Duration of the Out Of Africa Migration, it was much slower than believed

Reading a paper about the peopling of Sundaland by Homo erectus (Husson et al., 2022, Javanese Homo erectus on the move in SE Asia circa 1.8 Ma, I learned something about the time it took these ancient hominins to people this southeastern part of Asia, and it was much longer than I expected.

The authors report that " H. erectus reached Java and dwelled at Sangiran, Java, ca. 1.8 Ma.... The dispersal of H. erectus across Sundaland occurred over at least tens to hundreds kyr."

Migration pathways of H. erectus across Sundaland during Early Pleistocene. (a) Reconstructed landscape ca. 1.8 Ma (model goSPL32). The underlying physiography, with overprinted modeled rivers, converts into a map of resistance to displacement that serves to compute the migration pathways, either using a least-cost path algorithm shown for the Myanmar entry point (purple), or a mechanistic movement model (SiMRiv30) exemplified for the Vietnam entry point (orange). Blue star locates the Thailand entry point.. Husson et al., 2022

The map above shows how different the landscape was around 1.8 million years ago, with an emerged area spanning not only the Indonesian islands but also a land bridge from Vietnam into Sundaland, separating the seas of China and Thailand. It also depicts in orange and purple, possible entry routes from the mainland Asian area to Sanigran in Indonesia.

The time spent walking

The point they studied is how long did the trek of these erectus people take. After all, there are remains in China and also in Georgia, at the Dmanisi site, around 1.9 to 2 million years old, and the Indonesian remains are just 100,000 years younger. Let's see what the authors say (my comments are underlined) and I highlight some relevant parts:

" How long did the journey take? While least-cost paths are 6,000 km long on average from mainland Asia to Java (Fig. 4a), the mean travelled distances are 50 times longer for Markovian walks (327 x 10³km), regardless of the north entry point, and almost always in excess of 100 x 10³ km. The corresponding time span depends on the migration speed. Estimates are available for more recent palaeolithic hominins, ranging from 1 to 10 km/yr [I will comment on this point further down]. At similar speeds, the least-cost paths convert into an optimal travel time of 0.6 to 6 kyr, which is thus the minimal time to cross Sundaland. Markovian walks imply much longer durations before hominins reached Java, between 25-40 kyr at fast moving rates, and 250-400 kyr at slow moving rates (Fig. 4b). These relatively long delays at average or slow moving rates are difficult to reconcile with the early arrival to Java ca. 1.8 Ma, following departure from Georgia or China only a few 100 kyr before. We suggest that because these characteristic travel times scale with the characteristic time scale of changes of the physical environment, external forcings impelled hominins to migrate and boosted their displacement.... We find that H. erectus migrated across Sundaland at remarkably fast rates (>10 km/yr) with respect to their prior journey across Eurasia, and to the displacement rates of other lineages of the genus Homo"

In other words, they estimate that the joureny could take between 25,000 and 400,000 years, to move from northern Thailand or Vietnam to Sanigran in Indonesia. And say that the erectus people walked faster than that because there were external forces pushing them forward (?).

They adopt a quicker pace to fit the timeline of not more than 100,000 years between the oldest Asian specimens in China and Georgia, and those found in Indonesia. So, the authors suggest a slow pace across Asia from Georgia to East Asia, and then a very quick entry into Indonesia.

The paper assumes a speed of 10 km per year for Homo erectus which is the speed at which the fastest "recent palaeolithic hominins" move at (see further up, where it places that speed at 1 to 10 km/y). In my opion, this is quick.

So, just taking straight lines as a course (this would assume the travelers knew in advance where they were going, which they didn't), the distance however from northern Thailand to Java is roughly 3,000 km (2,000 miles). Which they. Taking figure 4b in the paper, at 10 km per year this journey would have required between 7 and 62 ky, on average 32 ky. If these people would have moved at the slower limit found in more recent palaeolithic hominins, they would have required 320 ky to do so (70 to 620 ky).

Interestingly, the modern humans of the Out Of Africa wave are said to have trekked from the Sinai (starting 60 kya) to Beringia in only 40,000 years. That is a much longer distance (10,700 km - 6,700 mi) than the one into Sunda from northern Thailand. So, moving at the pace of the erectus that peopled Sunda, to cover this distance that is 3 times longer, it would have taken almost 100 ky on average. So if they set out 60 kya, they still haven't arrived in America! And walking at the lowere speed limit, it would have taken 1 million years.

Something is wrong in the dispersal speeds.

The paper seems to understand this implication, because it ends with an interesting conclusion: "At a more general scale, our TCN ages obtained for the first appearance datum of H. erectus in Sangiran advocate for an early peopling of Sundaland by H. erectus, contemporary with their Chinese and Georgian counterparts, whose endocranial capacity was much smaller than specimens from Sangiran. These results reignite the long-standing controversy over the origin and dispersal pathways of archaic humans and invites a reexamination of the out-of-Africa paradigm, which provides a global roadmap for the dispersal of the genus Homo, but which one-way direction may be questioned."

Directionality suggests Asia-To-Africa migrations, and the smaller-brain sizes of contemporary erectus in China and Georgia vs. Indonesia requires some explanation. Why would the more advanced erectus be those most distant from Africa, in Sangiran?

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2026 by Austin Whittall © 

Mosquitoes with a taste for human blood evolved first in Asia than in Africa, and did so 2.9 to 1.6 Ma

A research paper published in Nature on February 26, 2026, used some interesting genetic studies to support the idea that mosquitoes in Southeast Asia evolved a taste for human blood (anthropophily) at the time that the first hominins reached the area (Homo erectus). It is an interesting paper as it used data from the study of insects to confirm some ancient anthropological dates.

These are the details of this recent paper published in Nature on Feb. 26, 2026: Singh, U.S., Harbach, R.E., Hii, J. et al. Early hominin arrival in Southeast Asia triggered the evolution of major human malaria vectors. Sci Rep 16, 6973 (2026). https://doi.org/10.1038/s41598-026-35456-y

An ancient relationship: homo and mosquitoes. Source

In the abstract, it mentions that some species belonging to a group called Leucosphyrus, part of the Anopheles mosquitoes that live in Southeast Asia not only transmit malaria, they are also fine tuned to drinking human blood. Other mosquitoes in the region occupy different niches, feeding off other monkeys (non-human primates or NHP) and transmit NHP malaria variants. This study analyzed the genes of mosquitoes from 11 species (including their mtDNA) and built dated phylogenetic trees for them.

The paper found that the original, basal species of mosquitoes in this region were monkey-feeding ones, living in what is now Malaysia, Borneo, and the Indonesian islands of Java and Sumatra, when this area was linked by a now submerged continental shelf, into Sundaland. During that time (Pliocene) the area was covered by a lush rainforest. The authors believe that "anthropophily most likely evolved once, involving adaptive introgression, in the early Pleistocene in Sundaland, giving rise to multiple descendent anthropophilic species. Such early origination of anthropophily must necessarily have been in response to the arrival of early hominins (Homo erectus) rather than anatomically modern humans, likely associated with loss and fragmentation of rainforests during the early Pleistocene. The early origination of anthropophily also provides independent non-archaeological evidence supporting the limited fossil record of early hominin colonization in Southeast Asia around 1.8 Mya." This date is roughly when Homo erectus reached the area.

The article reveals that mosquitoes are host-specific and that many genes are involved in this specificity. So "multiple genetic changes at these and other genes are required for the evolution of anthropophily, i.e. a strong, evolved preference for human blood. It is not surprising therefore that anthropophily is uncommon amongst the ~ 3500 known mosquito species. It is therefore more parsimonious to consider that anthropophily within a taxon has a single origin. Accordingly, it is improbable that there were multiple independent switches to anthropophily in the human-preferring species of the Dirus and Balabacensis Clades, which diverged around 1.3–0.5 Mya. Even taking into account some uncertainty in the molecular clock used for these divergence estimates these clades far predate the arrival of anatomically modern humans in SE Asia 76,000–63,000 years ago. We therefore reject with confidence the hypothesis that anthropophily in the Leucosphyrus Subgroup evolved in response to the arrival of modern humans in SE Asia."

Therefore it was an ancient event, long before human beings appeared in Africa.

The article adds that "Using the same molecular clock rate as applied in this study and a species tree, anthropophily would be inferred to have evolved ~ 509,000–61,000 years ago in the lineage leading to the major African malaria vectors, An. gambiae and An. coluzzii... Since An. gambiae originates in West African forests the switch to anthropophilly may instead have occurred in response to modern humans entering this forested region ~ 150,000 years ago. The emergence of anthropophily in the domestic form of Aedes aegypti and the molestus ecotype of Culex pipiens, both date to within the last 10,000 years, apparently in response to growing human populations and environmental change." So, the African human-feeding mosquitoes took place independently, and later than the SE Asian event(

Finally, the article "indicates that anthropophily in the Leucosphyrus Subgroup emerged much earlier than in other anthropophilic mosquito species. If a strictly bifurcating tree were assumed, anthropophily could have evolved either: by the time of node N1 [95% CI: 2.9–1.8 Mya] and subsequently been lost in the lineage leading to An. nemophilous and An. introlatus; or evolved twice along the lineages leading to the Dirus Clade (N3, 1.6 Mya [95% CI: 2.0–1.2 Mya]) and the Balabacensis Clade (N5, 0.5 Mya [95% CI: 0.7–0.4 Mya]).... We therefore consider the most parsimonious argument for the evolution of anthropophily in the Leucosphyrus Subgroup to be that it evolved once only through the process of adaptive introgression at nodes N1/N2 as this accommodates the multiplicity of genes underlying the trait and negates any need to invoke loss of this trait. According to this hypothesis, anthropophily would have evolved between the extremes of the N1 and N2 confidence intervals i.e. between 2.9 and 1.6 mya (Fig. 3) when all the lineages were in Sundaland, and prior to the divergence of the Dirus Clade (node N3, 1.6 Mya [95% CI: 2.0–1.2 Mya]) further north in Indochina. This hypothesis could be tested against the above alternatives by identifying the genes underlying anthropophily and characterizing their evolutionary history."

The interesting part in the preceding text is that dates, which on the "older" tip of the scale reach 2.9 and 2 million years back. This is in line with the recent findings in Eurasia about an early Homo erectus dispersal across Asia.

The authors continue (highlight is mine): "Dating of the evolution of anthropophily in the Leucosphyrus Group to 2.9–1.6 mya overlaps with the earliest proposed date for the arrival of early hominins (Homo erectus) into Sundaland at 1.8 Mya, but not with the more recent proposed date of 1.3 Mya. Our findings suggest that anthropophily in the Leucosphyrus Group emerged in Sundaland in the early Pleistocene in response to the arrival of early hominins who must have not only been present in this region by this time but must have been in substantial numbers to drive adaptation to human host preference. This supports the hypothesis of Husson et al. that early hominins were present and abundant in Sundaland ~ 1.8 Mya, prior to their dispersal via land bridges to Java. Middle Pleistocene fossils of Homo erectus indicate their prolonged occupation on the exposed Sundaland landmass, likely associated with extensive river systems. In the context of the very fragmentary nature of the fossil record in tropical SE Asia our findings contribute an important piece of evidence to the broader puzzle of the colonization of hominins in insular Southeast Asia".

Closing Comments

These findings add up to the other evidence mentioned in recent posts, suggesting a very early out of Africa (or even an Eurasian origin of hominins!). I have tried to learn more about American mosquitoes that feed on humans, but research seems to be scarce in this area other than focusing on the main vectors of malaria, dengue, yellow fever, Zika, and chikungunya (see this 2022 paper as an example). If there are anthropophilic mosiquitoes in America that could be dated, what dates would they provide? (several strains reached America after the European discovery, from Africa and Eurasia, the Aedes aegypti vector of yellow fever and Zika) originated in Africa and arrived with the slave trade, and Aedes albopictus, the "Asian tiger mosquito" arrived 40 years ago, from Asia, in used tires! Would other local strains shed light on the date of the peopling of America?

---

Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2026 by Austin Whittall ©