This relationship study aimed at investigating the effects of the mnemonic keyword technique on recall of vocabulary items in comparison to a dictionary-meaning supplied group. The study was conducted at Middle East Technical University, Department of Basic English. The participants were thirty-two Pre-Intermediate level students in two intact classes. The research questions focused on the significant difference, if any, in short-term and long-term recall of vocabulary items with respect to the implementatidn of teacher-supplied mnemonic keywords and teacher-supplied dictionary definitions in the instruction of vocabulary items. The study was conducted on two Pre-Intermediate level classes and each class was randomly assigned to be treated with one of the two techniques. Thus, two groups were formed: "Pre-Intermediate Keyword Group" and "PreIntermediate Dictionary Definition Group". To determine the words to be used in the experiment, a checklist of sixty words was given to a randomly selected class of sixteen Pre-Intermediate students who did not further participate in the experiment. These sixty vocabulary items were chosen by the researcher with the help of the class instructors of the experiment and control groups. The guidelines for the selection of these words were the coursebooks and materials dealt with at the PreIntermediate level at the Department of Basic English, Middle East Technical University. The words were nouns which were similar to those that the students would learn at that level. However, the list was based on the assumption that the students were not familiar with these items. The 40 items that were most recognised by the sixteen students were eliminated, leaving 20 items to be used in the experimental process. The researcher dealt with the instruction of the vocabulary items for both the keyword group and the dictionary-definition group. The keyword group received copies of a list of the twenty vocabulary items with their Turkish equivalents and key words for each one. The dictionary-definition group received a copy of the same vocabulary items supplied with dictionary definitions along with an example sentence for each item. Both classes were then subjected to an immediate recall test designed as a matching test dealing with the vocabulary items. After a period of three weeks the students were given the same matching test in a different format to test long-term retention. T-test scores were calculated to analyse which technique was more effective on students' short-term and long-term retention of the target vocabulary through the comparison of the means of each group. The immediate posttest results found high raw scores in both groups, which did not differ significantly from each other. The results also support the finding that neither technique can be considered superior to the other in promoting vocabulary retention at the pre-intermediate level. In fact, both techniques were unsuccessful in promoting the long-term retention of most of the vocabulary items tested.