Biodiversity conservation is becoming more challenging and imminent due to rapid habitat loss and fragmentation under ever growing global natural resource demanding. Habitat loss and fragmentation can lower migration rate between populations of a species, thereby reducing gene flow and genetic variability, leading to increased risk of extinction. Because of the relationship between genetic diversity and landscape characters, biodiversity conservation should involve the study of landscape characteristics and their changes. Thus, conservation efforts should not only focus on single species but also should consider all components of its habitats. We here discussed the relationship between landscape structure and genetic diversity, and outlined the scale, major concepts and techniques of landscape genetics. Landscape genetics is the interdisciplinary of population genetics, landscape ecology, and spatial statistics. It is used to quantify the effects of landscape characters on population genetic structures. Results from such studies may have great implications for biodiversity conservation and reserve management. There are five major research categories: (1) quantifying influence of landscape variables on genetic variation; (2) identifying barriers to gene flow; (3) identifying source-sink dynamics and movement corridors; (4) understanding the spatial and temporal scale of ecological processes; and (5) testing species-specific ecological hypotheses. Nowadays, landscape genetics is becoming a popular research area, because it opens the possibility to investigate ecological processes through genetic data and to analyses how these processes operate in the real world. Landscape genetics have heuristic, as well as practical, values in encouraging landscape ecologists to think more about biological processes rather than spatial patterns, and in encouraging population geneticists to consider the quality of a landscape instead of mere spatial distance. The use of molecular genetic is a new research method in testing landscape ecological hypotheses. It is new to use molecular genetic data to test specific landscape ecological hypotheses. This paper presented some suggestions for landscape ecologists in conducting their researches on landscape genetics: (1) do not confound neutral and adaptive genetic variation; (2) do not confound recent and historic gene flow; (3) test hypothesis and make it a truly landscape-related hypothesis; (4) check whether suitable genetic markers already exist or can easily be developed. The paper also explained to population geneticists the wealth of analyses that can be done with landscape ecological data, and then made these suggestions: (1) a landscape approach should go beyond testing the effect of distance; (2) disturbance and landscape change can be incorporated into the study design; (3) simulation model may help establish a mechanistic link; (4) the spatial and temporal variability of site conditions may be as important for explaining quantitative traits as are differences in their means. Under the influence of social and economic development, natural ecosystems are increasingly threatened by anthropogenic disturbances, such as habitat degradation, climatic changes, and invasive species etc. It is believed that landscape genetics will bridge researchers from micro- to macro-ecology, and it is conducive to articulation between academic arena and natural resource management community. The current focus of the research is landscape connectivity combining with genetic data. Interdisciplinary communication should be encouraged and facilitated in the study process.