It is commonplace to assert that the Japanese university entrance examinations dominate the whole educational system and practice in Japan. There are critics who even go so far as to claim that the exams are so powerful that if the exams were to change, innovation in education would automatically follow therefrom. Despite a large number of assertions and claims. however. surprisingly little empirical research has been conducted to look into the nature of the impact to date. The purpose of the present thesis is to report an empirical exploration into the washback effects of the Japanese university entrance examinations of English on pre-college level education. In the present research. an attempt was made to test the validity of various predictions derived from general public opinions by directly observing the classrooms. as well as gathering information from teacher interviews. The samples were taken from regular and special exam preparatory classes of high schools, and the summer and winter intensive courses of yobiko (a special exam preparatory institution). The results provided very little evidence that the predicted types of washback were present either on the exercise types employed in the classrooms, contents of the lessons, or teaching methods. Other factors than the exams, such as teachers' beliefs, their familiarity with teaching methods, school cultures, and availability of support-materials, seemed to be playing a greater role in determining what happens in the classroom. On the basis of the results, sets of suggestions were advanced for teachers, test-constructors, material developers, and future researchers. The thesis concludes with a progressive re-conceptualisation of the notion of washback in the light of the research findings.