4 Comparison with Plachinda's (#!Plachinda!#) results

Figure 3: Representative LSD mean Stokes I and V signatures of $\eta$ Aql from spectra obtained on 10 July 2002 (JD 2452101.480, phase 0.765). No significant circular polarisation is detected. Note that Stokes V has been enlarged by a factor of 500 and shifted upward by 1.1 for display purposes.

Many classes of radially-pulsating stars are known to exhibit atmospheric shock waves (e.g. Breitfellener & Gillet 1993; Chadid & Gillet 1996). Typically, two strong shocks are observed, the first around phase 0.9 (the primary shock originating as a result of $\kappa$-mechanism expansion) and the second around phase 0.65 (a weaker secondary "rebound'' shock). By convention, phase 0.0 represents the phase of maximum lumonosity. Plachinda (2000) observed an apparent rapid increase in the strength of the longitudinal magnetic field corresponding to the phase of the rebound shock, and developed a schematic model to explain this phenomenon.

We note a typo in the ephemeris as cited by Plachinda (2000). He uses the emphemeris of Kiss (1998), which should read ${\rm JD} = 2432926.749 + 7.176641\cdot E$. A more recent ephemeris is provided by Kiss & Vinkó (2000; ${\rm JD} = 2450100.861 + 7.176726 \cdot E$); this ephemeris is employed for all further analysis. Now, as indicated in Sect. 3, not one of the 14 LSD Stokes V profiles or longitudinal field measurements of $\eta$ Aql provides any strong evidence for a magnetic field. This is consistent with the results of Borra et al. (1981). These authors, using a measurement procedure conceptually similar to LSD, failed to detect a longitudinal magnetic field in $\eta$ Aql from 4 measurements with formal uncertainties in the range 8-23 G. To provide a direct comparison with the results of Plachinda (2000), we have phased our new data, as well as those of Borra et al. (1981), according to the pulsation ephermeris of Kiss & Vinkó (2000) and we show them, along with those of Plachinda, in Fig. 4. Clearly, these measurements are in strong conflict with those of Plachinda, most notably around phase 0.8.

We therefore conclude that neither our longitudinal field measurements nor our Stokes V spectra are consistent with the presence of the $\sim$100 G magnetic field reported by Plachinda (2000) to exist in the photosphere of $\eta$ Aql.

Figure 4: Comparison of our longitudinal magnetic field measurements (solid circles) of $\eta$ Aql, and those of Borra et al. (1981, solid triangles), with the field variation reported by Plachinda (2000, open circles and curve). Around phase 0.8, our results are strongly inconsistent with those of Plachinda.