The use of spatial stimulus features seems to facilitate both intramodal and crossmodal temporal order judgments (TOJ). For example, TOJ between two tactile stimuli are easier when the two stimulated hands are held far apart rather than close together (1), suggesting automatic coding of tactile location in external coordinates. Similarly, a TOJ between two stimuli from different modalities is easier when the two stimuli are separated in space (2). We have previously shown that use of spatial features for intramodal TOJ becomes evident by the age of six years (3). Here, we tested whether the advantage of spatial separation in crossmodal comparisons is observable at the same age. Fifty-nine children between 4 and 12 years as well as 13 young adults performed a modality TOJ task for simple tactile and visual stimuli. Stimuli were presented either within the same or in different hemifields. Spatial separation improved TOJ performance only for children aged 10 years and older. However, crossmodal TOJ performance was worse than intramodal TOJ performance starting at the age of 6 years. Crossmodal TOJ performance comparable to adults was not observed before the age of 12 years. We speculate that the ability to redundantly code sensory input in modality-specific and supramodal (external) spatial coordinates facilitates intramodal temporal processing. Further refinement of the processes providing external spatial coordinates (e.g., touch remapping) then results in integrated use of space and time allowing for more precise assignment of sensory inputs to the same or to different events.