Dyslexic children have been shown to have poorer phonological awareness and phonological memory skills than normal readers. Evidence from a number of studies shows that normal readers show impaired recall for phonologically similar lists whereas dyslexic children are more resistant to this effect (Shankweiler, Liberman, Mark, Fowler and Fischer, 1979). Further research has shown that dyslexic children often show impaired recall when items are orthographically similar (Rack, 1985). This has led to the belief that dyslexic individuals are impaired in their use of phonological processing, and that specifically, they encode items in memory using orthographic rather than phonological coding. 6 dyslexic children aged 9-11 years were compared with 6 chronologically-age-matched controls. Subjects had to remember a list of target words presented either visually or auditorily. Each target word was then presented paired with either an orthographically similar or phonologically similar distracter word and subjects had to decide which word was in the original list. It was hypothesised that dyslexic children would make more errors than controls due to a phonological memory deficit. It was also hypothesised that when the words were orthographically similar, dyslexic children would make a higher proportion of errors, whereas controls would make more errors when the pairs were phonologically similar. We expected this pattern of results to be obtained in both presentation modes. A mixed design ANOVA found no significant effects or interactions of any of the comparisons, so we did not find support for our hypotheses. In our study, both dyslexic and control children made a similar number and pattern of errors regardless of pair type or presentation type. We suggest that the lack of significant findings in our study was a result of our small sample size, and that further research is needed to establish whether dyslexic and control children differ in their method of encoding items in verbal short term memory.