An assay system is described in which a number of single tumour cells are injected intravenously into recipient mice and grow in the lungs of these mice to give macroscopic tumour nodules in 16 to 20 days. The relationship between the number of tumour nodules observed and the number of cells injected has been found to be linear up to a mean of 50 colonies/lung for both non-irradiated and irradiated cells, and the inter-experiment consistency has been shown to be good. The number of tumour nodules observed has been found to be affected by the simultaneous injection of a large number of heavily-irradiated cells, provided they are injected intravenously. An aerobic radiation survival-curve has been established and agrees very well with one obtained for the same tumour using the end-point dilution assay method. The assay has the advantage that the time required for completion is short, compared with that required for end-point dilution assays. © 1969 Informa UK Ltd All rights reserved: reproduction in whole or part not permitted.