Differentially expressed morphological characters depend on sex and ontogenetic stage in the crab Goniopsis cruentata (Crustacea: Grapsidae)

Study area and laboratory procedures

Goniopsis cruentata specimens were collected in a tropical estuarine mangrove system at Tamandaré beach (08°78’28”S 035°10’52”W), northeastern Brazil, a floodplain (Duarte 1993Duarte R.X. 1993. Mapeamento do quartenário costeiro do extremo sul de Pernambuco: Área 05 - Tamandaré. Relatório do curso de geologia. Curso de Geologia. UFPE. p. 86.), with semidiurnal tides and mean amplitudes of 2.4 m and an adjacent seawater temperature ranging between 24°C and 31°C (Moura 1991Moura R.T. 1991. Biomassa, produção primaria do fitoplâncton e alguns fatores ambientais na Bahia de Tamandaré, Rio Formoso, Pernambuco, Brasil. Recife: Universidade Federal de Pernambuco. Dissertação de Mestrado. 290 pp.). The study area is within a marine protected area, the “Costa dos Corais” (Fig. 1). Specimens were captured monthly between February 2019 and January 2020 at their display and/or food capture time, always during the daytime low tide, using nylon line and hookless bait. After capture, the crabs were packed in plastic containers and transported alive to the laboratory.

In the laboratory, the specimens were cryoanesthetized, identified and sexed according to the identification guide proposed by Melo (1996)Melo, G.A.S. 1996. Manual de identificação dos Brachyura (caranguejos e siris) do litoral brasileiro. São Paulo, Plêiade/FAPESP. 604 pp.. Afterwards, they were measured with a precision caliper (0.01 cm) for the dorsal and ventral regions and the cheliped (Fig. 2) and weighed with a precision scale (0.01 g). After the measurements, the specimens were placed on a modelling bed (to remove the influence of their bulging on the image parallelism) opposite a camera with a 50 mm focal length lens attached to a tripod parallel to the foreground and photo-documented in their dorsal, ventral, and cheliped (right and left) regions for further geometric morphometry analysis, as shown in Figure 2, which also illustrates their anatomical landmarks. Images of individuals with malformations or injuries (4% of the images) were discarded from the analyses. Ten landmarks were scanned on the carapace, ten on the pleon and five on the chelipeds (Fig. 2). The landmarks were strategically distributed to better obtain the shape of the animal using the TPSDig program (version 2.10, Rohlf 2006Rohlf F.J. 2006. tpsDig, software program. Version 2.10. Department of Ecology and Evolution, State University of New York at Stony Brook.), based on previous work with other Brachyura (Silva et al. 2018Silva L.N., Almeida P.R.S., Shinozaki-Mendes R.A. 2018. Dimorfismo sexual e alometria ontogenética em Goyazana castelnaui (Crustacea, Brachyura). Iheringia 108: e2018008. https://doi.org/10.1590/1678-4766e2018008, Marochi et al. 2018Marochi M.Z., Costa M., Leite R.D., et al. 2018. To grow or to reproduce? Sexual dimorphism and ontogenetic allometry in two Sesarmidae species (Crustacea: Brachyura). J. Mar. Biol. Assoc. U.K. 1: 1-14.).

Statistical analyses

Initially, the specimens were separated into four categories: juvenile female (JF), adult female (AF), juvenile male (JM), and adult male (AM). The determination of morphological maturity (juveniles vs. adults) was based on the maximum inflection point (linear regression) of the relationships between the CW×GL (carapace width × gonopod length) structures for males, and CW×W5 (carapace width × width of the fifth abdominal segment) for females, considering that the GL and W5 variables were taken as dependent variables. Thus, males with sizes greater than 2.59 mm CW and females with sizes greater than 2.69 mm were considered adults.

To determine the growth allometry, the linear function y = β0 + β1x and the potential function y = β0x^β1 were used, where β0 is the intercept constant on the ordinate axis, β1 is the allometric growth constant, x is the independent variable and y is the dependent variable.

Linear relationships were considered negative allometric when β1<1, isometric when β1=1, and positive allometric when β1>1. In the potential relationships, the reference value for β1 was 3. The following relationships were analysed: CW×CL, CW×W4, CW × pleon length (PL), CW × carapace height (CH), CW × right chelipede height (RCH), CW × left cheliped height (LCH), CW × right cheliped length (RCL), CW × left cheliped length (LCL), CW×GL (the latter only in males) using linear regression; and CW×W using potential regression. The first variable (CW) was considered the independent variable for all relationships for all groups. The parameters β1 and β0 were compared using the t test of model comparison between and among species for all linear relationships. The relationships were considered different when at least one parameter showed a statistical difference. The coefficient of determination (r²) was used as an indicator of the quality of linear regression.

To analyse the influence of linear measurements on the morphometric variation of the different groups, a principal component analysis (PCA) was conducted. To remove the influence of ontogeny on the data, all body dimensions were treated as proportional to CW (a variable less sensitive to distortion between size classes) (Harrison and Crespi 1999Harrison M.F., Crespi B. 1999. A phylogenetic test of ecomorphological adaptation in Cancer crabs. Evolution 53: 961-965. https://doi.org/10.1111/j.1558-5646.1999.tb05390.x). To test for statistical differences between the group separations, a one-way PERMANOVA test with Bonferroni’s correction was conducted (p<0.05).

For morphological exploration with the geometric morphometry technique, 475 images of the dorsal view, 469 of the ventral view, and 861 of the chelipeds were analysed. The landmarks’ coordinates were aligned to the centroid, and generalized procrustes analysis (GPA) was performed using the program MorphoJ version 2.0 (Klingenberg 2011Klingenberg C.P. 2011. Evolution and development of shape: integrating quantitative approaches. Nature Reviews Genetics 11:623-635. https://doi.org/10.1038/nrg2829), where differences in orientation, position, and scaling were removed (Bookstein 1991Bookstein F.L. 1991. Morphometric tools for landmark data: geometry and biology. Cambridge University Press, New York, NY. https://doi.org/10.1017/CBO9780511573064, Adams et al. 2004Adams D.C., Rohlf F.J., Slice D.E. 2004. Geometric morphometrics: ten years of progress following the “revolution”. Italian J. Zoology 71: 5-16. https://doi.org/10.1080/11250000409356545). From the waste matrix generated in the GPA, a PCA was conducted to identify the main characteristics of the shapes. Then, canonical variance analysis (CVA) was performed, with 10000 permutations, to find the shape characteristics that best distinguished and separated the groups, using Procrustes distance. Differences between the forms were tested applying the Hotelling test (T²) with Bonferroni correction in discriminant analysis. Thin-plate splines functions (Klingenberg 2008Klingenberg C.P. 2008. Software MorphoJ. Faculty of Life Sciences, University of Manchester, UK.) were generated between groups using the MorphoJ program. A multivariate analysis of variance (MANOVA) with probability adjustment for Bonferroni multiple comparisons (Fornel and Cordeiro-Estrela 2012Fornel R., Cordeiro-Estrela P. 2012. Morfometria geométrica e a quantificação daforma dos organismos. In: Marinho J.R., Hepp L.U., Fornel R. (Org.). Temas em Biologia: Edição comemorativa aos 20 años do Curso de Ciências Biológicas e aos 5 anos do PPG-Ecologia da URI Campus de Erechim. 1ed. Erechim: EDIFAPES, pp. 101-120.) was also conducted to identify possible differences in shape (pleon, carapace and chelipeds) and centroid size. Past software, version 3.07 (Hammer et al. 2001Hammer Ø., Harper D.A.T., Paul D. R. 2001. Past: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica, vol. 4(1), 1-9.), was used for these analyses. Finally, CV1 and CV2 regression scores were related as a dependent variable of centroid size to demonstrate the ontogenetic trajectory graphically.

Linear morphometry

Males and females showed negative allometry (β0<1) for all linear morphometric relationships, and there was a difference (p=0.0001 for all comparisons) in ontogeny (Table 2). Adult males showed a higher growth rate in CL, CH, RCH, LCH, RCL and LCL than juveniles. Adult females showed a higher growth rate than juveniles in the variables CL, W4, W5, PL, RCH, LCH, RCL and LCL. For potential ratios (CW×W), young specimens of both sexes showed negative allometry (β0<3) and adults an isometric ratio (β0=3). All models were statistically different when model parameters were compared using the t test (p=0.0001). For the coefficient of determination (R²), the juvenile males showed the lowest values (CW×GL-R²=0.355), whereas the adult males showed the highest values (CW×CL-R²=0.936) (Table 2).

In the PCA, the first two components explained 75.16% of the variation, accounting for 58.46% and 16.69%, respectively. The variables of greatest influence in separating the groups were RCH, LCH, RCL and LCL for component 1 (PC1) and PL, CH and CL for component 2 (PC2). Note that PC1 showed a tendency to separate individuals according to the stage (juveniles and adults), thus appearing as an age dimorphism, while PC2 separated them according to sexual dimorphism, as can be seen in Figure 3.

The one-way PERMANOVA test for comparison between groups showed a statistical difference (p<0.005) between all groups.

Geometric morphometry

When the Procrustes distance from the CVA of the dorsal view was compared, no statistically significant difference was observed for the ontogenetic comparisons: AF×JF (p>0.05), and AM×JM (p>0.05). However, for the sexual dimorphism comparisons, all of them showed a statistically significant difference (p=0.0001) (Fig. 4). The same result was observed in the discriminant analysis, with a correct allocation of 59.75% for AF and 70% for JF and of 58.62% for AM and 68.18% for JM. The results obtained by MANOVA with Bonferroni correction for the dorsal view also showed no significant difference in centroid size between the AF×JF and AM×JM groups (p-value ranging from 0.41 to 0.92).

The morphology of the abdominal region showed a difference in AF×JF interactions (p=0.0001), with no difference between AM×JM (p=0.800). The correct allocation for AM×JM was 63.0% and 63.6% for AM and JM, respectively. The results obtained by MANOVA with Bonferroni correction also showed a significant difference in shape and size for AF×JF, as shown in Figure 5, and only for centroid size (p=0.0001) in the AM×JM comparison (Fig. 5).

When cheliped shape was analysed, a morphological separation was observed through CVA regarding sexual comparison (F×M) and laterality (R×L), with p-value=0.0001 for both interactions. However, no statistically significant morphological separation was observed when ontogenetic comparison (A×J) was conducted (p=0.72). These differences were also observed in the discriminant analysis of F×M and R×L (p=0.0001) and A×J (p=0.59) (Fig. 6). For the A×J comparison, the allocations were 52.75% for adults and 59.39% for juveniles in females. In males, they were 52.44% and 52.29% for adults and juveniles, respectively.

In chelipeds, the results obtained by MANOVA with Bonferroni correction also showed no significant difference in the ontogenetic comparison form (A×J), with p>0.05. However, in the sexual (F×M) and laterality (R×L) interactions, significant separation was observed, with p=0.0001.

Linear morphometry

The allometric analysis showed that males acquired a larger average body size than females for the G. cruentata population captured in a tropical estuarine mangrove, and the same growth pattern was also described by Botelho et al. (2004)Botelho E.R.O., Andrade C.E.R., Santos M.C.F. 2004. Estudo da população de aratu-do-mangue Goniopsis cruentata (Latreille, 1803) (Crustacea, Decapoda, Grapsidae) no estuário do rio Camaragibe (Alagoas - Brasil). Bol. Téc-cien. Cep. Tamandaré. 12: 91-98.. According to Masunari et al. (2017)Masunari S., Martins S.B., Marochi M.Z., Serra W.S., Scarabino F. 2017. Morphological variability in populations of the fiddler crab Leptuca uruguayensis (Nobili, 1901) (Crustacea, Decapoda, Ocypodidae) from South America. Braz. J. Oceanogr. 65: 373-381. 10.1590/s1679-87592017136606503., the size variation between the sexes can be explained by differentiated reproductive needs, as size dominance in male crabs increases opportunities for victory during intraspecific competition, courtship and handling the female during copulation.

The body size of male specimens of G. cruentata captured on the southern coast of Pernambuco seems to be determinant in the sexual dimorphism of the population because the variables pleon length (PL), CH, and carapace length (CL) were responsible for the separation of males and females in both juveniles and adults. These morphometric characteristics suggest that both sexes of G. cruentata have a life history that prioritizes reproduction and survival, as rapid growth may provide selective advantages. For example, larger females of this species show a higher egg production per clutch (Cobo and Fransozo 2003Cobo V.J., Fransozo A. 2003. External factors determining breeding season in the red mangrove crab Goniopsis cruentata (Latreille) (Crustacea: Brachyura: Grapsidae) on the São Paulo State northern coast, Brazil. Rev. Bras. Zool. 20: 213-217. https://doi.org/10.1590/S0101-81752003000200007), while larger males probably intimidate smaller ones in competition for receptive females and are accepted more often. This behaviour is commonly observed in other Brachyura species (Brockerhoff and McLay 2005Brockerhoff A.M, Mclay C.L. 2005. Comparative analysis of the mating strategies of grapsid crabs with special reference to the intertidal crabs Cyclograpsus lavauxi and Helice crassa (Decapoda: Grapsidae) from New Zealand.J. Crustac. Biol. 25: 507-520. https://doi.org/10.1651/C-2548), indicating a natural selection pattern.

As for the differentiation between juveniles and adults (ontogenetic variation), this study revealed that adult individuals of G. cruentata grew in carapace width (CW) and weight in the same proportion (isometry). In contrast, juveniles grew proportionally more in CW than weight (negative allometry). This difference may result from a differentiated diet influenced by success in capturing food, which is higher in large-sized individuals because the presence of larger undesired individuals inhibits the presence of smaller ones, causing them to dismiss or give up capturing the baits (personal observation). Moreover, the crab’s weight after maturity increases as a result of the action of the androgenic gland (Pinheiro and Fiscarelli 2009Pinheiro M.A.A., Fiscarelli A.G. 2009. Length-weight relationship and condition factor of the mangrove crab Ucides cordatus (Linnaeus, 1763) (Crustacea, Brachyura, Ucididae). Braz. Arch. Biol. Technol. 52: 397-406. Available at: https://doi.org/10.1590/S1516-89132009000200017) and the development of reproductive cells. This change in the weight/growth model through ontogeny in Brachyura species was also found by Mantelatto and Fransozo (1992)Mantelatto F.L.M., Fransozo A. 1992. Relação peso/largura da carapaça no caranguejo Hepatus pudibundus (Herbst, 1785) (Crustacea, Decapoda, Callapidae) na região de Ubatuba, SP, Brasil. Braz Arch Biol Technol. 35: 719-724. in Hepatus pudibundus, by Pinheiro and Fransozo (1993)Pinheiro M.A.A., Fransozo A. 1993. Biometric relationship analysis of the wet weight by carapace width to the swimming crab Arenaeus cribarius (Lamarck, 1818) (Crustacea, Brachyura, Portunidae). Braz. Arch. Biol. Technol. 36: 331-341. in Arenaeus cribarius and by Peiró et al. (2011)Peiró D.F., Pezzuto P.R., Mantelatto F.L. 2011. Relative growth and sexual dimorphism of Austinixa aidae (Brachyura: Pinnotheridae): a symbiont of the ghost shrimp Callichirus major from the southwestern Atlantic. Lat. Am. J. Aquat. Res. 39: 261-270. https://doi.org/10.3856/vol39-issue2-fulltext-7 in Austinixa aidae, thus supporting the above hypothesis.

However, diverging from the results found in this study, Reis et al. (2015)Reis C.R., Taddei F.G., Cobo V.J. 2015. Growth and reproduction of the mangrove crab Goniopsis cruentata (Latreille, 1803) (Crustacea: Decapoda: Grapsidae) in southeastern Brazil. An. Acad. Bras. Ciênc. 87: 699-711. https://doi.org/10.1590/0001-3765201520130387 presented results for Goniopsis cruentata in which juvenile specimens showed negative allometry and adults positive allometry for the weight/growth relationship. Individual allometries for different locations observed in these crabs indicate that geographic variation (factors such as temperature) may influence the body size determination of Brachyura such as G. cruentata because, according to Lira et al. (2015)Lira J.J.P.R., Calado T.C.S., Rezende C.F., Silva J.R.F. 2015. Comparative biology of the crab Goniopsis cruentata: geographic variation of body size, sexual maturity, and allometric growth. Helgol. Mar. Res. 69:335-342. https://doi.org/10.1007/s10152-015-0441-8, local conditions influence variations in their biological characteristics. Therefore, geographic variations may be associated with differences in body size, maturity and variation in reproductive characteristics, a behaviour that is present and well defined in other organisms (Olalla-Tárraga et al. 2009Olalla-Tárraga M.A., Diniz-Filho J.A.F., Bastos R.P., Rodríguez M.A. 2009. Geographic body size gradients in tropical regions: water deficit and anuran body size in the Brazilian Cerrado. Ecography 32: 581-590. https://doi.org/10.1111/j.1600-0587.2008.05632.x).

Geometric morphometry

The combination of morphological methods with morphometric data proved useful for this study, supporting the investigation of the main causes of variation in the shape of the pleon, carapace and chelipeds in this case. Furthermore, studies show that if the complete shape of a structure is considered, the quantity and quality of variables for statistical analysis increases greatly, so geometric morphometry analysis has become an important tool for multidimensional statistical comparison of morphological characters (Bichain et al. 2007Bichain J.M., Gaubert P., Samadi S., Boisselier-Dubayle M.C. 2007. A gleam in the dark: phylogenetic species delimitation in the confusing spring-snail genus Bythinella Moquin-Tandon, 1856 (Gastropoda: Rissooidea: Amnicolidae). Molecular Phylogenetics and Evolution, 45: 927-941. https://doi.org/10.1016/j.ympev.2007.07.018).

The variation found in the carapace region of G. cruentata reveals a greater sexual dimorphism because no ontogenetic dimorphism was found in same-sex individuals. The sexual dimorphism consisted of the lateral displacement of landmarks 3 and 9, which caused a widening of the females’ posterolateral region. This widening resulted in an angular decrease in the region with a smaller displacement of landmarks 1 and 2. It is known that the wider carapace shape in female specimens emerges as a reproductive advantage because females need to develop strategies in this region to accommodate the gonads, and these differences may also reflect an adaptation for internal growth (e.g. gonad and muscle) (López-Greco et al. 2004López-Greco L.S., Viau V., Lavolpe M., et al. 2004. Juvenile hatching and maternal care in Aegla uruguayana (Anomura, Aeglidae). J. Crustac. Biol. 24(2): 309-313. 10.1651/C-2441.). Thus, a synoptic study with other physiological and functional methodologies can help to better understand the species’ life history.

Because of the lack of variation in the cephalothorax region’s shape during the ontogenetic trajectory, we can infer that this region suffers a greater influence of sexual aspects, in which more enlarged females have greater reproductive success (the existence of hierarchy, for example). According to Reis et al. (2015)Reis C.R., Taddei F.G., Cobo V.J. 2015. Growth and reproduction of the mangrove crab Goniopsis cruentata (Latreille, 1803) (Crustacea: Decapoda: Grapsidae) in southeastern Brazil. An. Acad. Bras. Ciênc. 87: 699-711. https://doi.org/10.1590/0001-3765201520130387, G. cruentata has a life history that prioritizes reproduction over survival, and rapid growth provides selective advantages for both sexes, with fast-growing individuals reaching sexual maturity with larger bodies.

Several authors have already observed differences in cephalothorax shape between sexes in crustaceans (Trevisan et al. 2012Trevisan A., Marochi M. Z., Costa, M., Santos, S., Masunari, S. 2012. Sexual dimorphism in Aegla marginata (Decapoda: Anomura). Nauplius 20: 75-86. https://doi.org/10.1590/S0104-64972012000100008), reporting that this distinction is most obvious in the posterior apex of the cephalothorax, specifically in the lateral region, where the posterior region is wider in females than in males. In addition, Marochi et al. (2018)Marochi M.Z., Costa M., Leite R.D., et al. 2018. To grow or to reproduce? Sexual dimorphism and ontogenetic allometry in two Sesarmidae species (Crustacea: Brachyura). J. Mar. Biol. Assoc. U.K. 1: 1-14. showed, for example, that there is sexual dimorphism of Aratus pisonii in carapace shape during the juvenile phase (before the pubertal moult).

In the abdominal region, only the females showed morphological variation (displacement of landmarks 2, 3, and 4 and their respective homologues) involving widening in the sixth, fifth and fourth abdominal segments. This widening caused a reduction at the end of the telson (displacement at landmark 5) and is closely related to egg storage capacity, well discussed by scholars, in which allometric and morphological growth in the female pleon is associated with directional sexual selection arising from an increase in fat reserves and an enlargement of the area to accommodate offspring or eggs (Hartnoll 1978Hartnoll R.G. 1978. The Determination of Relative Growth in Crustacea. Crustaceana 34: 281-293. https://doi.org/10.1163/156854078X00844). In both sexes, directional selection is likely related to increased reproductive success (Kodric-Brown et al. 2006Kodric-Brown A., Sibly R.M., Brown J.H. 2006. The allometry of ornaments and weapons. Proc. Natl. Acad. Sci. U.S.A. 103: 8733-8738. https://doi.org/10.1073/pnas.0602994103).

The relationship between one-dimensional measurement such as size (represented by the centroid size) and measurement of the entire shape of the pleon (represented by landmarks) allows growth to be studied as an integral approach, thus making it possible to identify the degree of shape change during growth in the pleon region for females. In addition to reflecting internal growth (e.g. gonad and muscle) (Williner and Collins 2013Williner V., Collins P.A. 2013. Feeding ecology of the freshwater crab Trichodactylus borellianus (Decapoda: Trichodactylidae) in the floodplain of the Paraná River, southern South America. Lat. Am. J. Aquat. Res. 41: 781-792. https://doi.org/10.3856/vol41-issue4-fulltext-15), these changes are linked to the specimens’ characteristics because larger females produce a greater number of eggs per clutch for G. cruentata (Cobo and Fransozo 1998Cobo V.J., Fransozo, A. 1998. Relative growth of Goniopsis cruentata (Crustacea, Brachyura, Grapsidae) on the Ubatuba region. São Paulo. Bazil. Iheringia, Ser. Zool., v. 84, pp. 21-28.).

For the male specimens, no morphological variations between developmental stages were found, implying that morphological characteristics are not related to ontogenetic development and that this region does not have a biological need that justifies a reproductive effort for variation in the pleon shape. Similar results were found in freshwater crabs by Silva et al. (2018)Silva L.N., Almeida P.R.S., Shinozaki-Mendes R.A. 2018. Dimorfismo sexual e alometria ontogenética em Goyazana castelnaui (Crustacea, Brachyura). Iheringia 108: e2018008. https://doi.org/10.1590/1678-4766e2018008 and Almeida et al. (2021)Almeida D.S., Menezes A.N.C., Shinozaki-Mendes R.A. 2021. Ontogenetic variation of the Goyazana castelnaui H. Milneedwards, 1853 (Brachyura, Trichodactylidae), crab in the semiarid region of Brazil. An. Acad. Bras. Ciênc. 95: 2-11. https://doi.org/10.1590/0001-3765202320200929. The lack of noticeable variation in the pleon of males may due to the fact that the structure is used only for the support and protection of the two pleopod pairs responsible for sperm transfer during mating (Hartnoll 1978Hartnoll R.G. 1978. The Determination of Relative Growth in Crustacea. Crustaceana 34: 281-293. https://doi.org/10.1163/156854078X00844).

The G. cruentata population showed significant sexual dimorphism but no ontogenetic variation in cheliped morphology and laterality. Regardless of the sexual influence, we observed that the morphology found in the G. cruentata chelipeds undergoes a greater expression of laterality (principal component 1) than the sexual aspects (principal component 2) and shows no ontogenetic expression in the shape influence. This variation has been recorded for other Brachyura species, and decapods are classified as either homochelous or heterochelous (the larger cheliped can be either the right or the left). For this study, despite the influence of sex on the cheliped morphology, both sexes showed similar laterality. Heterochely was therefore not influenced by sex, thus diverging from the findings of Silva and Paula (2008)Silva I.C., Paula J. 2008. Is there a better chela touse for geometric morphometric differentiation inbrachyuran crabs? A case study using Pachygrapsus marmoratus and Carcinus maenas. J. Mar. Biol. Assoc. U.K. 88: 941-953. https://doi.org/10.1017/S0025315408001483 in studies with Pachygrapsus marmoratus and Carcinus maenas, in which they state that differences in the chelipeds can also be influenced by gender.

The morphological variation of chelipeds in G. cruentata may be related to the different functions of the right and left chelipeds. The right (slimmer) cheliped can be used for food handling, while the left (more robust) cheliped is a larger and more powerful grinding tool for breaking down exoskeletons of smaller crabs and other food components. This “function division” was pointed out by Schenk and Wainwright (2001)Schenk S.C., Wainwright P. 2001. Dimorphism and the functional basis of claw strength in six brachyuran crabs. J. Zool. 255: 105-119. https://doi.org/10.1017/S0952836901001157, who compared the morphology underlying cheliped force production and intraspecific cheliped dimorphism in six species of brachyuran crabs, finding that in many crab species males and females exhibit dimorphic chelipeds, as in the case of G. cruentata.

The ability to break the shell is directly related to the cheliped’s closing force, which varies among species depending on size, leverage and dentition properties, and the type and angle of the muscle fibres closest to the cheliped (Schenk and Wainwright 2001Schenk S.C., Wainwright P. 2001. Dimorphism and the functional basis of claw strength in six brachyuran crabs. J. Zool. 255: 105-119. https://doi.org/10.1017/S0952836901001157). However, increased size and variation of the cheliped has also been associated with success in sexual and agonistic interactions (Warner 1982Warner G.F., Chapman D., Hawkey N., Waring DG. 1982. Structure and function of the chelae and chela closer muscles of the shore crab Carcinus maenas (Crustacea: Brachyura). J. Zool. 196: 431-438. https://doi.org/10.1111/j.1469-7998.1982.tb03514.x).

The shape of the right and left chelipeds was different in males and females and between sexes, a behaviour not yet recorded for an arboreal species that does not exhibit a high degree of heterochely, such as the species G. cruentata, A. pisonii and A. angustipes. The fixed finger of the slender and longer cheliped propodus found in females may appear as a facilitator in reaching the abdominal chamber and sensory structures in the egg cleaning and larval release process (Almerão et al. 2010Almerão M., Bond-Buckup G., Mendonça J.R.M.S. 2010. Mating behavior of Aegla platensis (Crustacea, Anomura, Aeglidae) under laboratory conditions. J. Ethol. 28: 87-94. https://doi.org/10.1007/s10164-009-0159-7). However, to support the hypothesis presented here, studies on the species’ behaviour are necessary. According to Warner (1970)Warner G.F. 1970. Behaviour of Two Species of Grapsid Crab during Intraspecific Encounters. Behaviour 36: 9-19. https://doi.org/10.1163/156853970X00024, species exhibitingslight sexual dimorphism in chelipeds belong to the category in which ritualized agonistic behaviour is also observed in females. Therefore, the morphological expression existing between males and females of G. cruentata should not be limited to ritualized agonistic behaviour, directing female morphology solely towards purely reproductive activities, since the cause of sexually dimorphic chelipeds may offer potential advantages that facilitate sex recognition and therefore courtship, as described in crabs with high sexual dimorphism (Uca, Crane 1958; Hemiplax, Griffin 1968Griffin D.J.G. 1968. Social and maintenance behaviour in two Australian ocypodid crabs (Crustacea: Bracliyura). J. Zool. 156: 291-305. https://doi.org/10.1111/j.1469-7998.1968.tb04353.x) and low sexual dimorphism (Goniopsis, Herma and Schõne 1963; Pachygrapsus, Bovbjerg 1960Bovbjerg R.V. 1960. Courtship behaviour of the lined shore crab Pachygrapsus crassipes Randall. Pac. Sci. 14: 421-422.).

Overall, the results presented here provide new information on the development of secondary characters and their consequences on the shape and size of the carapace and the propodus of the chelipeds in males and females of Goniopsis cruentata. Therefore, these functions could be considered morphological characters with sex-dependent differential expression (Hartnoll 1978Hartnoll R.G. 1978. The Determination of Relative Growth in Crustacea. Crustaceana 34: 281-293. https://doi.org/10.1163/156854078X00844, Barría et al. 2011Barría E.M., Sepúlveda R.D., Jara C.G. 2011. Morphologic variation in Aegla Leach (Decapoda: Reptantia: Aeglidae) from Central-Southern Chile: interspecific differences, sexual dimorphism, and spatial segregation. J. Crustac. Biol. 31: 231-239. https://doi.org/10.1651/10-3324.1). These consequences reflect the evolutionary sexual tendency toward reproduction. Furthermore, traditional morphometry and geometric morphometry proved efficient in investigating and recognizing variations between juveniles and adults and males and females in the carapace, pleon and cheliped with refined detail. This information is extremely important for maintaining and conserving invertebrate species, especially those with commercial value, even if their conservation status is not critical.