The purpose of this study was to determine whether there was a significant difference between students who were taught to write descriptive paragraphs using "semantic mapping strategies" and taught using conventional strategies for tenth-grade students MA Nurul Huda Sukaraja. This research uses experimental method, experimental queasy design. The population in this research is 103 students, and the sample of 50 students is taken by using purposive sampling, that is 25 students of experiment group and 25 students of a control group. To analyze data using the independent t-test, and the calculation results show that t-obtained is higher than t-table (4.57> 2.0106). Thus, the alternative hypothesis (Ho) There is no significant difference between students taught using semantic mapping strategies and students taught using conventional strategies to tenth-grade. (Ha) There is a significant difference between students taught using semantic mapping strategies and students taught using conventional strategies to tenth-grade. That means that Ha is a significant difference between the experimental group and the control group.