5 General properties

   

Table 4: CO data from the literature.

ESO-LV	Sample	$I_{\rm CO(1-0)}$	References
name		K kms-1
1060120	CS	2.2	Combes et al. (1994)$^{\dagger}$
1570050	HDS	<1.2	Horellou & Booth (1997)$^{\dagger}$
3570190	HDS	<0.6	Horellou & Booth (1997)$^{\dagger}$
4780060	CS	$5.4\pm 1.8$	Andreani et al. (1995)$^{\dagger}$
4840250	CS	$3.5\pm 0.7$	Andreani et al. (1995)$^{\dagger}$
5450110	HDS	$12.2\pm 0.8$	Elfhag et al. (1996)$^\ddagger$
5480380	HDS	4.4	Combes et al. (1994)$^{\dagger}$

$^{\dagger}$ Using SEST; $^\ddagger$ using Onsala 20 m.

 

 

Table 5: Average values.

Sample	log $L_{\rm B}$	log $L_{\rm FIR}$	log $M_{\rm H_2}$	EW(H$\alpha$)$^{\dagger}$
	$L_{\odot}$	$L_{\odot}$	$M_{\odot}$	Å
HDS mean	9.94 $\pm$ 0.33	9.59 $\pm$ 0.40	8.86 $\pm$ 0.39	15.9 $\pm$ 11.3
HDS median	9.94 $\pm$ 0.12	9.65 $\pm$ 0.28	8.91 $\pm$ 0.40	14.2 $\pm$ 11.2
CS mean	10.08 $\pm$ 0.29	9.85 $\pm$ 0.35	9.18 $\pm$ 0.39	8.7 $\pm$ 3.4
CS median	10.11 $\pm$ 0.14	9.82 $\pm$ 0.21	9.29 $\pm$ 0.22	8.2 $\pm$ 3.3

$\parbox{14cm}{ $^\dagger$\space Without LINERs.}$

Figure 3: Left panel a): total molecular gas normalized by the blue luminosity as a function of blue luminosity. Right panel b): the same as in the left panel. Additional samples of ultraluminous infrared galaxies, galaxies in clusters, and spiral galaxies are included. Symbols are the same as in Fig. 1. Luminosity is in $L_{\odot}$ and mass is in $M_{\odot}$.

The FIR emission together with the molecular gas provide unique information in terms of fuel and star formation. The FIR luminosity was calculated using the relation (Lonsdale & Helou 1985)

$$\begin{eqnarray*}L_{\rm FIR} = 5.9\times 10^{5}D^{2}(2.58\times F_{60}+F_{100}) \end{eqnarray*}$$

where F₆₀ and F₁₀₀ are fluxes in Jy at 60 and 100 $\mu$m detected by IRAS and D is the distance in Mpc corrected for the Virgo infall.

H$_{\rm 2}$ masses were estimated from the velocity integrated CO(1-0) emission, using a $N_{\rm H_{2}}/I_{\rm CO}$ conversion ratio of $3\times 10^{20}$ cm^-2 (K kms^-1).

We are assuming that the conversion factor is the same in all galaxies in our sample. This assumption is reasonable since our sample do not contain any later-type systems (Sd, Sm, Ir) which, despite the ongoing star formation, show weak CO emission (e.g. Rubio et al. 1991).

Average and median values of $L_{\rm B}$, $L_{\rm FIR}$, $M_{\rm H_2}$, and H$\alpha$ equivalent width are presented in Table 5.

Figures 1a and 2a show the total amount of molecular gas as a function of FIR and blue luminosities. Figure 1a confirms the known correlation between $L_{\rm FIR}$ and the H$_{\rm 2}$ masses (correlation $\rm coefficient= 0.80$ and 0.84 for the HDS and CS, respectively). From Fig. 2a we verify that galaxies in the CS are on average more luminous than those in the HDS (a distance bias in our subsample). In order to eliminate this effect, CO intensities were normalized by the blue luminosity, $L_{\rm B}$, in the analysis presented in Paper II. Given our morphological selection criteria, we assumed that the mass/$L_{\rm B}$ ratio is approximately the same for our galaxies (Roberts & Haynes 1994) and $L_{\rm B}$ is thus a measure of the total mass.

We have plotted the $M_{\rm H_{2}}/L_{\rm B}$ as a function of $L_{\rm B}$ (Fig. 3a) in order to compare whether the bias in blue luminosity present in our subsample may cause a bias in our analysis. The correlation found for HDS and CS is very similar (correlation $\rm coefficient= -0.03$ and 0.06 for the HDS and CS, respectively) suggesting no evident bias. We have compared our sample properties with samples observed by others, such as normal spiral galaxies (Young et al. 1989; Braine et al. 1993), the ultraluminous FIR galaxies (Sanders et al. 1991), and galaxies in the Coma and Fornax clusters (Casoli et al. 1991; Horellou et al. 1995). As it is shown in Figs. 1b, 2b, and 3b the 47 spiral galaxies of our sample (HDS and CS) have correlations between global parameters which are similar to those in other samples. The ultraluminous FIR galaxies (Sanders & Mirabel 1996), as expected, are overall brighter and more massive than our subsample. The other samples include spirals of all types which explains the large dispersion found in luminosities and masses.

Figure 4: Left panel a): FIR luminosity normalized by the total molecular gas as a function of the FIR luminosity normalized by the blue luminosity. Right panel b): the same as in the left panel. Additional samples of ultraluminous infrared galaxies, galaxies in clusters, and spiral galaxies are included. Symbols are the same as in Fig. 1. Luminosity is in $L_{\odot}$ and mass is in $M_{\odot}$.

As previously mentioned, only intermediate Hubble types (Sb, Sbc, and Sc) were selected in order to avoid any bias due to the correlation between general properties and morphology. However, even in this sample the uncertainties in morphological classification should be taken into account when making any firm statement. Galaxies in dense environments can have their morphology distorted by tidal effects which makes them difficult to classify. One should refer to Appendix A (only available in electronic form) in order to visually check the morphology of each individual galaxy in more detail. We also refer to the detailed morphological classification taken from RC3 presented in Table 1 which gives a general idea on the complexity of the morphologies.

In Table 3 we give both the CO(1-0) and CO(2-1) integrated line intensitites. In order to estimate the CO(2-1)/CO(1-0) intensity ratios we need to convolve the CO(2-1) data to the same angular resolution as the CO(1-0) data. Since we observed only a single position for most galaxies, we can not do this. However, taking the values in Table 3 at face value, the average CO(1-0) to CO(2-1) line intensity ratio is $0.93\pm 0.47$. This is an upper limit to the line ratio. In the case of a molecular gas distribution more extended than both the CO(1-0) and CO(2-1) telescope beams (45'' and 23'', respectively), the correction for different angular resolutions would be 1.0. In the other extreme, with the CO emission originating in a point source, the correction for different angular resolutions would be 0.25. Since our telescope beam in almost all cases is large with respect to the optical extent of the galaxies, and since the molecular gas is likely to be centrally concentrated, the correction for different angular resolutions should be $\sim$0.5. Our average line ratio is thus ${\sim} 0.5 \pm 0.4$. This value is lower than that found by Braine et al. (1993) of $0.89\pm 0.34$ for normal spiral galaxies. The lower value is characteristic of optically thick and subthermally excited molecular gas and most likely reflects the lower star formation activity in our environmentally selected sample as opposed to far infrared bright selected samples.

In Fig. 4 we verify that the HDS and CS are also very similar to the galaxies in other samples in terms of SFE. We conclude that the intermediate type spirals in the HDS and CS do not belong to a separate class of objects but contain objects with properties similar to galaxies in clusters, nearby spiral galaxies and infrared luminous galaxies.