Growth and reproductive cycle of Donax trunculus L., (Mollusca: Bivalvia) off Faro, southern Portugal PDF

Preview

Summary

Fisheries Research 41 (1999) 309±316 Growth and reproductive cycle of Donax trunculus L., (Mollusca: Bivalvia) off Faro, southern Portugal M.B. Gaspara,*, R. Ferreirab, C.C. Monteiroa a Instituto de Investigac,aÄo das Pescas e do Mar/Centro Regional de Investigac,aÄo Pesqueira do Sul, Av. 5 de Outub...

Description

Fisheries Research 41 (1999) 309±316

Growth and reproductive cycle of Donax trunculus L., (Mollusca: Bivalvia) off Faro, southern Portugal M.B. Gaspara,*, R. Ferreirab, C.C. Monteiroa a

Instituto de Investigac,aÄo das Pescas e do Mar/Centro Regional de Investigac,aÄo Pesqueira do Sul, Av. 5 de Outubro s/n, 8700, OlhaÄo, Portugal b Universidade do Algarve, UCTRA, Campus de Gambelas, 8000, Faro, Portugal Received 15 December 1997; accepted 27 November 1998

Abstract The shell growth and reproductive cycle of Donax trunculus (Mollusca: Bivalvia) from southern Portugal were studied from December 1993 to November 1994. Both the acetate peel method and quanti®cation of surface growth rings were used to establish age and estimate growth rate. Widely spaced growth bands were formed in spring and summer whilst narrowly spaced bands were deposited in late summer and early autumn. No cessation of growth was observed during the winter. The von Bertalanffy growth curve estimated from the internal bands did not differ signi®cantly from that obtained from the surface rings. However, the age of clams can be determined more accurately from acetate peels than directly from surface rings. Laboratory experiments demonstrated that periodicity of the deposition of microgrowth bands was related to an innate rhythm. Gonadal development was monitored using standard histological methods. Both males and females showed synchronism in gonadal development. The gametogenic cycle began in late November and ended in late August. Spawning occurred between March and August with two major spawning phases: March and May±August. D. trunculus reached maturity during the ®rst year of life. # 1999 Elsevier Science B.V. All rights reserved. Keywords: Donax trunculus; Bivalvia; Growth; Age determination; Reproductive cycle; Age at ®rst maturity

1. Introduction Donax trunculus is an Atlantic±Mediterranean warm-temperate species. In the Atlantic it has been recorded from Senegal to the French coast (Tebble, 1966). It inhabits the high-energy environment of exposed sandy beaches, where it forms extensive, dense beds. It is the most inshore surf clam species in Portuguese waters, occurring to 6 m depth, with higher densities between 0 and 3 m depth. It is an *Corresponding author. Tel.: +351-89-700-503; fax: +351-89700-535; e-mail: [email protected]

excellent burrower, which makes it well adapted to life in the swash zone. Along the south coast of Portugal, D. trunculus is one of the most important commercial species exploited by an artisanal ¯eet. Actually a total of 60 boats are involved in this ®shery. These boats vary in length from 4 to 15 m, with engines of 17±150 Hp. Unfortunately no reliable statistics concerning catches are available since ®shermen were not obliged to declare their catches until 1998. However in 1997, a catch of 425 t of D. trunculus, worth PTE128 million was estimated. Despite its economic importance, its biology in this region has not hitherto been described.

0165-7836/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 7 8 3 6 ( 9 9 ) 0 0 0 1 7 - X

310

M.B. Gaspar et al. / Fisheries Research 41 (1999) 309±316

This study reports on the rate of growth, life span and age/size at ®rst maturity, of a population of Donax trunculus, off Faro, southern Portugal, by way of a comparison with other populations at different geographical locations.

To determine the periodicity of deposition of the growth bands, 30 living clams were marked with cold shock (Richardson et al., 1979) and then transferred to experimental cages in the laboratory. The experiment was extended over 28 days. The marking technique was designed to date mark one band in the shell so that all subsequent growth could be related to the mark.

2. Materials and methods 2.2. Reproduction Samples were collected by dredging off Faro, from December 1993 to November 1994, at approximately monthly intervals. 2.1. Growth Twenty shells of D. trunculus (size of 20±44 mm shell lengths) were measured for length and aged by shell-sectioning techniques (see Rhoads and Pannella, 1970; Richardson et al., 1979). An examination of the growth patterns visible in acetate peels revealed an alternating pattern of widely and narrowly spaced growth bands. In some regions of the shell, bands were so close together that they constituted a clearly marked darker coloration along the shell up to the umbo. To validate this growth pattern, so that it could be used for estimating age, we counted the number of bands within a given distance (0.5 mm) recently deposited in the growth margin (Richardson, 1987). The annual growth rate of each clam was determined by measuring the distance between the umbo and each annual ring using an ocular micrometer. Data from internal lines were converted into length using the length±height allometric relation: Lt

…mm† ˆ 1:6  H

…mm†1:02

The unembedded right valves of the clams were examined for the presence of surface annual rings; the length of each ring was measured using vernier callipers. Von Bertalanffy growth curves were ®tted to length-at-age data derived from surface shell rings and internal growth patterns, using the NLIN procedure of SAS (1985). This interactive curve ®tting procedure employs a non-linear least-square regression (the Gauss±Newton method) and estimates the von Bertalanffy growth constant k, asymptotic maximum L1 and age at length zero, t0 (Von Bertalanffy, 1938) Lt

…mm† ˆ L1 ‰1 ÿ expÿk…tÿt0 † Š:

The sex ratio of D. trunculus was established according to the colour of the gonad, violet in the females, yellowish-orange in the males. Sex ratio could not be determined from September to November as clams were in a resting stage than, making sexes indistinguishable macroscopically. The internal structure of the gonad was examined by histological preparation (Gaspar and Monteiro, 1998). Microscopic examinations of each section permitted assigning each specimen to a stage of gonadal development as described by Gaspar and Monteiro (1998): stage 0 ± inactive; stage I ± early active; stage II ± late active; stage III ± ripe; stage IV ± partially spawned; and stage V ± spent. The period of reproduction was de®ned as the time when the clams were in stage IV. To determine the age/length at ®rst maturity an additional sample of 200 individuals, shell length of 13±21 mm, was collected in April. All the individuals in stages III or IV were considered mature. In addition to this histological study, the evolution of the reproductive activity was observed through the ash-free dry weight/dry shell weight ratio (Walne and Mann, 1975) variations of a standard individual (27 mm). The use of a standard individual eliminates the results of growth, so that accumulation or loss of organic matter associated with the reproduction can be shown.

3. Results 3.1. Growth Fig. 1 illustrates acetate peels from sections through the shell of D. trunculus. According to Taylor et al. (1973) these shells are composed of three layers, an outer composite prismatic layer, middle crossed lamellar layer, and an inner homogeneous/complex crossed lamellar layer (Fig. 1(A)).

M.B. Gaspar et al. / Fisheries Research 41 (1999) 309±316

Fig. 1. D. trunculus. Photomicrographs of acetate peels of shell sections. (A) Shell layers: PL ± outer prismatic layer; CL ± middle crossed lamellar layer; I ± inner homogeneous/complex crossed lamellar layer; (B) growth bands (Gb) along the outer and middle shell layers; (C) first annual ring. Scale barsˆ100 m.

The growth bands in the outer layer run parallel to the growing edges of the shell, and appear as thin dark lines, with wider transparent regions, i.e. the growth increments, between (Fig. 1(B)). In some areas within the shell the width of the increments diminished progressively, resulting in a number of narrowly spaced bands clearly marked with a slight dark coloration which allowed the band to be traced through all layers to the umbo (Fig. 1(C)). We believe that these lines are related to long-term (annual) seasonal effects. Examination of the growing shell showed that

311

the period of rapid shell growth occurred during spring and summer, while the period of slow growth occurred in autumn especially during October, when the growth bands are deposited closer to each other. Fig. 2 shows the percentage of clams in monthly samples with rings present at the shell margin. By October a growth mark had been formed at or close to the shell margin in nearly 100% individuals examined. After October a transition occurs and clams with wider increments in shell margin predominate. Therefore, the narrowing of the bands and consequent ring formation occurring at the shell margin during late summer and early autumn is an annual growth line. Growth curves estimated from the measurements of the distance separating the surface growth rings and the internal bands are given in Fig. 3. Visual observation allows us to note that growth curves estimated from the internal bands are similar to those obtained from the surface rings. However, the growth rates estimated from the analysis of the surface growth rings are slower (kˆ0.40) than those determined from the annual narrowing of the internal microgrowth patterns observed in shell sections (kˆ0.508). The difference observed in this parameter might have resulted from the fact that it is impossible to distinguish externally disturbed rings from seasonal rings. Therefore, for the ®rst two years, when two very close hyaline rings were observed, it was assumed that the ®rst one was the annual ring. This assumption may have contributed for the slower growth rate estimated when the surface rings were used. Our laboratory results showed that fewer bands were produced than could be accounted for by simply daily/tidal periodicity. An average of 9.81.7 bands were counted, signi®cantly different from the expected daily/tidal frequency. The fact that fewer bands were observed than would have been expected con®rmed that the light±dark cycle is not important for band deposition in these species. 3.2. Reproduction Proportions of males and females in samples taken monthly are shown in Table 1. Males usually made up a greater proportion, and overall results, indicated a signi®cant departure from 1:1 sex ratio at 5% level (2-test). For all size classes, males outnumbered females.

312

M.B. Gaspar et al. / Fisheries Research 41 (1999) 309±316

Fig. 2. Seasonal variation in the number of annual rings at the shell margin determined from monthly collections of D. trunculus.

D. trunculus is gonochoric; no hermaphrodism or sex reversal was encountered. In this study both females and males showed a synchronism in the gametogenic evolution and spawning throughout the period studied. Thus, it did not appear necessary to do a separate analysis of the gametogenic cycle by sex. The gonadal cycle is illustrated in Fig. 4. The gametogenic cycle had already began in December 1993 as

58% of the individuals were in the early active stage, the remainder (42%) were in a late active stage. The late active stage dominated the samples of January and February. Partially spawned clams (75%) comprised the bulk of the sample in March. Spawning activity continued until August when 80% of the clams were spent. In September and October the entire population was in an inactive stage and by the end of November a

Fig. 3. A comparison of the growth rate of D. trunculus determined from (A) acetate peel replicas of shell sections and (B) external surface growth rings.

313

M.B. Gaspar et al. / Fisheries Research 41 (1999) 309±316 Table 1 Distribution of sexes of D. trunculus by age classes during the studied period (Mˆmales; Fˆfemale) Month

Age at length 42 F

M

Total F

35

10

45

0

0

M

F

100 90 60 60 100 110 65 75 10 0 0 0

40 40 80 55 115 95 55 60 10 0 0 0

670

550

new gametogenic cycle began. It is important to note that during the spawning period gametes within follicles were observed in several stages of maturation. Both the condition index and the degree of gonadal development showed the same pattern of variation (Fig. 5). A steady increase in the index of the population was observed after the initiation of gametogenesis in early December. This increasing trend continued until the beginning of spawning. The condition index decreased sharply as the percentage of spawning individuals reached maximum. During the inactive stage the condition of specimens increased slightly until October, when there was a decrease in the index. Between February and August the condition index

showed an occurrence of two main spawning periods: in March and from May±August. In the study of gonadal status in 200 small individuals (ranging from 13 to 21 mm) the gonads of all specimens were differentiated, indicating that sexual maturity was reached during the ®rst year of the life.

Fig. 4. Percentage of individuals of D. trunculus in each phase of the reproductive cycle during each month of this study.

Fig. 5. D. trunculus: changes of the condition index of a standard animal of 27 mm length during the studied period.

4. Discussion 4.1. Growth Rapid shell growth occurred in spring and summer, when wide microgrowth increments were deposited,

314

M.B. Gaspar et al. / Fisheries Research 41 (1999) 309±316

whilst slow growth occurred during late summer and early autumn with the deposition of narrow growth increments and the formation of a well-de®ned annual growth ring. Surprisingly, we did not note a period of cessation of growth through the winter months. In fact, D. trunculus renewed growth in November which then continued until the following October. This is a different pattern of seasonal growth than that found in other Donax populations in Europe. The relation between growth patterns and environmental factors was not studied. However, the slowing of growth and the deposition of the annual growth ring in D. trunculus might be related to the decreasing of water temperatures. The seasonal pattern of the sea surface temperature on the south coast of Portugal is characterized by low values during mid-winter, increasing in late winter, stabilizing during the summer and decreasing in early autumn. Therefore, it is probable that the deposition of the annual growth ring in D. trunculus in this region might be related to the onset of the decreasing of water temperature (in late September, early October). The estimated growth curves from the surface growth rings and internal growth patterns show that the two methods provide different estimates of growth rates for the ®rst two years. The age of clams can be determined more accurately from acetate peels than directly from surface rings. The growth curve estimated in our study indicated a rapid growth during the ®rst two years of life, reaching the minimum legal size of ®shing (25 mm) in about one year. This work is the ®rst attempt to estimate growth parameters for Atlantic populations using the acetate peel method. For Mediterranean populations only RamoÂn et al. (1995) have used this method to estimate the age of D. trunculus. All other studies on the growth

of this species were carried out using the size frequency analysis. The analysis of both asymptotic maximum length (L1) and von Bertalanffy growth constant (k), included in Table 2, suggests that these two growth parameters are not affected by latitude. Thus, the differences observed in these parameters are probably due to some extent to the growing conditions of the environment. On the contrary, life span is related to latitude; life span decreasing with latitude: 5 years along French Atlantic coast, 2±3 years on the Moroccan coast and Mediterranean. Laboratory experiments carried out in the present study showed that the production of growth bands is related to an endogenous rhythm of shell formation rather than some exogenous factors, i.e. diurnal ¯uctuations, tide, or current ¯ow. Richardson (1991) and Gaspar et al. (1995) showed in a series of experiments with continuously immersed clams that the total number of bands in the shell was correlated with the growth of the shell rather than with daily or tidal control. Therefore, bands cannot be used in this species to measure tidal or daily growth rates. 4.2. Reproduction The sex ratio in D. trunculus population from southern Portugal was found to be in favour of males in most of the years. Lucas (1965), Badino and Marchionni (1972), and MoueÈza and Frenkiel-Renault (1973) mentioned that males of this species usually made up a slightly higher proportion but they did not show a statistically signi®cant difference from a sex ratio of 1:1. Sex ratios may also be a function of the size classes of bivalve populations. In the present study we found that the proportion of males increased. This might be explained as differential growth or

Table 2 Von Bertalanffy growth curve parameters estimated for different areas throughout geographical range of D. trunculus Area

L1

k

t0

Author

Atlantic (France) Atlantic (France) Atlantic (Spain) Atlantic (Morrocco) Atlantic (Portugal) Mediterranean (France) Mediterranean (Spain)

48.9 35.5 52.8 42.5 47.3 35.99 41.8

0.38 0.79 0.55

0.29 0.52

0.58 0.96 0.71

0.52 0.67 0.35

Guillou and Moal (1980) Ansell and LagardeÂre (1980) Maze and Laborda (1988) Bayed and Guillou (1985) Present study Bodoy (1982) RamoÂn et al. (1995)

M.B. Gaspar et al. / Fisheries Research 41 (1999) 309±316

differential mortality probably as a result of spawning effort. We observed that the gametogenic cycle in D. trunculus began in late November and the spawning occurred between late March and late August. The analysis of the condition index used in this study suggests that spawning in D. trunculus has two peaks: in March and May±August. MoueÈza and FrenkielRenault (1973), and Ansell et al. (1980) found, at Azur Plage, that spawning occurred in summer. Ansell and Bodoy (1979) in Camargue, reported that this species spawned from April±September, with two massive periods of gamete release. Similar observations on the reproductive cycle of this species were made by Badino and Marchionni (1972), for the Mediterranean, and by Lucas (1965) and Poli (1972) for the French Atlantic coast. At Medhia, Moroccan Atlantic coast, D. trunculus spawns between March and August, in two phases: spring and summer (Bayed, 1990). Thus, the reproduction period is very similar throughout the range of this species. The gametogenic development of this species indicated a continuous spawning activity. The presence of gametes at all stages of maturity indicates the occurrence of successive individual spawning. This may also indicate an existence of an endogenous mechanism that delays the development of some gametes. Thus, we believe that the objective of this reproductive strategy is to extend the spawning period. We have found that D. trunculus reached maturity during the ®rst year of life. In fact all the individuals examined (from 13 to 21 mm) showed differentiated gonads, suggesting that sexual maturity is a function of age, not size. Only MoueÈza and Frenkiel-Renault (1973) reported the length of ®rst maturity of D. trunculus, recording this on the Algerian coast as 16 mm.

Acknowledgements We would like to thank three anonymous referees whose suggestions greatly improved the manuscript. We also thank the staff of IPIMAR/CRIPSul for collecting the samples and the crew of NI/DONAX for their skilful handling of the boat and ®shing gears. M.B. Gaspar was funded by JNICT ± PROGRAMA CIEÃNCIA and PRAXIS XXI.

315

References Ansell, A.D., Bodoy, A.,...