INTRODUCTION
Razor clams are edible bivalve molluscs that fetch high prices in international markets (Baron et al., Reference Baron, Real, Ciocco and Re2004). Solen dactylus is one of the most important species of Solenidae, which like other species of Solenidae inhabits high and mid intertidal sandy–muddy beaches (Bruyne, Reference Bruyne2003) along the Oman Sea and the Persian Gulf (Bosch, Reference Bosch1982) (Figure 1). Solen dactylus has an elongated and straight shell (Bosch et al., Reference Bosch, Dance, Moolenbeek and Oliver1995) with one cardinal tooth in each valve (Barnes et al., Reference Barnes, Calow, Olive, Golding and Spicer2001). On the Iranian coast of the Persian Gulf, this species can introduce as a commercial and vulnerable razor clam, yet there is not enough information on growth and reproduction of S. dactylus to provide a suitable plan for conserving and managing this stock.
Study area in the Persian Gulf. Inset shows the sampling sites: ‘Dolat park’ and ‘Qadir park’ on Bandar Abbas coast.
Growth and reproduction of bivalvia especially razor clams have been studied in several different countries (e.g. Holland & Dean, Reference Holland and Dean1977; Breen et al., Reference Breen, Gabriel and Tyson1991; Gasper et al., Reference Gasper, Ferreira and Monterio1999; Fahy et al., Reference Fahy, Norman, Browne, Roantree, Pfeiffer, Stokes, Carrol and Hannaffy2001; Laudien et al., Reference Laudien, Brey and Arnttz2002; Sejr et al., Reference Sejr, Sand, Jensen, Peterson, Christensen and Rysgard2002; Zeichen et al., Reference Zeichen, Agnesi, Mariani, Maccaroni and Ardizzone2002; Baron et al., Reference Baron, Real, Ciocco and Re2004; Darriba et al., Reference Darriba, Sanjuan and Guerra2004; Gribben, Reference Gribben2005; Pinn et al., Reference Pinn, Richardson and Thompson2005; Ferreira et al., Reference Ferreira, Paixao, Alcantara, Santos and Rocha2006; Remacha & Anadon, Reference Remacha and Anadon2006). Holland & Dean (Reference Holland and Dean1977) worked on the Creek population of Tagelus plebeius and reported spring unimodal recruitment for this species; furthermore, they stated that larger length of these clams mainly occur in low intertidal areas. Fahy et al. (Reference Fahy, Norman, Browne, Roantree, Pfeiffer, Stokes, Carrol and Hannaffy2001) reported some growth and reproductive parameters for Ensis arcuatus in Ireland and introduced this species as a spring spawner. Baron et al. (Reference Baron, Real, Ciocco and Re2004) provided the first estimation of growth rate in Ensis macha in Argentine Patagonia. Darriba et al. (Reference Darriba, Sanjuan and Guerra2004) estimated a gonadal condition index (GCI) for Ensis arcuatus in Spain demonstrating the gonadal development of this species in the reproductive cycle. Gribben (Reference Gribben2005) studied the reproduction of Zenatia acinaces in New Zealand providing valuable information on the timing of spawning events in this razor clam. Using histological methods and macroscopic observation on Solen marginatus in three different parts of Spain, Remacha & Anadon (Reference Remacha and Anadon2006) discussed the various times of reproduction for this species.
This study provides the first data on growth rate, reproductive cycle and size at the first maturity for this species on the northern coast of the Persian Gulf, Iran, Bandar Abbas. This region is characterized by high temperatures in summer (annual range 16–45°C). Seasonal winds, current regimes and availability of nutrients make this area suitable for living marine animals, especially macrobenthic organisms. In Iran, fishermen harvest this species as food and bait for fishing; while, there are no regulations in place to protect this species. Consequently, there is a threat for the stock of Solen dactylus as a result of overfishing in this area.
For the efficient management of fishing practices of Solen dactylus in the study area, it is important to have accurate information on their growth and reproductive cycle. This study tries to provide a correct pattern of growth and reproductive cycle in Solen dactylus to control the fishing practices of this species in Bandar Abbas.
MATERIALS AND METHODS
Study area and sampling
Specimens of razor clams (Solen dactylus) were collected monthly from April 2007 to March 2008 along two perpendicular transects to the shore (30 × 30 m) in three different intertidal zones (high, mid and low tidal zones) of the Golshahr coast in Bandar Abbas (first transect, Dolat park, 56°21′E 27°12′N and second transect, Qadir park, 56°20′E 27°11′N) during the low tide (Figure 1). Zonation of intertidal areas was defined according to the depth. These stations were selected because they were the only sites which razor clams found on the Golshahr coast of Bandar Abbas; furthermore, clams in this area were abundant and available for sampling and the effect of overfishing can be studied by selecting these areas as a sampling site. Samples were captured by hand with a 0.5 m long metal wire (2 mm diameter) forming the V shape at one end. Three different sizes of this tool ranging from 3–10 mm in width of V shape part were used to pull out the different size-classes of razor clams from their hole. All specimens were fixed in 10% formaldehyde immediately after capturing (Figure 2).
The sampling tool and method for collecting razor clams on Bandar Abbas coast.
Data analysis
BIOMETRY AND GROWTH
The biometric measurements were undertaken including anterior–posterior length (length), dorsal–ventral length (width), the distance between two valves (diameter), total weight (TW), wet weight of the soft parts (SPW), dry weight of both valves (VW) by putting valves in 60°C for 3 hours in an oven (Darriba et al., Reference Darriba, Sanjuan and Guerra2004) and gonad weight (in the reproductive cycle). Clams were measured to the nearest 0.1 mm with Vernier calipers. Parameters of the relationship between length and the total weight of Solen dactylus were estimated by the following regression analysis and Excel program.
where W is the total weight (g), L is the length (mm), a and b are the constant (Park & Oh, Reference Park and Oh2002), and the length–weight relationship of b was tested by the Student's t-test.
The Von Bertalanffy growth function (Baron et al., Reference Baron, Real, Ciocco and Re2004) was used to estimate growth parameter as follows:
where Lt is the shell length at time t, L∞ is the asymptotic shell length and k is the annual growth coefficient.
For determining t0 the following function was used (Pauly, Reference Pauly1979):
REPRODUCTIVE PATTERNS AND TEMPERATURE MEASUREMENT
Around 200 specimens, ranging from 35 to 125 mm in total length, were studied by histological preparation. Gonads in this species are located inside the foot. A small section of the foot (including the gonad tissue) was taken, fixed in Bouin's fixative for 24 hours, preserved in 70% alcohol, dehydrated in an ethanol series and infiltrated with paraffin. Sections of 5 µm were cut and stained with haematoxylin, contrasted with eosin (Darriba et al., Reference Darriba, Sanjuan and Guerra2004). Sex was determined by observation of gonadal smears under binocular photomicroscope.
For estimating Lm50 the following function was used by Logistic program (Cherif et al., Reference Cherif, Zarrad, Gharbi, Missaoui and Jarboui2007).
where P is the proportion of matured specimens, b is the slope of the graph and Lm50 is the first maturity length (Cherif et al., Reference Cherif, Zarrad, Gharbi, Missaoui and Jarboui2007).
Sex ratio was calculated and tested by the Chi-square test (Cherif et al., Reference Cherif, Zarrad, Gharbi, Missaoui and Jarboui2007).
A gonadal condition index was determined (Darriba et al., Reference Darriba, Sanjuan and Guerra2004) as follows:
Sea surface temperature was recorded monthly by thermometer to the nearest 0.1°C. The correlation between GCI and sea surface temperature was tested by Pearson coefficient at 95% confidence limit.
RESULTS
Distribution, biometric analysis and growth
Razor clams in this study inhabit intertidal zones where water remains on coast after a low tide. The first transect was on mid tidal and low tidal zones. The sizes of razor clams were smaller (62.6 mm±15.74) (mean±SD) but were more abundant (8 ind. m−2±0.7) than the second transect. In the second transect specimens were larger (73.26 mm±16.96) while distributed in high and mid tidal zones (6 ind. m−2±0.4). The slope of the coast in the first transect was lower than the second transect and the coast became dry sooner than the second transect. Figure 3 shows abundance of razor clams in the intertidal pools in different intertidal zones.
Diagram of abundance of razor clams in different intertidal zones in two transects of Bandar Abbas coast, the northern Persian Gulf.
The largest razor clams in the first and second transects were 120 and 119 mm, respectively. The smallest specimens in the first and second transect were 25 mm in length. The average size of individuals in the first transect and the second transect was 62.6 mm±15.74 and 73.26 mm±16.96 (mean±SD), respectively. The average total weight of individuals in the first transect and the second transect was 4.99 g±3.1 and 7.79 g±4.49 (mean±SD), respectively. The relationship between length and total weight of Solen dactylus is W = 0.0001 L2.5921. Correlation coefficient (r2 = 0.96) for all specimens analysed in this study was significant at P < 0.001 (Figure 4). The mean of b was 2.57±0.1 (mean±SD) during one year. The result of the Student's t-test determined that the length–weight relationship of this species was negative allometric at 95% confidence limit of b.
The relationship between length and total weight of Solen dactylus for both transects.
Length–frequency distributions of Solen dactylus from the monthly sampling (April 2007 until March 2008) showed that 4 cohorts were distinguishable with one recruitment period per year (between April and May) indicated by small size-groups in April (Figure 5). In both transects, 65–75 mm size-groups were more abundant than other size-groups. This species grows about 22–29 mm every year.
Length–frequency distributions of Solen dactylus from April 2007 to March 2008 in two transects. The frequency scale is the number of individuals in each length-class.
In accordance to Von Bertalanffy's growth equation parameter estimation, the following four parameters were obtained for transects 1 and 2: 101 and 108 mm for L∞, 0.27 and 0.28 year−1 for k and −0.99 and −0.94 for t0, respectively.
REPRODUCTION BIOLOGY
A total of 200 individuals were studied to explore reproduction biology, in which sexes of 76 clams were not distinguishable, 65 clams (52.4%) were male and 59 (47.58%) were female. The ratio of males to females was marginally biased toward males in both transects (1.1 M/1 F). The result of the Chi-square test showed that sex-ratio was not significant from 1:1 (χ2 = 0.29, df = 1, P > 0.05). There was no size difference between sexes. The two sexes were synchronic in the evolution of the GCI during the reproductive cycle with minimum values in late spring and summer (sexual rest stage), a sharp increase in late autumn and early winter and after late January and early February (the period after spawning) decreased and became minimum again. Gametogenic cycle started when the sea-surface temperature recorded about 27°C in early September and clams spawned when temperature was about 20°C in January. Temperature and gametogenic development showed a negative correlation (Pearson coefficient = –0.759) at 95% confidence limit. In the second transect GCI was higher than the first transect during the year (Figure 6). Length at the first maturity was Lm50 = 46.1 for both sexes (Figure 7). It suggests that 50% of clams enter to the reproductive cycle when their size is 46.1 mm in length.
Variation of gonadal condition index (GCI) and sea-surface temperature from April 2007 to March 2008 in both sexes of Solen dactylus in two transects of Bandar Abbas coast, the northern Persian Gulf (mean±SE).
The first maturity length (Lm50) of Solen dactylus for both sexes in Bandar Abbas coast, the northern Persian Gulf.
The results of gonad development study showed the following characteristic for seven stages:
Stage 0 (sexual rest)
Gonads were not observed inside the foot. Connective tissues were observed in most of the specimens and the sexes were not distinguishable. At the macroscopic level, the gonadal tissue was not present and in some specimens was reduced to a thin lamina along the foot (Figure 8).
Photomicrograph showing the rest stage (stage 0) in Solen dactylus. Scale bar: 50 µm.
Stage I (proliferation of gonads and start of gametogenesis)
Mitotic activity is started. Gonad volume was small but follicle size increased with thick walls. In males, spermatocytes and spermatids developed in middle of the follicle. Spermatozoa, with tails pointing into the lumen in the centre of follicle were observed. In females, oogonia and previtellogenic oocytes appeared on the follicular wall. At the microscopic level, a canal was observed along the foot and the gonadal tissue appeared (Figures 9A & 10A).
Photomicrographs showing stages in the development of male gonad: (A) stage І; (B) stage ІІ; (C) stage ІІІA; (D) stage ІІІB; (E) stage ІІІC; and (F) stage ІV. Scale bar: 100 µm in (A, E and F), 50 µm in (B and C) and 25 µm in (D).
Photomicrographs showing stages in the development of female gonad: (A) stage І; (B) stage ІІ; (C) stage ІІІA; (D) stage ІІІB; (E) stage ІІІC; and (F) stage ІV. Scale bar: 100 µm in (A, B, D, E and F), 50 µm in (B and C) and 25 µm in (D).
Stage II (advanced gametogenesis)
Follicles were large with a thin wall. In males, spermatogonia were limited to follicle wall. Spermatids and spermatocytes were abundant. Spermatozoa were arranged radially with the tails toward the follicles. In females, follicle consisted of a few oogonia, previtellogenic oocytes and vitellogenic oocytes. Polygonal mature oocytes were free in the centre of the follicle, whereas immature oocytes were connected to the follicle wall by thin stalk. At the macroscopic level, the gonadal tissue was spreading along the foot and partly was covering the digestive gland. Connective tissues thickness of the foot was reduced. Gametes were observed along the foot (Figures 9B & 10B).
Stage III A (ripeness)
Follicles with thin walls were very large and located tightly together. In males, a few spermatogonia and spermatids were present. Dense spermatozoa were abundant which was restricted to the follicle walls. In females, some vitellogenic oocytes occupied almost the entire follicle. A few oogonia and previtellogenic oocytes were observed. At the macroscopic level, the gonad completely covered the digestive gland and connective tissues of the foot were very thin (Figures 9C & 10C).
Stage III B (start of spawning)
Follicles have contracted with space appearing between them and their walls were thin. In males, spermatozoa were fewer than previous stages and empty space was observed between follicles. A few spermatids and spermatocytes were present. In females, empty spaces were observed in the follicular lumen and a few previtellogenic and vitellogenic oocytes were observed. Very few oogonia were present. At the macroscopic level, the gonad was full of gametes which were larger than the other stages. Matured gametes started to release (Figures 9D & 10D).
Stage III C (restoration and the last emission)
Some follicles were empty. Follicle size was decreased and thickened. In males, spermatocytes were more than spermatozoa and spermatids. In females, follicles contracted and developed thick walls. Previtellogenic oocytes were more than ripe oocytes. Macroscopically, the gonad tissue was decreasing while thickness of the connective tissue was increasing. Digestive gland appeared after previous stages (Figures 9E & 10E).
Stage IV (exhaustion)
After several partial spawnings, a successful spawning led to this stage. In this stage follicles were very small and most of them were empty. In males, a few spermatozoa were observed and in females a few ova were present. Cytolysis in some oocytes happened. A few ripe oocytes and spermatozoa were evacuating. At the macroscopic level, connective tissues of the food became thick and the gonadal tissue was disappearing and returned to a very thin lamina inside the foot. Some follicles were absorbing to the connective tissue of the foot (Figures 9F & 10F).
DISCUSSION
Solen dactylus, like other razor clams, inhabits intertidal zones (Fahy et al., Reference Fahy, Norman, Browne, Roantree, Pfeiffer, Stokes, Carrol and Hannaffy2001; Darriba et al., Reference Darriba, Sanjuan and Guerra2004). This species is not able to close its shell completely; therefore, the posterior and anterior of the shell always remain open. In addition, this razor clam uses hydraulic power for digging. For these reasons, this species is adapted to lives in intertidal pools.
In transect 1, the slope of the coast was lower and became dry sooner than the second transect. Samples were in the middle and lower intertidal zones in the first transect because higher intertidal zone was dried and there were no water remains on the coast so that clams could live there. In the second transect, samples only lived in higher and middle intertidal zones because intertidal pools were present there. On the other hand, at the time of submersion in high tide, some predators like shore birds have no chance to hunt these clams. In conclusion, clams adapted to live in some places on coast with a lower exposed time in low tide. Razor clams in the first transect were smaller than the ones in the second transect. It can suggest that in the second transect older clams were found which produced the younger clams in the first transect. Probably, similar cohorts lived in each transect that is related to settlement of these clams near each other at the juvenile stages. The correlation coefficient between length and total weight–length in the first and second transect was r2 = 0.97. In Ensis arcuatus and Solen strictus this parameter was r2 = 0.97 and r2 = 0.91, respectively (Fahy et al., Reference Fahy, Norman, Browne, Roantree, Pfeiffer, Stokes, Carrol and Hannaffy2001; Park & Oh, Reference Park and Oh2002). This parameter shows the relationship between length and the total weight of samples; furthermore, the mean of b was 2.57±0.1 and like Solen strictus, the length–weight relationship of this species was negative allometric (Park & Oh, Reference Park and Oh2002).
Table 1 shows the different Von Bertalanffy's values and reproduction time in many countries. In the first and second transect, k were 0.27 and 0.28, respectively, which are close to Ensis macha in Argentina and Chile (k = 0.2–0.25) (Baron et al., Reference Baron, Real, Ciocco and Re2004); however, the weather in this study is different from Argentina. Undoubtedly, several factors play remarkable roles in growth rate of these species like climate, longitude, latitude and individual's difference. There are several notable ecological and environmental factors in the northern parts of the Persian Gulf that can provide a good place for growth and reproduction of sea shells. These factors are current regimes, seasonal winds, high concentration of nutrients and phytoplankton availability as a result of water mixing. The smallest individuals were about 25–28 mm in April. In late winter, these clams had grown to about 48–53 mm. This species, every year, grows about 22–29 mm in length, which is similar to Ensis arcuatus (25–28 mm; Fahy et al., Reference Fahy, Norman, Browne, Roantree, Pfeiffer, Stokes, Carrol and Hannaffy2001). Growth rate in razor clams after the first year of age became increasingly slower, for these clams enter to the reproductive cycle after this year when they are two or three years old. For this reason, they should spend a lot of energy on reproduction to spawn and reproduce that it can inhibit a fast growth at this time. Only one recruitment was observed in spring (April) which is similar to Tagelus plebusis (Holland & Dean, Reference Holland and Dean1977). This was in contrast to the twice-yearly recruitment reported for many temperate zone bivalves (Holland & Dean, Reference Holland and Dean1977). Specimens of Solen dactylus emit a large amount of gametes in a successful spawning and it can lead to a synchronized recruitment in one time.
Parameters of Von Bertalanffy growth function and reproductive seasons of different species of razor clams in different countries.
K, growth constant; L∞, asymptotic length; t0, age at length zero.
The total length at the first maturity was 46.1 mm, corresponding with an age of 1–1.5 years. This was in contrast to Ensis arcuatus that matured at the age of 2–3 years at a total length of 85 mm (Fahy et al., Reference Fahy, Norman, Browne, Roantree, Pfeiffer, Stokes, Carrol and Hannaffy2001) which has larger L∞ than Solen dactylus.
The gonadal condition index was determined because it was highly efficient in distinguishing between the different stages of gametogenic development in contrast to the other gonadal condition indices (Darriba et al., Reference Darriba, Sanjuan and Guerra2004). The GCI analysis was completely similar to the sexual cycle. In the first stage (stage 0), gonad was completely empty and GCI value was zero. From stage I to stage IIIA, GCI increased because gonads were full of sexual gametes, after spawning (stage IV) decreased to zero at the beginning of the other rest stage. These stages and GCI values are similar to Ensis arcuatus in Spain (Darriba et al., Reference Darriba, Sanjuan and Guerra2004). Maximum of GCI value in the reproductive cycle of Ensis arcuatus was about 0.23 and for Solen dactylus was about 0.3 (Darriba et al., Reference Darriba, Sanjuan and Guerra2004). In the present study a strong significant inverse relationship was found between gametogenic development and water temperature which agrees with other findings in different geographical locations (Arneri et al., Reference Arneri, Giannetti and Antolini1998; Darriba et al., Reference Darriba, Sanjuan and Guerra2004; Remacha & Anadon, Reference Remacha and Anadon2006) (Table 1). This species, like Solen strictus, from Hong Kong and Solen marginatus from Spain spawn only once a year (Gribben, Reference Gribben2005; Remacha & Anadon, Reference Remacha and Anadon2006). For intertidal bivalves, food and temperature are two important factors determining reproductive development (Darriba et al., Reference Darriba, Sanjuan and Guerra2004). In autumn phytoplankton bloom happens and the temperature begins to decrease, at that time this species enters the reproductive cycle like Ensis arquatus in Spain (Darriba et al., Reference Darriba, Sanjuan and Guerra2004) and Solen marginatus in Terron estuary in Spain (Remacha & Anadon, Reference Remacha and Anadon2006). For specimens like Solen dactylus which live in tropical places temperature's effect can play an important role in stimulating the clams to reproduce as a temperature shock. Spawning in Ensis arquatus took place in winter, like Solen dactylus, but in Solen marginatus occurred in May to July. However, in Solen dactylus, this stage took place in February. In the summer, most of the razor clams are in their sexual rest stage. For this species, like other razor clams, seven stages were identified (Darriba et al., Reference Darriba, Sanjuan and Guerra2004; Remacha & Anadon, Reference Remacha and Anadon2006). Solen dactylus is a gonochoric razor clam like Solen marginatus and Ensis arquatus. Gametogenic development in males and females was synchronized. Study on Solen dactylus in the Persian Gulf was necessary; because there is not an accurate pattern for fishery management of this species it has harmful effect on its stock and makes it endangered. Fishermen in this region can capture these clams easily without any controls. Provided that this condition continues, we will lose a vulnerable population in this region. To sum up, this study can help fishery managers to establish essential rules and a correct management in this area to conserve stocks of Solen dactylus.
ACKNOWLEDGEMENTS
Special thanks to E. Kamaly and K. Jokar and other people at fishery research laboratory of Bandar Abbas, The Persian Gulf and Oman Sea Ecology Research Center. Thanks to Dr M.R. Shokri at Shahid Beheshti University for critical review of an earlier version of this manuscript. Special Thanks to Dr M. King for the final review version of this manuscript.
