The data for this study came from the National Longitudinal Survey of Youth (NLSY). The NLSY began in 1979 as a household probability sample of 11,406 U.S. adolescents between the ages of 14 and 21 (a 90-percent response rate). Blacks, Hispanics, and lower socioeconomic groups were oversampled. The sample was reinterviewed yearly between 1979 and 1986. In 1980, a set of indepth delinquency questions was administered as part of the overall interview. This study used the subset of NLSY respondents who also had siblings in the study (5,863). The dependent variable was a constructed delinquency scale with items typical of self- report delinquency scales that involved a range of minor to more serious acts. The independent variables used in the regression analyses were measures of relationships between siblings within the family environment. The study compared sibling correlations for birth orders one to four in family sizes of two, three, and four siblings. The correlational results largely supported a shared (environment and/or heredity) model of familial resemblance rather than one based on unshared environmental influences, except for the brothers. The study demonstrates that shared familial influences exist and that they can account for a substantial part of the total variation in self-reported delinquency. Results, for male siblings, however, place a caveat on this conclusion, because males in birth orders one to three were less alike than males in adjacent birth orders. Additional data are needed to decide whether or not an unshared environmental model applies to brothers or whether the result for the male sibling groups of size three was accidental. 1 figure, 4 tables, and 30 references