Polyamine analogues such as bis(ethyl)norspermine and N1-(cyclopropylmethyl)-N11-ethyl-4,8-diazaundecane (CPENSpm) act as potent modulators of cellular polyamine metabolism in vitro and possess impressive antitumor activity against a number of cell lines. Some of these polyamine analogues appear to produce their cell-type-specific cytotoxic activity through the superinduction of spermidine/spermine N1-acetyltransferase (SSAT). However, there are several analogues (e.g., N1-(cycloheptylmethyl)-N11-ethyl-4, 8-diazaundecane (CHENSpm)) which are effective cytotoxic agents but do not superinduce SSAT. We have previously demonstrated that CPENSpm and CHENSpm both initiate the cell death program, although by different mechanisms, and that CHENSpm (but not CPENSpm) induces a G2/M cell cycle arrest. We now report that one potential mechanism by which some polyamine analogues can retard growth and ultimately produce cytotoxicity is through interference with normal tubulin polymerization. In these studies, we compare the effects of the polyamine analogues CHENSpm, CPENSpm, and (S)-N1-(2-methyl-1-butyl)-N11-ethyl-4,8-diazaundecane (IPENSpm) on in vitro tubulin polymerization. These spermine analogues behave very differently from spermine and from each other in terms of tubulin polymerization rate, equilibrium levels, and time of polymerization initiation. These results demonstrate that structurally similar polyamine analogues with potent antitumor effects can produce significantly different cellular effects. The discovery of polyamine analogues that can alter tubulin polymerization provides a series of promising lead compounds that may have a similar spectrum of activity to more difficult to synthesize compounds typified by paclitaxel.