4 Results

Tables 3, 4 and 5 (given only in electronic form) present the extensive photometry in the near-infrared H band of a sample which is complete in all the types of magnitudes listed in Table 2, in an area of about 400 arcmin² toward the Coma cluster of galaxies. Figures 1 and 7 show the studied region and the detected objects. An important point to mention is that $H\sim 17$ mag is six magnitudes below the characteristic magnitude of galaxies, well into the dwarfs' regime at the distance of the Coma cluster. To our knowledge, this is the first complete sample of galaxies in a nearby cluster for which the photometry is published.

Figure 8 shows the classification parameter as a function of H_10''. According to the Sextractor criteria, the classification is secured for most objects in this field, at least down to the completeness magnitude. When $Class\_star <\ \sim0.3$, objects are identified as galaxies, whereas $Class\_star >\ \sim 0.7$ (for 0404 and 0504) and $Class\_star >\ \sim 0.8$ (for patch1) correspond to stars. The difference between fields is due to seeing conditions.

Figure 9 displays the result of this classification when comparing the isophotal area (at $\mu =22$ mag arcsec^-2) to the corresponding isophotal magnitude. Stars and galaxies easily separate in two different sequences in a large magnitude range. This visual criterion is similar to the one originally proposed by Godwin & Peach (1977). The difference between the two sequences tends to vanish at faint magnitudes. In our case, the separation between stars and galaxies seems to be reliable down the completeness limit.

Figure 10 shows a comparison between the different magnitudes obtained in this field. As expected, magnitudes are sensitive to the size of the apertures for bright objects ( $H \raisebox{-0.5ex}{$\,\stackrel{<}{\scriptstyle\sim}\,$ }14$), and apertures larger than $\sim 10''$give similar results for $H \raisebox{-0.5ex}{$\,\stackrel{>}{\scriptstyle\sim}\,$ }14$.

Tables 3, 4 and 5 give, for each object brighter than at least one of the limits indicated in Table 2:

- sky coordinates (epoch J2000);

- H-band magnitude within the apertures: 5'', 10'', 13'', 14.5'' and 18.5'', and their errors, as computed by Sex. Listed errors do not include photometric zero-point errors (see Sect. 3.7), those arising from the compositing, and implicitly assume a gaussian or poissonian error distribution;

- isophotal ($\mu =22$ mag arcsec^-2) and Kron magnitudes (with errors). Kron magnitudes are computed by integrating the flux in an area 2.5 times larger than the Kron area, with a minimun aperture radius of 1.8 arcsec. "9.99" or "99.99" means that no measure is available (usually because the object is too small or too faint for its flux to be measured through a large aperture);

- isophotal area (at $\mu =22$ mag arcsec^-2), in arcsec;

- ellipticity and position angle (from North to East);

- Kron radius, or 1.8 arcsec if the latter is larger;

- stellarity index as defined by Sextractor;

- an optical identification, usually taken from the Godwin, Metcalfe & Peach (1983, and private comunication).

This catalogue is the database used to compute the luminosity function in this field (Andreon & Roser, in press), and also to study the morphology and the photometric properties of these galaxies in details.

Acknowledgements

We thank Alain Klotz for obtaining our March data the TBL. We also thank the TBL technical staff for their support during the observing runs. The USNO catalog and the POSS image were obtained at the Canadian Astronomy Data Center (CADC), which is operated by the Herzberg Institute of Astrophysics, National Research Council of Canada.