text new page (beta)
English (pdf)
Article in xml format
Curriculum ScienTI
Automatic translation
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
uBio 
Permalink
INTRODUCTION
Sexual dimorphism refers to differences between males and females of a species in secondary sex-related features, like body size, color pattern, morphological details of specific body parts, and behavior. Sexual dimorphism may be present in any sexually-reproducing dioecious organism, including plants (e.g. Lloyd and Webb 1977; Barret 2002; Tsuji and Fukami 2018) and animals (e.g. Garcia et al. 2006; Loker and Brant 2006; Ceballos et al. 2013). Darwin (1871) described several examples of sexual dimorphism when proposing his theory of sexual selection, and Andersson (1994) postulated that sexual dimorphism would result from different sexual selection pressures acting on the two sexes.
For fish, secondary sexual dimorphism has been recorded in body size (e.g. Parker 1992; Erlandsson and Ribbink 1997; Neat et al. 1998; McMillan 1999; Morbey 2018), fin size and shape (e.g. Skjæraasen et al. 2006; Pires et al. 2016), color pattern (e.g. Robertson and Warner 1978; Karino and Someya 2007), and head morphology (e.g. Hastings 1991; Gramitto and Coen 1997; Cox Fernandes 1998; Cox Fernandes et al. 2002, 2009; de Santana and Vari 2010). In some species, jaws, mouth and snout are larger in males than in females (Goto 1984; Crabtree 1985). Dentition may also be sexually dimorphic, with differences between males and females in number, shape and arrangement of teeth (Gomes and Tomas 1991; Kajiura and Tricas 1996; Böhlke 1997; Rapp Py-Daniel and Cox Fernandes 2005; de Santana and Vari 2010). The shape of the urogenital papilla may also differ between males and females (Esmaeili et al. 2017).
Secondary sexual dimorphism may also be expressed in communication systems, such as in sound-producing mechanisms (Ali et al. 2016; Parmentier et al. 2018) or as differences in electrical signal repertoires of male and female electric fishes (Fugere and Krake 2009; Ho et al. 2010, 2013). Among Neotropical electric fishes of the order Gymnotiformes, the most common forms of sexual dimorphism occur in body size (Hilton and Cox Fernandes 2006; de Santana and Cox Fernandes 2012), snout shape (de Santana 2003; Albert and Crampton 2009; Evans et al. 2018), and caudal filament size and shape (Hopkins et al. 1990; Giora et al. 2008). Differences in mouth shape, position and shape of teeth (de Santana and Vari 2010; Cox Fernandes et al. 2010), and electric organ discharge (Nogueira 2006; de Santana and Crampton 2007; Smith and Combs 2008; Fugere and Krake 2009; Ho et al. 2010, 2013) have also been reported.
Rapp Py-Daniel and Cox Fernandes (2005) discuss the evolution of sexual dimorphism in Gymnotiformes by mapping sex-related features on phylogenetic hypotheses and presenting evidences that secondary sexual differences arose independently both among the gymnotiform families and inside Apteronotidae, where most cases of sexual dimorphism in electric knifefishes were reported (de Santana 2003; Hilton and Cox Fernandes 2006; Albert and Crampton 2009; Cox Fernandes et al. 2010; de Santana and Vari 2010; Ho et al. 2013). Sexual dimorphism in Hypopomidae (Hopkins et al. 1990; Hopkins 1999; Giora et al. 2008; Gavassa et al. 2013), Gymnotidae (Mendes-Júnior 2015) and in Sternopygidae (Zakon et al. 1991; Giora and Fialho 2009; Vari et al. 2012) has also been reported. However, for Rhamphichthyidae (Ramphichthys + Gymnorhamphicthys + Iracema + Hypopygus + Steatogenys; Carvalho 2013; Tagliacollo et al. 2015) we have found no recorded instances of sexual dimorphism in the literature.
Recently, we had the opportunity to study the reproductive biology and spatial distribution of individuals of Gymnorhamphichthys rondoni (Miranda Ribeiro, 1920), a strictly psammophilous electric knifefish widely distributed in the Amazon Basin and a common inhabitant of upland forest streams of the Brazilian Amazon (Zuanon et al. 2006; Carvalho 2013). During the study, we noted differences in the proportional size of the head, as well as in the conspicuouness of the urogenital papilla between male and female specimens, which suggested a possible case of sexual dimorphism. Therefore, our objective was to evaluate the occurrence of secondary sexual dimorphism in a population of G. rondoni in a Central Amazon forest stream by analyzing external morphometric parameters.
MATERIAL AND METHODS
We collected 45 adult individuals (36 females and nine males) of Gymnorhamphichthys rondoni using an electric fish detector (Crampton et al. 2007) and hand nets in a terra firme forest stream at Fazenda Dimona of the Biological Dynamics of Forest Fragments Project (BDFFP - http://pdbff.inpa.gov.br/), located about 80 km north of Manaus, Amazonas state, Brazil. The studied forest stream is a tributary of the Cuieiras River in the Negro River basin, in the central Brazilian Amazon. The studied stream section (2º21’1.41”S, 60º5’44.31”W) has a width of 3 - 5 m, maximum depth of 1.5 m, a predominantly sandy substrate with coarse litter deposits, and the channel almost completely shaded by riparian forest canopy. The water was clear, acidic (pH ~5.0), with low electric conductivity (~10 µS*cm-1), and temperature of 23-24 ºC. In addition to the collected fish, we also used preserved specimens from the Fish Collection of the Instituto Nacional de Pesquisas da Amazônia (INPA-ICT). All specimens had the abdominal cavity opened for identification of sex via gonadal examination. We retained for subsequent analyzes only the adult specimens (i.e. those with gonads classified as in late maturation, spawning or regenerating, according to definitions by Brown-Peterson et al. 2011). Combining the 45 specimens collected by us with 81 adult specimens from INPA’s Fish Collection we had a final sample of 63 females and 63 males (Supplementary Material, Table S1).
To quantify morphological characteristics, we used digital calipers and measured (in mm) the length of snout to posterior end of anal-fin (LEA), length of anal-fin length (LAF), distance from anus to anal-fin origin (DAAF), distance from the genital papilla to anal-fin origin (DPAF), body height (BH), and head length (HL) (Figure 1). Morphometric differences between males and females were tested with a Kruskal-Wallis test, as the data distribution lacked normality. The morphometric variables were also analyzed using Principal Component Analysis (PCA) via R statistical software (R Core Team 2016). Since the first component usually is strongly influenced by the size of the specimens, we plotted the data considering the first principal component (PC1 x PC2) and the next two components (PC2 x PC3) to depict the ordination without the effect of body size.
Figure 1 Schematic drawing of Gymnorhamphichthys rondoni in lateral view showing the morphological measurements used in this study: DAAF = distance from anus to anal-fin origin; DPAF = distance from urogenital papilla to anal-fin origin; HL = head length; LAF = length of anal-fin; LEA = length from snout to posterior end of anal-fin.
To check for occurrence of sexual dimorphism in the urogenital papilla, we used an extended focus stereomicroscope to produce lateral and ventral images of the papillae of adult male and female G. rondoni. All the procedures in this study involving animals were in accordance with and duly approved by the Ethics Committee on Animal Use (CEUA/INPA, protocol #022/2016).
RESULTS
The morphometric analysis showed that female G. rondoni had a longer anal-fin (LAF), a larger distance between the urogenital papilla and the anal-fin origin (DPAF) and a larger distance from the anus to the anal-fin origin (DAAF), whereas males presented a longer head (HL) (Figure 2, Table 1).
Figure 2 Images of a female and male Gymnorhamphichtys rondoni collected in a forest stream tributary of the Cuieiras River in the Negro River basin, central Brazilian Amazon. LEA female = 123.74 mm; LEA male = 117.3 mm. Scale bars = 10 mm
Table 1 Summary of morphometric measurements (median (minimum - maximum)) in mm, and statistics of the Kruskal-Wallis rank sum test for females (n=63) and males (n=63) of Gymnorhamphichthys rondoni. For all tests, df = 1. Values followed by * indicate significant differences between genders.
| Measurements | Females | Males | H | p-value |
|---|---|---|---|---|
| Length from snout to posterior end of anal-fin (LEA) | 129.99 (77.19 - 185.15) | 126.3 (102.1 - 153.3) | 0.9006 | 0.3426 |
| Length of anal-fin (LAF) | 103.54 (60.32 - 143.50) | 94.56 (65.38 - 120.87) | 54.736 | 0.01931* |
| Distance from the urogenital papilla to the anal-fin origin (DPAF) | 13.65 (2.24 - 22.14) | 4.32 (2.42 - 5.88) | 82.678 | 2.20E-16* |
| Distance from the anus to the anal-fin origin (DAAF) | 11.070 (0.94 - 19.310) | 4.82 (2.67 - 6.64) | 67.635 | 2.20E-16* |
| Body height (BH) | 4.05 (2.05 - 6.65) | 4.03 (3.07 - 5.09) | 0.45012 | 0.5023 |
| Head length (HL) | 28.18 (12.13 - 40.53) | 32.01 (26.17 - 38.79) | 14.042 | 0.0001788* |
The first three morphometric-based PCA components explained 64.5%, 21% and 8.4% of observed variance, respectively (Figure 3). The first principal component (PC1) was strongly influenced by negative values of variables related to body size of the specimens, such as length from snout to posterior end of anal-fin (LEA) and LAF. The second component (PC2) was positively influenced by head length (HL) and negatively by DPAF and DAAF. The third component was negatively influenced by HL and positively by LAF (Table 2). PCA ordination separated males and females of G. rondoni mainly along the second principal component (Figures 3a and 3b). Females were larger than males, had a shorter head and body heigth, and a wider distance between the urogenital papilla and the anal fin origin, whereas males were smaller, had a longer head and a higher body height, and a smaller space between the urogenital papilla and the anal fin origin.
Figure 3 Principal Component Analysis of morphometric data of male and female specimens of Gymnorhamphichthys rondoni showing (a) the first and second components (PC1 x PC2), and (b) the second and third components (PC2 x PC3). Blue dots = males, red dots = females. HL = head length; BH = body height; LEA = length from snout to posterior end of anal-fin; LAF = length of anal-fin; DAAF = distance from anus to anal-fin origin; DPAF = distance from urogenital papilla to anal-fin origin. This figure is in color in the electronic version.
Table 2 Variable loadings on the first two principal components (PCs) for Gymnorhamphichthys rondoni (n= 126)
| Measurements | PC1 | PC2 | PC3 |
|---|---|---|---|
| Length from snout to posterior end of anal-fin (LEA) | -1.995 | 0.4436 | 0.5030 |
| Length of anal-fin (LAF) | -1.944 | 0.1023 | 0.8152 |
| Distance from the urogenital papilla to the anal-fin origin (DPAF) | -1.559 | -1.3665 | -0.4481 |
| Distance from the anus to the anal-fin origin (DAAF) | -1.790 | -1.0331 | -0.4671 |
| Body height (BH) | -1.718 | 0.7164 | 0.0462 |
| Head length (HL) | -1.147 | 1.4522 | -0.9878 |
| Explained variance | 64.5 | 21.0 | 8.4 |
| Cumulative variance (%) | 64.5 | 85.5 | 93.9 |
We found differences in the shape and position of the urogenital papilla between males and females (Figure 4). Female papillae were more horizontally elongated and approximately 10 times larger than those of males, and were located on a vertical line below the eye, while male papillae were located on a vertical line below the operculum. In females, papillae may expand remarkably during oocyte passage (Figures 4g and 4h).
Figure 4 Urogenital papilla of female and male Gymnorhamphichthys rondoni. In all images, the anterior portion of the body is towards the left. (a) Side view of head of a female showing urogenital papilla (blue arrow). (b) Side view of head of a male showing urogenital papilla (blue arrow). (c) Ventral view of head of a female showing urogenital papilla (blue arrow) and anus (green arrow). (d) Ventral view of head of a male showing urogenital papilla (blue arrow) and anus (green arrow). (e) Side view of female urogenital papilla. (f) Side view of male urogenital papilla. (g) Side view of urogenital papilla of a female with an oocyte (red arrow) inside. (h) Ventral view of head of a female with oocyte at the end of urogenital papilla. Blue arrow = urogenital papilla, green arrow = anus, red arrow = oocyte position. LEA female = 145.83 mm; LEA male = 144.83 mm. This figure is in color in the electronic version.
DISCUSSION
The observed sexual dimorphism in G. rondoni was related to body size, anal fin length, head length and to urogenital papilla shape and relative position on the body. Males had a proportionally larger head than females, whereas females had a longer anal fin, a larger distance between the urogenital papilla and the anal-fin origin, and a larger distance from the anus to the anal-fin origin. In females the papilla was elongated horizontally, longer than that of males and located on the vertical line below the eye. In males the papila was vertically elongated, smaller than that of females and located on a vertical line below the opercular opening. As far as we searched the scientific litterature, this is the first recorded case of sexual dimorphism in a species of Rhamphichthyidae.
In Gymnotiformes, it is relatively common to find sexual dimorphism in head shape and snout size (de Santana 2003; Albert and Crampton 2009). Tooth shape, size and position also differ between genders of several species of the apteronotid genus Sternarchorhynchus (de Santana and Vari 2010), and in “super-males” of Sternarchogiton nattereri (Cox Fernandes et al. 2010), in which males have hypertrophied and partially exteriorized teeth that seems to be related to male-male conflicts, or to their use when courting females (Cox Fernandes et al. 2010). However, the ecological or behavioral meaning of the larger head in male G. rondoni is not clear. Contrary to the obvious potential use of a hyperthophied mouth and teeth during aggressive encounters or agonistic displays by male apteronotid knifefishes, the small mouth and delicate tubular snout of male G. rondoni seems of little value during a male-male conflict. A possible alternative explanation for such a sexually dimorphic characteristic may be related to differences in foraging tactics or microhabitat use between genders, which remains to be verified.
Sexual dimorphism of the Gymnotiform urogenital papilla position was reported for 15 species of Sternarchorhynchus (Santana and Vari 2010). However, unlike G. rondoni, in Sternarchorhynchus species the urogenital papilla of males is located in a more anterior position on the body compared to females (de Santana and Vari 2010). In addition to this form of dimorphism, Cox Fernandes et al. (2014) also found a difference in the size of the urogenital papillae in Procerusternarchus pixuna (Hypopomidae), with male papillae smaller than those of females, as recorded here for G. rondoni.
When describing Gymnorhamphichthys rosamariae, Schwassmann (1989) reported mature males and females with elongated urogenital papillae and located at the vertical line passing through the eye, and that papilla growth and position are related to gonad development. In this way, papillae larger and closer to the eye line would indicate reproductively mature individuals, regardless of sex. There are records in Gymnotiformes species of the anus and urogenital papilla changing position on the body during ontogeny, moving gradually from the posterior region of the abdominal cavity to the cephalic region (e.g. Apteronotus caudimaculosus: de Santana 2003; Archolaemus blax: Vari et al. 2012; Distocyclus conirostris:Dutra et al. 2014; Eigenmannia besouro: Peixoto and Wosiacki 2016; E. meeki: Dutra et al. 2017; and E. sayona:Peixoto and Waltz 2017). This change in the relative position of the anus and urogenital papila was not detected in other examined apteronotids (Apteronotus eschmeyeri: de Santana et al. 2004; Sternarchogiton labiatus: de Santana and Crampton 2007; S. nattereri: de Santana and Crampton 2007, and Crampton 2007; Apteronotus anu: de Santana and Vari 2013; A. baniwa: de Santana and Vari 2013). We did not find evidence of an ontogenetic change in the position of the urogenital papilla of G. rondoni, as our study was limited to the analysis of adult specimens, which prevented the detection of ontogenetic variations.
The presence of larger urogenital papillae in females than in males may be related to the size of the gametes to be released, so that this characteristic should be more apparent in those species whith proportionately larger oocytes, such as G. rondoni (Garcia and Zuanon, unpublished data). On the other hand, the elongated urogenital papilla of Gymnotiform females may be related to some tactic of oocyte deposition. Gymnorhamphichthys rondoni lives only in places where the substrate is composed largely of sand, in which individuals remain buried during the day, emerging only at night to forage and perform reproductive activities (Zuanon et al. 2006). It is therefore possible that the horizontally elongated papilla aid in selection of oviposition sites, which remains to be studied. At the moment, we lack a functional explanation for the difference in the position of the urogenital papilla and anus observed in male and female G. rondoni. An anatomical study involving a complete ontogenetic series from the larval phase to sexually mature adults, would likely help to better understand the process and the biological significance of the differences reported here.
In a review of the Rhamphichthyoidea, Carvalho (2013) did not mention the occurrence of sexual dimorphism in species of Rhamphichthys, Gymnorhamphichthys or in Iracema caiana (Rhamphichthyinae sensu Carvalho 2013). However, our study found sexual dimorphism in relation to head length and the shape, size and position of urogenital papila for G. rondoni. Accordingly, it is possible that this species might also show sexual dimorphism in other characteristics, such as electric organ discharge patterns or in behavioral aspects, which deserve to be investigated.
CONCLUSIONS
The description and quantification of secondary sexual differences in a rhamphichthyid species may help providing important new information for the understanding of the evolution of sexual dimorphism among Gymnotiformes. Moreover, the known occurrence of external morphological differences could allow sex identification of living individuals in ecological or behavioral studies, avoiding the unnecessary sacrifice of fish and reducing impacts in natural populations, which may be specially important in protected areas.
