Cells from the spleen, lymph node, peripheral blood and peritoneal exudate of mice treated with C. parvum were tested for their ability to inhibit tumour growth in vitro. The peritoneal exudate cells from C. parvum treated mice were extremely effective in inhibiting tumour growth whereas the spleen and peripheral blood cells were only moderately so. In contrast, the lymph node cells caused only a modest inhibition of tumour growth at a very high effector to target cell ratio. Spleen cells from normal mice also exerted a moderate anti-tumour effect.

Severe depletion of thymus-derived lymphocytes does not reduce, and may significantly increase, the resistance of mice to syngeneic tumours, although it grossly impairs their ability to reject allotransplants. Injection of killed C. parvum causes further inhibition of tumour growth in T-cell deprived mice to an extent comparable to that resulting from the same dose of C.parvum in normal mice. This appears to lend support to the hypothesis that the antitumour effect of C. parvum depends primarily on macrophage stimulation.

Summary.-Local injection (i.e. injection at the site of tumour inoculation) of strains of C. Parvum which have a significant anti-tumour effect when given systemically (i.e. intravenously or, in the case of subcutaneous tumour transplant, intraperitoneally) strongly inhibits the growth of isogeneic transplants of a fibrosarcoma in intact CBA mice but has little or no effect on subcutaneous transplants of the same tumour in T-cell deprived mice. The anti-tumour effect of local injection of C. parvum, unlike that of systemic injection in this particular tumour system, thus appears to be T-cell dependent.PRELIMINARY observations of our own with a mouse fibrosarcoma, and experiments with other mouse tumours recently reported by Likhite and Halpern (1974) and Scott (1974), have shown that intratumour injection of C. parvum may strongly inhibit tumour growth and under some conditions cause complete regression.The present experiments were designed to investigate this phenomenon further, using both subcutaneous and intraperitoneal tumour transplants. To exclude purely mechanical effects we have compared the degree of tumour inhibition resulting from local injection of 3 anaerobic coryneforms which differ markedly in their effect on tumour growth when given systemically, and of an aerobic organism, C. diphtheriae.We have used as tumour hosts both intact and T-cell deprived mice, because in our own experience (Wooodruff and Dunbar, 1972;Woodruff, Dunbar and Ghaffar, 1973), the anti-tumour effect of systemic injections of active strains of C. parvum in respect of cholanthrene induced sarcomata is maintained in T-cell deficient mice, whereas Scott (1974) has reported that the growth of a mastocytoma, is inhibited by intratumour injection of C. parvum in intact mice but not in T-cell deprived mice. MATERIALS AND METHODSMice.-The recipient mice were either intact adult CBA females (20-22 g) or T-cell deprived CBA females prepared as described previously (Woodruff et al., 1973).Tumour. The tumour was originally induced in a female CBA mouse with methylcholanthrene. It was stored in liquid nitrogen after 1]5 transplant generations and was transplanted once more before being used in the experiments. The properties of this tumour have been summarized in a recent review (Woodruff, 1975). It was transplanted subcutaneously (right foreleg) or intraperitoneally, or by both these routes, in the form of a cell suspension prepared with pronase in a dosage (unless otherwise stated) of 104 cells.The mice were inspected and weighed every 2 days. If a subcutaneous tumour was present, its width in 2 directions at rightangles was measured with a caliper and the mean was recorded. Where appropriate, the diameters of subcutaneous tumours were summed, as described previously (Woodruff et al., 1973), up to the day when some of the tumours in control mice were 18-20 mm in diameter, and the group mean sums were compared by Student's t-test. In many cases, however, the differences between groups were sufficiently clearcut to make this unnecessary.

Summary.-The inhibitory effect of an i.v. or i.p. injection of C. parvum on intrastrain transplants of a mammary carcinoma in A/HeJ mice has been confirmed, and it has been shown further that C. parvum inhibits the growth of transplants of sarcomata induced with methylcholanthrene both in this strain (members of which lack the fifth component of complement) and in CBA mice (which are not complement deficient). In experiments with the mammary carcinoma, 2 injections of C. parvum on days + 3 and + 9 were more effective than a single injection on day + 3; injections on days + 3 and + 6, or + 3 and + 12, appeared to be marginally less effective than on days + 3 and + 9, but the difference was not statistically significant.Development of the CBA sarcoma was inhibited to about the same extent if, instead of treating the mouse with C. parvum, the tumour cells were pre-incubated with anti-tumour globulin (ATG) in the absence of complement prior to inoculation, and the effect of combining these procedures was much greater than that of either alone. Pre-incubation with ATG had a similar but less marked effect on the development of the mammary carcinoma but had no effect on the A/HeJ sarcoma. Injection (i.v.) of ATG did not inhibit the growth of any of the tumours in these experiments and possible reasons for this are discussed.

Summary.-S.c. injection of tumour cells or small pieces of tumour irradiated to a dose of 22,000 rad evoked resistance to live challenge with the same tumour (a CBA strain fibrosarcoma induced with methylcholanthrene) 14 days later. This resistance was, however, over-ridden if the challenging inoculum was sufficiently large, and did not develop if the cells were irradiated to 100,000 rad.The resistance evoked by injection of 106 irradiated tumour cells was impaired by i.p. injection of 1 4 mg C. parvum 5 days before, and virtually abolished by a similar injection 11 days after, the irradiated cells. The effect of s.c. injection of a mixture of 106 irradiated cells and C. parvum 14 days before live challenge depended on the dose of C. parvum. With 0*7 mg the development of resistance was largely but not completely abrogated; 0-35 mg resulted in a lesser degree of abrogation, and 0-09 mg or 0*02 mg had little or no effect.