To examine the relationship between the level of recorded crime and the use of custodial sentences by the courts, correlations were calculated for all police force areas in England and Wales between the percentage change in the proportion of offenders awarded immediate custodial sentences from 1989 to 1990 and the percentage increase in recorded crime from 1989 to 1990 and from 1990 to 1991. No statistically significant correlations were found. These findings also held when recorded violent crime, theft, and handling offenses were analyzed separately. Neither were statistically significant relationships found between changes in the level of recorded crime between 1989 and 1990 and changes in the level of cautioning during this period. These findings indicated that cautioning practice and sentencing practice were not related to changes in the level of recorded crime. There were several reasons why it was difficult to demonstrate such a link in the longer term. First, not all crimes recorded by the police are cleared through a caution to or the sentencing of the offender. Secondly, changes in the level of recorded crime can reflect the increased willingness of the public to report crimes; and thirdly, there were many factors extraneous to the criminal justice system that had an effect on the level of recorded crime. 2 tables and 5 references