3. HI properties of Kazarian galaxies

In Figs. 1a-d the histograms of the following derived parameters are shown: luminosity L, maximum of rotational velocity $V_{{\rm m}}$, neutral hydrogen mass $M_{{\rm H}}$ and indicative total mass $M_{{\rm i}}$. It must be noted that $V_{{\rm m}}$ values are available only for 14 galaxies of our sample (see Table 2 in Paper I).

Figure 2 displays the distribution of the relevant ratios $M_{{\rm H}}/L$, $M_{{\rm H}}/M_{{\rm i}}$, $M_{{\rm i}}/L$ as well as $\sigma_{{\rm h}}$, for our sample galaxies. The arrows in Fig. 2 indicate extremal values of the same parameters for classical galaxies which were taken from the earlier works of Balkowski (1973), Shostak (1978), Bottinelli et al. (1982) and more recent data given by Roberts & Haynes (1994). It must be noted that with the aim to compare the homogeneous data sets, throughout this paper we tried to use for comparison the data based on observations with Nançay radiotelescope when ever such data were available.

Figure 1: Distribution of the following parameters for Kazarian galaxies: luminosities, in $10^{9} L_{\odot }$ a); H I masses, in $10^{9} M_{\odot }$ b); maximum of rotational velocities, in ${\rm km~s}^{-1}$ c); indicative total masses, in $10^{9} M_{\odot }$ d)

An overview of data given in Table 1 shows that the hydrogen masses in Kazarian galaxies vary in the wide range from $0.3~10^{9} M_{\odot}$ to $29~10^{9} M_{\odot}$ while indicative total masses change from $5.4~10^{9} M_{\odot}$ to $619~10^{9} M_{\odot}$. Compared to this, the ratios $M_{{\rm H}}/L$, $M_{{\rm H}}/M_{{\rm i}}$, $M_{{\rm i}}/L$ and mean projected H I density $\sigma_{{\rm h}}$ occupy a somewhat smaller range.

Wide dispersion of these values shows clearly that Kazarian galaxies do not constitute a homogeneous sample by none of commonly used H I related parameters.

Figure 2: Distribution of the following parameters for Kazarian galaxies: HI mass to luminosity ratio, in solar units a); H I mass to indicative total mass ratio b); indicative total mass to luminosity ratio, in solar units c); H I mean projected density, in $10^{-3}~{\rm g~cm}^{-2}$ d)

Distributions presented in Fig. 2 show that the vast majority of our sample galaxies lie within the classical limits, and just a few of them are situated outside. Data from Table 1 indicate that these galaxies are Kaz 63 and 211 for $M_{{\rm H}}/L$ values; Kaz 3, 63, 65, 69, 211 and 579 for $M_{{\rm i}}/L$; Kaz 65, 69, 211 and 228 for $M_{{\rm H}}/M_{{\rm i}}$. Finally, Kaz 512 has a peculiar value of $\sigma_{{\rm h}}$. They are all listed in Table 2. It must be noted, however that Kaz 3 and 579 cannot be considered with certainty as peculiars due to their $V_{{\rm m}}$ and $M_{{\rm i}}$ data ambiguity (caused by small inclination) as well as Kaz 512 whose signal is poorly detected with S/N ratio less than 3 (see Paper I).

  

Table 2: Peculiar galaxies

Parameter	Kaz number
$M_{{\rm H}}/L$	63, 211
$M_{{\rm H}}/M_{{\rm i}}$	65, 69, 211, 228
$M_{{\rm i}}/L$	3, 63, 65, 69, 211, 579
$\sigma_{{\rm h}}$	512

It is well known that neutral hydrogen is rare in elliptical galaxies, and a lack of the meaningful average H I parameters do exist for these systems which in fact contain proportionately lower H I masses. Also, they demonstrate a much larger range in H I content and $\sigma_{{\rm h}}$ relative to the later spirals (Roberts & Haynes 1994). In this view, the location of Kaz 228 and 579 outside the usual limits shown in Fig. 2 must be taken with caution.

From the earlier studies it is also known that certain general relationships do exist between H I related parameters for normal galaxies. We used two well established correlations described by Shostak (1978) for unconfused and non-peculiar objects, namely linear diameter (A) versus corrected 21 cm line width ( $\Delta V_{0}$) and H I mass ( $M_{{\rm H}}$) versus linear diameter plots, in order to check whether or not Kazarian galaxies follow them.

These relationships are:

$$\log A = (1.67\pm0.39) \log \Delta V_{0} - 3.00$$

$$\log M_{{\rm H}} = (1.85\pm0.40) \log A + 7.22.$$

Taking into account the physical significance of the corrected profile width $\Delta V_{0} = \Delta V /\sin i$ and assuming purely circular motions we obtain $\Delta V_{0} \approx 2V_{{\rm m}}$ where $V_{{\rm m}}$ is maximum of rotational velocity whose values for 14 Kazarian galaxies are given in Paper I.

In Figs. 3a-b the mean regression lines are shown along with the points representing our sample galaxies.

Figure 3: Linear diameter versus maximum of rotational velocity a) and hydrogen mass versus linear diameter plots b) taken from Shostak (1978). Mean regressions are drawn by solid line (see formulae given in the text)

It is clearly seen that all Kazarian galaxies follow the trend indicated by the mean regression line in Figs. 3a and 3b. However, in Fig. 3a our sample appears to be located above the mean regression line; this is most probably due to the fact that our $V_{{\rm m}}$ values were derived by correcting the observed line widths for turbulent motions and thus are, on the mean, smaller than the non-corrected parameter used in Shostak's relation. In Fig. 3b all Kazarian galaxies are dispersed around the mean regression line and only one galaxy (Kaz 512) clearly deviates from that line. In the context of this paper it is more important to demonstrate the location of all Kazarian galaxies (which indicate they generally follow the normal galaxies) than to perform a quantative comparison since our sample galaxies do not comply with of the same restrictions applied by Shostak. Taking into account all of the aforementioned, the galaxies Kaz 3, 63, 65, 69, 211 and 512 can be considered as peculiars possessing H I related parameters outside the limits representing the classical objects. Some of these objects will be briefly commented on in Sect. 5.