The smooth clam, Callista chione (Linnaeus, 1758), is a shallowburrowing filter feeder, inhabiting sandy sediments, from just offshore to a depth of about 130m. It is widely distributed in Mediterranean and Atlantic waters, from the British Isles to the Moroccan coast. In the Mediterranean Sea, C. chione is among the most abundant bivalve species inhabiting shallow softbottomed shores and, in some areas, the most prominent suspension-feeding bivalve species in terms of biomass. The smooth clam is economically important in France, Portugal, Spain (Catalan Sea) and Italy (Adriatic Sea), and part of an extensive clam fishery carried out by artisanal dredging. Although it is not a very marketable species, C. chione is currently widely exploited as a consequence of the overexploitation of other clams, such as Ensis siliqua, Spisula solida, Chamelea gallina and Venus striatula. However, a decline of populations due to overfishing has been observed over a number of years, thus a reduction of fishing effort is now suggested. Despite the fact that C. chione is a common inhabitant of the sandy shores of the Thracian Sea, no systematic fishery exists similar to that present along the coasts of the western Mediterranean. Smooth clams are collected almost circumstantially by divers and are usually landed in small quantities together with other bivalves (Galinou-Mitsoudi, personal communication). Thus, no reliable fishery statistics are available. As C. chione is the object of an intensive uncontrolled fishery, mainly in the western Mediterranean, the study of its population dynamics is important for developing appropriate fishery management strategies. However, research on the biology and ecology of C. chione is limited. In the eastern Mediterranean, records refer only to its geographical distribution. The purpose of the current study is to contribute to the knowledge of the biology of the species in the Thracian Sea by providing information on shell banding, morphometric relationships, age and growth rates. Smooth clams were collected during December 2001 and June 2002 from two sites (St S and St K) along the western coast of Thassos Island, Greece, which are free of any source of major pollution. The two sites differ in wave exposure, anthropogenic activity and clam population density. At both sites the sediment mainly consisted of fine and medium sand, which together accounted for 78–92% of the particles. The redox-potential discontinuity (RPD) layer lay significantly closer to the surface at St K and within the depth range inhabited by C. chione. This resulted in a black biofilm on the shell periostracum, probably of anaerobic bacteria, which disappeared within 2–3 days when the clams were kept in aerobic conditions (immersed in continuously renewed seawater). Sampling of clams was carried out at depths between 1 and 3m at 0.5 m intervals, using a metal box quadrat (100 £ 100 £ 10 cm), which was pushed into the sediment and sifted by hand in situ. A random sample was also collected by hand from shallower depths at both sites in order to increase the number of captured animals. Fifteen clams from each site at each sampling season, chosen to cover the full size range of each sample, were pooled (shell length: 16–72mm) and size parameters, such as shell length (maximum distance of the anterior–posterior axis), shell height (distance of the ventral–dorsal axis, across the shell middle axis), shell width (maximum distance on the lateral axis, between both valves of the closed shell; to the nearest 0.1mm), shell weight, wet flesh weight and dry flesh weight (to the nearest 0.0001 g) were measured. Linear regressions were fitted to the log10-transformed data and the constants a (intercept 1⁄4 the initial growth coefficient) and b (slope 1⁄4 relative growth rate of size parameters) were estimated. t-Tests (H0 1⁄4 1, H0 1⁄4 3) with a confidence level of 95% (a 1⁄4 0.05) were also applied in order to determine whether the relative growth of size parameters is allometric. In order to identify spatial or seasonal fluctuations in the morphometric relationships of the clam, an analysis of variance with a covariate (ANCOVA) was carried out. Morphometric relationships between the shell-size parameters were compared only between the two sampling sites. The relationships between the wet and dry flesh weight against the shell length, and that of the wet flesh weight versus the dry flesh weight (which examines the flesh water content) were compared spatially and seasonally, as these parameters in bivalves are known to vary throughout the year. Characteristic ring-like patterns were noticed on the external surface, consisting of a wide opaque area of shell accretion followed by a transparent narrow area. It was assumed that they form annual increments. Acetate peel replicas of polished and etched shell sections were additionally prepared in order to examine the general shell structure and microgrowth patterns. The patterns of growth lines (see later) were used to identify the position of each growth ring. Distance measurements (to the nearest 0.1mm) were taken from the umbo to the end of each ring along the axis of maximum growth. Both external rings and microgrowth patterns were used for age determination as it soon became apparent that all growth lines identified on the shell surface were also clearly visible in the peel replicas of shell sections, and that measurements taken in the shell surface were in full correspondence with those taken on the peels. The von Bertalanffy growth model was applied to the size-at-age data and growth parameters (L1, K, t0) were estimated using a nonlinear regression under the FiSAT II software package. Morphometric relationships between several size parameters in C. chione are presented in Table 1. The allometric coefficient (b, slope) of these relationships is related allometrically in all cases and reflects ontogenetic changes in the shell and flesh. The shell of C. chione tends to become proportionally higher, wider and heavier, possibly providing better anchorage in the mobile sediment for larger animals. The flesh grows relatively faster than shell length, while the water content is reduced as the clams grow. The spatial comparison of the relationships between shell variables did not show any significant differences (ANCOVA, Correspondence: P. K. Leontarakis; e-mail: fri@otenet.gr