A single cell isolated from the sexual progeny of mutagenized parents gave rise to a clone of cells with an abnormal, conical shape. Breeding analysis revealed that this shape results from the action of a single recessive gene, co (conical). Homozygous mutant cells are shorter and wider than wild type cells, and have their widest point at a more posterior position. Nonetheless, cortical parameters such as number of ciliary rows, number of ciliary units within these rows, and positions of contractile vacuole pores remain essentially unchanged in conical cells, suggesting a considerable degree of mutual independence of pattern and form. Shape changes prior to cell division bring about some convergence in form of dividing conical and wild type cells. However, in conical cells the new oral apparatus and fission line form well posterior to the cell equator, so the opisthes are invariably smaller than proters. Macronuclei nonetheless undergo constriction at the normal central location, and the characteristic inequality in the DNA content of the two macronuclear division products is not increased by the conical condition. Generation times are, on the average, nearly the same in the two wild type daughter cells and in conical proters, while the small conical opisthes have generation times averaging one-third longer. This prolongation explains why population doubling times are always somewhat longer in cultures of conical cells than in parallel cultures of wild type cells. The unusually long generation times of conical opisthes allows for the compensation of initial size differences. Observations on shape changes in conjugating cells of various genotypic combinations suggest that the wild type gene product is not freely exchangeable across the conjugation bridge. The implications of the conical phenotype for problems of cellular patterning and positioning of organelle systems are considered in the discussion.