The palaeoanthropological site of Trinil is located in the island of Java in Indonesia. It is flat area, about 60 meters above sea level, with the Kendeng Hills to the north and the Lawu volcano to the south. The plains of Trinil form the ancient floodplain of the Solo and are shaped as a terraced landscape [12].

The Solo River, which is about 15 meters deep, runs through this area. The river deepened and widened its valley floor, under gravel-bearing, meandering conditions. It exposed a series of volcanic sediments that are mostly from the Pleistocene. East of Trinil, the Solo River turns northward and passes through the Kendeng Hills. Here, the river banks reveal layers of marine sediments from the Tertiary period [12].

The exposed pre-terrace series which are approximately 230 m thick along with the terraces, encapsulates a ~3 my (~3m to ~350 ka) record of tectonic, volcanic, climate, and sea-level fluctuations. The series is subdivided into five stratigraphic units each reflecting different depositional environments: Kalibeng Formation, Padas Malang Formation, Batu Gajah Formation, Trinil Formation, and Solo Formation [12].  

History

In 1887, Dutch physician Eugène Dubois decided to joined the Dutch army to go to the Dutch Indies, now the country of Indonesia. His goal was to find fossil evidence of a transitional form or the ‘missing link’ between apes and humans. Dubois spent a decade, from 1891 to 1900, conducting large-scale excavations along the Solo River near the village of Trinil in East Java [1][13][14].

Dubois excavated a total area of 2,317 m² for the left bank and 106 m² for the right bank. Thousands of vertebrate remains were recovered from the fossil-rice deposits along the riverbanks. In 1891, there was a significant discovery during the excavations: a right upper upper third hominin molar was found. A month later, at a distance of about 1 m from where the hominin molar was found, was the discovery of a hominin skullcap. Again, in August 1892, a hominin femur was discovered about 12 m downstream of the skullcap. The same month, a left upper second molar was also found [15][16][17].

Dubois conducted his last excavation of Trinil in 1900. This last year, a very large area was excavated, spanning an area of 75x6 m directly east of the trenches excavated in 1899. No further hominin remains were found during this operation. However, in 1932, four incomplete femora (Femora II-IV) were discovered in previously unopened boxes of the Dubois collection. These are currently housed at Naturalis Biodiversity Center, Leiden, the Netherlands and are believed to have originated from the 1900 excavations [15][16][17][18][14][13].

Dubois identified these hominin fossils as belonging to the species Pithecanthropus erectus [19][20][21][14], currently Homo erectus [22][23], with the Trinil skullcap serving as its type fossil [24]. Dubois believed that all the Pithecanthropus discoveries, including the four femora found later, came from the same fossil-bearing layer. He initially referred to this layer as the ‘Lapillischicht’ (lapilli layer [20][25][15][17], and later renamed it the ‘Hauptknochenschicht’ or H.K. (main bone layer [26])[14].

The Trinil hominin fossils have been instrumental in understanding human evolution and the emergence of paleoanthropology as a discipline. Alternately, the fossils caused intense debates within the scientific community. Dubois’ assertion that the archaic-looking skullcap and the modern looking femur (Femur I) belonged to the same species was strongly contested [27].

Following Dubois’ completion of his excavations at Trinil, a team led by Lenore Selenka conducted fieldwork from 1906 to 1908 to continue the search for more hominin fossils. The team carried out extensive work on the right bank (Grube I), and a smaller excavation on the left bank (Grube II), extending Dubois’ previous work. While the expedition did not find additional hominin remains, it produced a significant number of vertebrate fossils and valuable contextual data due to its systematic and multidisciplinary approach, as well as detailed records [4][18][14][13].

In the decades following the Selenka expedition, a number of smaller-scale fieldwork were conducted; however, very little information about those efforts has been published to date. In 1976 and 1977, an Indonesia-Japanese team made new geological observations on the setting of the fossil bearing levels primarily through coring studies [28][18]. 

Two additional hominin fossil femora labeled Trinil 9 and Trinil 10 have been traced back to Terrace T1 deposits and directly dated by U-series to a minimum age of 31-34 ka. Discovered by Prof. Teuku Jacob in 1978 at the right bank in cross-bedded sands, these small-bodied H. sapiens fossils morphologically align with Late Pleistocene specimens [29].

In 2016, 2018, and 2019, new fieldwork was conducted at Trinil, involving geological surveys, logging, and sampling of the sections dug into the Solo River’s current river banks. Additionally, test pits were excavated to target the low-lying remnants of find-rice deposits that are located 0–2 m above the current low water level. This work led to a revised stratigraphic framework for the Trinil area [12].

In addition to early hominin remains, the site also produced an enormous amount of vertebrate fossils [3][4] enhancing our knowledge of regional Pleistocene vertebrate biostratigraphy [5][6][7][8] and biogeography [9][10][12].

The Dubois collection includes fossil shells showing evidence that the H. erectus that lived in the region collected freshwater mussels for consumption and then utilized their shells for tool production. Interestingly, among these is a fossil shell carved with a geometric pattern dating from 380–640 ka, making it the world’s oldest hominin-made engraving [11].

Dates 

The Trinil fossils were previously thought to date to approximately 0.9 Ma. This date was based on the similarities between the Trinil and the ~0.9 Ma Grenzbank fauna in nearby Sangiran [30]. However, a recent chronological study [14] of the site using  ⁴⁰Ar/³⁹Ar, paleomagnetic, and luminescence (pIRIR290) dating revealed that the fossil layers are actually much younger, and that they belong to at least three distinct highly fossiliferous units with different ages.

The study reveals that Trinil has three distinct fossil-rich layers situated at low water level, BBC-1, BBC-2, and T2, instead of a single H.K. layer as Dubois proposed. Two distinct, fossil-rich channels fill the site. The stratigraphically lower Bone-Bearing Channel 1 (BBC-1) dates to 773–830 ka, while Bone-Bearing Channel 2 (BBC-2) is much younger at 380–560 ka. The Batu Gajah Formation is associated with the pre-terrace strata [12][13] and contains both fossil-bearing channel infills. Moreover, at low elevations comparable to BBC-1 and BBC-2, considerably younger T2 terrace deposits are found [14]. 

The findings indicate the existence of Early and Middle Pleistocene, and possibly even late Middle to Late Pleistocene, fossil-bearing sediments in the previously excavated region. This implies that Dubois discovered fossils from at least three distinct periods, each with its own unique age [14][13]. 

The skullcap, the two molars, and the 1897 premolar were most likely recovered from BBC-1 (773–830 ka) and/or BBC-2 (380–560 ka) deposits [14][13]. However, it is important to note that the dates are for the sediments where the fossils were found and not the fossils themselves. The fossils were not in situ and instead were deposited in a fluvial context and therefore can be much older than the sediments. 

Additionally, given that the morphology of the Trinil 2 skullcap resembles that of the early Pleistocene H. erectus fossils from Grenzbank. The molars and skullcap likely belonged to an older reworked fauna from BBC-1 or even much older deposits. Despite the younger age of these deposits, the fossils could still be as old as those discovered in Grenzbank [14][13].

Femora II-V, which likely originated from BBC-1 and/or BBC-2, have been considered morphologically similar to the H. erectus from Zhoukoudian in China [31], while others have pointed out similarities with modern humans [32][33]. However, a study using CT analysis on Femora II-V suggested that while they can be classified as H. erectus, Femur I on the other hand which Dubois claimed to be the same species as the skullcap, does not exhibit any characteristics of early hominins. Instead, it fits comfortably within Homo sapiens [34][13].

Due to these new stratigraphic and dating analyses [14][13], as well as morphological comparisons [34], it now appears that Femur I is significantly younger than the skullcap and Femora II-V, and it may even be contemporaneous with the sediments it was deposited in. If we consider a T2 age, which spans from the terminal to the Middle to Late Pleistocene, Femur I would be the oldest known modern human fossil known in Southeast Asia [29][13].