4 Discussion

4.1 The HI properties of LSB Giants

Not surprisingly, all of the LSB Giant galaxies observed in the present program are found to be very H I-rich systems, having ${M}_{\rm HI}$ values ranging from $2.5\; 10^{9}$ ${M}_{\odot}$-7.1$\;10^{10}$ ${M}_{\odot}$(Table 4). The bulk of the sample galaxies are not only H I rich in an absolute sense, but are also H I-rich both for their Hubble types and for their optical luminosities - six of the galaxies in the present LSB Giant sample have $M_{\rm HI}/L_{B}\gtrsim$1 (in solar units; see Table 4). More typical values of $M_{\rm HI}/L_{B}$ for high surface brightness spirals are: $\sim$0.2 (type Sab-Sb); $\sim$0.3 (type Sbc-Sc); $\sim$0.4 (type Scd-Sd) (Roberts & Haynes 1994).

As expected for massive, H I-rich disks, broad H I linewidths are also hallmarks of LSB Giants. Because a number of our program galaxies have rather low inclinations ( $i\le 40^{\circ}$), it is difficult to estimate the true rotational velocities in some cases, but even many of the uncorrected linewidths tend to be quite broad (e.g., $W_{20,{\rm c}}=398$ kms^-1 for UGC 1752, which is seen nearly face-on). Only PGC 71626 seems to have a surprisingly narrow linewidth ( $W_{20,{\rm c}}=115$ kms^-1) given its inclination ( $i=61^{\circ}$) and Hubble type (SBb or SBc).

The majority of the LSB Giants in our study exhibit relatively normal, classic, double-horned rotational profiles (Fig. 1). However, we do see a few unusual cases as well; UGC 568 shows a peculiar, strongly asymmetric profile; PGC 45080 and NGC 7589 both have rather squarish global profiles with no clear rotation horns; and finally, F530-1 has a double-horned profile that is strongly ``cleft'' in the middle. PGC 135754 may also have a similar cleft morphology, but our detection of this source is relatively weak. Among the 11 relatively normal, double-horned profiles, 3 cases show modest asymmetries (UGC 2936, NGC 2770, and F568-6). Thus we find no evidence that strong H I asymmetries are a general feature of LSB Giants. We comment further on our individual spectra below.

4.2 Comparison of new HI data with previous measurements and comments on individual objects

In order to compare the H I parameters derived in the present work with past measurements, we have compiled global H I parameters from the literature for the 13 of our sample galaxies which have previously been observed in the 21-cm line. Table 5 summarizes previously measured recessional velocities, linewidths, and integrated line fluxes for each case, and indicates the telescope at which the data were obtained, as well as the reference for the quoted parameters.

An examination of Table 5 shows that for LSB Giants observed by more than one group, significant discrepancies frequently exist between the measured parameters, particularly the global integrated line fluxes. We discuss each of the 16 targets observed in the present work in more detail below, including comments on these discrepancies. As part of our investigation of our LSB Giant targets, we also searched the vicinity of each of the 16 objects (i.e., an area about 1.5 times the Nançay HPBW) for nearby galaxies which could possibly give rise to confusion in our observed H I profiles. For this we employed the NED and LEDA databases, as well as optical images extracted from the Digitized Sky Survey (DSS). Results of these searches are also described below.

UGC 568: No H I data have been previously published for this object. Our global spectrum appears quite peculiar, having a strong asymmetry, with a broad, single ``horn'' on the low velocity side of the profile. Published optical redshifts show a discrepancy of over $\rm 400~km\,s^{-1}$ ( $V_{\rm opt}=13\,600$ kms^-1, Melnick & Quintana 1981; $V_{\rm opt}=13\,198\pm 275$ kms^-1, Impey et al. 1996). Our new H I velocity corresponds to the mean of the optical redshifts. Several other galaxies were found within our search area, but they do not seem candidates for an H I confusion that could explain our peculiar global H I profile: CGCG 384-029 is a $B_{\rm T}=15.0$ mag Sbc spiral with an optical velocity 886 kms^-1 higher than the LSB Giant (Dale et al. 1998; Katgert et al. 1998), while PGC 3280 is an S0 with a recessional velocity 1524 kms^-1 lower than the LSB; MCGG +00-03-027 is a 15.5 mag galaxy without known velocity, but it is classified as S0/a (LEDA) or Sa (NED), and is therefore expected to be gas-poor.

UGC 1752: Our new total integrated H I flux ( $S_{\rm HI}=5.13$ Jykm s^-1) is somewhat larger than that measured at Arecibo by Giovanelli & Haynes (1989; $S_{\rm HI}=3.93$ Jykm s^-1), suggesting this galaxy has extended H I emission.

PGC 135657: No H I data have been previously published for this object. Our H I detection has a recessional velocity 407 kms^-1 larger than the optical velocity of $V_{\rm opt}= 12\,701\pm 275$ kms^-1 reported for PGC 135657 by Impey et al. (1996). In the present survey we detect additional marginal features at the edges of our bandpass, near $V_{\rm HI}=$12100 kms^-1 and $V_{\rm HI}=13\,300$ kms^-1. There are a number of small angular size galaxies in this field, but none appear as obvious candidates for interlopers.

UGC 2936: The Arecibo spectrum of UGC 2936 published by Sprayberry (1995b) is highly asymmetric; however only a modest asymmetry is seen in our new data and in the VLA spectrum of Pickering et al. (1999). Our new integrated line flux of 11.23 Jykm s^-1 agrees to within errors with that of Tifft & Cocke (1988) ($\sim11.7$ Jykm s^-1). In comparison, the Arecibo spectra by Hewitt et al. (1983), Haynes & Giovanelli (1984), and Sprayberry et al. (1995b) all yield considerable lower integrated line fluxes ($\sim8.4$-9.0 Jykm s^-1), while our new integrated flux value is somewhat lower than the Arecibo measurement of Haynes et al. (1999; 13.29 Jykm s^-1) and the VLA measurement of Pickering et al. (1999; 13.6 Jykm s^-1).

CO(1-0) and CO(2-1) line emission were detected from UGC 2936 at SEST and IRAM by Chini et al. (1996), making this one of the rare galaxies classified as an LSB Giant that has been detected in CO (see also Knezek 1993, 1998). However, Schmelz et al. (1986) failed to detect OH line absorption from the object.

UGC 3140 (= NGC 1642): This galaxy forms a wide pair ( $8'\,\hspace{-1.7mm}.\hspace{.0mm}3$ projected separation; $\Delta V_{\rm opt}=332\pm 85$ kms^-1) with 15.0 mag Sb spiral UGC 3141. Their E-W separation is $3'\,\hspace{-1.7mm}.\hspace{.0mm}3$ and our Nançay spectrum does not appear to be confused. Our new linewidth measurements are in good agreement with the Jodrell Bank measurements of Staveley-Smith & Davies (1988) and measurements we performed from new HIPASS data (see Barnes et al., in prep. for a description of HIPASS), but are lower than the values published by Bottinelli et al. (1982). No published H Iobservations are available for UGC 3141.

NGC 2770: Short Westerbork radio synthesis 21-cm line observations (Rhee & van Albada 1996; Broeils & Rhee 1997) indicate that the H I distribution of NGC 2770 at a surface density level of 1 ${M}_{\odot}$ pc^-2 is about 1.7 times as large as the optical (D₂₅) diameter and that its mean H I surface density averaged over the entire H I disk < $\sigma_{\rm HI}$> is 4.5 ${M}_{\odot}$ pc^-2, which is higher than typical values for an Sc. Our new spectra show a slightly higher peak flux density on the low-velocity side of the global line profile, hinting at a possible weak asymmetry not seen in the Westerbork observations.

PGC 135754: No H I data have been previously published for this object and it was only weakly detected in the present study. Our new H I velocity is 434 kms^-1 higher than the optical value reported by Impey et al. (1996; $21335\pm 275$ kms^-1).

F568-6 (= Malin-2 = PGC 86622): H I flux appears to have been missed in the Arecibo observations of F568-6 by Schombert et al. (1992). Our new value of $S_{\rm HI}=4.20$ Jykm s^-1 shows good agreement with the value derived from the VLA observations of Pickering et al. (1997; 4.4 Jykm s^-1).

UGC 6614: Our new integrated H I flux of 15.94 Jykm s^-1 is significantly larger than previous measurements from Arecibo (Bothun et al. 1985a; Haynes & Giovanelli 1989). However, Pickering et al. (1997) measured $S_{\rm HI}=15.0$ Jykm s^-1 from VLA data, consistent with our results. Our H I diameter estimate for this object is very poorly constrained, as we find a Gaussian H I model gives a poor fit to the data. This is reaffirmed by the VLA data of Pickering et al. (1997), which show a significant central depression in the H I distribution of this galaxy.

Malin-1 (= PGC 42102): One published Arecibo observation of Malin-1 (see Table 5) gives an integrated H I line flux (2.5 Jykm s^-1; Bothun et al. 1987) comparable to the VLA value derived by Pickering et al. (1997), while a re-reduction of the same Arecibo data yielded $S_{\rm HI}=3.5$ Jy kms^-1 (Impey & Bothun 1989); finally Impey & Bothun (1989) measure $S_{\rm HI}=4.6$ Jykm s^-1 from a Green Bank 43 m spectrum. Our new data indicate that the Arecibo and Green Bank fluxes previously reported by Impey & Bothun (1989) are likely to be significantly overestimated. In contrast our new integrated flux (1.80 Jykm s^-1) is somewhat lower than the VLA measurement of Pickering et al. (1997), although our values marginally agree to within errors, since at the redshifted frequency of Malin-1 ($\sim1312$ MHz), our absolute calibration uncertainties are expected to be as high as $\sim\pm$25% (see Sect. 3.2). In terms of spectral morphology, our new global spectrum shows excellent agreement with that derived by Pickering et al.

No CO(1-0) line emission was detected in the central regions of Malin-1 by Radford (1992), with an estimated upper limit of 4.7$\, 10^9$ ${M}_{\odot}$ to the H₂ mass, and Braine et al. (2000) recently reported an upper limit of $M_{\rm H2}/M_{\rm HI}<0.03$.

PGC 45080: Our new integrated line flux is smaller than the values reported by Sprayberry et al. (1995b), and our new spectrum lacks the distinct rotation horns seen in the Sprayberry et al. data. Given the small optical size of this object (D₂₅=  $0'\,\hspace{-1.7mm}.\hspace{.0mm}79$) it seems unlikely significant flux extended outside our beam. There are also no obvious interlopers in this field that would have caused confusion in the Sprayberry et al. spectrum, hence the discrepencies in the two spectra are probably due to the lower signal-to-noise of our present data.

UGC 9024: Our new integrated flux value of 11.41 Jykm s^-1 is higher than past values measured at Arecibo or with single-pointing Nançay measurements (Bothun et al. 1985; Lewis 1987; Haynes et al. 1997; Theureau et al. 1998), suggesting this galaxy has extended H I emission.

F530-1 (= PGC 87136): We detect only a slightly higher integrated flux (2.07 Jykm s^-1) than the Arecibo value of Schombert et al. (1992; 1.7 Jykm s^-1). Our spectrum shows an unusual ``cleft'' morphology, although our signal-to-noise is only modest due to the broad linewidth of this source.

F533-3 (= PGC 68495): No flux is clearly detected at 2'E in our mapping observations of this galaxy; this may be due to the broadness and relative weakness of the line profile and/or an H I distribution that is very lopsided relative to the optical center of the galaxy. As a result, our H I diameter estimate for F533-3 very uncertain.

NGC 7589: NGC 7589 was mapped in H I at the VLA by Pickering et al. (1997). Our new H I profile hints at a possible small amount of additional flux on the low-velocity side of the profile compared with the VLA global spectrum; this is probably caused by confusion with F893-29, which was also detected in H I at the VLA. F893-29 is a 17.1 mag Sb? spiral (NED) or dwarf (Pickering et al. 1997), $3'\,\hspace{-1.7mm}.\hspace{.0mm}3$ SW of the LSB Giant, with a E-W separation of $2'\,\hspace{-1.7mm}.\hspace{.0mm}2$. F893-29 has a systemic H I velocity of 8768 kms^-1 (Pickering et al. 1997), consistent with the velocity of the ``extra'' flux we observe associated with our global profile of NGC 7589. Our new integrated line flux (2.96 Jykm s^-1) agrees well with that reported by Pickering et al (1997; 2.7 Jykm s^-1), hence the flux contamination from this second source appears to be minimal. However our line profile widths appear to be overestimated by $\sim$100 kms^-1 due to this second source, hence corrections have been applied to the W₂₀and W₅₀ measurements in Tables 2 and 4.

PGC 71626: PGC 71626 was detected at a recessional velocity consistent with that recently reported by Theureau et al. (1998), but somewhat offset from the published optical velocity of 9520 kms^-1 reported by Sprayberry et al. (1995b) and Impey et al. (1996). Our new integrated flux value of 3.94 Jykm s^-1 is higher than the value of 2.6 Jykm s^-1 reported by Theureau et al. (1998) based on a single-pointing Nançay observation. An additional H I source was detected in our spectrum at $V_{\rm HI}=9753\pm4$ kms^-1, with W₂₀=73 kms^-1 and $S_{\rm HI}=1.52\pm 0.48$ Jykm s^-1. This may be contamination from the S? galaxy MCG -01-59-026 (of unknown redshift), at a projected distance of $4'\,\hspace{-1.7mm}.\hspace{.0mm}5$ from PGC 71626, or due to an uncatalogued LSB dwarf visible on the DSS.

4.2.1 Optical parameters for the LSB Giants

All of the LSB Giant galaxies observed in the present survey have optical surface photometry available in the literature. To offer a more complete picture of the nature of the objects in our sample, we have compiled a summary of several optical and photometric parameters for each galaxy in Table 6.

Radial velocities quoted in Table 6 are heliocentric values derived from optical measurements. $B_{\rm tot}$ values are the total B-band magnitudes derived from extrapolated exponential disk fits unless otherwise noted. $\mu(0)$ is the extrapolated central disk surface brightness in the B-band (unless otherwise specified) and is uncorrected for inclination and internal extinction. $h_{\rm r}$ is the exponential disk scale length in kpc. M is the absolute magnitude (B-band, unless otherwise noted) derived from the total magnitude in Col. 3, and assuming $D=V_{\rm opt}/H_{0}$ if no distance was given in the original reference; no internal extinction corrections were applied. Corrections to the values in Table 6 for Galactic extinction, k-corrections, and cosmological surface brightness dimming were generally applied by the original authors.

The values given in Table 6 show that there is a great deal of scatter in the photometric parameters reported for LSB Giant galaxies. Surface photometry of LSB galaxies is always challenging, but it is further complicated for LSB Giants by the fact that many of these galaxies have significant bulges, leading to ambiguity in the bulge-disk decompositions and hence vastly different derived scale lengths and central surface brightness values from different workers. Offsets between recessional velocities derived via optical and H I observations are also common (cf. Table 5).

We include Table 6 partly as a caution to readers interested in understanding the global properties of Giant LSB galaxies that there still exist large uncertainties in the measured parameters for such objects, and to highlight the need for additional surface photometry (and H I measurements) of such galaxies. Such data will be necessary before we can distinguish whether all LSB Giants such as the ones in the present study form a distinct class of object, as suggested by Hoffman et al. (1992), or whether some or most of these objects may form a natural and continuous extension of other low and moderate surface brightness disk galaxies. Well-established photometric parameters will also be key to placing the LSB Giant galaxies on the Tully-Fisher diagram.

Acknowledgements

LDM acknowledges the financial support of a Jansky Postdoctoral Fellowship from the National Radio Astronomy Observatory (NRAO), which is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. We have made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the U.S. National Aeronautics and Space Administration, as well as the Lyon-Meudon Extragalactic Database (LEDA) supplied by the LEDA team at the CRAL-Observatoire de Lyon (France), and the Digitized Sky Surveys (DSS), which were produced at the Space Telescope Science Institute under U.S. Government grant NAG W-2166. The Unité Scientifique Nançay of the Observatoire de Paris is associated as Unité de Service et de Recherche USR No. B704 to the French Centre National de Recherche Scientifique (CNRS). Nançay also gratefully acknowledges the financial support of the Région Centre in France.