This paper disaggregates the term "biodiversity" into components (landscapes, ecosystems, communities, species/populations, and genes) and attributes (structure, composition, and function). It then disaggrgates "logging" by detailing the vast range of activities subsumed under the term including variation of logging intensities, logging methods, collateral damage, and silvicultural approaches. Using the richness present in both terms, a framework for considering the impacts of logging and other forest management activities on the various components and attributes of biodiversity is presented. This framework is, in turn, used to evaluate the extensive literature covering different studies of logging in tropical forests. This paper does not conclude with uncritical support for sustainable forest managmement of timber as a conservation strategy. Such an endorsement is unwarranted given widespread illegal logging in the tropics, widespread frontier logging and logging of areas of high priority for biodiversity protection, the persistence of poor logging practices despite substantial efforts in research and training, and the generally slow rate at which most loggers are transforming themselves from timber exploiters into forest managers. Rather the authors assert, from a biodiversity maintenance perspective, that natural forest management is preferable to virtually all land-use practices other than complete protection.