This dissertation investigates the commodity currency argument of primary and secondary effects of the terms of trade on exchange rates and real income, respectively. The Johansen procedure of cointegration testing is applied to dynamic models for a set of four developed countries (New Zealand, Australia, Norway and Iceland) and five less developed countries (Colombia, India, Malaysia, Thailand and Venezuela) each against it's major trading partner and the United States. The stationarity of the real exchange rates as well as cointegration between the nominal exchange rates and the ratio of national price-levels (price-ratio) are analyzed for two sets of data (annual and a relatively shorter quarterly) and for two different price measures (GDP deflator and CPI). The hypothesis of the terms of trade effects is investigated by including the terms of trade variable in the models of real exchange rates and models of nominal exchange rates and price-ratios. For developed countries, using annual data, real exchange rates are found to be stationary without the terms of trade in eleven cases, and on including the terms of trade evidence of cointegration is found in three further cases. For the quarterly data of the developed countries, there is some evidence of the real exchange rate being stationary without the terms of trade and some evidence of cointegration between the real exchange rates and the terms of trade for both price indices. Analysis of the long-run equilibrium relationship between the nominal exchange rate and price-ratio without the terms of trade showed some evidence of a cointegrating relationship. On including the terms of trade strong evidence of cointegration is obtained for New Zealand and Austra1ia but not for Norway. Moreover, while evidence for the long-run equilibrium relationships of purchasing power parity are mixed there is strong evidence of improvement in the terms of trade leading to appreciating exchange rates. Also, the terms of trade are found to be exogenous between small countries (New Zealand-Australia). Only quarterly data and CPI are used for the less dev.~loped countries. Results on stationarity of the real exchange rate, the equilibrium relationship between the exchange rates and price-ratio, and the role of terms of trade are again mixed. Finally, the short-run effects of the terms of trade on real income are investigated for New Zealand, Australia and Norway using quarterly data. Dynamic models of first, the real income and terms of trade and real income, terms of trade, and, second, the real exchange rates are analyzed. Validity of the commodity currency argument is evidenced only in some of the three-variable models. / Ph. D.