---

Thanks to the David and Lucille Packard Foundation and the Symposium Overseas Travel Fund for assistance with travel funds

A 14-year saturation-tagging programme of loggerhead sea turtles Caretta caretta has been conducted on the Greek Island of Kefalonia. A high encounter rate (100%) with nesting adults on Mounda Beach has allowed an accurate life history to be compiled for a number of animals. Fairly accurate predictions have been made for remigration, inter-nesting period, spatial fidelity and possible site infidelity. The position of all nests was recorded using triangulation. Most nests were exhumed after hatchling emergence activity ceased, which produced accurate nest success-rates. Intermittent hatchling emergence, possibly preceded by intermittent hatching, is commonplace. A detailed examination of undeveloped eggs was performed, following nest excavation. The microclimate within five developing nests was studied, using temperature data loggers above and below the clutch. The sediment grain-size and water content was determined for all nests.

The 2.8 kilometre beach is situated at the south-eastern tip of Kefalonia and is where the main nesting effort occurs. The loggerhead rookery is small (n =148 turtles between 1985-1998) but biologically important. No predators have been observed at the site and there is no indigenous egg-trade. Most nests are laid well above the strand line. Intermittent hatchling emergence is common, with some nests being active for 21 days. A resident population of male Caretta caretta seems to be in the lagoon at Argostoli throughout the year.

Three definitions were used in this study to describe nesting sites: Primary - a turtle's usual nesting site, which is used in most years; Secondary - a beach that is used when access to the primary site is prevented; Emergency - a broadly suitable beach that is used rather than dumping the eggs at sea.

The period between successive nesting seasons at Mounda Beach.

The interval between successive clutches laid in the same season at Mounda Beach.

The date that a turtle laid her first clutch of a season at Mounda Beach.

The particular point on the beach that a turtle laid her nests. Substrate differences exist between the Potomakia and Kaminia ends of the beach. Turtles that have not stranded at both ends may not know that Potomakia is much easier to excavate.

Results are available from the author and cover the following areas:

Sediment grain-size and water content for all nests and control sites.

Life histories for most females, which include the number of clutches produced, remigration period and inter-nesting interval. The success rate of nests and total egg numbers. The duration of an individual's nesting season and possible site infidelity. Intra-beach spatial fidelity; only two out of 15 turtles nested at both ends of the beach during 1998.

Surveys of secondary nesting beaches were carried out weekly. These were used in conjunction with radio telemetry examining migration and inter-nesting habitat use.

Temperature data loggers recorded the incubating nest microclimate every 2.5 hours and the effective depth of diurnal temperature fluctuations on the substrate.

Intermittent hatchling emergence: 2-13 days during 1998.

The stage of development for unhatched eggs and numbers of neonate mortalities.

One turtle [C66][552] has nested in 10/14 years and now is in an annual remigration cycle, laying two clutches each season (1995-1998). The life history for a turtle can be used to predict possible nesting years. The frequency of remigration can change but is fairly consistent, with most animals in a biennial or triennial cycle. Most turtles are never seen again after their initial nesting year (67%). It is not known when an animal has died because no tagged mortality has been reported to date.

Only one turtle [C158] could have laid a nest elsewhere during 1998, as there was a 38-day inter-nesting period between successive nests at Mounda Beach. A turtle could have laid nests at other sites before arriving at Mounda, or after leaving.

Experienced animals tend to nest if they strand and cover the nesting site more efficiently than neophytes. The assumption that the year of tagging is the first nesting year is probably erroneous. Turtles returning to Mounda Beach on a subsequent occasion are termed as resident, regardless of the remigration period. All residents laid two or more clutches during 1998; whilst all single nests were produced by neophytes.

Most turtles have an individual but regular start date to their nesting season. This can alter according to the number of clutches that an animal produces. An individual may arrive earlier in years that she lays three clutches than when she only produces two.

The high encounter rate (100%) with nesting adults enabled accurate records to be produced. Most nests (32 out of 33 nests in 1998) were excavated after hatchling emergence ceased and all recovered eggs were examined to determine possible reasons for non-development. An unusually high number (22% of all eggs) were yolked and undeveloped in 1998; compared with about 10% in other years. Most of these eggs had solidified yolks. This was possibly due to the exceptionally high temperatures experienced during 1998. The substrate surface temperature was >50C by 11:00 each morning during July and August. My colleague (JDR Houghton) is investigating the effect of elevated temperatures on the sex ratio of developing embryos.

One nest 26M had a foreshortened incubation period of 47 days, from oviposition on 23^rd/24^th July 1998 until first emergence 8^th September (mass emergence 20-30 hatchlings, followed by 2 on 9^th Sept, >29 hatchlings on 10^th Sept, 13 on 12^th Sept, 5 on 14^th Sept). (Turtle's tag numbers [C113][580]). Several other nests had the first emergences at 49/50/51 days. Although other places around the world experience hotter temperatures, the population is probably adapted to local conditions, and maintaining an incubation period of around 58 days.

Intermittent hatchling emergence was less prominent during 1998, with several mass-emergences occurring. These were virtually unknown at Mounda until 1998. It is possible that intermittency is more widespread than currently reported, because nests are excavated too early. A number of research sites excavate nests the day following first emergence. At Mounda Beach a nest may not be excavated until five weeks after the first hatchling appears. Live and vigorous neonates are often found in the egg-chamber or the sand column on their way to the surface, during excavations.

Turtles that nest on secondary beaches may be from a different population or, they may use different nesting sites in different years depending upon long-term weather and hydrographic cycles.