The discovery of a broad, $\sim$1.5$^{\circ}$ long filamentary [OIII] 5007 emission $\sim$1.2$^{\circ}$ south-east of the M31 nucleus has recently been reported. More than 100 hours of exposures of a wide field (3.48$^{\circ} \times 2.32^{\circ}$) have allowed this pioneering detection based on 30 \AA\ narrow-band filters and several small refractors equipped with large cameras. We report a first velocity measurement in this extensive [OIII] emission line region. We used the low-resolution spectrograph MISTRAL (R $\sim$ 750), a facility of the Haute-Provence Observatory 193 cm telescope. The velocity measurement is based on the H$\alpha$, [NII], [SII] and [OIII] lines. The best solution to fit the spectrum indicates that the H$\alpha$ and [OIII] emissions are at the same heliocentric line-of-sight velocity of -96$\pm$4 km s$^{-1}$. This was measured within an area of $\sim$250 arcsec$^2$ selected on a bright knot along the long filament of $\sim$1.5$^{\circ}$, together with a [OIII]5007 surface brightness of 4.2$\pm$2.1 10$^{-17}$ erg s$^{-1}$ cm$^{-2}$ arcsec$^{-2}$. This agrees moderately well with the previous measurement. We also estimated the H$\alpha$/[NII] line ratio as $\sim$1.1. The radial velocities at which the H$\alpha$ and [OIII] lines were detected seem to show that these hydrogen and oxygen atoms belong to the same layer, but we cannot exclude that another weaker [OIII] line, belonging to another structure, that is, at another velocity, is below our detection threshold. Different scenarios have been considered to explain this filamentary structure...
Comment: 8 pages, 5 figures, A&A letters, Accepted