This study has explored the possibility of applying the metaphase-arrest method with colchicine t o two models of induced liver growth in the rat, regenerative growth and phenobarbital-induced growth. At a dosage of 0.5 mg/kg body weight (BW), colchicine caused a linear accumulation of mitoses for up to 90 min when administered at 3 days after the start of phenobarbital treatment; however these mitoses included a number of anaphases and telophases. No anaphase escape was seen when this dose of colchicine was given at various times after partial hepatectomy, though the arrested mitoses were invariably more fragmented and some may have even degenerated beyond recognition as early as 90 min after injection. It is concluded that the optimal dose of stathmokinetic agent is heavily dependent on the relative liver weight, and thus would change continuously during compensatory hyperplasia.In the adult rodent, the vast majority of hepatocytes are considered to be in a state of proliferative quiescence called G , (Wright & Alison, 1984). However, these cells retain the ability to re-enter the proliferative cycle upon the application of an appropriate stimulus. Such stimuli include partial hepatectomy (Fabrikant, 1968), cytoplasmic atrophy induced by a portaprival state (Weinbren, Stirling & Washington, 1972), xenobiotic compounds which are necrogenic (Schultze, Gerhard & Maurer, 1975) and other drugs and chemicals which cause a functional overload (Schulte-Hermann, 1974).At present, all these hyperplasias have been largely assessed by techniques based on tritiated thymidine ([3H]TdR) and the frequency of mitoses. It is well known, however, that measurements of the flash labelling index (I,) and the mitotic index (I,) reflect not only changes in the rate of cell proliferation, but also changes in the respective phase durations, viz. the duration of DNA synthesis (t,) and the duration of mitosis ( f , ) .The metaphase-arrest technique provides a useful kinetic parameter, the rate of entry of cells into mitosis, which is not dependent upon t,, and it has been widely applied in situations of continuous cell renewal and perturbative change (Wright & Appleton, 1980). Surprisingly, the technique has not previously been adopted in studies of liver cell growth, and the purpose of the present study was to examine the feasibility of metaphase-arrest analysis in the liver. Comparison is made with the more conventional methods of [3H]TdR labelling, and problems of data analysis with regard to the age distribution and the presence of polyploid cells are briefly discussed.