The origin of reflections in the x-ray diffraction pattern from striated muscle and their use for understanding the structural organization of the contractile machinery are presented and discussed. Results of x-ray diffraction experiments obtained by a number of research groups using a variety of protocols revealed structural changes in contracting muscles which are interpreted in terms of molecular movements that underlie force generation. Some of these data are in line with the widely accepted 'lever arm' hypothesis which links force generation to a tilt of the light chain domain of the myosin head with respect to its motor domain. However, changes in the layer line intensities observed in response to various perturbations cannot be explained by tilting of the lever arm. Such changes, first revealed in response to temperature jumps, are interpreted as a transition of non-stereo-specifically attached myosin heads to a stereo-specifically bound state. The new 'roll and lock' model considers force-generation as a two-stage process: initial stereo-specific locking of myosin heads on actin is followed by the lever arm tilt.