Efficacious vaccination against fast growing pathogens results in a rapid, secondary immune response on natural infection; this provides protection to the vaccinated individual in the race between developing effective immunity and the rapid multiplication of the pathogen. In certain chronic diseases, due to slow growing pathogens, cellmediated immunity alone can contain the infection, and yet an antibody response is sometimes induced, at the expense of the cell-mediated response, upon natural infection. Such situations arise in leprosy and the leishmaniases and most probably in tuberculosis. AIDS and syphilis. In these cases, the purpose of vaccination must be to ensure that a stable, protective, cell-mediated immune response is inevitably induced upon natural infection. We believe we have developed a general strategy for causing a pathogen-specific imprint upon the immune system so that a stable, protective, cell-mediated response is inevitably induced in all individuals upon natural infection. BALBIc mice are ≪susceptible ≫ to Leishmania major in the sense that they mount a non-protective antibody response on substantial infection, and consequently suffer chronic and progressive disease. We have demonstrated that infection with low doses of parasites induces only cellular immunity, and establishes the desired imprint. Mice exposed to low doses and challenged some months later with a substantial, normally pathogenic dose of parasites, mount a stable, protective, cell-mediated response and the vaccinated «susceptible» mice withstand the infection. We have recently managed to achieve a similar lock of the immune response of BALB/c mice to BCG into a cell-mediated mode by low-dose exposure. We are thus in a position to examine whether such a lock confers resistance to a normally lethal challenge of Mycobacterium tuberculosis. Genetic diversity of people and animals presents a problem for the development of universally efficacious vaccination procedures, as this diversity tneans individuals will in general mount different immune responses upon exposure to standard vaccination. The observations of others as well as our own support the generalization that an antigen, capable of inducing antibody when given at a certain dose, will inevitably induce an exclusive cell-mediated response upon challenge with an appropriate lower dose. We suggest it is possible to choose a very low dose of an antigen that does not induce antibody in any individual but either induces cell-mediated immunity or is below the threshold dose of antigen required to do so. If the antigen is chosen to be a slowly growing microorganism, such as BCG, the very low inoculum is expected to grow unless restrained by a cell-mediated response; a very low and appropriate dose will thus in time induce cell-mediated immunity in all individuals and hopefully establish the required imprint in all individuals.