Objective      To explore whether brain volume ratios such as gray-to-white matter volume ratio

gray matter-to-global volume ratio

and white matter-to-global volume ratio are stable endophenotypes which can be used as key endophenotype indexes of children and adolescents’ brain development.       Methods      Using cross-sectional design

totally 58 pairs of twins aged 12 to 18 years were recruited according to the school registration information. DNA was extracted from venous blood samples to identify zygosity. Intelligence quotients (IQ) of participants were measured with C

-Wechsler intelligence scale for children (C-WISC). With 3.0T MRI (GE)

we examined all 58 twin pairs to obtain brain structure MR image for each person.       Results      (1) The 58 twin pairs included 31 monozygotic twin pairs (MZ group) and 27 dizygotic twin pairs (DZ group). (2) Gray matter volume

white matter volume

cerebrospinal fluid volume

global cerebral volume

gray-to-white matter volume ratio

gray matter-to-global volume ratio and white matter-to-global volume ratio were positively associated (

P

0.01) in the MZ group. Cerebrospinal fluid volume and gray matter-to-global volume ratio showed positive correlation (

P

0.05) in the DZ group while other 5 indexes showed non-significant difference (

P

0.05). Heritabilities estimated by Holzinger’s method of gray-to-white matter volume ratio

gray matter-to-global volume ratio

and white matter-to-global volume ratio were 0.828

0.596

and 0.741

respectively.       Conclusion      The heritability of gray-to-white matter volume ratio is highest in all volume ratios and is not affected by cerebrospinal fluid volume. Thus

gray-to-white matter volume ratio may be used as a stable endophenotype indicator for brain development research of children and adolescents.