Neo-Darwinian theory explains speciation as the gradual accumulation of genetic differences between isolated populations. Yet changes in allele frequency within an interbreeding population do not by themselves create a barrier to interbreeding, and the full transition from one animal population to two completely isolated species has never been observed. Here I propose the Twin Zygotes Model, in which a new sexually reproducing species arises within a single generation. The model rests on four observed phenomena: a single pregnancy or clutch can contain zygotes of both sexes; zygotes can carry gross chromosomal mutations and can divide into identical embryos; some carriers of such mutations become healthy adults; and siblings can mate with one another. When opposite-sex zygotes sharing the same environment acquire similar gross mutations, the resulting siblings are fertile with each other but not with the parental species, because their chromosomes fail to pair with the parental set at meiosis. Because the two mutational events are correlated, their joint probability far exceeds that of similar mutations in unrelated individuals. I estimate how often such events could found new species, review consistent observations, including species-specific karyotypes, polyploid speciation, and rapid chromosomal divergence in island mice, and state testable predictions that could falsify the model.