Purpose: The word learning of preschool-age children with developmental language disorder (DLD) is improved when spaced retrieval practice is incorporated into the learning sessions. In this preregistered study, we compared two types of spacing--an expanding retrieval practice schedule and an equally spaced schedule--to determine if one of these approaches yields better word learning outcomes for the children. Method: Fourteen children with DLD aged 4-5 years and 14 same-age children with typical language development (TD) learned eight novel nouns over two sessions. Spacing for half of the novel words was expanded gradually during learning; for the remaining novel words, greater spacing remained at the same level throughout learning. Immediately after the second session and 1 week later, the children's recall of the words was tested. Results: The children with TD recalled more novel words than the children with DLD, although this difference could be accounted for by differences in the children's standardized receptive vocabulary test scores. The two groups were similar in their ability to retain the words over 1 week. Initially, the shorter spacing in the expanding schedule resulted in greater retrieval success than the corresponding (longer spaced) retrieval trials in the equally spaced schedule. These early shorter spaced trials also seemed to benefit retrieval of the trials with greater spacing that immediately followed. However, as the learning period progressed, the accuracy levels for the two conditions converged and were likewise similar during final testing. Conclusion: We need a greater understanding of how and when short spacing can be helpful to children's word learning, with the recognition that early gains might give a misleading picture of the benefits that short spacing can provide to longer term retention.