This paper deals with the analysis of the relationship between CO2 emissions and temperatures. For this purpose, global CO2 emissions and four measures of global temperatures (land, land and ocean, northern and southern temperatures) are used. We used techniques based on fractional integration and cointegration. The results indicate first that the orders of integration differ in the two variables. Thus, while emissions are I(1) or I(d) with d higher than 1, temperatures display orders of integration strictly smaller than 1 and thus invalidating the hypothesis of cointegration between the two variables. Due to this, another approach is conducted where we suppose that the emissions are weakly exogenous in relation to the temperatures. The results using this approach show a significantly positive relationship between the two variables with a long memory pattern.