4 Discussion

Such a value for the distance to V705 Cas ($r= 1.67 \pm{0.34}$ kpc) is higher than the relevant values presented by Elkin (1995) and Lynch et al. (1997). In contrast, it is substantially lower than those derived by Hauschildt et al. (1994) on the basis of the UV spectroscopic observations with the IUE satellite and by Eyres et al. (1996) from radio observations. The method of the distance determination to a nova through the measurement of the equivalent widths of the Ca II K interstellar line is handicaped by the sensitivity of this particular line to the contamination by the circumstellar lines (Chochol et al. 1997). Thus, we suggest to view the distance determination of V705 Cas resulting from this approach (880 pc; Elkin 1995) as underestimated. For better understanding of the contamination by the circumstellar dust envelope, we give the following suggestion. Observed Ca II K interstellar line is created by two sources of absorption. One is connected to the dust envelope around the nova and the second one is produced by the interstellar medium (IM) itself. Resulting equivalent width of such profile doesn't reflect the true distance to the nova. Moreover the composite profile is produced not only by the absorption in the IM but by the emission from the dust envelope as well.

On the other hand, the CCD observations have revealed that V705 Cas is surrounded by an older shell (or shells), too (Boyle 1994). This shell is substantially larger than the shell ejected during the 1993 outburst. Therefore, it is disputable whether the direct radio images show only the latter shell. This fact indicates that the star may have undergone multiple explosions with mass ejection in the past. It is highly probable that the expanding envelope ejected by the nova in 1993 has been fragmented due to its collision with the material still residing in the immediate circumsystem vicinity as the relic of the previous outburst(s). For this reason, one can assume that the angular diameter of the outer border of the latter shell observed by Eyres et al. (1996) is smaller than it would be if the collision of the two structures would have not occurred. As a direct consequence of this, the distance to the nova should be lower than that value of 2.5 kpc which has resulted from radio observations. Thus, we suggest to consider this kind of distance determination as overestimated. Moreover, during the same outburst the ejection of matter could be characterized by different episodes of mass ejection. This yield higher unaccuracy in the distance determination from the radio method.

It should be noted here that the distance determination to a nova strongly depends on the adoption of the correct amount of the interstellar extinction. However, the relevant value could be overestimated just due to the continuing presence of the relic dust shell around the nova at the time of its new outburst. This is also probably the case for V705 Cas. Therefore, the distance of 1.67 kpc can be underestimated in part. For this reason, we have carried out the extinction determination more extensively, using several methods. We have obtained the lowest value of the colour excess E_B-V= 0.38 as the result of the intercomparison of the colour indexes of the stars surrounding the nova selected from the SAO catalogue. Nevertheless, the generally known great internal error of the photographic photometry does not secure the sufficient accuracy of this method. The corresponding distance is 2.16 kpc. Next method (Miroshnichenko 1988) gives the value E_B-V= 0.43 mag and the distance of 2.01 kpc. We can demonstrate similar scatter of the values of the colour excess also on the basis of papers by Hauschildt et al. (1994, 0.5 mag); Elkin (1995, 0.67 mag) and Lynch et al. (1997, even 1.09 mag). Especially the latter two values significantly differ from the ones published by the other authors and also from the values used throughout the present paper. The review of all colour excesses and corresponding distance values is presented in Table 2.

  

Table 2: Review of the dependence of distance on various adopted values of colour excess

It is also interesting to note the behaviour of the (B-V) index since shortly after the outburst until the dissipation of the dust shell. Immediately after the outburst, the value of the index was virtually constant (the values marked SP in Table 1) and the observed differences were within the measurement error. The index has only begun to change in relation to the formation of the dust shell. But what is especially interesting, its value was still increasing even during the time when the dust shell has already been so rarefied that it virtually did not manifest itself in the visual light curve anymore (Fig. 3). The further observational fact is discernible in the visual light curve at around $\rm JD= 2\, 450\, 250$ when there is a brightness decrease by about 1 magnitude. Through this the (B-V) index has probably returned to its original value and the physical influences within the shell deforming previously the overall light curve were definitely off. Although we have no UBV data from this period, we can speculate on the basis of the facts presented earlier here that after further years of the development of the relic shell its extinction influence will be of the order of 0.1 mag above the rough estimate. In such a case the presence of the shell is the cause of the underestimate of the distance by 0.4 kpc at most.

The problems with the determination of the correct colour excess of this nova can be caused by the overabundance of the CNO elements. It is probable that novae of the DQ Her subclass, to which V705 Cas unequivocally belongs, characterized by the deep minimum and the dust shell formation, are fundamentally different from the other types of classical novae. As the further support for this, we can point out also the colour index at maximum $(B{-}V)_{\rm 0,max}= 0.05$ mag we have determined for V705 Cas because its value significantly differs from the mean colour index of classical novae at maximum, 0.23 mag, derived by van den Bergh & Younger (1987).

Some results published in this paper have already been presented in preliminary form by one of us (K.P.) during the IAU Coll. No. 157 Cataclysmic Variables and Related Objects in Keele, UK, in 1995 (Petrík et al. 1996). The differences in the derived values of some nova parameters between that preliminary version and the present paper are caused by the use of the different statistical empirical relations.