The urban heat island phenomenon has been documented for many cities all over the word. The absorption of short wave radiation due to canyon geometry, increased long wave radiation from the sky due to air pollution, anthropogenic heat sources, increased sensible heat storage and decreased evapotranspiration due to construction materials are only some of the numerous factors that contribute to this phenomenon. Buenos Aires (Argentina) is one of the most populated cities in South America. Some authors analyzed the urban heat island effect specifically for this city, and they found that yearly mean urban to rural temperature differences and rural temperatures are negatively correlated in almost every case, suggesting that urban heat island intensity depends, among other parameters, on the temperature itself. The same authors showed that this characteristic was observed at United States and Australian cities too. In this work, temperature variability at Buenos Aires city is analyzed for the 1960-2007 period together with the observed urban-rural differences. Three meteorological stations were taken into account for this purpose: Ortuzar (urban station), Ezeiza (neighboring station, here considered as “rural”) and Aeroparque (urban station next to the river). The climatology of the three stations analyzed at the four principal observational hours (00, 06, 12 and 18 UTC) showed the most relevant changes at nocturnal hours. Positive trends were observed in these cases, with a reduction at the urban-rural temperature differences. Other meteorological variables are studied in order to explain these aspects. Changes in the percentage of days with clear sky, covered sky and windiness were considered. These results allow an analysis of the stability conditions (Turner classification) related with the observed urban/rural temperature changes.