In this paper we briefly describe preliminary data from two experiments that we have carried out to investigate the relationship be-tween visual encoding and memory for objects and their locations within scenes. In these experiments, we recorded participants' eyemovements as they viewed a photograph of a cubicle with 12 objects positioned pseudo-randomly on a desk and shelves. After view-ing the photograph, participants were taken to the actual cubicle where they undertook two memory tests. Participants were asked toidentify the 12 target objects (from the photograph) presented amongst 12 distraetora. They were then required to place each of theobjects in the Iocation that they occupied in the photograph. These tests assessed participants' memory for identity of the objects andtheir locations. In Experiment 1, we assessed the influence of the encoding period and the test delay on object identity and locationmernory. In Experiment 2 we manipulated scanning behaviour during encoding by　"boxing" some of the objects in the photo. Weshowed that using boxes to change eye movement behaviour during encoding directly affected the nature of memory for the scene.The results of these studies indicate a fundamental relationship between visual encoding and memory for objects and their locations.We explain our findings in terms of the Visual Memory Model (Hollingworth & Henderson, 2002).