This study investigated the relationship between working memory and language in typically developing young children. The aim was to gain a better understanding of language development, in particular, the involvement of visual and verbal short-term memory in language acquisition and its influence on vocabulary size. It explored possible underlying causes of why some children have problems in the process of learning to talk, whereas other children acquire language easily. A total of 51 New Zealand English speaking children aged two to five completed a battery of assessments measuring receptive and expressive vocabulary and visual and verbal short-term memory. The standardized tests administered included the Receptive One Word Picture Vocabulary Test (Brownell, 2000b), the Expressive One Word Picture Vocabulary Test (Brownell, 2000a), the Visual Patterns Test (Stokes, Klee, Cruickshank, & Pleass, 2009), and the Test of Early Nonword Repetition (Stokes & Klee, 2009a). Receptive vocabulary knowledge was strongly associated with visual (r = .75) and verbal (r = .60) short-term memory performance and age (r = .72). The relationship of expressive vocabulary to visual short-term memory (r = .80) was stronger than to verbal short-term memory (r = .62) but significant for both and also for age (r= .83). Significant unique predictors for expressive vocabulary were age (R2 change = .60) as well as visual (R2 change = .04) and verbal (R2 change = .04) short-term memory. However, age appeared to be the only unique predictor for receptive vocabulary (R2 change = .54). In addition, the findings suggested that visual and verbal short-term memory increases as children get older. Hence, the Visual Patterns Test and Test of Early Nonword Repetition seem to be good predictors, over and above age, of expressive vocabulary knowledge.