INTRODUCTION
Skates have been identified as one of the most vulnerable groups of marine fish to intensive fishing pressure, particularly due to life-history traits such as low fecundity, slow growth, delayed maturity and long incubation period (Holden, Reference Holden1973; Brander, Reference Brander1981). In addition to the concern over the collapse of the fisheries and possible extinction of some species, a further problem is the negative effects that drastic population declines can have on ecosystems (Holden, Reference Holden1973, Reference Holden and Harden Jones1974; Brander, Reference Brander1981; Walker, Reference Walker1998; Castro, Reference Castro1999; Dulvy et al., Reference Dulvy, Metcalfe, Glanville, Pawson and Reynolds2000; Shepherd & Myers, Reference Shepherd and Myers2005; Dudley & Simpfendorfer, Reference Dudley and Simpfendorfer2006).
The genus Psammobatis is endemic to the south-western Atlantic Ocean and it is represented by 7 species (McEachran, Reference McEachran1983; Menni & Stehmann, Reference Menni and Stehmann2000). Among these, systematic, morphological and biological aspects have been studied in the following species: Psammobatis bergi, P. extenta, P. rudis, P. normani and P. rutrum (McEachran, Reference McEachran1983; Mabragaña & Cousseau, Reference Mabragaña and Cousseau2004; Braccini & Chiaramonte, Reference Braccini and Chiaramonte2002a, Reference Braccini and Chiaramonteb; Sánchez & Mabragaña, Reference Sánchez and Mabragaña2002; Braccini & Perez, Reference Braccini and Perez2005; Martins et al., Reference Martins, Martins and Leme2005; San Martín et al., Reference San Martín, Perez and Chiaramonte2005, Reference San Martín, Braccini, Tamini, Chiaramonte and Pérez2007; Cousseau et al., Reference Cousseau, Figueroa, Díaz de Aztarloa, Mabragana and Lucífora2007; Mabragaña, Reference Mabragaña2007; Mabragaña & Gilberto, Reference Mabragaña and Giberto2007).
The skate Psammobatis lentiginosa is distributed between 32° and 45°S at depths ranging from 70 to 170 m. (Menni & Stehmann, Reference Menni and Stehmann2000; Mabragaña, Reference Mabragaña2007; Cousseau et al., Reference Cousseau, Figueroa, Díaz de Aztarloa, Mabragana and Lucífora2007; Di Giacomo et al. in press). Available information deals with species description and ecological and biological aspects (McEachran, Reference McEachran1983; Menni & Stehmann, Reference Menni and Stehmann2000; Cousseau et al., Reference Cousseau, Figueroa, Díaz de Aztarloa, Mabragana and Lucífora2007; Mabragaña, Reference Mabragaña2007; Di Giacomo et al., in press). In the San Matías Gulf (41°–42°S 64°–65°W), a small-scale trawl fishery targeting the common hake, Merluccius hubbsi, has been developing since 1971 (Di Giacomo & Perier, Reference Di Giacomo and Perier1992). Psammobatis lentiginosa is part of the by-catch and discarded at sea due to its small size. The objective of the present study is to determine life-history parameters of P. lentiginosa, with particular emphasis on its reproductive biology.
MATERIALS AND METHODS
Sampling
Skates were caught during bottom trawl surveys performed with commercial vessels and random selected sampling from commercial catches by onboard observers. The surveys were carried out in San Matías Gulf, northern Patagonia (Figure 1), one in October–November 2004 and the other in November–December 2005, aboard the fishing vessels ‘Chiarpesca 59’ and ‘Marina Z’, respectively. These were equipped with 96-ft commercial bottom trawls. In concordance with the methodology described by Di Giacomo & Perier (Reference Di Giacomo and Perier1991, Reference Di Giacomo and Perier1992) a variable number of hauls (35–40) of 30 minutes each were performed at a depth-range of 40–175 m. The total number of skates captured per haul and the total length (L, in cm) of each specimen were recorded during each survey.
San Matías gulf. Bottom trawl survey. Random stratified sampling. Strata depths 1, 2, 3 = 90–130 m; 4 > 130 m; 5 = 50–89 m; 6 = 20–49 m.
Skates caught by commercial vessels were bimonthly sampled by onboard observers (Instituto de Biología Marina y Pesquera Almirante Storni, Onboard Observers Program) between March 2005 and May 2007. In each commercial trip, when observers were onboard, all skates caught were kept for sampling. Samples of two sources were properly labelled and kept cold onboard before being sent to the laboratory for the morphometric and reproductive analyses described below.
Morphometric analysis and sex-ratio
Sex, total length (TL, cm), disc width (DW, cm) and total weight (TW, g) were recorded for each fish. Total length and disc width were measured to nearest cm from the top of the rostrum to the tip of the tail and the maximum distance between the edges of pectoral fins, respectively. The relationships between TL and DW (TL = a + bDW) and between TL and TW (TL = a. TWb) were calculated for each sex; data were log-transformed and a Student's t-test was used for comparisons of slopes and intercepts. Differences in the proportions of sexes for each season were evaluated by the Chi square (χ2)-test (Zar, Reference Zar1984).
Reproductive analysis
Sexual maturity degree was determined following the classification of Holden & Raitt (Reference Holden and Raitt1975) and Stehmann (Reference Stehmann2002). In males, sexual maturity was determined by the rigidity and length of the myxopterygia or claspers (Holden & Raitt, Reference Holden and Raitt1975; Stehmann, Reference Stehmann2002), which were measured to nearest mm from the caudal end of the cloaca to their tip (CL). Specimens were dissected to expose the reproductive system. Liver and reproductive organs were drained and the mass determined. Testes and epididymis were weighed (g).
Males were classified as virginal or immature (claspers not reaching margin of pectoral fin), maturing (semi-calcified claspers exceeding the pectoral fins but flexible) and mature (fully calcified and rigid claspers longer than pectoral fins).
In females, ovaries, oviducal glands and oviducts were weighed (g). All the oocytes of diameter larger than 1 mm where counted and the yolked oocytes were measured (mm). Females were classified as virginal or immature (oviducts, oviducal gland and ovaries undeveloped), maturing (oviducts and oviducal gland developing and ovarian follicles without yolk) or mature (oviducal glands fully developed, oviducts with or without egg cases and ovarian follicles with yolk). Differences in the proportions of mature to immature specimens for the entire dataset and for each season were evaluated by the χ2-test (Zar, Reference Zar1984).
In mature females, differences in the number of yolked oocytes between right and left ovaries were assessed by the Wilcoxon paired-sample test. Ovarian fecundity was determined as the number of follicles with yolk in the ovaries. Regression analyses were performed to evaluate the possible relationships between the number and diameter of mature oocytes and female length. The number of females carrying egg cases and total length (without horns), width, thickness (all mm) and weight (g) of completely formed egg cases were determined on a seasonal basis.
The length of the smallest mature specimen was recorded for each sex. Size at 50% of maturity (LT50%) was estimated using the FISHPARM. 3.05 program (Prager et al., Reference Prager, Saila and Recksiek1987).
The reproductive activity was indirectly assessed using several indices. The gonadosomatic index (GI = ovaries or testes weight * 100/total weight) and hepatosomatic index (HI = liver weight * 100/TW) were calculated for both sexes. The oviducal gland index (OGI = oviducal gland weight * 100/TW) and uterus index (UI = uterus weight * 100/TW) indices were calculated for females and the epididymis index (EI = epididymis weight * 100/TW) for males. The condition factor (CF = TW/TLb, where b is the exponent of the relationship between TL and TW obtained for each sex) was calculated for males and females.
Comparisons of TL–TW relationship and TW versus liver weight (LW) were analysed for seasonal and individual variations and related to the reproductive cycle.
The mean values of the indices were analysed for each season in order to characterize the reproductive cycle of the skate. Differences were tested by the Kruskal–Wallis non-parametric ANOVA, followed by multiple comparisons (Q) when they were necessary (Zar, Reference Zar1984).
RESULTS
Morphometric analysis
Female length ranged from 17 to 52 cm, while male length ranged from 13 to 55 cm (Figure 2). A total of 1033 skates, 514 females and 519 males, were sampled in the two surveys. The DW–TL relationship was DW = 0.56 * TL –0.34 (N = 269, r2 = 0.93) for females, and TW = 0.56 * TL –0.44 (N = 268, r2 = 0.94) for males. There were no significant differences between the sexes in either the slopes or intercepts of these relationships (t-test, df = 552, P > 0.05 for slopes and t-test, df = 553, P > 0.05 for intercepts). The TL–TW relationship differed significantly between sexes, with females being heavier than males (t-test, df = 1028, P > 0.05 for slopes and t-test, df = 1029, P < 0.05 for intercepts) (Figure 3).
Size–frequency distribution of Psammobatis lentiginosa. Female (black bars), N = 514; male (white bars), N = 519.
Total length–total weight relationship of Psammobatis lentiginosa. Female, N = 508, (-), R2 = 0.94; male, N = 518, (- -) R2 = 0.96.
The LW–TL relationship was LW = 7 * 10–05 TL3, 20 (r2 = 0.67) for females, and LW = 1* 10–4 TL2, 66 (r2 = 0.61) for males. These relationships differed significantly between sexes (t-test, df = 394, P < 0.05 for slopes) with livers of females being heavier than those of males at given TL. There was a noticeable individual variation in hepatic mass for each length-class (class interval = 1 cm) as males and females reach sexual maturity (Figure 4).
Total length–liver weight relationship of Psammobatis lentiginosa. Female (o), N = 223, (-), R2 = 0.71; male (•), N = 183, (- -) R2 = 0.61.
Sex-ratio
No significant differences from the expected 1:1 ratio were found for the overall female:male ratio (χ2 = 0.001; df = 1; P > 0.05; N = 1032), the immature female:male ratio (χ2 = 3.6; df = 3; P > 0.05; N = 452) and the mature female: male ratio (χ2 = 7.65; df = 3; P > 0.05; N = 580). Seasonal significant difference in female : male ratio was found in spring (χ2 = 12.39; df =3; P > 0.05; N = 407).
Maturity and reproductive analysis
The smallest mature male was 38 cm L and all males > 46 cm L were mature showing well-calcified claspers (Figure 5) and fully developed testis and epididymis (Figure 6). Transition from immaturity to maturity was observed in males between 38 and 46 cm L. For males, L50% was estimated to be 41.38 cm, representing 75.3% of the largest male sampled (Figure 7A). All females < 37 cm L were immature or maturing, while those > 45 cm L were mature having fully developed oviducts, oviducal glands and ovaries with mature yolked oocytes. In females the transition from immaturity to maturity was observed between 37 and 45 cm L. The L50% for females was estimated to be 41.06 cm L, which represented 79% of the largest female sampled (Figure 7B). The smallest female carrying egg cases, was 39 cm L.
Total length–clasper length relationship by maturity stages of males of Psammobatis lentiginosa. Immature ( ) N = 83; maturing (o) N = 44; mature (•) N = 202.
Dissected reproductive of mature male of Psammobatis lentiginosa. T, testis; Ep, epididymis. White bar = 3 cm.
(A) Size at first maturity (L50%T) of males of Psammobatis lentiginosa; (B) size at first maturity (L50%T) of females of P. lentiginosa.
There were no significant differences in the number of mature yolked oocytes between the left and right ovaries (Mann–Whitney, N = 43, P > 0.05). The mean number of yolked oocytes per skate (ovarian fecundity) was 4.22 (N = 43, SD = 2.41). A linear relationship was found between the mean number of yolked oocytes and the L of the female (F, N = 50, P > 0.05). The diameter of yolked oocytes ranged between 6 and 24 mm.
Dimensions of the egg cases are shown in Table 1 (Figure 8). There were no significant differences in the length and width of the egg cases (Mann–Whitney, N = 9, P > 0.05), between the left and right oviducts. Mature females carrying egg cases represented 13.4% (Figure 9) of all mature females; they were found all year round, with the highest percentage being recorded in autumn (Figure 10).
Egg case dimension of Psammobatis lentiginosa.
Dissected reproductive female tract of Psammobatis lentiginosa with egg case in uterus. Ov. : ovaries; O.g., oviducal glands; U, uterus; e.g., egg case. Black bar = 3 cm.
Seasonal variation of percentage of females with egg-cases in uterus of Psammobatis lentiginosa.
Length, width and thickness (mm) of egg case of Psammobatis lentiginosa.
–, not available.
The male IH did not vary significantly among seasons (Kruskal–Wallis, df = 3, P > 0.05) and the highest value was observed in summer (Figure 11A). The GI showed significant differences among seasons (Kruskal–Wallis, df = 3, P < 0.05) and pairwise comparisons revealed that IG values were significantly lower and higher in summer and spring, respectively (Q, df = 4, P < 0.05; Figure 11B). The IE did not vary significantly among seasons (Kruskal–Wallis, df = 3, P > 0.05), with the maximum value being obtained in spring and summer (Figure 11C). The CF showed lowest and highest values in autumn and winter, respectively (Figure 11D).
Seasonal variation of mature male reproductive indices of Psammobatis lentiginosa. (A) Hepatosomatic index; (B) gonadosomatic index; (C) epididymis index; (D) condition factor. Black squares represent mean values and bars ±1 standard deviation.
The female HI showed significant differences among seasons (Kruskal–Wallis, df = 3, P < 0.05) but contrasts were non-significant (Q, df = 4, P > 0.05) (Figure 12A). The highest HI value was obtained in summer. The GI differed significantly among seasons (Kruskal–Wallis, df = 3, P < 0.05; Figure 12B), although contrasts showed no significant differences. The highest GI value was recorded in spring.
Seasonal variation of mature female reproductive indices of Psammobatis lentiginosa. (A) Hepatosomatic index; (B) gonadosomatic index; (C) oviducal gland index; (D) uterus index; (E) condition factor. Black squares represent mean values and bars ±1 standard deviation.
The mean oviducal gland index (OI) did not vary significantly among seasons (Kruskal–Wallis, df = 3, P > 0.05) and the higher values were reached in summer and spring (Figure 12C). The mean uterus index (UI) varied significantly among seasons (Kruskal–Wallis, df = 3, P < 0.05) and pairwise comparisons indicated significant differences between summer autumn, autumn–spring and spring–winter (Q, df = 4, P < 0.05). The highest and lowest values of this index were recorded in autumn and spring, respectively (Figure 12D). The condition factor showed its highest value during spring. The lowest value was observed during autumn (Figure 12E).
DISCUSSION
The present study reports for the first time basic information on the reproductive cycle of the skate Psammobatis lentiginosa, endemic to the south-western Atlantic, and on the reproductive aggregation of this species in San Matías Gulf. Individuals reproductively active—female with egg case in formation—were reported for the north of the Argentinean continental shelf by Mabragaña (Reference Mabragaña2007).
Sexual size dimorphism favouring larger females is common in chondrichthyans (Mellinger, Reference Mellinger1989). In P. lentiginosa, females are heavier than males of the same size. The relationship between total length and total mass was allometric for both sexes. This feature has been reported for almost all species of the genus Psammobatis studied so far except for P. rudis (Mabragaña & Couuseau, Reference Mabragaña and Cousseau2004). Many species of Arhynchobatidae show allometry (P. extenta, Braccini & Chiaramonte, Reference Braccini and Chiaramonte2002a; P. normani, Mabragaña & Cousseau, Reference Mabragaña and Cousseau2004; genus Atlantoraja, Odone & Amorin, Reference Odone and Amorim2007).
In the studied P. lentiginosa population, overall sex-ratio was 1:1, similarly to that observed for P. rudis (Mabragaña & Cousseau, Reference Mabragaña and Cousseau2004; Mabragaña, Reference Mabragaña and Giberto2007), and seasonal sexual segregation was detected, in concordance with P. bergi, P. normani and P. extenta (Bracini & Chiaramonte, Reference Braccini and Chiaramonte2002b; Mabragaña & Cousseau, Reference Mabragaña and Cousseau2004; San Martín et al., Reference San Martín, Braccini, Tamini, Chiaramonte and Pérez2007).
In comparison with the reproductive and morphometric patterns of other species of Psammobatis from the south-western Atlantic, P. lentiginosa attains sexual maturity at a value of TL50% lower than the maximum one reported in literature. This feature suggests that P. lentiginosa is the least vulnerable species of the genus because more than 60% of the size–length distribution was adults.
Some species of Rajiformes, P. lentiginosa included, show significant differences in the hepatic weight between sexes (Mabragaña et al., Reference Mabragaña, Lucifora and Massa2002; Odone & Vooren, Reference Odone and Vooren2004), while others do not (Braccini & Chiaramonte, Reference Braccini and Chiaramonte2002b; Ruocco et al., Reference Ruocco, Lucifora, Díaz de Astarloa and Wholer2006). The analysis of the hepatic weight–total length relationship of P. lentiginosa during its ontogeny shows a variation in the interval of body length corresponding approximately to the stage of sexual maturity, possibly related to changes in morphology and physiology. This may also explain the large dispersion of hepatic weight values in individuals of the studied species at the onset of sexual maturity. This result has also been observed for A. cyclophora.
Storage of lipids in the liver and their mobilization during vitellogenesis have been described in chondrichthyans by Rossouw (Reference Rossouw1987) and mentioned in some rajiform species by Odone & Velasco (Reference Odone and Velasco2006). In P. lentiginosa, the maximum hepatic index values were obtained in summer. This result is possibly explained by the fact that the maximum egg case production is found in the following season (autumn), when lipid storage is expected to be the highest.
The variation of the condition factor may help to elucidate the mobilization of energetic reserves (Nikolsky, Reference Nikolsky1963) and in Rajidae, this index has been used as a complementary indicator of energy status (Du Buit, Reference Du Buit1975). In P. lentiginosa, the lowest CF of females was obtained in autumn, when, as mentioned above, there is a peak in egg case production. This activity would be achieved at the expense of increased energetic cost, which may ultimately result in a decline in the general condition.
The existence of functional symmetry of the female reproductive system is a common feature among species of Rajiformes (Braccini & Chiaramonte, Reference Braccini and Chiaramonte2002b; Mabragaña & Cousseau, Reference Mabragaña and Cousseau2004; Martins et al., Reference Martins, Martins and Leme2005; San Martín et al., Reference San Martín, Perez and Chiaramonte2005; Odone & Vooren, Reference Odone and Vooren2005; Ruocco et al., Reference Ruocco, Lucifora, Díaz de Astarloa and Wholer2006). In P. lentiginosa both the right and left ovaries showed to be functional and no significant differences in the number of oocytes were found between them. Likewise, no functional asymmetry was detected between the two uteri of females carrying eggs (one egg capsule per uterus). The maximum diameter of the mature oocyte in the ovary was 24 mm. The lack of data on this subject in other species of the genus prevents any further comparison.
The suitability of the indices of reproductive activity for defining reproductive cycles or to distinguish immature from mature individuals has been confirmed for several fish species (Wilk et al., Reference Wilk, Morse and Stehlik1990; Di Giacomo & Perier, Reference Di Giacomo and Perier1994; Perier & Di Giacomo, Reference Perier and Di Giacomo2002; Sulikowski et al., Reference Sulikowski, Kneebone, Elzey, Jurek, Danley, Howell and Tsang2005). In the present study, they proved to be useful tools to investigate temporal variations in the reproductive pattern of P. lentiginosa.
The following two reproductive cycles have been identified in species of Rajiformes: an annual reproductive cycle with one or two seasonal peaks in Atlantoraja cyclophora (Oddone & Vooren, Reference Odone and Vooren2005), Psammobatis bergi (San Martín et al., Reference San Martín, Perez and Chiaramonte2005), Raja clavata (Holden, Reference Holden1975) and an annual reproductive cycle with no marked seasonality in P. extenta (Martins et al., Reference Martins, Martins and Leme2005) and Bathyraja albomaculata (Ruocco et al., Reference Ruocco, Lucifora, Díaz de Astarloa and Wholer2006). Taking into account the percentage of females with egg cases in development or ready to be laid, P. lentiginosa has an annual reproductive cycle with a peak in autumn and a decline in winter.
The peak of egg capsule production was synchronous with the maximum development of the glandular caudal portion of the uterus, as indicated by the highest value of UI obtained at that time. Seasonal variations in UI values may reflect an active role of the caudal portion during the formation, storage and release of the egg capsules, although in Rajiformes, it is not known if this uterine glandular portion secretes substances for capsule production (Hamlett & Koob, Reference Hamlett, Koob and Hamlett1999).
In the uterus of P. lentiginosa, a fecundated ovule was found in egg capsules over 50% of development, in agreement with that found in many oviparous fish species, particularly Rajiformes and holocephalans (Di Giacomo & Perier, Reference Di Giacomo and Perier1994; Hamlett et al., Reference Hamlett, Knight, Pereira, Steele, Sever and Hamlett2005). In Rajiformes, spermatophores have been detected in the cephalic portion of the nidamental gland (Holden, Reference Holden1975; Hamlett & Koob, Reference Hamlett, Koob and Hamlett1999), but their presence could not be confirmed in P. lentiginosa because of the impossibility of performing serial histological sections of the reproductive system.
Another particularity in the reproductive cycle is related to the mating period. Our results suggest that although mating showed a seasonal pattern for most of the population, it may take place throughout the year in some individuals. On the other hand, the highest percentage of mature males and females in summer is related to an intense mating period previous to the maximum production of eggs in autumn. This observation is supported by the highest values of IE and IH shown by males in summer, before the peak in the number of egg capsules appearing in females in the next season. A similar pattern has been observed for Raja clavata by Holden (Reference Holden1975). In addition, the high variability of the indirect indicators of reproductive activity in each length-class suggests the coexistence of individuals at different maturity stages, with some females producing capsules and others undergoing mating.
The frequency of mating could not be established for P. lentiginosa due to the lack of data on the physiological duration of spermatophores in the reproductive system of female skates after mating. In the different Rajiformes, the egg-laying areas vary according to the bathymetric distribution of each species. For example, egg capsules of the species Bathyraja albomaculata, which is distributed up to slope waters, have been found in deep waters simultaneously with females carrying developing or fully developed egg-cases (Ruocco et al., Reference Ruocco, Lucifora, Díaz de Astarloa and Wholer2006), while species of genera Atlantoraja and Raja lay their egg capsules in coastal areas (Holden, Reference Holden1975; Mabragaña et al., Reference Mabragaña, Lucifora and Massa2002; Odonne et al., Reference Odone and Amorim2007). No egg capsules of P. lentiginosa were found in the present work, probably because the sampling area did not overlap the oviposition area of the species (between 50 and 171 m), or because the fishing gear used to catch skates (trawl nets) was not appropriate to collect egg capsules from the bottom.
These constraints in sampling egg capsules have been pointed out by Odone & Vooren (Reference Odone and Vooren2004), who also mentioned that egg capsules could be severely damaged or buried in the sediment while the nets are being dragged. The permanent presence of the species and the reproductive aggregation revealed by the seasonal samplings, together with the almost complete absence of the species on the continental shelf—only is reported in the north of San Jorge Gulf, 54°S and in the fishing area common to Argentina and Uruguay (McEacrhan, Reference McEachran1983; Cousseau et al., Reference Cousseau, Figueroa, Díaz de Aztarloa, Mabragana and Lucífora2007; Mabragaña, Reference Mabragaña2007)—indicate that San Matías Gulf is the main distribution area of P. lentiginosa in the south-western Atlantic.
The negative impact of fishing over the exploitation of Rajiformes species has been documented by Holden (Reference Holden1973, Reference Holden and Harden Jones1974), Brander (Reference Brander1981) and Dulvy et al. (Reference Dulvy, Metcalfe, Glanville, Pawson and Reynolds2000). Psammobatis lentiginosa is not targeted by the fishery in San Matías Gulf and all catch of this species is discarded. The lack of capture records makes it difficult to estimate fishing mortality of the population, and although experiments have been made to ascertain the likely survival rate of another species of skates discarded during commercial fishing (Chiaramonte, Reference Chiaramonte2006), information on P. lentiginosa is needed for the appropriate management of the local fishery.
ACKNOWLEDGEMENTS
This work is a contribution of the CONDROS group. Thanks are due to the ‘Programa de Observadores a Bordo’, for allowing sampling on board of fishing vessels, to the crews of ‘Chiarpesca 59’ and ‘Marina Z’ for help and assistance at sea and to Miguel Camarero for sampling assistance. This work was supported by the Universidad Nacional del Comahue and the Ministerio de la Producción of Provincia de Río Negro.
