@Damocles: Yes indeed. However if you put a filter between the slits and the target-screen to find out which slit a photon has taken you won't see any interference anymore (although the filter should not influence it).
Together with some more experiments you can show that the more information you try to collect about the way a photon travelled the less inference pattern you'll get.
At least that's what we were thought in school, if I remember correctly.