The West-African neighbouring Gambia and Saloum estuaries function in different ways, the first being normal (saltier waters in the lower river, salinity range 0-38) and the second being inverse (saltier waters in the upper river, salinity range 36-90). The common cichlid species Sarotherodon melanotheron was collected in both systems over 16 months (June 2001 to September 2002) at five locations. We quantified reproductive traits from macroscopic examination of the gonads, oocyte counts and measurements, and growth from interpretation and measurements of an otolith sub-sample. There was a clear seasonal cycle of reproduction in the estuaries with a peak at the beginning of the wet season (May-July). Size at maturation was smaller in the more saline environment, the Saloum, for both females (170 mm in Gambia and 131 mm in Saloum) and males (162 and 113 mm respectively). The relative fecundity was lower and the oocyte size was bigger in the freshwater location of the Gambia river. Growth, estimated by counting the annual opaque zones on whole otoliths, was reduced in the hypersaline environment of the Saloum (L ∞ = 198 mm, K = 0.26). Differences are less obvious between the growth in the Gambia (L ∞ = 238 mm, K = 0.21) and that in the Saloum with a salinity of less than 60 (L ∞ = 244 mm, K = 0.22). All these observations suggest that S. melanotheron is able to withstand saltier environments by limiting its growth, reducing the size-at-maturity, and changing its fecundity, but the most profound changes are only visible in hypersaline conditions.

International audienceThe influence of salinity on life-history traits was tested using two adjoining West African estuaries: the Gambia with a 'normal' salinity gradient (salinity always <40 and decreasing from the mouth upstream) and the Sine Saloum (Senegal) with an inverse gradient (from 35 at the estuary mouth up to >130 in the upper reaches). The breeding seasons and subsequent fork length (LF) at first maturity (LF50) were estimated for different fish species reproducing both in the Sine Saloum and in the Gambia River estuaries using a database built from experimental fish samplings between 1990 and 2003 with a purse seine (30 locations in the Sine Saloum and 44 in the Gambia). The database contained 30 553 individuals belonging to 60 different species among which only 20 species reproduced in both estuaries. The breeding seasons peaked just before, or at the beginning of the rainy season (June to July), and there were almost no sexually mature fishes at the beginning of the dry season (November to December). Patterns of differences between LF50 of the two estuaries did not follow a general trend (positive or negative), but varied in the same way for females and males of a given species. The LF50 was only systematically reduced with increasing salinity in species living in high-salinity waters (>70). For species living below 70, differences in sizes at first maturity between the two estuaries did not show any clear relationship with salinity. The smallest mature individual found in an environment was a good indicator of the size at first maturity reached in a particular ecosystem because the trend of the species differences between the two ecosystems generally followed that of the differences in size at maturity

The common West African bonga shad represents a large part of the fish biomass in 2 neighbouring estuaries that function in different ways. The Gambia estuary has a normal salinity gradient, while the Saloum has an inverse gradient. Bonga shad Ethmalosa fimbriata were collected in both ecosystems during a 16 mo period (June 2001 to September 2002) at 5 locations, to investigate the role of salinity on life history traits. The main traits were studied at a spatio-temporal scale: reproduction from macroscopic examination of the gonads, oocyte counting and measuring, and growth from interpretation and measurements of a sub-sample of otoliths. Analysis of genetic differentiation at 3 intronic and 1 anonymous nuclear gene loci was also carried out to investigate differences between estuaries and among locations. The results did not show any allelic frequency heterogeneity between populations, indicating that populations of both estuaries represent 1 single panmictic unit, and that selection is not significantly acting on these loci. Hence, the response of the different traits to environmental variation may primarily represent phenotypic plasticity. The seasonal cycle of reproduction was clearer in the Saloum, occurring during a long period (January to August). The calculated size at maturity was reduced for both sexes in the upper Saloum, where the salinity was highest. The relative fecundity and the oocyte size were larger in the Saloum. On the otoliths, translucent zones, formed each year at the end of the rains (September to October), were used to estimate the age in months. Growth rates were reduced in the hypersaline environment of the Saloum, whereas growth differences were smaller between the Gambia and the pooled Saloum data, with a salinity < 60 psu. Growth was faster in the lower parts of the Saloum, related to better conditions for fish. The results illustrate that an environment with high salinity (> 60 psu) affects the growth, reduces the size-at-maturity and increases the fecundity of E. fimbriata.