The records of 912 men and 188 women residents of Rome and vicinity who had received at least 1 pardon (conditional or unconditional) were examined. They represented a random sample of approximately 1 percent of the total number of persons processed in the police records for the year and 4 percent of those who had been given pardons. The records were analyzed by sex and age and by number and type of previous offenses and number and type of release or pardon. Of the 1,100, 31.2 percent, or 344 persons, recidivated after receiving a pardon, with the proportion of women recidivating being 1 in 5 while for men it was 1 in 3. For the same period, the rate of recidivism in the same jurisdiction, for all offenders, was higher -- 35.2 percent for men and 25.1 percent for women. Ages of those in the sample varied between 20 and 70, but two-thirds were over 50 years of age and the largest group (25.8 percent) were 50-59 years old. However, the age when each received the first pardon was much lower (two-thirds were first pardoned before reaching 30). The chance of committing another crime after being pardoned was much lower for those who received pardons late in life than for those who received them earlier. The most significant variable in predicting recidivism among those who receive pardons was found to be that of prior criminal activity. Although the sample was fairly evenly divided between first offenders and repeat offenders, only 125 first offenders recidivated following receipt of a pardon, whereas 219 of the repeat offenders recidivated. It is recommended that legislators limit the benefits of pardons to first offenders. Note is taken, however, that no cause and effect situation has been proved regarding granting of pardon to habitual offenders and their subsequent recidivism. A brief review of a research project on the effects of clemency in Rome is also presented. A sample report form used in the pardon study is appended.