Two independent laboratories have generated Ahr "null" or "knockout" mice that share some common characteristics but also have distinct phenotypes. In this Commentary, we will discuss our view of the candidate variables that might account for these differences. More importantly, we hope that this discussion can identify important parameters to be assessed by investigators in the process of characterizing their own modified loci. The variables that we have considered include the possibility that different targeting strategies can result in altered products with unsuspected function or that the targeting event itself can alter the function of neighboring genes. Further, genetic background can have an important influence on phenotype, and differences in genome can be introduced during derivation by the type of embryonic stem cells used and by the random segregation of parental genes in the F2 generation of line propagation. In addition, phenotype may be acutely sensitive to environmental variables, such as pathogen and chemical exposure and stress introduced by crowding and disease. Finally, we discuss approaches to resolving differences between null mice and propose a partial solution, the institution of a repository for detailed information on targeted alleles that may not typically be allowed in today's "fast paced" scientific publications.