Doxycycline (Dox) controlled Tet systems provide a powerful and commonly used method for functional studies on the consequences of gene overexpression/downregulation. However, whereas Dox delivery in tissue culture in vitro is relatively simple, the situation in vivo is more complex. Several methods of Dox delivery in vivo have been described-e.g., in drinking water containing alcohol, in drinking water containing various concentrations of sucrose, and in feed. Unfortunately there are no reports directly comparing the advantages and disadvantages of these diverse methods, and there is no generally accepted standard. We therefore compared four non-invasive methods of Dox delivery in vivo-in drinking water, by gavage, as a jelly, and in standard feed. To assess the delivery of Dox by these methods, we used a subcutaneous xenograft model based on colorectal carcinoma cells engineered for Dox-inducible expression of an activated mutant of c-Src and the luciferase reporter gene. Our results indicate that feed represents the most favorable method of Dox administration.