The possibility to perform the frequency analysis of the light variation of V798 Cyg on the basis of a more accurate set of data and the Fourier decomposition of the new light curve of V831 Tau allowed us to improve the knowledge of the HADS. The following points are now better defined: 1) V798 Cyg is confirmed to be a double-mode pulsator and its similarity with V1719 Cyg definitely established: the peculiar light curve on f1 (descending branch steeper than the observed one) and the double-mode pulsation with ratio 0.800 have to be considered together in modelling these stars. 2) The detection of additional periodicities in light curves is confirmed to be a delicate matter. A thorough discussion would be needed to invalidate previous results estalished on the basis of better time--series, as otherwise confusing results are obtained, as in the case of V798 Cyg (Hintz and Joner 1997, PASP 109, 639) 3) The short-period variable V831 Tau shows a light curve very similar to other stars having the same period; the previously reported sinusoidal light curve is not confirmed. The light curves of HADS towards short periods have a very regular progression in the Fourier parameters space. 4) The gap around 0.25-0.30 in the R21 distribution of HADS light curves is confirmed. Hintz and Joner (1997) claimed that it can be filled by the values obtained from double-mode pulsators, but the available photometry does not support that claim which seems the result from improper handling of the data.

The photometric properties of the variable stars located in the lower part of the classical instability strip are discussed. The importance of the determination of some light curve parameters and their connection with the stellar models are stressed, with a particular emphasis on large amplitude Delta Sct (DSCT) stars.