1. University of North Carolina, Department of Biology, 607 Fordham Hall, CB #3280, Chapel Hill, North Carolina 27699-3280, USA
2. Harvard Medical School, Department of Cell Biology, 250 Longwood Avenue, Boston, Massachusetts 02115, USA
3. Stanford University, Department of Biological Sciences, Stanford, California 94305, USA
4. Laboratory of Chemistry and Cell Biology, The Rockefeller University, New York, New York 10021, USA
5. Present address: University of Oregon, Institute of Molecular Biology, 1370 Franklin Blvd., Eugene, Oregon, USA
6. These authors contributed equally to this work

Correspondence to: E. D. Salmon¹ Correspondence and requests for materials should be addressed to E.D.S. (Email: tsalmon@email.unc.edu).

Proper positioning of the cell division plane during mitosis is essential for determining the size and position of the two daughter cells—a critical step during development and cell differentiation¹. A bipolar microtubule array has been proposed to be a minimum requirement for furrow positioning in mammalian cells, with furrows forming at the site of microtubule plus-end overlap between the spindle poles^{2, 4, 4}. Observations in other species have suggested, however, that this may not be true^{5, 6}. Here we show, by inducing mammalian tissue cells with monopolar spindles to enter anaphase^{7, 8}, that furrow formation in cultured mammalian cells does not require a bipolar spindle. Unexpectedly, cytokinesis occurs at high frequency in monopolar cells. Division always occurs at a cortical position distal to the chromosomes. Analysis of microtubules during cytokinesis in cells with monopolar and bipolar spindles shows that a subpopulation of stable microtubules extends past chromosomes and binds to the cell cortex at the site of furrow formation. Our data are consistent with a model in which chromosomes supply microtubules with factors that promote microtubule stability and furrowing.