We explored the role of age of acquisition in picture naming with a group of unbalanced, late bilinguals and a group of monolinguals. We hypothesised that we would find effects of L2 age of acquisition on L2 picture naming performance in late bilinguals if the age of acquisition effects we and others have found in L1 picture naming are not limited to language capabilities acquired early in the lifespan. In Experiment 1, late bilingual Spanish–English participants named a large set of pictures in their L2 (English). The most important predictor of naming ability was L2 age of acquisition. In Experiment 2, monolingual English participants named the same pictures. Naming speed was predicted by L1 age of acquisition. Hence speed of picture naming in a given language was predicted by age of acquisition values for that language, that is, L2 values predicted L2 performance (Experiment 1) and L1 values predicted L1 performance (Experiment 2). On the basis of these results we conclude that age of acquisition effects are not restricted to items learned before any putative critical period, but should be observed for items learned at any age. That is, age of acquisition effects are more likely to be due to the relative order in which items are acquired within a language.