Can the fact of 'belonging' to different groups be gathered from religious practice or religious absenteism? Or, in other words, is the fact that one goes to church on sunday, a criterion which permits us to distinguish different forms of behaviour in different spheres of life? Research carried out in Seraing, Belgium, indicates that it is easy to place too much reliance on the discriminating value of this criterion. In the spheres in which the hypothesis has been tested, the fact of going or not going to church is not linked to different forms of behaviour. This is especially true where spontaneous social behaviour is concerned. As far as participation in social organisations is concerned, which, in Belgium, are often opposed on the basis of a double philosophical political religi ous system, the situation is slightly different. But even in these cases another criterion seems to be more discernible: i.e. being very active in voluntary christian or non christian organisations. This characteristic distinguishes those engaged from those not engaged, if they go to church or not. On the other hand, on certain points militant christians and militant non christians are more opposed than church goers and non church goers.