In a series of animal and in vitro eyperiments on the virulency of double-resistant tubercle bacilli and in clinical and statistical observations concerning the virulency, the author obtained the following results. 1. In the virulency tests of tubercle bacilli by subcutaneous injection to guinea pigs the virulency of each doubleresistant tubercle bacilli of H(37)Rv strain for SM plus INH and PAS plus INH, the effect is represented by the virulency of INH-resistant bacilli, showing a marked diminution of the virulency. Moreover, the degree of such a decrease is in a proportion to the resistance against INH. 2. Likewise the virulency of SH plus INH-tow drug resistant bacilli isolated from patient, resembling closely to that of H(37)Rv-(SM+INH resistant) strain, is markedly diminished. Similar result is obtained by the bacilli isolated from those cultured in the medium containing SM. This suggests the existence of a high proportion of true-doubleresistant bacilli, and the virulency of this-true double-resistant bacilli seems to play a main role in the diminution of the virulency. 3. The catalase activity of the double-resistant bacilli, including the INH-resistant, is weakened or obliterated according to the degree of the resistance against INH, and consequently a close correlation can be recognized between the activity and the virulency. 4. Viewing the changes of symptoms in the so-called two-drug resistant cases clinically and statistically, their clinical progress is far better than that of SM-single resistant cases, and just as in the case of animal experiments this seems to be due to a high proportion of the true double-resistant bacilli being contained in the strain. In other words, it suggests that the virulency of the double-resistant bacilli is also somewhat diminished even in the human body. 5. Therefore, it is clinically quite important to know accurately the existence of the true-double-resistant bacilli that may be contained in the so-called two-drug resistant strain of patients, and the author advocates that the "crossculture test" devised by himself is well suited for such a purpose. 6, The author has been able to verify the existence of a high proportion of the true double-resistant bacilli in the so-called two-drug resistant strain of patients by his "crossculture test", and he calls attention to this method of approach. 7. As the results of treatment attempted systematically on the so-called two-drug resistant cases using various drugs, some satisfactory result can be expected in the treatment with different drugs to which no resistance has been shown, but the majority of cases with the concurrent use of the drug to which the resistance is already acquired fail to be effective. This fact clinically corroborates the inclusion of a high, percentage of the true double-resistant bacilli, and this agrees well with the result of previous animal experiments. The author points out that this point has a clinical significance worthy of attention.