This dissertation presents a sequential multiscale material modeling methodology and a multiscale field theory (MFT). In the sequential method, I developed a methodology to link atomic to nonlocal microcontinuum theories through phonon dispersion relations. The nonlocal constitutive relations for isotropic elastic solids are derived. The numerical algorithm to determine the material constants is presented. The material constants for silicon and diamond are determined by fitting the phonon dispersion relations obtained from atomistic calculations. Based on the field theory, numerical simulations are performed to investigate the material behaviors of bcc iron. The dislocation motion, phase transformation, size effects have been investigated. The mechanical wave propagation has been simulated. With the MFT, elastic properties have been obtained, which are in good agreement with experimental data or first principles calculations. Molecular dynamics (MD) simulations are performed to investigate the thermo-electromechanical response of ferroelectric perovskites. The microscopic paths by which homogeneous polarization switching process takes place in ferroelectric perovskites are characterized. The phenomena of hysteresis loop and the butterfly electric-field-strain curves are observed, the temperature dependent piezoelectric constants d₃₃ and d₃₁ are obtained. Meanwhile, I studied the polarization switching process from the MFT for ferroelectric nanoparticles. The MFT has also been applied to simulate layered superlattice structures. In the MFT, both MD and continuum modeling techniques can be utilized. For the purpose of verification, I created a multiscale model to simulate dynamic crack propagation, with regions away from the crack tip are analyzed using finite element analysis and the region near the crack tip is analyzed using atomistic simulations, the dynamic crack propagation process demonstrated that when the element size reduces to the size of a unit cell, the field theory is automatically reduced to an atomic theory. This thus naturally leads to a concurrent atomic/continuum model without the need for scale decoupling or a region of hand-shaking. As a result, the field-theory-based simulations are computationally much more efficient for statistical, finite size and finite temperature problems, for simultaneously large length/time scale phenomena, and especially, for dynamic, time-dependent and non-equilibrium phenomena and systems. This demonstrates the advantage and applicability of the field theory.