"Robinson has snatched away the comfort of the urban thesis. The idea that at the time of the Edict of Milan (313) some 10 per cent of the population of the Roman empire was Christian, 6 million persons out of a population of 60 million, has been (he observes) commonly put forward as a working assumption for the size of the Christian movement. But the Roman empire was largely rural with as few as 10 per cent of the population living in cities. If the rural Christian population approximated to zero, the cities would then have to be thought to be (at least) largely Christian - which ample evidence shows that they were not. Therefore, the numbers usually cited in recent work for the size of the Christian population cannot stand; and neither can the claim of no substantial number of Christians in rural areas. The model normally drawn on in studying the growth of early Christianity, therefore, is no longer plausible. Like Gregory, the historian must leave the 'safe and peaceful city' behind."