What was found at Liang Bua
The skeleton, catalogued LB1, was described by Peter Brown, Mike Morwood and colleagues in Nature in October 2004. She was female, around 30 years old at death, and would have weighed about 25 kilograms. Liang Bua has since produced more than 100 fossils representing at least 14 individuals, spread across tens of thousands of years.
Was she just a diseased modern human?
This was the central argument of the following decade. Microcephaly, Laron syndrome, cretinism and Down syndrome were all proposed in print. Each had the same appeal: it let the existing story of human evolution stand.
Two things ended the debate. First, no known condition in modern humans produces this combination of features — a small skull and a rearranged wrist and a reshaped shoulder and a lengthened foot. Second, and decisively, LB1 is not alone. A disease can explain one skeleton. It cannot explain a population that persists for tens of thousands of years.
The wrist that settled it
The most decisive evidence is a bone most people have never heard of. In the modern human wrist, the trapezoid is shaped roughly like a boot — a specialised form that appears late in our lineage. In chimpanzees and early hominins it is a simple wedge.
In 2007, a team led by Matthew Tocheri examined the Liang Bua wrist bones and found wedges. The wrist forms very early in development, long before birth; it is not a structure that a childhood illness goes back and redesigns. Alongside it: a shoulder blade set forward, giving a hunched carriage; long flat feet with a short big toe; and no chin. This is not a modern human with something wrong. It is an older body plan that got small.
The island rule
Islands do predictable things to animals. Large mammals with limited food and no serious predators get smaller; small animals whose predators are absent get bigger. Flores shows both in the same sediment: Stegodon elephants shrunk to roughly the size of a large buffalo, rats the size of cats, a marabou stork nearly two metres tall — and Komodo dragons three metres long, which are still there. Homo floresiensis is not an anomaly on that island. It is the island rule applied to us.
Mata Menge — 700,000 years of being small
At Mata Menge, in the open So’a Basin about 70 km east of Liang Bua, fossils published in 2016 pushed the story back to around 700,000 years ago. In August 2024, the same project published three more specimens, including SOA-MM9, an adult humerus 9–16% shorter and thinner than LB1’s — the smallest adult hominin humerus ever reported, implying an adult stature of 93–108 cm.
Their teeth and humeral shaft resemble Javanese Homo erectus. Hominins were on Flores by a million years ago; within roughly 300,000 years they had shrunk to about a metre — and then the body size simply held, for more than 600,000 years. A minority of researchers still argue for an older, pre-erectus ancestor on the strength of the wrist, foot and shoulder. The 2024 fossils weaken that case without ending it.
How did they get there?
Flores has never been connected to mainland Asia. Even at maximum ice-age sea-level drop, deep-water channels east of Java, Sumatra and Borneo never closed. Reaching Flores has always required crossing open ocean.
The usual explanation is accident — a tsunami or a vegetation raft. But in August 2025, stone tools at the Calio site on Sulawesi were dated to at least 1.04 million years, and possibly 1.48 million. Hominins were on multiple islands across this archipelago, separated by water that never dried, a million years ago. One crossing can be a freak wave. A pattern is harder.
What they actually did — the 2026 correction
For twenty years the picture was firelit hunters bringing down dwarf elephants. In July 2026, Veatch and colleagues published a taphonomic study in Science Advances that dismantled it.
They fed a goat carcass to a captive Komodo dragon at Zoo Atlanta and recorded exactly what its teeth do to bone, then mapped the marks on the Liang Bua Stegodon remains. Dragon tooth marks dominate the meatiest parts; stone-tool cuts cluster where almost no meat was left. That is the signature of arriving second. They were scavenging carcasses the dragons had already opened.
And no fire
The same study found no burnt Stegodon bone in the H. floresiensis layers, and no burning among more than 4,000 rodent bones from the same deposits. The dark patches previously read as hearths are manganese staining left by groundwater. The real fires in Liang Bua belong to a later occupation by modern humans. Homo floresiensis made stone tools and, on current evidence, never controlled fire.
The date that disappeared
The original announcement dated the remains to about 18,000 years ago — recent enough to overlap substantially with modern humans, and recent enough for people to link them to the Flores folk tradition of the ebu gogo, small hairy cave-dwellers. In 2016, re-excavation showed the young dates came from a younger sediment layer that had been eroded and re-deposited above the fossils. The remains are around 60,000 years old; the last stone tools about 50,000. The ebu gogo remain folklore, and there is no evidence they record a surviving population.
Why they disappeared
In December 2025, Gagan and colleagues published a climate record from a stalagmite in Liang Luar, 660 metres from Liang Bua, dated with 32 uranium-thorium measurements and read at 50-year resolution. Mean annual rainfall fell about 37% — from ~1,560 mm at 76,000 years ago to ~990 mm at 61,000 — with summer rainfall collapsing to ~450 mm. H. floresiensis remains cluster between 87,000 and 61,000 years ago; Stegodon makes its last appearance at 57,000. The authors say the findings increase the likelihood that climate was a driver. They do not claim it as the cause, and neither should anyone else.
Frequently asked questions
Were the hobbits real? Yes. Homo floresiensis is a distinct species, known from more than 100 fossils and at least 14 individuals. “Hobbit” is a nickname, not a scientific term.
How tall were they? LB1 stood about 1.05 m. The older Mata Menge individuals may have been slightly smaller, at 93–108 cm.
How big was their brain? About 380–420 cc, compared with roughly 1,300 cc in modern humans.
Did they make tools? Yes — stone flakes and points, from about 190,000 to 50,000 years ago.
Did they use fire? On current evidence, no. A 2026 study found no combustion evidence in their layers at all.
Did they hunt elephants? No. The same study concluded they scavenged carcasses opened by Komodo dragons.
When did they die out? They disappear from the record around 50,000 years ago. The remains cluster earlier, between 87,000 and 61,000.
Did modern humans kill them? There is no evidence of contact or conflict. H. sapiens reached Liang Bua by about 46,000 years ago.
Has anyone recovered their DNA? No. Flores is hot and wet, the bones are not fully fossilised, and extraction attempts have failed.
Are the ebu gogo surviving hobbits? There is no evidence for this. It is folklore.
Watch the full Echoes of Earth documentary, “The Hobbits Were Real — And Everything We Were Told About Them Was Wrong,” embedded below — and subscribe for more stories of the forgotten worlds that shaped the ancient Earth.
