By 1957, it was clear that genetic changes in the virulence of myxoma virus were the major factor in the changing epidemiology of the disease and made possible changes in the genetic resistance of rabbits. However, it was also clear to me that although myxomatosis was a superb ‘natural experiment’ in evolution, myxoma virus was not a good virus with which to study viral genetics. I therefore initiated work on this with a survey of various marker properties of several orthopoxviruses, mostly different strains of vaccinia and cowpox viruses, which were ideal agents for laboratory investigations (Fenner, 1958). Following selection of two with contrasting characters, I demonstrated, for the first time, intramolecular genetic recombination between animal viruses (Fenner, 1959). Travelling across USA on study leave in 1957 and discussing this work, as mentioned earlier, I was quickly convinced by Salvador Luria, then at Urbana, Illinois, that it would be impossible to delve deeply into mechanisms of recombination if I used two different wild type viruses. As with bacterial viruses, which he had studied, it was essential to use a suite of viruses derived from a single parent. Fortunately, I had such material on hand, the white pock mutants of rabbitpox virus (Gemmell and Fenner, 1960), and initiated work on these which later extended to the use of host-cell dependent and temperature-sensitive conditional lethal mutants (reviews, Fenner and Sambrook, 1964; Fenner, 1970).