Memory difficulties in autism are observed mainly on measures like free recall, where test procedures provide no support for memory. When support is present, such as in cued recall, difficulties are less evident. Such observations may explain the mixed findings on source memory in autism. Bennetto, Pennington and Rogers (Child Development, 67, 1816-1835) found increased earlier-list intrusions in a multi-list free-recall paradigm (support absent), yet Farrant, Blades and Boucher (Journal of Autism and Developmental Disorders, 28, 43-50) reported no impairment in identification of who had spoken a particular word at study (support present). We tested the effects on source memory of presence or absence of support for source in participants with Asperger's syndrome. The Asperger participants' overall deficit in source memory was largely eliminated when source was supported at test.