Crawling motion is characteristic of most animal cells and is principally based on actin-myosin II cytoskeletal system. The major components and reactions contributing to motility have been identified, but the overall picture of how these molecular events are orchestrated to result in the motion of the cell is still missing. We used a combination of experimental imaging, image analysis, and numerical modeling to analyze at a quantitative level how the elements of the actin-myosin machinery are coordinated in several important aspects of the crawling motion: the assembly of the characteristic pattern of actin network in the leading lamellipodium, the interaction between retrograde flow of the actin network and substrate adhesions, and the coordination of motion and assembly of actin and myosin II polymers over the entire cell during steady-state migration. First, we used enhanced phase contrast microscopy correlated with fluorescence and electron microscopy to investigate the supra-filament organization of actin in the lamellipodium of motile fish epidermal keratocytes. We showed a close correspondence between individual filament orientations and optical criss-cross network pattern by quantifying the orientational distribution of features in electron microscopy and optical microscopy images. We tested the hypothesis that the criss-cross pattern observed by optical microscopy results from variation of actin density, which is due to stochastic events of branching and capping of filaments. Based on this hypothesis, we produced simulated images of lamellipodium similar to experimental images. These simulations provided quantitative tools to estimate structural and dynamical parameters of the actin network in lamellipodia, notably the filament length, the actin concentration, the curvature of filaments and capping rate. Protrusion at the edge of the cell is coupled to the substrate adhesion, but at the same time is associated with the retrograde flow of the actin network away from the edge. We observed that focal adhesions were mostly localized at the boundary between the fast retrograde flow zone (the lamellipodium) and the slower flow zone (the lamellum). We analyzed simultaneously the actin flow and the formation of focal adhesion. Based on these results, we proposed a multi-stage model for the advance of the leading edge of migrating cell: advance of the lamellipodium is followed by the formation of the nascent adhesion complexes, which results in the reduction of the flow rate, advance of the lamellum, and then protrusion of the lamellipodium from a new base. We finally focused on the dynamics of the acto-myosin system in entire migrating cell. For this purpose, we developed a tracking program which allowed quantifying motion, contraction and assembly of cytoskeletal components. We measured the dynamics of actin, myosin II and their relative motion in migrating fish keratocytes. These results supported the dynamic network contraction model suggesting that the actin network deformed by myosin II drives the translocation of cell body. Focusing on the different aspects of the crawling motion, and balancing experimental and theoretical approaches, we obtained original results, which contributed to a better understanding of the dynamics of the lamellipodium and crawling motion in general.