The purpose of this study is to examine whether learning a known-and-unknown word combination is superior in terms of retention and retrieval of meaning to learning a single unknown word. The term "combination" in this study means a two-word collocation of a familiar word and a word that is new to the participants. Following the results of Kasahara (2010), the study predicted that attaching a known word to a word to be remembered could help learners to retain and retrieve the meaning of the target word. Sixty-six university students took part in the experiment. They were divided into two groups with the same vocabulary size. One group was told to remember the 20 target words; the other was told to remember the 20 two-word collocations (known words + the target words). Each group was asked to define the meaning of the target words or the collocations respectively in immediate and delayed tests. The results of both tests revealed that the collocations allowed significantly better retention and retrieval of the meanings than the single target words. (Contains 2 figures and 2 tables.)