Unless you've been holed up in your hut in Idaho for the last thirty years, slowly working your way through your stockpiles of canned goods and ammunition, you've undoubtedly heard of the famous 1967 experiment by Stanley Milgram in which he handed envelopes to a bunch of people with a name and address on them. The object was to get that envelope to the target by passing them on to someone they knew personally who seemed likely to be closer to that person, and asking them to do the same, attempting to advance the envelope with each connection. Milgram's finding that it took an average of six people to reach the target is the basis for the famous "Six degrees of separation" theory that later turned into an award-winning play and movie.

It seems, however, that the experiment was not quite as decisive as you might have thought:

When Kleinfeld began sifting through Milgram’s original data at Yale, she was surprised to find how much that data seemed to conflict with what Milgram had reported. Only 3 of the 60 envelopes in the original study had reached the divinity student’s wife—a completion rate of just 5 percent. The second study reported a completion rate of only 29 percent. Moreover, Milgram recruited subjects for two of his studies by buying mailing lists, which tend to be biased in favor of high-income people with high numbers of connections. Other sociological work has shown that low-income people are generally able to reach other individuals with low incomes, but not those with high incomes.

But how most of those chains failed may be even more interesting than how the minority succeeded:

In 2003 Watts published the results of an e-mail version he did of Milgram’s experiment. He set up a Web page and recruited 18 targets in 13 countries. In the end, 61,168 starters signed on, and 24,163 chains were begun. Of those, only 384 were completed. Those who finished their chains did so within slightly more than four links, on average. Watts, unlike Milgram, included a survey with his study, and one of the questions asked people who hadn’t finished to give the reason why. Less than one-half of 1 percent of respondents said they had failed to pass the e-mail on because they didn’t know who to send it to. Watts believes the majority failed because of other problems, such as e-mail spam blocks that diverted their requests. Other times, he suspects, chains failed because the people who received an e-mail weren’t as interested in continuing the chain as the people who’d started it.

Lack of interest, Watts says, points to the underlying complexity of networking. The question is not just whether we are closely connected, but how we navigate those connections—and whether we choose to do so at all. “People can find these paths as long as they’re motivated to do so and able to motivate people to help them,” he says. “But no matter how motivated you are, you have to be able to motivate the other person, who can put you in touch with the next person, and the next person has to do it too.”

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.