Provision of multiple-choice (MC) glosses, which combines the advantages of glosses and inferring, has recently gained its share of supporters as a potential technique for enhancing L2 texts and increasing word gain for L2 learners. Upon taking an actual TOEFL, the participants underwent a vocabulary pretest to ensure that the target words were unfamiliar to all of them. Half of the participants received the passage which included L2 MC glosses and half of them received the same passage embedding L1 MC glosses. They were then tested through both immediate and delayed production and recognition posttests. The results indicated that both L1- and L2-gloss groups significantly outperformed in immediate than in delayed testing condition. However, in both conditions, L2 glosses produced significantly higher scores than L1 glosses did suggesting that L2 glosses far outweigh for fostering learning and retaining target words. The findings of the study, therefore, could hold appeal for materials developers, syllabus designers as well as teachers and instructors who are primarily engaged in vocabulary development courses.