The present study was carried out to find out whether regular administration of cloze test improved the students' knowledge of grammar more than the multiple choice one. Subjects participating in this study were 84 Iranian pre-university students of Allameh-Gotb-e Ravandi University, aged between 18 and 35 and enrolled in a grammar course. To investigate the research, the candidates were screened through a NELSON test. This test enabled the researchers to have homogeneous subjects; therefore, among 84 subjects, 75 students were selected to take part in this research. Among this number of students, 39 students were to be in one group and the other 36 were the other group members. The two groups were provided with 10 consecutive weeks, each session 100 minutes long. Every week one experimental group was treated by a 25-item cloze test of grammar while the other experimental group was provided with a 25-item multiple choice one. The time allowed for each multiple choice test was about 25 minutes. Each week the grammatical point which had been taught the previous session, was tested. After the treatment, a 60-item test consisting of a 30-item cloze passage and a 30-item multiple-choice grammar section were administered to find out whether there was any difference in the grammatical knowledge of both experimental groups. The analysis of data indicated that there was no significant difference between the average performance of the two groups. In another words, the findings proved that the cloze procedure did not improve the subjects' knowledge of grammar any more than the multiple-choice grammar test did.