Fish conservation in South America is a pressing issue. The biodiversity of fishes, just as with all other groups of plants and animals, is far from fully known. Continuing habitat loss may result in biodiversity losses before full species diversity is known. In this review, the main river basins of South America (Magdalena, Orinoco, Amazon and Paraná-La Plata system), together with key aquatic habitats (mangrove-fringed estuaries of the tropical humid, tropical semi-arid and subtropical regions) are analysed in terms of their characteristics and main concerns. Habitat loss was the main concern identified for all South American ecosystems. It may be caused by damming of rivers, deforestation, water pollution, mining, poor agricultural practice or inadequate management practice. Habitat loss has a direct consequence, which is a decrease in the availability of living resources, a serious social and economic issue, especially for South American nations which are all developing countries. The introduction of exotic species and overfishing were also identified as widespread across the continent and its main freshwater, coastal and marine ecosystems. Finally, suggestions are made to find ways to overcome these problems. The main suggestion is a change of paradigm and a new design for conservation actions, starting with integrated research and aiming at the co-ordinated and harmonized management of the main transboundary waters of the continent. The actions would be focused on habitat conservation and social rescue of the less well-off populations of indigenous and non-indigenous peoples. Energy and freshwater demands will also have to be rescaled in order to control habitat loss.

In channel and floodplain habitats of the Cinaruco River, Venezuela, Cichla temensis was more abundant and larger than C. intermedia and C. orinocensis. Seasonal variation in hydrology influenced habitat use, spawning, and predator-prey interactions. The three piscivores partitioned habitat, with C. intermedia showing a strong affinity for structured habitats in the main channel during all water level fluctuations. C. orinocensis was most abundant in shallow areas with submerged structure in lagoons and, to a lesser extent, in low velocity regions of the channel, and C. temensis occupied a wide range of lotic and lentic habitats. During the low-water period, the feeding frequency and body condition of all three species declined, and this was related, in part, to preparation for spawning near the end of the low-water season. The diet of C. intermedia was least similar to its two congeners during falling and rising water. C. orinocensis and C. temensis had lowest diet overlap during the low-water conditions, the period when many individuals of these two species move into lagoons for nesting. Prey in stomachs were significantly larger during the falling-water than the rising-water period, and predation by Cichla and other large piscivores during the falling-water period may have reduced the abundance of large prey, particularly Semaprochilodus kneri. These migratory detritivorous fish were important prey for C. temensis during the falling-water period and probably contributed a substantial fraction of the annual energy intake for this species. Together, the three Cichla species consume a wide spectrum of prey from a diverse fish assemblage, but prey are subdivided based on habitat, prey type, and season.

No abstract

Aim The aim of this study was to determine whether the Casiquiare River functions as a free dispersal corridor or as a partial barrier (i.e. filter) for the interchange of fish species of the Orinoco and Negro/Amazon basins using species assemblage patterns according to geographical location and environmental features.Location The Casiquiare, Upper Orinoco and Upper Negro rivers in southern Venezuela, South America.Methods Our study was based on an analysis of species presence/absence data and environmental information (11 habitat characteristics) collected by the authors and colleagues between the years 1984 and 1999. The data set consisted of 269 sampled sites and 452 fish species (> 50,000 specimens). A wide range of habitat types was included in the samples, and the collection sites were located at various points along the entire length of the Casiquiare main channel, at multiple sites on its tributary streams, as well as at various nearby sites outside the Casiquiare drainage, within the Upper Orinoco and Upper Rio Negro river systems. Most specimens and field data used in this analysis are archived in the Museo de Ciencias Naturales in Guanare, Venezuela. We performed canonical correspondence analysis (CCA) based on species presence/absence using two versions of the data set: one that eliminated sites having < 5 species and species occurring at < 5 sites; and another that eliminated sites having < 10 species and species occurring at < 10 sites. Cluster analysis was performed on sites based on species assemblage similarity, and a separate analysis was performed on species based on CCA loadings. ResultsThe CCA results for the two versions of the data set were qualitatively the same. The dominant environmental axis contrasted assemblages and sites associated with blackwater vs. clearwater conditions. Longitudinal position on the Casiquiare River was correlated (r 2 = 0.33) with CCA axis-1 scores, reflecting clearwater conditions nearer to its origin (bifurcation of the Orinoco) and blackwater conditions nearer to its mouth (junction with the Rio Negro). The second CCA axis was most strongly associated with habitat size and structural complexity. Species associations derived from the unweighted pair-group average clustering method and pair-wise squared Euclidean distances calculated from species loadings on CCA axes 1 and 2 showed seven ecological groupings. Cluster analysis of species assemblages according to watershed revealed a stronger influence of local environmental conditions than of geographical proximity.Main conclusions Fish assemblage composition is more consistently associated with local environmental conditions than with geographical position within the river drainages. Nonetheless, the results support the hypothesis that the mainstem

No abstract