Transcranial magnetic stimulation (TMS) induces phosphenes and disrupts visual perception when applied over the occipital pole. Both the underlying mechanisms and the brain structures involved are still unclear. In the first part of this study we show that the masking effect of TMS differs to masking by light in terms of the psychometric function. Here we investigate the emergence of phosphenes in relation to perimetric measurements. The coil positions were measured with a stereotactic positioning device, and stimulation sites were characterized in four subjects on the basis of individual retinotopic maps measured by with functional magnetic resonance imaging. Phosphene thresholds were found to lie a factor of 0.59 below the stimulation intensities required to induce visual masking. They covered the segments in the visual field where visual suppression occurred with higher stimulation intensity. Both phosphenes and transient scotomas were found in the lower visual field in the quadrant contralateral to the stimulated hemisphere. They could be evoked from a large area over the occipital pole. Phosphene contours and texture remained quite stable with different coil positions over one hemisphere and did not change with the retinotopy of the different visual areas on which the coil was focused. They cannot be related exclusively to a certain functionally defined visual area. It is most likely that both the optic radiation close to its termination in the dorsal parts of V1 and back-projecting fibers from V2 and V3 back to V1 generate phosphenes and scotomas.