In a previous study, we showed that the blast stem cells of acute myeloblastic leukemia (AML) were more sensitive to cytosine arabinoside (ara-C) when growing in suspension culture than during colony formation in methylcellulose. We suggested that the difference might be explained by considering the cellular mechanisms responsible for growth in suspension and colony formation. In the former, the clonogenic cells increase in number (self-renewal); in the latter, most of the divisions are terminal. The increased sensitivity to ara-C in suspension might then be attributed to its ability to inhibit self-renewal to a greater degree than cell division generally. A test of this hypothesis would be to compare the survival curves in suspension and in methylcellulose using a drug that spared or stimulated self-renewal. Such an agent is 5-azacytidine (5-aza) and has the additional advantage that its analogue, 6-azacytidine (6-aza) has no effect on self renewal. The data supported the hypothesis, since clonogenic AML blasts were much less sensitive to 5-aza in suspension than in methylcellulose. The effect of 6-aza, while qualitatively similar, was much less marked. Controls showed that the difference in survival curves could not be explained on a kinetic basis or by the secretion of growth factors by 5-aza-treated cells. We suggest that a comparison of the effects of drugs in suspension and in methylcellulose may be useful in preclinical screening of putative anti-AML compounds.