Although numerous attempts had been made to identify recombination in bacteria, it is only when Lederberg and Tatum started to use strains with multiple auxotrophic requirements that reversion to prototrophy was reduced sufficiently to allow the unambiguous detection of recombinants. The original K-12 strain carries the F episome, a plasmid that encodes all the function required for DNA transfer between cells. Fortunately, many of the strains derived from K-12 have lost the F and can receive DNA from a donor strain. The difference between this system of DNA exchange, later called conjugation, and transformation was clearly demonstrated. Attempts to map mutations on the E. coli chromosome were not very successful, essentially because the recombination frequency was extremely low. The isolation of Hfr male strains (for high frequency of recombination) by Cavalli-Sforza and Hayes was a tremendous improvement for chromosome mapping. Indeed, a reproducible recombination frequency could now be obtained. Using a blender to separate mating pairs, Wollman and Jacob could determine the minimum contact time necessary for a given marker to be transferred from a donor to a recipient. When matings involved one strain lysogenic for phage lambda, there was a striking difference depending on which strain was lysogenic. When the recipient strain was lysogenic, nothing happened to the mating cells. In contrast, when the donor strain was lysogenic, the recipient started to produce viral particle with a high efficiency. In this system, more than 50% of the recipient could be induced to produce phage upon mating. Finally, mating pairs were sufficiently frequent to be observed in the electron microscope and a small cellular bridge was detected between the recipient and donor cells.