In order to investigate the phenomenon of the distribution of gains from international trade, Arghiri Emmanuel's ideas are firs t critically discussed, particularly in relation to the traditional Ricardian framework as applied to labor-surplus economics. It is found that Emmanuel's concept of unequal exchange, which has been termed non-equivalent exchange by Jan Otto Anderson, has certain theoretical drawbacks. In particular, it has been pointed out that the question involved is not one to prove that the poor countries are actually worse off through trade as suggested by Emmanuel. The question involved is rather one of redistribution of gains from trade as has been voiced in the search for a new international economic order by the members of some developing countries in the U.N. Such an approach leads to the adoption of the concept of a generalized asymmetric exchange as the measure of unequal exchange. This generalization has been achieved in terms of Leontief 's input output analysis. Such a measure coincides with the disjunctive exchange approach when the input-output coefficients are modified over time. The Leontief input-output analysis leads to an aggregation problem which has been solved by taking labor as the only primary factor of production - an approach standardized by Leontief himself. According to this measure, the extent of unequal exchange can be quite different from those obtained by the measures suggested by Emmanuel, Shanin and others. It has been pointed out that there is no a-priori reason to believe that a poor country necessarily gains less than its rich counterpart. Indeed, the test that has been made of the measure in the case of trade between Ecuador and the USA shows that it is Ecuador rather the U. S. which gains more from trade between them. The study also suggests some policy recommendations for reducing unequal exchange with special reference to labor-surplus economics.