search for: station_buffer

...then go through the second column of TABLE 2 and find the corresponding buffer size from TABLE 1, then rep() the value of TABLE 1 in the column S of TABLE 2 corresponding to the right combination... One other (quick) method would be to create a unique factor from the two first columns of TABLE 1 (station_buffer) but maybe there is something cleaner ? Do you have an idea ? Thanks ! Mel PS: those tables are supposed to be very large datasets of several thousands of rows. ___________________________________________________ Mélanie HAMEL Institut de Recherche pour le Développement (I.R.D.) U227 - CoRéUs...