Abstract

In order to test the habitat harshness hypothesis (HHH) the reproductive biology of Donax hanleyanus was studied histologically, comparing populations from three beaches with contrasting morphodynamics (dissipative, intermediate and reflective) over 25 months. The reproductive phase of D. hanleyanus was extended at the reflective beach compared to the other two. Males and females from the dissipative and intermediate beaches were significantly smaller and had lower biomass at maturity than those at the reflective beach. Recruits were significantly more abundant and the recruitment period was extended significantly at the dissipative beach. Spawning events took place twice each year at the dissipative (early spring and spring-summer) and the intermediate beach (winter and summer), whereas continuous gamete releases were noted at the reflective beach. Size and biomass at first maturity were lower at the dissipative beach, whereas monthly mean abundance of D. hanleyanus was higher at the reflective beach. The gametogenic cycle correlated significantly with sea-surface temperature, relative spermatozoon abundance, condition index, ash-free shell-free dry mass, and mean size and abundance of oocytes, for all three populations. At the population level, many of these reproductive responses to physical variables were opposite to those predicted by the HHH, including: greater abundance, extended reproductive cycle, extended period with spawning individuals, and larger size and higher biomass at first maturity at the reflective beach. This suggested that the hypothesis of habitat safety (HHS), originally proposed for supralittoral species, may be extended to intertidal species; a combination of narrow swashes and steep slopes makes reflective beaches a safer and more stable environment for intertidal species such as D. hanleyanus.

INTRODUCTION

Macrobenthic communities populating exposed sandy beaches demonstrate clear increases in ecological diversity, species richness, abundance and biomass from reflective to dissipative beach morphotypes (e.g. McLachlan, 1990; Ricciardi & Bourget, 1999; Defeo, Lercari & Gomez, 2003). In physically controlled environments such as sandy beaches, communities are structured by constituent species responding independently to the physical environment rather than by biological interactions, giving strong support to the ‘autoecological hypothesis’ (AH) (Noy-Meir, 1979). In agreement with this hypothesis, but restricted to the intertidal of sandy beaches, the ‘swash exclusion hypothesis’ (SEH) (McArdle & McLachlan, 1991, 1992) predicts a consistent increase in species richness, abundance and biomass from reflective to dissipative beaches. Furthermore, previous studies have shown that organisms on harsh reflective beaches need to invest more energy in maintenance processes than in growth and reproduction (Gómez & Defeo, 1999; Defeo, Gomez & Lercari, 2001). Defeo et al. (2001, 2003) combined the community level AH and SEH, to postulate the ‘habitat harshness hypothesis’ (HHH). The HHH predicts that (1) at the community level, reflective beaches will exhibit lower species richness, diversity and abundance, while (2) at the population level, they will be characterized by lower abundance, growth, fecundity, reproductive output and higher mortality rates. However, other recent investigations suggest that populations co-occurring on sandy beaches with a range of contrasting morphodynamics do not conform consistently to these predictions.

Veloso & Cardoso (2001) demonstrated no significant changes in abundance of macrobenthic communities between intermediate and reflective beaches. At the population level Defeo et al. (1997) recorded that the isopod Excirolana braziliensis exhibited higher abundance at a reflective beach than at a dissipative one, a finding that was confirmed by Defeo & Martínez (2003). In further contrast to the predictions of the HHH, the amphipod Pseudorchestoidea brasiliensis presented greater abundance, egg production potential and recruitment levels and lower natural mortality at a reflective beach (Gómez & Defeo, 1999). Following a 2-year study of seven Uruguayan sandy beaches with contrasting morphodynamics, the same authors recently demonstrated that the amphipod Atlantorchestoidea brasiliensis shows clear population responses to physical variables in direct opposition to those predicted by the HHH, including an increase in abundance and individual size from dissipative to reflective beaches (Defeo & Gómez, 2005). Furthermore, abundances of the decapod Emerita brasiliensis did not vary between dissipative and reflective beaches and at the latter beach type displayed higher male growth rates as well as lower natural mortality (Defeo et al., 2001). In summary, these studies suggest that beach morphodynamics might not be the primary factor affecting abundance, population dynamics and life history traits of macrobenthic species. However, the HHH has not been tested comprehensively with respect to reproductive biology. The only investigation dealing with this subject to date was a 13-month study in which 150 wedge clams from two different beach morphotypes in Uruguay were examined. The results in that instance appeared to confirm the HHH (Delgado & Defeo, 2007).

The wedge clam Donax hanleyanus Philippi, 1847 is a numerically dominant intertidal warm-temperate species on South American sandy beaches and is distributed from tropical (17°S Caravelas, Brazil) to temperate regions (37°S Punta Mogotes in Mar del Plata, province of Buenos Aires) (de Castellanos & Fernandez, 1965; Penchaszadeh & Olivier, 1975; Narchi, 1978; Cardoso & Veloso, 2003). Wedge clams inhabit a range of Argentinean intertidal habitats (Marcomini et al., 2002; Herrmann, 2009), providing an opportunity to assess responses to contrasting morphodynamic regimes. The present study tests the HHH at the population level, comparing the reproductive biology of D. hanleyanus from three Argentinean sandy beach habitats, one dissipative, one intermediate and one reflective. Following predictions of the HHH (Defeo et al., 2001, 2003), we sought to determine whether D. hanleyanus at the dissipative beach does indeed exhibit greater abundance of recruits, larger size at first sexual maturity, larger maximum individual size and mass, and extended periods of reproduction, recruitment and spawning.

MATERIAL AND METHODS

Study area

The reproductive biology of Donax hanleyanus was studied in the Province of Buenos Aires at Santa Teresita (36°32′S, 56°41′W), Mar de las Pampas (37°19′S, 57°00′W) and Faro Querandí (37°29′S, 57°07′W); the last named locality is the southernmost extent of the species' distribution. These three open ocean sandy beaches show contrasting morphodynamics and flow continuously into one another with a north–south shoreline orientation, which is stable in the long term (Marcomini & López, 1993).

Abiotic factors

According to McLachlan's (1980) rating scale for exposure and Short & Wright's (1983) classifications of beach types, Santa Teresita is sheltered and dissipative, Mar de las Pampas exposed and intermediate, and Faro Querandí exposed and reflective (Table 1). The three sampling sites receive continuous wave action and are subject to semidiurnal tides, with a maximum tidal range of 1.6 m, a spring tide average of 1.7 m and a neap tide mean of 0.2 m. The sea-surface temperature (SST) (mean ± SE) varies between 11 ± 0.14°C in winter and 23 ± 0.21°C in summer. The study sites are composed of fine (Santa Teresita), medium (Mar de las Pampas) and coarse sands (Faro Querandí), with a mean particle diameter of 0.21, 0.37 and 0.48 mm, respectively (Table 1). All three beaches are affected by freshwater seepage, as confirmed by satellite images, and a southward current bringing water masses from the estuary of the Río de la Plata, 290 km long and up to 220 km wide. Mean salinity ranges between 31‰ and 34‰. All three beaches are well drained and oxygenated.

Table 1.

Characterization of physical and biological attributes of the three studied localities on north Argentinean Atlantic coast.

Beach features Santa Teresita Mar de las Pampas Faro Querandí 
Latitude 36°32′S 37°19′S 37°29′S 
Longitude 56°41′W 57°00′W 57°07′W 
Beach width (m) <80 <70 <100 
Intertidal zone width (m) <70 <60 <60 
Tidal range (m) 1.8 1.7 1.7 
Mean grain size (phi/mm)* 2.26/0.21 1.43/0.37 1.05/0.48 
Median grain size (phi)* 2.28 1.39 0.99 
Sorting*,†,‡ Good (0.36) Moderate (0.68) Moderate (0.51) 
Skewness*,† −0.11 +0.07 −0.11 
Textural group§ Fine Medium Coarse 
Mean slope of intertidal (%) 1/43 1/16 1/14 
Exposure Sheltered Exposed Exposed 
Morphodyn. type§ Dissipative Intermediate Reflective 
Dean parameter (Ω)#,** 1.92–2.53 1.50–1.86 3.28–3.93 
D. hanleyanus belt (m) 30 12 10 
Macrofauna richness (species) 
Beach features Santa Teresita Mar de las Pampas Faro Querandí 
Latitude 36°32′S 37°19′S 37°29′S 
Longitude 56°41′W 57°00′W 57°07′W 
Beach width (m) <80 <70 <100 
Intertidal zone width (m) <70 <60 <60 
Tidal range (m) 1.8 1.7 1.7 
Mean grain size (phi/mm)* 2.26/0.21 1.43/0.37 1.05/0.48 
Median grain size (phi)* 2.28 1.39 0.99 
Sorting*,†,‡ Good (0.36) Moderate (0.68) Moderate (0.51) 
Skewness*,† −0.11 +0.07 −0.11 
Textural group§ Fine Medium Coarse 
Mean slope of intertidal (%) 1/43 1/16 1/14 
Exposure Sheltered Exposed Exposed 
Morphodyn. type§ Dissipative Intermediate Reflective 
Dean parameter (Ω)#,** 1.92–2.53 1.50–1.86 3.28–3.93 
D. hanleyanus belt (m) 30 12 10 
Macrofauna richness (species) 

After Inman (1952).

#After Dean (1973).

**Calculated for specific seasonal SST from 9°C to 25°C.

In order to characterize the physical parameters of the Donax belt, particle size analyses from all studied sites were carried out by sampling sediments to a depth of 10 cm with a plastic corer of 3.5 cm diameter. Sand samples were washed with freshwater overnight to remove salt and then dried at 70°C. Afterwards, any shell fragments were extracted and the remaining part of the samples was analysed using a MacroGranometer™ settling tube (e.g. Flemming & Thum, 1978; Flemming & Ziegler, 1995) and the SedVar™ V6.2p software package (Brezina, 1997), which is part of the system. The data processing software makes use of the more versatile equation of Brezina (1979) rather than that of Gibbs Matthews & Link (1971), which is applicable only to smooth glass spheres to calculate settling velocities. All textural parameters presented in this study were calculated using the percentile statistics of Inman (1952), while sediments were classified according to the Wentworth scale (1922).

Wave height was approximated by measuring the height of breaking waves (n = 10) with graduated poles against the horizon, and adding the result to the height difference between the location of the observer and the lowest point at which the backwash met the next incoming swash bore. The wave period was measured as the time interval between breakers (N = 50). The morphodynamic state of each site studied was described by the Dean parameter (Ω) (Dean, 1973):  

Eq.1
formula
which is based on mean wave height Hb (m) divided by wave period T (s) and sand fall velocity W (m/s). The slope of the beach face was measured by the height difference (Emery, 1961) between the drift and the water line. The swash period was estimated according to McArdle & McLachlan (1991). Salinity (Optech portable refractometer, model RSM) as well as the SST (hand-held digital thermometer) were measured monthly in situ at the three sample locations. For additional detailed information the SST was measured daily at Santa Teresita between October 2005 and December 2006 (Herrmann et al., 2008a) and was supplemented with data of the Argentinean Marine Institute (CEADO: Centro Argentino de Datos Oceanográficos, Servicio de Hidrografía Naval, Buenos Aires, Argentina), which operates a permanent weather station at this beach.

Sampling and histological examination

Following the systematic approach described by Herrmann (2009), a total of 22,519 D. hanleyanus were collected from the three study beaches between December 2004 and December 2006. Quantitative sampling of clams was carried out at monthly intervals (Santa Teresita and Mar de las Pampas: December 2004 to December 2006; Faro Querandí: March 2005 to December 2006) at a series of survey stations set at 4 m intervals along three transects. The transects ran perpendicular to the shoreline between the spring tide high water mark and the spring tide low water mark and were separated by 20 m intervals. At each station, three replicate sand samples (40 × 40 cm) were excavated to 35 cm depth using a 0.16 m2 steel corer, and sieved individually on a 1-mm mesh. Specimens of D. hanleyanus were measured to the nearest 0.1 mm with a digital vernier calliper and returned alive to their habitat. For biomass determinations, as well as for calculations of the condition index (CI), 2,205 specimens were preserved in 70% ethanol and subsequently analysed in the laboratory. For histological examinations a total of 35 clams, covering the full range of antero-posterior shell lengths (apSL) (Santa Teresita: 5–28 mm, Mar de las Pampas: 9–38 mm and Faro Querandí: 9–39 mm), was collected monthly from the three beaches (Ntotal = 2,275 specimens). Clams with severed adductor muscles were fixed in Bouin's solution for 2 h, then transferred to 70% ethanol and later processed in the laboratory using standard histological methods, i.e. embedding in paraffin, sectioning at 5 µm and staining with haematoxylin–eosin (following Howard et al., 2004). Gonads were examined with a light microscope and assigned to one of five developmental stages [rest, pre-active (Fig. 1A, E), active (Fig. 1B, F), spawning (Fig. 1C, G) and cytolysed (Fig. 1D, H); Table 2]. Images of each sample were captured using a Sound Vision digital camera and processed using the imaging software AxioVision version 4.4 (2008). For all developmental stages except sexual rest, the mean oocyte diameter was determined by measuring 30 oocytes per specimen (Ntotal = 17,286 oocytes measured). In addition, the abundance of oocytes in each sectioned female gonad was examined by counting oocytes from 1 mm2 surface area (Ntotal = 15,450 oocytes counted). SST was measured daily (at 13.00) at Santa Teresita using a digital thermometer with a precision of 0.1°C. Since monthly SSTs taken at Mar de las Pampas and Faro Querandí did not vary from those of Santa Teresita (one-way ANOVA, F2,69 = 0.089, P > 0.05), daily SSTs from the latter were used for all three beaches.

Table 2.

Qualitative descriptions of histological preparations of Donax hanleyanus gonad used to assess gametogenic stages.

Stage Definition Brief description of gonad Figure 1 
Sexual rest Follicles few and small, sex indistinguishable, protogonia and gonia in mitosis – 
Pre-active Reproductive material is scant and intersected by transverse muscular fascicle cells, alveoli appear loosely arranged, phagocytes are common A, E 
II Active Germ cells are in various stages of development and fill the alveoli, alveoli are large with complete and closed walls B, F 
III Spawning Clear loss of gametes, reproductive material varies in quantity but it is fairly abundant, alveolar pattern is disturbed, walls are broken and alveoli appear flattened C, G 
IV Cytolysed Reproductive material is completely degenerated; alveoli are very small and wide apart, massive numbers of phagocytic cells D, H 
Stage Definition Brief description of gonad Figure 1 
Sexual rest Follicles few and small, sex indistinguishable, protogonia and gonia in mitosis – 
Pre-active Reproductive material is scant and intersected by transverse muscular fascicle cells, alveoli appear loosely arranged, phagocytes are common A, E 
II Active Germ cells are in various stages of development and fill the alveoli, alveoli are large with complete and closed walls B, F 
III Spawning Clear loss of gametes, reproductive material varies in quantity but it is fairly abundant, alveolar pattern is disturbed, walls are broken and alveoli appear flattened C, G 
IV Cytolysed Reproductive material is completely degenerated; alveoli are very small and wide apart, massive numbers of phagocytic cells D, H 

Individual stages of male and female gonads are shown in Figure 1.

Figure 1.

Microphotography of male (A–D) and female (E–H) gonad stages of Donax hanleyanus: pre-active (A, E), active (B, F), spawning (C, G) and cytolysed (D, H). Abbreviations: aw, alveolar wall; al, alveolus; sp, spermatozoa; sd, spermatids; tf, transverse fibre. Scale bars = 100 µm. This figure appears in colour in the online version of Journal of Molluscan Studies.

Figure 1.

Microphotography of male (A–D) and female (E–H) gonad stages of Donax hanleyanus: pre-active (A, E), active (B, F), spawning (C, G) and cytolysed (D, H). Abbreviations: aw, alveolar wall; al, alveolus; sp, spermatozoa; sd, spermatids; tf, transverse fibre. Scale bars = 100 µm. This figure appears in colour in the online version of Journal of Molluscan Studies.

Data analysis

Estimation of relative spermatozoon abundance

Relative spermatozoon abundance (RSA) is a measure of male gonad activity. RSA values were used to simplify the classification of male gonadal tissue and to analyse seasonal variations in the gametogenic cycle. RSA was calculated as:  

Eq.2
formula
where the grey value per mm2 was measured from greyscale digital images of histological sections using the digital imaging software package Adobe Photoshop version CS3 Extended (2008). The grey value is equal to the brightness of pixels in a digital image, commonly expressed in integers ranging from 0 (black) to 255 (white) on an 8-bit digital signal.

Condition index, biomass and recruitment

CI was calculated to detect seasonal variations in the mass of the visceral mass of D. hanleyanus. The total shell-free, wet mass (SFWM) of each ethanol-preserved clam was recorded to the nearest 0.1 mg immediately after dabbing on blotting paper. Mantle, siphons, retractor and adductor muscles were then removed, and the SFWM of the resulting unit comprising the visceral mass and foot was recorded. Following the equation of de Villiers (1975), CI was calculated as:  

Eq.3
formula
where Mv is the wet ethanol-preserved visceral mass (including the foot) and Mt is the total ethanol-preserved SFWM. Additionally the ash-free, shell-free dry mass (AFDM) was estimated as an indicator of clam condition, using the conversion factor 0.186 provided by Brey, Rumohr & Ankar (1988).

Clams were measured monthly at each beach (data from Herrmann, 2009) and a pooled 2-year length–frequency distribution was plotted for each of the three populations (‘population’ in this paper refers to all specimens of D. hanleyanus inhabiting the geographic location without any genetic implication). On the basis of the histological results, these length–frequency distributions represented three discrete ontogenetic stages: (1) recruits (<11 mm), (2) juveniles (11–22 mm, the size class where sex can be differentiated for the first time) and (3) adults (>22 mm, size where individuals were 100% mature).

Size and biomass at sexual maturity

The size at which 100% of the population is mature was estimated from the proportion of mature females and males, respectively, in different size classes, using the logistic equation of McGullagh & Nelder (1997):  

Eq.4
formula
where BapSL is the proportion of females and males, respectively, with mature gonads in each size class (apSL), and a, b and x0 are parameters. The biomass at which 50% of the population is mature was estimated using to the same formula, whereby BSFWM is the proportion of females and males, respectively, with mature gonads in each biomass class (SFWM).

Mature clams were allocated to the developmental stages ‘active’ and ‘spawning’, while immature individuals were deemed to be in the sexual rest stage. Equation (3) was fitted by non-linear least squares, using the quasi-Newton algorithm of the software package SigmaPlot version 11 (2008) to estimate the standard error of parameters. Size at 50% population maturity (apSL50%) was estimated as:  

Eq.5
formula

and the biomass at which 100% of the population is mature (SFWM50%) was estimated accordingly.

Statistical analysis

Sex ratios (male:female) of D. hanleyanus were calculated according to the presence of oocytes and spermatozoa using chi-square analysis. The influence of SST on the gametogenic cycle, as well as its relationship with RSA, CI, AFDM, abundance and size of oocytes within the population inhabiting each of the sampled beaches were statistically analysed by Spearman's rank order correlation. Spatial and temporal differences in the gametogenic cycle and oocyte abundance as well as mean and modal sizes of oocytes were tested by one-way analysis of variance (ANOVA). For all beaches the three ontogenetic groups were tested for differences by two-way ANOVA using the factors ‘beach’ and ‘ontogenetic group’. Based on the Durbin–Watson coefficient, residuals of the logistic functions of size and biomass at sexual maturity were tested for autocorrelation. The closer the coefficient to value ‘2’ (within the range 0–4), the less significant the autocorrelation (SigmaStat, 2004). To compare results of size and biomass at 50% population maturity, as estimated for both sexes and for each of the three beaches, analyses of the residual sum squares (ARSS) were performed (Chen, Jackson & Harvey, 1992). All statistical analyses were carried out using the statistical package SPSS version 16.0.1 (2007). Differences were considered significant at the level of α = 5% (Zar, 1999).

RESULTS

Abiotic factors

Results of grain size and sand fall velocity analyses, as well as associated parameters are summarized in Table 1.

Gametogenic cycle

Histological examinations of gonadal tissue from a 25-month sampling series demonstrated that sex ratios did not significantly deviate from 1:1 at the intermediate and reflective beaches (Table 3a), but that there was a greater proportion of females at the dissipative beach in December 2004 (χ2 = 8.758, df = 1, P < 0.05), January 2005 (χ2 = 11.765, df = 1, P < 0.05) and April 2006 (χ2 = 4.571, df = 1, P < 0.05). No case of hermaphroditism was found. Histological analysis revealed that the reproductive cycle of both male and female Donax hanleyanus undergoes a distinct seasonality at all three beaches (Fig. 1; Table 3b). Sexual activity was detectable mainly during spring and summer (Fig. 2; Table 3t) at all three beaches. At the dissipative beach, the spawning period of wedge clams was restricted mainly to spring and summer, while gamete release continued all year round at the intermediate (except in July, August 2005 and July, December 2006) and reflective beach (except July, August 2005 and August, November, December 2006) populations, respectively (Fig. 2). In both years of the study, cytolysed specimens dominated the population in autumn and winter at all three beaches (Fig. 2; Table 3u). Wedge clams of indistinguishable sex were found at the reflective and intermediate beaches during all months except during summer, while a high proportion of individuals in the sexual rest stage appeared at the dissipative site in early autumn 2005 and in late summer 2006 (Fig. 2; Table 3v).

Table 3.

Results of statistical analysis of the gametogenic cycle of Donax hanleyanus in relation to abiotic and biotic factors at the three beaches.

 Parameter Statistical results at the beaches
 
  Santa Teresita (dissipative) Mar de las Pampas (intermediate) Faro Querandí (reflective) 
Sex ratio (females:males)* For values see text χ2 = 0.718, df = 1, P> 0.397 χ2 = 1.029, df = 1, P= 0.310 
Gametogenic cycle vs time course F11,9=5.644, P= 0.007 F11,10 = 15.595, P= 0.000 F11,10 = 8.861, P= 0.001 
Gametogenic cycle vs SST rs= 0.726, P= 0.000, n = 21 rs= 0.787, P= 0.000, n = 22 rs= 0.799, P= 0.000, n = 22 
Gametogenic cycle vs CI rs = -0.556, P= 0.025, n = 16 rs= 0.480, P= 0.024, n = 22 rs= 0.575, P= 0.005, n = 22 
Gametogenic cycle vs AFDM rs= 0.557, P= 0.016, n = 18 rs= 0.669, P= 0.001, n = 22 rs= 0.445, P= 0.043, n = 21 
Gametogenic cycle vs mean size of oocytes rs= 0.721, P= 0.004, n = 14 rs= 0.501, P= 0.021, n = 21 rs= 0.443, P= 0.039, n = 22 
Gametogenic cycle vs abundance of oocytes rs= 0.708, P= 0.000, n = 21 rs= 0.779, P= 0.000, n = 22 rs= 0.639, P= 0.001, n = 22 
CI vs SST rs= 0.507, P= 0.032, n = 18 rs= 0.337, P = 0.125, n = 22 rs= 0.673, P= 0.001, n = 22 
AFDM vs SST rs= 0.574, P= 0.016, n = 17 rs= 0.600, P= 0.003, n = 22 rs= 0.581, P= 0.005, n = 22 
Oocyte mean size vs SST rs= 0.741, P= 0.002, n = 14 rs= 0.334, P = 0.139, n = 21 rs= 0.448, P= 0.037, n = 22 
Abundance of oocytes vs SST rs= 0.504, P= 0.020, n = 21 rs= 0.743, P= 0.000, n = 22 rs= 0.701, P= 0.000, n = 22 
Oocyte modal size vs seasons (spawning) F1,12 = 10.154, P= 0.008 F1,19 = 4.777, P= 0.042 F1,12 = 5.680, P= 0.027 
Oocyte abundance increases vs seasons F1,19 = 10.864, P= 0.004 F1,20 = 6.214, P= 0.022 F1,20 = 5.609, P= 0.028 
Oocyte abundance decreases vs seasons F1,19 = 10.142, P= 0.005 F1,20 = 9.346, P= 0.006 F1,20 = 14.107, P= 0.001 
Settling period of recruits vs SST rs = -0.327, P = 0.119, n = 24 rs = -0.123, P = 0.584, n = 22 rs = -0.356, P = 0.104, n = 22 
Appearances of recruits F1,22 = 224.266, P= 0.000 F1,20 = 16.064, P= 0.001 F1,20 = 25.997, P= 0.000 
Gametogenic cycle vs RSA rs = 0.804, P= 0.000, n = 21 rs = 0.856, P= 0.000, n = 22 rs = 0.822, P= 0.000, n = 22 
RSA increases vs seasons F1,19 = 6.889, P= 0.017 F1,20 = 11.661, P= 0.003 F1,20 = 35.507, P= 0.000 
RSA decreases vs seasons F1,19 = 15.366, P= 0.001 F1,20 = 30.139, P= 0.000 F1,20 = 55.621, P= 0.000 
Gonad in active and spawning stage F1,40 = 33.454, P= 0.000 F1,42 = 8.032, P= 0.007 F1,42 = 17.539, P= 0.000 
Gonad in cytolysed stage F1,19 = 17.898, P= 0.000 F1,20 = 4.542, P= 0.046 F1,20 = 9.767, P= 0.005 
Gonad in sexual rest stage F1,19 = 46.143, P= 0.000 F1,20 = 5.301, P= 0.032 F1,20 = 4.484, P= 0.047 
 Parameter Statistical results at the beaches
 
  Santa Teresita (dissipative) Mar de las Pampas (intermediate) Faro Querandí (reflective) 
Sex ratio (females:males)* For values see text χ2 = 0.718, df = 1, P> 0.397 χ2 = 1.029, df = 1, P= 0.310 
Gametogenic cycle vs time course F11,9=5.644, P= 0.007 F11,10 = 15.595, P= 0.000 F11,10 = 8.861, P= 0.001 
Gametogenic cycle vs SST rs= 0.726, P= 0.000, n = 21 rs= 0.787, P= 0.000, n = 22 rs= 0.799, P= 0.000, n = 22 
Gametogenic cycle vs CI rs = -0.556, P= 0.025, n = 16 rs= 0.480, P= 0.024, n = 22 rs= 0.575, P= 0.005, n = 22 
Gametogenic cycle vs AFDM rs= 0.557, P= 0.016, n = 18 rs= 0.669, P= 0.001, n = 22 rs= 0.445, P= 0.043, n = 21 
Gametogenic cycle vs mean size of oocytes rs= 0.721, P= 0.004, n = 14 rs= 0.501, P= 0.021, n = 21 rs= 0.443, P= 0.039, n = 22 
Gametogenic cycle vs abundance of oocytes rs= 0.708, P= 0.000, n = 21 rs= 0.779, P= 0.000, n = 22 rs= 0.639, P= 0.001, n = 22 
CI vs SST rs= 0.507, P= 0.032, n = 18 rs= 0.337, P = 0.125, n = 22 rs= 0.673, P= 0.001, n = 22 
AFDM vs SST rs= 0.574, P= 0.016, n = 17 rs= 0.600, P= 0.003, n = 22 rs= 0.581, P= 0.005, n = 22 
Oocyte mean size vs SST rs= 0.741, P= 0.002, n = 14 rs= 0.334, P = 0.139, n = 21 rs= 0.448, P= 0.037, n = 22 
Abundance of oocytes vs SST rs= 0.504, P= 0.020, n = 21 rs= 0.743, P= 0.000, n = 22 rs= 0.701, P= 0.000, n = 22 
Oocyte modal size vs seasons (spawning) F1,12 = 10.154, P= 0.008 F1,19 = 4.777, P= 0.042 F1,12 = 5.680, P= 0.027 
Oocyte abundance increases vs seasons F1,19 = 10.864, P= 0.004 F1,20 = 6.214, P= 0.022 F1,20 = 5.609, P= 0.028 
Oocyte abundance decreases vs seasons F1,19 = 10.142, P= 0.005 F1,20 = 9.346, P= 0.006 F1,20 = 14.107, P= 0.001 
Settling period of recruits vs SST rs = -0.327, P = 0.119, n = 24 rs = -0.123, P = 0.584, n = 22 rs = -0.356, P = 0.104, n = 22 
Appearances of recruits F1,22 = 224.266, P= 0.000 F1,20 = 16.064, P= 0.001 F1,20 = 25.997, P= 0.000 
Gametogenic cycle vs RSA rs = 0.804, P= 0.000, n = 21 rs = 0.856, P= 0.000, n = 22 rs = 0.822, P= 0.000, n = 22 
RSA increases vs seasons F1,19 = 6.889, P= 0.017 F1,20 = 11.661, P= 0.003 F1,20 = 35.507, P= 0.000 
RSA decreases vs seasons F1,19 = 15.366, P= 0.001 F1,20 = 30.139, P= 0.000 F1,20 = 55.621, P= 0.000 
Gonad in active and spawning stage F1,40 = 33.454, P= 0.000 F1,42 = 8.032, P= 0.007 F1,42 = 17.539, P= 0.000 
Gonad in cytolysed stage F1,19 = 17.898, P= 0.000 F1,20 = 4.542, P= 0.046 F1,20 = 9.767, P= 0.005 
Gonad in sexual rest stage F1,19 = 46.143, P= 0.000 F1,20 = 5.301, P= 0.032 F1,20 = 4.484, P= 0.047 

Bold indicates significant correlation (P < 5%).

*Chi-square test.

One-way ANOVA.

Spearman's rank order correlation.

Figure 2.

Gametogenic cycle of Donax hanleyanus indicating proportions of distinct gonad stages of individuals sampled at the dissipative (A), intermediate (B) and reflective (C) beaches. White bars: no individuals found. N, numbers of analysed clams per beach.

Figure 2.

Gametogenic cycle of Donax hanleyanus indicating proportions of distinct gonad stages of individuals sampled at the dissipative (A), intermediate (B) and reflective (C) beaches. White bars: no individuals found. N, numbers of analysed clams per beach.

The gametogenic cycle (active and spawning) of D. hanleyanus was correlated significantly with SST (Fig. 3A, C; Table 3c). In both years, the size of oocytes (Tables 3f, 4) and their abundance (Fig. 3D; Table 3g) was correlated significantly with the gametogenic cycle (Fig. 3C) at all three beaches. Furthermore, both parameters correlated significantly with SST at the dissipative and reflective beaches (Table 3j, k). At all three beaches, oocytes showed increased abundance in spring (Fig. 3D; Table 3 m), and decreased abundance in autumn–winter (Fig. 3D; Table 3n), when most specimens were in the cytolysed stage (Fig. 1D, H).

Table 4.

Donax hanleyanus: monthly SST (°C), mean oocyte size (µm) ( ± SD in µm), number of females per month (nf) and number of measured oocytes (no) from individuals inhabiting one of the three sampling localities Santa Teresita, Mar de las Pampas and Faro Querandí.

Year Month SST (°C) Santa Teresita (dissipative)
 
Mar de las Pampas (intermediate)
 
Faro Querandí (reflective)
 
    ± SD (µm) no nf  ± SD (µm) no nf  ± SD (µm) no nf 
2004 Dec 18.42 35.52 (9.53) 177 25 39.10 (11.23) 141 17 No sampling 
2005 Jan 20.79 37.52 (8.45) 176 27 40.14 (10.67) 187 19 
 Feb 21.08 41.79 (10.15) 185 16 35.53 (10.09) 106 13 
 Mar 20.49 39.78 (9.45) 237 21 40.96 (10.07) 233 12 38.09 (10.05) 116 14 
 Apr 16.75 – – – 36.97 (10.58) 117 13 30.49 (11.19) 61 16 
 May 14.14 – – – 39.41 (9.40) 102 15 32.55 (10.43) 53 21 
 Jun 12.67 – – – – – – 39.34 (12.69) 38 19 
 Jul 12.31 – – – – – – 37.51 (12.13) 53 16 
 Aug 12.74 – – – 35.65 (11.89) 122 35 33.97 (11.12) 71 41 
 Sep 14.84 7.62 (2.41) 73 29.36 (11.08) 65 12 23.01 (11.42) 49 14 
 Oct 16.18 28.14 (9.85) 454 16 – – – 29.52 (9.89) 591 20 
 Nov 18.53 38.24 (4.64) 570 20 36.22 (5.83) 510 17 40.14 (5.60) 514 18 
 Dec 20.87 35.32 (5.66) 540 18 36.88 (5.99) 612 20 40.09 (4.46) 660 22 
2006 Jan 21.59 – – – 38.10 (5.04) 540 18 36.34 (4.59) 390 13 
 Feb 22.55 – – – 34.84 (7.14) 360 12 37.61 (5.33) 330 11 
 Mar 21.29 – – – 30.78 (8.19) 540 18 26.54 (8.21) 359 12 
 Apr 19.51 34.89 (5.74) 90 29.70 (12.07) 390 13 32.78 (9.89) 390 13 
 May 14.99 – – – – – – 30.91 (10.87) 457 17 
 Jun 13.24 – – – 32.56 (9.62) 30 29.98 (11.53) 556 20 
 Jul 12.40 28.49 (12.61) 66 31.03 (11.89) 554 19 21.16 (12.40) 460 14 
 Aug 10.69 – – – 18.91 (12.44) 480 16 14.79 (8.29) 237 
 Sep 12.86 9.29 (4.22) 208 12 12.90 (7.65) 347 13 12.90 (7.80) 460 17 
 Oct 17.07 21.56 (9.28) 410 15 15.52 (5.62) 162 20.73 (9.27) 180 
 Nov 19.54 38.64 (5.06) 420 14 34.11 (8.36) 480 16 37.45 (4.27) 540 18 
 Dec 21.84 36.28 (4.14) 17  35.26 (8.10) 570 19 39.66 (7.68) 450 16 
Year Month SST (°C) Santa Teresita (dissipative)
 
Mar de las Pampas (intermediate)
 
Faro Querandí (reflective)
 
    ± SD (µm) no nf  ± SD (µm) no nf  ± SD (µm) no nf 
2004 Dec 18.42 35.52 (9.53) 177 25 39.10 (11.23) 141 17 No sampling 
2005 Jan 20.79 37.52 (8.45) 176 27 40.14 (10.67) 187 19 
 Feb 21.08 41.79 (10.15) 185 16 35.53 (10.09) 106 13 
 Mar 20.49 39.78 (9.45) 237 21 40.96 (10.07) 233 12 38.09 (10.05) 116 14 
 Apr 16.75 – – – 36.97 (10.58) 117 13 30.49 (11.19) 61 16 
 May 14.14 – – – 39.41 (9.40) 102 15 32.55 (10.43) 53 21 
 Jun 12.67 – – – – – – 39.34 (12.69) 38 19 
 Jul 12.31 – – – – – – 37.51 (12.13) 53 16 
 Aug 12.74 – – – 35.65 (11.89) 122 35 33.97 (11.12) 71 41 
 Sep 14.84 7.62 (2.41) 73 29.36 (11.08) 65 12 23.01 (11.42) 49 14 
 Oct 16.18 28.14 (9.85) 454 16 – – – 29.52 (9.89) 591 20 
 Nov 18.53 38.24 (4.64) 570 20 36.22 (5.83) 510 17 40.14 (5.60) 514 18 
 Dec 20.87 35.32 (5.66) 540 18 36.88 (5.99) 612 20 40.09 (4.46) 660 22 
2006 Jan 21.59 – – – 38.10 (5.04) 540 18 36.34 (4.59) 390 13 
 Feb 22.55 – – – 34.84 (7.14) 360 12 37.61 (5.33) 330 11 
 Mar 21.29 – – – 30.78 (8.19) 540 18 26.54 (8.21) 359 12 
 Apr 19.51 34.89 (5.74) 90 29.70 (12.07) 390 13 32.78 (9.89) 390 13 
 May 14.99 – – – – – – 30.91 (10.87) 457 17 
 Jun 13.24 – – – 32.56 (9.62) 30 29.98 (11.53) 556 20 
 Jul 12.40 28.49 (12.61) 66 31.03 (11.89) 554 19 21.16 (12.40) 460 14 
 Aug 10.69 – – – 18.91 (12.44) 480 16 14.79 (8.29) 237 
 Sep 12.86 9.29 (4.22) 208 12 12.90 (7.65) 347 13 12.90 (7.80) 460 17 
 Oct 17.07 21.56 (9.28) 410 15 15.52 (5.62) 162 20.73 (9.27) 180 
 Nov 19.54 38.64 (5.06) 420 14 34.11 (8.36) 480 16 37.45 (4.27) 540 18 
 Dec 21.84 36.28 (4.14) 17  35.26 (8.10) 570 19 39.66 (7.68) 450 16 
Figure 3.

Gametogenic cycle of Donax hanleyanus in relation to abiotic and biotic factors at the dissipative (I), intermediate (II) and reflective (III) beaches; (a) mean SST (°C), (b) mean RSA (mm−2), (c) percentage of gonad stages ‘ripe’ and ‘spawning’, (d) mean abundance of oocytes (numbers/mm2), (e) mean CI, (f) mean AFDM (g) and (g) numbers of recruits (<11 mm) per transect. Grey bars indicate seasons where Carreto et al. (1995) observed chlorophyll a maxima in the Buenos Aires shelf region. Error bars are standard deviation (SD). Note different scale on Y axes in graphs IIf and IIIf.

Figure 3.

Gametogenic cycle of Donax hanleyanus in relation to abiotic and biotic factors at the dissipative (I), intermediate (II) and reflective (III) beaches; (a) mean SST (°C), (b) mean RSA (mm−2), (c) percentage of gonad stages ‘ripe’ and ‘spawning’, (d) mean abundance of oocytes (numbers/mm2), (e) mean CI, (f) mean AFDM (g) and (g) numbers of recruits (<11 mm) per transect. Grey bars indicate seasons where Carreto et al. (1995) observed chlorophyll a maxima in the Buenos Aires shelf region. Error bars are standard deviation (SD). Note different scale on Y axes in graphs IIf and IIIf.

Analyses of monthly oocyte size classes showed a unimodal distribution, with size ranges from 2 to 70 µm (Fig. 4), a situation which did not vary significantly between beaches (ANOVA, F2,54 = 0.194, P > 0.05). At the dissipative beach the modal oocyte size decreased twice each year, in early spring (September) and in spring–summer (December). A twice-annual decrease was also recorded at the intermediate beach each year, in winter (August) and in summer (twice in February, once in December 2006), while at the reflective site decreases were observed during all seasons (autumn: April 2005 and March 2006; winter: June and July 2006; spring: September 2005; and summer: January 2006) (Table 3l), suggesting two spawning events at the dissipative and intermediate beaches and more continuous gamete releases at the reflective beach (Fig. 4).

Figure 4.

Monthly oocyte size–frequency distribution of Donax hanleyanus. Grey bars show modal values indicating the record prior to and after a sudden reduction of oocyte sizes. N, numbers of measured oocytes per month.

Figure 4.

Monthly oocyte size–frequency distribution of Donax hanleyanus. Grey bars show modal values indicating the record prior to and after a sudden reduction of oocyte sizes. N, numbers of measured oocytes per month.

Relative spermatozoon abundance

RSA correlated significantly with the gametogenic cycle (active and spawning gonad stages) of D. hanleyanus at all three beaches (Fig. 3B; Table 3q), whereby RSA showed significant increases at the dissipative and intermediate beaches in spring and at the reflective beach in spring–summer (Fig. 3B; Table 3r). At all three beaches RSA decreased significantly in autumn–winter (Fig. 3B; Table 3s).

Condition index, biomass and recruitment

At all three study beaches, the annual reproductive cycle of D. hanleyanus correlated significantly with CI (Table 3d). CI was correlated significantly with SST at both the dissipative and the reflective beach (Table 3 h). Seasonal variations in clam biomass (AFDM) within the respective beach populations correlated significantly with SST (Table 3i) and with mature gonad stages (ripe and spawning) at all three beaches (Table 3e), but did not vary between beaches (ANOVA, F2,59 = 0.425, P > 0.05). Recruits of D. hanleyanus were found in all seasons at all three beaches during both sampling years (except at the dissipative beach in January 2006 and at the intermediate beach in June, July and October 2005). However, the peak of settlement was recorded at the intermediate beach in summer 2005 and in summer–autumn 2006 [Fig. 3 (IIg); Table 3p], at the reflective beach in autumn 2005 and autumn–winter 2006 [Fig. 3 (IIIg); Table 3p], and at the dissipative beach for extended periods during autumn–winter 2005 and summer-autumn 2006 [Fig. 3 (Ig); Table 3p]. Despite this apparent seasonality, the settlement period of recruits was not significantly correlated with SST (Table 3o). The length–frequency distributions covering 2 years revealed that the intermediate and reflective beaches were populated by considerably larger clams than the dissipative site (F2,21836 = 63.618, P < 0.05). The mean apSL of juveniles found at the dissipative beach was significantly higher than that of both other populations, but adult apSL was considerably smaller (F2,21836 = 11,302.263, P < 0.05). The analysis of the population structure indicated that at the dissipative beach, recruits represented 22.8% of the population and adults accounted for only 1.5%, whereas the reflective beach population comprised a mere 10.5% recruits and 65% adult wedge clams (Figs 4, 5).

Figure 5.

Length–frequency distribution (pooled from 25 monthly samples) of Donax hanleyanus at the dissipative (A), intermediate (B) and reflective (C) beaches, classified into three ontogenetic groups: recruits (<11 mm), juveniles (11–22 mm) and adults (>22 mm). Note the different scale on the Y axes.

Figure 5.

Length–frequency distribution (pooled from 25 monthly samples) of Donax hanleyanus at the dissipative (A), intermediate (B) and reflective (C) beaches, classified into three ontogenetic groups: recruits (<11 mm), juveniles (11–22 mm) and adults (>22 mm). Note the different scale on the Y axes.

Size and biomass at sexual maturity

Single individuals of D. hanleyanus matured when smaller (apSL) and lighter (SFWM) at the dissipative beach (males: 8.61 mm, 0.02 g; females: 9.35 mm, 0.04 g) than at the intermediate (males: 12.72 mm, 0.10 g; females: 13.21 mm, 0.12 g) or reflective (males: 22.92 mm, 0.55 g; females 22.44 mm, 0.39 g) beaches. At all three study locations, the relationship between size and biomass at 50% population maturity was explained convincingly (P < 0.05) by the non-linear regression given as Eq. 5, for both males and females (Fig. 6; Table 5). At the dissipative beach, clams were 100% mature on reaching an apSL of c. 23 mm (both sexes) and an SFWM of 0.4 g. Females at the intermediate beach reached up to 20 mm apSL and 0.9 g SFWM before attaining 100% maturity, while for males apSL and SFWM at 100% maturity were even higher, 27 mm and 0.5 g, respectively. In contrast, at the reflective beach clams were largest and heaviest (c. 30 mm apSL and 1.8 g SFWM, both sexes) when reaching 100% maturity (Fig. 6; Table 5). The logistic function of male wedge clam apSL was significantly steeper at the dissipative (ARSS F2,21 = 324.631, P < 0.05) and intermediate beaches (ARSS F2,23 = 418.321, P < 0.05), indicating that 50% population maturity was reached at a significantly larger size (apSL50%) at the reflective beach (Fig. 6A; Table 5). A similar pattern was found in female specimens, where the slope of the logistic function was also significantly steeper at the dissipative (ARSS F2,20 = 658.950, P < 0.05) and intermediate beach (ARSS F2,23 = 973.554, P < 0.05), resulting in significantly larger female clams in a 50% mature population (apSL50%) at the reflective beach (Fig. 6B; Table 5). Size and biomass (both sexes) at 50% population maturity (SFWM50%) were significantly higher at the reflective beach (Fig. 6C, D), leading to significantly steeper logistic SFWM function at the dissipative (ARSS: males F2,50 = 253.423, P < 0.05, females F2,53 = 579.924, P < 0.05) and intermediate beaches (ARSS: males F2,152 = 481.643, P < 0.05, females F2,125 = 761.464, P < 0.05) (Table 5).

Figure 6.

Logistic function (Eq. 4) indicating size (apSL) and biomass (SFWM) at sexual maturity in male (A, C) and female (B, D) Donax hanleyanus, fitted by non-linear regression, showing sex-specific differences. Statistical results are provided in Table 5.

Figure 6.

Logistic function (Eq. 4) indicating size (apSL) and biomass (SFWM) at sexual maturity in male (A, C) and female (B, D) Donax hanleyanus, fitted by non-linear regression, showing sex-specific differences. Statistical results are provided in Table 5.

Table 5.

Parameters (a, b and x0) estimated values and associated statistics relating to size (apSL) and biomass (SFWM) at sexual maturity function (Eq. 3) and size of 50% population maturity (apSL50%) and biomass of 50% population maturity (SFWM50%), respectively (Eq. 5).

 Males
 
Females
 
 Santa Teresita
 
Mar de las Pampas
 
Faro Querandí
 
Santa Teresita
 
Mar las Pampas
 
Faro Querandí
 
 apSL SFWM apSL SFWM apSL SFWM apSL SFWM apSL SFWM apSL SFWM 
a 102.96 (3.32)* 101.62 (1.39)* 97.66 (1.92)* 97.94 (0.73)* 101.03 (2.45)* 98.84 (0.75)* 100.61 (2.09)* 100.63 (1.04)* 99.41 (1.23)* 99.12 (0.66)* 101.74 (2.52)* 98.62 (0.80)* 
b 1.95 (0.27)* 0.05 (0.01)* 1.93 (0.24)* 0.03 (0.01)* 1.14 (0.20)* 0.12 (0.01)* 1.49 (0.16)* 0.04 (0.00)* 0.72 (0.09)* 0.08 (0.01)* 1.61 (0.23)* 0.11 (0.01)* 
x0 15.81 (0.32)* 0.15 (0.01)* 17.57 (0.27)* 0.19 (0.01)* 24.40 (0.23)* 0.92 (0.01)* 15.77 (0.19)* 0.15 (0.00)* 16.37 (0.11)* 0.24 (0.01)* 23.54 (0.27)* 0.89 (0.01)* 
50% 15.69 0.15 17.66 0.19 24.38 0.92 15.75 0.15 16.38 0.25 23.49 0.89 
R2 0.97 0.91 0.97 0.86 0.97 0.89 0.99 0.96 0.99 0.92 0.97 0.90 
DW 1.64 1.62 2.49 1.99 2.32 1.84 2.11 1.73 2.39 1.99 2.43 2.18 
 Males
 
Females
 
 Santa Teresita
 
Mar de las Pampas
 
Faro Querandí
 
Santa Teresita
 
Mar las Pampas
 
Faro Querandí
 
 apSL SFWM apSL SFWM apSL SFWM apSL SFWM apSL SFWM apSL SFWM 
a 102.96 (3.32)* 101.62 (1.39)* 97.66 (1.92)* 97.94 (0.73)* 101.03 (2.45)* 98.84 (0.75)* 100.61 (2.09)* 100.63 (1.04)* 99.41 (1.23)* 99.12 (0.66)* 101.74 (2.52)* 98.62 (0.80)* 
b 1.95 (0.27)* 0.05 (0.01)* 1.93 (0.24)* 0.03 (0.01)* 1.14 (0.20)* 0.12 (0.01)* 1.49 (0.16)* 0.04 (0.00)* 0.72 (0.09)* 0.08 (0.01)* 1.61 (0.23)* 0.11 (0.01)* 
x0 15.81 (0.32)* 0.15 (0.01)* 17.57 (0.27)* 0.19 (0.01)* 24.40 (0.23)* 0.92 (0.01)* 15.77 (0.19)* 0.15 (0.00)* 16.37 (0.11)* 0.24 (0.01)* 23.54 (0.27)* 0.89 (0.01)* 
50% 15.69 0.15 17.66 0.19 24.38 0.92 15.75 0.15 16.38 0.25 23.49 0.89 
R2 0.97 0.91 0.97 0.86 0.97 0.89 0.99 0.96 0.99 0.92 0.97 0.90 
DW 1.64 1.62 2.49 1.99 2.32 1.84 2.11 1.73 2.39 1.99 2.43 2.18 

Values are mean ± SE, mean size at first maturity in millimetres (apSL) and biomass at first maturity in grams (SFWM), Durbin–Watson statistic (DW).

*P < 0.05.

DISCUSSION

Comparison of reproductive biology on morphodynamically distinct beach types

Histological analyses from the 25-month sampling series revealed differences in the reproductive biology of Donax hanleyanus populating three morphodynamically distinct beaches. However, in contrast to the 13-month data set of Delgado & Defeo (2007), which documented the reproductive cycle of Uruguayan D. hanleyanus from two sandy beaches and supported the HHH, the present study does not confirm all predictions of this hypothesis for the reproductive cycle of the species in Argentina.

Consistency with predictions of the HHH

In accordance with the HHH, the settlement period of D. hanleyanus recruits was extended and recruits were more abundant at the dissipative beach compared to either the intermediate or reflective beach. Although there have been several previous studies on the reproduction biology of Donax species (Argentinean D. hanleyanus: Penchaszadeh & Olivier, 1975; Peruvian D. marincovichi: Huaraz & Ishiyama, 1980; Portuguese D. trunculus: Gaspar et al., 1999; Spanish D. semistriatus and D. venustus: Tirado & Salas, 1999; Brazilian D. hanleyanus: Gil & Thomé, 2004), data on recruitment at beaches with different morphodynamics are scarce and inconsistent. The present results are in keeping with those of Delgado & Defeo's (2007) conclusions regarding recruitment in the Uruguayan D. hanleyanus population, in that both abundance of recruits and recruitment period are greater at a dissipative beach. The results of the current study suggest a shorter recruitment season for D. hanleyanus in reflective beach populations and so coincide with a prediction of the HHH. However, recruits of Donax do not always show the same pattern. For instance, Laudien, Brey & Arntz (2001) observed a longer period of recruitment in Namibian D. serra at a reflective beach than a dissipative one, a finding that directly contradicts the predictions of the HHH.

Contrasts to predictions of the HHH

The ‘spawning’ stage of the clams was more restricted in the dissipative beach population than in those from the beaches exhibiting intermediate and reflective morphodynamics. Furthermore, the proportion of specimens in the sexual rest stage varied from beach to beach. While at the reflective and intermediate beach gonads of indistinguishable sex appeared in low percentages from autumn to spring, large numbers of specimens in the sexual rest stage were only found twice (early autumn 2005 and late summer 2006) at the dissipative beach. Comparable size classes were present at all three beaches at the same time (Herrmann, 2009). Additionally, RSA was lower at the dissipative and intermediate beaches than at the reflective one. Further evidence for an extended reproductive cycle at the reflective beach, contrary to the predictions of the HHH, was derived from monthly oocyte size–frequency distributions. Modal oocyte values indicate two spawning events per year at the dissipative and intermediate beaches and continuous gamete releases at the reflective beach. Size and biomass at first maturity of both sexes were lower at the dissipative beach, suggesting an abrupt transition from sexual rest to reproductive activity at this site. Monthly mean abundance of D. hanleyanus was significantly higher at the reflective beach (ANOVA, F2,69 = 14.675, P < 0.05), compared to the dissipative and intermediate locations (Herrmann et al., 2008b, 2009a), which is also counter to the predictions of the HHH. Consequently, the present study supports the results of previous investigators documenting increases in abundance of isopods (E. braziliensis: Defeo et al., 1997; Defeo & Martínez, 2003), amphipods (P. brasiliensis: Gómez & Defeo, 1999; A. brasiliensis: Defeo & Gómez, 2005) and decapods (E. brasiliensis: Defeo et al., 2001) at Uruguayan reflective beaches.

Comparison of the reproductive biology among the three beaches

Gametogenic cycle

The histological examination of D. hanleyanus gonads suggests several reproductive events per year (Fig. 2), with an underlying seasonality, confirmed by a significant correlation with SST at all three beaches. Gonad classifications revealed two annual spawning events at the dissipative beach (spring and summer) and the intermediate beach (spring and summer–autumn) and a continuous gamete release over the year at the reflective beach (Table 6A). These are in line with the results of Delgado & Defeo (2007) and Gil & Thomé (2004), who observed two spawning events at dissipative beaches and continuous spawning at reflective sites in Uruguay and Brazil respectively. However, spawning events can be determined more exactly by oocyte measurements (e.g. Penchaszadeh & Olivier, 1975; Sarkis, Couturier & Cogswell, 2006; Morriconi, Lomovasky & Calvo, 2007). Modal values of the oocyte size–frequency distributions (Fig. 4) indicate two spawning events at the dissipative beach (September and December), and three each at the intermediate beach (February, August and December) and the reflective beach (January, April–July and September), with a more extended period at the latter (Table 6B). Similarly, four decades ago Penchaszadeh & Olivier (1975) detected two spawning events (January–February and August–September) of D. hanleyanus by measuring oocyte sizes at Villa Gesell, 10 km north of the intermediate beach, Mar de las Pampas. In agreement with previous investigations on the reproductive biology of D. hanleyanus (Penchaszadeh & Olivier, 1975; Gil & Thomé, 2004; Delgado & Defeo, 2007), the present study confirms that there is no period of complete gonadal inactivity in this species. The same is true in the Peruvian D. marincovichi (Huaraz & Ishiyama, 1980) and the Namibian D. serra (Laudien et al., 2001), but not in the Portuguese D. trunculus (Gaspar et al., 1999) and the Spanish D. venustus and D. semistriatus (Tirado & Salas, 1999). The discrepancy may be caused by the significant differences in seasonal ranges in SST range routinely experienced by different species (Urban & Campos, 1994; Sasaki, Ota & Saeki, 1997; Laudien et al., 2001). However, SST is not the only parameter influencing the reproductive cycle of suspension feeders (Sastry, 1968, 1979; Giese, 1974). Changes in phytoplankton biomass and species composition are also key factors. Thus, the increase of chlorophyll a concentrations in the Buenos Aires shelf region observed during winter and summer by Carreto et al. (1995), with a main peak in spring and a secondary peak in autumn, corresponds convincingly with the dominance of ripe and spawning stages of D. hanleyanus (Fig. 3). This suggests that phytoplankton abundance may also have a direct impact on the reproductive cycle of D. hanleyanus.

Table 6.

Annual reproductive events of Donax hanleyanus from Argentinean (AR), Uruguayan (UY) and Brazilian (BR) beaches of different morphodynamic types (M): dissipative (D), intermediate (I) and reflective (R). A, spawning events (grey) indicated by classifying gonad tissue into different gametogenic stages; B, spawning events (grey) derived from oocyte measurements.

graphic 
graphic 

Relative spermatozoon abundance

As documented and discussed above, RSA was lower at the dissipative and intermediate beach (spring), than at the reflective one (spring–summer). The measurement of spermatozoon abundance turned out to be a good reflection of male gonadal tissue condition and thus a useful indicator in the investigation of seasonal variations in the gametogenic cycle between beaches with contrasting morphodynamics. This method negates the need for time-consuming microscopical examination of gonads to assign samples to different developmental stages. RSA may be used to analyse the relationships between abiotic and biotic parameters and the gametogenic cycle. Furthermore, RSA may be calibrated species-specifically; counting spermatids per surface area allows estimations of absolute spermatozoon abundance.

Condition index, biomass and recruitment

CI and AFDM are useful tools in describing the reproductive biology of D. hanleyanus. Both indicate gonadal mass changes throughout the year, with highest values when gonads are in the mature stage. The results presented here are in line with studies of other surf clams, where the CI was also successfully used to describe changes in the gametogenic cycle (D. trunculus: Gaspar et al., 1999; D. serra: Laudien et al., 2001; Mesodesma donacium: Riascos et al., 2008). Recruitment pattern and abrupt changes in oocyte size appear to indicate a meroplanktonic phase of c. 3 months, assuming that collected recruits originated from the studied adult population. Recruitment patterns observed in the present study provide clear evidence that juvenile wedge clams occur only sporadically and recruitment varies between years, agreeing with the findings of other studies (e.g. Arntz et al., 1987; Laudien et al., 2001; Herrmann et al., 2009b). It should be noted that the absence of recruits does not necessarily indicate a lack of spawning activity (Caddy & Defeo, 2003). Environmental conditions may strongly influence recruitment of marine invertebrates (e.g. hydrodynamic processes: Roughgarden, Gaines & Possingham, 1988; food limitation: Olson & Olson, 1989; predators: Sale, 1990), and unfavourable conditions may lead to failure of recruitment at the parent beach.

Size and biomass at sexual maturity

Single individuals of D. hanleyanus in the present study matured with an average shell length of 9 mm (and 0.02 g SFWM) and reached gonadal maturity apSL50% at 16 mm (and 0.15 g SFWM50%). The present results coincide well with estimates of 40 years ago (15 mm: Penchaszadeh & Olivier, 1975), and with data from the Uruguayan D. hanleyanus population (minimal length with gonad development 10 mm, gonad maturity 12 mm; Delgado & Defeo, 2007).

CONCLUSION

In conclusion, the results of the present study obtained from three beaches over 25 months demonstrate that at the population level Donax hanleyanus respond systematically to beach morphodynamics in a manner opposite to that predicted by the HHH. The population at the dissipative beach exhibited a greater abundance of recruits and an extended recruitment period, but spawning specimens were greatly outnumbered by those in sexual rest stages. Spawning events were recorded twice each year at the dissipative (early spring and spring–summer) and intermediate beaches (winter and summer), whereas continuous gamete release was noted at the reflective beach. The onset of maturity was observed in single D. hanleyanus individuals of c. 9 mm apSL and 0.02 g SFWM. 50% of the population attained maturity at 15 mm apSL and 0.15 g SFWM, and 100% were mature at a size of 23–27 mm apSL and 0.4–0.9 g SFWM. Size at first maturity and biomass at first maturity were lower at the dissipative beach, whereas monthly mean abundance of D. hanleyanus was higher at the reflective beach. Finally, the current study demonstrated that the ‘hypothesis of habitat safety (HHS)’, originally postulated by Defeo & Gómez (2005) for supralittoral species, may be extended to intertidal species; the combination of narrow swashes and steep slopes make reflective beaches a safer and more stable environment for supralittoral and intertidal species such as D. hanleyanus.

ACKNOWLEDGEMENTS

This paper is part of the doctoral thesis of Marko Herrmann, partly supported by ‘Deutscher Akademischer Austauschdienst (DAAD)’, the PADI foundation and by the University of Bremen. The authors express their deepest gratitude to Dr Omar Defeo for critical reading of and valuable suggestions on the final manuscript. The authors thank José Alfaya, Sonia Cabrera and Soledad Zabala for field assistance, as well as the youngest helper Belén Alvela who measured water surface temperature off Santa Teresita each day. Thanks are due to the staff of the Laboratorio de Microscopia, especially to Sara Orrea for assistance in preparing the histological sections. We are also very grateful to Sandra Noir and Ingrid Albrecht, who were a great help in counting and measuring thousands of oocytes. Additional thanks are also expressed to all Argentinean colleagues from laboratory 19 at the Facultad de Ciencias Exactas y Naturales (UBA) and from laboratory 80 at the Museo Argentino de Ciencias Naturales Bernardino Rivadavia (MACN).

REFERENCES

ARNTZ
W.E.
BREY
T.
TARAZONA
J.
ROBLES
A.
Changes in the structure of a shallow sandy-beach community in Peru during an El Niño event
South African Journal of Marine Science
 , 
1987
, vol. 
5
 (pg. 
645
-
658
)
AXIOVISION
Digital Image Processing Software. Carl Zeiss MicoImaging GmbH
2008
BREY
T.
RUMOHR
H.
ANKAR
S.
Energy content of macrobenthic invertebrates: general conversion factors from weight to energy
Journal of Experimental Marine Biology and Ecology
 , 
1988
, vol. 
117
 (pg. 
271
-
278
)
BREZINA
J.
Particle size and settling rate distributions of sand-sized materials
2nd European Symposium on Particle Characterisation (PARTEC)
 , 
1979
Nürnberg
pg. 
12
  
Germany
BREZINA
J.
SedVar 6.2p: computer program for the calculation of particle settling velocities as a function of temperature and salinity
 , 
1997
Neckargemuend, Germany
Granometry
CADDY
J.F.
DEFEO
O.
Enhancing or restoring the productivity of natural populations of shellfish and other marine invertebrate resources
FAO Fisheries Technical Paper
 , 
2003
, vol. 
448
 (pg. 
1
-
168
)
CARDOSO
R.S.
VELOSO
V.G.
Population dynamics and secondary production of the wedge clam Donax hanleyanus (Bivalvia: Donacidae) on a high-energy, subtropical beach of Brazil
Marine Biology
 , 
2003
, vol. 
142
 (pg. 
153
-
162
)
CARRETO
J.I.
LUTZ
V.A.
CARIGNAN
M.O.
CUCCHI COLLEONI
A.D.
DE MARCOS
S.G.
Hydrography and chlorophyll a in a transect from the coast to the shelf-break in the Argentinian Sea
Continental Shelf Research
 , 
1995
, vol. 
15
 (pg. 
315
-
336
)
CHEN
Y.
JACKSON
D.A.
HARVEY
H.H.
A comparison of von Bertalanffy and polynomial functions in modelling fish growth data
Canadian Journal of Fisheries and Aquatic Sciences
 , 
1992
, vol. 
49
 (pg. 
1228
-
1235
)
DE CASTELLANOS
Z.A.
FERNANDEZ
D.
Sobre la presencia de Donax hanleyanus en la costa Argentina
Neotropica
 , 
1965
, vol. 
11
 (pg. 
1
-
58
)
DE VILLIERS
G.
Reproduction of the sand mussel Donax serra Roding
Investigational Report, Republic of South Africa, Department of Industries, Sea Fisheries Branch
 , 
1975
, vol. 
103
 (pg. 
1
-
33
)
DEAN
R.F.
Heuristic models of sand transport in the surf zone
Proceedings of engineering dynamics in the surf zone
 , 
1973
Sydney, Australia
Institute of Engineers
(pg. 
208
-
214
)
DEFEO
O.
GÓMEZ
J.
Morphodynamics and habitat safety in sandy beaches: life-history adaptations in a supralittoral amphipod
Marine Ecology Progress Series
 , 
2005
, vol. 
293
 (pg. 
143
-
153
)
DEFEO
O.
MARTÍNEZ
G.
The habitat harshness hypothesis revisited: life history of the isopod Excirolana braziliensis in sandy beaches with contrasting morphodynamics
Journal of the Marine Biological Association of the United Kingdom
 , 
2003
, vol. 
83
 (pg. 
331
-
340
)
DEFEO
O.
BRAZEIRO
A.
DE ALAVA
A.
RIESTRA
G.
Is sandy beach macrofauna only physically controlled? Role of substrate and competition on isopods
Estuarine, Coastal and Shelf Science
 , 
1997
, vol. 
45
 (pg. 
453
-
462
)
DEFEO
O.
GOMEZ
J.
LERCARI
D.
Testing the swash exclusion hypothesis in sandy beach populations: the mole crab Emerita brasiliensis in Uruguay
Marine Ecology Progress Series
 , 
2001
, vol. 
212
 (pg. 
159
-
170
)
DEFEO
O.
LERCARI
D.
GOMEZ
J.
The role of morphodynamics in structuring sandy beach populations and communities: what should be expected?
Journal of Coastal Research
 , 
2003
, vol. 
35
 (pg. 
352
-
362
)
DELGADO
E.
DEFEO
O.
Tisular and population level responses to habitat harshness in sandy beaches: the reproductive strategy of Donax hanleyanus
Marine Biology
 , 
2007
, vol. 
152
 (pg. 
919
-
927
)
EMERY
K.O.
A simple method of measuring beach profiles
Limnology and Oceanography
 , 
1961
, vol. 
6
 (pg. 
695
-
710
)
FLEMMING
B.W.
THUM
A.B.
The settling tube – a hydraulic method for grain size analysis of sands
Kieler Meeresforschungen Sonderheft
 , 
1978
, vol. 
4
 (pg. 
82
-
95
)
FLEMMING
B.W.
ZIEGLER
K.
High-resolution grain size distribution patterns and textural trends in the backbarrier environment of Spiekeroog Island (Southern North Sea)
Senckenbergiana Maritima
 , 
1995
, vol. 
26
 (pg. 
1
-
24
)
GASPAR
M.B.
FERREIRA
R.
MONTEIRO
C.C.
Growth and reproductive cycle of Donax trunculus L., (Mollusca: Bivalvia) off Faro, southern Portugal
Fisheries Research
 , 
1999
, vol. 
41
 (pg. 
309
-
316
)
GIBBS
R.J.
MATTHEWS
M.D.
LINK
D.A.
The relationship between sphere size and settling velocity
Journal of Sedimentary Petrology
 , 
1971
, vol. 
27
 (pg. 
3
-
26
)
GIESE
A.C.
Giese
A.C.
Pearse
J.S.
Introduction: general principles
Reproduction of marine invertebrates
 , 
1974
New York
Academic Press
(pg. 
1
-
49
)
GIL
G.M.
THOMÉ
J.W.
Descrição do ciclo reprodutivo de Donax hanleyanus (Bivalvia, Donacidae) no sul do Brasil
Iheringia
 , 
2004
, vol. 
94
 (pg. 
271
-
276
)
GÓMEZ
J.
DEFEO
O.
Life history of the sandhopper Pseudorchestoidea brasiliensis (Amphipoda) in sandy beaches with contrasting morphodynamics
Marine Ecology Progress Series
 , 
1999
, vol. 
182
 (pg. 
209
-
220
)
HERRMANN
M.
Population dynamics of the surf clams Donax hanleyanus and Mesodesma mactroides from open-Atlantic beaches off Argentina
Reports on Polar and Marine Research
 , 
2009
, vol. 
585
 (pg. 
1
-
235
)
HERRMANN
M.
ALFAYA
J.E.F.
PENCHASZADEH
P.E.
LAUDIEN
J.
Sea surface temperature at station Santa Teresita and data from Mesodesma mactroides (Bivalvia: Mesodesmatidae) from northern Argentina, dataset
PANGAEA – Publishing Network for Geoscientific & Environmental Data
 , 
2008
 
HERRMANN
M.
LAUDIEN
J.
PENCHASZADEH
P.E.
FISCHER
S.
ARNTZ
W.E.
Population structure, growth and production of the wedge clam Donax hanleyanus (Bivalvia: Donacidae) from northern Argentinean beaches, data set
PANGAEA – Publishing Network for Geoscientific & Environmental Data
 , 
2008
 
HERRMANN
M.
CARSTENSEN
D.
LAUDIEN
J.
PENCHASZADEH
P.E.
FISCHER
S.
ARNTZ
W.E.
Population structure, growth and production of the wedge clam Donax hanleyanus (Bivalvia: Donacidae) from northern Argentinean beaches
Journal of Shellfish Research
 , 
2009
, vol. 
28
 (pg. 
1
-
16
)
HERRMANN
M.
ALFAYA
J.E.F.
LEPORE
M.L.
PENCHASZADEH
P.E.
LAUDIEN
J.
Reproductive cycle and gonad development of the Northern Argentinean Mesodesma mactroides (Bivalvia: Mesodesmatidae)
Helgoland Marine Research
 , 
2009
 
HOWARD
D.W.
LEWIS
E.J.
KELLER
B.J.
SMITH
C.S.
Histological techniques for marine bivalve mollusks and crustaceans
NOAA Technical Memorandum NOS NCCOS
 , 
2004
, vol. 
5
 (pg. 
1
-
218
)
HUARAZ
F.L.
ISHIYAMA
V.C.
Madurez sexual de la ‘concha mariposa’ (Donax peruvianus) de la playa de Jahuay, Ica, Peru
Revista de Ciencias U.N.M.S.M.
 , 
1980
, vol. 
72
 (pg. 
47
-
56
)
INMAN
D.L.
Measures for describing the size distribution of sediments
Journal of Sedimentary Petrology
 , 
1952
, vol. 
22
 (pg. 
125
-
145
)
LAUDIEN
J.
BREY
T.
ARNTZ
W.E.
Reproduction and recruitment patterns of the surf clam Donax serra (Bivalvia, Donacidae) on two Namibian sandy beaches
South African Journal of Marine Science
 , 
2001
, vol. 
23
 (pg. 
53
-
60
)
McARDLE
S.B.
McLACHLAN
A.
Dynamics of the swash zone and effluent line on sandy beaches
Marine Ecology Progress Series
 , 
1991
, vol. 
76
 (pg. 
91
-
99
)
McARDLE
S.B.
McLACHLAN
A.
Sandy beach ecology: swash features relevant to the macrofauna
Journal of Coastal Research
 , 
1992
, vol. 
8
 (pg. 
398
-
407
)
McGULLAGH
P.
NELDER
J.A.
Generalized linear models
 , 
1997
Edn 2
London
Chapman and Hall
McLACHLAN
A.
The definition of sandy beaches in relation to exposure: a simple rating system
South African Journal of Marine Science
 , 
1980
, vol. 
76
 (pg. 
137
-
138
)
McLACHLAN
A.
Dissipative beaches and macrofauna communities on exposed intertidal sands
Journal of Coastal Research
 , 
1990
, vol. 
6
 (pg. 
57
-
71
)
McLACHLAN
A.
BROWN
A.C.
The ecology of sandy shores
 , 
2006
Amsterdam
Elsevier
MARCOMINI
S.C.
LÓPEZ
R.A.
Coastal protection effects at Buenos Aires, Argentina
Coastal Zone: Proceedings of the Symposium on Coastal and Ocean Management
 , 
1993
, vol. 
93
 (pg. 
2724
-
2738
)
MARCOMINI
S.C.
PENCHASZADEH
P.E.
LÓPEZ
R.A.
LUZZATTO
D.C.
Beach morphodynamics and clam (Donax hanleyanus) densities in Buenos Aires, Argentina
Journal of Coastal Research
 , 
2002
, vol. 
18
 (pg. 
601
-
611
)
MORRICONI
E.
LOMOVASKY
B.J.
CALVO
J.
Reproductive cycle and energy content of Tawera gayi (Hupé 1854) (Bivalvia: Veneridae) at the southernmost limit of their distribution range
Journal of Shellfish Research
 , 
2007
, vol. 
26
 (pg. 
81
-
88
)
NARCHI
W.
Functional anatomy of Donax hanleyanus Philippi 1847 (Donacidae-Bivalvia)
Boletim de Zoologia, Universidade de Sao Paulo
 , 
1978
, vol. 
3
 (pg. 
121
-
142
)
NOY-MEIR
I.
Structure and function of desert ecosystems
Israel Journal of Botany
 , 
1979
, vol. 
28
 (pg. 
1
-
19
)
OLSON
R.R.
OLSON
M.H.
Food limitation of planktotrophic marine invertebrate larvae: does it control recruitment success?
Annual Review of Ecology and Systematics
 , 
1989
, vol. 
20
 (pg. 
225
-
247
)
PENCHASZADEH
P.E.
OLIVIER
S.R.
Ecología de una población de ‘berberecho’ (Donax hanleyanus) en Villa Gesell, Argentina
Malacologia
 , 
1975
, vol. 
15
 (pg. 
133
-
146
)
PHOTOSHOP
Digital imaging software package Adobe Photoshop version CS3 Extended. Adobe Systems Inc
2008
RIASCOS
J.M.
HEILMAYER
O.
OLIVA
M.E.
LAUDIEN
J.
ARNTZ
W.E.
Infestation of the surf clam Mesodesma donacium by the spionid polychaete Polydora bioccipitalis
Journal of Sea Research
 , 
2008
, vol. 
59
 (pg. 
217
-
227
)
RICCIARDI
A.
BOURGET
E.
Global patterns of macroinvertebrate biomass in marine intertidal communities
Marine Ecology Progress Series
 , 
1999
, vol. 
185
 (pg. 
21
-
35
)
ROUGHGARDEN
J.
GAINES
S.
POSSINGHAM
H.
Recruitment dynamics in complex life cycles
Science
 , 
1988
, vol. 
241
 (pg. 
1460
-
1466
)
SALE
P.F.
Recruitment of marine species: is the bandwagon rolling in the right direction?
Trends in Ecology and Evolution
 , 
1990
, vol. 
5
 (pg. 
25
-
27
)
SARKIS
S.
COUTURIER
C.
COGSWELL
A.
Reproduction and spawning in calico scallops, Argopecten gibbus, from Bermuda
Journal of Shellfish Research
 , 
2006
, vol. 
25
 (pg. 
503
-
508
)
SASAKI
K.
OTA
H.
SAEKI
M.
Morphological development of veliger larval and juvenile stages of the surf clam Spisula sachalinensis
Fisheries Science
 , 
1997
, vol. 
63
 (pg. 
81
-
89
)
SASTRY
A.N.
The relationships among food, temperature, and gonad development of the bay scallops Aequipecten irradians Lamarck
Physiological Zoology
 , 
1968
, vol. 
41
 (pg. 
44
-
53
)
SASTRY
A.N.
Giese
A.C.
Pearse
J.S.
Pelecypoda (excluding Ostreidae)
Reproduction of marine invertebrates
 , 
1979
New York
Academic Press
(pg. 
113
-
292
)
SHORT
A.D.
WRIGHT
L.D.
McLachlan
A.
Erasmus
T.
Physical variability of sandy beaches
Sandy beaches as ecosystems
 , 
1983
The Hague, The Netherlands
W. Junk
(pg. 
133
-
144
)
SIGMAPLOT
Software package for exact graphs and data analysis, Version 11
 , 
2008
Erkrath, Germany
Systat Software GmbH
SIGMASTAT
Software package SigmaStat 3.1: user's manual
 , 
2004
Point Richmond
Systat
SPSS
Statistical package for the social sciences, version 16.0.1
 , 
2007
Chicago
SPSS
TIRADO
C.
SALAS
C.
Reproduction of Donax venustus Poli 1795, Donax semistriatus Poli 1795 and intermediate morphotypes (Bivalvia: Donacidae) in the littoral of Málaga (Southern Spain)
Marine Ecology
 , 
1999
, vol. 
20
 (pg. 
111
-
130
)
URBAN
H.-J.
CAMPOS
B.
Population dynamics of the bivalves Gari solida, Semele solida and Protothaca thaca from a small bay in Chile at 36°S
Marine Ecology Progress Series
 , 
1994
, vol. 
115
 (pg. 
93
-
102
)
VELOSO
V.G.
CARDOSO
R.S.
Effect of morphodynamics on the spatial and temporal variation of macrofauna on three sandy beaches, Rio de Janeiro State, Brazil
Journal of the Marine Biological Association of the United Kingdom
 , 
2001
, vol. 
81
 (pg. 
369
-
375
)
WENTWORTH
C.K.
A scale of grade and class terms for clastic sediments
Journal of Geology
 , 
1922
, vol. 
30
 (pg. 
377
-
392
)
ZAR
J.H.
Biostatistical analysis
 , 
1999
Edn 4
Prentice-Hall
Upper Saddle River