Instructional media is an important instrument for teaching and learning. Computer assisted instruction can be used to help each individual student achieve educational learning and to satisfy personalized learning need by requiring an active participation in the process of education. The researcher has led the good points of the computer assisted instruction, and developed computer assisted instruction, for use as a teaching and learning instructional media. The purposes of this research were 1) to construct a computer assisted instruction for fundamental English on " Eat Well and Stay Healthy " for Mathayomsuksa 2 students. 2) to compare learning achievement between the experimental group by using the computer assisted instruction and the control group by using the conventional method. The samples were 80 Mathayomsuksa 2 students of Bunharnjamsaiwittaya 5 School, Songpeenong District, Suphanburi Province during the second semester of 2002 academic year. They were selected by using single stage cluster sampling technique and assigned as an experimental group (n=40) using computer assisted instruction, and a control group (n=40), using the conventional method. The treatment duration was eight periods, 50 minutes each. The instrument consisted of : 1)Computer Assisted Instruction for Fundamental English on " Eat Well and Stay Healthy " for Mathayomsuksa 2 Students, 2)Conventional Lesson Plans, 3) An achievement test. The data were statistically analyzed by using percentage , arithmetic mean ( ), standard deviation (SD) and t-test using SPSS for windows. The results of the study were as the following : 1. The efficiency of Computer Assisted Instruction for Fundamental English on " Eat Well and Stay Healthy " for Mathayomsuksa 2 Students met the efficient standard criterion of 84.0 / 76.22, which was higher than the selected efficient standard criterion of 75/75 2. The comparison of the learning achievement of the experimental group after using Computer Assisted Instruction ( =23.73) was significantly higher than that of the conventional method group ( =21.55) at the 0.01 level