Based on personal acquaintances and experience dating back to the early 1950s, the senior author reviews the history of viral therapy of cancer. He points out the difficulties encountered in the treatment of human cancers, as opposed by the highly successful viral therapy of experimentally maintained tumors in laboratory animals, especially that of ascites carcinomas in mice. A detailed account of viral therapy of human tumors with naturally oncolytic viruses follows, emphasizing the first clinical trials with viral oncolysates. The discrepancy between the high success rates, culminating in cures, in the treatment of tumors of laboratory animals, and the moderate results, such as stabilizations of disease, partial responses, very rare complete remissions, and frequent relapses with virally treated human tumors is recognized. The preclinical laboratory testing against established human tumor cell lines that were maintained in tissue cultures for decades, and against human tumors extricated from their natural habitat and grown in xenografts, may not yield valid results predictive of the viral therapy applied against human tumors growing in their natural environment, the human host. Since the recent discovery of the oncosuppressive efficacy of bacteriophages, the colon could be regarded as the battlefield, where incipient tumor cells and bacteriophages vie for dominance. The inner environment of the colon will be the teaching ground providing new knowledge on the value of the anti-tumor efficacy of phage-induced innate anti-tumor immune reactions. Genetically engineered oncolytic viruses are reviewed next. The molecular biology of viral oncolysis is explained in details. Elaborate efforts are presented to elucidate how gene product proteins of oncolytic viruses switch off the oncogenic cascades of cancer cells. The facts strongly support the conclusion that viral therapy of human cancers will remain in the front lines of modern cancer therapeutics. It may be a combination of naturally oncolytic viruses and wild-type viruses rendered oncolytic and harmless by genetic engineering, that will induce complete remissions of human tumors. It may be necessary to co-administer certain chemotherapeutic agents, advanced cancer vaccines, or even immune lymphocytes, and targeted therapeuticals, to ascertain, that remissions induced by the viral agents will remain complete and durable; will co-operate with anti-tumor host immune reactions, and eventually will result in cures of advanced metastatic human cancers.