An outlaw accosts (three) sisters, demanding that one of them marry him on pain of death. As all refuse, he kills all but the youngest. She accidentally learns that he is their brother. The outlaw usually then kills himself in remorse.
According to a study cited by Matt Ridley (_The Red Queen_, p. 281), "two siblings reared apart are surprisingly likely to fall in love with each other if they meet at the right age" (cf. Ridley's _The Agile Gene_, p. 173). The reference is to M. Greenberg and R. Littlewood, "Post-adoption incest and phenotypic matching: Experience, personal meanings, and biosocial implications," in the _British Journal of Medical Psychology_, 68:29-44, 1995.
There does seem to be anecdotal evidence for this; newspaper reports say that Britain in 2008 started to work on laws to make sure adopted children knew about any relatives they had. This was in response to a case of two twins separated in infancy; they met when they grew up, fell in love, and were married before anyone realized they were siblings. But this is just an isolated incident, not a rule.
I have not seen Greenberg and Littlewood to know if Ridley is describing it correctly, let alone to know if the conclusions are justified. But it may be less surprising than it sounds. Evolutionary success consists in conserving one's genes. This means that the evolutionary ideal is to marry someone related at about the first or second cousin level -- close enough to share a lot of genes, not so close as to have a particularly high risk of reinforcing dangerous recessives.
(There does seem to be one side footnote to this: Olivia Judson, in _Dr. Tatiana's Sex Advice to All Creation_, Henry Holt/Owl Books, 2002, pp. 52-53, notes that there are many variants in the genes of the MHC, or major histocompatibility index -- and that people apparently can tell, by smell, who shares their MHC genes; women don't want to be involved with men who are too close in MHC. But, of course, brother and sister need not share MHC genes -- given the size and complexity of the gene group, they very likely will not -- it's just that the odds are higher than among strangers.)
It is interesting to note that surveys have shown that beauty is in the eye of the beholder, but ugliness is not -- that is, almost everyone agrees that certain people are ugly, but not everyone agrees on who is attractive. It is further interesting to note that -- insofar as this has been studied -- we seem to find attractive people who appear to share our own genetic traits. (This is not from Ridley; I can't remember where I read this.)
On the other hand, people don't tend to fall in love with people they grow up with. Presumably this is a semi-instinctive incest taboo: The deep-down emotional assumption seems that these people are siblings or parents or offspring (so Edward Westermarck; again see Ridley, _The Red Queen_, p. 283; _The Agile Gene_, pp. 171-173).
Obviously a sibling is the closest relative we can find within our generation. If siblings are raised separately, they will not feel the raised-together taboo, so the shares-my-genes attraction will produce a tendency to fall in love. At least, that seems the logical implication of the data. And hence songs such as this and "Sheathe and Knife" and "Lizie Wan."
Though the siblings, it appears, would have to be separated by the age of three. But Ridley adds that the aversion seems to be stronger in females. If the brother is older (as seems to be the case, e.g., in "Lizie Wan," and probably here), he might have left the household before the girl reached the "aversion threshold."
In that context, it's worth remembering that sons of noble families were often sent away from their homes to be raised and trained in arms. In England, noble siblings were rarely raised together in the Middle Ages. So -- assuming all this theorizing is correct -- incestuous love affairs would be much more common among the nobility than the common folk. Indeed, there was a rumor that Ernest, Duke of Cumberland, the fifth son of George III who later became King of Hanover, fathered a child on his sister Sophia; see Christopher Sinclair-Stevenson, _Blood Royal: The Illustrious House of Hanover_, p. 123, 128. Sinclair-Stevenson thinks it impossible that Cumberland was actually the father, but it hardly matters if he was; the point is that he could have been. (A *really* dirty part of my mind notes that George III -- like his descendant Nicolas II of Russia -- long forced his daughters of marriageable age to stay at home with him. But George's daughters, at least, managed affairs -- see Sinclair-Stevenson, p. 124).
An even stronger instance of brother-sister incest occurs in the Bible, no less. Very few female members of the Davidide royal family are mentioned in the Bible -- except one. 2 Samuel, chapter 13 (one of the chapters that seems to have been written by an immediate witness -- some suspect the priest Abiathar), details the rape of David's daughter Tamar by her half-brother Amnon; the next several chapters are devoted to the dreadful after-effects of that rape.
The ultimate example of incestuous royal families, though, is surely the Ptolemaic Dynasty, which ruled Egypt from the time of Alexander the Great until the Roman conquest. Ptolemy II, late in life, would marry his sister Arsinoe II, and Ptolemy IV took up with his sister Arsinoe III.
And then there are the children of Ptolemy V. The older son, Ptolemy VI Philometer (which means "loving his mother"!), married his sister Cleopatra II; they had a daughter Cleopatra III. The second son of Ptolemy V was Ptolemy VIII Physcon, who in his turn married Cleopatra II and then, while she was still alive, her daughter Cleopatra III. Their children were Ptolemy IX Lathyrus, Cleopatra IV, and Ptolemy X Alexander. Ptolemy Alexander would later marry Cleopatra Berenice, the daughter of Ptolemy Lathyrus and Cleopatra IV. (This did have genetic effects, to be sure. The later Ptolemies were mostly immensely, grotesquely fat and diseased. On the other hand, Cleopatra VII -- "the" Cleopatra, of Julius Caesar and Mark Antony fame, whose mother and grandmother were non-Ptolemies -- was certainly accomplished and probably quite beautiful.)
Later, Cleopatra VII would marry a couple of her brothers, but that was political. In the cases of Arsinoe II and Cleopatra III, their royal brothers and uncles married for love, or at least lust. Thus, historically, royal incest seems not to have been all that uncommon. Probably more common than the above would imply, given how strongly it would be hushed up! - RBW