Brazilian Amazon deforestation rate is found to display a unit root and to be cointegrated with Brazilian GDP and its square – An Environmental Kuznets Curve (EKC). Although, it is not the first time that such an EKC is detected, this may be the first such time-series evidence. Detecting an EKC is hampered by several econometric issues that have been shown to lead to possibly spurious results in cross-section and panel contexts, but are satisfactorily addressed in a cointegrated (time-series) framework. Alternative theories for explaining the deforestation path are rejected. There is evidence that the " Action Plan " of the Brazilian government against deforestation had an important effect. These results are in contrast to the economics literature on an EKC in emissions such asCO2, but appear to be consistent with a geographical sciences literature that considers that deforestation declines when alternative activities become available.