text in 
English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Similars in
SciELO 
Permalink
INTRODUCTION
Some population parameters need to be identified in order to understand the biology of a species (Granados-Lorencio 1996). Reproduction is an important aspect of an organism’s biology, and studies on reproduction in fish populations exploited by fishing are crucial when developing assessment models and managing resources, as they analyze essential variables such as average length at maturity and reproductive potential, among other (Sley et al. 2012).
Many species of the Serranidae family are considered of great economic value, especially in coastal fisheries in tropical and subtropical areas. Paranthias colonus is a serranid that is distributed from the Gulf of California to Peru, including islands (Heemstra and Randall 1993). Studies on this species have reported on spinal column malformations (Rodríguez-Romero et al. 2001), parasitic fauna (Mendoza-Cruz et al. 2013), important aggregations, and reproductive behavior in the Gulf of California (Sala et al. 2003).
Paranthias colonus is captured by the commercial fishery off the coast of Jalisco and Colima (Rojo-Vázquez et al. 2001; Espino-Barr et al. 2003, 2004). Many species have biological characteristics that make them vulnerable to intense fishing (Ralston 1987), such as spawning aggregations at specific times and places (Burton et al. 2005, Nemeth et al. 2007, Starr et al. 2007) and method of reproduction. This research aims to generate information on reproductive aspects of P. colonus, a species that is commercially captured on the Pacific coast of Mexico.
MATERIALS AND METHODS
The specimens were obtained from the commercial fishery catches in Navidad Bay (19°10′30″ N, 19°14′50″ N; 104°49′45″ W, 104°41′30″ W), southern coast of Jalisco. The bay has rocky bottoms, soft bottoms, and sand/rock mixed bottoms, with reef structures that stand out in semi-protected areas (Rojo-Vázquez et al. 2001).
From January 1999 to December 2008 (except 2001), for 5 days a month, we sampled organisms captured by fishermen in the region, who used gillnets with mesh sizes of 7.62, 8.89, 10.16, and 11.43 cm. The total length (TL, 0.10 cm), total weight (TW, 0.10 g), and sex of each individual were recorded, and gonads were removed and weighed (gonad weight, GW; 0.01 g). Gonads were classified following Everson et al. (1989) and they were preserved in neutral formalin.
Samples 0.50 cm thick were taken from ovaries and testes for histological analyses. Samples were dehydrated (ethyl alcohol), embedded (Paraplast), sliced (5.00 μm thick), stained (hematoxylin-eosin), and preserved (Canada balsam). Ten sections per maturity stage were chosen for each sampling month. From these sections we measured the diameter of 10 oocytes (with visible nucleus; oocyte diameter, OD) in each stage found and the thickness of the ovarian wall at 4 random points. Measurements were made with a digital camera (AxioCam ERc5s, Zeiss) mounted on a microscope (Axiostar, Zeiss). Oocytes were classified according to Yamamoto and Yamazaki (1961), Lucano-Ramírez et al. (2001), and Brown-Peterson et al. (2011); testes, according to Uribe et al. (2014).
We pooled the monthly data obtained for all years to determine a typical year. This was done because of the observed trend and the low sample sizes in some months. We determined the sex ratio for the total number of organisms by month and by length class, and we used the chi square (χ2) test to verify significant differences in the 1:1 ratio. We determined the gonadosomatic index (GSI = [GW/TW] × 100), relative condition factor (RCF = [TW/TL3.06] × 100), and mean length at sexual maturity (L50). This last one was estimated by performing a nonlinear fit to the logistic model using PTL = 1/[1 + e (bTL + a) ] × 100, where PTL is the percentage of mature individuals at TL, and a and b are parameters of the fitted curve.
We performed one-way analyses of variance to examine the differences between monthly values (GSI, RCF, and OD). When significant values were obtained, the Student-Newman-Keuls post hoc test was applied to identify homogeneous groups. Spearman’s correlation analyses (r s ) were performed between GSI, RCF, and OD. For the statistical analyses, we used STATISTICA v.7.1 (StatSoft 2006) and a significance level of 0.05 in all cases.
RESULTS
Length distribution and sex ratio
In the 9 sampling years, a total of 1,541 P. colonus organisms were collected, which measured between 19.50 and 38.00 cm long, with an average value (± standard error of the mean) of 30.30 ± 0.07 cm. Average length for females was 30.20 ± 0.10 cm; for males, 30.40 ± 0.10 cm (Table 1). The highest frequency of organisms for both sexes occurred at 30.00 and 32.00 cm TL (Fig. 1). Of all the sampled organisms, 782 (51%) were female and 759 (49%) male; the sex ratio was 1.00:0.97 (F:M) and did not differ from the expected theoretical ratio (χ2 = 0.34, P = 0.559). Significant differences in sex ratio were observed in February (1.00F:2.40M; χ2 = 6.08, P = 0.014) and August (1.00F:0.60M; χ2 = 8.85, P = 0.003). No length class (20.00 to 38.00 cm) showed differences in the sex ratio.
Table 1 Number of females (F) and males (M), total length (mean ± standard error of the mean), and sex ratio by year for Paranthias colonus caught in Navidad Bay, Jalisco, Mexico.
| Year | F | M | Female TL (cm) | Male TL (cm) | Range | F:M ratio | χ2 | P |
| 1999 | 63 | 43 | 30.50 ± 0.25 | 30.60 ± 0.25 | 26.10 - 35.40 | 1.00:0.68 | 3.77 | 0.052 |
| 2000 | 46 | 59 | 30.80 ± 0.25 | 30.40 ± 0.26 | 25.80 - 35.40 | 1.00:1.28 | 1.61 | 0.204 |
| 2002 | 23 | 23 | 31.90 ± 0.30 | 32.10 ± 0.42 | 27.70 - 37.00 | 1.00:1.00 | 0.00 | 1.000 |
| 2003 | 143 | 162 | 28.30 ± 0.29 | 28.40 ± 0.25 | 19.50 - 36.00 | 1.00:1.13 | 1.18 | 0.277 |
| 2004 | 50 | 58 | 29.20 ± 0.47 | 30.60 ± 0.31 | 20.70 - 35.00 | 1.00:1.16 | 0.59 | 0.442 |
| 2005 | 72 | 68 | 31.10 ± 0.35 | 30.70 ± 0.45 | 20.60 - 35.00 | 1.00:0.94 | 0.11 | 0.740 |
| 2006 | 206 | 194 | 30.40 ± 0.16 | 30.50 ± 0.14 | 21.70 - 38.00 | 1.00:0.94 | 0.36 | 0.548 |
| 2007 | 53 | 45 | 30.80 ± 0.32 | 31.60 ± 0.27 | 24.10 - 36.50 | 1.00:0.85 | 0.65 | 0.420 |
| 2008 | 126 | 107 | 31.40 ± 0.23 | 32.20 ± 0.21 | 20.40 - 36.60 | 1.00:0.85 | 1.55 | 0.213 |
| Total | 782 | 759 | 30.20 ± 0.10 | 30.40 ± 0.10 | 19.50 - 38.00 | 1.00:0.97 | 0.34 | 0.559 |
Gonadal maturity stages
We identified 4 maturity stages in both sexes. The mature stage showed high percentages from March to June and from November to December in females (30% to 78%) and males (67% to 95%). The immature stage showed high percentages in January, July, and August in both sexes (Fig. 2).
Monthly variation in the gonadosomatic index
The overall average GSI value for males was higher than that for females (F 1, 1,254 = 374.88, P < 0.001). The highest average values occurred in March-June and November-December and differed significantly with respect to the rest of the months for both females (F 11, 631 = 82.27, P < 0.001) and males (F 11, 604 = 93.24, P < 0.001). Pairwise comparisons showed 5 groups in both sexes and 2 trends, averages with low values (females 1, 2, 3; and males A, B, C) and high values (females 4, 5; and males D, E) (Fig. 3). The GSI variation for females and males followed a similar trend, which indicated high correlation (r s = 0.97, P < 0.001, n = 12).
Monthly variation of the relative condition factor
Average RCF for females (0.98 ± 0.003) and males (0.97 ± 0.003) showed no significant differences (F 1, 1,539 = 0.20, P = 0.684). Monthly RCF averages varied in both sexes. Maximum values were recorded in April and December (1.00 ± 0.01 and 1.01 ± 0.01, respectively) for both females and males, and minimum values were recorded in July (0.93 ± 0.01) for females and in June and September (0.93 ± 0.01 and 0.93 ± 0.02, respectively) for males. Despite little monthly variation, significant differences were observed for both females (F 11, 770 = 5.06, P < 0.001) and males (F 11, 747 = 6.51, P < 0.001) (Fig. 4). The multiple comparisons test detected 2 groups for females and 3 for males, with several overlaps between months; however, maximum and minimum values appeared separated from the rest of the monthly values. RCF monthly averages for females and males were significantly correlated (r s = 0.92, P < 0.001, n = 12). The joint variation of GSI and RCF was significant for females (r s = 0.63, P = 0.026, n = 12) but not for males (r s = 0.53, P = 0.071, n = 12), although it was very close to the marginal probability value.
Monthly variation in oocyte diameter
The monthly variation in OD showed a significant difference (F 11, 4,290 = 40.23, P < 0.001). The pairwise test distinguished 3 groups: oocytes with the smallest diameters in July, August, and September; intermediate diameters in June, October, and November; and oocytes with the largest diameters in May and December, with an overlap between medium- and large-sized diameters from January to April (Fig. 5). OD and GSI correlated (r s = 0.64, P = 0.026, n = 12). The high values for mature stage, GSI, and OD allowed us to indicate that the reproductive period of P. colonus occurs from March to June and from November to December.
Gonad description
The gonads, paired structures joined at the narrowest part to the urogenital orifice, were located below the swim bladder. Ovaries, in general, showed a cylindrical shape and shades of orange. Immature ovaries weighed 2.10 ± 0.20 g (average ± standard error of the average) and had oocytes that measured 53.07 ± 1.31 µm (range: 28.60-163.90 µm); ovaries in the maturing stage weighed 11.80 ± 0.40 g and had oocytes that measured 159.79 ± 2.94 µm (range: 28.80-522.70 µm); mature ovaries weighed 18.50 ± 0.40 g and had oocytes that measured 180.12 ± 2.83 µm (range: 28.80-498.41 µm); and spent ovaries weighed 9.10 ± 0.60 g and had oocytes that measured 165.30 ± 4.03 µm (range: 29.20-417.40 µm). On the other hand, testes were white. Immature testes weighed 2.90 ± 0.40 g; developing testes, 12.70 ± 0.90 g; mature testes, 31.20 ± 0.40 g; and spawned testes, 9.790 ± 0.80 g. On average, the weight of mature testes significantly exceeded the weight of mature ovaries (F 1, 1,254 = 312.89, P < 0.001). The largest testicle weighed 57.81 g and belonged to a 33.00-cm long organism with a GSI of 13.56, and the heaviest ovary weighed 44.99 g and belonged to a 35.10-cm long female with a GSI of 10.18.
Immature ovaries had only oocytes in primary growth measuring 54.30 ± 0.35 µm in diameter, which were well organized within the lamellae (Fig. 6a, b). The following stages of gonadal maturation showed oocytes with cortical alveoli (113.20 ± 0.75 µm), primary vitellogenesis (203.79 ± 1.83 µm), secondary vitellogenesis (305.37 ± 2.00 µm), late migration of the germinal vesicle (323.86 ± 2.83 µm), and a degraded germinal vesicle (336.13 ± 5.07 µm). The mature ovary showed 6 phases (Fig. 6c, d), and the largest oocytes were the mature ones (336.13 ± 5.07 µm); the presence of these different phases suggests that oocytes develop asynchronously. The arrangement in the testicle is of the unrestricted lobular type. The lobules were formed by cysts, which showed different degrees of spermatogenesis development (Fig. 7a). We also identified folds or septa (lamella-like) that delimited the cysts and, within this delimitation, a sperm concentrate (Fig. 7b). No main duct was observed in the center of the testes, but sperm concentrations were identified in the testicle periphery, very close to the testis layer; a clear lumen was also observed (Fig. 7c, d). Since there were no microscopic characteristics that indicated hermaphroditism, P. colonus is a gonochoric species.
Figure 6 Paranthias colonus ovary with asynchronous oocyte development. (a, b) Immature phase with ovarian wall (OW), primary growth oocyte (PG), and lamellae (La). (c) PG, cortical alveolar (CA), primary vitellogenic oocyte (Vtg 1), and secondary vitellogenic oocyte (Vtg 2). (d) Oocytes in late germinal vesicle migration (GVM), Vtg 2, and germinal vesicle breakdown (GVBD).
Mean length at maturity
The smallest female with mature gonads measured 22.70 cm; and the smallest male, 23.50 cm. Mean length at sexual maturity, estimated by nonlinear adjustment, was 26.79 cm for females and 26.86 cm for males (Fig. 8). Mean commercial catch lengths for females (30.30 cm) and males (30.40 cm) indicated that, respectively, 79.70% and 83.30% of organisms had already reproduced at least once.
DISCUSSION
Knowing the interval and/or average length of organisms caught by fisheries is relevant because it helps understand the structure of the population in an area of interest (Gallardo-Cabello et al. 2007). The maximum and average length values for P. colonus found in this work are close to those reported by Espino-Barr et al. (2004) (maximum: 38.31; mean: 28.67 cm) on the coast of Jalisco and by Mendoza-Cruz et al. (2013) in El Sargento, Baja California Sur (maximum: 39.00 cm). Espino-Barr et al. (2003) recorded shorter lengths (maximum: 35.30 cm; mean: 27.90 cm) for the coast of Colima. The slight differences between these values could be related to the sample size, catch period, collection site, and fishing gear used (Cruz-Romero et al. 1987, Sadovy and Shapiro 1987, Craig et al. 1999).
Sex ratio is an important element in the population structure of species. The difference in the number of individuals in each sex can vary between populations and during the life cycle because of changes in environmental factors, food availability, and mortality, among other factors (Nikolsky 1963). The sex ratio recorded in this study for P. colonus did not show differences in the number of females and males. In contrast, for Paranthias furcifer, Posada-López (1996) found that the number of females was greater than the number of males (1.3:1.0), and this difference in sex ratio is attributable to the influence of depth at the sampling site. Similarly, studies have found that the genera Epinephelus (Caballero-Arango et al. 2008, Erisman 2008, De Martini et al. 2011, Ruiz-Ramírez et al. 2018) and Mycteroperca (Caballero-Arango et al. 2011), also in the Serranidae family, are represented by a greater number of females; these genera, however, are protogynous hermaphrodites.
Identifying the reproductive season is an essential objective in studies similar to the present one. To achieve this, studies have frequently analyzed the percentages of gonadal maturity stages; for example, this has been done for different serranid species, such as P. furcifer (Posada-López 1996), Mycteroperca tigris (Caballero-Arango et al. 2011), Epinephelus acanthistius (Díaz-Madrid et al. 2012), and Epinephelus labriformis (Ruiz-Ramírez et al. 2018). As in previous works, the present study used a morphochromatic maturation scale to define the P. colonus reproductive season. In addition, the monthly analysis of GSI is a useful method to delimit the reproductive season of the species; an increase in GSI indicates gonad maturity, whereas a decrease reveals spawning or inactivity (Htun-Han 1978). On the other hand, studies can analyze the variation in the average OD throughout the year to enhance the delimitation of a fish reproductive season (Lucano-Ramírez et al. 2016; Ruiz-Ramírez et al. 2018; Lucano-Ramírez et al 2019a, b). Therefore, according to the maximum values of female and male GSI, maximum values of average OD, and highest percentages of mature gonads, P. colonus has 2 reproductive periods, one from March to June and the other from November to December; both had equal intensity, but the first was somewhat longer and more marked. Other serranid species, such as M. tigris (in Campeche, Caballero-Arango et al. 2011) and E. labriformis (in Jalisco; Erisman 2008, Ruiz-Ramírez 2017) showed the same pattern (2 reproductive periods). In the case of P. furcifer in southwestern Puerto Rico, spawning occurs in 2 periods associated with the full moon in November and December (Posada-López 1996), and these months coincide with one of the reproductive periods observed in the present study for P. colonus. In the study region, intense upwelling occurs from February to May and causes a high concentration of chlorophyll a and an increase in productivity (Ambriz-Arreola et al. 2012). This period of enrichment benefits species that spawn at this time, as is the case for P. colonus. In November and December, the period of cyclones and rain ends, temperature begins to drop, and oceanographic conditions are intermediate. The influence here is not clear, but these months of climatic transition could bring a benefit to some species that has yet to be demonstrated.
The RCF evaluates the general welfare of organisms associated with the change in corpulence, growth, and sexual maturity (Rodríguez-Gutiérrez 1992). This factor depends on age, sex, season, maturation stage, diet, fat tissue reserves, and muscle development (Barnham and Baxter 1998). In P. colonus, the RCF showed a positive correlation with GSI, which indicates that reproduction is certainly related to the condition of the organism but that the reproductive process does not significantly affect the condition of the organism. A similar trend was reported by Ruiz-Ramírez et al. (2017), who pointed out that the condition factor for E. labriformis was relatively constant as the reproduction period progressed. Furthermore, Estrada-Godínez et al. (2011) mentioned that Mycteroperca rosacea, as well as other serranids and tropical fish species, do not stop feeding during the maturation and spawning period.
Many species of the Serranidae family are hermaphrodites, including those of the genera Cephalopholis, Epinephelus, Mycteroperca, and Plectropomus (Sadovy de Mitcheson and Liu 2008). According to Sadovy and Shapiro (1987), to determine if a species is a hermaphrodite, we must observe individuals in the process of changing sex (individuals in transition), for example, a regressing ovary that has cysts with cells in some phase of spermatogenesis appearing in the germinal epithelium (Smith 1965). Although P. colonus belongs to the Serranidae family, we observed no organisms in transition from female to male or vice versa, found no differences in the length interval between females and males, and found both small and large females and males, so we can assume that this species is not a hermaphrodite. Sadovy de Mitcheson and Liu (2008) mentioned that some species are known to be gonochoric and cited, among them, the species of the genus Paralabrax, Epinephelus striatus, M. rosacea, and P. furcifer.
No studies have been found that describe the development of P. colonus oocytes. Considering the results obtained in the present work, the development is of the asynchronous type. Nagahama (1983) and Maack and George (1999) indicated that species that show oocytes with asynchronous development can have a long reproductive season, during which they spawn several times. In view of the above, we agree with these authors, since P. colonus showed a prolonged reproductive period with 2 maximum stages of 4 and 2 months each.
Few studies have analyzed testis development in fish, since changes in the cells are difficult to distinguish (West 1990). The P. colonus testis was organized by lobes, and these, by cysts within them, which are traits of lobular-type development. This type of testicular organization has been recorded in different fishes (Grier et al. 1980; Grier 1993; Albieri et al. 2010; Ruiz-Ramírez et al. 2012, 2017; Lucano-Ramírez et al. 2014, 2016, 2017). In addition, we observed folds in the testis of P. colonus (similar to the lamellae of the ovaries) where the sperm cells were organized, something also observed in E. labriformis (Ruiz-Ramírez 2018), but we never observed oocytes.
To rationally manage fishing on populations subject to exploitation, identifying L50 is essential because it allows proposing regulation measures (minimum catch length, certain fishing gear) (Jennings et al. 2001, Ruiz-Ramírez et al. 2018). This and other studies have estimated L50 for different species in order to learn if they reproduce at least once before being captured (Ruiz et al. 1999, Guirao et al. 2005, Lucano-Ramírez et al. 2016, Ramírez et al. 2018), a highly recommended estimation (Beverton and Holt 1957). In the present work, we observed that the L50 of females and males were close (26.80 and 26.90 cm, respectively). This observation is common and has been reported for Diodon holocanthus (Lucano-Ramírez et al. 2011) and Lutjanus argentiventris (Lucano-Ramírez et al. 2014); although Posada-López (1996) did not mention the L50 for P. furcifer, the smallest female and male with mature gonads had similar lengths (14.40 and 13.80 cm fork length, respectively). In the study region, commercial catches of P. colonus include a large percentage of individuals (80%-83%) that have already had at least one reproductive event, which indicates that a large part of the population manages to reproduce and give continuity to the species. It is very important to continue studies bearing basic information on this species that generates medium- and long-term benefits.
