Tumors may be regarded as arising from the conversion of one or, at most, a few normal cells. A permanent change, presumably in the cell genome, might result whenever cells are affected by a tumorigenic virus or chemical carcinogen. The work described in this paper is an attempt to observe such changes of cells in vitro, and to correlate what is seen with their ability to cause tumors when injected into animals. To reduce the morphological and karyological variability encountered with heterogeneous cell populations, such as those derived from trypsinized whole embryos, all experiments were begun with the same clone of a continuous line of cells which does not produce tumors. Colonies derived from single cells of this clone can easily be selected after growth in soft agar.' It thus became possible to examine the descendants of a single cell before and after treatment with carcinogens and viruses, and to determine what cellular changes were associated with the acquisition of an ability to grow into tumors in vivo.Materials and Methods.-The cells used were obtained from Dr. T. C. Hsu in,1963. In 1964, this pseudodiploid, male Chinese hamster lung tissue cell line was cloned and adapted to grow as colonies in a semisolid agar medium as described previously" 2 Thereafter, routine passages in agar culture had a plating efficiency of approximately 10%. Traditional carcinogens were used: 7,12-dimethylbenzanthracene (DMBA), 3-methylcholanthrene (MCA), and 3,4-benzo(a)pyrene (BP). Fresh stock solutions in either dimethylsulfoxide (DMS0),3 or dimethylformamide (DMF) (1-2 mg/ml), were added to the agar medium at the time the plates were poured, at an initial concentration of 0.3-0.5 ,g/ml for the hydrocarbons and not above 0.2% for the solvents. Cells (ca. 10' in molten soft agar) were poured onto an agar underlay containing hydrocarbon. They were grown in the continuous presence of hydrocarbon for 12 days and then transferred to agar medium without carcinogen.SV40 virus was obtained from Dr. J. Fogh, and a stock prepared using secondary cultures of green monkey kidney cells (titer of 5 X 108 PFU/ml4). A polyoma virus stock of similar titer (1 X 108 PFU/ml) was prepared on mouse embryo kidney cells as already described.5 Virus stocks were kept frozen at -200C. Tumor-inducing viruses were incubated overnight with 2 X 108 washed cells (multiplicity of about 1000 PFU/cell) in a total volume of 4 ml, the cells washed, and then plated in the soft agar medium.For the analysis of chromosome complements, several single colonies were transferred individually from agar layers to liquid medium in Petri dishes containing coverslides. Colchicine (0.5 /Lg/ml) was added 4 hr before fixation to accumulate metaphases. The cells were caused to swell in hypotonic saline (phosphate-buffered saline 1:5), fixed in acetic acid-alcohol (1:3 v/v), and rapidly air-dried. Preparations were stained with McNeal's tetrachrome or aceto-orcein (Gurr) and mounted.The following cytological data were then obtained: (a) the number of metaphase chromosomes per ...