Two hundred and forty mildly myopic schoolchildren aged 9-11 years were randomly allocated to three treatment groups and the progression of myopia was followed-up for three years. The treatment groups were: (1) minus lenses with full correction for continuous use (the reference group), (2) minus lenses with full correction to be used for distant vision only, and (3) bifocal lenses with +1.75 D addition. Three-year refraction values were received from 237 children. The differences in the increases of the spherical equivalents were not statistically significant in the right eye, but in the left eye the change in the distant use group was significantly higher (-1.87 D) than in the continuous use group (-1.46 D) (p = 0.02, Student's t test). There were no differences between the groups in regard to school achievement, accidents, or satisfaction with glasses. In all three groups the more the daily close work done by the children the faster was the rate of myopic progression (right eye: r = 0.253, p = 0.0001, left eye: r = 0.267, p = 0.0001). Myopic progression did not correlate positively with accommodation, but the shorter the average reading distance of the follow-up time the faster was the myopic progression (right eye: r = 0.222, p = 0.0001, left eye: r = 0.255, p = 0.001). It seems that myopic progression is connected with much use of the eyes in reading and close work and with short reading distance but that progression cannot be reduced by diminishing accommodation with bifocals or by reading without spectacles.