2 Observations and data reduction

The observations were carried out with the Wide Field Camera (WFC) at the Prime Focus at the INT 2.5 m telescope located at Observatorio de El Roque de los Muchachos (La Palma), on April 26th and 28th 2000, under photometric conditions. The average seeing ranged from 1.5 to 2 arcsecs during both nights. Given the mean velocity of the galaxies in the two clusters under study, $6555~\pm~684$ kms^-1 and $6990~\pm~821$ km s^-1 for Abell 1367 and Coma respectively (Fadda et al. 1996), the narrow-band [S II] filter ( $\lambda_{0} = 6725$ Å, $\delta \lambda \approx 80$ Å) was used to isolate the H$\alpha$ line and the r' Sloan-Gunn broad-band filter ( $\lambda_{0} = 6240$ Å, $\delta \lambda \approx 1347$ Å) to recover the continuum. Figure 1 shows the transmitance profiles of both filters. Given the width of the [S II] filter, the [N II] $\lambda\lambda$6548, 6584 Å lines are included in the high transmitance pass-band of this filter, so in what follows we will refer to the combined H$\alpha$ + [N II] flux and equivalent width, as H$\alpha$ flux and equivalent width respectively.

The WFC is composed of a science array of four thinned AR coated EEV 4K$\times$2K devices, plus a fifth acting as autoguider. The pixel scale is 0.333 arcsec pixel^-1, giving a total field of view of about $34\times 34$  arcmin². Given the arrangement of the detectors, a square area of about $11 \times 11$ arcmin² is lost at the top right corner of the field. The top left corner of detector #3 is also lost because of filter vignetting.

Figure 2: Projected positions of our exposures in the Abell 1367 (left plot) and Coma (right plot) clusters. Superimposed contours correspond to the X-ray maps from Donnelly et al. (1998) and White et al. (1993) respectively. Filled dots represent the galaxies showing H$\alpha$ emission.

Four fields near the center of each cluster were observed. Three different exposures, slightly dithered to remove cosmic rays, were obtained for each position in each filter, except for the second exposure of the Coma cluster where only one exposure per filter was obtained. Figure 2 shows our surveyed area. Our observations cover mainly the North-East region of the Coma cluster as described by Colless & Dunn (1996), coinciding with the central part of the Godwin catalog of the Coma cluster (Godwin et al. 1983). One of our fields of Abell 1367 (number 1 in Fig. 2) is not covered by the Godwin catalog (Godwin & Peach 1982). For comparison the X-ray contour maps of the two clusters (White et al. 1993 for Coma, and Donnelly et al. 1998 for Abell 1967) are plotted in the figure. The galaxies detected in H$\alpha$ are marked with filled dots. The diary of the observations is presented in Table 1.

The data reduction was carried out using standard tools in the IRAF environment. The astrometric solution was found with the USNO catalog of stars. The accuracy of this solution was found to be better than 3 arcsecs throughout the frames. Several exposures of standard spectrophotometric stars were taken during both nights. The chip-to-chip differential responses were derived by direct comparison of the photometry measured for the objects, non-saturated stars and galaxies, present in the overlapping regions. Zero-points and extinction coefficients were derived from the calibration equations. Overall, our photometric uncertainty is less than 10%.

In order to properly subtract the continuum from the H$\alpha$ frames, we scaled the counts of the continuum frames until (unsaturated) stars and elliptical galaxies reached an average H$\alpha$ + [N II] equivalent width of 0 Å. The net H$\alpha$ + [N II] photometry of the selected galaxies was performed using the QPHOT command of the APPHOT package in IRAF. Aperture photometry was carried out, in both the ON-band and continuum frames, for each galaxy and subtracted to get the net H$\alpha$ + [N II] fluxes.