At 6 years of age, a boy with bilateral sensorineural deafness, lateral displacement of inner canthi, a bulbous nasal tip, synophrys, and cryptorchidism was clinically diagnosed as having Waardenburg's syndrome type I (WS-1). In addition, he had a lumbar spina bifida with hydrocephalus shunted on the second day of life and severe mental retardation with a head circumference at the fifth percentile. Neither parent showed signs of WS-1, and the family history was negative. Because of the WS-1 features, attention was focused on the PAX3 location in 2q, at which time a de novo paracentric inversion of 2q23–q37.1 was noted. Subsequent high-resolution chromosome analysis 8 years later indicated a complex rearrangement involving regions 2q31–q35 and 2q13–q21. Whole chromosome painting and high-resolution comparative genomic hybridization yielded negative results for any translocation, duplication, or deletion of any chromosome segments. Sequencing of the PAX3 gene yielded no detectable mutation. Fluorescent in situ hybridization (FISH) studies with human BAC clones revealed five breakpoints in chromosome 2q resulting in two paracentric inversions and one insertion, the karyotype being interpreted as 46,XY,der(2)inv(2)(q13q21)inv(2)(q21q24.2)ins(2)(q24.2q33q35). In this extremely rare chromosomal rearrangement, the FISH result showed a breakpoint at 2q35 being proximal to and without involvement of the PAX3 gene. While further studies continue, possible interpretations include involvement of a regulatory gene(s) for PAX 3 and other genes at the other breakpoints related causally to the spina bifida and mental retardation.