consolidated or even deepened their underdevelopment.There is little evidence to suggest that this process is likely to be reversed in the foreseeable future except insofar as the provincial poor migrate and become the poor of the metropolitan cities. Externally,the evidence is that although the initial development of Sao Paulo's industry was relatively autonomous it is being increasingly satellized by the world capitalist metropolis and its future development possibilities are increasingly restricted.(5)This development,my studies lead me to believe,also appears destined to limited or underdeveloped development as long as it takes place in the present economic,political,and social framework. We must conclude,in short,that underdevelopment is not due to the survival of archaic institutions and the existence of capital shortage in regions that have remained isolated from the stream of world history.On the contrary,underdevelopment was and still is generated by the very same historical process which also generated economic development:the development of capitalism itself.This view,I am glad to say,is gaining adherents among students of Latin America and is proving its worth in shedding new light on the problems of the area and in affording a better perspective for the formulation of theory and policy.(6) D The same historical and structural approach can also lead to better development theory and policy by generating a series of hypotheses about development and underdevelopment such as those I am testing in my current research.The hypotheses are derived from the empirical observation and theoretical assumption that within this world-embracing metropolis-satellite structure the metropoles tend to develop and the satellites to underdevelop.The first hypothesis has already been mentioned above: that in contrast to the development of the world metropolis which is no one's satellite, the development of the national and other subordinate metropoles is limited by their satellite status.It is perhaps more difficult to test this hypothesis than the following ones because part of its confirmation depends on the test of the other hypotheses. Nonetheless,this hypothesis appears to be generally confirmed by the non-autonomous and unsatisfactory economic and especially industrial development of Latin America's national metropoles,as documented in the studies already cited. The most important and at the same time most confirmatory examples are the metropolitan regions of Buenos Aires and Sao Paulo whose growth only began in the nineteenth century,was therefore largely untrammeled by any colonial heritage,but was and remains a satellite development largely dependent on the outside metropolis, first of Britain and then of the United States. A second hypothesis is that the satellites experience their greatest economic development and especially their most classically capitalist industrial development if and when their ties to their metropolis are weakest.This hypothesis is almost diametrically opposed to the generally accepted thesis that development in the underdeveloped countries follows from the greatest degree of contact with andconsolidated or even deepened their underdevelopment. There is little evidence to suggest that this process is likely to be reversed in the foreseeable future except insofar as the provincial poor migrate and become the poor of the metropolitan cities. Externally, the evidence is that although the initial development of Sao Paulo's industry was relatively autonomous it is being increasingly satellized by the world capitalist metropolis and its future development possibilities are increasingly restricted.(5) This development, my studies lead me to believe, also appears destined to limited or underdeveloped development as long as it takes place in the present economic, political, and social framework. We must conclude, in short, that underdevelopment is not due to the survival of archaic institutions and the existence of capital shortage in regions that have remained isolated from the stream of world history. On the contrary, underdevelopment was and still is generated by the very same historical process which also generated economic development: the development of capitalism itself. This view, I am glad to say, is gaining adherents among students of Latin America and is proving its worth in shedding new light on the problems of the area and in affording a better perspective for the formulation of theory and policy.(6) V The same historical and structural approach can also lead to better development theory and policy by generating a series of hypotheses about development and underdevelopment such as those I am testing in my current research. The hypotheses are derived from the empirical observation and theoretical assumption that within this world-embracing metropolis-satellite structure the metropoles tend to develop and the satellites to underdevelop. The first hypothesis has already been mentioned above: that in contrast to the development of the world metropolis which is no one's satellite, the development of the national and other subordinate metropoles is limited by their satellite status. It is perhaps more difficult to test this hypothesis than the following ones because part of its confirmation depends on the test of the other hypotheses. Nonetheless, this hypothesis appears to be generally confirmed by the non-autonomous and unsatisfactory economic and especially industrial development of Latin America's national metropoles, as documented in the studies already cited. The most important and at the same time most confirmatory examples are the metropolitan regions of Buenos Aires and Sao Paulo whose growth only began in the nineteenth century, was therefore largely untrammeled by any colonial heritage, but was and remains a satellite development largely dependent on the outside metropolis, first of Britain and then of the United States. A second hypothesis is that the satellites experience their greatest economic development and especially their most classically capitalist industrial development if and when their ties to their metropolis are weakest. This hypothesis is almost diametrically opposed to the generally accepted thesis that development in the underdeveloped countries follows from the greatest degree of contact with and