WELLS,1in his classical review of 250 malignant neoplasms present at birth, expressed the belief that the induction of congenital neoplasia "must differ in some fundamental way from the ordinary sorts of cancer, which require so long a period for development, and this difference may throw some light on the mystery of malignancy." Reported here is a further exploration into the origins of congenital neoplasia through a study of the US national mortality from cancer within the first 28 days of life over a five-year interval. Methods The National Vital Statistics Division, United States Public Health Service, provided us with 21,929 copies of death certificates (international list numbers 140-205) for all children under 15 years of age who died of cancer in the United States during the five-year period, 1960 to 1964. Excluded from analysis were 270 children, including five neonates, who died with malignant histiocytosis. Of the remaining