The present study used event-related potentials (ERPs) to examine the time-course of location-specific and location- independent orthographic priming. In a masked sandwich priming experiment, changes in the relative positions of letters in prime and target stimuli combined with shifts of prime location relative to target location were manipulated. In particular, relative-position primes formed by concatenated subsets of the target stimuli (e.g., ‘grdn/GARDEN’) and absolute-position primes formed by hyphenated equivalent subsets (e.g., ‘g-rdn/ GARDEN’) were presented either centrally or displaced by two letter positions to the right or to the left (targets were always central). ERP waveforms were modulated starting at around 100 ms post-target onset and extending into the N400 component. Early priming effects were seen between 100-200 ms post-target onset, where priming effects were only apparent with centrally presented hyphenated primes. By 200-300 ms post-target onset, priming effects were present for both concatenated and hyphenated primes, with the latter still showing sensitivity to prime location. Finally, on N400 amplitude, both prime types revealed priming of similar size and scalp distribution independently of prime location. These results are consistent with an early activation of location-specific letter detectors, between 100 and 200 ms after stimulus onset, that subsequently map onto location-independent orthographic representations.