Urban morphology plays a significant role in shaping the spatial distribution of crime. This study takes an environmental criminology perspective on crime and examines how residential burglary is related to two typical morphology features—housing density and composition, which were rarely concerned by previous research. Wuhan, the largest city in central China, was selected as the case study. We first applied a new urban morphology approach to identify the morphology category of each neighborhood based on its housing density and composition. Negative binomial regression models were adopted to evaluate the impacts of morphology factors on the residential burglary at the neighborhood level while controlling for socio-demographic features, transport facilities, housing price and age. Results suggest that both housing composition and density are significantly associated with residential burglary. In particular, one unit increase in Floor Space Index, an indicator of housing density and Ground Space Index, an indicator of housing composition could lead to an 11.9% and 9.1% increase in the incident rate of residential burglary. The ‘block’ and ‘strip’ composition exert more substantial impacts than ‘point’ composition; neighborhoods with ‘high’ and ‘medium’ residences tend to be more dangerous than neighborhoods with ‘low’ residences. Results of this study reveal that communities must be designed with the relationship between risk levels of residential burglary and the ways by which communities are designed in mind. Implications regarding burglary prevention and neighborhood planning practices are discussed.