Phylogeography of the green turtle, Chelonia mydas, in the Southwest Indian Ocean

Patterns of mitochondrial DNA (mtDNA) variation were used to analyse the population genetic structure of southwestern Indian Ocean green turtle (Chelonia mydas) populations. Analysis of sequence variation over 396 bp of the mtDNA control region revealed seven haplotypes among 288 individuals from 10 nesting sites in the Southwest Indian Ocean. This is the first time that Atlantic Ocean haplotypes have been recorded among any Indo-Pacific nesting populations. Previous studies indicated that the Cape of Good Hope was a major biogeographical barrier between the Atlantic and Indian Oceans because evidence for gene flow in the last 1.5 million years has yet to emerge. This study, by sampling localities adjacent to this barrier, demonstrates that recent gene flow has occurred from the Atlantic Ocean into the Indian Ocean via the Cape of Good Hope. We also found compelling genetic evidence that green turtles nesting at the rookeries of the South Mozambique Channel (SMC) and those nesting in the North Mozambique Channel (NMC) belong to separate genetic stocks. Furthermore, the SMC could be subdivided in two different genetic stocks, one in Europa and the other one in Juan de Nova. We suggest that this particular genetic pattern along the Mozambique Channel is attributable to a recent colonization from the Atlantic Ocean and is maintained by oceanic conditions in the northern and southern Mozambique Channel that influence early stages in the green turtle life cycle.     

Spatial distribution of turtle barnacles on the green sea turtle, Chelonia mydas

Distribution patterns of epibiotic barnacles on green sea turtles were investigated in waters neighboring Okinawa, Japan. A number of barnacle species were found to coexist on the turtles and were classified into three genera: Chelonibia, Platylepas and Stomatolepas. Attachment sites on the turtles varied among the barnacle species, suggesting that there is niche partitioning with respect to their microhabitat selection. Turtle bodies offer a “patchy” environment for barnacles, so we also analyzed coexistence patterns in the context of an aggregation model. Within each genus, individual barnacles showed a clumped distribution. The different genera do not have mutually exclusive distribution patterns, but instead occur on the same turtle to various degrees. However, when turtles were divided into two size classes, both the level of aggregation and the degree of interspecific overlap among the barnacles was significantly lower on large turtles. We suggest that obtaining basic information on turtle epibionts will shed light on the biology of wild turtles, which is still largely unknown.     

The Green sea turtle (Chelonia mydas): mating, nesting and hatching on a farm

The collection of eggs from wild rookeries for incubation and hatching on the Mariculture Ltd farm on Grand Cayman Island, BWI, was described in the previous paper. It was realized, however, that this method of stocking the farm must sooner or later be superseded by the production of farm-laid eggs, and at an early stage a start was made on the task of establishing a breeding colony. The necessary collection of adult turtles and progress in breeding to the end of 1973 is described in the present paper.     

The biology of terns nesting at Aldabra Atoll, Indian Ocean, with particular reference to breeding seasonality

Five species of tern breed on Aldabra Atoll (09° 24' S; 46° 20' E). The Caspian tern Sterna caspia and Crested tern S. bergii feed exclusively in very shallow reef/lagoon water, the Fairy tern Gygis alba and Brown noddy Anous stolidus feed out at sea, and the Black-naped tern Sterna sumatrana is intermediate in its foraging. Both of the shallow-water species lay during the south-east monsoon season, the Caspian tern from April to August and the Crested tern from June to August, but the Crested tern also lays in December and January. The remaining three species have extended laying periods largely circumscribed by the north-west monsoon season from November to March. Breeding population size of the Caspian tern is in the low tens and of the Brown noddy in the low thousands, with the other species each numbering in the hundreds. The distribution and abundance of the nine species of tern breeding within the Seychelles ( sensu lato ) vary on the different island groups in a manner interpretable in terms of depth of the surrounding waters. Systematic differences between the central Seychelles and Aldabra groups in timing of breeding by terns which feed out at sea may be associated with seasonal latitudinal movement of the divergence zone between the South Equatorial Current and the Equatorial Counter-current, acting via correlated latitudinal shifts of prey species and game-fish abundance.     

The Green sea turtle (Chelonia mydas): nesting on Ascension Island, 1973–1974

The collection of eggs of Chelonia mydas from beaches on Ascension Island and elsewhere for hatching at the Mariculture Ltd farm on Grand Cayman was described in a previous paper (Simon, 1975). Improved organization of these collections has made it possible to record many incidental observations which throw light on the nesting behaviour and performance of the female Chelonia , and Ascension Island, because of its small size, numerous beaches and good communications, proved to be particularly favourable for such research. The present paper records observations made in February and March 1973 and 1974. The second author has contributed only by the analysis of the data and the preparation of the paper.     

Pulmonary function of the green sea turtle, Chelonia mydas

Lung volumes, oxygen uptake (VO2), end-tidal PO2, and PCO2, diffusing capacity of the lungs for CO (DLCO), pulmonary blood flow (QL) and respiratory frequency were measured in the green sea turtle (Chelonia mydas) (49-127 kg body wt). Mean lung volume (VL) determined from helium dilution was 57 ml/kg and physiological dead space volume (VD) was about 3.6 ml/kg. QL, determined from acetylene uptake during rebreathing, increased in proportion to VO2 with temperature. Therefore, constant O2 content difference was maintained between pulmonary arterial and venous blood. DLCO, measured using a rebreathing technique, was 0.04 ml X kg-1 X min-1 X Torr-1 at 25 degrees C. Several cardiopulmonary characteristics in C. mydas are advantageous to diving: large tidal volume relative to functional residual capacity promotes fast exchange of the alveolar gas when the turtle surfaces for breathing: and the concomitant rise of pulmonary blood flow and O2 uptake with temperature assures efficient O2 transport regardless of wide temperature variations encountered during migrations.     

Characterization of polymorphic microsatellite markers for the green turtle (Chelonia mydas)

We describe primers and polymerase chain reaction conditions to amplify 12 microsatellite loci from the green turtle (Chelonia mydas), including one dinucleotide, four trinucleotide and seven tetranucleotide loci. The primers were tested on 78 individuals from a Pacific population nesting in the Hawaiian Islands. The primer pairs developed in this study yielded an average of 8.33 alleles per locus (range of 3-15 alleles), an average observed heterozygosity of 0.668 (range 0.309-0.910), and an average polymorphic information content of 0.647 (range 0.287-0.894).     

Control of salt gland activity in the hatchling green sea turtle, Chelonia mydas

 We studied the control of salt gland secretion in hatchling Chelonia mydas. The threshold salt load to activate salt secretion was between 400 μmol NaCl 100 g bodymass (BM)⁻¹ and 600 μmol NaCl 100 g BM⁻¹, which caused an increase in plasma sodium concentration of 13% to 19%. Following a salt load of 2700 μmol NaCl 100 g BM⁻¹, salt gland secretion commenced in 12 ± 1.3 min and reached maximal secretory concentration within 2–7 min. Maximal secretory rate of a single gland averaged 415 μmol Na 100 g BM⁻¹ h⁻¹. Plasma sodium concentration and total osmotic concentration after salt loading were significantly higher than pretreatment values within 2 min. Adrenalin (25 μg kg BM⁻¹) and the cholinergic agonist methacholine (1 mg kg BM⁻¹) inhibited salt gland activity. Atropine (10 mg kg BM⁻¹) reversed methacholine inhibition and stimulated salt gland secretion when administered with a subthreshold salt load. Arginine vasotocin produced a transient reduction in sodium secretion by the active gland, while atrial natriuretic factor, vasoactive intestinal peptide and neuropeptide Y had no measurable effect on any aspect of salt gland secretion. Our results demonstrated that secretion of the salt gland in C. mydas can be modified by neural and hormonal chemicals in vivo and that the cholinergic and adrenergic stimulation of an exocrine gland do not appear to have the typical, antagonist actions on the chelonian salt gland. Accepted: 28 September 1999     

Bile salts of the green turtle Chelonia mydas (L.)

1. Bile salts of the green turtle Chelonia mydas (L.) were analysed as completely as possible. 2. They consist of taurine conjugates of 3 alpha, 7 alpha, 12 alpha, 22 xi-tetrahydroxy-5 beta-cholestan-26-oic acid (tetrahydroxysterocholanic acid) and 3 alpha 12 alpha, 22 xi-trihydroxy-5 beta-cholestan-26-oic acid, with minor amounts of 3 alpha, 7 alpha, 12 alpha-trihydroxy-5beta-cholan-24-oic acid (cholic acid), 3alpha, 12 alpha-dihydroxy-5beta-cholan-24-oic acid (deoxycholic acid) and possibly other bile acids. 3. Cholic acid and deoxycholic acid represent the first known examples of bile acids common to chelonians and other animal forms: they may indicate independent evolution in chelonians to C24 bile acids. 4. The discovery of a 7-deoxy C27 bile acid is the first evidence that C27 bile acids or their conjugates have an enterohepatic circulation.     

The rate of predation by fishes on hatchlings of the green turtle (Chelonia mydas)

This study addresses the need for empirical data on the survival of sea turtle hatchlings after entry into the sea by (1) developing a method for measuring marine predation; (2) estimating predation rates while crossing the reef; and (3) investigating the effect of environmental variables on predation rates. Predation rates were quantified by following individual hatchlings, tethered by a 10m monofilament nylon line, as they swam from the water's edge towards the reef crest. Predation rates under particular combinations of environmental variables (tide, time of day, and moon phase) were measured in separate trials. Predation rates varied among trials from 0 to 85% with a mean of 31% (SE=2.5%). The simplest logistic regression model that explained variation in predation contained tide and moon phase as predictor variables. The results suggest that noctural emergence from the nest is a behavioral adaptation to minimize exposure to the heat of the day rather than a predator-escape mechanism. For the green turtle populations breeding in eastern Australia, most first year mortality is caused by predation while crossing the reef within the first hour of entering the sea.     

The primary structure of myoglobin from pacific green sea turtle (Chelonia mydas caranigra)

The amino acid sequence of the main component myoglobin from skeletal muscle of Pacific green sea turtle (Chelonia mydas caranigra) has been determined. The globin is 153 residues in length and has a free amino-terminus. The heme-binding and internal residues are as found in mammalian myoglobins. Ten substitutions are observed between this myoglobin and that from map turtle. About 38, 52, 47 and 86 substitutions are noted in comparison with the myoglobins of other reptiles, mammals, birds and fish, respectively. The inferred pattern of structural stabilization and conservation of two loci are typical of tetrapod myoglobin.     

Cutaneous fibropapillomas and renal myxofibroma in a green turtle, Chelonia mydas

A debilitated 7 kg juvenile green turtle (Chelonia mydas mydas) with multiple ulcerated and infected cutaneous fibropapillomas was clinically evaluated and found to have a nonregenerative anemia, hypoproteinemia, hypoalbuminemia and several electrolyte abnormalities. Surgery was performed to remove the larger tumors. The turtle did not eat postsurgically, and an attempt was made to place a pharyngostomy tube utilizing endoscopy. Edematous esophageal papillae, the angulation of the gastroesophageal junction, and a S-shaped configuration of the esophagous prevented successful placement of the tube. The animal was found dead the next day and necropsied. Multiple large white firm nodules were seen bulging from both kidneys. Microscopic examination of the nodules resulted in a diagnosis of renal myxofibroma.     

Conformational changes of alpha-macroglobulin and ovomacroglobulin from the green turtle (Chelonia mydas japonica)

Green turtle plasma alpha-macroglobulin and ovomacroglobulin underwent conformational changes when they were treated with proteinases or methylamine. Their conformational changes were studied by HPLC gel chromatography, circular dichroism, and electron microscopy. The Stokes radii of native green turtle alpha-macroglobulin and ovomacroglobulin were estimated to be 84.3 +/- 0.5 A, and 93.0 +/- 0.5 A, respectively, by means of an HPLC experiment. After reaction with methylamine or proteinases, the Stokes radius of alpha-macroglobulin changed to 83.0 +/- 0.5 A or 85.4 +/- 0.5 A, respectively, and that of ovomacroglobulin to 93.0 +/- 0.5 A or 87.1 +/- 0.5 A. The circular dichroic spectra of native alpha-macroglobulin and ovomacroglobulin exhibited a negative band at around 215 nm, indicating the presence of beta-structure. Reaction of the two macroglobulins with methylamine resulted in a slight decrease in the ellipticity and reaction with proteinases led to a slight increase. The electron micrographic images of native alpha-macroglobulin and ovomacroglobulin can be described as deformed rings for the former and rugby balls for the latter. A common characteristic feature of the two molecules was that the central parts of the molecules were only thinly occupied by subunit. After reaction of macroglobulins with proteinases, the void spaces became partially filled and their overall shape more rectangular. Methylamine treatment caused a structural change only in alpha-macroglobulin but not in ovomacroglobulin. The difference in the susceptibility of the macroglobulins to methylamine was taken as an indication of evolutional divergence of the two homologous proteins within the last 300 million years.     

Thermal dependence of tissue metabolism in the green turtle,Chelonia mydas

To better understand how tissue specific metabolic rates might contribute to the maintenance of elevated body temperatures in green turtles (Chelonia mydas), we determined the effect of temperature on oxygen consumption of green fat, small intestine, nonswimming skeletal muscle, pectoralis muscle, liver, heart, and kidney tissues from 5–35°C. We found a direct relationship between tissue metabolic rate (microliters of O₂/g wet mass per hour) and temperature in all tissues measured except for green fat. The Q₁₀ values ranged from 0.65 to 3.38. There were significant differences in metabolic rate among tissues as well as in how temperature influenced tissue metabolic rates. Tissue metabolic rates were highest in kidney and heart tissues. Green fat and small intestine had the lowest and most temperature-insensitive values. Muscle tissue had a high oxygen consumption relative to other reptiles, and this elevated metabolism may provide a functional advantage for long distance swimming and heat production.     

Seasonality and numbers of green turtles Chelonia mydas nesting on the Pitcairn Islands

During 1991 and 1992 the first regular field observations of green turtles, Chelonia mydas , on the Pitcairn Islands (South Pacific) revealed that nesting occurs only on Henderson where about 10 females laid eggs (c. 1% of the French Polynesian population). Substrates on Pitcairn and Ducie Atoll are unsuitable for nesting and no signs of nesting were found on Oeno Atoll where the terrain appears suitable.     

Characterization of single nucleotide polymorphism markers for the green sea turtle (Chelonia mydas)

We present data on 29 new single nucleotide polymorphism assays for the green sea turtle, Chelonia mydas. DNA extracts from 39 green turtles were used for two methods of single nucleotide polymorphism discovery. The first approach employed an amplified fragment length polymorphism technique. The second technique screened a microsatellite library. Allele-specific amplification assays were developed for high-throughput single nucleotide polymorphism genotyping and tested on two Pacific C. mydas nesting populations. Observed heterozygosities ranged from 0 to 0.95 for a Hawaiian population and from 0 to 0.85 for a Galapagos population. Each of the populations had one locus out of Hardy-Weinberg equilibrium, SSCM2b and SSCM5 for Hawaii and Galapagos, respectively. No loci showed significant genotypic linkage disequilibrium across an expanded set of four Pacific nesting populations. However, two loci, SSCM4 and SSCM10b showed linkage disequilibrium across three populations indicating possible association.     

Patterns of lipid storage and mobilisation in the female green sea turtle ( Chelonia mydas)

Reproductive data from southern Queensland indicate that vitellogenesis in female Chelonia mydas takes approximately 8 months and is followed by a migration to a breeding area. At Heron Island, females lay multiple clutches over approximately 3 months. To investigate how females mobilise and store lipid during the breeding season we collected plasma, yolk, and fat tissue samples from females at a variety of stages during the nesting season. In breeding females, concentrations of plasma triglyceride increased seasonally. They reached peak concentrations during vitellogenesis and courtship, remained high throughout the nesting season, and then declined to a nadir after the last clutch. Plasma protein concentration increased throughout the breeding season, peaking following the last clutch for the season. Yolk lipids were highest during courtship and were similar throughout the nesting season, suggesting that uptake of lipid by ovarian follicles is completed prior to the beginning of the nesting season. Plasma triglyceride decreases in females with prolonged periods of unsuccessful nesting, and total lipid levels in adipose tissue and follicle yolks were significantly lower in atretic females. It appears that: (1) endogenous energy reserves can be reduced by stochastic environmental events (such as those reducing nesting success), and (2) a metabolic shift signalling the end of the nesting season is characterised by a drop in plasma triglycerides and slight increase in total plasma protein.     

Single paternity of clutches and sperm storage in the promiscuous green turtle (Chelonia mydas)

Paternity of 22 green turtle ( Chelonia mydas ) clutches from 13 females of the southern Great Barrier Reef breeding population was determined through microsatellite analyses at five loci, including the analysis of successive clutches for nine of the females. A large number of alleles per locus (10–40) provided probabilities of detecting multiple paternity that were quite high, particularly at all loci combined (99.9%). Although green turtles are promiscuous breeders and there was an expectation of finding extensive multiple paternity, only two clutches were multiply sired and, in these, very few eggs had been fertilized by a secondary male. The rarity of multiple paternity may reflect either a low proportion of multiple matings by females in this population, or sperm competition, possibly resulting from a first-male sperm preference. Additionally, the analysis of > 900 offspring provided data on mutations, which included 20 mutation events that were observed in 27 offspring and involved both maternal and paternal lineages. Most mutations ( n = 16) occurred at a single highly variable locus and their presence emphasizes the need to use multiple loci in paternity studies.     

Mitochondrial DNA diversity and phylogeography of endangered green turtle (Chelonia mydas) populations in Africa

We analysed the genetic structure of seven nesting sites of the endangered green turtle (Chelonia mydas) in Africa using mitochondrial DNA control region sequences. Tissue samples were collected from 188 nesting females at six sites in West Africa and one in the Indian Ocean. A 488 bp fragment of the control region revealed 14 different haplotypes, 10 of which are previously undescribed. The most common haplotype (CM8) was observed in 157 individuals. All other haplotypes were closely related, except two divergent lineages: CM38, removed by four substitutions, and the three Indian Ocean haplotypes, distinguished by 31 substitutions. Significant differences in haplotype and nucleotide diversity were observed between Atlantic rookeries and among ocean basins. Analysis of molecular variance revealed high levels of differentiation between the Atlantic and the Indian Ocean populations but a much shallower Atlantic substructuring. Green turtle population genetic structure is thought to have been shaped by a dynamic succession of extinction and recolonisation of rookeries, by natal homing and occasional breakdown in nest-site fidelity. Mismatch distributions of pairwise differences between haplotypes at each rookery were found to be consistent with recent population expansion. We argue that demographic histories can be explained by scenarios at several temporal scales, including geological events, sea level fluctuations and more recent patterns of exploitation. We discuss management and conservation implications of our results for these threatened populations, identifying two ESUs (one in the Atlantic and one in the Indian ocean) and three MUs within the Atlantic.