“…Because of the limited number of observations available (e.g., only three H-band measurements of Aql), we used the shape of the radial velocity curve derived from the fit to the optical data (i.e., by using the same Fourier coefficients); the IR data were only used to determine an overall amplitude of the velocity curve. For the IR points we used an effective p-factor of 1:41 AE 0:03, as recommended by D. Sasselov (2001, private communication) and based on an analysis by Sabbey et al (1995), taking into account both the use of a constant p-factor and the use of parabolic line fitting. The resulting best-fit parameters are very similar to those based on optical radial velocities (i.e., Table 5 (top) and Gem (bottom) as a function of pulsational phase, together with a model based on radial velocity data, but fitting for distance, mean radius, and phase shift.…”