An early criticism of the experimental background of special relativity theory was that the Michelson-Morley experiment was bidirectional in the moving frame. This experiment did not require that the velocity of light is constant in all Galilean frames, but only that the average to-and-fro velocities are constant. These objections were removed with the experimental verification of the time dilation effect predicted by relativity theory. It is coneptually simple to devise a unidirectional ether drift experiment based on a measure of fringe shifts. The experiment could not be performed in Michelson's time because of the large coherence lengths required for the observation of the fringe shifts. For an interference arrangement similar to that used to produce Young's fringes, a coherence length greater than a meter is required in order to observe a displacement of 0.01 fringes. Such coherence lengths are available today from continuous wave lasers.