1 Introduction

In the last decade, the problem of the effective temperature determination of the chemically peculiar (CP) stars was considered by different authors. Two recent papers by Lanz et al. (1993) and by Hauck & North (1993) contain the temperature calibrations of Geneva photometric indices for various kinds of CP stars. By CP stars, the authors mean the Am (CP1 stars) as well as the Ap stars, the latter including the He-wk (CP4 stars), MgMn (CP3 stars), Si, Cr and SrCrEu types (CP2 stars). These studies show once more the difficulty to derive photometrically the effective temperatures for most of the CP stars. Since the photometric methods are used, the color temperature is evidently derived and the color temperature is not always equal to the effective one. Therefore for any peculiar stars the $T_\mathrm{eff}$ determination from the photometry is much more complex than for normal stars owing to the fact that the colors of these stars can be affected by the blanketing effect.

In the literature there are a few determinations of $T_\mathrm{eff}$for CP stars independent of the photometry. The infrared flux method (IRFM) fist proposed by Blackwell & Shallis (1977) was applied to five CP stars by Shallis & Blackwell (1979) and to seven CP stars by Shallis et al. (1985). The method requires a measurement of the total integrated flux from the star at the Earth, an absolute flux at wavelength in the infrared region and the flux on the star's surface at the same wavelength. The latter flux is given by interpolation in Kurucz models. The ratio of the total integrated flux to the monochromatic one is used to determine the angular diameter and effective temperature simultaneously by means of an iterative procedure. This method is quite promising, because the difficulties resulting from blanketing could be discarded. Lanz (1985) and Glushneva (1987) determined the effective temperature for 28 Bp stars and for 13 peculiar B2 - F0 stars respectively by using IRFM. Glushneva found that effective temperatures obtained from I and J magnitudes were systematically lower up to 300 K than $T_\mathrm{eff}$ obtained from K magnitude. Megessier (1988) determined the effective temperatures for several standard stars in order to compare them with the fundamental values of Code et al. (1976) as well as for twelve CP2 stars by using the IRFM with integrated fluxes given by Lanz (1984). She got the closest agreement with Code et al. (1976) values, when restricting the method to J and K magnitudes. However, she found that the Hayes' (1979) calibration of the infrared photometry should be preferred.

Adelman (1985) estimated temperatures of 16 CP3 stars and 49 CP2 stars from fitting separately ultraviolet data, visual data and Balmer jump with the predictions of solar composition fully-line blanketed model atmospheres (Kurucz 1979). The comparison of the mean ultraviolet region temperatures shows substantial differences with visual color temperatures and Balmer jump temperatures. This may be due to the redistribution of the flux from the far-UV spectral region to the optical one.

Stepien & Dominiczak (1989) suggested an independent method of $T_\mathrm{eff}$ determination for chemically peculiar stars. This method assumes that the first order difference between normal and peculiar energy distributions is due to the extra blocking of the flux in the far-UV and the redistribution of it in the longer wavelengths. From the fit of observed visual energy distribution of CP stars with a solar-composition model, they derived the UV flux deficit relative to the model, and then the temperature correction to the model temperature.

The review of the various methods of $T_\mathrm{eff}$ determination for the CP stars shows once more the difficulty of deriving the effective temperatures for these stars. Since one use the methods taking into consideration the blanketing effect, the obtained temperature is close to the effective one. In the infrared flux method a monochromatic flux is taken in the infrared region to minimize a blanketing effect. The method proposed by Stepien & Dominiczak take into account the blanketing effect as well. The photometric methods may be useful, as it is possible to apply the correction of the color (or model) temperatures following the procedure proposed by Stepien & Dominiczak and to give relatively good estimates of $T_\mathrm{eff}$.Another way is to use the observational parameter which is not affected by peculiarities and can be applicable both to the "normal" main sequence stars and to CP stars. The Balmer continuum slope near the Balmer jump can be a useful tool for determination of $T_\mathrm{eff}$ for CP stars (Sokolov 1995).

In this paper the determination of the effective temperatures of CP stars using the Balmer continuum slope near the Balmer jump is discussed. This study is based on the ground-based observations of the stars measured by Adelman, Pyper, Shore and White (Adelman et al. 1989) and by Pulkovo astronomers (Alekseeva et al. 1996), as described in Sect. 2. The determination of $T_\mathrm{eff}$ from the Balmer continuum slope is described in Sect. 3. The principal results and discussion are presented in Sect. 4.