INTRODUCTION
Squat lobsters (Decapoda: Anomura: Galatheidae) comprise more than 200 small-sized species distributed worldwide, ranging from sublittoral to abyssal environments. Because of their high abundance, some galatheids represent an important food source, supporting large-scale fisheries (Longhurst, Reference Longhurst1967; Castilla & Becerra, Reference Castilla, Becerra and Valle-Levinson1976; Roa et al., Reference Roa, Gallardo, Ernst, Baltazar, Cañete and Enríquez-Brionnes1995) and populations of several marine resources like crabs, squids, fish, whales, albatrosses and penguins (Romero, Reference Romero2003; Etnoyer et al., Reference Etnoyer, Canny, Mate and Morgan2004; Longhurst, Reference Longhurst2004). Munida is the only galatheid genus present in the south-western Atlantic. Fourteen species live between the equator and the La Plata River (36°S) (Melo-Filho & Melo, Reference Melo-Filho and Melo2001), and only two, Munida spinosa (Henderson, 1885) and Munida gregaria (Fabricius, 1793) have been reported to occur on the continental shelf from the La Plata River to the southern tip of South America (Cape Horn, 55°S) (Spivak, Reference Spivak1997; Arntz et al., Reference Arntz, Gorny, Soto, Lardies, Retamal and Wehrtmann1999). Munida spinosa inhabits deep waters on the shelf break and M. gregaria (=M. subrugosa) (Pérez-Barros et al., Reference Pérez-Barros, D'Amato, Guzmán and Lovrich2008) inhabits coastal and shelf waters from Cape Horn to Valdes peninsula (43°S), and deeper shelf waters up to Uruguay (35°S) (Spivak, Reference Spivak1997). Munida gregaria is simultaneously a deposit feeder and a predator on components of the fito and zoo-benthos, and is considered a key species in the Patagonian shelf marine community representing ‘the direct link between the detritus and the top predators’ (Romero et al., Reference Romero, Lovrich, Tapella and Thatje2004) and a potentially exploitable fisheries resource (Lovrich et al., Reference Lovrich, Casalinuovo, Molina, Cárcamo and Pierotti1998; Wyngaard et al., Reference Wyngaard, Iorio and Boschi2001). Larvae of M. gregaria constitute an important fraction of the zooplankton biomass during late winter and spring (Lovrich, Reference Lovrich1999; Dellatorre, Reference Dellatorre2009).
Reproduction of M. gregaria has been studied along its latitudinal distribution (Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002, Reference Tapella, Valiñas, Lovrich, Vinuesa and Romero2005; Vinuesa, Reference Vinuesa2007; Dellatorre & Barón, Reference Dellatorre and Barón2008). Fecundity (considered as number of eggs per clutch) is strongly correlated with female size. In the southern limit of its distribution (Beagle Channel (BC), 55°S), fecundity exceeds 10,000 eggs per brood in females 27 mm in carapace length (CL) (Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002; Vinuesa, Reference Vinuesa2007) while in San Jorge Gulf (SJG) (Central Patagonia, 46–47°S) fecundity is higher than 7000 eggs per brood in females 22–23 mm in CL (Vinuesa, Reference Vinuesa2007). Tapella et al. (Reference Tapella, Valiñas, Lovrich, Vinuesa and Romero2005) reported a similar relationship between female size (CL) and fecundity in those locations (without reporting the regression model) but greater egg size in females from the BC, stating consequently that reproductive output (the ratio between the organic matter of the brood and the female) is higher in the BC than in SJG. However, this study did not consider the possibility of multiple spawning events per female in a single reproductive season (reported later by Vinuesa, Reference Vinuesa2007; Dellatorre & Barón, Reference Dellatorre and Barón2008) to estimate the overall reproductive investment. The seasonal sea surface temperature (SST) regime differs markedly between BC (monthly average ranging 4.2–9.8°C) and SJG (monthly average ranging 6.8–14.2°C) (Dellatorre & Barón, Reference Dellatorre and Barón2008) and probably explains the greater size of eggs in southern populations. Also females reach 27 mm in CL in BC and only 23 mm in CL in SJG (Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002; Vinuesa, Reference Vinuesa2007). This work is the first report about fecundity of M. gregaria from the northern limits of its coastal distribution.
The breeding period of M. gregaria spans from May to September in BC (Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002). In SJG ovigerous females are found in high proportions from June to October (Vinuesa, Reference Vinuesa2007), while in Nuevo Gulf (NG) (42–43°S) they are present from June to November–December, in a surface temperature regime ranging seasonally from 9.6°C to 17.2°C in the coldest and warmest month respectively (Dellatorre & Barón, Reference Dellatorre and Barón2008). Evidence of multiple spawning during the same reproductive season was reported by Dellatorre & Barón (Reference Dellatorre and Barón2008) based on the observation of ovaric re-maturation during the breeding, and preliminary measures of embryonic development duration at fixed temperature conditions. Contrasting temperature regimes along the species range of distribution may affect the embryonic development time (Wear, Reference Wear1974) and other reproductive traits like reproductive seasonality, reproductive investment, fecundity and breeding cost (Fernandez & Brante, Reference Brante, Fernández, Eckerle, Mark, Pörtner and Arntz2003).
Embryonic development has not been extensively studied in galatheid crabs (Wear, Reference Wear1974; Van Dover & Williams, Reference Van Dover, Williams, Wenner and Kuris1991) probably because most of the species belonging to this family inhabit deep habitats (Melo-Filho & Melo, Reference Melo-Filho and Melo2001) and because abortion of the entire egg masses during stressing manipulation is a common behaviour (G. Lovrich, personal communication). Even though the embryonic development of M. gregaria has not been described in detail, incubations in aquaria have allowed to estimate a duration of approximately 29 days at constant temperature of 11°C (Dellatorre & Barón, Reference Dellatorre and Barón2008). The description of morphological changes of embryos and its chronology can assist in future estimations of breeding period, number of spawnings and hatching periods of M. gregaria, in different locations (Nagao et al., Reference Nagao, Munehara and Shimazaki1999).
The present work has two objectives: (1) to analyse the size–fecundity relationship of M. gregaria from northern Patagonia and to compare it with those observed in areas with contrasting thermal regimes: BC and SJG; and (2) to describe the complete embryonic development of the species, illustrating the major embryonic features along with the chronology of their appearance.
MATERIALS AND METHODS
Munida gregaria specimens were captured in August 2004, from June to September 2005, in July 2006, and in July 2007, using cylindrical collapsible crayfish traps deployed for 1–2 days on the muddy sea bottom in two different sites (with 18 and 25 m depth and separated by 2 km) in Nueva Bay (42.75°S 65.00°W, within Nuevo Gulf) (Figure 1). Fecundity, considered as the number of eggs per clutch, was measured in 64 ovigerous females from all the sampled months, except in July 2007. Total number of eggs in the clutch was counted under binocular microscope on 41 females. In the other 23 specimens, the egg clutch was carefully separated from the female by cutting the base of each pleopod, and eggs were individually removed from the pleopod setae with fine tipped forceps. Eggs were sieved and placed on blotting paper to remove excess water. For each female, the total clutch was weighed to the nearest 1 mg using an analytical balance. Then, 10 to 30% of the eggs in the clutch were taken apart, weighed and counted under binocular microscope, and the number of eggs in the entire clutch was extrapolated. Only recently spawned females (first two stages of embryonic development) captured during the first half of the reproductive season (from June to September) were analysed, since multiple spawning may occur during the reproductive season (Dellatorre & Barón, Reference Dellatorre and Barón2008) and there might be intra-seasonal variations in fecundity.
Fig. 1. Southern coast of South America and Argentine Continental Shelf. Latitude (south) and longitude (west) are at right and lower margins respectively. LPR, La Plata River; VP, Valdes Peninsula; NG, Nuevo Gulf; NB, Nueva Bay; SJG, San Jorge Gulf; MS, Magellan Strait; BC, Beagle Channel; CH, Cape Horn.
To study the embryonic development, 18 ovigerous females recently spawned (10 females sampled in July 2006 and another 8 females sampled in July 2007) were maintained in 10 l plastic aquaria filled with seawater, and with constant aeration. Seawater temperature was maintained at 11°C (±1°C) because the mean surface temperature on coastal northern Patagonia (43°S) during the breeding season (June to December) is 11.5°C (SST data obtained from satellite estimations for the period 1987–1998; AVHRR Oceans Pathfinder NOAA–NASA). The crabs were fed ad libitum with hake (Merluccius hubbsi) meat every 2–6 days, and seawater was replaced every 4–6 days.
Every 2–6 days, a sample of five to ten developing embryos (representative of the entire clutch) was removed from the pleopods of each female using fine-tipped forceps, examined under a dissecting microscope at 25X magnification, and classified into five categories of embryonic development (Table 1) (modified from Dellatorre & Barón, Reference Dellatorre and Barón2008). The moment of change of embryonic stage was considered in the middle of the period between observations in which such change was produced. Stage duration was expressed in days and the average of stage duration was computed from all the females that survived each entire stage (the average of the complete embryonic development duration was computed from three females that survived until hatching).
Table 1. General features of Munida gregaria embryos at each stage of embryogenesis.
In order to describe morphological variations of embryos in detail, some of them were observed under a Leica DFC280 light microscope equipped with differential contrast interference optics. The external envelope (chorion) of embryos was removed with fine-tipped forceps to obtain the best view of the embryonic structures. Some of the appendages were removed for observation at 200X in the stereoscopic microscope. Digital images were obtained with Leica Application Suite (Version 2·5 R1) software.
Another twenty-five females were captured in July 2007 breeding embryos in different stages of development, and approximately fifteen living eggs from each of these females were measured using an ocular micrometer under 50X magnification. For those measures the following dimensions were defined: (1) dorso-ventral diameter: between animal pole (where embryo develops) and vegetal pole (the opposite); and (2) antero-posterior diameter: perpendicular to dorso-ventral diameter, anterior part of the embryo is identified by the ocular globe. The egg volume was estimated using the formula of an ellipsoid volume:
where V is the egg volume, r 1 the half of the major diameter (antero-posterior) and r 2 the half of the minor diameter (dorso-ventral) (García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2004). On each stage, the antennules, antennae, eyes, mouth parts, abdomen, telson, carapace, rostral spine, posterior spines and chromatophores were observed and described. The terminology used for embryonic appendages follows that used by Clark et al. (Reference Clark, De Calazans and Pohle1998) and Roberts (Reference Roberts1973) for larval appendages.
RESULTS
Ovigerous females of M. gregaria ranging 9.3–20.5 mm in CL (N = 64) carried between 32 and 7065 eggs attached on their pleopods. The total number of eggs in the clutch (F) was strongly correlated with carapace length (CL, mm) (R = 0.638) (Figure 2). The following model was fitted to the relationship by least square regression:
Fig. 2. Relationship between size and fecundity. The black regression line represents the model adjusted to data from Nuevo Gulf, while the dotted line shows the regression model fitted by Tapella et al. (Reference Tapella, Lovrich, Romero and Thatje2002) to data from the Beagle Channel. The extension of both lines on the abscise axis represents the size-range sampled.
The complete embryonic development lasted on average 27.9 days (Table 2). Freshly spawned oocytes were adhered to the setae along each one of the four female pleopods. Stages of the embryonic development observed in Munida gregaria (Table 1) were as follows:
Table 2. Duration of embryonic development stages of Munida gregaria and size of the eggs during embryogenesis. The range of stage duration is based on estimations from different breeding females (number of females in parentheses). *, major diameter (antero-posterior) is reported. The egg volume was estimated from antero-posterior and dorso-ventral measures of diameter assuming an ellipsoidal shape. Egg diameter measures were taken in eggs from different breeding females (number of females in parentheses). (1) and (2) egg diameter (µm) from San Jorge Gulf (Vinuesa, Reference Vinuesa2007) and Beagle Channel (Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002) respectively, standard deviation in parentheses.
Stage I
This stage lasts between 2 and 3.5 days (Table 2). It begins immediately after fertilization, when the first divisions of the oocyte occur (Figure 3A). Two mutually adherent coverings surround the embryo, forming the funiculus that holds the embryo to the pleopod setae (Figure 3A). The egg is spherical in shape, with an average diameter of 530 µm (Table 2) and presents a uniform dark green coloration. In the first hours after extrusion the egg chorion does not recover a spherical shape easily when crumpled (during manipulation with forceps). Later, the cellular division forms the blastomers (Figure 3A). By the end of this stage, the morula can be observed (Figure 3B). Within the oocyte, vitelline granules are not common, and the embryo is not detectable.
Fig. 3. Photographs of embryonic stages I (A, B), II (C, D) and III (E, F) of Munida gregaria. Bt, blastomers; Ep, embryonic primordium; An, antennule; AnI, antenna; Og, ocular globe; Mxp, first maxilliped; Ab, abdomen; T, telson; V, vitellum.
Stage II
This stage lasts 7–10 days and rounded eggs have a diameter similar to the previous stage (Table 2). The embryonic primordium appears as a small transparent zone at the animal pole of the zygote (Figure 3C), representing the ventral region of the embryo. This area begins to deepen and extends in an antero-posterior direction. At the beginning of this stage, the outlines of the ocular globe can be discerned as flat smooth structures projecting dorso-laterally and the bud of the antennules and antennae can be observed as two pairs of extensions (Figure 3D). At the middle of this stage two pairs of mouth parts can be observed, all the appendages are flattened and uniramous (Figure 3E). The cephalic zone then arises, bearing four pairs of appendages. The posterior thoraco-abdominal plate presents a short, wide projection, with its distal extremity bifurcated into two lobes of the future telson.
Stage III
The duration of this stage is between 5 and 8 days (Table 2). The beginning of this stage is defined as the moment at which embryonic tissues occupy more than half of the egg perimeter in lateral view (Figure 3F). The embryo grows along the antero-posterior axis, the egg covering aquires an ellipsoidal shape and increases the maximum diameter (Table 2). The cephalothorax and abdominal regions can be clearly differentiated (Figure 3F). The appendages have grown and differentiated markedly. The antennule and antenna grow caudally parallel with each other (Figure 4A). The antennule is uniramous and tubular with a short middle seta and three long terminal setae on the exopod, while the endopod is absent. The biramous antenna is the most developed of the appendages; hence it is longer than the antennule. The endopodite is short, wide, with a single, subterminal, thickened seta. The exopodite is long, wide with nine setae, short and firm, forming a straight line on the lateral–ventral border toward the terminal zone. The mouth appendages: mandibule, maxillule and maxilla, are visible in this stage of development between the antenna and the first maxilliped. All of the mouth parts have the same length and they do not overlap each other. The two pairs of maxillipeds are biramous and have grown markedly in length toward the anterior zone of the embryo, presenting endopodite and exopodite of the same length. The endopodite possesses three long terminal setae (Figure 3F). One rostral and two posterior spines can be observed. The rostral spine is curved ventrally toward the caudal region along the mid-line of the embryo. The posterior spines arise from the posterior–ventral border of the cephalothorax in a cephalic direction. At the end of this stage, the abdomen remains folded ventrally over the thoracic zone growing between the appendages toward the insertion of the mouth appendages. Five segments can be observed, the sixth is still fused to the telson. The lobules of the telson present a sharp end and six short terminal spines.
Fig. 4. Photographs of embryonic stages III (A), IV (B) and V (C, D) of Munidagregaria. F, funiculus; Ct, chromatophores; Sp, telson spines; Op, ocular pigments; *, indicates chromatophores. Other letters represent the same as in Figure 3.
Stage IV
This stage lasts between 4 and 6 days, eggs are ellipsoidal and maximum diameters continue increasing (Table 2). It begins with the appearance of pigmentation in the ocular globe and the chromatophores (Figure 4B). This pigmentation can be observed in the area posterior to the ocular globe, as a dark curved line (Figure 4B). The reddish chromatophores are located in different parts of the embryos (Figure 4B). It can be observed with a star-shape at the bases of the mandibule and at the base of the maxilliped. Elongated chromatophores can be observed around the digestive tube in the 2° and 5° segment of the abdomen at the beginning of this stage (Figure 4C), and in all abdomen segments at the end. The telson presented two pairs of star-shaped chromatophores around the anus (Figure 4C).
All the buccal appendices overlap. The mandibule is surrounded by the maxillule and maxilla. The mandibule is tubular and short, and presents molar process. The maxillule shows an endopodite with four terminal setae, a basal endite with six setae and a coxal endite with five setae. The maxilla presents four endites with respectively 6, 4, 4 and 4 setae. Seven setae can be observed on the palp and five terminal setae on the scaphognathite. Segmentation can be observed on the maxillipeds. The endopodite presents five segments and the exopodite two segments. In the telson, the six pairs of spines are increasing in length at the same rate along the external border (Figure 4C). Cardiac movements can be clearly observed in the dorso-posterior region of the cephalothorax.
Stage V
The duration of this stage is between 3 and 4 days (Table 2). The embryo is almost completely formed and ready to hatch (Figure 4D). Green vitellum droplets are located in the dorso-medial region of the cephalothorax and its diameter in lateral view is shorter than half of the egg diameter in the same axis (Figure 4D). The eyes are large and sessile, and the reddish-brown pigmentation of the ocular globe is oval in shape (Figure 4D). By the end of this stage the embryos are easily separated from the pleopods and their external covering becomes labile. Hatching is induced by touching the covering with a dissecting needle. The uniramous antennule has no segmentation and displays a plumose subterminal seta and 5 terminals setae (2 short and 3 long). The biramous antenna has a short and wide endopodite with a plumose subterminal seta and 8 lateral spines. The nine setae of the exopodite are plumose, forming a straight line on the lateral–ventral border toward the terminal zone. The first maxilliped presents a short coxa with 2 plumose setae, a long basis with 4 groups of 3 plumose setae in the internal margin. The endopodite presents 3, 2, 1, 2 and 5 plumose setae in the internal margin. The exopodite presents 4 terminal plumose setae. The second pairs present a short coxa without setae and a long base with 3 distal plumose setae. The endopodite presents 2, 2, 2 and 5 plumose setae in the internal margin. The exopodite presents 4 terminal plumose setae. The third pair is rudimentary. No pereopods can be observed.
The abdomen extends to the insertion of the antennule (Figure 4D). It has 5 segments with the sixth fused to the telson. The segments present a row of small posterior–dorsal spines. The third and fourth segment presents a pair of lateral spines. The telson is bi-lobulated with 6 pairs of posterior plumose spines. Sporadic movement of the appendages and abdomen can be observed, becoming more frequent near the time of hatching. At the end of the embryonic development the eggs are ellipsoidal in shape, their maximum diameter is about 30% greater than in recently extruded eggs, and the volume has increased by approximately 50% (Table 2).
DISCUSSION
The embryonic stages of the decapods can be distinguished based on different criteria (García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2004, Reference García-Guerrero and Hendrickx2006; González-Pisani et al., Reference González-Pisani, Rubilar and Dupré2009). To differentiate stages in this study, we used the appearance of different morphological characters, primordia of appendages, pigmentation on the ocular globe and chromathophores, as well as the complete formation of the embryo. Other authors have defined embryonic stages of development based on: (1) the index of ocular globe pigmentation, which is useful in species where pigmentation occurs gradually during a long period (over more than 70% of the total incubation period) (Perkins, Reference Perkins1972); and (2) the changes in the form of some structures or appendages (Tavonatti, Reference Tavonatti1998; Tavonatti & Dupré, Reference Tavonatti and Dupré2001; Dupré, Reference Dupré2003). In M. gregaria, those criteria were not used because the ocular pigmentation does not appear until the end of stage III, when more than half of the total incubation period has passed and also because the ocular pigmentation completes in approximately 5 days (less than 20% of the total development time). On the other hand, development of the appendages of this species does not allow defining embryonic stages of development as in the case of the antennae of Jasus frontalis (Tavonatti & Dupré, Reference Tavonatti and Dupré2001).
Some of the embryonic structures of M. gregaria are similar to those observed in embryos of other anomuran crabs, as for example the rostral and postero-lateral spines of the carapace (Boschi et al., Reference Boschi, Scelzo and Goldstein1967; García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2004; Fujita & Osawa, Reference Fujita and Osawa2005; García-Guerrero et al., Reference García-Guerrero, Cuesta, Hendrickx and Rodríguez2005; Hernández et al., Reference Hernández, Graterol, Magán, Bolaños, Lira and Gaviria2005; Cuesta et al., Reference Cuesta, García-Guerrero, Rodríguez and Hendrickx2006) and the I, II and III maxillipeds, the last appearing as rudiments at the end of the embryonic development (García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2006). The eggs have an oval shape at the end of the development, probably caused by the increase in length of the appendages and the intake of water to the intra-chorionic space (Pandian, Reference Pandian1970; Lardies & Wehrtmann, Reference Lardies and Wehrtmann1996; López et al., Reference López, Jeri, González and Rodríguez1997; Hernández & Palma, Reference Hernández and Palma2003; García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2006; Surot-Navarro, Reference Surot-Navarro2006), in contrast to other brachyuran embryos that have spherical eggs with minor variations in shape throughout development (Nagao et al., Reference Nagao, Munehara and Shimazaki1999; Pinheiro & Hattori, Reference Pinheiro and Hattori2003; García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2004).
The embryonic development of species belonging to the families Galatheidae and Chirostylidae has not been investigated to date. Of the other two families included in the superfamily Galatheoidea (Martin & Davis, Reference Martin and Davis2001), the Aeglidae are strictly freshwater organisms (Martin & Davis, Reference Martin and Davis2001) and osmoregulatory adaptations (Charmantier, Reference Charmantier1998) may generate development features that will not be compared with those observed in this work. The embryonic development of the Porcellanidae has been extensively studied (Gore, Reference Gore1968, Reference Gore1971; Hernández et al., Reference Hernández, Graterol, Alvarez and Bolaños1998; García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2004, Reference García-Guerrero and Hendrickx2006; Hernández et al., Reference Hernández, Graterol, Magán, Bolaños, Lira and Gaviria2005; González-Pisani et al., Reference González-Pisani, Rubilar and Dupré2009). Embryonic development of M. gregaria differs from that of porcellanids in some aspects. Antennae exopodites of M. gregaria are wide and flat with short and firm setae, becoming long and flexible by the end of development. In contrast, in the porcellanids exopodites are long and tubular, with short and firm setae forming a straight line on the lateral–ventral border toward the terminal zone (González-Pisani et al., Reference González-Pisani, Rubilar and Dupré2009). In M. gregaria all chromatophores appear simultaneously, when approximately 70% of the embryonic development time has elapsed. In change, porcellanid chromatophores appear a few days after the eye pigments (Gore, Reference Gore1968, Reference Gore1971; Hernández et al., Reference Hernández, Graterol, Alvarez and Bolaños1998; García-Guerrero & Hendrickx, Reference García-Guerrero and Hendrickx2004; García-Guerrero et al., 2006; González-Pisani et al., Reference González-Pisani, Rubilar and Dupré2009). At the end of development of M. gregaria third maxillipeds are rudimentary while in porcellanids these appendages are not detectable. Moreover, in porcellanids the initially bilobulated telson acquires a spatulate shape, with a convex distal border, while in M. gregaria it remains bilobulated, each lobule displaying a sharp distal portion at the end of development. This last feature, and the cephalothoracic spines, allows the clear distinguishing of larval forms of both families in plankton samples (Dellatorre, Reference Dellatorre2009).
The length of the embryonic development of M. subrugosa (=M. gregaria) has been a matter of speculation for several years. Rodríguez & Bahamonde (Reference Rodríguez, Bahamonde and Arana1986) expected that it would last 8–9 months in the Magellan Strait (South America, 54°S) and Tapella et al. (Reference Tapella, Lovrich, Romero and Thatje2002) stated that females hold their eggs over a period of three to four months in the Beagle Channel. Based on the monthly variations of the proportions of ovigerous females and stages of embryonic development, Vinuesa (Reference Vinuesa2007) proposed that embryonic development would take 90 days in females spawning in July and 70–80 days in females spawning in September in San Jorge Gulf. Dellatorre & Barón (Reference Dellatorre and Barón2008) provided the first data on the length of the embryonic development, ranging in the laboratory from 26 to 29 days at 11°C constant temperature. Our results confirm that M. gregaria embryos take approximately 28 days from fertilization to hatching at 11°C constant temperature.
The embryonic development of galatheid crab species has not been described until present, and little information has been published in regards to its length (Wear, Reference Wear1974; Van Dover & Williams, Reference Van Dover, Williams, Wenner and Kuris1991; Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002; Vinuesa, Reference Vinuesa2007). Temperature is a key abiotic factor affecting the incubation of embryos. Typically, increased temperatures shorten the duration of embryogenesis in decapod crustaceans (Perkins, Reference Perkins1972; Wear, Reference Wear1974; Petersen, Reference Petersen1995; Tong et al., Reference Tong, Graeme, Pickering and Paewai2000; Smith et al., Reference Smith, Ritar, Thompson, Dunstan and Brown2002; Hamasaki, Reference Hamasaki2003; Wehrtmann & López, Reference Wehrtmann and López2003; Stevens et al., Reference Stevens, Swiney and Buck2008). Some of those authors developed predictive models of embryonic development time as function of water temperature (Wear, Reference Wear1974; Hamasaki, Reference Hamasaki2003; Wehrtmann & López, Reference Wehrtmann and López2003). The work of Wear (Reference Wear1974) is the only one dealing with galatheid species in temperate waters, and this makes the work adequate for comparison with our results. This author developed an equation to estimate the duration of embryonic development of Galathea dispersa at different temperatures based on aquarium incubations at four different constant thermal regimes (at 9, 12, 18 and 21°C). According to this equation, G. dispersa would incubate their embryos during 37 days at 11°C constant temperature. Assuming that the curve of the temperature–incubation period relationship has a similar shape for M. gregaria, we can recalculate the ‘biological zero’ or alpha parameter (Wear, Reference Wear1974) and estimate the duration of embryonic development of M. gregaria at average ambient temperatures in the species habitat along its latitudinal distribution range along the south-western Atlantic coast. Under this assumption, the embryonic development of M. gregaria would last between 35 and 42 days in the San Jorge Gulf, where sea surface temperature (SST) ranges between 8.5 and 9.5°C during reproductive season, and approximately 75 days in the Beagle Channel where SST averages 5.5°C during reproductive season (SST data obtained from satellite estimations for the period 1987–1998; AVHRR Oceans Pathfinder NOAA–NASA).
Egg size (diameter and volume) is smaller in Nuevo Gulf than in San Jorge Gulf (Vinuesa, Reference Vinuesa2007) and the Beagle Channel (Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002, Reference Tapella, Valiñas, Lovrich, Vinuesa and Romero2005). This probably reflects species adaptations to contrasting temperature regimes in those locations (Fischer, Reference Fischer2009). Egg size increments in higher latitudes would result in a shortened larval development and would reflect an adaptation to a short temporal window with nutritional and physical parameters optimal for larval survival. This is generally known as the Thorson's rule (Thorson, Reference Thorson1950). On the other hand, the reduction in egg size increases surface:volume ratio, and this could compensate for the increases in breeding cost (mostly derived from ventilation of egg clutch) when temperature increases (Fernandez & Brante, Reference Fernández and Brante2003). Intraspecific latitudinal variations of egg size has been recorded in Cancer setosus (Brante et al., Reference Brante, Fernández, Eckerle, Mark, Pörtner and Arntz2003), Pinnaxodes chilensis (Lardies & Castilla, Reference Lardies and Castilla2001) and other brachyuran species (Brante et al., Reference Brante, Cifuentes, Pörtner, Arntz and Fernández2004). Smaller and lighter embryos at the lower latitudes may reflect thermal adaptation to higher mean temperatures (Brante et al., Reference Brante, Fernández, Eckerle, Mark, Pörtner and Arntz2003; Fischer, Reference Fischer2009). Those increments in egg size with latitude would increase the length of embryonic development and bias our previous estimations (Wear, Reference Wear1974). Clutch size of decapods is limited morphometrically by the space available for yolk accumulation in their rigid cephalothorax (Hines, Reference Hines1982, Reference Hines1992), which predicts a positive relationship between fecundity and female size. This prediction is true for M. gregaria along its latitudinal distribution (Tapella et al., Reference Tapella, Lovrich, Romero and Thatje2002, Reference Tapella, Valiñas, Lovrich, Vinuesa and Romero2005; Vinuesa, Reference Vinuesa2007; present study). Considering this morphological limitation for the clutch size and assuming the same proportion of energy invested in egg production (for a single clutch), an inverse relationship between egg size and fecundity can be expected (Hines, Reference Hines1982; Clarke, Reference Clarke1987). This tradeoff between egg size and fecundity occurs in M. gregaria, as in other decapod species like porcellanid crabs (Lardies & Wehrtmann, Reference Lardies and Wehrtmann1996), commensal crabs (Lardies & Castilla, Reference Lardies and Castilla2001) and shrimps (Lardies, Reference Lardies1995); but in some other cases increments in egg size and fecundity are observed simultaneously (Jones & Simons, Reference Jones and Simons1983).
Temperature is the most important factor affecting embryonic development and reproductive strategies in marine invertebrates (Vernberg, Reference Vernberg1962). High temperature may increase growth rate and other metabolic activities reducing available energy for reproduction of some decapod species in lower latitudes (Jones & Simons, Reference Jones and Simons1983) impeding the tradeoff between egg size and fecundity predicted by Hines (Reference Hines1982). This may not be the case of M. gregaria and fecundity (estimated as the number of eggs per clutch) seems to be constrained only by morphological factors. Nevertheless, more information is necessary to estimate the overall reproductive investment of M. gregaria along its latitudinal distribution considering successive spawnings during the same reproductive season (Dellatorre & Barón, Reference Dellatorre and Barón2008) and the breeding costs (Fernández et al., Reference Fernández, Bock and Pörtner2000).
ACKNOWLEDGEMENTS
Experiments were performed in the Centro Nacional Patagónico, and partially funded by projects PICT 14700 (ANPCyT) and PIP 5835 (CONICET). The authors were fully supported by doctoral fellowships from the CONICET. We are grateful to our advisor Dr Pedro Barón, who has guided our doctoral research and improved an early version of this manuscript. An anonymous referee greatly improved this manuscript with constructive comments.
