本研究旨在探討「傳統字義生詞教學法」和「構詞法輔以常用詞根之生詞教學法」對台灣的法語學習者學習和記憶法語生詞的差異，希望能在生詞教學上，提供另一個不同於傳統生詞教學而且更為有效的方法。 本研究的主要發現如下：1)接受「構詞法輔以常用詞根之生詞教學法」的受試者(實驗組)比接受「傳統字義生詞教學法」的受試者(對照組)在生詞的短期記憶和長期記憶上，均有較佳的表現。2)實驗組的辨別字義和拼字能力也比對照組佳。3)在字彙量、短期記憶與長期記憶的相關性上，短期記憶較佳者，也有較好的長期記憶表現。 / The purpose of the study is to investigate the different effects of the traditional definition-based teaching method and the morphological approach via commonly-used roots on Taiwanese French learners’ vocabulary acquisition and memorization. It is hoped that the results of the study can provide an alternative to traditional vocabulary teaching in French. First of all, an on-line questionnaire which investigated the teaching situation of French vocabulary in Northern Taiwan was administered. Then, the subjects in this study involve 115 nonnative French majors at National Central University and Fu Jen University, including 57 seniors and 58 sophomores. The seniors were divided into experimental and the control groups, while the sophomores were also divided into experimental and the control groups. The experimental groups received the morphological approach via commonly-used roots, and the control groups received the traditional definition-based teaching method. In the beginning, the experimental and control groups took the same French vocabulary pre-test to examine their vocabulary size. Then, after receiving different instruction, the subjects were immediately asked to take post-test 1, which investigated their short-term memory for words. Two weeks later, post-test 2 was conducted to check the subjects’ long-term memory for words. The results of the study reveal that the experimental groups have better short-term memory for words and long-term memory for words than the control groups, regardless of the subjects’ language proficiency. With regard to the effect of learning new words, both the experimental groups and the control groups made progress after receiving the instruction. After two weeks, the experimental groups showed no regression on the vocabulary post-test. Furthermore, the experimental groups performed better than the control groups on both word spelling and word meaning, and in general, the subjects’ short-term memory for words was correlated with their long-term memory for words.