Evidence of hourly variations in the deuterium Lyman line profiles toward ɛ Persei.

The idea that the observed deuterium abundance toward Epsilon Persei is caused by blending of the D I Lyman lines by H I lines due to high velocity gas varying with time is directly checked by separating the Copernicus data for each line in successive sets. Four possible explanations of the time fluctuations observed in the Lyman spectra are considered: statistical fluctuations, instrumental effects, the data correction process, and variations in the Lyman profiles. The first three of these are ruled out, and all the varying features are confirmed as H I components. The fact that these components vary within hours shows that they are located on the stellar wind and/or near the stellar surface. Profile fits show them to have a column density of a few 10 to the 14th per square cm.