Skinner’s (1957) pioneering behavioural account of language identified verbal behaviour as fundamentally functional. This account of language has been the basis for effective teaching programmes for children with developmental disabilities (Sundberg & Michael, 2001). However, despite the well documented efficacy of language-training in applied settings, these programmes have been criticised for producing language that is rigid in nature and not readily generative (McEachin, Smith, & Loovas, 1993). It has been postulated that the reason for this inflexibility of language is due to the lack of focus on ‘emergent’ (i.e., untrained or derived) language responding (Luciano, et al., 2009). To address this issue, recent research has seen the integration of Skinner’s Verbal Behaviour theory with Relational Frame Theory (Hayes, Barnes-Holmes & Roche, 2001), a modern behavioural language theory. A body of research now demonstrates the effective facilitation of derived relational responding in children with developmental disabilities (e.g. Barnes-Holmes & Murphy, 2007). The current research sought to add to RFT literature on the relational responding skills of children with developmental disabilities. The studies reported in the current thesis were concerned with testing and training relational responding skills using protocol based on RFT (Rehfeldt & Barnes-Holmes, 2009), with two groups of children. Study 1, presented in Chapter 2 of the current thesis, sought to test and train relational responding in accordance with co-ordination and distinction with seven children with autism. Studies 2, 3 and 4, presented in Chapter 3, were concerned with testing the relational responding skills of five typically-developing children. Specifically, Study 2 tested relational responding in accordance with co-ordination and distinction; Study 3 tested relational responding in accordance with comparison; and Study 4 tested relational responding in accordance with opposition. A secondary objective of the current research was to assess the potential relationship between relational responding skills and verbal ability, as measured by two standard verbal assessments. Thus, the study assessed each participant’s expressive and receptive verbal abilities through the administration of two standardised verbal assessments; the Peabody Picture Vocabulary Test (PPVT-4; Dunn & Dunn, 1997) and the Kaufman Brief Intelligence Test (K-BIT; Kaufman & Kaufman, 2004), respectively. The verbal assessments were conducted with both groups of participants both before and after relational responding testing/training. The overall aim of the research was thus to, 1) test and train relational responding skills using protocol based on RFT (Rehfeldt & Barnes-Holmes, 2009), with children with autism and typically developing children; 2) to investigate whether higher verbal ability scores correlate with higher relational responding skills; 3) to compare the verbal ability and relational responding skill of children with autism to those of typically developing children, and 4) to investigate whether relational responding training resulted in enhanced verbal ability scores. The findings of these studies demonstrate the diversity in verbal ability and relational responding skills among typically-developing children and those with autism. Specifically, participants with autism produced very weak performances on both verbal assessments. While all the children with autism reached criterion during co-ordination relations, only one of the seven participants reached criterion for relational responding in accordance with distinction. In comparison, all of the typically developing participants produced average performances on both verbal assessments and reached criterion for all relational frames tested; co-ordination, distinction, comparison and opposition. These findings, in line with RFT predictions, suggest, that higher verbal ability is correlated with higher relational responding skills. Overall, no significant changes in verbal ability were detected post relational testing/training for any of the participants in the current research. Chapter 4 of the current thesis synthesises the empirical work presented in the preceding chapters and addresses a number of theoretical and clinical issues that arise from this work.