It was a grand experiment. Psychiatrist Thomas Wehr gathered a group of people together in the 1990s and left them in the dark 14 hours a day, for a month. After four weeks, people had developed a particular sleeping pattern.
They slept for four hours at the beginning of their 14 dark hours around 8 p.m., woke for an hour or so some time after midnight, then slept again for four more hours.
This pattern of sleeping, waking up then sleeping again is known as biphasic sleep. Wehr believed that this would be the normal pattern for most people if artificial light were not a part of our lives.
Publication of these results didn't change anything in society at large, where we've all been pretty much convinced that we're supposed to be getting an eight-hour block of sleep every night.
Historian Roger Ekirch of Virginia Tech published a paper in 2001 that supported Wehr's findings from a decade earlier. His research on historical sleep patterns was extensive, spanning 16 years.
Ekirch later published a book called "At Day's Close: Night in Times Past" which contains more than 500 references drawn from court records, diaries, literature and medical books.
He found that people would drift into first sleep two hours after the sun went down. They would awake and stay up for an hour or so, and then go back to sleep.
During the wakeful period in the middle of the night, people tended to get up out of bed, perhaps read, write, pray, visit, and sometimes have sex, with each other.
One medical book from the 1500s recommended that sexual intercourse be engaged in after they woke from the first sleep. The participants would be rested and would enjoy it more, the doctor suggested.
Late in the 1600s, Ekirch found that there were fewer allusions to first and second sleep in the writings he researched. Over the next two centuries, the ideal of eight solid hours of sleep gradually became the norm.
By early in the 20th century, biphasic sleep had become a notion that was unheard of.