Ok, guys, you persuaded me. I guess I was misled by the charge separation being induced by the external electric field, so naively I thought: excess charge present at the two ends of the conductor = potential gradient, but obviously it is not like that.
Thanks a lot for your time, much...