which, especially in countries with a large peasant population, favoured high fertility ( Gauvreau et al. 1986). The combination of various factors such as the deepening of the crisis in the agricultural sector--especially subsistence agriculture--declining infant mortality, the spread of primary schooling, but also new values introduced by returning or visiting emigrants, all would have led to the emergence of a new reproductive timetable, rejecting the idea of large families in favour of smaller ones. These factors would, to some degree, have come to maturity in most of the countries of the region during the 1960s, probably leading to a drop in fertility, first among the more favoured social sectors, then fairly quickly among the remaining social classes, as has been demonstrated in the case of Guadeloupe ( Poirier 1989).
All in all, recent societal changes in the region, linked as much to social progress (reductions in mortality and extension of schooling) as to crisis situations (the non-viability of subsistence agriculture, high unemployment rates and massive under-employment), would have opened the way to an even more rapid diffusion of contraception to the extent that socio-economic conditions lent themselves to such diffusion and to the extent that the family planning programmes in the countries concerned were well organized.
This attempt to interpret the various types of demographic transition occurring in the Caribbean stresses the impact of economic and social factors on the initiation, the speed, and the breadth of fertility decline. The various types of transition seem, in particular, to have been conditioned largely by the existence of larger or smaller peasant populations, and by the degree of their marginalization since the end of the Second World War.
The economic system of plantations and the existence of peasant masses (campesinos, minifundistas) living on the fringe of this system are also to be found in several Latin American countries. However, comparisons between the Caribbean and Latin America should be made cautiously. In fact, with the exception of Haiti, Latin American countries have been independent for much longer than those of the Caribbean. They are also larger, generally less homogeneous, and it is not always clear that they can easily be studied as national entities. As far as fertility is concerned, before the First World War, no country of the Caribbean experienced the kind of fertility decline found in Argentina and Uruguay. On the other hand, the declines in fertility that began in the late 1950s in several countries of the English-speaking Caribbean have no equivalent in Latin America.
In spite of these differences, in Latin America, as in the Caribbean, the decline in infant mortality preceded the decline in fertility, and in both cases the latter is the result of the diffusion of contraception, as well as of a probably non-negligible use of abortion in many countries. Furthermore, the observed declines in fertility