Search

Severe droughts can lead to fires that cause massive tree mortality in even the wettest and most isolated Amazonian forests. After repeated fires, blackwater floodplain forests can remain in an open vegetation state of arrested succession that facilitates a transition towards a white-sand ‘savanna-like’ vegetation. These vegetation shifts, from closed-canopy floodplain forests to open fire scars and eventually white-sand savannas, may have profound implications for fish communities that depend on floodplain forests for food and recruitment. In turn, changes in fish communities may contribute to the arrested forest recovery after fires as fish are important seed dispersal agents for many floodplain tree species. To explore the impacts of floodplain forest fires on fish communities, we sampled fish in unburnt forests, fire scars, and white-sand savannas in the middle Rio Negro basin (Brazil) during two consecutive flooding seasons. We compared the abundance, species richness, and the taxonomic and trophic composition of fish assemblages across the three habitat types. We found significant shifts in fish assemblages in fire scars compared to unburnt floodplain forests. Also, as fire scars increased in size, total fish biomass decreased strongly. Fish communities in unburnt floodplain forests seem to be characterized by a higher proportion of smaller and omnivorous fish species than fish communities in burned floodplain forests and white sand savannas. Fish assemblages in fire scars and white-sand savanna were not significantly different. Synthesis and Applications. Amazonian fish communities change after floodplain forest fires. Unburnt forests have diverse fish communities, with a large proportion of unique and small omnivorous species. In contrast, carnivores and detritivores become more common in fire scars and white-sand savannas formed after forest fires. Less omnivore fish after forest fires can reduce tree seed dispersal and forest regeneration, affecting ecosystem functioning and the services provided by Amazonian forests. Fish are also a primary source of food and income for people living in Amazonian floodplains. Preventing forest fires is therefore of fundamental importance for the conservation of Amazonian aquatic and terrestrial ecological communities as well as for the livelihood of people.

Climate change is a major threat to biodiversity. Rainfall patterns over recent decades have changed and extreme events are intensifying. These are expected to affect the hydrological dynamics and structure of forest streams in the Amazon basin, which strongly depend on the local rainfall regime, with potential consequences for local fish assemblages. Here we used four repeated surveys of stream fish assemblages across 10 × 10 km area of a pristine forest reserve in central Amazonia over an 18‐year (2001–2018) period to investigate the temporal dynamics of fish taxonomic and functional composition in relation to changes in stream habitats and local climatic regime. We investigated changes in magnitude, direction, and congruence of fish assemblage trajectories over time. Total annual rainfall increased over time, with stormy days (i.e. daily rainfall above 20 mm) becoming more frequent during recent decades and a concomitant change of stream substrates from litter‐dominated to free sand. The overall taxonomic and functional composition of fish assemblages changed consistently over time, with stronger differences between the first and last surveys due to alterations in relative abundances of some of the most abundant species. Functional composition moved from open water species (compressed‐bodied species with less developed fins and terminal‐oriented mouths) to species adapted to lateral pools and shallow marginal areas (species with fusiform body shape, more developed fins, and superior‐oriented mouth) We provide evidence of a concerted directional change of fish assemblages associated with increased precipitation and a higher frequency of stormy days over the last 2 decades. Considering the short time interval and the subtleness of the climatic changes assessed in this study, as well as the historically assumed structural stability of Amazonian streams, these findings are surprising and indicate those stream fish assemblages may respond quickly to climate changes.

Aspidoras azaghal n. sp. was discovered during a multitaxonomic scientific expedition to the remote Amazon Terra do Meio region in tributaries to the rio Xingu basin, Para, Brazil. The new species can be promptly distinguished from its congeners by the following combination of features: (I) absence of the first dorsal-fin element; (II) parieto-supraoccipital fontanel located medially on bone; (III) absence of a longitudinal dark brown or black stripe along flank midline; (IV) ventral surface of trunk covered by clearly smaller, irregular and/or roundish platelets; (V) inner laminar expansion of infraorbital 1 well developed; (VI) relatively wide frontal bone, with width equal to half of entire length; (VII) absence of a thick, longitudinal conspicuous dark brown stripe along dorsal portion of flank; and (VIII) poorly developed serrations on posterior margin of the pectoral-fin spine. Besides morphological evidence, the molecular analyses indicated significant differences between the new species and its congeners, with A. albater and A. raimundi as its closest species, showing 6.53% of genetic differentiation in both cases. The intraspecific molecular data revealed gene flow (FST = 0.21641 for the COI marker, and FST = 0.53213 for the Control Region) between specimens from upstream and downstream a 30-meters height waterfall at the type-locality, which therefore represent a single population (P = 0.001 for COI, and P = 0.007 for the Control Region). Furthermore, it was possible to observe a unidirectional gene flow pattern, with genetic diversity increasing in the downstream direction.

The extent and intensity of impacts of multiple new dams in the Amazon basin on specific biological groups are potentially large, but still uncertain and need to be better understood. It is known that river disruption and regulation by dams may affect sediment supplies, river channel migration, floodplain dynamics, and, as a major adverse consequence, are likely to decrease or even suppress ecological connectivity among populations of aquatic organisms and organisms dependent upon seasonally flooded environments. This article complements our previous results by assessing the relationships between dams, our Dam Environmental Vulnerability Index (DEVI), and the biotic environments threatened by the effects of dams. Because of the cartographic representation of DEVI, it is a useful tool to compare the potential hydrophysical impacts of proposed dams in the Amazon basin with the spatial distribution of biological diversity. As the impact of Amazonian dams on the biota of both rivers and periodically flooded riparian environments is severe, DEVIs from different Amazonian tributary basins are contrasted with patterns of diversity and distribution of fish, flooded forest trees and bird species. There is a consistent relationship between higher DEVI values and the patterns of higher species richness and endemism in all three biological groups. An assessment of vulnerability at the scale of tributary basins, the assessment of biodiversity patterns related to DEVI, and the analysis of teleconnections at basin scale, demonstrate that recent construction of dams is affecting the biota of the Amazon basin. The evidence presented here predicts that, if currently planned dams are built without considering the balance between energy production and environmental conservation, their cumulative effects will increase drastically and represent a major threat to Amazonian biodiversity. Nanyang Technological University Published version This research was supported partially by the Council for Scientific and Technological Development-CNPq, CAPES Foundation, and the Earth Observatory of Singapore.

Resource partitioning is a stabilising mechanism known to maintain species diversity in a variety of environments. Assemblages of stream shrimp species are structured by habitat features and predation. Therefore, segregation along habitat dimensions could facilitate coexistence among species in shrimp assemblages even when segregation is a result of predation pressure by fish species. These ecological interactions take place on a background modulated by biogeographic features, such as connectivity among drainages. However, these generalisations are mainly based on studies undertaken in temperate regions. We investigated whether abundances of rainforest shrimp species are related to habitat dimensions, and whether habitat–abundance relationships might be mediated through fish‐assemblage structure and the effect of drainages on connectivity. We detected effects of habitat variation on densities of shrimp species, but the magnitudes of the effects were larger for some species than others. Fish‐assemblage composition also affected shrimp densities. Two of the three species of shrimp showed some degree of habitat specialisation, but only along current‐velocity, depth, and pH gradients. Habitat segregation among species occurred along the current‐velocity and pH gradients. Relationships between density and environmental gradients differed between catchments for only one species and only along the pH gradient. Our findings provide evidence that rainforest‐stream shrimp species respond differently to environmental gradients and this could facilitate coexistence among species. However, interactions with fish seem to have a stronger effect on species densities, and consequently species segregation, than direct effects of the environmental gradients, resulting in apparent competition for these resources.

Amphibious habits are present in a variety of fish and are allowed by a set of physiological, morphological and behavioural adaptations. The most intriguing question, however, is what motivates an aquatic organism to venture into the land? Unfavourable abiotic and biotic conditions in the aquatic environment have been reported as main stimuli for fish to emerge from water. However, few attempts have been made to disentangle the effects of different social contexts on the propensity of amphibious fishes to leave the water. Trying to separate the effects of predator threat, intraspecific competition and mate searching on overland movement of amphibious fish, we ran a mesocosm experiment with the Amazonian killifish Anablepsoides micropus. Overland movement was quantified based on fish leaps between artificial pools set in enclosures in a primary forest area. Males moved overland more frequently than females and their movement depended on the social context, being higher in the presence of predators and lower in the presence of females. Females, instead, showed context‐independent movement rates. This suggests that males move between pools searching for mates, stopping when successful in this search. On the other hand, occasional encounters with predators may force males to abandon the pool as a defence mechanism. Female aggressiveness towards the predator may influence its low movement in the presence of the predator. Such differential responses may play important ecological and evolutionary roles, allowing A. micropus to occupy a wide range of environmental conditions under varied ecological and social contexts.

International audience; Agricultural land use is a primary driver of environmental impacts on streams. However, the causal processes that shape these impacts operate through multiple pathways and at several spatial scales. This complexity undermines the development of more effective management approaches, and illustrates the need for more in-depth studies to assess the mechanisms that determine changes in stream biodiversity. Here we present results of the most comprehensive multi-scale assessment of the biological condition of streams in the Amazon to date, examining functional responses of fish assemblages to land use. We sampled fish assemblages from two large human-modified regions, and characterized stream conditions by physical habitat attributes and key landscape-change variables, including density of road crossings (i.e. riverscape fragmentation), deforestation, and agricultural intensification. Fish species were functionally characterized using ecomorphological traits describing feeding, locomotion, and habitat preferences, and these traits were used to derive indices that quantitatively describe the functional structure of the assemblages. Using structural equation modeling, we disentangled multiple drivers operating at different spatial scales, identifying causal pathways that significantly affect stream condition and the structure of the fish assemblages. Deforestation at catchment and riparian network scales altered the channel morphology and the stream bottom structure, changing the functional identity of assemblages. Local deforestation reduced the functional evenness of assemblages (i.e. increased dominance of specific trait combinations) mediated by expansion of aquatic vegetation cover. Riverscape fragmentation reduced functional richness, evenness and divergence, suggesting a trend toward functional homogenization and a reduced range of ecological niches within assemblages following the loss of regional connectivity. These results underscore the often-unrecognized importance of different land use changes, each of which can have marked effects on stream biodiversity. We draw on the relationships observed herein to suggest priorities for the improved management of stream systems in the multiple-use landscapes that predominate in human-modified tropical forests.

This study assessed the width of the trophic niche of four characid species (Bryconops giacopinii, Bryconops inpai, Hyphessobrycon aff. melazonatus and Iguanodectes geisleri) found under different co-occurrence circumstances in Amazonian upland streams. The study was conducted during the rainy season of 2011 at eight sites of two micro-basins of the Ducke Reserve, Manaus, Amazonas, Brazil. The four species were studied in the following circumstances: only one of the species occurring in the stream; two species co-occurring; three species co-occurring. The relative volume of the food items in the fish stomachs was used to calculate Hurlbert's trophic-niche breadth for the individuals of each species in the different co-occurrence circumstances. Hyphessobrycon aff. melazonatus changed their diet when occurring in syntopy with other characid species of similar feeding habits, as shown by a significant narrowing of its trophic niche. The opportunistic habits and great feeding flexibility of these characid species make the partitioning of food resources possible and act as an important ecological mechanism that facilitates the coexistence of different species, possibly by attenuating the effects of direct competition for food. In addition, the low carrying capacity of these upland forest streams may be an important environmental factor influencing the results of this study.

Teleocichla preta nov. sp. inhabits the rapids along the Rio Xingu and lower portion of the Rio Iriri. It is the largest species in the genus, reaching 121·3 mm standard length (LS ) while others do not reach more than 87·8 mm LS . Teleocichla preta is distinguished from all other species of Teleocichla by the unique blackish (in live specimens) or dark brown (preserved specimens) overall colouration of the body, which masks the faint vertical bars or zig-zag pattern of blotches on the flanks. Teleocichla preta also has a deeper body and a deep laterally compressed caudal peduncle, unlike any other congener, as well as a stout lower pharyngeal tooth plate bearing molariform teeth on its median area.

Climate change is poised to have fundamental impacts on the freshwater environments of the Amazon River Basin, and protected areas are routinely proposed as a possible management strategy to conserve freshwater fishes. However, there remains a paucity of information regarding the sensitivity of threatened fish species to climate‐induced changes in water quantity and quality. An expert‐based survey was used to address the following questions: (1) Are currently threatened fish species in Brazil also sensitive to projected effects of climate change? (2) Does the current conservation status of fish species also reflect their degree of sensitivity to climate change? and (3) What are the specific aspects of climate change that are likely to contribute the most to species sensitivity? Survey respondents evaluated 35 species (11 families) representing 50% of the threatened species in the Brazilian Amazon. The results suggest that the majority of threatened Brazilian fish species are considered highly sensitive to climate change impacts. Climate‐induced changes in water quality were, on average, considered a greater threat to species persistence than potential changes in water quantity. Survey results also suggest that fishes exhibit high sensitivity to changes in temperature and dissolved oxygen, and moderate to high sensitivity to changes in high‐flow (i.e. flood) and low‐flow (i.e. drought) regimes. A considerable mismatch was found between species conservation status and sensitivity to climate change, suggesting that perceptions of present‐day extinction risk do not necessarily provide insight into future risks associated with climate change. Species sensitivity to climate change showed no relationship to dispersal ability, indicating that protected areas may serve as important refugia for those species unable to keep pace with climate change. Despite this, the number and size of protected areas in Brazil have decreased over the past decade, largely to support the exploitation of hydropower and mining. Strategic conservation planning that involves existing and new protected areas for those species most at risk to climate change is warranted. Copyright © 2016 John Wiley & Sons, Ltd.

The temporal effect of discharge and limnology on fish composition and species diversity in a floodplain lake at the confluence of the Amazon and Negro Rivers was evaluated. Species richness, abundance and assemblage composition were strongly influenced by seasonal discharge of the Amazon and Negro Rivers, which affects lateral connectivity, water conductivity and temperature. As a consequence, temporal β-diversity was high in the lake and the assemblage was dominated by seasonally transient species. Relatively large species known to feed on resources within the floodplain were captured almost exclusively during the flood period. During the dry season, the assemblage was dominated by fishes adapted to harsh conditions of high temperature and low dissolved oxygen concentrations. An open system with high spatial and temporal heterogeneity created by the meeting of two large rivers with different water chemistry, Lago Catalao has a dynamic fish assemblage. Given its high temporal β-diversity and abundance of fishes, many of great importance in local fisheries, Lago Catalao and other floodplain lakes in this region merit special attention for conservation.

Monthly (April 2009 to May 2010) bottom-trawl sampling for Brachyplatystoma species along the rapids stretch of the Madeira River in Brazil revealed that Brachyplatystoma rousseauxii larvae and juveniles were present in low abundances in all areas and during all hydrological periods. The presence of larvae and juveniles throughout the hydrological cycle suggests asynchronous spawning in the headwaters of the Madeira River.

SUMMARY 1. The geographical ranges of species are influenced by three components: spatial distribution ofenvironmental conditions, biotic interactions and the dispersal capacity of species. The scarcity of dis-tributional records in vast regions such as the Amazon impedes understanding of fish distribution.Predictive distribution models have emerged as a better alternative to surpass this problem, but theabsence of large-scale maps for aquatic variables has been suggested as an important limitation.2. We aimed to evaluate the use of macroclimatic variables as surrogates for local limnological vari-ables in the Brazilian Amazon. Ordinary least squares model were used to predict the local habitatvariables from climatic and geomorphological information as macroscale variables. Models for sixstream-dwelling fish were built in MaxEnt and validated using area under curve and true skill statis-tics (TSS).3. All local variables were predicted successfully (R 2 > 0.39), and MaxEnt models had good suitabil-ity using the macroscale variables (TSS higher than 0.70). We conclude that macroscale variables canbe effective surrogates for local habitat variables, at least for large-scale analyses on poorly sampledregions such as the Brazilian Amazon.Keywords: Amazon, aquatic variables, macroscale variables, modelling species distribution, stream fish

Hypotheses about phylogenetic relationships among species allow inferences about the mechanisms that affect species coexistence. Nevertheless, most studies assume that phylogenetic patterns identified are stable over time. We used data on monthly samples of fish from a single lake over 10 years to show that the structure in phylogenetic assemblages varies over time and conclusions depend heavily on the time scale investigated. The data set was organized in guild structures and temporal scales (grouped at three temporal scales). Phylogenetic distance was measured as the mean pairwise distances (MPD) and as mean nearest-neighbor distance (MNTD). Both distances were based on counts of nodes. We compared the observed values of MPD and MNTD with values that were generated randomly using null model independent swap. A serial runs test was used to assess the temporal independence of indices over time. The phylogenetic pattern in the whole assemblage and the functional groups varied widely over time. Conclusions about phylogenetic clustering or dispersion depended on the temporal scales. Conclusions about the frequency with which biotic processes and environmental filters affect the local assembly do not depend only on taxonomic grouping and spatial scales. While these analyzes allow the assertion that all proposed patterns apply to the fish assemblages in the floodplain, the assessment of the relative importance of these processes, and how they vary depending on the temporal scale and functional group studied, cannot be determined with the effort commonly used. It appears that, at least in the system that we studied, the assemblages are forming and breaking continuously, resulting in various phylogeny-related structures that makes summarizing difficult. © 2014 The Authors. Ecology and Evolution published by John Wiley & Sons Ltd.

Summary Macroecological comparisons have shown that habitat characteristics have a major influence on the evolution of life-history strategies of fishes in large river systems. Specifically, habitat variability and predictability seem to forge the reproductive strategies of fishes into a trilateral continuum with three main types: seasonal, opportunistic and equilibrium. Despite the usefulness of such a broad categorisation, large-scale patterns of trait distribution should ultimately be influenced by selection at smaller (i.e. individual) spatial scales. However, the scarcity of studies developed at the organismal scale leaves gaps in our understanding of how habitat characteristics affect life history and behavioural tactics of stream fishes. We examined the fine-scale spatial distribution of fishes in pristine Amazonian headwater streams over a full seasonal cycle to test whether segregation into main channel and associated temporary pond habitats was predictably related to hydrological variation and species’ life-history traits. Most species could not be unambiguously assigned to one of the three end points of the trilateral continuum, but seemingly occupied instead the intermediate multivariate space between the end points. However, species were clearly ordered along a gradient of relative habitat use, ranging from strict channel dwellers throughout the hydrological cycle to pond specialists, with some species occupying the ponds at high water levels. Hydrological effects on habitat use were significantly related to species’ reproductive strategies, with the strongest responses found in species with large eggs and seasonally concentrated reproduction, which were most abundant in ponds at high water levels. We suggest that trophic limitations and hydrological characteristics of Amazonian headwater forest streams may severely limit the seasonal strategy, one of the end points of the trilateral model. The species-specific habitat segregation reported here adds to the list of mechanisms, such as miniaturisation and isolation by distance, previously postulated to contribute to high diversity of stream fish faunas in the Amazon basin.

The recovery of giant otter populations after the hunting prohibition and restriction of the pelt trade resulted in more frequent conflicts with fishermen. In this study, fisherman–giant otter conflicts were analyzed in the Uacari Sustainable Development Reserve, where giant otters are accused of interfering with fisheries by eating the fish (predation), frightening the fish away (local interference), and damaging fishing equipment (direct interference). Interference by predation was analyzed by evaluating overlap in fish species consumption between humans (measured by subsistence and commercial catches) and giant otters. The giant otter diet was assessed from fecal samples, and the human diet through questionnaires. Local and direct interferences were analyzed through fish samples using gillnets and comparing capture efficiency with and without giant otters’ presence. The overlap between human and giant otter diets was low (0.37), varied seasonally, and was smaller in the low water (0.24) than in the high water period (0.60), when both species tend to be more generalists. Overlap between fish species consumed by giant otters and those exploited by commercial fisheries was small (0.34). Giant otter presence during the experimental fishing was low (9.5%), restricted to the high water period, and did not significantly reduce the captures (U = 13, P = 0.61). The low overlap in diet may be a result of differences in preferences and fishing strategies. The conflict between giant otters and fishermen is greater in the high water period, when the income of the fisheries decreases; however, the conflict seems to be mainly motivated by the resident's prejudice against giant otters. Resumo A recuperacao de populacoes de ariranhas, apos a proibicao da caca e do comercio de peles, tem tornado mais constantes os conflitos com pescadores. Neste estudo analisou-se o conflito entre pescadores e ariranhas na Reserva de Desenvolvimento Sustentavel Uacari, onde as ariranhas sao acusadas de interferirem na pesca por comerem (predacao) e espantarem (interferencia local) os peixes, e danificarem apetrechos de pesca (interferencia direta). A interferencia por predacao foi analisada atraves da sobreposicao alimentar entre as especies de peixes consumidas por humanos (capturas comerciais e de subsistencia) e ariranhas. A dieta das ariranhas foi avaliada por amostras de fezes, e a dos humanos por entrevistas. As interferencias local e direta foram analisadas atraves de pescarias experimentais com malhadeiras, comparando-se a eficiencia de captura entre eventos com e sem a presenca de ariranhas. A sobreposicao entre as dietas de humanos e ariranhas foi baixa (0,37), variou temporalmente, e foi menor na seca (0,24) que na cheia (0,60), quando ambas as especies tendem a ser mais generalistas. A sobreposicao entre os peixes consumidos pelas ariranhas e aqueles explorados pela pesca comercial local foi pequena (0,34). A presenca de ariranhas durante as pescarias experimentais foi baixa (9,5%), so ocorreu na cheia, e nao interferiu significativamente na eficiencia das capturas (U = 13, P = 0,61). A baixa sobreposicao nas dietas pode ser resultado de diferencas nas preferencias alimentares e estrategias de pesca. O conflito entre ariranhas e pescadores e intensificado na cheia, quando o rendimento das pescarias diminui naturalmente, mas parece ser motivado principalmente pelo preconceito dos moradores com relacao as ariranhas.

Collecting freshwater zoological material for research purposes is common and necessary, but its impact on biological communities is often overlooked. We tested the impacts of a conventional ichthyological collection in small forest streams and showed that fish assemblages recover after a few months. We highlight ecological traits and habitat characteristics that might exacerbate the impacts of repeated removal of fish in long-term investigations, and advocate for the use of a capture-release approach whenever possible.

– Habitat connectivity is considered a central factor shaping ecological communities, and the effects of waterway barriers such as natural waterfalls on fish movements are expected to produce differing assemblage structures in riverine ecosystems. Here, we evaluate the influence of a sequence of waterfalls on the compositional dissimilarity of fish assemblages along the Madeira River, the largest tributary of the Amazon River. We found significant differences in species composition between rivers stretches located upstream and downstream of Teotonio waterfall and, to a less extent, Jirau waterfall, independently of the hydrological period. After accounting for the relative roles of local and regional factors in explaining fish compositional dissimilarity, we still observe a significant effect of the waterfalls. We conclude that these waterfalls act as natural ecological barriers limiting fish dispersal processes and discuss aspects of these ecological filters and the potential effects of two dams currently under construction in the Madeira River.

– We compared fish abundance, diversity and species composition between lakes open (fished) and closed (no-take) to fishing activities in Mamiraua Sustainable Development Reserve in the Central Brazilian Amazon, in order to investigate potential influences of the common-based management. We sampled 1483 fishes from 70 species through gillnet fishing during the low-water season, in seven fished and seven no-take lakes. Contrary to expected, the mean values for abundance, size, diversity and species-richness of fish did not differ between fished and no-take lakes. There was no difference between fished and no-take lakes considering only the abundance of the 14 fish species more intensely targeted by fishermen. However, the abundance of an important commercial fish, the tambaqui (Colossoma macropomum) was higher in no-take lakes. Such data from a rapid assessment may be useful to monitor this and other fishery co-management schemes.

SUMMARY 1. The effects of seasonal inundation on the biology of fishes on floodplains of large Amazonian rivers are well studied. However, the small seasonal changes in headwater streams are generally considered to have little effect on fish assemblages. 2. In this study, we analysed seasonal changes in the species composition and abundance of fish in small Amazonian forest streams. We sampled fish with hand and seine nets in headwater streams in a 10 000 ha terra-firme forest reserve near Manaus, Brazil. Each stream was surveyed at the end of the 2005 dry season, at the beginning of the 2006 rainy season and at the beginning of the 2006 dry season, by means of a standardized sampling effort. 3. The numbers of individuals and species caught were higher in the dry season, but rarefaction analyses indicated that greater species numbers could have been due simply to the larger number of individuals caught. 4. Between the dry and rainy season, the direction of changes in species composition in multivariate space varied among sites, especially for quantitative (abundance) data. However, the observed variation among sites was the less than expected if the directions of change were random. 5. Fish assemblages in the second dry season were more similar to those in the previous dry season than expected if changes in species composition among seasons were random. This indicates that a general seasonal pattern in fish assemblages can be detected, despite the existence of some erratic site-specific changes. 6. Most of the species that showed large seasonal variations in density occupy temporary ponds during the rainy season, when much of the valley is inundated and pond networks form adjacent to streams. Short-duration lateral migrations to these ponds may play an important role in the seasonal fish-assemblage dynamics in Amazonian headwater streams. 7. Our results contrast with previous studies on small Amazonian streams, which have found little seasonal change in fish assemblages, and highlight the importance of the abundance of common species as an indicator of general fish assemblage structure in biological monitoring programmes.

SUMMARY 1. The effect of habitat structural features and physicochemical characteristics of the water on the composition and richness of fish assemblages in temporary ponds near streams were examined at three spatial scales: among ponds, among streams and between drainage basins, in a ‘terra-firme’ (not subject to long-term flooding) forest reserve in Central Amazonia. 2. The fish assemblage in temporary ponds was composed of subsets of 18 small-bodied species widely distributed in the reserve. The assemblages had a nested subset structure, where smaller ponds contained subgroups of the species found in larger ponds. 3. Species composition and richness in temporary ponds were similar between drainage basins, although the fish assemblages in streams differed between basins. 4. Fish assemblage structure was influenced by local factors related to habitat structure, such as pond area and depth, canopy cover and hydroperiod. Physicochemical characteristics of the water in the ponds were similar between drainage basins and had little detectable effect on the structure of pond fish assemblages. 5. No correspondence was found between the composition, richness or abundance of fishes in the ponds and in stretches of the streams adjacent to the ponds. Therefore, it is not possible to predict the composition of these temporary pond fish assemblages from the fish assemblages found in adjacent streams.

The food habits of the giant otter, Pteronura brasiliensis, were determined by analyzing fecal samples collected year–round from communal latrines at Xixtiau Creek (00°48'S, 61°33'W), Jauaperi River (central Brazilian Amazon) between 1993 and 1994. Communal latrines were also sampled during the low water season at Aquidauana River (19°32'S, 56°40'W), Pantanal of Mato Grosso do Sul (central Brazil). Fragments of vertebrae, mandibles, teeth, scales, and fin spines provided an estimate of the diet composition. Remains of fish were present in all of the samples. The main fish groups found at Xixuau Creek were Perciformes, represented specifically by Cichlidae (97.3% of all samples), Characiformes (86.5%) and Siluriformes (5.4%). The Characiformes were represented mainly by E, rythrinidae (Hoplias sp. 90.6%), followed by Serrasalmidae (28%). The Anostomidae occurred with a frequency of 18.7 percent, while Characinae, Bryconinae, and Cynodontidae were only observed in 3.1 percent of the Characiformes records. At Aquidauana River, the Characiformes were the most frequent fish group, represented in 100 percent of all samples, followed by Siluriformes (66.6%) and Perciformes (33.3%). The fish from the main groups identified in the diet of the giant otters prefer riverbanks, lakes, and flooded forests. The feeding habits of Pteronura brasiliensis are probably influenced by the vulnerability and abundance of the prey consumed by this carnivore.

The metacommunity framework has greatly advanced our understanding about the importance of local and regional processes structuring ecological communities. However, information on how metacommunity structure and the relative strengths of their underlying mechanisms change through time is largely lacking. Dynamic systems that undergo environmental temporal changes and disturbances, such as floodplains, serve as natural laboratories to explore how their metacommunity structure change in time. Here we applied the Elements of Metacommunity Structure framework and variation partitioning analysis to assess how temporal changes in the local environmental factors and regional dispersal processes in the rain season influence a seasonal floodplain-fish metacommunity. Across four months, relevant environmental factors were measured across 21 patches where over 3500 individual fish were sampled. Connectivity was measured using landscape resistance-based metrics and additional spatial variation in metacommunity structure was assessed via spatial autocorrelation functions. The metacommunity structure changed from nestedness, at the beginning of the flood season, to a quasi-Clementsian gradient at the end. Our analyses show that connectivity is only important in the beginning of the flood season whereas environment is only important at the end. These results suggest that this metacommunity is structured by changes between dispersal limitation and environmental filtering through time.