Objective: To study the differences in children's psychiatric symptoms and child mental health service use at two time points: 1989 and 1999. Method: Two cross-sectional representative samples of 8- to 9-year-old children from southern Finland were compared. The 1989 sample consisted of 985 children, of whom 95% participated, and the 1999 sample consisted of 962 children, of whom 86% participated. Information was gathered from parents and teachers using Rutter's questionnaires and other related determinants of service use and from children using the Child Depression Inventory. The sampling, procedure, and methods were similar at both time points. Results: The overall rate of children's problems assessed by parents and teachers had not increased during the period 1989 to 1999. Boys had fewer psychiatric symptoms in 1999 than in 1989, whereas no clear change had occurred in girls' symptoms, except that, according to parents, girls in 1999 had more hyperactive symptoms. However, children themselves reported more depressive symptoms in the 1999 than in the 1989 sample. In 1989, 2.3% and in 1999, 5.3% of children had used child mental health services. The increase in service use among girls was fourfold. Parental evaluations of child psychopathology and teacher evaluations whether the child was psychologically healthy were the strongest determinants for referral at both time points. Parents preferred to seek help for their children's problems from teachers, school nurses, and school psychologists rather than from specialized child psychiatric services. Conclusions: There has been an increase in mental health service use especially among girls as well as a convergence of symptom levels by gender. It is important to develop child psychiatric services that are as close to the child's living environment as possible to further reduce the threshold for seeking help and to promote early detection and intervention. J. Am. Acad. Child Adolesc. Psychiatry, 2004;43(9):1134-1145. Key Words: mental health, service use, time trend.