Standard reduced form models are estimated for Canada to examine the relationships between real per capita GDP and four measures of environmental degradation. Of the four chosen measures of environmental degradation, only concentrations of carbon monoxide appear to decline in the long run with increases in real per capita income. The data used in the reduced form models are also tested for the presence of unit roots and for the existence of cointegration between each of the measures of environmental degradation and per capita income. Unit root tests indicate nonstationarity in logs of the measures of environmental degradation and per capita income. The Engle-Granger test and the maximum eigenvalue test suggest that per capita income and the measures of environmental degradation are not cointegrated, or that a long-term relationship between the variables does not exist. Causality tests also indicate a bi-directional causality, rather than a uni-directional causality, from income to the environment. The results suggest that Canada does not have the luxury of being able to grow out of its environmental problems. The implication is that to prevent further environmental degradation, Canada requires concerted policies and incentives to reduce pollution intensity per unit of output across sectors, to shift from more to less pollution-producing-outputs and to lower the environmental damage associated with aggregate consumption.