Abstract: In the present study, some reproductive parameters including sex ratios, maturity stages, length and age at first maturity gonadosomatic and gonad indices, egg diameter and fecundity of Mormyrus kannume from the Nile, Egypt, were estimated in relation to standard length, weight, age and months. Other population parameters describing such population were also estimated. Different patterns of variations were recorded in these parameters. Overall sex ratio was significantly balanced in the Nile but not in the lacustrine locality: Lake Nasser. The females and males of M. kannume reached maturation at 31.7 and 31.5 cm with reproductive load of 0.754 and 0.748, respectively. Monthly variations in ED were evident with similar variability in GSI and GI. ED has its highest values in April and May and its lowest values in November and December (corresponding to highest fecundity). No association between ED and age groups whereas ED showed variability among maturity stages. Fecundity exhibited variability with age, length and months in the same locality. Fecundity highest average value was recorded in December (12623±6811).There is no association between fecundity and maturity stages and fish condition factor. The fecundity increased with age groups and has its highest values in age group I and II and lowest one in age VI. Fecundity also showed variations with time and locality when compared with other studies. In spite of estimation of its relation with other variables in univariate term, fecundity and variables controlling it were best modeled by stepwise multiple linear regression which isolated fish size, egg diameter and gonad index to be the best significant factors influencing fecundity and valid for prediction. In conclusion, M. kannume is a fecund fish with variations in time and localities.
INTRODUCTION
Estimation of reproductive potential (RP) and recruitment of different species is important and at the same time, difficult process (Hilborn and Walters, 1992). So, the reproductive parameters represent a powerful tool in fisheries management and the related modeling processes. Fishery scientists considered fecundity as one of the most interested reproductive biology aspects in stock assessment based on egg production methods (Saville, 1964; Koslow et al., 1995; Morgan, 2008; Mekkawy and Hassan, 2011). In addition, Wang et al. (2003) and Morgan (2008) stated that the size and age at sexual maturity and the sex ratio are fundamental biological parameters in stock assessments and fishery management modeling. In this respect, estimates of body size and /or age at first sexual maturity are also, necessary parameters for age- and size-structured models, such as the spawner biomass per recruit model (Gabriel et al., 1989), per recruit model (Foale and Day, 1997) egg and other size-or age-structured models (Derios et al., 1985; Quinnll et al., 1990). Morgan (2008) stated that reproductive biology play an important role in determining productivity and in turn a population’s resiliency to exploitation by fisheries as well as perturbation caused by other human activities. These findings provide sound scientific advice for fisheries management
Estimation of the reproductive aspects in terms of fecundity, egg size, size and length at sexual maturity, sex ratio, gonadosomatic index are essentials in consideration of life history theory. The goal of current life history theory is to predict how an organism’s finite resources, time and energy, are allocated towards growth, maintenance and reproduction (Cody, 1966). In this concern, Williams (1966), Pianka (1976) and Willbur (1977) emphasized on a compromise between the organism’s immediate prospects of reproductive success versus lifelong future prospects.
Mormyrids are endemic to Africa represented by about 100 species in two subfamilies (Aly, 1993) and represented in Lake Naser, Egypt by 7 species. Twenty species under genus Mormyrus were recorded in Africa (Froese and Binohlan, 2000). M. kannume is the only economic species represented downstream in the Nile with few small sized species of other mormyrid species after the construction of the High Dam in 1969 at Aswan, Egypt. Reviewing literature at hand on M. kannume and /or M. caschive (Ashour et al., 1990a; Zaher et al., 1991; Aly, 1993), no detailed research has been done on the reproductive biology of these species in the Nile. However, the reproductive biology of many Nile fish species was the focus of many authors (Mekkawy and Hassan, 2011).
According to the scientific information on M. kannume species of the Egyptian sector of the Nile, at Assiut, Egypt, the present work aimed at estimation of some of the reproductive parameters of Mormyrus kannume including sex ratios, size/age at first maturity, maturity stages, gonadosomatic and gonad indices and different fecundity relationships which can be included in fishery and stock assessment models of that species.
MATERIALS AND METHODS
One hundred and sixty specimens of M. kannume (16-47 cm in SL) were collected from the Nile at Assiut, Egypt, dissected and examined. Fish and gonads were weighed to the nearest gram and standard length was measured to the nearest centimeter. Ovaries were preserved in 10% formalin for later estimation of fecundity. For age determination of specimens used in this study, the first five vertebrae of each fish were cleaned and examined using binocular microscope.
Sex ratio: Sex ratio was determined for the two species on fish size and monthly bases. Sex ratio departure from the expected 1:1 ratio was testified. Moreover, sex distribution through length groups and months was also considered.
Maturity stages: The maturity stages of ovaries and testes were identified and distinguished according to the following basis (Aly, 1993):
Stage I (immature): gonads are small, ovary is transparent. Eggs are small and cannot be detected by the naked eye. Testis is thread-like.
Stage II (resting stage): gonads are still small and extend along 1/3-1/2 the length of the abdominal cavity. The ovary is still transparent and not thick. The eggs still cannot be distinguished with the naked eye. Testis is still without well-defined side lobes.
Stage III (maturation): the gonad begins to develop and extend along ¾ of the abdominal cavity length. Ovary is yellowish. Eggs are distinguished to naked eye. Testis starts to have side lobes.
Stage IV (maturity): sexual products are clear. The gonads nearly extend along the whole length and reach their maximum weight. Ovaries are orange but the eggs still not extruded while the testis lobes are clear, finger-like and full of sperms.
Stage V (reproduction): Ovaries are voluminous with large viable eggs which could be seen through the thin ovarian wall. Testes are whitish in color with stout lobes which are distended with sperms.
Stage VI (Spent): The ovary is flaccid, reddish black in color and much reduced in size, the testes are yellowish, white in color, soft, empty and fleshy in its appearance.
In the analysis, the maturity stages I and II were together considered as immature, stage III as maturing, while stage IV and V are considered as mature/ripe. Length (Lm) and age (Tm) at first maturity were estimated for males and females.
Gonadosomatic (GSI) and Gonad Indices (GI): The gonadosomatic index (GSI) was predicted by the following formula (Khallaf, 1986): GSI = GW *100/W where GW is the gonad weight in g and W is body weight in g. The Gonad Index (GI) was predicted by the following formula (Scrimgeour and Eldon, 1989):
GI = GW *108 / SL3. Two condition Factors (Kn = W* 100/SL^n and Kc= W*100/SL^3; n=allometric coefficient) were estimated.
Fecundity: Fecundity (F) is the number of eggs per ovary of mature fish (Bagenal and Tesch, 1978), whereas relative fecundity is the number of egg per unit length (RFSL) or weight (RFW) of fish. Three pieces per ovary were cut and weighed, then the eggs were counted and fecundity and relative fecundity were calculated. Egg Diameter (ED) was measured to the nearest micron by an ocular micrometer fixed in the eye-piece of a light microscope.
Statistical analysis: Length and size at first maturity (Lm and Tm) were estimated using probit analysis. Length-weight power function equation parameters (a and n) and SL-TL linear equation (a and b) and Bertalanffy equation parameters (L∞, K, T0 ) as well as reproductive load (Lm/L∞) were estimated. Monthly and size variations in the reproductive parameters were analyzed with one way ANOVA and Post Hoc Tukey-test; homogeneity of variance was evident by Levene’s Test in most cases and assumed in few others since no transformations were valid. Fecundity relationships with size, age, weight maturity stages and egg diameters were analyzed and fitted. The fecundity in relation to SL, W, GW, ED, GSI, GI, Age, M (maturity stage), Kn and Kc was subjected to stepwise multiple linear regression (forward and backward) to determine the best variables that control fecundity in a multivariate sense. Correlation, multiple correlation and partial correlation coefficients were estimated. Sex ratio was analyzed and fitted by Chi-square test. All tests were considered at 0.05-level of significance. SPSS-package release 9 (SPSS, 1998), Statistica package release 5 (Statsoft, 1995) and FiSAT-II package (Gayanilo et al., 2005) were used in these statistical analyses and models.
RESULTS
Some biological parameters: The males and females of Mormyrus kannume population studied grow positively allometricly and isometricly respectively and are characterized by the following general biological parameters (Table 1).
Sex ratio: The sex ratio of M. kannume of Assiut (male: female) was 1: 0.84 and did not depart from the expected 1:1 rate (Table 2). However there was significant association between sex distribution and length groups (χ2 = 21.28; p = 0.0034) (Table 3). On the contrary, no significant association between sex and months (χ2 = 15.08; p = 0.1788). Sex ratio of March (1: 0.25) was the only ratio that departed from the expected ratio 1:1 (χ2 = 5.4; p<0.05) whereas only in length groups 20 and 32, sex ratios (1: 0.16 and 1: 2.06, respectively) were significantly different from the expected one 1:1 (χ2 = 11.64 and 5.56 at p<0.05). Such ratios were responsible for the significant association between sex distributions and length groups. In spite of these findings, fishes were numerically insufficient to permit a reliable conclusion in some length groups and some months.
Size and age at first sexual maturity (Lm and Tm): The age and size-distribution of the mature and immature M. kannume are given in Table 4 and 5. From such distributions, Lm and Tm of M. kannume were estimated to be:
Reproductive load: The reproductive loads (Lm/L∞) of males, females and combined sex of M. kannume from the Nile at Assiut were 0.748, 0.754 and 0.595, respectively.
Maturity stages: The distribution of maturity stage categories (i.e., I+II, III, IV+V) showed variations among length groups in both males and females. The maturity increases with increase of fish size (Table 6). On other hand, all maturity stages categories of females were represented in March and July-October period. However, IV+V category was represented along the period of study which means a long period of spawning (Table 7). These female findings were different from those of males (Table 7). Seasonal variations were evident (p<0.0001).
| Table 1: | Parameters and their standard errors of Bertalanffy and length-weight power function equations (coefficient of variation is in parenthesis; SL (cm) and W (g) |
| Table 2: | Sex ratio of M. kannume in different length groups from Assiut (2000-2001 periods) |
| Table 3: | Seasonal variations in sex ratio of M. kannume from Assiut (2000-2001 periods) |
| Table 4: | Distribution of mature and immature individuals of M. kannume from the Nile at Assiut, Egypt. Among different age groups in the period November 2000-October 2001 |
| Table 5: | Distribution of mature and immature individuals of M. kannume from the Nile at Assiut, Egypt among different length groups in the period November 2000-October 2001 |
| Table 6: | Length groups variations of different maturity stages of males and females of M. kannume from the Nile at Assiut, Egypt |
| Table 7: | Monthly variations of different maturity stages of females and males of M. kannume, from the Nile at Assiut, Egypt |
| Table 8: | Gonadosomatic index and gonad index of males and females in different length groups of M. kannume from the Nile at Assiut, Egypt |
| Table 9: | Monthly variation with gonadosomatic index and gonad index of males and females of M. kannume from the Nile at Assiut, Egypt |
| Table 10: | Egg diameter, GSI and GI in different length groups of M. kannume from the Nile at Assiut, Egypt |
Gonadosomatic (GSI) and Gonad (GI) Indices: The basic statistics of GSI and GI are shown in Tables 8 and 9 for males and females of M. kannume. The distributions of GSI and GI categories (Table 8) show variability among length groups. For females (N = 85), GSI (ungrouped data) was significantly correlated with SL (R = 0.77) and insignificantly with body weight (W) (-0.16), whereas GI (ungrouped data) exhibited no significant correlation (R = 0.04) with SL and positive significant correlation with body weight (R = -0.42). GSI and GI were weakly correlated (R = 0.43). Gonad weight (GW) was significantly correlated with SL (R = 0.69) and insignificantly correlated with W (R = -0.06). For males (ungrouped data, N = 101), GW, GSI, GI were significantly correlated with SL (R = 0.38, -0.50 and -0.53, respectively) and W (R= 0.42, -0.46 and -0.46, respectively). GSI and GI were highly correlated (R = 0.97). The last categories with higher magnitudes of GSI and GI were represented by low or zero percent for females. Similar situation was recorded for males especially for GSI (Table 9). In addition to these findings, GSI and GI exhibited monthly variations with regards to their values (p<0.0001) and categories distribution.
Fecundity: Basic statistics of Egg Diameters (ED) of mature M. kannume females and the corresponding GSI and GI are given in Table 10. No significant variability in average Egg Diameter (ED) with increase of fish size (R2 = -0.13) for ungrouped data (N = 59); ED was significantly correlated with GW, GSI and GI (R = 0.50, 0.62 and 0.62, respectively). Monthly variations in ED were evident (p<0.0001) with similar variability in GSI and GI (Table 11). ED has its highest values in April and May and its lowest values in November and December (corresponding to highest fecundity). No association between ED and age groups (p<0.05) whereas ED showed variability among maturity stages (p<0.0001).
Fecundity (F) and relative fecundity with Respect to Length (RFSL), were significantly correlated with SL for grouped (R = 0.995) and 0.984, respectively) and ungrouped data (R = 0.653 and 0.450, respectively, N = 59).
| Table 11: | Monthly average with egg diameter of females of M. kannume from the Nile at Assiut, Egypt |
| Table 12: | Fecundity variation with Length groups of M. kannume from the Nile at Assiut, Egypt |
| Table 13: | Fecundity variation with weight groups of M. kannume from the Nile at Assiut, Egypt |
Fecundity, also showed significant association with W for grouped and ungrouped data (R = 0.958 and 0.579, respectively) whereas the Relative Fecundity (RFW) exhibited no significant correlation (R = -0.887 and -0.135 for grouped and ungrouped data, respectively). SL-F, SL-RFSL, W-F and W-RFW relationships were estimated for the grouped and ungrouped data as follow:
These findings referred to the correlation of fecundity with SL even after standardization but no significant correlation with W and RFW. For transformation, the relationship between SL and TL are given by the following Equations:
TL = 0.7392+ 1.092639 SL and SL= -0.19276 +0.091776
TL (N = 59) |
Fecundity and relative fecundity showed variability among length groups (Table 12), weight groups (Table 13) and months (Table 14). Absolute fecundity has its lowest values in June and July and its highest value in December (p<0.0001). There are no association between fecundity and maturity stages and fish condition factor (p<0.05). The fecundity increased with age groups and has its highest values in age group I and II and its lowest one in age VI (p<0.0001) (Table 15).
Stepwise multiple linear equation of fecundity: The relationship between absolute fecundity and SL, W, Age, GW, M, ED, GSI, Kn, Kc and GI were best described in a multivariate sense. The backward stepwise multiple linear regression referred to SL, ED and GI to be the best factors controlling the fecundity. So, fecundity and relative fecundity (per cm) can be estimated by the following equation (p<0.0001 for all parameters):
| F = -9699.3+514.7*SL-2.89*ED+174.3*GI (multiple R2
= 0.657) |
with partial correlations of 0.72, -0.49 and 0.69 for SL, ED and GI versus normal correlation of 0.59, -0.06 and 0.38, respectively.
| RFSL = -111.067+10.083*SL-0.098*ED+5.583*GI (multiple
R2 = 0.631) |
with partial correlations of 0.70, -0.57 and 0.74, respectively. The other variables have no significant contribution in these models and so are excluded.
| Table 14: | The relationship between the months and absolute and relative fecundities of M. kannume, from the Nile at Assiut, Egypt |
| Table 15: | Relationship between age and some characters of M. kannume from the Nile at Assiut, Egypt |
| Table 16: | Comparisons of fecundity and relative fecundity length and weight equation of M. kannume and caschive with Aly (1993) from Lake Nasser, Egypt |
| Table 17: | The relationship between average fecundity with length groups of M. kannume and M. caschive from different locality |
DISCUSSION
In the case of choosing a freshwater species for farming, sex ratio must be on the top of production considerations such as dividing resources between male and female reproduction in terms of sex allocation theory (Charnov, 1982). West et al. (2000) reported that Individuals facultatively adjust their offspring sex ratios in response to local conditions. Such sex ratios are generally dealt with reproduction studies for identifying the sexual behaviour of the fish understudy during different months of the year (Shenouda et al., 1994). The ideal value of sex ratio is 1: 1 (Hashem, 1972, 1973). But such value may vary yearly (Latif and Shenouda, 1973), seasonally (Shenouda, 1985; Downs and White, 1997), monthly (Nural-Amin et al., 2005; Laleye et al., 2006), or according to gear types, species of the same family (El-Shafey and Selim, 1998), length group (Dulcic and Kraljevic, 1996; Dulcic et al., 2003) and environmental factors such as temperature (Bohlen et al., 2008). In general, male may prevail in some populations (Mckenzie, 1974; Downs and White, 1997; Bohlen et al., 2008) or females in other ones (Lukens, 1978; Salem and Mohamed, 1982; Dulcic and Kraljevic, 1996; Dulcic et al., 2003; Nural-Amin et al., 2005; Ilkyaz et al., 2006). Sex ratios are balanced in other fish species (Hashem, 1972, 1973; Bohlen et al., 2008). In an adaptive hypothesis, Fisher (1930) suggested that equal sex ratios may be the results of natural selection when the sexes cost the same to produce resulting in a balance of investment between males and females.
Some Mormyrus populations from two different Egyptian localities, the Nile at Assiut (present study) and a lacustrine one, Lake Nasser (Aly, 1993) show variability in sex ratio to a certain degree. M. kannume population of Assiut exhibited overall sex not departed from the expected ratio 1:1 whereas those of Lake Nasser departed significantly; significant association was recorded between sex ratio and length groups in both populations. Such association was recorded between sex ratio and months for Lake Nasser population (p<0.05) but not for Assiut population. Sex ratio of Lake Nasser ranged from 1: 0.12 to 1:1.58 with departure from expected value (1:1), >48. Such ratio ranged from 1:0.45 to 1: 1.09 with departure from expected value in May-June and Sept.-Oct. M. caschive of Lake Nasser reflected similar variability (1:0.58-1:3) with departure from expectation in length group >48 and May-June only with significant association between sex and both length groups and months; no deviation in the overall sex ratio (1:0.99) from the ratio 1:1. These patterns of variations between Assiut and Lake Nasser population reflect time and locality factors and referred to the action of natural selection on the overall sex ratio.
The assessment of fecundity is the corner stone of the reproductive biology since it is not a stable character due to changes in enviromental conditions and species specific factors (Nikolsky, 1963; Khallaf and Authman, 1991; Shenouda et al., 1995). Sometimes, there are variability in fecundity for the same species in the some fisheries and in different years (Latif and Shenouda, 1973). Reproductive biology of M. kannume show variability under the spectrum of variations in the ecological and environmental condition of the aquatic ecosystem. Some authors (e.g., Ashour et al., 1990a) studied the reproductive cycle of that species in terms of the seasonal histological analysis of its ovary. Scott (1974) referred to M. kannume to be monoestrous. Okedi (1970), Scott (1974), Ashour et al. (1990b), Zaher et al. (1991) and Aly (1993) referred to variability of spawning season length of M. kannume. Okedi (1970) considered Mormyrids as anadromous spawners and the stimuli for annually twice spawning periods to be flood-associated. In Lake Victoria, ripe gonads of M. kannume in all months except July and August (Okedi, 1970). Ashour et al. (1990b) stated that spawning season of that species is fractional and extend for a long period ( from May to August with peaks in June and July) in the Nile at Beni suef, Egypt. Zaher et al. (1991) reported a spawning stage extending from May to July. Aly (1993) reported a summer spawning season for M. kannume and M. cashive (especially June-July) in Lake Nasser, Egypt. In the Nile at Assiut (present study), the spawning season of M. kannume extend allover the year as reflected by maturity stage distribution and gonadosomatic index with overall increasing magnitudes (Table 9).
Size at maturity (Lm) of M. kannume show variability among sex and localities. In Lake Nasser, Lm of that species was calculated by the author based on Aly (1993) data to be 46.6, 31.8 and 36.7 for males, females and combined sex, respectively. In the Nile at Assiut, Egypt (present study), Lm was 31.5, 31.7 and 31.6 for males, females and combined sex, respectively. Reviewing Lm recorded for different Mormyrus species in Fishbase web site data variability in size at first maturity was evident even for species of the same length range. Such variability in Lm may be affected by several physical and biological factors in time and fisheries in addition to the fishing efforts. Froese and Binohlan (2000) suggested that fish generally optimize their mean Lm to coincide with the length class of maximum fecundity. Binohlan (1998) also suggested that fishes of about 200 cm maximum size mature at about 100 m (i.e., reproductive load of Lm/L∞ to be 0.5). In the present work, the reproductive load was 0.748, 0.754 and 0.595 for males, females and combined sexes which is relatively higher. Such reproductive load is suggested to be determined by the species-specific gill surface area (Pauly, 1998). The low area explains the slow growth and is correlated with breathing difficulties especially in warm waters.
The relationship between fecundity, relative fecundity and fish size and weight of M. kannume and M. cashive according to the present author and Aly (1993) may be described by a power function or a linear regression (Table 16). Fecundity-age relationships of M. kannume were: F = 1561.9 A1.24 (R2 = 0.99) and F = 2065.7 A-728.9 (R2 = 0.95) from Lake Nasser and the Nile at Assiut.
Fecundity of M. kannume of the Nile at Assiut ranged between 648-19380 (6794±4339). The average means of fecundity within different length groups of Table 17 reflects the high fecundity of M. kannume at the Nile, in comparison with Lake Nasser. Aly (1993) recorded a fecundity mean of 1601 to 6363 corresponding to age groups of III to V, respectively in Lake Nasser in comparison to 2329 to 11530 for the groups I to VI, respectively for M. kannume of the Nile at Assiut.
The egg diameter of M. kannume exhibited monthly variability as well as variations with fish size and age (Tables 10, 11,15). The overall range of egg diameter of M. kannume of the Nile was 0.69 to 2.84 mm. El-Etriby (1985) reported a maximum egg diameter of 2.38 mm for the same species in Lake Nasser. Ashour et al. (1990a) and Zaher et al. (1991) refered to a maximum diameter of 1.03 mm with an average of 0.95 mm for mature egg of M. kannume from the Nile at Beni-Suef. Aly (1993) recorded a maximum egg diameter of 2.55 mm for M. kannume and of 2.46 mm for M. cashive from Lake Nasser. ED represents one of the important factors that influence fecundity. So, in multivariate sense, fecundity and relative fecundity are found to be affected by fish size, egg diameter and gonad index whereas other characters considered have insignificant effects on controlling fecundity of Mormyrus kannume; models are given for their future prediction.