Background: Grammatical morphology, particularly the use of the past tense in English, has been identified as a behavioural marker for specific language impairment. However, there is a dearth of instruments that assess key verb features, are short and easy to administer for clinical use, and have normative data for the UK population. Aims: The present study aimed to develop a short version of a past-tense task (named the PTT-20) and to provide UK normative data for a large control sample of primary school-age children. Methods & Procedures: The task consisted of 20 items taken from a well-known 52-item task developed by Marchman, Wulfeck, and Ellis Weismer in 1999, retaining the key features and variety of verbs of the original design. It was administered to 424 typically developing children aged 5;0 to 11;6 years. Outcomes & Results: The short version reduced administration time by two-thirds to around 5-8 minutes. Normative data revealed developmental progression in performance throughout the primary school years, with the task exhibiting ceiling effects in 29% of children by age 11 years. Expected and impaired performance indicators are presented by age band, that is, 5, 6, 7, 8, 9, 10, and 11 years. Comparisons between identified children with specific language impairment and typically developing children revealed significant differences in performance. The PTT-20 correlated highly with the full 52-item task and had good internal consistency. Conclusions & Implications: The PTT-20 can be a useful tool in the identification and assessment of young school-age children with suspected language difficulties. It can also be used in research and practice as a benchmark against which to compare the ability of children with identified language impairments. © 2010 Royal College of Speech & Language Therapists.