1 Introduction

The strong morphology segregation observed in rich clusters of galaxies (Dressler 1980) testifies to the fundamental role played by the environment on the evolution of galaxies. Which physical mechanisms are responsible for such transformations is however still a matter of debate. Several processes might alter the evolution of cluster galaxies. Some of them refer to the interaction of the galaxies with the intracluster medium (Gunn & Gott 1972) and others account for the effects of gravitational interactions produced by the gravitational potential of the cluster (Merritt 1983) or by galaxy-galaxy interactions (Moore et al. 1996, 1998, 1999). All these mechanisms can produce strong perturbations in the galaxy morphology with the formation of tidal tails, dynamical disturbances which appear as asymmetries in the rotation curves (Dale et al. 2001) and significant gas removal (Giovanelli & Haynes 1985; Valluri & Jog 1990).

Some of these processes are expected to produce changes in the star formation rates of galaxies in clusters. Several studies have addressed the issue of the influence of the cluster environment on the SFR of disk galaxies, however no agreement has been established so far: while some authors proposed similar or even enhanced star formation in cluster spirals than in the field (Donas et al. 1990, 1995; Moss & Whittle 1993; Gavazzi & Contursi 1994; Moss et al. 1998; Gavazzi et al. 1998; Moss & Whittle 2000), some others claim quenched SFRs in cluster spirals (Kennicutt 1983; Balogh et al. 1998; Hashimoto et al. 1998). This discrepancy could arise from non-uniformity of the adopted methods (UV vs. H$\alpha$ vs. [O II] data) or from real differences in the studied clusters (Virgo, Coma, Abell 1367, clusters from Las Campanas Redshift Survey, clusters at z > 0.18).

In particular, an enhanced fraction of spirals with circumnuclear H$\alpha$ emission was found in the highest density regions of some nearby clusters (Moss et al. 1998; Moss & Whittle 2000), whereas no such difference was found for galaxies with diffuse emission. The compact H$\alpha$ emission seems associated with ongoing interactions of galaxies, but numerical simulations by Bekki (1999) showed that mergers between clusters and subclusters might produce central starbursts in cluster spirals.

Existing studies of the H$\alpha$ properties of galaxies in clusters suffer from various biases: the photoelectric data by Kennicutt et al. (1984) and Gavazzi et al. (1991, 1998) are based on samples of galaxies selected on the basis of their optical properties, independent of their H$\alpha$ properties. On the other hand, the objective-prism surveys by Moss et al. (1988, 1998) and Moss & Whittle (2000) are H$\alpha$ selected but they are too shallow to allow a determination of the H$\alpha$luminosity function as deep as desired.

With the aim of obtaining a reliable determination of the current SFR in nearby clusters of galaxies and to study the spatial distribution of the star formation regions, we undertook a deep imaging survey of a one degree $\times$ one degree area of the Coma and Abell 1367 clusters.

Our work provides the first deep and complete study of galaxies in clusters based on their H$\alpha$emission properties.

This paper is arranged as follows: Sect. 2 contains a description of the observations, of the data reduction and the detection procedures. The H$\alpha$ data are presented in Sect. 3. The H$\alpha$luminosity function and a brief discussion of the contribution of both clusters to the local star formation rate density are presented in Sect. 4. Conclusions are presented in Sect. 5. Comments on the most interesting objects as well as the H$\alpha$ images of the detected galaxies are given in the Appendix.

Figure 1: Transmitance of the filters used for the observations.