The composition of fish communities in a certain area depends on the diverse environmental features of the habitats within that area. Anthropogenic changes to the environment lead to alterations in aquatic ecosystems that impact both the biodiversity of freshwater fish and the composition of their communities. Despite Croatia being one of the richest countries in Europe in terms of freshwater ichthyofauna, very little research focuses on fish ecology and the distribution of fish communities, and even fewer on the human impact on changes in fish communities and their distribution. To resolve this, a two-year research (2019-2020) was carried out to determine the most significant natural and anthropogenic habitat features that influence the composition of freshwater fish species within the Sava River basin of Croatia, and how these features affect the distribution of these species. Additionally, the occurence of two invasive fish species in relation to habitat features in the studied area was analyzed. In the Introduction, the question of the ever-changing habitats’ negative impact on the freshwater ichthyofauna was discussed. Croatia, specifically Danube basin, was discussed in a perspective of European freshwater fish diversity and the problem of the lack of specific ecological studies of this area was highlighted. History of the freshwater ichthyofauna ecology research and current knowledge on the matter was presented in the beginning of the Overview of the previous research chapter. The most important habitat features that influence the freshwater fish assemblages on the global scale were listed and discussed in detail. The list included presence of predators and competitors, elevation, water velocity, depth, bed supstrate, aquatic plants, water temperature, oxygen concentration and water quality (pH, conductivity). Second part of this chapter was focused on the current knowledge on the stressors which are impacting fish assemblages. Pollution, hydromorphological modifications (dams, weirs, channelization, embankment) and invasive species were covered. Also, “urban streams” as a result of several simoultaneous stressors were mentioned and discussed. Final part was specifically aimed towards the state of fish assemblages in Sava river basin in Croatia and related studies. Those studies were mostly inventory ones; some are covering impacts of pollutants on specific species and there are several regarding alien freshwater fish species. In this area, topmouth gudgeon (Pseudorasbora parva) and pumpkinseed (Lepomis gibbosus) were recognized as the most abundant invasive fish species from a List of invasive alien species of Union concern. Their habitat preferences, according to literature, were presented. In the Material and methods chapter the research period, area, and methodology was presented. Research was conducted from June 2019 to September 2020. Fish sampling using electrofishing was carried out at a total of 111 locations across 84 different watercourses for which there were insufficient available data on freshwater ichthyofauna. Simultaneously, basic physico-chemical water characteristics and morphological stream features were recorded. Biodiversity was calculated as Shannon-Wiener index and it was compared between different types of habitats with Mann-Whitney-U test. Impact of the habitat features and stressors on fish asssemblages was analised using the canonical correspondance analysis or redundancy analysis, depending on the structure of the dependant data. Bray-Curtis similarity was used to compare the fish assemblages on the same watercourses in different state. It was also used to generate similarity matrices in cluster analysis, analysis of similarities and similarity percentages. Cluster analysis was done on the complete set of data to divide them into interpretable subunits, regarding to the habitat features and stressors. Analysis of similarities and similarity percentages were used to determine the difference between fish assemblages of different types of watercourses (regarding their habitat features and present stressors). The probability of occurrence of invasive fish species was modelled using a generalized linear model. Statistical data processing was performed using the PAST, CANOCO, and R software packages. Results were presented in the Results chapter and compared to similar studies in the Discussion chapter, finally resulting with a set of conclusions. A total of 42 fish species were recorded, of which 20 rare species were excluded from further analysis. The research determined that the most significant factor influencing community distribution was the upstream-downstream gradient, encompassing stream elevation, streambed substrate composition, stream depth and width, and water temperature. Along this gradient, fish communities shifted from cold-water rheophilic specialists to warm-water generalists, including non-native species. Hydromorphological alterations to watercourses influenced by humans disrupted such natural continuity of community changes in certain watercourses. European chub (Squalius cephalus) and Danube gudgeon (Gobio obtusirostris) were the most numerous species in the study and had the greatest impact on the separation of fish communities in different habitat types. The probability of occurrence of invasive topmouth gudgeon was highest in rich communities of slow watercourses, while the probability of occurrence of pumpkinseed increased with stream depth. The results of this research contribute to a better understanding of the distribution of fish communities in the watercourses of the continental Croatia in relation to natural and altered habitat features. This study also describes the habitats in which the invasive alien fish species are most likely to occur and provides the basis for the future research.