The authors investigated the reliability of a body knowledge tool utilizing a test-retest design. Subjects were 86 children, 6-12 years old, who attended a rural elementary school in Southwestern Pennsylvania. Data consisted of children's perceptions of internal body parts as measured by responses to the Modified Gellert Body Knowledge Interview. A second set of body knowledge scores was generated when the children were retested 7 days after the original testing. A significant relationship (r = .70, p < .001) was found between the two body knowledge scores. Analysis of the body knowledge scores indicated that children's knowledge of the human body increases with age. A significant correlation was also found between grade level and body knowledge scores. Gender was unrelated to level of performance on the body knowledge instrument. The most commonly listed body parts were heart, brain, bones, veins, blood, and muscles. The reliability established for this instrument indicates that studies could be pursued to validate its effectiveness in determining school-age children's body knowledge.