The series of experiments reported in this thesis concern the ability to make perceptual-motor judgements of distance (Ex. 1 to Ex. 7) and size (Ex. 8). Experiments 1 and 2 indicated that visual judgements of maximum step length were effected by; distance from the site of action, the angle at which the obstacle was presented and whether monocular or binocular vision was used. This suggested that perceived maximum ability was not based on a body scaled invariant as suggested by Gibson (1979). Experiments 3 and 4 were designed to investigate the effect of altering the length of distance to-be-remembered, and compared performance across both visual and kinaesthetic conditions. The results suggested that the reproduction of distance is normally based on memory for the location of the end point, rather than the extent of the distance. No support was found for the claim that differences between the accuracy of recall of location and extent was due to the differential rehearsability of visual and kinaesthetic codes. Instead, it was proposed that changes in the procedure may have influenced performance by reducing the usefulness of a 'landmark' based form of coding in the extent trials. Experiments 5 and 6 were designed to investigate predictions arising from one of the dominant models of cross-modal performance (Connolly and Jones, 1970). Connolly and Jones's model postulated that differences between intra- and cross-modal performance could be explained in terms of the characteristics of modality specific short-term storage codes, and that translation between codes occurs prior to short-term storage. In general the results obtained were supportive of the pattern of accuracy reported by Connolly and Jones. However, the effect of delaying until the end of the retention interval knowledge of the reproduction mode was inconsistent with the model, that is, withholding information about the required reproduction mode appeared to increase the accuracy of judgements. One explanation for this effect is that pre-translated information was held in a form which was associated with high levels of both accuracy and attention. This speculative explanation was seen to have parallels with the Working Memory model (Baddeley and Hitch, 1974). Experiments 7 and 8 used an interference task paradigm to investigate whether a separate visuo-spatial store could be demonstrated to exist in relation to perceptual-motor information. The results failed to find conclusive support for such a store. The cumulative findings of Experiments 1 to 8 are discussed in relation general models of perceptual-motor performance.