A random sample of 520 inmates from a penitentiary population of 1074 are classified into the typology groups according to procedures developed by Megargee, et al. (1977). A variety of demographic and criminal history variables are analyzed to determine subject characteristics for each of the ten typology groups. Institutional adjustment is assessed by data on inmate behavior routinely collected and recorded on each inmate by prison staff. Analysis of the data shows that: (1) the Megargee typology could classify inmates into ten groups not proportionately different than those in a medium-security institution, (2) the groups did not differ significantly from one another on age, race, or offense type, and (3) contrary to predictions by Megargee et al. For a medium-security prison, the typology groupings are not effective in predicting which inmates would be antisocial or aggressive in a penitentiary setting. (Publisher abstract)