Akin's remarks -- made as part of a larger attack on abortion -- immediately drew widespread criticism, mostly for being unsupported by any kind of empirical evidence. There is no solid evidence in medical literature to support the theory that a woman's body has natural defenses against being impregnated by a rapist.
There is, however, much evidence that many rape victims are impregnated by their rapists. In a 1996 paper in the American Journal of Obstetrics and Gynecology, three University of South Carolina researchers surveyed more than 4000 American women and found 34 cases of rape-related pregnancy, leading them to estimate that there are about 32,000 pregnancies from rape each year.
"Rape-related pregnancy occurs with significant frequency," the authors wrote.
A more controversial 2001 paper from St. Lawrence University researchers Jon and Tiffany Gottschall published in the journal Human Nature suggested that rape results in a higher rate of pregnancy than consensual intercourse. The Gottschalls found that the pregnancy rate among agroup of 405 rape victims was 6.4 percent, as compared to a 3.1 percent pregnancy rate for intercourse in the general population.
"Our analysis suggests that per-incident rape-pregnancy rates exceed per-incident consensual pregnancy rates by a sizable margin, even before adjusting for the use of relevant forms of birth control," the Gottschalls wrote.
However, some animals actually do have methods to avoid pregnancy from rape. In their book, "A Natural History of Rape," University of New Mexico biologist Randy Thornhill and University of Colorado anthropologist Craig T. Palmer describe a process known as "ejaculate dumping" in red junglefowl, a kind of Asian pheasant thought to be the wild ancestor of the domestic chicken.
Female red junglefowls prefer to mate with males higher up on the pecking order, so subordinate males often resort to rape. But the female junglefowl can cut her risk of becoming pregnant by her rapist by immediately expelling some of the unwanted semen, according to Thornhill and Palmer.
Sperm dumping has also been observed in fruit flies. Female fruit flies typically mate with more than one male, instigating a sperm competition between her various suitors. Some scientists have theorized that a male's ejaculate can incapacitate rival sperm, but it's difficult to imagine how such a mechanism would avoid damaging the user's sperm as well.
Two University of Sheffield researchers dismissed the idea that a male's sperm can directly take out the competition, saying in a 2004 Nature paper that female fruit flies eject stored sperm from a first mating after taking on a new lover.
So sperm dumping does happen on occasion in nature, but there's little evidence to suggest that human women have the same defense mechanisms as a red junglefowl.
Akin, for his part, has backed down from his remarks, saying that he "misspoke."