Given that learning by doing has always been recommended, it was anticipated that learning certain topics in Science class can be better learned while doing activities in physical education class. To verify this, a group of sixty 7th graders was divided into two equal subgroups based on the girls’ pretest scores. One group was taught science in a regular classroom, while the other was taught the same topics through the activities in the physical education class. After eleven sessions both groups were tested on the subject as well as on attention using instruments that were assumed to be valid and reliable. Data analysis yielded the expected results as the experimental group showed significantly better scores on all measures.