Thanks Hassan / Joshua >As above, set a UNIQUE index on the field, but use INSERT IGNORE to >convert duplicate-row-errors to warnings. > > Yes, that worked quite nicely; I guess whats its doing is overwriting whats already there. Interestingly, when I take the info out of the sql and into array, it takes about 0.7 secs longer for the 110,000 entries than for the 0.8 secs (average) uptake straight from the file. I wonder if there were many people accessing the site for this function, whether fileopen would lose this speed and sql be a better way of populating an array. At the moment, Im assuming that for any number of users fileopen & fgets is better than mysql_query & mysql_fetch_assoc. would this be correct? A.