They are among the smallest of our ancient relatives and also one of anthropology's biggest mysteries.
Now, a reanalysis is shedding light on the tiny, upright creatures, nicknamed "hobbits," that apparently lived on the Indonesian island of Flores even as our own Homo sapiens ancestors were beginning their relentless expansion around the globe.
The new evidence comes from the Liang Bua cave, where the skeleton of a 1.1-metre-tall individual, thought to be a 30-year-old female, was first unearthed in 2003. By excavating farther into the cave, researchers say, they have clarified how long the pint-sized hominins, scientifically classified as Homo floresiensis, could have persisted in the area. There is no indication that the hobbits co-existed with anatomically modern humans in the region for very long, if at all, they say.
"If we had known this at the time of the discovery, then the scientific debate and controversy surrounding these remains probably would have been significantly less," said Matthew Tocheri, a paleoanthropologist at Lakehead University in Thunder Bay and co-director of the Liang Bua excavation.
Initially, layers of sediment associated with the skeleton as well as stone tools and bones from several other individuals were found to have been deposited as recently as 12,000 years ago. That would mean that Homo floresiensis probably overlapped with the ancestors of present-day Indonesians for tens of thousands of years on Flores, a scenario that has fuelled alternative theories that the skeleton is really that of an ordinary human with a disorder that stunted her growth.
Instead, Dr. Tocheri and his colleagues now say it is clear that all the bones of Homo floresiensis found to date were deposited no later than 60,000 years ago. Stone tools that have been attributed to the species are no more recent than 50,000 years ago. That is about the time that anatomically modern humans are first thought to have reached the Indonesian archipelago.
The team's revised chronology, published on Wednesday in the journal Nature, is based on something that previous studies overlooked: layers of volcanic ash that blanketed the region starting around the same period. The ash, which was eroded away and is not seen at the precise spot where the hobbit skeleton was found, overlays the sediment associated with the hobbit bones and tools.
"That essentially removes the overlap," said Mark Collard, an anthropologist and archeologist at Simon Fraser University in British Columbia, who was not involved in the new analysis.
While it does not rule out the possibility of interaction between hobbits and our own species 50,000 years ago, or even later, the new find removes the need to accept a lengthy co-existence and bolsters the view that Homo floresiensis is a separate, if surprising, offshoot of the Homo family tree.
"It's kind of hard to argue that they're modern humans, given the time frame," said Karen Baab, a biological anthropologist at Midwestern University in Glendale, Ariz.
Dr. Baab called the new analysis from Liang Bua "very thorough," but she added that arguments based on anatomy are still the most convincing. Studies that claim some of the hobbit bones are from diseased individuals have been strongly challenged over the years in what has at times been a heated debate within the community of researchers who study human evolution.
Researchers said the new dates, which suggest that Homo floresiensis may have vanished at about the time Homo sapiens arrived in the area, raise the possibility that modern humans hastened their demise.
Dr. Collard noted that competition with modern humans is thought to have played a role, perhaps a decisive role, in the extinction of Neanderthals in ice-age Europe. On Flores, "we may well be looking at the same situation," he said.
Dr. Tocheri, who is the Canada Research Chair in Human Origins and who divides his time between Thunder Bay and Indonesia, added that the reanalysis still leaves open the question of exactly how Homo floresiensis relates to us and when it branched off from other human relatives in the remote past.
One possibility is that the species is descended from Homo erectus – long thought to be the first hominin to have left Africa – and that it evolved into a smaller creature when it was confined by the limited resources of a relatively small island territory.
But anatomists also note that some of the features of Homo floresiensis, including shinbones and a small brain size, seem more primitive than Homo erectus. This raises the possibility that an even earlier species, with fewer of the adaptations needed for walking upright and running, somehow made its way out of Africa and left a remnant population on Flores.
The only way to distinguish between these theories is with the uncovering of more fossils – a strong possibility, given that up to three-quarters of the Liang Bua cave remains to be excavated, not to mention other candidate sites around the island.
"We're at the tip of the tip of the iceberg," said Dr. Tocheri, who added that the most important result to come out of the work is the recognition that "shows us how much more diverse our human family tree was even in the relatively recent past."