The author has attempted to investigate in Chile the causes that promote in some years a heavy attack of Puccinia graminis tr. and the factors that reduce the attack in the southern part of the country. Climaterical dat a of Santiago from 1924 to 1942 was colected. This data corresponds to climatic factors probably related to the intensity of attack according to literature and observations of Chilean farmers. The named years were ranked from 1 to 4, according to the intensity of attack: observed in each year in the central zone of the country. The correlation coefficients between each of the considerad climatic factors and intensity of attack were calculated. Five of the factors were significantly correlated with intensity of attack. The highest correlation coefficient was that between the average minimum winter temperature and the intensity of attack (r = + 0,753, significant for P: 1%). Applying the concept of partial correlation it is iniered that, of those five factors correlated with intensity of attack, the only one which presents a direct correlation is the average minimum winter temperature. This means that, when the mean of all the daily minimums during winter is high, there are more chances that a strong attack of the disease in spring time will be produced. This correlation do not explain the lack of attack in the south, where the average mínimum temperature is higher than in the central zone (Santiago).