We have studied 12 40 angstrom mm-1 spectra of HD 50896 taken between 1980 October 2 and 1982 October 20, and we are unable to demonstrate binary motion about a collapsed companion in a period of 3.763 day. The He II lines are displaced longward by approximately 134 km s-1, while N IV lambda-4058 is displaced shortward by about -85 km s-1 and N V lambda-4945 by -249 km s-1 from the estimated line-of-sight velocity of the star. We argue that we are seeing inflow from a disk in the case of the He lines and outflow in a wind in the case of the N IV and N V lines. The shapes of the line profiles change systematically in a period of 3.763 day. The star HD 50896 appears to be a single star, like HD 191765, WN6, which lies within a ringlike, rotating disk which is connected to the central star by a few ever-changing filaments which are supported by magnetic field lines. We doubt that the line profiles, which may change in 5 minutes according to Ebbets, give evidence of blobs moving in a dense wind or of the presence of a collapsed companion. The period of 3.763 day found by Firmani et al. may represent the rate of rotation of the disk.