This paper describes an experiment that examines the influence of visual realism on reported presence. 33 participants experienced two different renderings of a virtual environment that depicts a pit in the centre of a room, in a head-tracked head-mounted display. The environment was rendered using parallel ray tracing at 15fps, but in one condition ray casting (RC) was used achieving a result equivalent to OpenGL based per-pixel local illumination, and in the second full recursive ray tracing (RT). The participants were randomly allocated to two groups – one that experienced RC first followed by RT, and the second group in the opposite order. Reported presence was obtained by questionnaires following each session. The results indicate that reported presence, in terms of the ‘sense of being there’ was significantly higher for the RT than for the RC condition. Categories and Subject Descriptors I.3.7 [Three-Dimensional Graphics and Realism]: ray tracing, virtual reality. General ...