Until very recently, the Las Campanas surveys of the Virgo cluster (Binggeli et al. 1985) and the Fornax cluster (Ferguson 1989) were about the only galaxy catalogs complete down to low luminosities. The Local Group and the catalog of nearby galaxies (Kraan-Korteweg & Tammann 1979; KKT) are easily affected by incompleteness (e.g., Irwin et al. 1990), especially in the zone of avoidance (e.g., Kraan-Korteweg et al. 1994; Huchtmeier et al.\ 1995; McCall & Buta 1995; Huchtmeier et al. 1997). Studies of nearby groups of galaxies were usually confined to only a small part of the sky and, due to the groups' small distances, are not sensitivity limited.
Recently, Côté (1995) has surveyed the nearby Sculptor and Centaurus groups, and discovered nine new dwarf members. This adds support to the notion that the census of nearby galaxies, particularly low surface brightness galaxies, is far from complete (e.g., Shade & Ferguson 1994).
There are several good reasons to look for more nearby dwarf galaxies. The first, obvious reason is to improve completeness of the sample of nearby galaxies (e.g., KKT) for the luminosity function of the "field sample''.
Figure 1: a)
HI profiles of the detected galaxies as seen with the 100-m
radio telescope at Effelsberg which has a half power beam width of
9.3
at 21-cm wavelength. As local HI is observed all over the
sky our (
) observing procedure produces the difference of the
local emission between the source and the reference position at local
velocities around zero 
Figure 1: b)
HI profiles of the detected galaxies as seen with the 100-m
radio telescope at Effelsberg which has a half power beam width of
at 21-cm wavelength. As local HI is observed all over the
sky our () observing procedure produces the difference of the
local emission between the source and the reference position at local
velocities around zero
A second goal is a study of group dynamics. HI spectra provide the only practical method for obtaining radial velocities of the very low surface brightness galaxies. A large population of dwarf galaxies with known radial velocities will allow a new estimate of the mass distribution of the M 81 group. The finding of Bothun et al. (1987) that a previously cataloged dwarf galaxy in the Virgo cluster is indeed a massive low surface brightness background galaxy stresses both the need for radial velocity measurements in such studies and the opportunity for important serendipitous discoveries.
A further motivation is to find additional very low mass galaxies. Without velocity information it is impossible to distinguish between truly low mass galaxies and low surface brightness background galaxies. HI line-widths give an additional clue to discriminate against higher mass systems. Skillman et al. (1988, 1989a,b) have shown that extremely low mass galaxies hold the most promise for finding regions of near-primordial abundances. Follow-up CCD H-alpha imaging of the most promising candidates can then lead to the identification of HII regions for chemical abundance studies. The lowest mass galaxies are also important for helping to determine the trend of mass-to-light ratio with total luminosity (Côté 1995).
In his review of nearby groups of galaxies, de Vaucouleurs (1975) identified the M 81 group as the second closest group to ours at a distance of 2.5 Mpc. Modern Cepheid distance measurements of M 81 place it at a distance of 3.6 Mpc (Madore et al. 1993). de Vaucouleurs noted that the precise extent of the M 81 group was not well determined, as it appeared that other nearby galaxies (e.g., NGC 2403, NGC 2366, NGC 4236) could be outlying members of the M 81 group, which, itself, might be a condensation in a larger "cloud". Geller & Huchra (1983) identified the M 81 group (their group No. 52) among the bright galaxies contained in the CfA catalog as a smaller, more concentrated distribution than de Vaucouleurs.
Tully (1988) produced a catalog of galaxies judged to be reasonably
complete out to 
Mpc. Using this catalog, Tully (1987)
identified groups (thought to be gravitationally bound and
virialized entities), associations, and clouds. From his
analysis, the M 81 group is found to be more extended (in agreement
with de Vaucouleurs original definition) and lying in the
Coma - Sculptor Cloud, which contains many familiar nearby groups
(M 81, the Local Group, the Sculptor Group, and the Centaurus Group).
Given the extended nature of the M 81 group, it was clear that a
successful search would need to cover a large area of the sky.
There are several catalogs containing optical information on M 81 dwarf
galaxies; we have used the catalogs by Karachentseva (1968),
Börngen et al. &
Karachentseva (1982) and Börngen et al. (1982) who list 39 faint probable
members of the M 81 group excluding the known bright galaxies that usually
are taken to identify this group (e.g., de Vaucouleurs 1975). These dwarf
galaxies have been further classified concerning probable membership
by Karachentseva et al. (1985).
Lo et al. (1986) conducted a deep IIIa-J
survey of the M 81 group covering 36 fields using the 48 inch Palomar Schmidt.
After visually searching the plates for low surface brightness objects,
137 candidate dwarf galaxies
were detected of which 57 were judged to be group members.
We composed a search list from the above named catalogs, which amounts to
136 candidate dwarf galaxies roughly in the range 65
Dec.
80 and 
.
Most of these dwarfs have crude visual magnitudes of 17 to 19 mag (absolute
magnitudes of -11 to -9 at the distance of the M 81 group), diameters around
1 arcmin or less, and classifications of type Im with a few E type or
spheroidals (Börngen spheroidals were not detected in HI). A few galaxies
are very close to brighter galaxies of this sample. This confusion could not be
resolved as most probably the detections are due to the brighter galaxies.