1 Introduction

1.1 The discovery of giant low surface brightness galaxies

In 1987, Bothun et al. reported the serendipitous discovery of the extraordinarily large low surface brightness (LSB) galaxy now known as Malin-1. In spite of having a projected B-band central surface brightness of only 26.5 mag arcsec^-2, Malin-1 is the largest spiral galaxy known, with a disk scale length of 73 kpc (assuming H₀=75 kms^-1 Mpc^-1), and an exceptionally high H I mass ( $\sim10^{11}~\mbox{${M}_\odot$ }$) and optical luminosity ( M_B=-23.1). Subsequently, through systematic searches of photographic survey plates, other galaxies with similar (albeit slightly less extreme) properties to Malin-1 have been uncovered (Bothun et al. 1990; Sprayberry et al. 1993; Sprayberry et al. 1995b). We hereafter refer to these as ``LSB Giants''. A handful of LSB Giants are also found in the UGC (Nilson 1973), NGC (Dreyer 1953), and ESO (Lauberts & Valentijn 1989) catalogues (see Gallagher & Bushouse 1983; Impey & Bothun 1989; Walsh et al. 1997; Pickering et al. 1997; Schombert 1998). Nonetheless, while recent photographic and CCD surveys have uncovered large numbers of new small and medium-sized, moderate-to-low surface brightness spiral galaxies (e.g., Schombert et al. 1992; Impey et al. 1996; O'Neil et al. 1997), LSB Giants have remained relatively rare. Having faint, diffuse disks, but sizes, H I masses, and luminosities at the high end for disk galaxies, the LSB Giants occupy a unique realm of physical parameters space and may share evolutionary histories distinct from other LSB galaxies (e.g., Hoffman et al. 1992).

Since a continuum of values exists for galaxy properties such as surface brightness, luminosity, and scale length, Sprayberry et al. (1995b) proposed to define LSB Giants as those objects meeting a ``diffuseness index'' criterion: $\mu_{B}(0)+5{\rm log}(h_{\rm r})>27.6$, where $\mu_{B}(0)$ is the extrapolated, deprojected B-band disk central surface brightness in magnitudes arcsec^-2, and $h_{\rm r}$ is the disk scale length in kpc. Among seven LSB Giants described by Sprayberry et al. (1995b), mean properties include: $<B-V>=0.73\pm0.05$, $<\mu_{B}(0)>=23.23\pm0.19$ mags arcsec^-2, and $<h_{\rm r}>=13.0$ kpc. The colors of these LSB Giants are thus comparable to those of normal spirals (Sprayberry et al. 1995b), but are redder than typical colors of many small and moderate-sized LSB disks (e.g., McGaugh & Bothun 1994; Matthews & Gallagher 1997; de Blok et al. 1996; Beijersbergen et al. 1999). In addition, the LSB Giants are distinct from other more common LSB spirals in that they often have a significant bulge component (e.g., Gallagher & Bushouse 1983; Knezek 1993, 1998), and frequently their centers harbor an active nucleus (e.g., Schombert 1998).

The origin and evolutionary histories of LSB Giant galaxies are still enigmatic. Hoffman et al. (1992) have proposed a formation scenario whereby these systems form in very low density regions from rare, 3$\sigma$ density fluctuations. They predict these galaxies should exhibit quiescent, unevolved, gas-rich disks, with rotation curves that flatten near $V_{\max}\sim300~\mbox{km\,s$^{-1}$ }$. Knezek (1993) has suggested an alternative scenario, based on Kormendy (1989), whereby LSB Giants may have dissipatively formed from massive, metal-poor dark matter halos.

1.2 The need for new H I observations

Testing formation and evolution scenarios for LSB Giants requires an accurate knowledge of the neutral gas properties and linewidths of these galaxies. And only by combining such measures with optical data can we begin to build a picture of the star-formation histories of these systems and their relationship to other types of LSB galaxies.

Other motivations also exist for improved H I observations. Hoffman et al. (1992) have argued that for the enormous disks of LSB Giants to remain quiescent over a Hubble time, they must be very isolated. Yet studies hint that LSB Giants are in fact less isolated than other LSB spirals, although redshift surveys in the vicinities of these objects are still incomplete (Sprayberry et al. 1995b). Pointed H I observations in the vicinity of LSB Giants can thus reveal if these galaxies have any yet-undiscovered gas-rich neighbors.

Another important use of H I data is for exploring the Tully-Fisher (TF) relation for giant LSB spirals. Sprayberry et al. (1995a) have shown that at least two of the presently known LSB Giant galaxies are extreme outliers from the TF relation defined by normal galaxies. This is unlike the bulk of moderate-sized, moderate luminosity LSB galaxies, which tend to follow TF (Sprayberry et al. 1995a; Zwaan et al. 1995; Verheijen 1997). It is of considerable interest therefore to assess from a larger sample whether LSB Giants deviate systematically from the TF relation.

While previous H I observations have established that LSB Giants are in general very gas-rich ( ${M}_{\rm HI}$ $\;\gtrsim10^{10}$ ${M}_{\odot}$; e.g., Sprayberry et al. 1995b; Walsh et al. 1997; Pickering et al. 1997, 1999), unfortunately existing H I data for many LSB Giants are of dubious quality (i.e., the galaxy was confused or resolved by the telescope beam, the spectra are of low signal-to-noise, or measurements from different workers are highly discrepant; see also Table 5 and Sect. 4).

For example, based on Arecibo 21-cm observations of 3 objects, Sprayberry et al. (1995b) suggested that peculiar, asymmetric H I profiles may be commonplace for LSB Giants. However, since the H I extents of these galaxies were expected to be comparable to the size of the telescope beam, it is important to verify that these ``peculiar'' spectra do not result from some combination of source resolution and telescope mispointing. In addition, independent checks on derived H I parameters are valuable since it is more difficult to accurately measure integrated fluxes and linewidths when the global H I profiles are quite broad compared to the bandwidth used, and when sources are at large recessional velocities ( $V_{\rm r}\gtrsim$ 15000 kms^-1), where flux calibration can become increasingly uncertain. Finally, there are still a handful of known LSB Giants for which no H I data have previously been obtained.

Quality single-dish H I spectra for LSB Giants are also a useful precursor and complement to H I aperture synthesis studies of these galaxies (e.g., Walsh et al. 1997; Pickering et al. 1997, 1999). H I mapping is of course a critical part of understanding the dynamics and gas distributions for these galaxies, but since many of the known examples of LSB Giants are rather distant ( $V_{\rm r}>10\,000$ kms^-1) and of relatively modest optical angular size (D₂₅<2'), such observations are challenging and benefit from careful planning based on prior H I measurements. Moreover, because LSB Giants are generally expected to have disks with relatively low H I surface densities (e.g., Pickering et al. 1997), diffuse emission can be missed in aperture synthesis measurements, and total flux and maximum rotational velocity measures from H I pencil beam observations serve as an important check.

Based on the above motivations, we have used the Nançay Radio Telescope to obtain new global observations of a sample of 16 LSB Giant galaxies for which existing global H I measurements were incomplete, required reconfirmation, or were nonexistent. We measure integrated H I line fluxes, linewidths, and recessional velocities, and attempt to clear up conundrums surrounding several of these objects in the literature.