller (1989) Sharma. Yemen Hirth and Can- (1970) Shanna, Yemen PA0 (1973) Masirah Is. Oman Ross and Barwani (1982) Ras a1 Hadd, Oman Ross and Barwani (1982) Hawks Bay, Pakistan Minton (1966)

Table 5. Diameters (mm) of green turtle eggs. Location Mean Range Clutches Eggs Reference Pacific Ocean Xisha Is., China Huang Chu-Chien (1982) Sarawak Hendriclaon (1958) Pangmbahan, Indonesia Suwelo and Kuntjom (1969) Wan-An Is., Taiwan Cben and Cheng (1995) Ogasawara Is. Fukada (1965) Heron Is., Australia Busrard and Greenham (1969) Heron Is., Australia Busrard (1972) Heron Is., Australia Limpus (1980) French Frigate Shoals, Hawaii Balazs (1980)

Atlantic Ocean Kennedy Space Center, Florida Ehriiait (I 979a) Hutchinson Is.. Florida Galltigher ex al. (1972) Tortuguero. Costa Rica Can- and Hirth (1962) Tortuguera, Costa Rica Hirth (1988) Tortuguero, Costa Rica Bjorndal and Can- (1989) Aves 1s. Rainey (1971) Shell Beach, Guyana Prit~harf(1969) Suriname Schulz ( 1975) Ascension Is. Can- and Hirth (1962) Ascension Ls. Hays et al. (1993) Indian Ocean Primeiras Is. Hughes (1974a) Europa Is. Hughes (1974a) Empa Is. Servan (1976) Mayotte Is. Frazier ( 1985) Moheli Is. Frarier (1985) Aldabra Atoll Brazier(1971) Tromelin Is. Hughes (1974a) Table 5. Continued. Location Mean Range Clutches Eggs Reference Kami Is., Saudi Arabia 43.2 33.8-49 58 580 Miller (1989) Abdul Wadi, Yemen 425 40-45 1 100 Hirth and Can- (1970) Shanna, Yemen 45.5 41-48 5 FAO(1973) . Hawkes Bay. Pakistan 50-55 Minton (1966)

Table 6. Weights (g) of green turtle eggs. Localion Mean Ranee Clutches Eees Reference Pacific Ocean Sarawak 36 28.6-44.7 3 30 HendricloÈ (1958) Sabah ca 40.3 de Silva (1970) Wan-An Is., Taiwan 53.2 44-65 35 Chen and dieng (1995) Long Is., Papua New Guinea 41.2 334.9 95 Spring (1983) Heron Is.. Australia 51.9. 44.7-60.4 20 Budand Greenham(1%9) Heron Is., Australia ~51.6 44-60.4 Busmd (1972) Heron Is., Australia 50 33.5-58.3 11 110 Limpus (1 980) French Frigate Shoals, Hawaii 50 45-54 1 99 Balazs (1980) Atlantic Ocean Kennedy Space Center, Florida 60.1 55.8-63.8 1 Ehihart (1979a) Tortuguero, Costa Rica 48.8 35-66 20 400 Hi& (1988) I Aves Is. 45.1 40.5-49.4 1 48 Mney (1971) Indian Ocean Maziwi Is. 35-48.7 Fcazier (1984b) Primeiras Is. 44.9 413-49.7 2 40 Hughes (1 974a) Europa Is. 47.9 38.1-58.6 28 280 Hughes (1974a) Empa ls. 45.8 40-50 Servan (1976) Mayotte Is. 46.3 21-60 4 40 Frazier (1985) TromeIin Is. 48 41 3-53 10 200 Hughes (1974a) Karan Is.. Saudi Arabia 44.1 33.2-56.3 48 480 Miller (1989) Abdul Wadi, Yemen 40.4 30-44 1 100 Hirfl and Can (1970) Shanna, Yemen 42.3 37.5-47.5 1 50 Hirth and Can- (1970) Sharma,Yemen 44.8 35-55 5 FA0 (1973)

Table 7. Number of clutches of areen turtles per nesting season. Location Mean Range N Reference Pacific Ocean Xisha Is., China fa3 Huang Chu-Chien (1982) Sarawak 4.1 1-9 447 Hendrickson (1958) Sukumade, Indonesia 2-4 Nuitja (1993-4) Ogasawara Is. 1-5 Fukada (1965) Ogasawara Is. 3.9 Horikoshi, pers. corn. in Mortimer and Can- (1987) Heron Is. Australia 4.5 3-6 Bustard (1972) I Heron Is. Australia ea 5.5 Limpus (1980) French Frigate Shoals, Hawaii 1.8 1-6 208 Balm (1980) Michoacan, Mexico 4 Marquez et al(1982) Coiola and Mmata, Mexico 2.8 1-9 379 Alvarado et a1 (1985) Colola and Maruata, Mexico 3.5 1-9 100 Alvarado and Figueroa ( 1986) Colola and Maruata, Mexico 2.5 1-7 Alvarado and Rgueroa (1990) Playa Naranjo, Costa Rica a42 2-4 32 Cornelius (1976) Galipagos Is. a3 1-7 Green (1994) Huang Chu-Chien (1982) Hendrickson (1958) Nuitja (1993-4) Trono (1991) Chen and Cheng (1995) Spring (1983) Busiari (1 972) Limpus (1980) Balazs (I 980) Marquez et al(1982) Alvarado and Figueroa (1986, 1990) Cornelius (1976) Green (1994)

Witherington and Ehriian (1989a) Gallagher et al. 1972

Can- and Giovannoli (1957) Can- and Own (1960) Can- and Hirth (1962) Can- et al (1978) Pritchanl(1969) Pri tchari ( 1969) Schulz (1975) OUT and Hirth (1 962) Simon and Parkes (1976) Mortimer and Can- (1987) Godley and Eroderick (1993)

Swan (1976) Bonnet et @ ( 1985a) Honiell(1927) Table 8. Continued, ------Location Mean Range N Reference Aldabra Atoll Frazier (1971) Koran Is., Saudi Arabia 9-15 Miller (1989) Sharmn, Yemen 7-13 5 Hi& and Cair (1970)

Table 9. Remigration intervals (years) of green turtles. Predominant

Location Mean Range N Reference Pacific Ocean Sarawak 3 Hendricbon (1958) sabah 3 2-4 102+ de Silva (1982) Baguan Is., Philippines 2.5 24 Trono (199D Ogasawara Is. 2-4 7 Kiuata, undated, in Groombridge and Luxmoore (1989) Ogasawara Is. 4 2-7 Suganuma (1989) Raine Is., Australia 5 Limpus et al. (1993) Heron Is., Australia 4 Busiard (1972) Heron Is., Australia 4.65 2-7 31 Limpus (1 993b) French Frigate Shoals. Hawaii 2 2-6 21 Bates (1980) French Frigate Shoals, Hawaii 2-3 1-8 130 Bate (1983a) Michoaean, Mexico 1.8 Colola and Maruata, Mexico 3 1-5 26 1 Galiipagos Is. 3 2-6 85 Atlantic Ocean Cape Canaveral and 2 Witherington and Melbourne, Flonda Ehrhart (1989a) Melbourne, Florida 2 Bjomdal et al. (1983) Tortuguero, Costa Rica 3 Can- and Ogren (1960) Tmumero, Costa Rica 3 2-3 46 Can- and Hirth ( 1962) Tortuguero, Costa Rica 3 2-9 447 Can- and Can- (1970a) Tonuguero, Costa Rica 3 1-4+ 1,412 Canr et a1 (1978) Suriname 2.3 1-4 599 Schulz (1975) Ascension Is. 3-4 2-9 69 Mortimer and Can- (1987) Indian Ocean Europa and Tromelin Is. 3 Bonnet et at (1985a) North West Cape and Muiron Is., Western Australia 2-5 4 Prince (1993) Barrow Is., Western Australia 2-5 17 Prince (1993) Lacepede Is., Western Ausnalia 2-5 228 Prince (1993)

tween body size and clutch size (in 1986, 1987, 1989) Tortuguero. Fowler (1979) and Bjomdal and Can- (1989) nor between body size and overall seasonal fecundity (in found no significant trend in clutch size, at the individual 1987,1988) in Michoacan turtles. However, they did find level, over a season. Alvarado and Figueroa (1990) de- a positive correlation between carapace length and over- tected no significant increase or decrease in clutch size all seasonal fecundity in 1985 (Alvarado and Figueroa over the course of a season at Michoacan. 1986). At Tortuguero, recruits lay an average of 2.7 clutches Carr and Hirth (1962), Mortimer and Can" (1987) and with a mean clutch size of 11 1.4 eggs while remigrants Hays et al. (1 993) reported a tendency for earlier clutches lay an average of 3.4 clutches and their mean clutch size to be larger than later clutches on Ascension Island. At is 116.8 eggs (Can" et al. 1978; Bjorn dal 1980b).

35 Mean carapace length (cm)

Fig. 12. Mean clutch sizes and average carapace lengths of nesting green turtles from around the world (data from Tables 2 and 4). Statistical significance was not tested because the turtles were measured in several different ways and because sample sizes were very different-but the trend is obvious.

Hays et al. (1 993) discovered that the sizes (diameters) Predators of green turtle eggs include crabs, insects, of eggs within a clutch were variable on Ascension 1s- lizards, coatis, raccoons, foxes, jackals, dogs, pigs and land. Larger eggs are laid at the start of a clutch and birds (in already partially depredated nests) (section 3.3.4). smaller eggs at the end. This decline in egg size between Hatching success at a number of beaches bordering the the beginning and end of a clutch averaged 1.21 mm. Pacific, Atlantic and Indian Oceans are provided in Table Van Buskirk and Crowder (1994) found a significant 11. The mean and range listed for Sarawak (Groombridge trade-off between clutch size and egg size among the seven and Luxmoore 1989) in the Table are the overall average species of marine turtles. and the overall range of averages for the years 1970 Natural incubation periods as well as length of incuba- through 1985. Likewise, the mean and range given for tion in some egg hatcheries are listed in Table 10. The Sabah (de Silva 1982) are the overall average and overall overall average is about 57 days. range of averages for the years 1966 through 1978. Leh Average incubation times are about twenty days longer (1 994) reported that mean hatch rates in the egg hatcher- in the wet season than in the dry season in Sarawak ies on the Sarawak "nirtle Islands, between 1970-1990, (Hendrickson 1958). Servan (1976) stated that on Europa varied between 53 and 96% with higher percentages of Island, incubation time ranges from 50 days in the sum- hatching in the drier months and lower hatching success mer to 99 days in the winter. in the wet monsoon months (November to April). Table 10. Natural incubation periods (days) of green turtle eggs (from1 oviposition to emergence of hatchlings on the surface) *hatchery. Location Mean Pacific Ocean Xisha Is., China KO Khrain. Thailand Sarawak 54 (dry) * 70 (wet) * Sabah de Silva (1 970) Pangumbahiin, Indonesia 50.6 Suwelo and Kungom (1969) Baguan Is., Philippines 54.3 Trono (1991) Wan-An Is., Taiwan 49.3 Chen and Cheng (1995) ogasawara Is. ea 63 Fukada (1 965) Solomon Is. 60 McKeown (1977) Heron Is., Australia 56 Bustard (1972) French Frigate Shoals, Hawaii 64.5 Balazs (1980) Michoacan, Mexico Maquez et a1 (1982) Galapagos Is, 55 Green (1 994) Atlantic Ocean Cape Canaveral and 54 Withtringlon and Melbourne, Florida Birfiart ( 1989a) El Cuyo, Mexico 59 Rodriguez and Zambrano (1991) Tomguero, Costa Rica 57.5a Can- and Ogren (1960) Tortuguero, Costa Rica 55.6* Can- and Hirth (1 962) Tortuguero, Costs Rica 61.9 Fowler (1979) Bigi Santi, Suriname 58.3 Pritchaid (1969) Suriname 56.4 Schuiz (1975) Praia do Fane, Brazil 50 D'Amaro and Marczwski (1 993) At01 das Rocas, Brazil 61 Bellini et al. (19%) Ascension Is. 59.5 Can- an4 Hi& (1962) Indlan Ocean Maziwi Is. Frazier (1984b) Europa Is. Swan (1976) Aldabra Atoll ca 47 Hornell (1927) Aldabra Atoll ca 69 Frazier (1971) Kam I;;., Saudi Arabia 62.5 Miller (1989) Abdul Wadi. Yemen 49 Hirth and Can- (1 970)

a Natural and hatchery

The data in Table 11 indicate that hatching success in For five years the egg hatchery at Colola has had a higher the Galapagos can vary from about 2 to 78%. The low percentage hatch than the one at Maruata and this is be- emergence rate on Quinta Playa is due mostly to egg pre- lieved due to the higher moisture content of the sand at dation by a beetle (Trox suberosus) and nest destruction Colola (Alvarado and Figueroa 1990). In 64 green turtle by feral pigs. The extremely poor hatching success on nests in an egg hatchery in Sri Lanka, Hewavisenthi Espumilla is due to feral pigs (Green and Ortiz-Crespo (1994a) found a negative correlation between clutch size 1982). There is about a 20% higher hatching rate in the and percentage of live hatchlings. Most mortality was in natural nests on Colola and Maruata (Pacific Mexico) than the late embryonic and early hatching stages. The re- in hatchery nests, 83% versus 64%. The 64% rate is an duced oxygen supply, in nests with large clutches, during average for the hatcheries at Colola and Mamata in 1989. the late stages of development may play a role in these

37 Table 11. Percentage emergence of hatchlings from natural nests, 'hatchery ''unknown. Location Mean Range Nem Eggs Reference Pacific Ocean 73.3* Groombridge and Luxmoore (1989) salawak 47.1 * Hendridc.son (1 958) Sabah 67.3* ' de Silva (1982) Baguan Is., Philippines 85.7 ¥non(1991) Ogasewara Is. 54.6* (1 982) Sugamima (1985) 83.1 * (1983) 77.5* (1984) Long Is., Papua New Guinea 89 Spring (1983) Solomon Is. 68.4? McKeown (1977) Solomon Is. 78.9? Vaughan (1981) Raine Is. Australia 75-80 Lmipus et a1 (1 993) Heron Is., Australia 88 (1966-7) Bustard (1972) 85 (1967-8) 67 (1965-6)* 65 (1966-7)* 52 (1967-8)* French Frigate Shoals, Hawaii 70.8 Balm (1980) Colola and Maruata, Mexico 83 Alvarado and Figueroa (1990) Cdola and Mmata, Mexico 64* Alvarado andFigueroa (1990) Las Bachas, Gaiipagos Is. 78.4 Green and Odz-Crespo (1982) Bahia Barahow Galapagos Is. 72.9 Las Salinas. Galapagos Is. 64.8 47.2 Quinta Playa, GaMpgos Is. 41.7 Espumilla, GalApagos Is. 1.9 . Atlantic Ocean and Mediterranean Sea Melbourne, Florida 61.6 Wnherington and Ehrfuat (198%) Broward County, Florida 61.8 Broward County Erosion 64.6* Prevention District (1987) El Cuyo; Mexico 86.5 Rodriguez and Zainbrano (1 991) Tonuguero, Costa Rica 50.7* ( 1959) Cam and Hirth ( 1962) 50.8* (1960) Tortuguero, Costa Rica 83.1 Fowler (1979) Tonuguero, Costa Rica 46.3 (shaded) Horikoshi (1989) 57.3 (open) Suriname 84 Schulz (I 975) 58* Krofajapasi, Surinaine 80.4 Whimom and Dunon (1 985) Ascension Is. 54.4 Can- and Hirth (1962) Cyprus ea 75* Demetropoulot and Lamben (1986) Alagadi, Cyprus 85.3 Godley and Broderick (1 993) Indian Ocean Maziwi Is. 78? Frazier (1984b) Europa Is. 77.6? Hughes (1974b) Europa Is. 84 Servan (1976) 54* Tromelin Is. 69.8? Hughes (1974b) Koran Is., Saudi Arabia 81.7 Miller (1989) Abdul Wadi, Yemen 48 Hinh and Can ( 1970) findings (Hewavisenthi 1994a). perature directly affects the gonadal differentiation of In Suriname, Schulz (1975) reported a successful hatch embryos. Wanner temperatures produce females and of about 86% in styrofoam boxes but he also noted that cooler temperatures, males. TSD in sea turtles and in incubation time in boxes was longer than that in natural has been the subject of two reviews (respectively, nests and this was probably due to lower temperatures. Standora and Spotila 1985, and Janzen and Paukstis 1991). We now know that this practice probably produced more The proceedings of a recent symposium (Lance 1994)on male hatchlings. Whitmore and Button (1985) noted an environmental sex determination in reptiles includes average of 72.9% hatch of Suriname eggs in boxes. twelve papers on the subject (some specific papers are Can- (1984) discussed how the number of eggs laid by cited in this section of the synopsis). Spotila et al. (1983) a green turtle at each oviposition has adaptive value. Some have written a manual describing methodology for study- of these advantages include escape from some predation ing TSD. For measuring incubation temperatures on sea by swamping the predators on the beach and in the lit- turtle beaches, Godfrey and Mrosovsky (1994) recom- toral zone; metabolic heating of the nest by a mass of mend using a module that memorizes maximum and mini- eggs; social facilitation in the climb to the sand surface; mum temperatures. They report that for beach tempera- and, group orientation in the crawl to the sea. tures at marine turtle nest depth, the average of the maxi- mum and minimum temperatures over a 24 hr interval is 3.2 Embryonic and Hatchling Phase very close to the mean based on more frequent readings. The relatively inexpensive unit also provides flexibility 3.2.1 Embryonic phase in choice of recording site. Miller (1985) reviewed the earlier literature on embry- The adaptive significance of TSD in reptiles is unclear ology and provided a composite account, with photo- but is discussed by Bull and Charnov (1989), Ewert and graphs, of the embryology of six species of sea turtles, Nelson (1991), Janzen and Paukstis (1991) Burke (1993) including Chelonia mydas. He divided the embryology and Ewert et al. (1 994). On the other hand, what was adap into 3 1 stages of development (5 preovipositional and 26 tive when TSD evolved may no longer be pertinent. postovipositional stages). Stages 1 - 5 are preovipositional Davenport (1989) and Mrosovsky and Provancha (1992) cleavage stages. Stages immediately following oviposi- have discussed some possible effects of global wanning (the tion (6 - 10) were defined by changes in the shape of the greenhouse effect) on sex determination in sea turtles. Based blastopore and by differentiation of the notochord, neural on his empirical research with hatchling painted turtles, folds and head folds. The number of sornites and the dif- Chrysemys picta, in their shallow nests in Illinois, Janzen ferentiation of the heart and pharyngeal clefts primarily (1994) found that annual offspring sex ratio was highly defined stages 11 - 18. Stages 19 to hatching were deter- correlated with mean July air temperature (July corresponds mined by modifications in the limbs, formation of the shell to the developmental period when embryonic sex is deter- and development of scales and pigmentation. The fre- mined). He calculated that an increase in the mean July quency of occurrence of abnormal embryos among rna- temperature of 4'C would effectively eliminate production rine turtles is low (reviewed in Miller 1985). Lewis et al. of males in this population. He concluded by stating "popu- (1992) briefly review some reports of twinning and they lations of species with temperature-dependent sex detenni- describe in some detail the histopathologic and anatomic nation may act as bellwethers for the impending disruption relations of one case of omphalopagus twins. Improper to biological systems posed by global temperature change." handling of eggs at any time during their development The possible physiological and molecular bases for TSD in reduces hatching success. reptiles have recently been discussed by Wibbels et al. (1994) Ackerrnan (1 98 1b) found that green turtle embryos grow and Spotila et al. (1994), respectively. slowly during the first half of their incubation and rapidly Mrosovsky and Pieau (199 1) recommend standardiza- during the second half. The rapid phase slows prior to tion of terms used in describing sexual determination es- hatching resulting in a sigmoid-shaped overall growth pecially in view of the fact that the subject is now of great process. interest to a wide range of people, including geneticists, evolutionary ecologists and conservationists. They rec- 3.2.2 Hatchling phase ommend the following terms and definitions: transitional Sexual differentiation in the green turtle, as in other sea range of temperature-that range between male and fe- turtles, is determined by the substrate temperature (other male producing temperatures, within which both sexes environmental factors may be involved) during incuba- may differentiate among individuals of a population; tion. This is commonly referred to as TSD (temperature- masculinizing and feminizing limit temperatures-those dependent sex determination) or ESD (environmental- temperatures delimiting the transitional range, below, or dependent sex determination). The middle trimester of above, which masculinization, or feminization, are maxi- incubation appears to be the critical period in which tem- mum; pivotal temperature-that temperature within the transitional range giving 50% of each sex in experiments Tortuguero nests at the pivotal temperature, eggs near the in which eggs are incubated at constant temperatures. center of the nest produced females and those at the pe- Pivotal temperatures can be determined for a number of riphery, males, and this may have been due to metabolic clutches or for a particular clutch or for a sample from a heating (Standora et al. 1982a). Horikoshi (1992). work- clutch. Mrosovsky and Provancha (1992) discuss sane ing on Tortuguero Beach, estimated pivotal temperature of the precautions to take in the collection and in the analy- at between 28.5 and 29.0° and he calculated an overall sis of data pertaining to reptilian sex ratios, especially sex ratio, in one season, of about 40% female. Bjorndal those of sea turtles. Vogt (1994) describes how some con- and Bolten (1992) predicted that the primary sex ratio in servation managers' attempts to produce equal numbers the Tortuguero colony will vary from year to year because of females and males (by incubating eggs at the pivotal the nesting sites of individuals are not consistent. temperature) may not be in the best interests of turtles. In Suriname, Mrosovsky et al. (1984) reported that more He argues that, in some cases, it may be more useful to males were produced in the cooler, wetter months and produce more females than males in order to enhance the more females during the wanner, drier months of the nest- ' reproductive output of the population, and that incuba- ing season. The pivotal temperature was estimated at tion of eggs near the pivotal temperature has a higher prob- 28.8OC and the overall sex ratio was estimated at 53.9% ability of producing intersexes. However, Mrosovsky and female. Only 1.1% from the field sample were intersexes. Godfrey (1995) caution that careful planning is neces- Godfrey et al. (1996) estimated that 63.8% of the green sary before wildlife managers manipulate sex ratios. turtle hatchlings produced on Matapica Beach, Suriname, Lovich (1996) continues the cautious approach of in 1h3,were females. They further estimated that over Mrosovsky and Godfrey (1 995) and discusses how knowl- fourteen years the overall sex ratio on Matapica Beach edge of natural sex ratio variation, multiple paternity and averaged 68.4% females. sperm competition, fertility factors and intersexual and Using thermocouples along a beach transect, Alvarado intrasexual competition is needed before "jump-starting" and Figueroa (1990) estimated more females were pro- declining turtle populations by manipulating sex ratios. duced at Colola and more males at Maniata but the over- Miller and Limpus (1981) found that in a clutch of eggs all sex ratio in the Michoacan area was about 50:50. from Heron Island, incubation at 26OC resulted in 85.7% Eleven natural nests (1,089 eggs) were reburied in an males, 0% females and 14.3% intersexes; incubation at egg hatchery in Sarawak and produced 81.3 - 91.3% fe- 29OC resulted in 0% males, 90.2% females and 9.8% in- males when temperatures during the middle third of the tersexes; and incubation at 33OC produced 0% males, incubation period ranged between 29.5 - 30.3OC (Leh et 85.7% females and 14.3% intersexes. Limpus et at (1983) al. 1985). reported that the sex ratios of green turtles hatched on the The natural sex ratio on Baguan Island in the Philip- wanner side of Heron Island differed from that on the pines was calculated at about 90% female and the sex cooler side. Hays et al. (199%) found that inter-beach ratio in a partly shaded egg hatchery was computed at thermal variation on Ascension Island was largewith 38% female, 36% male, and 26% intersexes (Trono 199 1). the darkest beach (albedo, 0.16) being 4.2' C wanner than Mrosovsky (1994) reviewed the available literature and the lightest colored beach (albedo, 0.73)Ñan they show concluded that pivotal temperatures for sea turtles are clus- how these temperature data could be used to calculate tered close to 29OC. He also described how SSPPs (sea- hatchling sex ratios here. Limpus et al. (1993-94) briefly sonal sex production profiles) can show how similar over- summarized results of constant temperature incubation all sex ratios can be achieved in dissimilar ways. studies of green, loggerhead and flatback turtles in east- Many factors probably affect TSD of green turtles in- em Australia with the statement "The timing of the breed- cluding nest site, depth of body pit, depth of clutch, posi- ing season and the location of the rookeries used by a tion of egg in the clutch, weather during the incubation stock appear to be selected to provide a range of nest tem- period, sand color and beach topography. peratures above and below the pivotal temperature rather Needless to say, serious thought must be given to re- than temperatures coinciding with the pivotal tempera- sulting sex ratios when any manipulative project is un- ture." dertaken, such as egg transplanting and artificial ineuba- On the Tortuguero nesting beach Spotila et al. (1987) tion. For example, Mrosovsky (1982) reported on the found that temperatures >30,3OC during the middle third masculinization of hatchlings incubated in styrofoam of incubation produced females and temperatures <28.5OC boxes. Morreale et al. (1982) cautioned that the incuba- produced males. Thus, the pivotal temperature for tion of eggs in central beach hatcheries or in hatcheries Tortuguero green turtles is between 28.5 - 30.3OC. They aboveground should only be attemped after the appropri- estimated that the sex ratio produced on the Tortuguero ate TSD is defined. The taking of eggs by humans for nesting beach, which has sunny and shaded nesting sites, sustenance on easily accessible parts of the beach or at in one season, was 67% female and 33% male. In certain times of the nesting season could affect overall sex ratios in the nesting colony. Predators can affect natu- ditions (Ehrenfeld 1968; Mrosovslq and Shettleworth ral sex ratios by taking eggs laid in more shallow nests 1968; Mrosovsky 1970,1972, 1978b. Mrosovsky et al. under shrubs. Even with our knowledge of TSD, the least 1979; Mrosovsky and Kingsmill 1985). Mrosovsky manipulative management strategies are preferred by the (1978a) also demonstrated how hatchlings integrate author of this synopsis (see section 6.2). brightness information over time in their sea-finding be- The hatchling cuts through its eggshell with the aid of a havior. That is, a flashing light did not influence a turtle's horny protuberance, sometimes called the egg caruncle, behavior as much as a continuous light, so a hatchling or egg tooth, on the tip of its snout. This caruncle disap- should not be influenced by lightning flashes over a beach. pears after a few days. The histological structure of the However, Van Rhijn's (1979) experiments with hatchlings egg tooth is interesting for the arrangement of its collag- pointed to a redundant system of orienting mechanisms enous fibers of dermis and for its thick stratum corneum that has both optic and non-optic components. Based on (Bons and Bonaric 1971). laboratory experiments, Van Rhijn and Van Gorkom Over the course of several days, and in a synchronous (1983) concluded that hatchlings primarily orient visu- manner, the hatchlings climb to the surface. Emergence ally but a photic system may take over under conditions on the surface is synchronized with a certain sand thresh- that still have to be explained. Mrosovslq and Kingsmill old temperature or, more likely, with a sand temperature (1985) rejected, as unlikely, the possibility of a system gradient. Emergence is predominantly at night, usually redundant to a complex phototropotactic system. as a single unit, but sometimes afew individuals may pre- The sea-finding ability of hatchlings, as well as adult cede or follow nest exit of the main group. Nocturnal females, was further examined on the Tortuguero beach emergence is adaptive in that it eliminates exposure to by Ehrenfeld and Can- (1 967) and Ehrenfeld (1%8).. Spec- diurnal predators and it eliminates exposure to hot sand tacles containing special filters were fitted to the heads of surface temperatures which could be lethal or which could adults, and hatchlings were tested in a circular arena where slow down the hatchling in its crawl to the sea and thus the view of the horizon was unobstructed or where it was lengthen its exposure time to predators. Details of the blocked by a low wall. They substantiated the claim of aforementioned hatchling behavior can be found in: others that the water-finding orientation is primarily a vi- Hendrickson (1958), Can- and Hirth (1961). Bustard sual process (based on brightness rather than color), that (1967), Mrosovsky (1 968, 1980). Hirth (1971 b), and- it involves an appraisal of beach topography, and that de- Gyuris (I 993). polarizing filters did not affect orientation. They also Hatchlings crawl along the sand surface using their four concluded that there was no innate compass direction flippers in typical reptilian fashion (front member moved based on celestial cues, although Fischer (1964) did find in conjunction with hind member on opposite side), while a fixed direction preference related to the position of the adults on shore heave themselves along using sometimes sun in hatchlings tested in a water arena. only the front flippers simultaneously but usually all four Mrosovsky and Can- (1967) examined light preferences flippers simultaneously. Hatchling green turtles may ex- of hatchlings in a simple apparatus on the natural nesting hibit optokinetic responses as early as 30 minutes after beach and discovered that when given a two-choice situ- leaving the nest (Ireland 1979). Experiments with ation, they prefer blue and green stimuli over red. Light hatchlings kept in tanks for several days and then released preferences were analyzed further by Mrosovsky (1967) crawled slower down the beach than fresh hatchlings and and by Mrosovsky and Shettleworth (1968) and it was they had difficulty entering the sea against incoming waves found that while wavelength preferences of hatchlings are (Hewavisenthi 1994b). controlled primarily by brightness, the existence of some The hatchling's orientation to the sea is based prima- color preference cannot be ruled out. In the laboratory, rily on visual cues. In the 1960's. Archie Can- and his using a V-maze, Witherington and Bjopdal(1991) found colleagues set the agenda for much of the hatchling be- that hatchlings oriented toward near-ultraviolet (360 nm), havioral work that would ensue for the next three decades. violet (400 nm) and blue-green (500 nm) and chose a stan- Can- and Ogren (1960) conducted a series of tests on the dard light source over an adjustable light source. Tortuguero, Costa Rica, beach and concluded that the Salmon et al. (1992) described the behavior of hatchlings' fundamental goal sense involved visual stimuli hatchlings in an arena where manipulation of visual and and that the response was a modified phototaxis. The slope stimuli was possible. Hatchlings oriented toward 'openness of outlook" of a sea-sky horizon was impor- the more intensely illuminated sections of the arena and tant and would even draw the hatchlings away from a they also oriented away from dark silhouettes which simu- moon or sun where either was over land. Later, several lated an elevated horizon, typical of the view toward land. investigators described the crawl to the sea in terms of a When hatchlings were presented simultaneously with sil- positive phototropotaxis and they explained how this be- houette and photic cues, at eye level, in different diree- havior would be successful under almost all natural con- tions, they oriented away from the silhouette cues. Re- spouses of green turtles to slope cues, under near normal low water (cf. Prick, 1976). Lohmann et al. (1 990) coi nocturnal light conditions, were weak. The authors con- eluded that hatchlings sequentially employ two differei clude that green turtle hatchlings usually find the sea by orientation systems: visual cues are used to crawl fror orienting away from elevated silhouettes and that, eco- the nest to the surf and, then, in the ocean,hatchlings on logically, this is a reliable cue for nesters on continental ent by swimming into waves. beaches. On some relatively flat, island beaches photic Lohmann (1991, 1992) and hhmann and Lohmani cues may play a more important role. Working on a (1993) found that loggerhead and leatherback hatchling; Suriname beach, where green turtle eggs laid below the have a magnetic sense and this sense may complement high tide line are reburied on safer ground, Godfrey and or supplant, wave orientation in deep water. Labmtoq Bmeto (1 995) demonstrated that when reburied in dense experiments with loggerhead hatchlings demonstrated thai vegetation the subsequent hatchlings showed no signifi- visual cues available on land set the preferred direction cant orientation to the sea. Thus, they advise caution in of magnetic orientation (Lohmann and Lohmann 1994a). the selection of relocation sites. Light ep al. (1992, 1993) discovered that loggerhead Lab-reared yearlings retain the ability to find the sea hatchlings respond to inclination, but not polarity, of the and even cross complex terrain to reach it, and adult males earth's geomagnetic field, and, since inclination changes evidently possess the same ability, although once reach- with latitude, sea turtles may use inclination as one com- ing the surf as hatchlings, they never normally return to a ponent of the "map sense", hi laboratory experiments, terrestrial environment, except at those few locations Lopann and Lohmann (1994b) demonstrated that log- where terrestrial basking occurs (Can- and Hirth 1962). gerhead hatchlings can distinguish between different After conducting a variety of experiments with green magnetic inclination angles and perhaps derive from them turtles in the Gulf of California, Caldwell and Caldwell an approximation of latitude. They also hypothesized how (1962) concluded that the sea approach ability is present this ability to recognize specific inclination angles could in both sexes of all ages. help explain how adult sea turtles can identify their nasal When arriving at the wet. wave-washed sand, the beaches after years at sea. In later experiments, Lohmann hatchlings appear to crawl faster and immediately after and Lohmann (1996a) demonstrated that loggerhead entering the surf their swimming behavior is commonly hatchlings can also distinguish between different mag- described as a frenzy. The swim frenzy is strongly adap- netic field intensities found along their migratory route, tive in all sea turtle species in that it takes the hatchlings Possessing abilities to distinguish between different field quickly away from the predator-rich shallow water out to intensities and different magnetic inclination angles, log- deeper waters and into current systems where food and gerhead hatchlings (and presumably green turtle shelter exist. But, as already mentioned (section 2.2.1), hatchlings) have the abilities necessary to assess global the flatback turtle may lack a pelagic phase. position using a bicoordinate magnetic map. In a laboratory study, Wyneken and Salmon (1992) com- Perry et al. (1985) found magnetic remanence in the pared the swim frenzy of green, loggerhead and leather- head region (the greatest concentration in the anterior part back hatchlings from beaches in southeast Florida. All of the dura mater) of nine green turtles (four hatchlings, species swam almost continuously during their first 24 three juveniles and two adults). hours and then there was a decrease in swimming as they The sizes and weights of hatchlings are provided in became less active at night The species differed in levels Tables 12 and 13, respectively. of nocturnal activity, this being highest in leatherbacks The eggs and hatchlings of twenty recruits at Tortuguero and lowest in loggerheads. were analyzed and it was found that at their first oviposi- It has been shown that hatchling green, loggerhead and tion large recruits lay more eggs and produce more leatherback turtles can maintain a constant compass course hatchlings than do smaller recruits and that the hatchlings in shallow water by using waves as an orientation cue of the larger nesters were slightly smaller than the (Salmon and Lohmann 1989; Wyneken et al. 1990; hatchlings of the smaller nesters (Hirth 1988). Pinckney Lohmann et al. 1990; Lohman 1992; Lohmann and (1990) found an inverse relationship between clutch size Lohmann 1992). After entering the sea, hatchlings swim and hatchling length in loggerheads on Kiawah Island, into the waves and this could take them directly to the South Carolina. Chen and Cheng (1995) found a statisti- open ocean. In a laboratory setting, Lohrnann et al. (1995) cally significant relationship between straight carapace demonstrated that hatchling green and loggerhead turtles length of nester and straight carapace length of the can determine the propagation direction of ocean waves hatchling (r2sO. IS, N=6). Hatchling size may be influ- by monitoring the circular movements that occur as waves enced by maternal and genetic factors. Environmental pass over them. factors such as hydric, thermal and respiratory variables Lohmann et al. (1990) found that a crawl across the within the egg chamber may also affect the size of beach was not necessary for normal orientation in shal- hatchlings. In a laboratory setting, McGehee (1 990) found Table 12. Carapace lengths (mm) of hatchling green turtles.

Pacific Ocean Xisha Is., China Huang Chu-Chien (1982) Sabah 46-51 de Silva (1970) Wan-An Is.. Taiwan 41 '4-52.4 Chen and Cheng (1995) Ogasawara Is. Fukada (1965) Long Is.. Papua New Guinea Solomon Is. 45-52 McKeown (1977) Heron Is., Australia BUSW(I 972) Heron Is., Australia 40.2-51.9 Limpus (1980) French Rigate Shoals, Hawaii 48-59 Balazs (1980) Playa Naranjo, Costa Eta 50-52 Cornelius (1976) Galipagos Is. 41-49.5 Pntchard (1971) Atlantic Ocean and Mediterranean Sea El Cuyo, Mexico 48.5-53.6 Rodriguez and Zambrano

Tortugurn, Costa Rica 46-56 Can-and Hinh (1962) Tortuguero, Costa Rica 47-56 Hirth (1988) Aves Is. 52.8-56.5 Bigi Santi. Suriname 5 1-55 Pritchard (1969) Suriname 48-53 Ascension Is. 49.1-55 Can- and Hirth (1962) Ah@. Cyprus S.D. 0.4 Godley and Broderick (1993) Indian Ocean Maziwi Is. Frazier (1984b) Europa Is. 45.8-51.4 Hughes (1974a) Eumpa Is. 35-54 Swan (1976) Moheli Is. 47-52 Fmzier (1985) Aldabra Atoll £ 45-53 Frazier (1971) Tromelin Is. 45.2-51.9 Hughes (I 974a) KmIs., Saudi Arabia 45-52.1 Miller (1989) Abdul Wadi, Yemen 44-48.4 Hmh and Can- (1970) Hawkes Bay, Pakistan 48-53 Minion (1 966)

that loggerhead hatchlings' carapace lengths were strongly employ a foreflipper beat or a hindflipper action as the correlated with sand moisture content, and Broadwell dominant slow swimming stroke, green turtles use the dog- (1992) showed that larger loggerhead hatchlings were paddle (Davenport and Pearson 1994). Twelve hatchling produced on Florida beaches containing more moisture green turtles observed in captivity by Davenport and and greater pore spacing. Good reviews on this subject Pearson (1994) had difficulty diving at first but all were are Packard and Packard (1 988), Wilbur and Main (1988) able to dive and exhibited neutral or negative buoyancy and Steams (1992). Packard et al. (1992) review some by the time they weighed between 100 and 150 g. Labo- methods for measuring water potential in subterranean ratory studies show that young green turtles, weighing nests and they recommend using thermocouple psychrom- between 200 and 1,200 g, are excellent swimmers who ew- use their foreflippers like wings rather than oars (Daven- Virtually nothing is known about the survival rates of port et al. 1984). green turtle hatchlings once they enter the sea. Larger hatchlings may climb to the sand surface quicker, crawl 3.3 Juvenile, Subadult, and Adult Phase to the sea faster, be better swimmers, experience faster growth rates and have fewer predators than smaller 3.3.1 Longevity hatchlings, but all of this needs investigation. Little is known about the longevity of green turtles. At Green turtle hatchlings appear to be faster swimmers Tortuguero, Costa Rica, there are records of a female nest- and to use their foreflippers more than either ridleys or ing over a 19-year span, and of two others for over 17 loggerheads and unlike other sea turtle hatchlings who years (Can-et al. 1978). In 1985, the author of this syn-

43 Pacific Ocean Sarawak Hfurisson (1955) Sabab de Silva (1970) Wan-An Is.. "faiwan Chen and Cheng (1995) Solomon Is. McKeown (1977) Heron Is: Australia Bustud (1972) Heron Is., Australia Lunpus (1980) KtBoch Frigate Shoals, Hawaii Bates (1980) Atlantic Ocean Kennedy Space Cents, Florida Ehriian (1979a) Southeast Florida Ehrhan and WiAaington (1992) Tonu~pero,Costa Rica Hiith (1988) Aves ls. Rainey (1 97 1) Indian Ocean Maziwi Is. Raatr (1984b) Europa Is. Hughes (1974a) Euroja Is. Swan (1976) Moheli Is. Frazier(1985) AIdabni Atoll Frazier (1971) Tromelin Is. Hughes ( 1974a) Karan Is., Saudi Arabia Miller (1989) Abdul Wadi, Yemen Hirth and Can- (1970)

opsis found a turtle nesting on Tortuguero Beach that had Schwartz (1989) stated that sea turtles do not hibernate been tagged there in 1962, representing a reproductive and that muddying (digging) in is a response by the turtle life of at least 23 years. If we assume turtles in the Costa to keep from floating upwards. Gregory (1982) has writ- Rica population reach maturity in 12 to 26 years (Frazer ten a comprehensive review of reptilian hibernation and and Ladner 1986) then the turtles mentioned above may Penny (1987) has reviewed the major physiological have been between 29 and 49 years of age. The present mechanisms involved in the overwintering strategies of maximum reproductive lifespan of female green turtles frogs and turtles. at Heron Island, Australia, is about 22 years and one male wtherington and Ehrhart (1989b) reported on several green turtle here has been recorded over an 18 year repro- hypothermic stunning episodes that occurred in Mosquito ductive lifespan (Fitzsimmons et d.1995b). Frazer (1983) Lagoon, Florida, between 1977 and 1986. Average water estimated a maximum reproductive life span of 32 years depth here is only about 1.5 m. Of 342 green turtles col- for Georgia loggerheads. Some factors affecting longev- lected. there was an 11.5% mortality. Cloaca1 tempera- ity are described in sections 3.3.4 and 4.4.2. tures of 22 living turtles averaged 6.1°C Morning sur- face water temperatures during these cold-stunning events 3.3.2 Hardiness generally were below 8OC. Schroeder et al. (1990) re- In 1976, Felger et al. published a paper on the dormant, ported that 246 green turtles were recovered cold-stunned partially buried green turtles overwintering on the sea in the Mosquito Lagoon area in December 1989; 67 were bottom in the Gulf of California. Here, the Seri Indians dead or died within 12 hours. Minimum water tempera- harpooned the turtles buried at water depths of from 4 to ture was below 10° during the episode. A few cold- 8 m, and Mexican fishermen caught dormant turtles at stunned juvenile green turtles have been collected in New depths of from 10 to 15 m. The green turtle is dormant at York waters (Morreale et al. 1992). Schwartz (1978) stated water temperatures below approximately 15OC in the Gulf that the lethal temperature for Chelonia mydas is about 5 of California. Owens (1993-94) reported that at about - 6.5'C. Ogren and McVeay (1982) compared the appar- 15'C environmental temperature. C. mydm became qui- ent hibernation and hypothermic stunning of green and escent and at 10° they are quiescent, do not feed and loggerhead turtles. appear to be hibernating. In retrospect. Can" (1982) noted The body temperatures (thennoprobe inserted about 15 that some of the immature green turtles off the west coast cm into the cloaca) of fifty immature turtles with an aver- of Florida went into winter dormancy in the mud. age curved carapace length of 55.7 cm (range 42.1-85.1 cm) were taken in Moreton Bay, Australia. The body tem- method involves maintaining the turtle in a prone position, peratures did not deviate significantly from water tem- intubation with an endotracheal tube fitted with a low-pres- peratures throughout seasonal fluctuations in water tem- sure cuff, and ventilating with a manual resuscitator. peratures in the range of 15 to 22.7OC (Read et al. 1996). The current method being used to treat carapace inju- The authors raise the possibility that immature green ries at Sea World in Florida is a transparent wound dress- turtles in Moreton Bay are more tolerant of cold water ing known as tegaderm (Walsh et al. 1994). The healing than individuals in some other populations. sequence in these types of wounds starts with granulation On the other end of the thermal spectrum, it has been of healthy tissue, followed by re-epithelization and pig- found that when a hatchling's body temperature reaches mentation, and then calcification. about 36'C. it starts seeking shade (Bustard 1970) and Campbell (1996) describes rehabilitation of injured and that sea turtles may extend their normal ranges in response sick sea turtles. Some of the common broad-spectrum to warmer water temperatures (Radovich 1961). antibiotics used are chloramphenicol succinate, enmfloxin The normal resting and active body temperatures of and trimethoprin-sulfadiazine. -green turtles are discussed in section 3.4.4. As far as is known, the green turtle is the only sea turtle 3.3.3 Competitors that spends time on land for non-nesting purposes. This Heinsohn et al. (1977) briefly describe how, in some behavior has been observed in Australia (Garnett et al. Australian waters, food (seagrasses)competition between 1985a), Hawaiian Archipelago (Balazs 1980;Whittow and green turtles and dugongs is reduced by the former's reli- Balazs 1982). SocmIsland and Galapagos Islands prim ance also on algae while the latter's primary food sources 1981; Snell and Fritts 1983) and in Namibia (Tarr 1987). are seagrasses. In the Torres Strait where much seagrass Non-nesting emergences of both sexes have been seen in is eaten by dugongs, Garnett et al. (1985b) found that the daytime and at night and involve some small but mostly most common plants consumed by green turtles were five large individuals. In the Wellesley Group, Australia, bask- genera of algae and the seagrass Thalassia. ing solitary turtles or sometimes groups of up to 400 in a In the Caribbean grass meadows, sea urchins and cer- small embayment (made up mostly of internesting females tain fishes are the main competitors of green turtles. Three and some adult males) can periodically be seen (Lipus et species of sea urchins that graze extensively on seagrass al. 1994a). Hawaiian green turtles of all sizes regularly are Diadema antillarum, Tripneustes ventricosus and "bask" in captivity (Balazs and Ross 1974; Kam 1984). Lytechinus vanegatus. The sea urchins tend to graze the Garnett et al. (1985a) review some of the reasons for non- distal portions of the grass blades. The bucktooth nesting emergences, including avoidance of courting males parrotfish, Sparisoma radians, feeds primarily on by females, synthesis of vitamin D, acceleration of diges- Thalmsia testudinurn. Several other species of parrotfish tion, egg maturation, avoidance of predation by sharks and and surgeonfish feed on seagrasses, especially on the energy conservation. Congdon (1 989) reviews the basking epiphytized tips of the blades (Zieman et al, 1984). Along habit of turtles and in addition to the aforementioned with C. mydas, grazers in the tropical western Atlantic possible reasons for non-nesting emergences, he elimi- seagrass communities include gammarid amphipods, gas- nation of ectoparasites and epizoic algae and drying of tropods, echinoids and fish (references in Dawes et al. integument to reduce bacterial and fangal infections. 1991). The green turtles' sensitivities to parasites and diseases In the Arabian region, dugongs are among the larger are discussed in section 3.3.5. The fact that green turtles grazers of seagrasses. The smaller grazers include the can be raised and kept in captivity for years, albeit not urchin (Tripneustes grarilla), surgeonfish (Zebrasoma without major problems, attests to their hardiness. There xanthurum and Ctenochaetus striarus) and rabbitfish is some nipping and biting between males and between (Siganus rivulatus). In a quantitative study, urchin con- males and females in the courtship and mating repertoire sumption was equivalent to about 33% of the total seagrass (see section 3.1.3). growth and consumption-by fish amounted to less that Evidently because of the lack of food, little feeding 5% of the total plant growth (references in Sheppard et occurs off nesting beaches (but see section 3.4.1). Some al. 1992). long-range oceanic migrations during which adult turtles Intraspecific density-dependent nest destruction may are presumed not to feed are described in section 3.5.1. prevail on beaches, usually small island beaches, where Stabenau et al. (1993) review the suggested methods nesting space is limited (Bustard and Tognetti 1969). for resuscitation of comatose sea turtles (compression of Monk seals with pups sometimes compete with green plastron, electrical stimulation of pectoral region, inser- turtles for choice basking spaces on French Frigate Shoals tion of plastic tube into trachea followed by blowing into (Balazs 1980). the tube) and they recommend a method that has been Humans sometimes compete with marine turtles for used successfully with Kemp's ridley turtles. The field beaches. An example is the drive to acquire beachfront for the Archie Can- National Wildlife Refuge, in Florida, leatherback hatchIings and (hey attributed this to the in face of pressure to develop the beach for humans (Anon tective coloration provided by the counteishading of 1993b; Owen et al. 1994). green turtles. Most green and leatherback hatch! ignored fish and when a response occurred it 3.3.4 Predators frequently a change in course. Unlike some loggerhe Significant predators of eggs are various species of crabs green hatchlings did not become immobile when threatei and mammals (Table 14). Some crabs burrow into nests Sharks are major predators of large green turt and may provide routes for secondary predators, such as Hendrickson (1958) estimated that 4% of the adult a variety of insects. On several beaches, ants and fly lar- males on the Malaysian beaches showed signs of assur vae are associated with rotting eggs. Feral dogs are ma- shark attack -amputated flippers and missing piece; jor predators on some beaches. The elaborate camou- shell. In agreement with some of the islanders, he wa flaging of the nest site must have been naturally selected the opinion that sharks do apparently congregate in la to counter some egg predation. On the Tortuguero Beach, around the Sarawak Turtle Islands during Costa Rica, in 1977, between 24% and 38% of the nests ding season. Witzell(1987) reviewed some were destroyed by predators, mainly dogs, coatis and vul- the literature and reported that tiger sharks, Galeocei tures (Fowler 1979). In July 1993. about 63.8% of the cuvier, prey on large marine turtles, including green turti green turtle nests on Akyatan Beach, Turkey, were preyed around the world. upon by either red fox (Vulpes wipes) or golden jackal Autar (1994) recorded 82 nesting green turtles bei (Canisaurew)(Brown and Macdonald 1995). About 66% killed by jaguars (Panthers onca) on the beaches of the eggs were consumed in the vicinity of the predated Suriname between 1963 and 1973. Thirteen green turf nests. The tracks of the canids indicated that the preda- were killed in 1980 and more were still being killed tors were systematically searching for turtle nests. Pre- jaguars as recently as 1994 in Suriname. dation here occurred over at least four weeks of incuba- Stancyk (1982) reviewed some of the older literati) tion. Hatching success was only 10.8% and hatchlings on marine turtle predation and he describes some pred were preyed upon at seven of ten nests. tor control methods including chemical control, shootii Significant predators of hatchlings are crabs, including and trapping, nest transplants and egg hatcheries. at least six species of Ocypode. fishes and birds. Shallow 3.3.5 Parasites, commensals and diseases water predation by fishes is assumed to be high. "pie noc- There are several hundred scientific papers dealing d turnal emergence of hatchlings on the beach, usually but rectly or indirectly with the parasites, commensals ar, not always in one large mass, their innate orientation to diseases of green turtles. Some of the more recent an visual cues on the beach, their swim frenzy,and their coun- significant papers are cited in Table 15. Species from th tershading may be viewed as adaptations to predatorpres- five kingdoms occur as symbionts with Chelonia myda. sure. Some bacterial symbionts in the gut of the green turd Gyuris (1994) quantified hatchling predation rates by are associated with degradation of cellulose (see sectio tethering them on a 10m monofilament line and follow- 3.4.1). ing them as they swam across the water's edge to the reef The turtle barnacle, Chelonibia testudiwria, was founi crest on Heron bland, Australia. Predation rates under on the carapaces of 52.9% of 814 turtles on a feedin, different combinations of environmental variables (tide, pasture off Queensland,Australia (mean count of 2.6 bar moonphase, time of day) varied from 0 to 85% with a nacles per turtle) (Limpus et al. 1994b). PIaiyfepu. mean of 3 1%. Predation was lower during high tide than decorata occurred in appreciable numbers on the skin o during low tide. The most common predators were fishes almost every turtle. Other barnacles on individuals in this of the family Serranidae, followed by Lutjanidae and population were Tubicinella cheloniae, Stomtolepa: Labridae. Small sharks, lethrinids and eels occasionally transversa and Stephanolepas muricata. Ozobranchid preyed on the hatchlings. Most attacks were sudden rushes leeches, Ozobranchus maqoi, and/or their eggs also oc- by the predators and no hatchlings took evasive action to curred on almost every turtle in this feeding population avoid the predators. Gyuris (1994) stated that "for the Brock et al. (1976) described an outbreak of tuberculo- green turtle populations breeding in eastern Australia, most sis in captive turtles in Hawaii caused by tubercle bacilli first year mortality is caused by predation while crossing of the Mycobacieriurn avium complex. Glazebrook et al. the reef within the first hour of entering the sea." (1 993) describe a serious disease complex (ulcerative st* Wyneken et al. (1994) observed that hatchling green matitis-obstructive rhinitis-pneumonia) in captive turtles, as well as loggerhead and leatherback hatchlings, hatchlings and juveniles in Australia. Three bacteria in nearshore Florida waters, dove in response to overhead (Vibrio alginolyticus, Aeromonas hydrophila and Flu- threats. They also found that green turtle hatchlings en- vobacterium sp.) and four genera of fungi (Paecilomyces countered fewer aquatic threats than the loggerhead and sp., Peniclllium sp., Aspergillus sp., and Fusarium sp.) Table 14. Representative predatois of green turtles. Predator Location Reference

EGGS Crabs Birgus latm Aldabra Atoll Honegger (1967) Bmchyura sp. Aldabra Atoll Honegger ( 1967) Ocypode quadrata' Suriname Schulz (1975) Ocypode quadrata Costa Rica Fowler (1979) Ocypode sp. Yemen Hirth and Can- (1970) - crabs Karan k. Miller (1989) Insects Twx subemsus GaMpagm Is. deal and Orriz-Crespo (1 982) beetles Koran Is. Miller (1 989) Reptiles %rawsp. East Indies Raven (1946) Varanussp. Pakistan Minion (1966) Mammals W~Y Brown and Macdonald (1995) U. S. Virgin Is. Henty 1993 in Mackay (1994) Numa mrica Costa Rica Fowler (1979) P rocyon fotor Florida Bryam (1986) Vulpes vulpes Turkey Brown and Macdonald (1995) dogs Pakistan Minion (1966) dogs Yemen Hirth and Can- (1970) dogs Suriname Schulz (1975) dogs Costa JUca Fowler ( 1979) dogs Comoros Frazier (1985) fox Yemen Hirth and Can- ( 1970) jackal Pakistan Minion (1966) Pigs East Indies Raven (1946) ~ÈS GalApagos Is. Green and Orriz-Crespo (1982) Birds Caihanes aura Costa Rica Fowler (1979) Coragyps arraius Costa Rica Fowler (1 979) crows Pakistan . Minion (1966)

HATCHL1NGS Crabs Birgus lutm Gielop Is. Kolinski (1994b) Coenobita cavipes Europa Is. Hughes (I974b) Coenobira cavipes Tromelin Is. Hughes (1974b) Coenobita comprsssus Galipagos Is. Green (1983) Coenobita perfata Btkar Atoll Fosberg (1969) Coenobita rugosus Europa Is. Hughes (1974b) Coenobita rugosus Tmmelin Is. Hughes (1974b) Coenobita sp. Kami is. Miller (1989) Eriphia sebana Karan Is. Miller (1 989) Gmpsus grapsus Trindade Is. Moreira et al. (1995) Grapsus Lagostoma Trindade Is. Moreira et al. (1995) Grapsus tenuicrusratus Karan Is. Miller (1 989) Grapsus sp. Europa Is. Hughes (1974b) Grupsus sp. Tmmelin Is. Hughes (1974b) Ocypode ceratopthalmw Cocos-Keeling Is. Gibson-Hill (1 950) Ocypode cemroprhImus Malaysia HendriclÈo ( 1958) Ocypade ceratqrhalmus Heron Is. Bustard (1 966) Table 14. Continued. Redaior Location Reference. ... Ocypods cemopthaImus Aldabra Atoll Mill (1975) Ocypode cemopthaImus Hawaii Baltzs (1980) Ocypode ceraloplhttlimis Comoros Pnaier (1985) Ucypode gaudichasiss Gabspagos Is. Green (1983) Ocypoa'e lawis Hawaii Balazs (1980) Ocypode madagascariensis Comoros Raziw (1 985) Ocypodequ.adra!a Suriname Schdz (1975) Ocypode Costa Rica Bowler (1979) OcypWtt saratw Karan Is. Miter (1989) Ocypode SP. Yemen Hirth and Can- (1970) crabs Trindade Is. Olswi (1981) . Fishes Camignobilis Aldabra Atoll Honegger (1967) Camlumbru Trindade Is. Monani et al. (1995) Carelutriniisspallamni Heron Is. BusBri (1 966) Choerodon cyanodus Heron Is. Gyuris (1995) Coryphaena hippurus Horida Wih(W4) Cmmileples atliveiis Heron Is. Gyuxis (1994) Echidnae sp. Europe Is. Hughes (1974b) Echidnae sp. Tromelin Is. Hughes (1974b) Epinephelus labriformis GaUpagos Is. Epinephelus sp. Heron Is. Gyuris (1994) Epinephelus sp. Yemen Hilth and Can- (1970) Epimphetv sp. Trindade Is. Moreira et al. (1995) Eulamia spaltavani Fairfax Is. Booth and Peters (1972) Gemalbacora Europe Is. Hughes (1 974b) Hynnis cubensis Trindade. Is. Moreira et al. (1995) LuIianus aqenil'mculatus Europa k.. Hughes (1974b) Lutianus bohar Aldabra Aioll Honegger (1967) Lutianus sp. Heron Is. Gyuris (1994) Mycte~opercasp. Trindade Is. Moreira et al. (1995) Sphymenu barracuda Trindade Is. Moreira et al. (1995) black tip reef shark Heron Is. Gyuris (1 994) sharks Ascension Is. Can- and Hirth (1962) sharks Aldabra Honegger (1967) sharks Yemen With and Can- (1970) sharks Surinam Schulz (1975) sharks Galhpagos Is. Green (1983) Reptiles Boiga dendmphila Malaysia Mastic~Aisanthong Was Revillagigedo Python reticulatus Malaysia Vamnus sp. Malaysia Mammals Mas mu~ulus Kami Is. Miller (1989) Nasua narica . Costa Rica Fowler (1979) Ranus exulam Bikar Atoll Fostwg (1969) cats Ascension Is. Can- and Hilth (1962) cats GaUpagos Is. Green (1983) cats Aldnbra Atoll Seabrook (1989b) dogs snrimune Sdmlz (1975) dogs Costa Ria Fowler (1979) gray fox Florida Broward County Erosion Prevention District (1987) Thble 14. Continued. Predator Location Reference rats Malaysia Hendrickson (1958) rats Galiipagos Is. Green (1 983) Birds Ardea cinerea Aldabra Atoll Honegger (1967) Cathartes aura Costa Rica Fowler (1 979) Coragyps atratus Suriname Schuiz (1975) Coragyps atratus Costa Rica Fowler (1 979) Corvus albus Aldaba Atoll Honegger (1 967) Corvus albus Europa Is. Hughes (1 974b) Corvus albus Comoros Frazier (1985) Corvus cow lslas Revillagigedos Awbrey et al (1984) Dryolimnas cuvieri Aldabra Atoll Frith (1975) Fregata arief Aldabra Atoll Honegger (1967) Ffeeara minor Aldabra Atoll Honegger (1967) Fregma Minor Empa li. Hughes (1 974b) Fregata minor Tmmelin Is. Hughes (1974b) Laws novaehollandiae Heron Is. Bustard (1 966) Milvus migmns Comoros Pnizier (1985) Nycwaassa viotacea Galipagos Is. Green (1983) Nycticorax caledonicw Raine ls. Limpus et al. (1993) Phoenicopterus ruber Aldabia Atoii Honegger (1967) Sterna anueiherus Karan Is. Miller (1989) Threskiomis aethiopica Aldabra Atoll Frith (1975) burrowing owl lslas Revillagigedo Brattstmm (1 982)

IMMATURES AND ADULTS Fishes .d Carcharhinus longimanu.~ South Africa Hughes. (1 974b) Epinephelus tauvina Hawaii Balazs (1980) Gafeacerdocuvier Worldwide W1tzell(1987) Promicrops lanceolatw Tonga Witzell(1981) Cmodylus pomsus Ponape Is. Alien (1974) Mammal Pantlieru onca Suriname Autar (1994)

Table 15. Some symbionts (parasites, commensals, mutualists) of green tmiles. Complete geographic locations of some symbionts and their host are nrovided In the references. Species Location Reference

Captive (Australia) Glazebrook and Campbell (1990a), Glazebrook et al. (1993) Arizona hinshairi Captive (Australia) Glazebrook and Campbell (1 99Oa) Esdierichia coli Captive (Australia) Glaabrook and Campbell (1990a) Flavobacterium sp. Captive (Australia) Glazebrook and Campbell (I %@a), Glazebrook et al. (1993) Mycobacterium sp. Captive (Australia) Glazebrook and Campbell (1990a) Mycobacterium avium Captive (Hawaii) Brock et al. (1976) We15. Continued- Species Location Reference

Pseudomonas aeruginosa Captive (Australia) GIazebrook and Campbell (19%) PseUaOmonasfluorescens Captive (Australia) Glazebrook and Campbell (1990a) Salmonella enteritidis Captive (Australia) ~hzebrookand Campbell (1 990a) Streptococcus sp. Captive (Australia) Glazebrook and Campbell (1 990a) Vibrio alginolyticus Captive (Australia) Glazebrook and Campbell (1990a). Glazebrook c (1993) Unidentified Nicaragua Fenchel et al. (1979) Protist- Achnanthts sp. Captive (USA) Schwanz (1992) Sulanudimn baaeriophona Caribbean Fenchel(1980) Caryospom cheloniae Captive (Grand Cayman) Leibovirz et al. (1978) Caryospora cheloniae Australia Gordon et al. (1993) Eniamoeba inv'fldens Captive Frank et al. (1976) Lichmo~phaehmnbeqii Captive (USA) Schwam (1992) Octomitw sp. Caribbean Fenchel(1980) NItsiua sp. Captive (USA) Schwaitz (1992) Tvpanosom restundinis Not slated Ernst and Ernst (1979) Fungi Aspergillus sp. Captive (Australia) Glazebrook et d. (1993) Cladosporiumsp. Captive (Grand Cayman) Jacobson et al. (1 979) Fusariwn sp. Captive (Australia) Glazebrook et al. (1993) Paecilomyces sp. Captive (Grand Cayman) Jacobson et al. ( 1979) Piiecilomyces sp. Captive (Austdia) Glazebrook and Campbell (IMa) Paeciliimyces sp Captive (Australia) Glazebrook et al. (1 993) Peniciilimn sp Captive (Australia) Glazebrook et al. (1993) Sporotrichiumsp, Captive (Gab! Cayman) Jacobson et al. (1979) Plantae Acmchaeiium sp. Johnston Atoll Balazs (1985a) Acrochaetium gmcile Hawaii Balazs (1980) Chadophorasp. Johnston Atoll Balm (1985a) &tocarpus iiuiicus Hawaii Balazs (1980) Eittemmorpha ckuhraia Hawaii Balazs (1980) Falkenbergia rufolanosa Hawaii Balazs (1980) Lyngbya cinerescens Hawaii Bates (1980) Ly~byamajuscuia Hawaii Balazs ( 1980) Lyn&ya semipiens Johnston Atoll Balazs (1985a) Melobesiu sp. Hawaii Bate (1980) Oscillatoria sp. Hawaii Bdazs (1980) Pilina sp. 3ohn.uim Atoll Balazs (1985a) Polysiphonia dotyi Hawaii Bks(1 980) Potysipiwnia tsudana Hawaii Balazs (1980) Polysiphonia fsudana Johnston Atoll Balazs (1985a) Sphacelariafuivigeria Hawaii Balm (1 980) Sphacelaria novae-hollundiae Hawaii Balazs ( 1980) Sphacelaria tribuioides Johnston Atoll Balm (1 985a) Ulw fwciata Brazil Frazier et al. (I 992) Urospum sp. Johnston Atoll Balazs (1985a) Animalia Cnidaria Hydrozoa Tubuiuria sp. Brazil Platyheiminthes Trematoda Adenogasler serialis Mexico Ernst and Ernst (1977) Amphiorchis amphwrfhis Not slated Smith (1972) Table 15. Continued. Location Reference

Anguxllctyum Iongum Australia Blair (1 986) Angiodictyum Iongum Hawaii Dailey et aL (1993) Angiodiciyum pamllelum Puerto Rico Dyer et al. (1991) Angiodictyum parallelurn Egypt Sey (1977) Ang iodiayurn posterovitellatum Australia Blair (1986) Angiodictyum sp. Australia Glazebrook and Caropbell(1990b) Calycodes anthor Egypt Ernst and Emgt (1977) Carettacola hawaiiensis Hawaii Dailey et aL (1993) Charaxicephalus mbustus Egypt Sey (1977) Cricocephalus albus Trinidad Gupta (1961) Cricocephalus megastomus Taiwan ErnstandErnst(1977) Cricocephalus resectus Egypt Sey (1977) Cdcocephalus ruber Australia Ernst and Ernst (1977) Cricocephalus sp. Captive (Australia) Glazebrook and Campbell (1990a, b) Cymatocarpus solearis Not stated Ernst and Emst (1977) Desmogonius desmogonius Taiwan Ernst and Ernst (1977) Desmogonius sp. Australia Glazebrook and Campbell (I990b) Dewembarus chelonei Trinidad Gupta(1961) Dewembarus pmteus Egypt Sey (1977) Dewembarus proleus Puerto Rim Dyer et a1 (1991) Dewembanis viridis Not stated Ernst and Barbour (1972) Diaschistorchis lateralis Japan bstand Ernst (1977) Diaschistorchis pdus Australia Ernst and Ernst (1977) Distoma testudinis Not stated Ernst and Ernst (1977) Disromm constricmm Not stated. Ernst and Barbour (1972) Enodiotrema megachondrus Egypt Ernst and Ernst (1977) Glyphicephalnslobatus Pueno Rim Dyer et al. (1991) Glyphicephalus solidus Brazil Ernst and Ernst (1977) HaemoxSnicon chelonenecon Not suited Smith (1 972) Haemoxenicon stunkardi Panama Ernst and Emst (1977) Hapalotrema darsopora Hawaii Dailey et al (1993) Hapalorrema loossi Egypt Ernst and Ernst (1977) Hapalotrema pambanensis India Gupta and Mehiotta (1981) Hapalotrema postorchis Hawaii Dailey et a1 (1993) Hapalotrema sp. Australia Glazebrook and Campbell (1990b) Hapalotrema sp. Captive (Australia) Glazebrook and Campbell (1990b) Learedius europaeus Not stated Ernst and Ernst (1977) Learedius learedi Panama Ernst and Ernst (1977) Learedius learedi Captive (Grand Cayman) Greiner et al (1980) Learedius learedi Bermuda Rand and Wiles (1985) Learedius learedi Puerto Rim Dyer et aL (1991) Learedius learedi Hawaii Dailey et aL ( 1993) Learedius loochooensis Japan Ernst and Ernst (1977) Learedius orientalis Pakistan Ernst and Ernst (1977) Learedius orientalis Puerto Riw Dyer et al. (1995) Learedius similis Not stated Smith (1972) Learedius sp. Australia Glazebrook and Campbell (1990b) Medioporus cheloniae Japan Ernst and Ernst (1 977) Memcetabulum invaginaturn Brazil Ernst and Ernst (1977) Microscaphidim aberrans Egypt Sey (1977) Micro.scaphidium abermns Australia Blair ( 1986) Micmscaphidiwn reficulm Egypt Sey (1977) Micmcaphidium reticulare Puerto Rico Dyer et al. (1995) Microscaphidim nticulare Australia Blair (1986) Micivscaphidium warui Australia Blair (1986) Table 15. Continued. Species Location Reference

Moaostoma psemlamphisromum Not stated Bust and Ernst (1977) Monticellius idcum Pakistan Bust and Ernst (1977) Neoctangium tiavassosi Trinidad hpt~(19611 Neospimrchis schistosomatoi&s Not stated Smith (1972) Neospimi'dlis schistosoinatoides Bermuda Rand and Wiles (1985) Octmgium hasta Egypt Ernst and Ernst (1977) Octongium hyphalm Australia Blah- (1987) Octmgium sagina Egypt SCT (1977) 0-~t sagina Australia Blair (1987) Octmgium sagiza Puerto Rico Dyer ctal. 0991) OcqiummkDnoi Malaya ErnstandEmst(1977) Oaangium sp. Austraila Glazebrook and Cainpbell(1990b) Onchulasma amphiorchis Mexiw Caballero y Caballwo (1962) Pac~sdusirwatus Ranee Bust and Ernst (1977) Paralepodem acarlaeum Puerto Rico Dyer et al. (1995) Phyllodisfomm cymbiforme Not stated Ernst and BaAour (1972) Pleumgonius bibbus Egypt Ernst and 5st(1977) Pleumgonius chelonii Pakistan Ernst and Ernst (1977) Pleuwgonius linearis Brazil Ernst and Enist (1977) Pleumgonius longiusculus Egypt Ernst and Ernst (1977) Pleumgonius mehrai Trinidad Gupta (1961) Pleumgonius mehrai Captive (Grand Cayman) Greiier et aL (1980) Pleumgonius minutissirnus Egypt Ernst and Ernst (1977) Polyangiurn Unguatula Egypt Sey (1977) Pofyangium linguatula PuertoRico Dyer et al. (1991) Polyangiurn linguatula Hawaii Dailey et al. (1993) Poiyangium miyajimai Malaya Ernst and Ernst (1977) Polyangium sp. Australia Glazebrook and Campbell (1990b) Poiygorgym cholados Australia Blair (1986) Poiystoma my& Not stated Ernst and Bartour ( 1972) Pnmocephalu$ obiiguus Brazil Emst and Emst (1977) Pyelosomo cochlear Hawaii Dailey et ai. (1993) Pyelosomum cochlear Puerto Riw Dyer et al. (1991) Riytidodes gelatinosus Egypt Emst and Ernst (1977) Australia Glazebrook and Campbell (1990b) Rhylidodoides intestimlis Captive (USA) Price (1939) Rhytidodoides sirnilis Captive (USA) Price (1939) Schtzamphistomoities chelonei THnidad Gupa (1961) Schiwmphistomoidesspiimbsum Brazil Ernst and Ernst (1977) Schiwrnphutoimimerratum Australia Blair (1983) Schwmphisiomum sceloporurn Australia Blah- (1983) Schiwmphiswmum sp. Australia Glazebrook and Cainpbell {WOb) Sdmamphistomm sp. Puerto Rim Dyer et al. (1995) Spmrchis parvum Not stated Emst and Ernst (1977) Squuroacetabulum solus India Simha and Chattopadbyaye (1970) Ce8toda Ancimcephaius imbricarus Not stated Bust and Ernst (1977) Tentacularia co~~phaenae Not stated Ernst and Emat (1977) Nematoda Angusticaecurn holoptera Empe Erost and Ernst (1977) Anisakis sp. Captive (Australia) Burke and Rodgers (1982) Anisakis sp. Glazebrook and Campbell (1990a) Porrocaecum sulcatum USA Allison et al. (1973) Sulcascaris sulcata Not staled Spmt (1977) Tonaudia tonaudia Ceylon Emst and Ernst (1977) Species Location Reference i Mdlusca Mylilus edulis platencis Brazil Frazier et al. (1992) Pleumploca princeps Galapagos Frazier et a1 ( 1985) Bryozoa Electra sp. Pakistan Frazier et al. (1992) Unidentified Brazil Frazier et al. (1992) Annelida Hirudinea Ozobranchus branchiatus USA Nigrelli and Smith (1943) Ozobranchus branchiatus Malaysia Hendrickson (1 958) Ozobranchus branchiatus Costa Rica Sawyer et al. (1 975) Ozobranchus branchiatus Hawaii Balazs (1980) Ozobranchus margoi Captive (USA) Schwanz (1974) Ozobranchus margoi Captive (Hawaii) Balazs (1980) Ozobranchus sp. Comoros Frazier (1 985) Ozobranchus sp. India Frazier (1989) Arthrupoda Arachnids Unidentified mites Australia Glazebrook and Campbell (1990b) Cinipedia Chelonibia testudinaria Malaysia Hendrickson (1958) Chelonibia testudinaria Aldabra Frazier (1 97 1) Chelonibia testudinaria Hawaii Balazs (1980) 1 Chelonibia testudinaria Peru Brown and Brown (1982) Chelonibia testudinaria Comoros Frazier (1985) Chelonibia testudinaria India . Frazier (1 989) 4 Chelonibia sp. EWP~ . Hughes (1 974b) Chelonibia sp. Tromelin Hughes (1974b) Chelonibia sp. Papua New Guinea Spring (1983) 1 Chelonibia sp. Australia Glazebrook and Campbell (1990b) 1 Lepas sp. Peru Brown and Brown (1982) Platylepas hexastylos USA Schwartz (1960) Platylepas llexastylos Aldabra Frazier(1971) Platylepas hexastylos Hawaii Balazs (1980) Platylepas hex~~tylos Comoros Frazier (1985) Platylepas hexastylos Johnston Atoll Balazs (1985a) Platylepas hexastylos India Frazier (1 989) 1 Platylepas sp. Papua New Guinea Spring (1983) 1 Platylepas sp. Australia Glazebrook and Campbell ( 1990b) 4 Srephanolepus muricata Malaysia Hendrickson (1 958) 1 Stephanolepus ~nuricata Hawaii Balazs (1980) 1 Slephanolepus sp. Europa Hughes (1974b) 1

Stomatolepus elegans Not slated Ernst and Barbour (1972) ¥ Amphipoda Hyachelia tortugue Hawaii Balazs (1980) Hyachelia tortugae Johnston Atoll Balazs (1985a) Isopoda Eurydice sp. Malaysia Hendrickson (1958) Decapoda Planes cyuneus Peru Brown and Brown (1982) Insecta Eumacmnychia sternalis Mexico Alvarado and Figueroa (1990) are implicated and a method of treatment is suggested. severe cases, turtles have reduced vision and difficu Haines (1988) summarized what is known about gray eating and swimming. Affected individuals are alsc patch disease in cultured green turtles. The cause of the mic compared to normal animals. disease is a herpes-type virus and the result is a macera- Turtles with fibropapillomatosis have been rep tion and erosion of the skin and carapace. Stress factors from such widely scattered places as theCaribbean (P and water temperatures appear to play important roles Rico, U. S. Virgin Islands, Cayman Islands, Domil since crowding and increased temperatures increase oc- Republic, Antigua, Barbados, Trinidad. Nether1 currence of the infections. Treatment with metabolic in- Antilles, Bahamas. Mexico, Belize, Panama. Color hibitors is available. Jacobson et al. (1986) suggested Venezuela), Florida, California, Hawaiian Islands, Js that a herpesvirus was involved in the pathogenesis of Malaysia, Indonesia and Australia (Jacobson 1990; ' LETD (lung, eye and trachea disease) in captive green iams et al. 1994). Fibropapillomas were seen on at turtles. LETD is a respiratory disease characterized by two nesting green turtles in Yap State, Federated Stat gasping, buoyancy abnormalities and inability to dive Miponesia, in 1992 (Kolinski 1994a). The appear properly. The eyes are often covered with caseous exu- of fibropapillomas on some San Diego Bay green tu date, which is also seen around the glottis and within the is presumed to be a recent occurrence (McDonald trachea. The systemic diseases of fanned animals, in- Dutton 1990). Some fishermen informed Guada e cluding diseases of the integumentary, sensory, skeletal, (1991) that green turtles with fibropapillomas are c muscular, digestive, respiratory, cardiovascular and ex- monly caught, eaten and sold illegally in some pan I cretory systems, are discussed by Glazebrook and Venezuela The epizootic of fibropapillomas in the 4 Campbell (1990% b). ibbean has occurred since the mid- 1980s, about five y Sinderrnann (1988) briefly summarized coccidian dis- later than similar epizootics in Florida and Hawaii, ease in farmed green turtles. Gordon et al. (1993) re- all may be part of a panzootic (Williams et al. 1994) ported on a recent epizootic of coccidiosis among green 166 green turtles recaptured around the Cayman Isla turtles off the coast of Queensland, Australia. Clinically after being released as hatchlings or yearlings from the most consistent signs were pronounced weakness and Cayman Turtle Farm, 66% were infected with cutane depression. They speculated that a heavy concentration fibropapillomas. Seventy-two percent of the individi of infective stages on the feeding grounds may have pre- retaken within less than one year from release did cipitated the epizootic. Mycotic pneumonia has been di- exhibit any fibropapillomas while only 26% of the tur agnosed in captive juvenile turtles (Jacobson et al. 1979). recaptured after more than a year after release lac1 Green turtles are parasitized by a wide variety of trerna- fibropapillomas (Wood and Wood 1993b). Sixty-n lodes, including six species of Pleurogonius, at least five percent of 26 immature green turtles captured in Flor species of Learedius, and at least four species of Bay, Florida, in 1991, exhibited fibropapilloma (Schroe Cricocephalus, Hapalotrema, and Octangium. and Foley 1995). The prevalence of fibropapillom Glazebrook et al. (1989) found the incidence of cardio- the Indian River. Florida, green turtle population var vascular flukes and/or their eggs (Haplotrema spp. and from 40 to 60% between 1982 and 1992. In 1993 its pre Learedius spp.) in marine turtles from northeast lence was 20% (Ehrhart and Redfoot 1995). About h Queensland to be 4.8% (5 of 104 from turtle fanns). 33.3% of the turtles in Kaneohe Bay. Hawaii, now have turn (5 of 15 from an oceanarium) and 77.3% (17 of 22 wild and up to 10% of the nesting females on French Frig; turtles). Of the 27 turtles infected, 23 were green turtles Shoals, Hawaii, possess fibropapillomas (respective and 4 were hawksbills. The average number of flukes per Balazs et al. 1993 and Dailey et al. 1992). Fibropapilloir host was 47. Gross pathological changes associated with were recorded on 62 (7.9%) of 784 green turtles at afee the presence of flukes in some individuals included thick- ing site on Moreton Banks, Queensland,Australia (Limp ening of arterial walls, thrombus formation and an excess et al. 1994b). of pericardial fluid. Microscopically,the essential change The etiology of fibropapilloma is unknown, althou; was that of chronic inflammation. there are several hypotheses linking viruses, parasite The source of infection of Anisakis sp. in the captive pollutants, or combinations of these agents to the diseas turtles in the Torres Strait appeared to be raw sardines Jacobson et al. (1989) described fibropapillomas in s (Burke and Rodgers 1982). juvenile Florida green turtles. The cutaneoi Remoras (Echenesis sp.) have been seen attached to fibropapillomas were characterized by papillary proli green turtles in a wide variety of locales. eration of the epidermis on broad fibmvascularstalks, h. Fibropapillomatosis is a debilitating and life-threaten- trematode eggs were seen in any of the biopsy (N=2! ing disease of green turtles. Fibropapillomas of various specimens. Brooks et al. (1994) examining, histolog sizes can be found on several body sites, including the cally, the eyes of three stranded juvenile turtles fra skin, eyes and surrounding tissue, mouth and viscera. In Florida found ocular fibropapillomas composed of a overlying hyperplastic epithelium, a well-vascularized collagenous stmma and a population of reactive fibro- blasts. A herpesvirus was found in cutaneous fibropapillomas in two juvenile turtles from Florida (Jacobson et al. 1991). In preliminary experiments on the transmission of fibropapillomatosis, Herbst et al. (1994a) concluded that it is unlikely that trematode eggs are a primary cause of the disease. In more recent trans- mission experiments, Herbst et al. (1 994b; 1 995) have shown that Ebropapillomatosis can be experimentally transferred to disease free recipient turtles. Latency to tumor development is about four months. The experi- ments suggested that fibropapillomwis is caused by a subcellular agent and most likely is a virus. Although several field studies suggest that high green turtle fibropapillomatosis is associated with near-shore marine habitats that have been impacted by huxnan activities, Herbst and Klein (1995a) caution that the role of envi- ronmental cofactors in fibmpapillomatosis will require careful scientific study. Ten Hawaiian turtles with fibropapillomas were exam- ined by Dailey et aI. (1992) and were found infected with 232 worms comprising seven species of digenetic trema- todes. Examining tumors fiom Hawaiian green turtles, Dailey and Moris (1995) found that dl tumors examined (N=61) contained spirorchid eggs and they feel that the infdonsuggests a direct link between fib&papillornas and spirorchid trematode infections. Working with Florida pnturtles, Greiner (1995) stated that there is a possi- bility that trematode eggs are indirectly or directly in- volvd with fibropapillomas. All the tumors (N=39) from Caribbean green turtles that Williams et al. (1994) exam- ined histologically had spirorchid eggs, and the leech, Oz~branchusbrmchiutus, was associated with three dis- eased individuals. Aguirre et al. (i994b) evaluated Ha- waiian green turtles for potential pathogens associated with fibropapillomas but were unable to isolate the etio- logic agent. Selected tissues from juvenile Hawaiian gnxn turtles afflicted with fibmpapilloma did not contain any of the selected organochlorines, polychlorinated biphe- nyls, organophosphates, or carbamate insecticides in con- centrations above stated methods of detection limits (Aguirre et al. 1994%c). AgWet al. (1995) determined that subadult Hawai- ian turtles with fibmpapillomas were immunosuppressed and chronically stressed prior to being subjected to cap- ture stress. This determination was based upon raised corticosterone concentrations and a positive correlation with hetempKmymphocyte ratios. A flow cytometric DMcontent analysis indicated that fibropapillomas and visceral tumors have normal cell cycles (Papadi et al. 1995). Normal and tumor-bearing immam turtles were moni- tored with ultrasonic tmtsmitters in Kanmhe Bay, Oahu, lora. Green turtles not only change their diet during their mals consistently ingesting a mixed diet would almm ontogeny but their gut proportions are also modified. The cerrainly develop a microflora capable of &@ng tb large intestine of pt-haahlings is about half the leng~ various complex carbohydrates.'' Lanyon et al. (1989 of the small intestine while that of adults is more thm describe how .smonal fluctuations in total seagrass nu twice the ]en@ of the small intestine. These changes in trients myhave impmt consequences for the nutxi. gut proportions are correlated with a shift from a camive tional status and life history of green turtles. Limpus anc row, or omnivorous, to a herbivorous diet (Davenprt et NichoIls (1988) have demonstrated a linkage between tlx al. 1989). Herbivory is associated with a vol~nouslarge ENS0 (El Niiio Southern Osillation) and the number oi intestine - a place where fdspends a long time. BeIs green turtles that bdon eastem Australian beaches two and Renous (1992) analyzed film of the feeding behav- ye- lam. They suggest that the J%SO myregulate ior of captive, juvenile green turtles and stated that the nesting numbers via a nutritional pathway. Philander protraction-reuaction cycle of the forelimb is clearly as- (1990) describes some of the research going on regarding sociated with the gape cycle. the southern oscill&on. 1-R.Wd and F.E. Wood (1977) The adaptations of the Caribbean pnturtle to a diet and F.~.J&oodand J.FL Wood (1977) have determined the chiefly of seagrass, Thalassia testudinum, which is high quantitative requirements of hatchlings for lysine, tryp- in cellulose content and thus low in quality, are hindgut tophan, me~onine,valine, Ieucine, isoleucine and pheny- microbiil fermentation and selective grazing @jorndal lalanine. Wood and Wood (198 11 found that young captive 1982b). Bjorndal (I 979) has shown how cellulose is di- turtles fed on rations containing 35% protein grew fa- gested through microbial fernenation as efficiently in d~ turtles fed on rations with 30% and 25% protein. In the green. turtle as it is in dugongs and ruminants, and, captive pnturtles on the Cayman MeFarm, feed con- Bjorndal (1980a) found that green turtles on a E version varies from 1.2 to 6.5 units of diet to unit of body tesfudinurn grazing pasture in the Bahamas maintained weight, increasing with size of the individual (WtocI 1991). grazing plots of young leaves by constant recropping. The Bjomdd (1985) has reviewed the litemwe on nutrient di- young leaves are higher in protein and lower in lignin gestibility for organic matter, cellulose, nitrogen and car- than older leaves. Zieman et a1. (1984) hypothesized that bon and she points out the significant difference in some the foraging behavior of green turtles evolved to avoid digestibili~values between fmdand wild mtles. the epiphytic carbonate of the upper regions of seagrass leaves. On the other hand, Williams (I 988) observed that 3.4.2 Food green turtles ate all accessible K resrdbum in a stressed The most comprehensive studies of the food of green pasture in the Virgin Islands. turtles have primadly come from examinations of sm- Vicente and TaUevast (1995) surveyed green turtle for- ach contents of large individuals. Representative plants aging pastures mund six islands in the Cmonwealth eaten by green turtles in the Pacific, Atlantic and Indian of Puerto Rice and the U. S. Virgin Islands. Among their Ocean regions are given in Table 16. It is interesting to findings were that green turtles graze more frequently note that at least nine species of G~ciluriuand at least along extensive mntinuous or discontinuous bands be- eight species of Sargassum are eaten. At least eight spe- twwn deep coral reef habitats or barren mud bottom and cies of Caulerpa and six species of Codium are ingested, dense grass W;on extensive seagrass beds, green turtle with some species of each eaten in the Pacific, Atlantic grazing was always limited to the deeper zones of the and Indian Ocean regions. Some species of the seagrasses bed; when several species of seagrasses occur together, Halophila and Syringodium are also eaten in the three the turtles do not disri~nateamong the species; and, large ocean regions. juvenile turtles graze on shallow (lm) and on deep (15.2m) 7'hirty-four species of algae were recorded from the grass beds and on both exposed and protected beds. stomachs of three adult and one subadult Wle fkom the Green turtles at a Thlussiu iestudinum site in the Ba- Ogasawara Islands (Kurata et al 1978). Brown algae hamas consume the equivalent of about 0.24%to 0.33% formed the greatest bulk of the diet and was represented of their body weight each day (Bjorndal 1980a). The car- by 19 species. Some hydmzoa were found in the stom- rying capacity of a X restudinurn feeding pasture in the achs, as well as a piece of plastic. Caribbean was estimated at one turtle per 72 m2 mjorndal Gamett et al. (1985b) examjned the stomach contents 1982b). In a stressed Z testdinurn pasture in the Carib- of 44 turtles from Torres Strait, 34 of which were adult bean, Williams (1988) estimated a carrying capacity of females, and found that six genera contributed 73.5% of one turtle per 669 - 3,946 m2. the total dry matter weight: Hypnea (27.7%). hurenciu Garnett et al (1985b) found that green turtles in the (I 1.9%). Caulem (9.8%). Edalh (9.8%), Sargassum Torres Strait eat both algae and seagrass and their review (5.9%), and the seagrass Thulassia (8.8%), Red algae of the literature indicates that algae are not nu~tionally made up the bulk of most stomach contents; large mun~ superior to seagrass. Bjorndal eta!. (1991) state that "Ani- of brown or green algae were eaten by a few individuals, Table 16. Repcsenptive plant food of rndy subaduit nnd adult green ~1~.few exceptions, anidf~od is eatm only in small amounts by wild, lqegem tiutles and k dim& in he text. Food Mon Refemme CMorophyta Anadyomene sp. Gamen et d. C1985b) A~minvilieari&uensis ogasawara Is. Kurata et d. (1978) Avmhillea sp. Brazil Ferreh (I 968) B~psLspemta 3otmston Atoll Balazs (I 985a) Caulerpa bmchypus Tom Strait Gamen et d. (1985b) Cauletpa cupmssoides Brazil Fmirn (1968) Caulerpa cupmss&s Tom Strait Gmnet al. (1985b) Cadetpa lenfili#era Tom Strait emn et d,(1985b) Cadezp mexicam Br&?.il Ferreira (1 968) Caulerpa mexicana Cornoms Frrizier (I 985) Gulerpa prolgera Brazil Ferreim (1968) Gaulerpa pml#eru Mimgua Mortirnez (1981) Cauleqw mcemsa Hawaii Enlam (19801 Caulepa racemsa Johston Atoll Bb[1985a) Cauierpa racema Tom Strait Gamen et al. (1985b) Caulerpa sertuiarioides Br8zil Femira (1%8) Caulerpa senularioides Nicaragua Mortim (1981) Caulerpa sertuhrioides Tom Strait Gwen et al. (l985b) Cauletpa urvilliam Tom Strait Gamett et at. (1985b) Caulerpa sp. Tomes Strait Gmnu al. (198%) Cauleya sp. Aldnbra Atoil Frazier (1971) Chaetomrpha area Oman Ross (1985) Chaerotnorphn sp. Cornoms Frazier (1985) Chetomrph sp. Tom Strait Gamen et al. (1985b) Claxlophora sp. ...Cornoms. Fder(1985) C&um adhaemu O@Sa~araIs. Kurata et al. (1978) Codium urabicum Hawaii Balm (1 980) Codium edule Hawaii Balazs (1980) Codium i~thnwcladum Bd Ferreira (1 968) Codium is#bmi&m Nicaragua Mortirner (1981) Codium phasmicum Hawaii B&aa (1980) Codimtomentosum ogamwara Is. Kurata et al. (1978) Codlum sp. AIdabra Atoll Frrizier (1971) Codium sp. 'rbms strait Gamen et al. (I 98%) Dictyosp/uzeria cawmosa ogasawm Is. Kurata et d. (1978) Dicrpsphaeria wr.tltysii OgaSawaIn Is. Kurata et al. (1978) &femrpha$mosa GJmm Fmier (1985) Haiimeda gmciiis Frrizier (1985) Halimeda tuna Frazier (1985) Halimeda sp. Nicaragua Mortim(1981) Halimeda sp. Tom Strait Gamen et al. (1985b) Mowarum oq.rpermum Brazil kdra(1968) Pe~icilluscapilazus Nicaragua Mortimer (1981) Rhizoclonium sp. Tmstrait Garnen et ale(1985b) UhteuJabeEum Nicaragua Mortirner (1981) Udotea sp. Tom Strait Gamen ad.(1985b) Ulvafasciafa Bnuil Ferreh (1968) Ulwficiara ogawwara Is. Kurata et al. (1978) Ulwjmciata Hawaii l3alazs (1980) Ulw lacruca Ross (1985) Ufvalactuca Cornoms Fmier (1 985) Ulm penusa ogasawara Is. Kurata et d. (1978) khnia aegagrc~pila Tokelau Balm (1983b) OgmawaIa Is. Kmta et al. (i978) Tomes smit Gmm ex a]. (1985b) ogasawm Is. ffirata et al. (k 978) TomStrait Gmen et al. (1985b) Bd Femeira (1968) Mo~w(1981) Fmiex(1985) Kmet al. (1978) Fmh(1%8) Games et al. (1985b) Kwfa et al. (1978) Mer(1985) Ogasawam Is. Kurata et al. (1978) ogasawam Is. Kmta a al. (1978) Tom Strait Gatnett et al. (198%) Tones Strait Gmen et al. (1985b) ogesawm Is. * Kwta et al. (1978) Ogasawam Is. Kmta et a]. (1978) Ross (1985) Kurata ex al. (1978) Ferreira (1968) Fmk(1%8) og~waraIs. Kumta et al. (1978) Nicamgua Mottimer (1981) Nicamgua Mortjmer (1981) oman Ross (1985) opsawara Is. Kmta er al. (1978) ogesawaia Is. Kuraraet aL (1978) Brazil Fem'im (1968) Nicaragua Mortimer (1981) Torres strait Camen et al. (1985b) Brazil Ferreira (I 968) Tom Smit Gmen et al. (I 98%) Moher(1981) Opsawm Is. Kurata et al. (1978) ogasawaia Is. Kurara et a1 (1978) Hawaii Baiazs (1980) Tcklau Balm (1983b) TmStmil Gamett et al, (1985b) ogasawm Is. Kumet al. (I 978)

Brazil Fmeh(1968) TomStrait Gmen et al. (1985b) Torres Strait Gmmet at. (1985b) Brazil Fen-eim (1968) Hawaii Balazs (1980) T9rres strait Gmett et a1, (1985b) Feer(1985) Femh (1968) Nicaragua Mo~rne.r(l981) Tom Strait Gamett et al. (1985b) Tom Strait Gmen et a]. (I 985b) Brazil Femira (1968) B~thmimtn'quetrm Bra72 Fd(1968) Caulacantb sp. Toms Strait Gameu et aI. (198%) Cenfmeraselavulatm ogasawara Is. Kuma et al. (1978) Cenrmems clavukm Cornoms Fder(1985) Cemnzium sp. TomSmit Garnett et 81. (1985b) Chaqia parvuk comms F&er (19851 Champla sp. TomStrait Garrieu a aL (1985b) CMriasp. TomStrait Garma et d. (1985b) Coelothrixidea Cornow Frnzjer (1985) lot^ sp. Tom Strait Games eal. (1985b) Corallina cubemis Nicaragua Mdmer (1981) &rai~ina medifermnea ogasawm Is. Kurata et a!. (1978) Cryptonem'a crenulata Brazil Fd(1968) Cwjonemia crenuka Ni- Modm(1981) Cryptone& lmuriam Brazil Fenrim (1968) Dasya sp. Toms Strait Gmett e~a1. (1985b) bantimladia duperryi Bdl Fmim (1 968) MmWduperryi Ni Mod~(1981) Eu&um murieatum Tom Strait Gamett eta!. (1985b) Euckum sp. Bdl Femisa (1968) ElLChem sp. Toms Strait Game8 et al. (1985b) Cakimu~obtwafa Brazil Fmh(1968) Gaiaxaum veprecula Cornoms Fwier (1985) Calmaura sp. ogmwmIs. Kurata et al. (1978 Gal-m sp. TomSuait Gamett et d. (1985b) Gelidiella acema BriKil Femira (1968) Gelidiella acemsa * TomSmit Gamett et al. (1985b) Gelidiella frinitafmis Brazil Femira (1968) Gelidiopsis acmapa Tmstrait Gamett et d. (1985b) Gelidlopsis gracilis Brazil Fmeiai ( 1968) Gelidiopsis variabilis Tomes Shait Gmttet d. (1985b) Gelidm comeum Brazil Fmh(1968) Gefidiumsp, Aldabm Atoll Fder(1971) Gelidium sp. Oman Ross (1985) Giganina sp, Peru Brown and Brown (1982) Gmikriacervicornis Bdl Femh(1968) Gmcilaria cmssa Toms Smut Gama et al. (1985b) Gracikda cuneafa Brazil Fmim (1968) Gracilaria cylindrica Nicaragua Mortimer (1981) Gmciluria domingensis Brazil Fmh(I %8) Gmcikriafemx Brazil Femh(1968) Gracilariafofiijem Brazil Fmh( 1968) Gmcilaria mam'ilaris Nicaragua MorTimer(l981) Gmcikria verrucosa Nicaragua MorTimer (1981) Gracikria sp. Nicarapa Mortimez (1981) Gmcilariu sp. Tom Sirait Gameti et aI. (1985b) Gmcilariapsis sjoesfe&ii Bmil F& (1968) Gdfizhsia sp. Tom@ Strait Garnett et d. (1985b) Huhplegma dupemyi Brad1 Ferrem (1 968) Hulymenia jloresia Brazil Fd($968) Halymnia Joresiu Nicaragaa Modmer (19811 Iialymenia sp. Torres Strait Gamett et a1. (1985b) Hetemsiphonia sp. TomShait Game= et d. (1985b) Hypnea cervicornis Brazil Ferrek (1968) Table 16. Continued. Food Location Reference

Hypnea muscifonnis Brazil Ferreira (1968) Hypnea musciformis Nicaragua Mortimer (1981) Hypnea sp. Tones Strait Garnen et al. (1985b) Hypnea sp. Oman Ross (1985) Hypoglossum sp. Torres Strait Garnen et al. (1985b) Jania niponica ogasawara Is. Kurata et al. (1978) Jania sp. Cornoros Frazier (1985) Laurencia brongniarrii Tones Strait Gamen et al. (1985b) Luurencia sp. Brazil Feneira (1968) Luurencia sp. Aldabra Atoll Frazier (1 97 1) Luurencia sp. Tones Strait Gamett et aL (1985b) Lenormandiopsis lorentzii Tones Strait Gamen et al. (1985b) Lenormandiopsis sp. Tones Strait / Gamen et al. (1985b) Leveilla jungermanniouies Tones Strait Gamen et al. (1985b) Liagom setchellii Ogasawara Is. Kurata et al. (1978) Plaiysiphonia sp. Tones Strait Gamen et al. (1985b) Plenosporium pusillum Ogasawara Is. Kurata et al. (1978) Polysiphonia denudata Comoros Fker(1985) Polysiphonia sp. TorresStrait * Gamen et al. (1985b) Prionites obtusa Torres Strait Gamen et al. (1985b) Protokuerzingia schottii Brazil Ferreira (1968) Pterocladia capillacea Hawaii Balazs (1980) Rhodochorton howei Ogasawara Is. Kurata et al. (1978) Rhodopeltis borealis Ogasawara Is. Kurata et al. (1978) Rhodymenia intricate! Ogasawara Is. Kurata et al. (1978) Rhodymenia sp. Peru Brown and Brown (1982) Scinaia sp. Tones Strait Garnen et al. (1985b) Spyridia filamentosa Hawaii Balazs (1980) Spyridia filamentosa Nicaragua Mortimer (1981) Spyridia filamentosa Torres Strait Gamen et al. (1985b) Tolypiocludia glomerulara Torres Strait Gamen et al. (1985b) ' Viliaobtusiloba Brazil Ferreira (1968) Vtdalia obtusiloba Nicaragua Mortimer (1981) Hdalia sp. Torres Strait Garnett et al. (1985b) Helobiae Cymodocea serrulata Yemen Hinh et al. (1973) Cymodocea sp. Torres Strait Gamen et al. (1985b) Cymodocea sp. Aldabra Atoll Fmzier (1971) Cymodocea (or Thalassia) Comoros Frazier (1985) Halodule uninervis Yemen Hinh and Can- (1970) Halodule uninervis Oman Ross (1985) Halodule wrightii Brazil Ferreira (1968) Halodule wriglitii Nicaragua Mortimer (1981) Halophila buillonis Nicaragua Mortimer (1981) Halophila ovalis Tonga Hirth (1971a) Halophila ovalis Oman Ross ( 1985) Halophila spinulosa Torres Strait Garnett et al. (1985b) Syringodiumfiliforme Nicaragua Mortimer (1 98 1) Syringodium isoetifolium Fiji Hirth (1971a) Syringodium isoetifolium Tonga Hirth (1971a) Syringodiutn isoetifolium Yemen Hirth et al. (1973) Thalassia hemprichi Torres Strait Gamen et al. (1985b) Tlialassia restudinurn Nicaragua Mortimer (1981) Thalussodendron ciliutum Comoros Frazier (1 985) No differences in diet were detected between the sexes. Of the animal food identified, sponges were predominant, but apart from shell, no animal material contributed more than 5% to any one stomach. The most common mouth contents of a largely inma- ture feeding population off Queensland, Australia, were in order of frequency of occurrence, the seagrasses Halophila ovalis, Module uninervis, Zostera capricorhi and Halophila spinulosa, and algae, Hypnea cervicomis. Individuals in this population also occasionally were seen feeding on jellyfish, Catostylus mosaicus (Limpus et al. 1994b). The stomach contents of 518 green turtles feeding on algae on the reef around Heron Is, Australia, were retrieved by gastric lavage (Forbes 1994). The sample included juveniles, subadults, adults, females, males and individu- als of undetermined sex. The pooled diets of all turtles contained 38 species of Rhodophyta, 21 species of Chlorophyta and 10 species of Phaeophyta, Animal mat- ter, present in some samples, typically represented less than 1% of the diet volume. Green turtles in the Hawaiian Archipelago eat 56 spe- cies of algae (out of approximately 400 species present in the Archipelago), 1 marine grass, and 9 kinds of inverte- brates. However, nine species of algae are the principal foods (Balazs 1980). Codium and Ulva are the principal foods of juveniles, subadults and adults. Juveniles and subadults have been observed feeding on Physalia, Velella and Janrhina that occasionally drift into the coastal areas. A small, black sponge, Chondmsia chucalla, is also some- times eaten. Hawaiian green turtles, of all sizes, gener- ally bite off only small pieces of algae while foraging. The serrated edges of the beak appear well adapted to this purpose (Balazs 1980). Russell and Balazs (1 994) document how just three years after being introduced into Hawaii (from Florida) the red alga, Hypnea muscifomis, was being eaten by green turtles. Galapagos green turtles feed on at least 30 species of algae including Callithamnion, Gelidium, Gracilaria, Padim and Ulva (Green 1994). Brown and Brown (1982) recorded the stomach con- tents of 39 subadult and adult green turtles in Peru and found, in addition to algae, a significant amount of ani- mal matter (molluscs, polychaetes, jellyfish, amphipods, sardines and anchovies). Mortimer (1 98 1) found that the most important item in the diet of subadult and adult turtles in the feeding pas- tures off Nicaragua is turtle grass, Thalassia testudinum. which accounted for about 79% of the total dry weight of the samples (N=243). Other seagrasses and algae ac- counted for, respectively, 9.7% and 8.2% dry weight of the samples. Red algae made up most of the algae by dry weight, and brown algae the least. No differences were found in food preferences of the two sexes. Animal mat- Mortimer (1982b) and Garnett et al. (1985b) briefly In the sea off Queensland, Limpus (199%) determinec review the literature describing how some traditional turtle that the mean growth rate of 25 adult males was 0.046 hunters can detect differences in the flavor of the meat cdyear. between turtles feeding on seagrasses and those eating It is generally assumed that growth in wild and captive algae, with the seagrass-eating turtles being more tasty. green turtles, of both sexes, is negligible, or sharply re- Samples of a turtle's diet can be retrieved by stomach duced, once sexual maturity is reached (Carr and flushing. Forbes and Limpus (1993) describe a method Goodman 1970; Bjorndal 1980b; LeGaIl et al. 1985; that has been used successfully on green turtles between Limpus 1993b; Wood and Wood 1980,1993a). Statisti- 35 and 118 cm in carapace length. The technique, modi- cally significantcarapace length-weight relationships were fied from previous described methods, can be performed described for adult females and males at several nesting in less than ten minutes and involves use of a pry bar. sites and feeding pastures (Hirth 1982). water-injection tube, retrieval tube and collection bag. Mean growth rate for captive-reared yearlings released and recaptured in the sea around the Cayman Islands was 3.4.3 Growth rate 8.3 dy&, for the 30 - 40 cmsize class (Wood and Wood Many factors affect a green turtle's growth rate, includ- 1993b). In captive green turtles, the logistic growth equa- ing individual physiology, age, sex, diet and gagraphi- tion best describes the growth (Wood and Wood 1993a). cal location of the feeding habitat with its attendant water Bjorndal and Bolten (1995) and Bjorndal et al. (1995) quality and temperature. demonstrated how length-frequency analysis shows prom- As far as is known, there are no data on growth rates of ise @ a method for the study of growth in marine turtles. wild hatchlings, based upon marked and recaptured indi- The method may be especially useful for populations of viduals. Twelve posthatchlings varied considerably in immature turtles, in sea turtle tagging studies with low re- growth rates over 176dayswhen fed satiation rations (trout capture rates, and for work that involves terminal sampling. pellets) in captivity, but individuals had constant specific growth rates (Davenport and Scott 1993a). Growth rate 3.4.4 Metabolism was predominately controlled by efficiency of assimila- Ackennan (1980) found that sea turtle eggs exchange tion of nutrients, rather than by size of appetite or meta- respiratory gases with the surrounding substrate as their bolic level (Davenport and Scott 1993b). metabolic activity increases throughout the incubation Hadjichristophorou and Grove (1983) observed the fee+ period and that growth rate and mortality of the embryos ing behavior of captive one-year old turtles feeding on is related to respiratory gas exchange. The pattern of oxy- floating trout pellets. They found that diets containing gen uptake in eggs over the incubation period was sig- 40-50% protein and 4.2-5 kcal/g were assimulated with moidal (Ackennan 1981 a). Ackennan et al. (1985) also efficiencies of 76% Â 6SD and 86% k 6SD for energy described how the exchanges of respiratory gases, heat and protein nitrogen, respectively. Fourteen hatchling and water between the egg clutch and its surroundings Mexican green turtles raised in captivity for one year and were interactive. They viewed an egg clutch as a very fed a 1: 1 mixture of fresh fish and commercial dry pellet large egg which is much less sensitive to the hydric envi- food (38% protein) attained an average weight of 2,000 g ronment than a single, smaller egg. Booth and Thomp- and a mean curved carapace length of 26 cm (Godinez- son (1991) reviewed and compared the gaseous environ- Dominguez et al. 1993). This rate of growth was greater ment of sea turtle nests with that of other reptiles. than some other captive hatchling growth studies although Baldwin et al. (1989) demonstrated that Heron Island the feeding regimes in the studies did vary. hatchlings utilize anaerobic metabolism during their dig- Growth rates for different size wild green turtles are ging out from the nest, crawling across the beach, and given in Table 17. The data show that growth is slow and while swimming through offshore shallow water. that there is some geographic variability in growth. Davenport et al. (1 982) found that the heart rate of year- Samples from the Bahamas, Florida, Galfoagos, Texas lings in the laboratory during gentle activity was 46 - 48 and the Virgin Islands exhibit a trend toward decreasing beats per minute and that this rose to 64 - 68 beats per growth rate with increasing size. A sample from the Ba- minute during vigorous activity and slowed to 25 - 28 hamas indicates that immature females and males grow beats per minute during a ten-minute dive. at similar rates (Bolten et al. 1992). Smith et al. (1 986) discovered an uncoupling of heart The mean rates of growth of immature Hawaiian turtles, rate and temperature- dependent metabolic requirements at seven sites, ranged from 0.08 to 0.44 cm/month in during cooling of green turtles in thermoregulation ex- straight carapace length (Balazs 1982a). Growth rates at periments. two sites in the main islands were greater than growth During the course of feeding experiments on young rates at five sites in the northwestern segment of the Ar- turtles in the laboratory, Lutz (1990) determined that the chipelago. metabolic rates ranged from 47.9 to 73.8 rnl/kg/h. 30-40 5.3 Florida Mendon* ( 198 1 ) 30-40 8.8 Bahamas Bjomdal and Bolten (1088) 30-40 5.0 Virgin &. Boulon and Frazer (1990) 30-40 5.1 Puerto Rim Collazo et al. (1992) 304 8.9 Texas Shaver (1994) 40-50 0.8 Australia Limpus and Walter (1980) 40-50 4.9 Bahamas Bjomdal and Bolten (1988) 40-50 4.7 Virgin Is. Boulon and Frazer (1990) 40-50 6.0 Pueno Rico Collazo et al. (1992) 40-50 0.4 Galapagos Green (1 993) 50-60 1.0 Australia Lirnpus and Walter (1980) 50-60 3.1 Florida Mendonca (1981) 50-60 3.1 Bahamas Bjomdal and Bolten (1988) 50-60 3.5 Virgin Is. Boulon and Fra2er (1990) 50-60 3.8 Puerto Rim Collazo et al. (1992) 50-60 0.5 Ga-gos Green (1993) 50-60 6.6 Texas Shaver (1994) 60-70 1.4 Australia Limpus and Walter (1980) 60-70 2.8 Florida Mendon* (1981) 60-70 1.8 Bahamas Bjomdal and Bolten (1988) 60-70 1.9 Virgin Is. Boulon and Frazer (1 990) 60-70 3.9 Puerto Rico Collazo et el. (1992) 60-70 0.2 Galfipagos Green (1993) 70-80 1.5 Australia Limpus and Walter (1980) 70-80 2.2 Florida Mendonca (1981) 70-80 1.2 Bahamas Bjomdal and Bolten (1988) 70-80 0.1 G~~~PWOS Green (1993) 80-90 1.1 Australia Mpusand Walter ( 1980) 80-90 0.1 Galipagos Green (1993)

Studying adult females on the Tortuguero nesting beach, of wide temperature variations encountered during mi- Jackson and Prange (1979) found that active metabolism, grations. Kooyman (1 989) compared some aspects of the averaging 0.23 lkg-h, was about ten times the standard green turtle's diving physiology with that of other verte- resting level. Most of the active metabolism is aerobic. brates. The cardiovascular changes associated with in- Berkson (1966) has shown that green turtles are tolerant termittent ventilation at rest and with sustained swimming of anoxia while diving. The respiratory physiology of were determined by West et al. (1992). Prange (1976) diving in sea turtles was reviewed by Lutz and Bendey estimated that adult turtles making the round-trip breed- (1985) and they concluded that most dives appear to be ing migration between Brazil and Ascension Island (about aerobic with the lung serving as the principal oxygen store. 4,600 km) would require the equivalent of about 21% of Gatz et al, (1987) described several cardiopulmonary char- their body weight in fat stores to account for the ener- acteristics in C. mydas adaptive to diving, including a large getic cost of swimming. As already mentioned (section tidal volume relative to functional residual capacity and a 1.3.3) the energy required for migrations may come from concomitant rise of pulmonary blood flow and oxygen sub-carapace depot fat. Butler et al. (1984) determined uptake with temperature. The former trait promotes fast that green turtles can maintain high swimming speeds (at exchange of alveolar gas when the turtle surfaces for least 0.6 m s-1) and metabolize aerobically with little or breathing and the latter aids oxygen transport regardless no resort to anaerobiosis. Six juvenile green turtles off the east coast of Florida, gence from the nest and the hatchlings have the potent fitted with radio and sonic tags and with a PIT tag, were to obtain osmotically free water immediately on enterit at the surface an average of 7.4% and submerged 92.6% the sea by a combination of drinking seawater and st of the time, during a one weekstudy period (Nelson 1994). excretion by the salt gland. Using X-ray rnicroanalysi Working with nine immature turtles, fitted with radio and Marshall (1989) discovered that, in salt glands à sonic transmitters, at a jettied pass in Texas in July, Au- hatchlings, during secretion, intracellular Na+ concentr. gust and September, Renaud et al. (1995) found that 99% don in the principal cells increased while Cl' and K"*"mi of all the turtle submergence times were 40rnin. Brill centrations remained unchanged. The change in Na+ an et al. (1995) used depth-sensitive ultrasonic transmitters the high Cl" concentration suggested similarities with tk to monitor the movements of twelve immature turtles in elasmobranch rectal gland. Kaneohe Bay, Oahu, Hawaii. All turtles remained within Marshall and Saddlier (1989) studied the duct syster a small portion of the Bay where patch reefs and algae of the lachrymal gland and showed that the duct con" were common. Over 90% of the submergence intervals prises central canals, secondary ducts and a sac-like mai were 33 rnin. or less. duct ad it was suggested that the duct system is unlike\ Jackson (1985) reviewed some of the earlier literature to be merely a passive conduit, but that it may have a roll and described how the respiratory system of the green in the modification of the fluid secreted by the gland. turtle is well adapted to meet its diverse requirements, The possible role of a salt gland, along with other at such as vigorous swimming and deep diving and nesting tributes, in the evolution of marine reptiles from estua activities. Two of the special traits of the green turtle's ritae predecessors is discussed by Dunson and Mazzott respiratory system include rapid emptying of the lungs (1989) and Kinneary (1996). and a high capacity for oxygen exchange. In a comparative study of western Atlantic turtles. Internal body temperatures of green turtles on land have Bjorndal (1982b) explained how a Tortuguero-nesting been recorded under a variety of conditions, in a variety green turtle, feeding on seagrasses in the internesting of ways, and the sample sizes have differed, but all the years, channels approximately 10% of its annual energy means have registered between 27.4OC and 33.Z° (Hirth budget into reproduction while a Suriname-nesting turtle, 1962; Mrosvsky and Pritchard 1971; Brattstrom and feeding on algae in the internesting years, allocates about Collins 1972; Whittow and Balazs 1982; Standora et al. 24% of its annual energy budget to reproduction. 1982b; Snell and Fritts 1983). Using biotelemetry, Standora et al. (1982b) found an 3.5 Behavior actively swimming green turtle had an internal body tem- perature (pectoral region) of 37.1 OC in water at 29.1 OC 3.5.1 Migrations and local movements and they found that even when the turtles are inactive their Adult turtles make gametic migrations between their metabolism is sufficient to keep the body temperature as feeding pastures and nesting beaches, and some of these much as 2OC above the water and air temperatures. migrations may encompass thousands of km. Fig. 13 il- Standora et al. (1 982b) posited that heat is produced in lustrate some long-range movements and provide an idea the metabolically active tissues and then is slowly dis- of the total range of at least some members of the popula- tributed to the rest of the body, i.e., the green turtle is a tion. regional endothem. The dispersal of fifty female turtles tagged in western Prange (1985) reviewed some of the pertinent literature Australia is depicted in Fig. 13a. One turtle was recov- on sea turtle post-orbital salt glands - the primary means ered in the Aru Islands. Indonesia (Prince 1993). Two by which marine turtles secrets excess monovalent salts - hundred and seventy three recoveries have been made of and he reported hew the concentrations of the secretions green turtles nesting in eastern Australia (Limpus et al. from the salt glands can be twice that of seawater. 1992). International recoveries encompass New Nicolson and Lutz (1989) demonstrated that the salt Caledonia, Vanuatu, Papua New Guinea and Indonesia. gland secretion of juvenile turtles was free of protein and Four females were tracked by the Argos satellite dur- was mainly composed of chloride and sodium ions, in ing their post-reproductive migration from Pulau Redang, similar relative concentrations to those of sea water, but Malaysia, and three reached their feeding grounds (in that there werealso substantial amounts of potassium and Sabah, Indonesia and Philippines) some 923-1.616 krn magnesium ions and lesser quantities of urea and bicar- distant, in between 27 and 29 days (Luschi et al. 19%). bonate. As other investigators reported, they found that Eleven green turtles tagged in the Sabah Turtle Islands the tear flow from the two eyes frequently did not func- have been retaken in the Philippines and two were recap- tion in synchrony. tured in Indonesia (de Silva 1986). The most distant re- According to Marshall and Cooper (1988). the lachry- covery was from Bakkungan Kecil to Kai Kechil, Indo- mal salt glands of hatchlings are functional upon emer- nesia, a distance of 1,556 km. A female marked in the - 120 130 140 150 160 - Solomon b Islands

Vanuatu a

New Caledonia N= 15

d 150 170 170 150

Fig. 13. Long distance recoveries of green turtles. Arrows indicate spread from the nesting beach, or tagging area, and are not intended to suggest routes. a-the tagging sites are North West Cape and Barrow Island (solid lines) and 1 Islands (dashed lines) in western Australia (schematic adapted from Prince 1993); in eastern Australia, tagging sites are Raine Island-Pandora Cay Group (solid lines) and the Capricorn-~unkerGroup (dashed lines) (sche- matic adapted from Bustard 1976, Limpus and Parmenter 1986, and Limpus et al. 1992). bthe tagging locale is Scilly Atoll in French Polynesia and the three most distant recovery sites are indicated. Other, less distant recoveries, are identified in the text (schematic adapted from Hirth 1993).

"-^ Fig, 13, Long distance recoveries of green turtles. Arrows indicate spread from the nesting beach, or tagging area, and are not intended to suggest routes. c-the nesting site is French Frigate Shoals, in the middle of the Hawaiian Archi- pelago (schematic adapted from Balazs 1980). &tagging sites are eleven nesting and foraging grounds in the Galapagos Islands (schematic adapted from Green 1984). e-tagging sites are Colola and Maruata, Mexico (schematic adapted from Alvarado and Figueroa 1992) f-tagging site is Tortuguem, Costa Rica (schematic adapted from Can- 1984). g- tagging beaches are in eastern Suriname and western French Guiana (schematic adapted from Pritchard 1976). h- tagging beach is Aves Island (schematic adapted from Bainbridge 199 1 and Sol6 1994). i-tagging beach isAscension Island (schematic adapted from Can- 1984). j-tagging sites are Ras al Hadd and Masirah Island. Oman (solid lines) and Musa and Shanna beaches,Yernen (dashed lines) (schematic adapted from, respectively, Ross 1987; and Hirth and Can, 1970, FAO, 1973, and updated by Hirth). k-tagging sites are Tromelin Island (solid lines) and Europa Island (dashed lines) (schematic adapted from Hughes 1982 and Le Gall and Hughes 1987).

,

Sarawak Turtle Islands was found 800 km away in North tag recoveries demonstrate that the breeders at Frenc Borneo (Hdsson 198)). Five turtles nesting on Long Frigate Shoals, in the middle of the Archipelago, are re Island, Papua New Guinea, were taken later in Man Jaya, cruited from both ends of the island chain. The longes Indonesia (Spring 1983). distance between recovery points was about 1,100 kr A nester tagged on Gielop, Yap, Federated States of (Balm 1980). Micronesia, on 11 May 1991 was incidentally caught in Twenty-three turtles have been retaken from a total o fishing gear in Langob. Philippines on 17 January 1992. 5,844 individuals tagged in the Galapagos Islands (Greel Yap and Langob are separated by about 1,472tan- Green 1984). Recovery sites include Peru, Ecuador, Colombia turtles tagged in the Philippine Turtle Islands have been Panama and Costa Rica (Fig. 13d). The approximate recaptured in the Sabah Turtle Islands and vice versa maximum distance was 2,163 km, from Bahia Barahona (Rannirez de Veyra 1994a). Isabela Island to San Andres, Peru. Four females tagged while nesting on Gielop Island, Of 5,176 green turtles tagged at Colola and Mapata Yap State, Federated States of Micronesia, and one male Mexicoflere have been 47 recoveries more than 100 tar tagged while mating off Gielop Is., were subsequently from the nesting beaches (Fig. 13e). International recov- recovered in the Philippines to the west, between approxi- eries are from El Salvador, Guatemala, Nicaragua, Costa mately 1,550 and 1,950 krn distant from the tagging site, Rica and Colombia (Alvarado and Figueroa 1992). Five and from 139 to 530 days after tagging. However, it is nesters at Colola, Mexico, were fitted with transmitters noteworthy that one individual tagged while nesting on and tracked via the Tiros-Argos satellite system in 1991. Gielop Island was recovered at Majuro Island in the ~esultsrevealed that at least some of the turtles swam Marshall Islands, about 3,410 km to the east, after less hundreds of kilometers away from land in very deep wa- than 239 days at large. Two individuals marked while ter (Bytes et al. 1995). nesting on Ngulu Atoll, Yap State, were recovered in the Most of the Tortuguero, Costa Rica international tag Philippines, from 1,690 to 2.020 km distant, between 84 recoveries (N=1,005) have come from Nicaragua, espe- and less than 217 days after tagging. Of two females cially from the Miskito Cays region, where extensive tagged on Elato Atoll, Yap State, one was retaken in the seagrass pastures are located. Numerous recoveries have Philippines, 2,760 km distant, after 384 days at sea; and, also been made in Mexico, Panama, Colombia and Ven- the other was retaken wh3e feeding off Kavieng, Papua ezuela (Fig. 130. One important result of the Tortuguero New Guinea, about 1,270 km southeast of the tagging tagging work, now in its fortieth year and with over 55,000 site, and after 171 days at sea (Kolinski 1995). turtles tagged (Anon 1991b), is that turtles tagged at the Turtles tagged in the Ogasawara Islands have been re- Tortuguero nesting beach have not been seen renesting covered at various sites in the Japanese Archipelago anywhere else than at Tortuguero. (Kurata et al. 1978). After two decades of tagging nest- Long-distance recoveries of females tagged on their ers in the Ogasawara Islands it is concluded that these nesting beaches have ranged between 0.4% (Galapagos) females are migrating to feeding grounds off the main and 9.3% (Tortuguero) (Alvarado and Figueroa 1992). islands of Japan (Tachikawa et al. 1994). One nester seen Of 91 green turtles retaken after being tagged on nest- on Oroluk Atoll in the Caroline Islands on 2 June 1986 ing beaches in eastern Suriname and western French was seen in Taiwan on 18 April 1987 (Edson and Curren Guiana, all but one were from Brazil (Pig. 13g). Sixty- 1987). two were taken off the coastal state of Cearil (Pritchard Turtles tagged on Scilly Atoll in French Polynesia (Fig. 1976). These turtles mingle with tunles from Ascension 13b) have been recovered as far west as New Caledonia, Island on the Brazilian algal feeding pastures. Vanuatu and the Solomon Islands, distances of about 4,000 Aves Island, in the eastern Caribbean Sea, is only about krn. Other recapture sites have included Tonga, Fiji, Wallis 500 m in length, about 120 m wide at the widest point, and Futuna and the Cook Islands (Hirth 1993). Tuato'o- and approximately 3.3 m at its highest elevation. More Hartley et al. (1993) reported that two green turtles tagged than 4,000 green turtles have been tagged at Aves Island on Rose Atoll in American Samoa were recovered in Fiji, and 43 have been recaptured (Fig. 13h) (Sol6 1994). One also a westward movement. Three turtles on Rose Atoll tagged turtle was recorded nesting on Mona Island, Puerto were fitted with satellite transmitters in November 1993 Rico; the rest were recaptured at sea. Recaptures have and were recovered 1,600 km to the west in Fiji between been made from 12 islands in the Caribbean Sea and 5 34 and 45 days later (Craig 1994). However, one nesting continental nations bordering the Caribbean Sea and At- green turtle fitted with a satellite transmitter swam from lantic Ocean. Most of the recaptures (37.2%) were off Rose Atoll to the vicinity of Tahiti, a southeast direction, the coasts of Nicaragua and the Dominican Republic. The over a period of 36 days (Craig and Balazs 1995). most distant recapture was made in Maranhao, Brazil, Long distance recoveries of 52 female and male turtles about 2,870 tan from Aves Island. This turtle swam an in the Hawaiian Archipelago are shown in Fig. 13c. The average of 23 km daily over a period of 125 days. About 3,384 green turtles have been tagged on Ascen- and Dutton 1995). Several recent, long-range movements sion Island and there have been 66 recoveries, aB from have been recorded: Guseman and Ehrhart (1 992) report the coast of Brazil'CMortirner and Can- 1987). about 2,300 that two juveniles were tagged in Mosquito Lagoon, krn distant (Fig. 139. The shortest recovery interval was Florida, and one was recovered in Cuba and the other in 56 days. Nicaragua; and, Hirth et al. (1992) reported how one Six green turtles tagged at Ras a1 Hadd and at Masirah immature was tagged on Wuvulu Island, Papua New Island, Oman, have been recovered in Saudi Arabia, Guinea on 1 July 1989 and was retaken on 1 October, United Arab Emirates, Yemen, Eritrea and Somalia (Ross 1989, in northeastern Irian Jaya, Indonesia. In the latter 1987). Nine green turtles tagged on Musa and Shma example, the shortest straight line distance between points beaches, Yemen, have been retaken off the east coast of of contact was 305 km. Head-started green turtles, re- Somalia, and two more were captured on the feeding pas- leased in Florida, have been recorded along the Atlantic tures within Yemen (Hirth and Can- 1970; FA0 1973; coast of the U.S. as far north as New York, the eastern Hirth, updated) (Fig. 13j). Atlantic Ocean (Azores, Madeiria and Mauritania) and Out of 4,843 and 3,766 females tagged on Europa and South America (Colombia, Venezuela, Guyana, Brazil) Tromelin Islands there have been, respectively, 15 and 12 (Witham 1991). Of 399 captive-reared turtles that were international recoveries (Fig. 13k). Maximum distance released when 2 to 3 years of age from Puerto Morelos, between site of tagging (Europa) and site of recapture Mexico, 4 were recaptured in Cuban waters when they (Maurice) was about 2,250 km (LeGall and Hughes 1987). were between 7 and 9 years of age (Zurita et al. 1994). A Some of the green turtles taken by fishermen around the fifth turtle from this captive-reared cohort was retaken in Toliara (Madagascar) coral reef feeding grounds may be Mexico, about 45 km from the site of release, when it migrants from Europa Island (Rakotonirina and Cooke was 11 years old. It was placed in a pen, extending from 1994). the beach out 50m into the sea, and it nested four times Two turtles tagged on Karan Island, Saudi Arabia, were on the beach. recovered in Kuwait, approximately 250 km away (Miller Green (1993) cites several older papers describing long- 1989). distance movements of wild- and captive-reared juvenile One green turtle tagged in Cyprus was recaptured in green turtles of from 2,300 to 5,600 km. the Gulf of Gab6 in Tunisia (Laurent et al; 1990). A green turtle's migrations are an important and inte- Minimum average swimming speeds for migrating gral part of its life history strategy. The vast majority of green turtles have ranged between about 20 and 90 km/ migrating, adult, female green turtles exhibit philopatry day (Meylan 1982b). (regional homing) and strong nest site fixation in their Information on migrations of males is accumulating renesting episodes. However, the short- and long-range slowly. In the Hawaiian Archipelago, nine males have orienting and navigating mechanisms remain largely un- been recorded migrating between French Frigate Shoals known. The genetic, hormonal and environmental van- and Pearl and Hermes Reef, Oahu and Lisianski (Balazs ables influencing migration are just beginning to be un- 1983a). Green (1 984) reported that two males tagged in raveled. Furthermore, there must be some individual the Galapagos Islands were recaptured in Peru and an- behavioral plasticity in response to changing enviromen- other marked in the Galapagos was retaken in Costa Rica. tal conditions affecting oriented travel and more infor- The longest distance traveled, to Peru, was 2,150 tan. Two mation at this individual level is sorely needed. Telemet- males tagged on Scilly Atoll in French Polynesia were ric monitoring of individuals will certainly help in this recaptured in Fiji, almost 3,000km distant (Galenon 1979 regard. in Green 1984). Twenty-five breeding males tagged in Adult and hatchling green turtles probably have simi- the Capricorn and Bunker Groups of the southern Great lar sensory modalities and some of the behavioral research Barrier Reef were recaptured at various feeding areas: 21 with hatchlings, especially their seaward orientation in within the Capricorn Group. 3 elsewhere in Queensland, shallow water off the nesting beach, and their magnetic and 1 in New Caledonia. These recapture sites ranged detection abilities, may be applicable to the long-distance from 40km to 1,443 km from the tagging area (Limpus migrations of adults (see section 3.2.2). It is possible that 1993b). adults use wave propagation and a geomagnetic sense, Some information is available on the local and long- along with other cues, in their long-range oceanic navi- range movements of immature green turtles: In some lo- gation. calities, at least some immature individuals have a ten- Carr (1 965) considered celestial navigation as a pos- dency to remain in the vicinity of, or to return to, their sible guidance mechanism in the long-range travels ofthe foraging area (Schmidt 1916; Can and Caldwkll 1956; green turtle, but this hypothesis was untested after Ireland 1980; Balazs 1982a; Williams 1988; Manzella et Ehrenfeld and Koch (1967) found theAtlantic green turtle al. 1990; Hirth et al. 1992; Renaud et al. 1994; McDonald to be extremely myopic when its eyes are out of water.

69 However, it has now been shown that the eyes of Chelo- variety of cues, over a period of weeks, as they move away nia mydas are approximately emnieiropic in air and this from their natal beach. would allow them to view celestial objects with greater Using operant conditioning techniques, Manton et al. clarity than was previously thought possible (Northmore (1972% b) estabfished that small green turtles are capable and Granda 1991). of underwater chemoreception and they discuss use of Carr and Coleman (1 974) postulated that the Ascen- this ability for navigation purposes.Owens et al. (1986) sion Island - Brazil migration may have evolved gradu- reviewed some of the pertinent literature and research on ally over 70,'million years as the South Atlantic Ocean chemoreception in amphibians and reptiles and concluded widened via plate tectonic movements. But, based on that sea turtles can orient to specific chemical cues learned mtDNA analysis,Bowen et al. (1989) stated that the colo- early in their lives. In a laboratory setting, Grassman and nization of Ascension has been evolutionarily recent. Owens (1987) found that the chemosensory environment LeGall (1989) elaborated on the possible link between of nestling and hatchling green turtles affected their sub- plate tectonics and other long-range green turtle migra- sequent behavior, and more recently, Grassman (1993) tions. The widening Gulf of Aden was mentioned as a has provi

4. POPULATION 4.2.1 Average abundance and density 4.1 Structure The nflmbers of nesters on some of the well-known nest- ing beaches are given in section 1.2.2 and in Figs. 9, 10 4.1.1 Sex ratio and 1 1. Natural fluctuations in numbers of nesters have The sex ratios of hatchlings on somebeaches have been been recorded on several beaches (see following section). discussed in section 3.2.2, and as pointed out in that sec- Without historical records, it is impossible to determine tion. a number of temporal and ecological factors can af- if Anall nesting populations are inherently small, or if they fect natural, hatchling sex ratios. are remnants of a once larger nesting population, or if The sex ratio of a largely immature population (N=197) they are incipient colonizers. King (1982) has reviewed inhabiting the feeding grounds around Heron Reef was the decline and extirpation of green turtles in several lo- equivalent to 1:1 (Limpus and Reed 1985a). Of 784 indi- cales, especially in places where commercial exploitation viduals, mostly immature, sampled on the Moreton Banks, replaced subsistence take. As far as is known, no extir- Queensland, Australia feeding area, 65.6% were females pated nesting colony has been, or is being, recolonized, (Limpus et al. 1994b). The sex ratio of 56 immature turtles As Avise and Bowen (1994) point out in their discussion on the Bermuda foraging grounds was not significantly of the mtDNA research, green turtle nesting sites tend to different from 1: 1 (Meylan et al. 1992a). Of 120 imma- be strongly isolated from one another over ecological ture turtles sampled on a feeding ground in the Bahamas, timescales because of the propensity for natal homing by 46 were males. 65 were females and the sex of 9 was females (but weakly differentiated over evolutionary undetermined. The sex ratio is not statistically different timescales because of mistakes in natal homing), so that from 1: 1 (Bolten et al. 1992). Sixty six iminature turtles decline or loss of a specific nesting site is not likely to be that died during a cold-stunning episode in a developmen- compensated by natural recruitment of females hatched tal habitat in east central Florida were necropsied. Of these, elsewhere-at least over tirnescales germane to human 42 and 24 were females and males respectively-a signifi- interests. cantly female biased ratio (Schroeder and Owens 1994). Aerial surveys are sometimes used to obtain crude The sex ratio of a pooled sample (N=66) of immature estimates of mating aggregations or nesting activity and offer Hawaiian turtles caught in their feeding habitats did not the advantages of checking beaches on isolated atolls and differ significantly from 1: 1 (Wibbels et al. 1993). can cover much territory in a short time. But such estimates On the feeding pastures in Oman, Ross (1984) deter- always need to be verified by ground-truth observations mined a 1: 1 adult sex ratio (N=242). However, in com- preferably over a number of consecutive days. Hirth and mercial catches of mostly subadult and adult turtles else- Ogren (1987) have described how the visible aspects of a where, females have usually outnumbered males: Nica- sea turtle nest can change significantly over time. ragua (Carr and Giovannoli 1957; Mortimer 198l), Baja Long range, demographic studies on nesting sites are California (Caldwell 1962b), Yemen (Hirth and Carr 1970) important and can provide, among other things, informa- and South Africa and St. Brandon (Hughes 1974a). tion on aging, phenotypic plasticity, remigrations, nest site The optimal male: female sex ratio in the breeding herd selection and population cycles. in the Cayman Turtle Farm is about 1:4 (Wood 199 1). 4.2.2 Changes in abundance and density 4.1.2 Age composition The density of nesting can fluctuate dramatically from No complete age-sex pyramid has been constructed for year to year on some beaches. Limpus (1980) reported a any green turtle population. The epipelagic phase of the nesting population of about 1,100 on Heron Island in 1974- green turtle, and hence the turtle's age, has been estimated 75 and then about 50 the following nesting season. On to last for one to five years (see section 2.2.1). Estimates Raine Island in the peak nesting season of 1974-75 over 11,000 nesting turtles were ashore on one night on the cycle and 12.2% migrated at intervals from 5 to 8 years, 1.7 krn long beach yet the following year only about 100 One female nested after an interval of 1 year. The aver- turtles nested on Raine nightly (Limpus 1982.a). age remigration interval of males in the Capricorn and At Tortuguero, it was estimated that 3 1,211 green turtles Bunker Groups of the southern Great Barrier Reef is 2.08 nested in 1978,5,178 in 1979,52,046 in 1980 and 8,430 years (range 1-5, N s: 24). This is shorter than that re- in 1981 (Can- et al. 1982). Over a fifteen year period corded for females on Heron Island where the mean (1971-1985) on the regularly censused 8 km study beach remigration cycle is 4.65 years (range 2-7, = 31) at Tortuguero, the number of nesters ranged from 413 in (Limpus 1993b). Males, in the GaMpagos Is. most com- 1979 to 3,022 in 1980 (Anon 1987). monly remigrate annually while females here exhibit a Schulz (1982) cited some significant fluctuations in the predominately three year cycle (Green 1994). number of nests on Suriname beaches: there were 3,6 10 nests in 1975 and 8,080 in 1976. 4.3.2 Factors affecting reproduction The nesting density at Ras al Hadd changes from year Many factors affect reproduction and they have been to year (Ross 1987). During the August through hem- discussed in specific sections. Some differences in hatch- her peak of nesting in 1983 about 340 turtles nested ing success of eggs and temperature dependent sex deter- nightly. During the same peak months in 1984,1985 and mination are described in sections 3.1.7 and 3.2.2, respec- 1986, about 185,750and 740, respectively, nested nightly. tively. When nests are constructed in the shade of condo- Le Gall et al. (1986) recorded fluctuations in the nesting miniums the natural sex ratios may be altered, as population on Europa Island. Estimations of nesters range Mrosovsky et al. (1994) surmised for loggerheads on a between 2,000 and 11,000 during the peak nesting months. Florida beach. A male's fitness, at least during one re- In Michoacan. Alvarado and Figueroa (1990) estimated productive season, may be determined by whether he suc- that the numbers of females nesting annually from 1981 ceeds in copulation or is relegated to a role of "escort" through 1989 were approximately: 5,586; 4,483; 1,000; (section 3.1.3). Natural sex ratios in some populations 940; 1,200, 3,334; 1.993; 570 and 1,300.

4.3 Natality and Recruitment of turtles in these areas could decimate multiple nes 4.3.1 Reproduction rates .. populations simultaneously (discussed by Avise Green turtles are characterized by slow growth, delayed Bowen 1994). sexual maturity, high fecundity, itemparity, relatively high predation rates on eggs and hatchlings and a relatively long reproductive life. All of these parameters affect their reproductive rates. Numerous data concerning reproductive strategies are given in Tables 4,7,9, and 1 1. Using these data one can calculate the number of eggs and hatchlings that a nester, and nesting population, can produce over one or many nesting seasons. There are a lack of data on the repro- ductive lifespan of green turtles, but Frazer (1 983) calcu- lated a maximum reproductive life span of 32 years for female loggerheads in the Georgia population. nesting beach are flooded by a fluctuating ground Can- and Can- ( l97Oa) have shown how a few Tortuguero table. turtles have shifted from a 3 to a 2 year remigration cycle Fowler (1979) determined that beach erosion des and vice versa, and they postulated that this may be a reflection of feeding regimes. In contrast to the abundant data on female remigration intervals (Table 9). little is known about the migratory on Krofajapasi Beach, Suriname, in 1982, were laid cycles of males. Research at three localities does indi- low the spring high tide level, and were conside cate that males have shorter remigration intervals than "doomed". females at the same sites. Balazs (1983a) found that of Agardy (1990) reported how Hurricane Hugo i 52 recaptures of males at French Frigate Shoals, 42.3% Caribbean region in 1989 caused some erosion on returned after 1 year, 32.7% after 2 years, and 25% after 3 years. Of 130 females at the same site, 36.2% exhib- ited a 2 year cycle, 43.'1% a 3 year cycle, 7.7% a 4 year clutches in an egg hatchery in Mexico. As pointed out by Milton et al. (1994) hurricanes can adversely affect sea Tortuguero Beach, Costa Rica, in 1990. They found that turtle populations by, among other things, washing-out the presence and behavior of tourists resulted in distur- nests completely, flooding nests and suffocatine/drown- bance of nesters. Tourists visitation was concentrated on ing eggs and hatchlings, removing some sand from tops weekends, correlating with the times that one third fewer of nests, depositing extra sand over nests, mixing and turtles came to the, beach. A pilot (raining course and changing beach particle size. and depositing debris on the guide program involving Tortuguero residents was suc- nesting beach. Radically altered beach topography and cessful in helping to mitigate the impact of tourists by beach chemistry may affect the natal beach homing hy- controlling the number of people on the nesting beach at pothesis since the altered beach may have little resem- night and by controlling the use of flashlights and flash blance to the female's natal beach. photography (Jacobson and Robles 1992). Limpus (1993a) very briefly mentions how climate Witham (1982) discussed the affect of human activities change and sea level rise in the South Padfic region could on sea tupe nesting beaches, including impact from arti- affect sex ratios of incubating eggs, erode some existing ficial lights, physical barriers such as sea walls, groins nesting sites and affect the frequency of ENSO episodes and jetties, and vehicular traffic, with an emphasis on which, in turn, could affect green turtle breeding cycles. Florida beaches. Coston-Clements and Hoss (1983) re- Daniels et al. (1 993) have discussed how rising sea lev- viewed some of the literature on the impact of humans on els, due to the greenhouse effect, could reduce logger- sea turtle beaches and cite several papers dealing with head turtle nesting habitat in South Carolina. Their meth- oth*er species of turtles but whose conclusions are appli- odologies and scenarios are applicable to green turtle nest- cable to green turtles. ing on other coasts. Compaction of sand by human and vehicular traffic and It goes without saying, that mining beach sand can have beach nourishment may act to impair natural gas ex- devastating affects on green turtle nesting. Sella (1982) change, which Ackerman (1980) found could lower the described such a situation in Israel. effectivenessof incubation. Nesting green turtles avoid artificial lights at Tortuguero Use of vehicles on beaches by sea turtle researchers (Can and Can- 1972; Witherington 1992) and on Ascen- should be discouraged. Of all people, turtle researchers sion Island (Mortimer 1982a). Witherington (1992) has should walk gently on the beach, Transporting tourists in shown, where beach lighting is essential, that the use of vehicles along the beach is also discouraged. yellow, low pressure sodium vapor luminaires has no sig- Crain et al. (1 995) briefly review some of the sea turtle nificant effect on nesting but white, mercury vapor lumi- factors which should be considered in beach nourishment naires significantly reduces the number of green turtles schemes including escarpments which can impede the emerging and nesting. The sodium vapor luminaires emit crawl of nesters onto the beach; beach compaction which light near the peak of spectral sensitivity for green turtles can alter egg chamber architecture; and, changes in the (Granda and O'Shea 1972) and the light may represent gaseous, hydric and thermal microclimate of the nesting an inane color to nesters. arena which may affect hatchling embryology and survi- It is well known that reckless use of lights [e. g. flash- vorship. lights, flash photography) and excessive human move- Rirnkus and Ackerman (1995) assessed the impact of ments and barking dogs can cause emerging green turtles beach renourishment on the hydric microclimate of nest- to return to the sea. Green turtles are especially sensitive ing beaches on the east coast of Florida and found that to lights and movements up until oviposition. While some renourished beaches are wetter than natural beaches; that human interference (tagging, measuring. checking egg-water exchange due to water potential differences is epibionts, counting eggs, photographing, etc.) may not probably not affected; and, that the thermal conductivity cause a green turtle to abandon its nest after oviposition of renourished beaches is likely to be higher than natural has begun, it has been shown that some harassed turtles beaches and that this may influence the nest/egg water will complete the nest covering and camouflaging pro- exchange. cess in a hurried or desultory manner (Hirth and Samson Oolitic aragonite sand from the Bahama Islands is un- 1987; Campbell 1994). Such less-than-normal nests may der consideration as a source of fill for some Florida beach be subject to more predation and clutches may suffer from nourishment projects even though Florida beaches are abnormal temperature, moisture and gas exchange re- composed primarily of silicate sand (Shaw et al. 1995b). gimes. It is recommended that turtles be allowed to com- Aragonite sand was about 2Tcooler than Florida sili- plete their entire nesting repertoire under as natural con- cate on a loggerhead beach in Florida and extended the ditions as possible. Ecotourists should cooperate with incubation period by five days and possibly altered natu- guides when they are informed of these matters. ral loggerhead sex ratios. The same may be true for Florida Jacobson and Lopez (1994) conducted a study on the green turtle clutches in aragonite sand. effect of tourists on green turtle nesting activity on Ryder (1995) found that loggerhead nesting and hatchling success on a renourished beach and a control and diseases including tuberculosis, pneumonia, cocci&- beach on the east coast of Florida were similar although osis and fibropapillomatosis (section 3.3.5). Long dis- there were differences in compaction and sand tempera- tance migrations may be more hazardous than short mi- tures between the two beaches. These findings may be grations although data are lacking on this subject. Green applicable to green turtles which nest in the same area. turtles, and other species of marine turtles, may be acci- Large or heavy litter that has been washed up on the dentally speared by billfishes (Frazier et al. 1994). Colli- beach over a clutch of eggs can interfere with hatchling sions with coral and rock, especially at the entrance to emergence. Sharp objects can cut flippers. Weathering nesting beaches, sometimes result in injuries. of tar balls and human debris on the beach may affect the Factors that reduce the green turtles' ability to survive nest-site selection process. Litter of all kinds on the beach need to be studied in a quantitative fashion. Such studies impedes the crawl of hatchlings to the sea (Hirth 1987). should include biomgnification of pollutants in the food Anthmpogenic debris cm some beaches of Karan Island chain, synergistic effects of environmental pollutants and interferes with turtles nesting there (Miller 1989). chemical pollution in the algae/seagrass feeding pastures. For example, Abdellatif (1993) identified pesticide run- 4.3.3 Recruitment off (from locust spraying on the coastal plain) as prob- Recruitment is the influx of new members into a popu- ably the most serious threat to the Red Sea environment lation by reproduction or immigration. Crude hatchling off the Sudanese coast. recruitment rates can be computed for certain populations During a six-week study on Europa Island, it was ob- by referring to Tables 4,7 and 1 1. Recruitment rates for served that 50 green turtles perished by falling into reef juvenile and subadult size classes are unknown. Recruit- crevices on their return to the sea after nesting (Hughes ment into nesting populations is better known. For ex- 1974b). Heat exhaustion killed many turtles on Raine ample, at Tortuguero, from 1969 to 1974. recruits ac- Island when, due to the high numbers of nesters, some counted for 80.4 to 90.0% of the total number of turtles individuals resorted to nesting in the daytime (Low 1985). seen each year. From 1975 to 1978, recruits comprised In 1984, the storm surge of Cyclone Kathy stranded over 62.5 to 80.7% of the turtles seen. Whether this change 1,000 green turtles in Australia. It is estimated that more reflected a decrease in recruitment or an increase in adult than 500 would have died on the mudflats without human survivorship was not clear (Bjorndal 1980b). Reriiigrants aid (Limpus and Reed 1985b). accounted for 15.8% of the nesting populaiion at Green tunics are impacted by humans, directly or indi- Tonuguero in 1968 and 8.7% of the population in 1969 rectly, in all of their critical habitats: on the nesting beach (Can- and Can 1970b). and in the internesting habitat; in the epipelagic habitat; Relevant survivorship data are in section 4.5. on the developmental and adult foraging habitats; and, in the migrating routes between these habitats. Some of the 4.4 Mortality marine impacts are complicated by the fact that mitiga- tion will require international cooperation. 4.4.1 Mortality rates It was found that green turtle eggs collected on Ascen- Bjomdal(1980b) estimated instantaneous death rates of sion Island contained DDE and PCB residues (Thomp- between 0.2929 and 0.5538 for fourteen cohorts (1959-1972) son et al. 1974). Low levels of DDE and DDT were de- of adult females at Tortuguero. The estimates were made tected in green turtle eggs in Florida (dark and Krynitsky during a time period when the colony was heavily exploited. 1980). McKim and Johnson (1983) analyzed polychlori- See section 4.5 for discussion of survivorship rates. nated residues in the muscle and liver of juvenile gmn turtles collected on the east coast of Florida, finding only 4.4.2 Factors causing or affecting mortality low concentrations. In the muscle and liver, DDE resi- There are many factors affecting mortality and they have dues were less than 1 ppb and less than 10 ppb, respec- been referred to in specific sections. For example, growth tively. Total PCB levels ranged from 5.9-9.4 ppb in muscle rate and mortality of embryos is related to respiratory gas and from 43-80 ppb in the liver. exchange (section 3.4.4). Eggs are taken by a number of Green turtle hatchlings are disoriented by photopollution predators with various species of crabs and mammals tak- on the hatching leach (literature reviewed in Verheijen ing significant numbers. Major predators on hatchlings 1985). Raymond (1984) reviewed the problem of include crabs, fishes and birds. Sharks prey upon large hatchlings' disorientation on beaches due to a variety of individuals (section 3.3.4). Survivorship may be related man-made lights, including building lights, streetlights, to growth rates in the sense that faster growing individu- vehicular lights and flashlights. He recommended a num- als are subject to less predation (section 3.4.3). Hypoth- ber of solutions ranging from eliminating the problem ennic stunning episodes have been reported (section lights to preventing direct lights on the beach to reducing 3.3.2). Green turtles are host to a number of parasites the intensity of the lights. Chan and Liew (1988) reviewed some of the literature Quantitative studies are needed on the interactions of on the effects of oil pollution on sea turtles. They quoted humans and green turtles in the internesting habitat. Stud- some reports showing that fresh oil can induce embry- ies which come to mind include the numbers of turtles onic mortality, that hatchlings associated with pelagic drift taken by harpoons and nets while copulating offshore; lines are extremely vulnerable to the effects of oil and the number of collisions with, and injuries caused by, that dennopathologic changes were seen in subadults ex- boats; and the degree'of disruption of internesting move- posed to oil. Several green turtle deaths in the Gulf of ments caused by jetties, water pollution and divers. ~exicowere associated with an oil spill (Shabica 1982). The ingestion of tar balls and plastics by green turtles, Berger (1991) examined the potential environmental im- their entanglement in oceanic debris, and the effects of pact of offshore oil spills in the vicinity of Palawan. Phil- these factors on morbidity and mortality have been dis- ippines. The oil spill trajectories would be dependent upon cussed by Balazs (1985b). Turtles may be attracted to spill location and time of the year. Depending on these debris because it resembles natural food in size or shape circumstances the green and hawksbill nesting sites on or color, orfthe debris itself may be appealing to turtles. Palawan and the Calamian Islands would be at risk. The debris may be ingested incidentally with natural food Based upon gas chromatographic analysis of oil resi- items. The stage of appetite must also influence feeding dues scraped from four species of marine turtles, includ- behavior. The effects of ingested debris may be related to ing green turtles, stranded in the Gulf of Mexico region, the amount and kind ingested and possibly could range Van Vleet and Pauly (1987) concluded that the turtles were from mechanical blockage of the gut to subtle interfer- impacted by oil originating from tanker discharge. ence with the turtle's metabolism. Entanglement can in- Miller (1989) discussed the direct and indirect affects terfere with all aspects of a turtle's behavior. Bjomdal et of oil spills on turtles in the Gulf of Arabia. He said that al. (1994) noted that anthropogenic debris was found in the effect of the NOWRUZ oil spill (1983) was probably the digestive tracts of 24 of 43 juvenile green turtle car- severe-contributing to the death of numerous turtles and casses that washed ashore in Florida. The death of two impacting seagrass foraging areas and nesting teaches. turtles was attributed to ingestion of debris. Plastics were According to Miller (1989) the major threats to the sur- the most commonly ingested debris and fishing line, fish vival of green turtles in the Arabian Gulf are oil pollu- hooks, rubber, aluminum foil and tar were also swallowed. tion, habitat destruction and to a lesser extent fishing. The Thirteen of 784 turtles in a feeding population off 1991 Gulf War oil spill severely impacted the sandy Queensland,Australia, displayed evidence of having been beaches on Karan and Jana Islands. On Karan contrac- impacted by anthropogenic activities (e. g. entanglement tors removed 14,000 m3 of tar and oil sediment from the in fishing line or rope; boat or propeller injuries) (Limpus sandy beaches. Clean marine sand was used to re-estab- et a1 . 1994b). lish the original configuration of the beaches. Marine The Center for Environmental Education (1987) has turtles continue to nest in apparently normal numbers. In published an illuminating report on the types, sources and 1991,164 hawksbills and about 1,100 green turtles nested impacts of nondegradable plastics in the marine ecosys- on Karan and Jana Islands. In 1992, 150 hawksbills and tem, with emphasis on U. 5. waters. 700 greens nested on the two islands (Krupp and Jones Can (1987b) described how the post-hatchlings and 1993). The five coral islands of Harqus, Karan, Kurayn, young turtles may eat plastic scraps and other buoyant Jana and Jurayd form part of a proposed marine sanctu- debris as they occupy the same driftlines in the epipe- ary (Krupp and Jones 1993). Three species of sea grasses lagic habitat. Plotkin and Amoss (1990) found that 7 of (Halodule uninervis, Halophila ovalis and ff. stipulacea) 15 green turtles stranded on the south Texas coast from in the northwestern Gulf have not suffered acute or 1986 through 1988 had ingested marine debris. Uchida longterm degradation as a result of the 1991 Gulf War oil (1990) concluded that most of the sea turtles found in spill (Kenworthy et al. 1993; Durako et al. 1993). waters adjacent to Japan show evidence of plastic inges- Mosier (1994) describes how use of GIs (Geographi- tion. cal Information System- a system of hardware and soft- In laboratory experiments, Lutz (1990) stated that low ware which is designed to analyze spatially referenced levels of plastic ingestion had no significant effect on gut data) can aid in such marine turtle research as the rela- function, metabolic rate, blood chemistry, liver function tionships between oil spills and sea grass beds and nest or salt balance in young green turtles. Further studies sites. GIs can also help relate turtle strandings and geo- with latex indicated the sojourn of the latex material in graphic data to spatial patterns of turtle mortality. the gut ranged from a few days to four months. Later, There is circumstantial evidence to suggest that green studying four, captive immature turtles, Schulman and turtles, like other sea turtles, can suffer from underwater Lutz (1995) concluded that "Even small amounts of plas- explosions used to remove oil platforms (Klima et al. tic may remain in the gut for months, causing a distur- 1988). bance in gut function, lipid metabolism and resulting in excessive gas accumulation in the gut." with about 30% of these being green turtles (Robins 1995). Murphy and Hopkins-Murphy (1989) reviewed the rel- It is estimated that 1%of these drown in the nets and if all evant literature and estimated that up to 300 green turtles comatose turtles are assumed to die then up to 6.8% of all (10,000 loggerheads. 800 ridleys) may be drowned annu- trawl caught turtles die. The observed mortality rate for ally in U. S. waters as a result of shrimp trawling activi- the Queensland coast trawl fishery is lower than other trawl ties. The Sea Turtle Stranding and Salvage Network fisheries and this is believed due to the short tow dura- (STSSN) publishes regular reports on the numbers of tions (~80min). stranded sea turtles from Maine to Texas and in pans of It was estimated that 245 and 100 green turtles, respec- the U. S. Caribbean. Although exact causes of death or tively, were incidentally caught in 1984 and 1985 in trawl morbidity are frequently undetermined, many strandings nets and driftigill nets off Terengganu, Malaysia (Chan et can be linked to incidental capture in fishing activities. al. 1988). In 1990,1991,1992 and 1993 respectively, 329,248,225 The estimated of the Japanese large-mesh and 207 green turtles were reported stranded (Teas 1992% driftnet fishery in the North Pacific Ocean in 1990- 1991 b, 1993, 1994). Thirty -eight green turtles were recorded was 1,501 turtles, of which 248 were estimated to be green stranded in the northwestern Gulf of Mexico between 1986 turtles (Wetherall et al. 1993). It was further speculated and 1989 and during the same interval 51 5 loggerheads by the aforementioned authors that of the 248 green turtles and 357 Kernp's ridleys were stranded. Strong circum- entangled the minimum mortality was 74 turtles, mostly stantial evidence suggests a linkage between strandings juveniles and subadults, and that the likely sources of the and shrimping (Caillouet et al. 1991). The recently orga- green turtles would include the nesting colonies in the nized Caribbean Stranding Network, with participants Ogasawara Islands and French Frigate Shoals. from nine Caribbean countries, has the stated primary The incidental take of sea turtles during dredging op- objectives of uniting stranding efforts throughout the Car- erations has been documented for the Cape Canaveral En- ibbean region and coordinating assessments of marine trance Channel in Florida and the King's Bay Entrance vertebrate deaths (Pinto-Rodriquez et al. 1995). The Channel in Georgia (Dickerson et al. 1992). There were projects of this network include mortality assessments, 16 incidents in the Canaveral Entrance and 1 incident in rescue and rehabilitation, and education. the King's Bay Entrance between 1980 and 1991. Magnuson et al. (1990) reviewed the data linking shrimp Wershoven and Wershoven (1992) stated that since 1986 trawling and sea turtle mortality (mostly loggerheads and propeller injuries were cited in 34 of 56 stranding deaths Kemp's ridleys, some green turtles) in U. S. waters and of Florida juvenile and subadult turtles, and that fishing they concluded that shrimp trawling was the primary agent hooks and lines were responsible for another 7 deaths. for sea turtle mortality caused by humans. They recom- One male (curved carapace length, 78 cm) and one fe- mended the use of TEDs (Turtle Excluder Devices) (see male (curved carapace length, 70 cm) were caught in the section 6.2). tuna longline fishery off Costa Rica in the Pacific Ocean Two immature green turtles, 1 hawksbill, 30 Kemp's (Segura and Arauz 1995). One was hooked in the mouth ridleys and 50 loggerheads were verified caught in the and the other on a flipper. Squid, hening and sail fish summer flounder (Paralichthy~dentatus) trawl fishery off were the bait. Both survived. North Carolina between November 1991 and February Between 1981 and 1990 an average of 14 green turtles 1992 and because a total of 1.063 turtles was estimated to was caught annually in shark nets off Natal, South Af- have been caught (and 89-1 81 estimated to have died as a rica, and 35% were subsequently released (Dudley and result of the trawl fishery) Epperly et al. (1995~)recom- Cliff 1993). mended that sea turtle regulations are needed for this fish- The loss of green turtles to "ghost fishing" (lost or dis- ery- carded fishing gear continues to catch and drown/kill sea Chester et a]. (1 994) describe how sea surface tempera- turtles) should be addressed especially in view of the large ture imagery derived from the U. S. A.'s polar orbiting amounts and numerous types of fishing gear now in use satellite is being used, along with other data, to help re- and their construction from durable material. duce the impact of commercial trawl fishing on sea turtles off the east coast of the U. S. A. 4.5 Dynamics of Populations Green turtles, along with flatbacks, loggerheads and A green turtle survivorship curve is roughly concave ridleys, are incidentally taken in Australia's prawn fish- indicating high egg and hatchling mortality (under natu- ery, but the impact of trawl-induced drownings on the turtle ral conditions). populations there is probably not of such proportions as After examining fourteen cohorts of adult females from to create immediate concern according to Poiner et al. Tortuguero, Bjorndal (1 980b) computed that for the (1990). Approximately 5,295 sea turtles are caught an- Tortugueropopulationto maintain itself, one out of every nually in the trawling fishery of Queensland, Australia, 245.5 eggs. or one out of every 97.2 hatchlings reaching the sand surface, must live to sexual maturity and repro- same conclusion after a long-term study of common snap- duce. Working with data available at the time, Hirth and ping turtles ( serpentina) (Congdon et al. 1994). Schaffer (1974) estimated a range of 0.22 to 1% for Using a stage-class matrix model, Siddeek and Baldwin hatchling survival rates necessary to maintain a stable (1996) assessed the Oman green turtle stock and found population. Fram (1986) reviewed some of the earlier that (1) juveniles (1 -29 years of age) dominated the stable work dealing with gross survivorship from egg to matu- stage-class population vector (2) a maximum hunting rity in four species of sea turtles. He calculated that the quota of about 143 females maintained a stable popula- proportion of eggs surviving to adulthood ranges between tion (3) in addition to protecting eggs and hatchlings, 0.0009 and 0.0018 in a declining population of logger- reduction in juvenile mortality significantly increased the heads. population growth rate and (4) simulated reduction in the Iverson (1991) reviewed the literature on survivorship current annual 4,280 female fishing deaths to 268 pro- schedules of unties and he cited the following annualized duced a positive population growth rate within feasible survivorship for green turtles from egg to hatching: 0.86 stock padeter values. in Australia- 0.767 in Hawaii, 0.69 in Suriname, 0.558 in Using a series of deterministic matrix models for yel- the Galapagos, and 0.396 at Tortuguero (0.607 for adult low mud turtles (Kinostemonflavescensj and Kemp's rid- females at Tortuguero). leys (Lepidochelys kempij, Heppell et al. (1 996) argued Bustard and Tbgnetti (1969) produced a model show- that management efforts focused exclusively on improv- ing how nest destruction is dependent on population den- ing survival in the first year of life are unlikely to be ef- sity and how this provides a mechanism to regulate popu- fective for long-lived turtles; that population projections lation size. for both species predict that headstarting can augment Wilbur and Morin (1988) briefly point out how differ- increasing populations when adult survival is maintained ent kinds and levels of human predation, from little dis- at high levels; and, if subadult and adult survival is re- turbance to high predation on eggs and to high predation duced, headstarting cannot compensate for losses in later on eggs and adults, could influence life histories of iso- stages. lated green turtle populations. Dunham et al. (1 988) dis- In view of these aforementioned demographic studies, cuss how reliable life tables for turtles are much needed the best conservation strategy for green turtles, at the and how sound management programs for commercially present time, is the one that provides protection for all valuable or endangered species can be derived fromlife critical stages in the turtles' life cycle. tables. Although not dealing specifically with sea turtles, Many population models are based on the number and Soul6 (1987) discusses some of the minimum conditions fecundity of tagged migrants and untagged recruits that (population size and range) necessary for the long-term are observed nesting. Consequently the problem of tag viability of natural populations and metapopulations. loss has significant implications in studies of population One of the best sea turtle population models so far de- dynamics (assuming complete beach coverage). Nesters veloped is one formulated for loggerheads. Using demo- are tagged on the trailing edges of each front flipper with graphic data from the southeastern U. S. loggerhead popu- self-piercing strap tags made of various metals-monel, lation, Crouse et al. (1987) developed a stage-based popu- titanium or inconel, or with plastic tags. Based on long- lation model showing that survival in the juvenile and term multiple tagging studies in eastern Australia, Limpus subadult stages has the largest effect on population growth. (1992) determined that titanium tags outlasted monel tags In this model, annual survivorship of hatchlings (