The Denver Developmental Screening Test (DDST) was initially developed specifically to identify children with mental retardation. However, its use in screening low birth weight and other biologically at-risk infants for motor problems is widespread. In view of the absence of biologically vulnerable children in the DDST standardization sample, and the limited validational support available for its use with children under 30 months of age, the purpose of this study was to investigate the effectiveness of the DDST in screening 62 Neonatal Intensive Care Unit graduates during infancy. Categorical results (Normal, Questionable, Abnormal), obtained in the home setting by DDST, were compared with results of Neonatal Intensive Care Unit Follow-up Clinic evaluations. There was a marked underselection of the DDST's categorical results in identifying those infants with suspect or abnormal findings on the more detailed clinic evaluations. However, the effectiveness of the screeners' clinical impression ratings (based on observations made during administration of the DDST) in identifying infants with positive criterion results was encouraging. These findings suggest that, although the DDST's categorical results may be of limited value in screening biologically vulnerable infants, administration of the DDST may provide a useful framework for more systematically observing and documenting significant qualitative aspects of an infant's developmental status, particularly for the screener who is less experienced in neuromotor assessment. These observations need to be defined further, and validated.