Redescription of megalopa and juvenile development of Pachygrapsus gracilis ( Decapoda : Grapsidae ) from the Amazon region , reared in the laboratory

This present study re-describes the megalopa stage and provides detailed morphological descriptions and main growth changes observed in stages I through VII of the juvenile instars of the dark shore crab Pachygrapsus gracilis (Saussure, 1858), from the Amazon region. The specimens in this study were reared in the laboratory and the megalopae were collected at Ajuruteua beach in northeastern Pará, Brazil. Previous studies had described the megalopa of P. gracilis from Mexican waters, as well as those of Pachygrapsus transversus (Gibbes, 1850) and Pachygrapsus marmoratus (Fabricius, 1787). A comparison between the Mexican and Amazonian populations of P. gracilis revealed significant morphological differences. The main difference is the presence of 3 elongated setae on the 5th pereiopods of individuals of the Amazonian population. The setal number and their arrangement in the appendages also differed. In P. gracilis, the male and female genital openings are observed from the juvenile instar III, whereas differentiation in male pleopods is observed only in juvenile instar V. In females, the pleopods undergo rapid differentiation during juvenile instar VI. These morphological comparisons and other observations on development are briefly compared and discussed with reports for other species.

The dark shore crab Pachygrapsus gracilis (Saussure, 1858) has a wide distribution, occurring in the western Atlantic from the Caribbean to Texas, French Guiana, Brazil (Trinidad and from Ceará to Rio Grande do Sul) and Argentina, In the eastern Atlantic the species occurs from Senegal to Angola (MELO 1996, POUPIN et al. 2005).
Pachygrapsus Randall, 1839 megalopae was described in the past from individuals collected in the natural environment: P. marmoratus (Fabricius, 1787) by GUERAO et al. (1997), P. transversus (Gibbes, 1850) by FLORES et al. (1998) and P. gracilis by CHÁRAZO-OLVERA & ROCHA-RAMÍREZ (2007).These studies used specimens collected from the environment, probably due to the difficulty of obtaining the stage of megalopae under laboratory conditions.Only two studies on the juvenile growth and morphology of this group are available in the scientific literature: GUERAO et al. (1997) for P. marmoratus and FLORES et al. (1998) for P. transversus.GUERAO et al. (1997) described the morphology of the carapace, pereiopods and third maxillipeds of the juvenile instar I of P. marmoratus.Their data provide information on the size of the carapace up to the juvenile IV.FLORES et al. (1998) described in detail the megalopa, the first crab instar, and the juvenile development instars of P. transversus giving emphasis on the process of pleopod differentiation up until juvenile instar VII.
Studies on crab growth are important because some morphological characteristics, including secondary sexual characters, which arise at the juvenile phase, are useful in the identification of species, as well as age and sex determinations.
The morphological characteristics found in crustaceans have contributed significantly to the understanding of the taxonomic, systematic and phylogenetic nature of crustaceans.They have also generated information which has helped the understanding of their biology and assisted population management (BARUTOT et al. 2001).However, morphological descriptions of crustaceans must be accurate, and describe the setal arrangement, the number of the juvenile appendages and other structures correctly.
In some cases, intraspecific variation has been found in individuals of different populations.Nelice M. Batistelli (unpubl.data), for example, found discrepancies in setal number and setation arrangements in re-descriptions of various fiddler crab larvae.More studies on the morphological stages and new descriptions of larval and juvenile stages of crustaceans are necessary in order to clarify these differences.
Due to the scarcity of studies detailing the morphological development of crustaceans and the growth of these animals, it is necessary to conduct further studies on the growth and morphology of juvenile instars.The present study re-de-ZOOLOGIA 28 (4): 465-478, August, 2011 scribes the megalopa stage of P. gracilis inhabiting the Amazon region, and describes important morphological changes observed during the growth from juvenile instars I to VII, of.

MATERIAL AND METHODS
In February 2008, 116 megalopa were obtained at Ajuruteua (00º49'20.2"S,46º36'02.2"W)Beach, municipality of Bragança (northeastern Pará state, Brazil).The specimens were collected off of mangrove leaves, pieces of trunks and roots brought by the tide.They were identified in the laboratory when they reached the juvenile instar.Specimen identification followed RODRIGUEZ (1980), MELO (1996) and POUPIN et al. (2005).
In the laboratory, specimens were placed individually in plastic recipients with sea water (salinity 30) and a small rock shelter.The temperature was maintained at 26.5 ± 1.14ºC, photoperiod regime (12:12 h/light:dark cycle) and pH 8.1.Every two days the water of the recipients was changed.The individuals were fed live Artemia nauplii daily and were monitored twice a day.The exuviae and dead animals were preserved in glycerol+ethylic alcohol 70% (1:1) solution.
At least 10 specimens of each stage were dissected with fine needles (BD Ultra-Fine ® , 12.7 X 0.33 mm) under an optical Zeiss (Axioskop 40) microscope equipped with an ocular micrometric.The appendages were extracted and transferred to another slide where they were stained with a solution of methylene blue 1% (sometimes was used).The terminology used in the descriptions follows POHLE & TELFORD (1981), GUERAO et al. (1997), FLORES et al. (1998), GARM (2004), and CHÁRAZO-OLVERA & ROCHA RAMÍREZ (2007).The studied specimens were deposited at the crustacean collection of Museu Paraense Emílio Goeldi, under the catalog number MPEG-817.
The individuals were reared until juvenile instar VII.During rearing, data on survival and intermolt periods were obtained for individuals in each juvenile instar.The carapace width (CW) was considered as an independent variable or as a reference (in axis x) because of its greater amplitude and because it is the best representative dimension of organism size (HARTNOLL 1982).This variable was used for calculating the percentage of molt increment and absolute increment after each ecdise.The abdomen width of male and female specimens was obtained by measuring the 4 th abdominal somite.
The absolute increment (AI) after each ecdise was obtained using the following equation: AI = CW 2 -CW 1; where CW 1 is the carapace width in pre-molt and CW 2 the carapace width in post-molt.
The morphometric relationship between the carapace length (CL) and carapace width (CW) throughout the juvenile development was obtained by linear regression.

MEGALOPA
A total of 89 megalopae molted into the juvenile instar I.The remaining individuals died during the experiment.These were preserved.The first molting was recorded on the second day, and the last on the 18 th day of culture after collections in the field (Fig. 1).Large frequencies of metamorphosed megalopae were recorded on the third and fourth days of rearing.The mean intermolt period was 5.72 ± 4.45 days.

JUVENILE GROWTH
During the culture, the mortality of juveniles was quite low: only three specimens died in the juvenile instars I, II and III, respectively.The mean intermolt period gradually increased during the juvenile development and after consecutive molts (Fig. 2).In juvenile instar I, the intermolt period mean was 6 ± 3 days whereas in the juvenile instar VII it averaged 20 ± 12 days.
The carapace length of the juvenile instar I (1.96 ± 0.07 mm) was similar to that found for megalopa (1.97 ± 0.04 mm).The growth (carapace width) of each subsequent juvenile instar showed an increase during the days of culture and molts (Fig. 3).In the juvenile instar I, the mean carapace width was 2.21 mm and in the juvenile instar VII the mean carapace width was 7.42 mm.The carapace width of male and female crabs was similar during juvenile growth until the VII instar (Fig. 4).
The percentage of molt increment showed a reduction at each molt event (Fig. 5).From the juvenile instar I to juvenile instar II, the mean rate of increment was 31.13 ± 8.75%, whereas from the juvenile instar VI to VII it was 18.51 ± 4.16%.The absolute molt increment increased at each molt event; from juvenile instar I to II the mean was 0.68 mm and from the juvenile instar VI to VII the mean was 1.17 mm (Fig. 5).
The width of the abdomen in juvenile forms increased during juvenile growth (Fig. 6).However, the width of the abdomen of males and females presented differences only in the juvenile instars VI and VII (Fig. 6).In the juvenile instar I, the mean abdomen width in males and females was similar, 0.70 ± 0.04 mm and 0.60 ± 0.04 mm, respectively.In juvenile instars VI and VII, these values were very different for both sexes; females had 2.10 ± 0.4 mm and 3.20 ± 0.76 mm of abdomen width, while males presented 1.70 ± 0.19 mm and 2.10 ± 0.15 mm, respectively.

MORPHOLOGICAL DESCRIPTION Megalopa
Carapace (Fig. 7): length: 1.97 ± 0.04mm; width: 1.39 ± 0.04 mm; subretangular; surface showing four distinct lobes on the anterior portion and few small setae; middle and posterior portions fringed with numerous small marginal setae; anterolateral region showing a small enlargement; rostrum elongated, triangular and strongly deflected ventrally (Fig. 8); eyes large, ocular peduncle well-developed; the presence of the dorsal organ was not detected.

6
Figures 5-6.Pachygrapsus gracilis: (5) mean percentage of molt increment and mean absolute molt increment for each juvenile instar; (6) mean abdomen width of males and females in each juvenile instar.
Mandible (Fig. 13): uncalcified, symmetrical with bladelike margin, incisive and molar processes indistinguishable; mandibular palp 3-segmented with 10 setae on distal segment, some setae are evidently pappose and some appear to be plumodenticulate but such pattern of setules is not obvious.
First maxilliped (Fig. 16): coxal endite with four terminal simple setae, one terminal plumose setae and 12 subterminal simple setae; basal endite with 13 terminal simple setae and two subterminal simple setae; endopod unsegmented with two distal simple setae; exopod 2-segmented with three simple setae on proximal and four terminal plumose setae on distal segment; well development epipod with 10 long simple setae.
Pereiopods (Fig. 7): well-developed with few setae distributed on the segments (as illustrated); P1 -Chelipods similar in shape and size; dactyl lacking teeth; propodus showing two truncated teeth on the posteroventral margin, propodus with one strong ventral spine close to dactyl (Fig. 9); P2-P4 -Dactyl ventrally curved with acute extremity, showing strong teeth on the ventral margin; P5 -Dactyl with three long denticulate setae and a strong terminal spine, ventral margin lacking teeth (Fig. 10).
Mandible (Fig. 23): calcified, asymmetrical, left mandible with blade-like margin, incisive and molar processes indistinguishable, mandibular palp 3-segmented with four and 14 plumose setae on 2 nd and 3 rd segments; right mandible incise process showing two blunt teeth and blade-like molar process, mandibular palp 3-segmented with six and 16 plumose setae on 2 nd and 3 rd segments.
Abdomen (Fig. 29): 2 nd to 5 th somites with characteristic transverse striate; telson, a triangular in shape, bordered with 25 simple setae and four dorsal simple setae.Pleopods: were not observed in this stage.Uropods absent (not observed in all juveniles stages).

Development of other juvenile instars
The main morphological changes were observed in the developmental sexual dimorphism in which they are described as follow.No significant morphological changes were observed in the carapace of subsequently juvenile instars (Fig. 30).
Juvenile II.Females: showing four pairs of pleopods (2 nd to 5 th ), each one biramous and rudimentary, but due to the

V VII VI
fact that the illustrations were made from exuviae, the pleopods in juvenile II were damaged and they were not illustrated.Males: pleopods were not detected.
ing setae.The second pair of pleopods weren't detected in any juvenile instar (Fig. 33).
Juvenile V. Females: second, third, and fourth pairs of pleopods with endopod 3-segmented showing 7, 4, 5 simple setae on second pair, 4, 5, 3 on third pair and 3, 4, 5 on fourth pair; endopod of the fifth pair 2-segmented with 5, 4 simple Juvenile IV.Females: pleopods larger than in juvenile III; endopods 2-segmented, endopod of the second pair of pleopods with 3, 1 simple setae on distal and proximal segments, respectively, endopods of the third pair with one simple setae on each segment, endopods of the fourth pair lacking setae and endopod of the fifth pair with 2, 0 simple setae on distal and proximal segments, respectively; exopods unsegmented, exopod of the thrid pair of pleopods with two simple setae (Fig. 32).Males: pleopod larger than in the previous instar, lack- setae; exopod of the second, fourth and fifth pairs unsegmented with 10, 14, 15 simple setae; exopod of the third pair 2-segmented with 8, 5 simple setae on proximal and distal segments, respectively (Fig. 32).Males: pleopods entirely developed.Endopod larger than exopod with a small distal groove rounded by numerous setae and with a longitudinal groove; exopod rudimentary (Fig. 33).
Males: similar to previous instar.

DISCUSSION
The analysis and comparisons with previous morphological studies available in the literature for the genus Pachygrapsus (MELO 1996, POUPIN et al. 2005), clearly indicates that P. Gracilis, has some unique features, including gonopod not T-shaped, lateral margins convergent posteriorly and with tooth behind exorbital angle.
The number and type of setae in the appendages of already described species (P.marmoratus, P. transversus and P. gracilis) are quite useful to differentiate among them, mainly the setae on the endopod of maxillule, maxilla and maxillipeds.Morphologically, all Pachygrapsus species studied have large eyes and a well-developed ocular peduncle.Beyond this, other similarities are found in the carapace characteristics, i.e, anterolateral region with lower expansion reaching orbital region and rostrum with a conspicuous medial depression.The carapace of the P. marmoratus and P. transversus are smooth and do not have a protuberance, whereas P. gracilis specimens from the Amazon and Mexico bear distinct lobes on the carapace.
Most of morphological differences are generally observed in the setal number and arrangement of: peduncle and endopod of the antennule, antenna, basal and coxal endites and endopod of the maxillule, maxilla endopod, first maxilliped exopod, fifth segment of the second maxilliped endopod, first segment of the third maxilliped exopod, pleopods and number of long setae on p5 (Tab.I).
The carapaces of P. marmoratus (GUERAO et al. 1997) is larger (CW = 2.10-2.30mm, CL = 3.00-3.20mm), and P. transversus (FLORES et al. 1998) (CW = 1.90 mm, CL = 2.80 mm) megalopa are larger in average than those of our species.The Mexican specimens of P. gracilis (CHÁRAZO-OLVERA & ROCHA-RAMÍREZ 2007) are larger than the Amazonian specimens.The carapace width of specimens from the Amazon was 1.39 ± 0.04 mm and the carapace length was 1.97 ± 0.04 mm, while in Mexican specimens the carapace length and width were 2.20 ± 0.02 mm and 1.57 ± 0.04 mm, respectively.However, many similarities were found between Mexican and Amazonian megalopae (see Tab. I).
The most notable difference found in the morphology of the two populations being compared was the presence of three distinct elongated setae, or feelers, on the fifth pair of the pereiopods,and the absence of those in the Mexican population.The other two species of the genus, P. marmoratus e P. transversus also have these feelers.
Other differences found between specimens from Mexico and the Amazon region of Brazil were as follows: setal number on antennule peduncle and endopod: Amazonian P. gracilis had Juvenile VI.Females: pleopods fully developed; endopod of the second pair of pleopods 5-segmented with 4, 1, 1, 5, 6 simple setae; endopod of the third pair 6-segmented with 5, 1 2, 1, 6, 5 simple setae, respective; endopod of the fourth and fifth pair 3-segmented with 3, 5, 4 and 3, 4, 5 simple setae, JV JIII JIV a total of 11 setae on the peduncle and three distal setae on the endopod, while Mexican P. gracilis had only four setae on the peduncle and four distal setae on the endopod.Other important distinctions were also observed between Amazonian and Mexican species (see Tab. II).The morphological variations found within P. gracilis (Tab.II) when our data is compared with previous studies may be due to mistakes or omissions in other contributions, or variations between crab populations.
Few studies are available in the literature giving detailed description of the juvenile morphology.For instance, for P. marmoratus, only some structures of the juvenile instar I were observed by GUERAO et al. (1997).On the other hand, a comprehensive study, giving a detailed morphological description, was conducted on P. transversus from megalopa to the seven juvenile instar (FLORES et al. 1998).
The morphology of the juvenile instar I of P. gracilis is very similar to the morphology of the adult.The number of setae and their distribution pattern on the cephalic appendages and maxillipeds are typical for adult crabs, as also found by FLORES et al. (1998) for P. transversus.In this juvenile, the two anterolateral teeth of the carapace are already visible in P. gracilis.In P. transversus, the anterolateral tooth is present only in the juvenile instar II.In P. marmoratus, by contrast, the three anterolateral teeth which can be observed in the adult, are already differentiated in the juvenile instar I.This suggests that the anterolateral teeth can be used to differentiate the juvenile I of congenerics.Other similarities and differences found between the species studied are shown on table II.
Similarities in the morphological characteristics and development of pleopods of juvenile males were observed in P. gracilis and Pyromaia tuberculata (Lockington, 1877) (FLORES et al. 2002).In both species, only one pair of pleopods is present during the entire juvenile development.However, in P. transversus, as well as in Sesarma rectum Randall, 1840 (FRANSOZO 1987), Eriphia gonagra (J.C.Differences are observed in the first appearance of the genital orifice of P. gracilis, P. transversus and P. tuberculata (FLORES et al. 2002).In males and females, the genital orifice arises in juvenile instars III, II and IV, respectively.Unfortunately, no comparisons could be made with other species because the authors did not mention details on the sexual dimorphism in these stages.
Studies on morphological descriptions of Pachygrapsus and other brachyuran species are very scarce, although they have been shown to be of great importance for systematic and diagnosis of species at different life stages.Thus, more studies on this topic should be made to contribute to the knowledge of the species biology and growth process.
When juvenile growths are compared, the pattern observed in P. gracilis juveniles is more similar to that of other cultured decapods.Growth patterns ascertained under culture conditions should be interpreted with caution because laboratory conditions and the natural environment are different.Environmental factors not present in controlled conditions, may affect the juvenile growth rate.In such case, the molt increment of specimens reared in the laboratory may be equivalent, and probably smaller than those in the field (HARTNOLL 1982).Increases in the duration of the intermolt period and in carapace size at each subsequent juvenile instar accompanied a decrease in molt size increment during juvenile rearing.The mean intermolting periods of P. gracilis juveniles were: 6.2 ± 3. 4, 8.2 ± 3.4, 12.5 ± 3.2, 11.5 ± 4.9, 12.6 ± 7.5, 15 ± 8, 20.4 ± 12 days in juveniles I, II, III, IV, V, VI, and VII, respectively (Fig. 2).These results could be compared with other congeneric species.However, studies on juvenile growth are scarce, with only one study on this theme reported for a Brazilian species, P. transversus (FLORES et al. 1998).Fortunately, the authors, as in the present study, described the early seven juvenile instars of P. transversus.The mean intermolt duration in each juvenile instar of P. transversus was recorded as 17.9 ± 8, 17.8 ± 6.2, 18.6 ± 4.7, 22.5 ± 5.1, 20.5 ± 7.5, 25.8 ± 4.9, 28 days for juveniles I, II, III, IV, V; VI and VII, respectively.Such results are quite different from those found in the present study, in which intermolt values were considerably lower in all instars (Fig. 2).
A few more studies measuring carapace size in each juvenile instar have been conducted for other species of Pachygrapsus: P. marmoratus (GUERAO et al. 1997), and P. transversus (FLORES et al.ZOOLOGIA 28 (4): 465-478, August, 2011465-478, August, 1998)).The dimensions of carapace length of P. transversus (FLORES et al. 1998) and P. gracilis (present study) were compared.The latter species has a shorter carapace in the first six instars and a longer carapace in the juvenile VII (Fig. 3).On the other hand, P. transversus, showed carapace length of 2.3 ± 0.1 mm, 2.6 ± 0.2 mm, 3.0 ± 0.2 mm, 3.5 ± 0.3 mm, 4.1 ± 0.5 mm, 4.6 ± 0.5 mm, 5.0 ± 0.4 mm in juveniles I, II, III, IV, V, VI, and VII, respectively.In the Mediterranean species P. marmoratus (GUERAO et al. 1997), the largest carapace sizes of all studied species have been observed.In the 4 early instars of this crab, the carapace was 2. 90, 3.20, 4.20, and 5.20 mm long for juveniles I, II, III and IV, respectively.Important information could be made, comparing P. gracilis (present study) with the Mexican P. gracilis but, unfortunately, the data was not presented by those authors.The carapace size in males and females of P. gracilis were compared in present study and no significant difference in size was observed (Fig. 4).
Pachygrapsus gracilis have also been demonstrated to be resilient, having a survival rate up 75%.This result is considerably elevated when compared to Mexican populations which present a survival rate of approximately only 30% (CHÁRAZO-OLVERA & ROCHA-RAMÍREZ 2007).For P. transversus (FLORES et al. 1998), 17 megalopae were collected and only 1 individual survived until juvenile instar VII.
The sexual dimorphism on the abdomen development has not been fully investigated (FLORES et al. 1998(FLORES et al. , 2002)).Such study would be very important because abdominal development is associated with the reproductive activity of males and females.Such analysis is commonly performed in studies on the relative growth of brachyurans.The positive allometric growth of the female abdomen has generally been associated with morphological requirements for reproduction in adults.Thus, large abdomens protect and incubate a larger number of eggs, increasing the reproductive potential of females (HARTNOLL 1974).The abdominal sexual dimorphism in P. gracilis and P. transversus developed in juvenile VI.Females develop a larger abdomen with respect to males this instar (Fig. 6).

Figure 31 .
Figure 31.Pachygrapsus gracilis, genital openings of males and females in the juvenile instar III.Scale bar: 0.6 mm.