Criminal opportunity theory suggests that community economic deprivation has two countervailing effects on property crime; it causes strain and disorganization that may encourage some individuals to offend, but it also simultaneously lessens opportunities to engage in property crime by reducing the supply of worthwhile targets in an area. In testing this theory, data were obtained from the records of the municipal police departments of Austin and Seattle, as well as from the 1990 U.S. Census of the Population. The units of analysis were census tracts with populations greater than 1,000 for Austin and Seattle (n=136 and n=117, respectively). A variety of crime data for all years after 1995 were available from the police in both cities. The analysis focused on data related to property crime, particularly motor vehicle theft and burglary rates, both averaged for 1996 and 1997 and expressed in terms of crimes per 100,000 persons. Economic deprivation was measured by using the official poverty line designated by the U.S. Census Bureau. Multivariate regression analyses for burglary and motor vehicle theft rates showed a curvilinear relationship between levels of neighborhood deprivation and property crime in both cities. The positive effect of deprivation on property crime was stronger in the poorer neighborhoods (where the strain/disorganization side dominated) than it was in the more affluent neighborhoods (where the supply/net gain side significantly suppressed the strain/disorganization effect). Although the criminal opportunity saturation thesis is a reasonable explanation for the curvilinear relationship between poverty and property crime, other explanations cannot be ruled out. It is possible, for example, that this curvilinear relationship is a product of differences in reporting behavior between individuals in very poor neighborhoods and those in relatively affluent neighborhoods. Very poor neighborhoods may appear to have less property crime than expected because residents are more likely to abstain from reporting victimizations to the police. Implications are drawn for research and policy related to property crime. 1 table and 50 references