On the Origins of Homo floresiensis-New Fossils and New Implications

Introduction:

Homo floresiensis, a strange species of hominin from the Indonesian island of Flores, has had its origins debated since its first descriptions in 2004. Some new fossils, discovered in 2013 but just now described, have added some new insight on the topic of where this species came from. The fossils, consisting of 2 teeth and a partial humerus, dated to about 700,000 years old (700 kya), provide evidence that H. floresiensis was a dwarfed species descended from Homo erectus, which inhabited the nearby island of Java, rather than descending from more basal hominins from Africa such as Homo habilis like what was previously thought by most. 

Background on Homo floresiensis:

Homo floresiensis was a dwarf hominin species from the Indonesian island of Flores, which had an average height of 1 meter (~3 feet). It isn’t unusual for hominins to be this short, as many early hominins weren’t much larger than this, but this height is unusual considering the age of the species. The original specimens, including the type specimen, come from a site called Liang Bua cave, and were dated to about 80 kya, which is very recent. These remains from Liang Bua consist of 2 partial skeletons (LB1 and LB6/1), along with other postcranial elements like carpals and arm bones. 

Stone tools associated with H. floresiensis are known from as old as 190 kya to as recent as 50 kya, however older material is known from a second site, called Mata Menge, with remains as old as 700 kya. The fossil and archaeological record of H. floresiensis is quite complete, and our understanding of the species is good, but one aspect that is still heavily debated is its origins. It is unclear whether or not this species originated from Homo erectus, a species that inhabited the nearby island of Java, that simply underwent insular dwarfism (evolving to be smaller-bodies on an island with limited resources), or if it originated from an earlier hominin species, such as Homo habilis, which left Africa earlier. 

It makes more sense to think that H. floresiensis descended from H. erectus due to how close the 2 groups were to one another at around the same time, and the 2 species do share some traits such as a maximum cranial breadth at the supramastoid region, but certain traits found in the skeleton call this into question. The overall skeleton of H. floresiensis is very primitive looking, and is more similar to earlier Homo species or even earlier australopiths. 

It had a small cranial capacity of ~380 cc and a small body size. These traits can be explained by insular dwarfism, but its limb proportions and features of the wrist bones are traits only found in hominins earlier than H. erectus. Analysis of the skeleton of H. floresiensis suggests that it diverged from earlier hominins sometime around the time of H. habilis.

Newly described material from Mata Menge gives a clearer look into earlier H. floresiensis, showing some unexpected traits and giving further insight on the origins of this strange species. 

The New Fossil Material:

As mentioned, the original remains from Homo floresiensis come from a site called Liang Bua cave, dated to about 80 kya. This is very recent, but older fossils from Mata Menge may represent more basal H.  floresiensis at around 700 kya. Up until this point, the only fossils described from this site consisted of one mandible fragment and 6 isolated teeth. 

The new fossils consist of a partial adult humerus and 2 teeth. These remains come from the same sandstone layer (layer II) as the other remains from Mata Menge, and therefore are around the same age. The humerus (SOA-MM9) is remarkably small, measuring only 88 mm (~3 in) in length. However, it is a partial humerus, and so the length of the whole complete bone in life is estimated to be between 206-226 mm (~8-9 in). This is about 9-16% shorter than the humerus of LB1, the type specimen of H. floresiensis from Liang Bua. This humerus is the smallest known adult hominin humerus ever discovered. Despite its short length, it is clearly an adult humerus, making its size unique even for its species. 

When compared to other hominin humeri other than H. floresiensis, it is shown to be even more unique. SOA-MM9 is distinct from the distal humerus of Australopithecus as it lacks aspects in the supracondylar ridge and crest and curvature in the sagittal plane, traits characteristic of Australopithecus distal humeri. In the cross-sectional shape of the humerus, it is most similar to basal Homo, such as Homo naledi and later H. floresiensis.

As for the teeth, the first tooth described is a maxillary canine tooth called SOA-MM10. This tooth is also very small, and is similar to Homo erectus and Australopithecus in its low distal shoulder. The second and more important tooth (SOA-MM11) is a mandibular 3rd molar. It is small, and very close to the teeth of Javan H. erectus due to its mesiodistally short crown. This is different from what is seen in Homo habilis, which is known to have a more mesiodistally elongated crown. Along with being very similar to Javan H. erectus, it is also somewhat unique from later H. floresiensis from Liang Bua. It possesses 5 cusps described as being in a ‘+’ arrangement, whereas Liang Bua Homo floresiensis possesses 4 cusps, giving them a more derived dentition. 

Implications:

There are several implications from these new fossils. First off, their smaller size and older age demonstrates that Homo floresiensis was small very early on in their evolution, potentially smaller than later members of the species. These fossils represent a basal early population of H. floresiensis that is less derived than the later specimens from Liang Bua. The molar is very similar to Javan Homo erectus, more so than it is to Australopithecus and Homo habilis, so this population may represent a dwarfed H. erectus population before the more derived dentition appeared in later H. floresiensis.

Under this, if H. floresiensis was closer to early pre-H. erectus Homo, then it must have convergently evolved to be more similar to the neighboring H. erectus population. This makes it very convincing that H. floresiensis is a dwarfed species descending from H. erectus. 

This is plausible, as H. erectus is known in Java by at most 1.5 million years ago (1.5 mya). It’s very possible that H. erectus was in Flores by 1 mya (likely via accidental rafting), giving a 300k year period for the dwarfing to occur, leading up to the dwarfed Mata Menge material at 700 kya. Definitive H. erectus fossils from Flores must be found on Flores by at least 1 mya to further support this. 

Conclusion:

This research by itself presents a convincing case that Homo floresiensis is a dwarfed descendant of Homo erectus, but it doesn’t explain the more basal features found in the postcrania that associate it with earlier Homo species. This doesn’t 100% prove H. floresiensis is a descendent of H. erectus, but it is another piece of the puzzle, a very large and messy puzzle. As it stands right now, H. floresiensis is a messy mix of basal and derived characteristics, making its origins very unclear. Further evidence will be needed to ‘prove’ either hypothesis, but this research certainly brings us closer to figuring out the origins of H. floresiensis. 

Sources:

1. Kaifu, Y., Kurniawan, I., Mizushima, S., Sawada, J., Lague, M., Setiawan, R., Sutisna, I., Wibowo, P. U., Suwa, G., Kono, T. R., Sasaki, T., Brumm, A., van den Bergh, D. G. (2024). Early evolution of small bodied size in Homo floresiensis. Nature Communications, 15, 6381. https://doi.org/10.1038/s41467-024-50649-7
2. Kreier, Freda. “Tiny arm bone belonged to smallest ancient human ever found.” Nature. 08-06-24. https://www.nature.com/articles/d41586-024-02548-6 
3. Dorey, Fran. “Homo floresiensis”. The Australian Museum. 23-12-19. https://australian.museum/learn/science/human-evolution/homo-floresiensis/ 
4. Orr, M. C., Tocheri, W. M., Burnett, E. S., Awe, D. R., Saptomo, W. E., Sutikna, T., Jatmiko, Wasisto, S., Morwood, J. M., Jungers, L. W. (2013). New wrist bones of Homo floresiensis from Liang Bua (Flores, Indonesia). Journal of Human Evolution, 64(2): 109-129. https://doi.org/10.1016/j.jhevol.2012.10.003 
5. van den Bergh, D. G., Kaifu, K., Kurniawan, I., Kono, T. R., Brumm, A., Setiyabudi, E., Aziz, F., Morwood, J. M. (2016). Homo floresiensis-like fossils from the early Middle Pleistocene of Flores. Nature, 534, 245-248. https://doi.org/10.1038/nature17999 
6. Argue, D., Morwood, J. M., Sutikna, T., Jatmiko, Saptomo, W. E. (2009). Homo floresiensis: a cladistic analysis. Journal of Human Evolution, 57(5): 623-639. https://doi.org/10.1016/j.jhevol.2009.05.002 

Link to a short video on the topic: https://www.youtube.com/shorts/1NT5gco1Lf4