3 Light curves of individual stars

The adopted ephemeris elements of the infrared light curves for the program stars are listed in Table 3; they have been mainly taken from Catalano & Renson (1984, 1988, 1997); Catalano et al. (1991, 1993), and references therein. For each light curve a least square fit of all the data has been performed with a function of the type:

$$\begin{eqnarray} {\Delta}m & = & A_0 + A_1 \sin 2 \pi [(t-t_0)/P + \phi_1] \nonumber \\ & & \quad\ + A_2 \sin 2 \pi [2(t-t_0)/P + \phi_2].\end{eqnarray}$$ (1)

In this relation $\Delta m$ is the magnitude difference in each filter between the CP star and the comparison star, t is the JD date, t₀ is the adopted initial epoch, P is the period in days. This procedure can be partially accounted for by considering that within the accuracy of the measurements a sine wave and its first harmonic appear to be generally adequate to describe the light curves (North 1984; Mathys & Manfroid 1985) and the magnetic field variations (Borra & Landstreet 1980; Bohlender et al. 1993).

In the figures, where the infrared variations are plotted versus the ephemeris elements adopted for each program star (and summarised in Table 3), open squares represent the data collected in the 1986 run (i.e. the data with an accuracy lower than

  

Table 3: Ephemeris elements adopted to compute the phases of the variations, coefficients of the fits and error

0.01 mag), while the other symbols, listed in Table 1, represent the observations of the successive runs having an accuracy better than 0.01 mag. In the same figures the continuous line represents the fit to the observations obtained by means of Eq. (1): indeed this fit, whose coefficients and the sigma are presented in the last three columns of Table 3, has to be considered only as indicative, just to evidentiate the observed variations. It has to be noted here that, in spite of the fact that for some stars the infrared observations span long time intervals, the observed variations show lower amplitudes than in the visible and larger dispersions too, so that the selection between the aliases in the period values is not reliable and very often it does not remove at all the uncertainties. For these reasons we preferred to rest on the visible light curves, which generally show larger amplitudes and lower dispersion. However for most of the stars the individual fits of the J, H, and K variations did show very similar behaviour, and in many cases the first harmonic was sufficient to fit adequately the observations.

HD 10783 = GC 2141 = UZ Psc

The light variability of HD 10783 was detected by Van Genderen (1964) who preferred the value 4.1565 d of the period to the value 4.134 d derived by Steinitz (1964) from the analysis of Babcock's (1958) effective magnetic field data. UBV photometry and magnetic field measurements were carried out by Preston & Stepien (1968), who refined the period to the value 4.1327 d.

  

Figure 1: Infrared light curves of HD 10783. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

Additional photometric observations by Van Genderen (1967, 1971) have shown the photometric period to be consistent with the value determined by Preston & Stepien (1968).

Recent photometric UBV observations have been carried out by Hardie et al. (1990), who have derived the ephemeris elements given in Table 3, which have been adopted to plot our infrared observations in Fig. 1. From this figure a small variation (with an amplitude of the order of 0.02 mag) is evident in all three filters, although with quite a large dispersion.

HD 12447 = HR 596 = $\alpha^2$ Psc

HD 12447 is the brightest component of a visual binary system (ADS 1615) whose secondary (0.9 mag fainter) is only 3.6$^{\prime\prime}$ apart.

HD 12447 has been found variable in light by Winzer (1974) with a period of 0.7383 d, which later was inferred by Borra & Landstreet (1980) to be probably due to the variability of the comparison star HD 13467,

  

Figure 2: Infrared light curves of HD 12447. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

in fact Borra & Landstreet found their longitudinal magnetic field measurements to be compatible with period values such as: 0.5920, 0.5975, 0.6042, 0.7508, 1.45050, 1.49070, 1.5257, 2.7552, 2.8764, and 3.0395 d, and preferred the value 1.49070 d which gave the smallest $\chi^{2}/\nu$ ratio.

In spite of the fact that the infrared observations of HD 12447 span a longer time interval than the magnetic data, the small amplitudes and the quite large dispersion of the observed variations prevented any selection between the aliases in the period values. On the other hand the magnetic variation as observed by Borra & Landstreet is well defined, so that we adopted their period value.

Our infrared observations of HD 12447 are plotted in Fig. 2 versus the phase computed by means of Borra & Landstreet (1980) ephemeris elements listed in Table 3: the dispersion of the data is quite large and the period could also be somewhat uncertain. However the individual fits of the infrared variations appear better defined by a double harmonic in H and K.

HD 74521 = HR 3465 = 49 Cnc = BI Cnc

HD 74521 has been discovered to be a spectrum variable by Deutsch (1947). The first photometric observations were carried out in UBV by Stepien (1968) who derived a period of 5.43 d.

  

Figure 3: Infrared light curves of HD 74521. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

From subsequent UBV observations Winzer (1974) showed this period to be incorrect, and found the best representation of his data was obtained with a period of 4.2359 d. Later on Rakos & Fiedler (1978) found a nearby value of 4.239 d. From photometric observations in uvby and spectrophotometry Adelman & Pyper (1979) found the colour indices of HD 74521 to vary but they could not refine the period. On the basis of photometric observations in the Geneva system Lanz & Mathys (1991) found the most probable period to be 7.769 ($\pm$0.001) d.

The longitudinal magnetic field of HD 74521 was first measured by Babcock (1958), who found it variable in the range from -180 to +1450 gauss. Mathys (1991) has carried out further magnetic observations and by comparing his own data together with those by Babcock (1958) and by Bohlender & Landstreet (1991), found the best representation of the longitudinal magnetic field data to occur with periods of the order of 7.7730 or 1.1475 d (which is close to the 1 d^-1 alias of the former value). According to Mathys' analysis the longitudinal magnetic field varies in the range 500 to 800 gauss and does not show reversal of polarity.

From recent uvby observations Catalano & Leone (1993) got the best representation of all sets of data by means of the period value: 7.76851 ($\pm$0.00015) d, which also gives quite a good representation of the magnetic field data.

The infrared light curves of HD 74521 are plotted in Fig. 3 versus the phase computed by means of of Catalano & Leone (1993) ephemeris elements listed in Table 3. From this figure we see that the infrared variations are double-waved and all in phase with each other. The amplitude is almost the same ($\sim 0.02$ mag) in all filters. The behaviour of the light curves is also the same in all filters.

HD 90044 = HR 4082 = 25 Sex = SS Sex

The spectrum variability of HD 90044 has been discovered by Bonsack (1974), who found variations of the SrII and CaII lines with full observed range occurring on spectrograms taken two days apart.

  

Figure 4: Infrared light curves of HD 90044. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

The light variability of HD 90044 has been studied by Manfroid & Renson (1980, 1983) who found the period to be 4.37 d, value confirmed later by Mathys & Manfroid (1985).

Longitudinal magnetic field observations have been carried out by Bohlender et al. (1993) who considered the detection of a field fairly certain although they could not construct a phase diagramme because of the large uncertainty in the period.

Recent refinements of the period of HD 90044 have been performed by Catalano & Leone (1993) and Manfroid & Renson (1994) by using their own uvby observations together to the data collected in the Long-Term Photometry of Variables Project (Manfroid et al. 1991).

The infrared light curves of HD 90044 are plotted in Fig. 4 versus the phase computed on the basis of the ephemeris elements taken from Manfroid & Renson (1994) and listed in Table 3. From Fig. 4 we see that the infrared variations of HD 90044 are single-waved and all in phase with each other. The amplitude is almost the same ($\sim$0.03 mag) in all filters.

HD 116458 = HR 5049 = 67G. Mus

HD 116458 is a very interesting sharp-lined CP2 star: in fact it has one of the highest known absolute values of the $\lambda$5200 photometric index $\Delta a$ (0.054).

  

Figure 5: Infrared light curves of HD 116458. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

From the fact that very different spectral classification are available in the literature, Dworetsky et al. (1980) found entirely possible that this star is a fairly extreme spectrum variable.

Wood & Campusano (1975) measured a strong negative longitudinal magnetic field and supported evidence it be a spectroscopic binary with a period of 70.651 d. Albrecht et al. (1977) found the magnetic field do not differ much from a mean value of about - 1800 gauss except for two deviating values (out of twelve), and confirmed the spectroscopic binarity with the above mentioned value of the period. A preliminary orbit with a period of 70.651 d and a large eccentricity ($e \approx 0.4$) has been determined by Maitzen & Wood (1977) who also found HD 116458 to be constant in light and colour over a time scale of the order of 10 d, in accordance with the fact that longer periods tend to occur for larger $\lambda$5200 depression stars.

Mathys (1991) measured the longitudinal magnetic field of HD 116458 and supported evidence for this star having an essentially constant negative magnetic field of the order of - 2000 gauss, though low amplitude variations (of the order of ten percent) could be present in a long time scale.

More recently Hensberge (1993) inferred a period value of $147.9\pm0.6$ from light and spectrum variations. However by combining their uvby observations with those collected in the Long-Term Photometry of Variables Project (Manfroid et al. 1991); Catalano & Leone (1993) found all photometric data to be well represented with such a short period as 4.27349 ($\pm$0.00032) d.

The infrared light curves of HD 116458 are plotted in Fig. 5 versus the phase computed by means of Catalano & Leone (1993) ephemeris elements listed in Table 3. The infrared variations of HD 116458 have very small amplitudes, of the order of 0.01 mag and are all in phase with each other, but the period has to be confirmed.

HD 119419 = HR 5158 = V827 Cen

The light variations of HD 119419 have been first observed in the uvby photometric system by Manfroid & Renson (1980, 1983) who found a period of 2.605 d. Later Manfroid & Mathys (1985) confirmed this value of the period. By means of new measurements in the Geneva system Lanz & Mathys (1991) improved the period to the value: 2.6006 ($\pm$0.0003) d.

  

Figure 6: Infrared light curves of HD 119419. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

More recently Catalano & Leone (1993) got a period value of 2.60562($\pm$0.00008) d.

A strong polarity reversing longitudinal magnetic field has been discovered by Thompson et al. (1987). Indeed Mathys (1991) finds HD 119419 to be one of the few stars showing definite evidence of an anharmonic variation of the magnetic field. From line intensity measurements Mathys also finds the equivalent width of FeII$\lambda$5961 and SiII$\lambda$5978 lines to vary in phase with each other and nearly in anti-phase with respect to the magnetic field.

The infrared light curves of HD 119419 are plotted in Fig. 6 versus the phase computed by means of Catalano & Leone (1993) ephemeris elements listed in Table 3. The analysis of the infrared light curves shows that all of them are single-waved and in phase with each other, with amplitudes of the order of 0.02 mag.

HD 125630 = GC 19369 = BS Cir

The star HD 125630 has been discovered to be variable in light with a period P = 2.205 d by Manfroid & Renson (1980, 1983).

  

Figure 7: Infrared light curves of HD 125630. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

This value of the period was later on confirmed by Mathys & Manfroid (1985) who reanalysed the same set of data. Recent uvby observations have been carried out by Catalano & Leone (1993) leading to the improved period value: 2.20552 ($\pm$0.00006) d.

No spectroscopic study nor magnetic field determination are available for this star.

The infrared light curves of HD 125630 are plotted in Fig. 7 versus the phase computed by means of Catalano & Leone (1993) ephemeris elements listed in Table 3. As it is evident from Fig. 7, the infrared light curves of HD 125630 show a simple-waved behaviour, with quite large but different amplitudes amounting to 0.08 mag in J and H, and to 0.06 mag in K.

HD 147010 = GC 21960 = BD -19^o 4359 = V933 Sco

The star HD 147010 has been discovered to be spectrum variable by Kameswara Rao & Rajamohan (1982) who found drastic changes in the strength of the CrII $\lambda$4012.5 line and estimated the period to be of the order of 5.7 d.

The light variability of this star has been ascertained by North (1982, 1984) who determined the period to be 3.9210 ($\pm$0.0001) d from photometric observations in the Geneva system.

  

Figure 8: Infrared light curves of HD 147010. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

From observations in the v, b, and y filters of the Strömgren system Borra et al. (1985) found the period to be 3.99827 d. Recently, on the basis of several new measurements in the Geneva system, Lanz & Mathys (1991) have improved the period to the value 3.92076 ($\pm$0.00010) d. On the basis of recent uvby observations Catalano & Leone (1993) have refined the period to the value: 3.920676 ($\pm$0.000005) d.

A very strong longitudinal magnetic field of negative polarity was measured independently by Brown et al. (1981) and Glagolevskii et al. (1982). Subsequently Thompson et al. (1987) studied the magnetic field variation of HD 147010 and found the period to be consistent with the photometric one derived by North (1982) and Borra et al. (1985). Mathys (1991) has confirmed the quite sinusoidal character without polarity reversal of the longitudinal magnetic field variation, occurring with the same period as the photometric one, and extrema in the range - 2000 to - 5000 gauss, in partial agreement with the measurements of Thompson et al. (1987) which appear to be more negative by about 1200 gauss. Mathys also studied the equivalent width of the FeII $\lambda$5961 line and found large variations whose extrema coincide in phase with the light variations but do show no simple phase relation with the extrema of the magnetic field.

The infrared light curves of HD 147010 are shown in Fig. 8, where they are plotted versus the phase computed by means of Catalano & Leone (1993) ephemeris elements listed in Table 3. From Fig. 8 we see that all light variations are single-waved and in phase with each other. The amplitude is almost the same in all filters amounting to about 0.02 mag.

HD 166469 = HR 6802 = V4045 Sgr

The star HD 166469 has been discovered to be variable in light by Renson (1978), who derived a period of 2.90 ($\pm$0.04) d.

  

Figure 9: Infrared light curves of HD 166469. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

From careful reexamination of the same set of data Manfroid & Mathys (1985) derived a value of 2.8855 ($\pm$0.0008) d or other nearby values. From recent uvby observations carried out at ESO, Catalano & Leone (1993) have refined the period to the value: 2.88632 ($\pm$0.00015) d.

Nor spectroscopic neither magnetic studies are known for this star.

The infrared light curves of HD 166469 are plotted in Fig. 9 versus the phase computed by means of Catalano & Leone (1993) ephemeris elements listed in Table 3. From Fig. 9 we see that the infrared light variations are single-waved and in phase with each other with an almost constant amplitude of about 0.02 mag or slightly smaller.

HD 170397 = HR 6932 = V432 Sct

The star HD 170397 has been discovered to be variable in light by Renson (1978), who derived a period of 2.21 ($\pm$0.03) d.

  

Figure 10: Infrared light curves of HD 170397. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

From the same set of data Manfroid & Mathys (1985) derived a value of 2.1912 ($\pm$0.0005) d or other nearby values. Comparing their recent uvby observations with those of Manfroid & Mathys (1985); Catalano & Leone (1993) have refined the period to the value: 2.19133 ($\pm$0.00005) d.

The longitudinal magnetic field of HD 170397 has been measured and found to be variable by Borra & Landstreet (1980) who estimated the period to be 2.24 d. Some more magnetic field measurements by Bohlender & Landstreet (1991) and Mathys (1991) are well represented by the 2.1912 d period.

The infrared light curves of HD 170397 are plotted in Fig. 10 versus the phase computed by means of Catalano & Leone (1993) ephemeris elements given in Table 3. From Fig. 10 we see that all light variations are single-waved and in phase with each other and the amplitude is of about 0.01 mag in all filters.

HD 187473 = GC 27467 = V4064 Sgr

The variability of HD 187473 has been detected and studied by Hensberge et al. (1977, 1978), we found this star to vary with a period of 4.718 $\pm$ 0.001 d showing extra-ordinary light variability with such large amplitudes as $\Delta u \le 0.2$ mag, $\Delta v = 0.07$ mag, and $\Delta b = \Delta y = 0.1$ mag

  

Figure 11: Infrared light curves of HD 187473. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

However, in spite of such very large range of light variations, no spectrum nor magnetic field observations have been carried out yet.

The infrared light curves of HD 187473 are plotted in Fig. 11 versus the phase computed by means of Hensberge et al. (1978) ephemeris elements listed in Table 3. From Fig. 11 we see that the infrared light curves are in phase with each other and show the same double-wave behaviour. In spite of the few observations, it is to be noted the outstanding amplitude, of the order of 0.1 mag, the largest observed until now.

HD 223640 = HR 9031 = 108 Aqr = ET Aqr

The photometric variability of HD 223640 has been studied in the Strömgren system by Morrison & Wolff (1971) who found a period of 3.73 d. The light curves show quite the same behaviour in all filters with a fairly constant amplitude of about 0.06 mag peak to peak.

Megessier & Garnier (1972) contributed a few photometric data again in the Strömgren system and checked the spectroscopic variability;

  

Figure 12: Infrared light curves of HD 223640. The phases are computed according to the ephemeris elements given in Table 3. The solid line is a least-square fit of the observations by Eq. (1) as described in the text

these authors found the Ti and Sr lines strongly variable and the lines of Fe constant. The variation of Ti correlates with the photometric variation, Ti being strongest when the star is brighter. Later Megessier (1974, 1975) interpreted all the variations in terms of the oblique rotator taking also into account the sign changes of the longitudinal magnetic field as determined by Babcock (1958).

Recent photometric observations have been carried out by North & Burnet (1991), which have been supplemented with new magnetic field observations (North et al. 1992), and have led to an unambiguous and more precise value of the rotational period. In their detailed study of HD 223640 North et al. (1992) have confirmed the reversal of the effective magnetic field mentioned by Babcock (1958) and the nearly equatorial surface distribution of CrII and FeII, while TiII is mainly concentrated in patches at intermediate latitudes.

The infrared light curves of HD 223640 are plotted in Fig. 12 versus the phase computed by means of North et al. (1992) ephemeris elements listed in Table 3. From Fig. 12 we see that the light curves in all filters are in phase with each other and show the same simple-wave behaviour with amplitudes of the order of 0.02 mag but with a quite large dispersion.