Recently, a family of homeobox genes involved in brain and craniofacial development was identified. In light of this genetic background, we hypothesized that some functional characteristics of human brain (hand skill, cognition) may be linked to some structural characteristics of human skull (e. g., craniofacial width) in humans. Hand preference was assessed by Oldfield's Handedness Questionnaire. Hand skill was measured by Peg Moving Task. Face width was measured from the anteroposterior cephalograms (x-ray) using right (R) and left (L) zygomatic points. Intelligence "g" was assessed by Cattell's Culture Fair Intelligence Test; the perceptual-verbal ability was assessed by "Finding A's Test"; the spatial ability was assessed by the mental rotation task, in right- and left-handed men and women. The percentages of right-,left-, and mixed-faced subjects were close to those found for paw preference in cats. Women tended to be more right-faced (R - L > 0) and less left- faced (R - L < 0) than men, who tended to be more left- faced and less right-faced than women. R - L face width inversely correlated with L - R PMT (peg moving time) in left- handers; there was a direct relation between these variables in right-handers. Cattell IQ linearly increased with R - L face width in left- handers, negatively correlated in right-handed men and women. Verbal ability inversely related to R - L face width in right- and left- handed men, but directly correlated in right- handed women. The number of correct response on mental-rotation task positively and linearly correlated with R - L face width in left-handers and right- handed women. It was concluded that the structural-functional coupling revealed in the present work may have its origins in parallel development of the craniofacial skeleton and brain under the influence of homeobox genes.