This is probably somewhere in the archives but I couldn't find it: What's the R equivalent function to the Stata command: "Use v1 v3 v5 v7 v2 using fileX.dat" Where vn are variables named within the headed file fileX.dat I know I could do this crudely by just loading the file then creating a subset, but the file is pretty huge. The Import/Export Manual gives some examples of how to perform some low level operations using scan() but I'd rather use something more prosaic. Cheers, Jon Minton [[alternative HTML version deleted]]