[en] The thermodynamics, kinetics, and computer simulations of crystallization and melting is discussed. The thermodynamics is shown to be well understood, although for many specific crystals not enough details for full description are available. Experiments on the crystallization kinetics of poly(ethylene) and poly(oxyethylene) in the presence of crystal nuclei as a function of molecular mass revealed that with increasing mass, the crystallization behavior deviates increasingly from that of small, rigid molecules. Instead of showing a continuously changing, linear crystallization rate with temperature through the equilibrium melting temperature, T_m⁰, these flexible macromolecules show a region of practically zero crystallization rate between T_m⁰ and about (T_m⁰ - 15) K, creating a temperature region of metastability in the melt that cannot be broken by nucleation with pregrown crystals. Molecular Nucleation was proposed as a cooperative process to be of overriding importance for the description of polymer crystallization, and to be at the center of segregation of molecules of lower molecular mass by growing crystal fronts. Initial efforts to model sufficiently large crystals using Monte Carlo and molecular dynamics methods are presented. Some of the short-time intermediates in the melting, crystallization, and annealing processes seem to have little similarity to commonly assumed models of crystallization and melting and are presented as discussion topics