Did All Species Arise about 200,000 Years Ago?

Continuing the discussion from Dabar Conference Paper Confidentially Available:

By the way, I’ve come across a couple of papers this evening, which may be of interest to you:

Sweeping gene survey reveals new facets of evolution

The study’s most startling result, perhaps, is that nine out of 10 species on Earth today , including humans , came into being 100,000 to 200,000 years ago.

“This conclusion is very surprising, and I fought against it as hard as I could,” Thaler told AFP.

That reaction is understandable: How does one explain the fact that 90 percent of animal life, genetically speaking, is roughly the same age?

Was there some catastrophic event 200,000 years ago that nearly wiped the slate clean?

Environmental trauma is one possibility, explained Jesse Ausubel, director of the Program for the Human Environment at The Rockefeller University.

Viruses, ice ages, successful new competitors, loss of prey —all these may cause periods when the population of an animal drops sharply,” he told AFP, commenting on the study.

“In these periods, it is easier for a genetic innovation to sweep the population and contribute to the emergence of a new species.”

Here’s another related paper:

Far from special: Humanity’s tiny DNA differences are ‘average’ in animal kingdom

In genetic diversity terms, Earth’s 7.6 billion humans are anything but special in the animal kingdom. The tiny average genetic difference in mitochondrial sequences between any two individual people on the planet is about the same as the average genetic difference between a pair of the world’s house sparrows, pigeons or robins. The typical difference within a species, including humans, is 0.1% or 1 in 1,000 of the “letters” that make up a DNA sequence…

Another intriguing insight from the study, says Mr. Ausubel, is that “ genetically, the world is not a blurry place. It is hard to find ‘intermediates’ - the evolutionary stepping stones between species. The intermediates disappear.”

Dr. Thaler notes: “Darwin struggled to understand the absence of intermediates and his questions remain fruitful.”

“The research is a new way to show that species are ‘islands in sequence space.’ Each species has its own narrow, very specific consensus sequence, just as our phone system has short, unique numeric codes to tell cities and countries apart.”

Adds Dr. Thaler: “ If individuals are stars, then species are galaxies. They are compact clusters in the vastness of empty sequence space.”

At last, a clear scientific definition of what a species is! It seems the ID crowd are half-right: species don’t blur into one another, even over time, and they arise relatively suddenly. Where they go wrong is in failing to reckon with the power of neutral evolution, as the article makes clear.


Yes, I saw this too. Some puzzling features here. It will take some time to look through carefully. I’m not sure I’m convinced by the high level journalism here. I need to read the original paper, and perhaps even look at the data. Notably, this is mitochondrial data, which is not well calibrated as a molecular clock. So I would take many of the time estimates with a large helping of salt.


The pre-print abstract for this paper is Genesis 1:25.


My skim through the the version on the biology preprint server (I haven’t seen the published version in Human Evolution since it wasn’t yet available when I looked on Friday) had me very interested but immediately skeptical of the typical oversold marketing of the paper by popular outlets. Diversity of mtDNA can be a proxy for nuclear DNA diversity but certainly not always. It would not surprise me to have low mtDNA diversity (suggestive of recent origin) within a species but the species be much older than the mtDNA suggest for some of the very reasons mentioned here and elsewhere with respect to Adam and Eve and different rates of evolution of mtDNA genomes. I am also not surprised to see species as diversity islands when mtDNA alone is used. I fully expect the nuclear genome picture to be much messier. Just look at humans/neanderthal nuclear genome vs mtDNA genomes.

1 Like

My sense exactly.

That, also, is likely to not hold up when we start looking at genomic data. Because mtDNA is a single locus, without recombination, it is going to be subject to things like Muller’s ratchet and all the limits a stochastic single measurement.

For example, looking at mDNA for humans and Neanderthals, you’d think they were totally isolated species. Look at their genomic DNA, and you see interbreeding. No surprise, as genomic DNA is much more loci, with recombination, and a much more sensitive test.

Btw @Joel_Duff, great to see you. Looking forward to discussing the Genealogical Adam with you all.


Nope, not really, though many people will read it that way. That is what you get for irresponsible science journalism.

Mark Y. Stoeckle, David S. Thaler
More than a decade of DNA barcoding encompassing about five million specimens covering 100,000 animal species supports the generalization that mitochondrial DNA clusters largely overlap with species as defined by domain experts. Most barcode clustering reflects synonymous substitutions. What evolutionary mechanisms account for synonymous clusters being largely coincident with species? The answer depends on whether variants are phenotypically neutral. To the degree that variants are selectable, purifying selection limits variation within species and neighboring species have distinct adaptive peaks. Phenotypically neutral variants are only subject to demographic processes: drift, lineage sorting, genetic hitchhiking, and bottlenecks. The evolution of modern humans has been studied from several disciplines with detail unique among animal species. Mitochondrial barcodes provide a commensurable way to compare modern humans to other animal species. Barcode variation in the modern human population is quantitatively similar to that within other animal species. Several convergent lines of evidence show that mitochondrial diversity in modern humans follows from sequence uniformity followed by the accumulation of largely neutral diversity during a population expansion that began approximately 100,000 years ago. A straightforward hypothesis is that the extant populations of almost all animal species have arrived at a similar result consequent to a similar process of expansion from mitochondrial uniformity within the last one to several hundred thousand years.

I find it remarkable that they do not adequately deal with the fact that mtDNA in coding regions is not neutral. Maybe I am missing something hear, but there seems to be major oversights. Fig 7. also is particularly misleading. They use extant population size instead of historical population size.

Regardless, it is important to emphasize we do not even have a way of measuring mitochondria mutation rates yet. So the clocks are not well calibrated here. There are also also large open questions about mitochondrial transmission to offspring, and the effect of heteroplasmy. It seems like they are concluding too much from this paper, IMHO.


Hi Joshua,

Thanks for your comments, especially this one:

Yes, I saw this too. Some puzzling features here. It will take some time to look through carefully. I’m not sure I’m convinced by the high level journalism here. I need to read the original paper, and perhaps even look at the data. Notably, this is mitochondrial data, which is not well calibrated as a molecular clock. So I would take many of the time estimates with a large helping of salt.

I had a couple of questions of my own:

(i) the article says that most species today, including Homo sapiens, are between 100,000 and 200,000 years old. However, recent discoveries have pushed back the origin of Homo sapiens to 300,000 years ago, so it seems we’re an exception to the general trend. Or are the authors suggesting that fossil identifications of species are unreliable, and that only identifications which are based on mitochondrial DNA should be trusted?

(ii) From what I gather, the average lifetime of a species seems to be around 5 million years. So if, as the article says, most species are between 100,000 and 200,000 years old, then even today, it’s still pretty early on in their lifetimes. Curious.

1 Like

The rapid cycling in and out of Ice Ages is a pretty new phenomenon on the Earth. With the continued separation of the continental plates, the opening of the Atlantic Ocean creates the basis for a brand new pattern of ocean currents. But it wasn’t until the North and South American plates collided at the Isthmus that the Atlantic ocean was sealed off from the Pacific waters, and North/South cycling of currents became consistent and forceful. I believe that was about 3 million years ago.


800,000 years ago, probably because of the latest movements of the continents and/or a sufficient fixation of CO2 in the frozen sediments at the bottom of the ocean, the Earth found itself on a climate sweet spot!:

Shifts in atmospheric CO2 of only 100 ppm was enough to send the Earth into an Ice Age (with mile-high glaciers over the future New York City)… and a pendulum swing with 100 ppm less was enough to completely melt the bulk of the Earth’s glaciers!

This is not just supposition: 800,000 years = 8 ice ages. Each time there was an ice age, it correlated to CO2 of 180 ppm. And each time the ice age dissipated CO2 maxed out at 280 ppm.

So, assuming all of this is a valid analysis, 800,000 years is a pretty short blip in time. Prior to a million years ago, the Earth experience long shallow cycles of changes, punctuated by some important and dramatic ones.

And with each dramatic change came dramatic speciations.

The current article says that we see lots of populations producing variants from 100,000 to 200,000 years ago. This translates into the general idea that these new populations fall into two groups: a group that speciated 2 ice ages ago, and a group that speciated one ice age ago. This tells us that surviving unchanged for more than 2 ice ages is quite a trick!

If, instead of living after a quick succession of 8 Ice Ages, humans had internet and interesting articles to read 3 million years ago, we would probably find that 100,000 to 200,000 years “as an average” would be replaced by 1 million to 2 million years ago… or higher. While a steady stream of mutations is the heart of speciation, the bread and butter for the process is changing patterns of natural selection due to environmental factors (even if the factors might mean new food sources or new rivals for old food sources, and so on).

We aren’t looking at an average life cycle for all species at all times in Earth’s history… we are looking at a very special time in the history of the Earth … which produces speciation patterns based on global ice ages sweeping in, and knocking everybody’s card tables to the ground!

The jury is still out on that isn’t it? I mean even that link has plenty of hedges. Besides the re-measure on the dating, the face of those finds may have been more like us- sans the jaw- but the braincase was not. So is it our faces or our brains that set us apart? I’d say lots of monkeys have flatter faces.

1 Like

So, @vjtorley, this paper equivocates several things, or at least it seems so. I would go so far as to say that is methodologically flawed in the biorxiv form. I’d hope that the authors correct it in review.

  1. They equivocate TMRCA with species age. That is just false. To map to other conversations we have, they are essentially reproducing the well known results that Mito-Eve appears to arise between 200 kya and 100 kya, but this in itself tells us precisely zero information about when Homo sapiens arise.
  2. Mitochondrial clocks require a great deal of effort to calibrate and different species will have different rates. This is very difficult to do, and they have not done so adequately (or at all?). This means that they are equivocating “diversity” for “TMRCA”. They should have instead reported the substitution rate distance, not the TMRCAs, as the TMRCAs are certainly not definitive. The fact of the matter is that we do not have good estimates of mitochondrial mutation rates in most species.
  3. They appear to be using extant population size instead of historical population size. That is not warranted. The fact that humans have a few billion people in recent history, does not mean anything regarding our ancient history. The fact of the matter is that we do not have good estimates for past populations sizes.
  4. To get a TMRCA we also need to account for the degree of selection at this locus, because it is clearly not neutral. This was not adequately accounted for, and once again, will affect the results.
  5. They wildley over interpret measurements from a single loci. Such an effort is deeply flawed. For example, by mitochondria, Neandertals and Sapiens are different (but not more different than two chimps!) but when we look at autosomal genome, we see evidence of interbreeding. This directly contradicts the conclusions and reasoning of the paper.

So, to be clear, this seems to be very problematic study, with results that are wildly overstated. They have a provocative title, but are no way able to justify it.

It hard to describe this as anything other than an error of false equivolance. They take a very unreliable estimate of TMRCA, and then equivocate that “When a species arises.” Such a jump is just false. That is not correct or valid. TMRCA and speciation are not the same thing. It ignores everything else.

It seems to be falling hard for the streetlight fallacy. We can’t measure A, but we can measure B, so lets pretend by measuring B we know what A is.

The fundamental error here is summed up in an old joke scientists love to tell. Late at night, a police officer finds a drunk man crawling around on his hands and knees under a streetlight. The drunk man tells the officer he’s looking for his wallet. When the officer asks if he’s sure this is where he dropped the wallet, the man replies that he thinks he more likely dropped it across the street. Then why are you looking over here? the befuddled officer asks. Because the light’s better here, explains the drunk man.

They need to get out of the genetic streetlight, to consider a whole ton of other things. Their response would likely be “but that is hard!” to which most scientists would say: “Yes, science is hard. Get to it.”

Except they are not measuring the age of species. They are measuring TMRCA at a single locus, without even calibrating their clock.

Absolutely correct @gbrooks9. The equivocation between TMRCA and when a species arises is really poor. I’d be surprised and disappointed if they got that past peer review.




What do you think of my assessment @Joel_Duff?

What do you think of my assessment @Joel_Duff?

@swamidass I think your assessment is spot on however I have yet to see the version of this manuscript its final form. I don’t understand why it is not available from the J of Human Evolution so many days after the popular press coverage. I would assume version from the preprint server underwent editing/review prior to being published and so maybe the authors may have made some changes though the popular coverage would suggest otherwise.
To me the most critical takeaway is as you said:
“Except they are not measuring the age of species. They are measuring T_{\tt MRCA} at a single locus, without even calibrating their clock.”


By default, two consecutive Ice Ages creates a de-facto re-set to the existing species… which is mathematically unavoidable if the sentence “most species were created 100,000 to 200,000 years ago”.

8 Ice Ages in 800,000 years is quite an ordeal to put any terrestrial populations through!


No doubt. I think a “re-set” is a good way of putting it. Defining a species though raises its ugly head though. A severe bottleneck may coincide with the origins of a species but I expect that many species may look 200,000 years old but populations leading to that species may have existed in a form that we would recognize as the same species (in the morphological, phylogenetic and biological species concepts) for a million years or more. From a morphological species concept perspective the fossil record generally suggests that species have an average lifespan of 1 million years. Of course that that doesn’t preclude that recent historical events–such as ice age–may have placed a strong limit on species (or population) life spans.


I finally found a link to the article as it appears in Human Evolution. https://phe.rockefeller.edu/news/wp-content/uploads/2018/05/Stoeckle-Thaler-Final-reduced.pdf It looks like to be identical to the one found on the pre-print server.

1 Like

It is like a time capsule. It reads like it’s being written in the pre-genomic age.

Well, i’m surprised and disappointed. They got it past peer review, and this article is just full of absurd reasoning.

To the contrary, the last 2 million years has been very climate stable. The onset of ice ages was very gradual, with more than enough time for populations to adapt, which is why we do not see major extinctions during this time.


I think once a population survives the first ice age … 800,000 years ago… the odds are good (all things being equal) that it will survive 7 more. But will it survive 7 more without changing something? Species CAN change without necessarily warranting that there is a NEW species.

So… can the authors of the study … and you @swamidass… both be right…? Maybe ONLY in ways like this!:

Like we see in many cladograms, an ancient population is depicted as going all the way to an existing population (say, giraffes). And then there are all these branches breaking off from the main line. But this is deceptive! Dies this mean that the cladogram includes the fact giraffes have existed since the beginning of the cladogram’s time frame? Or should the clafogram have been more clear… that giraffes arrived later from a common ancestor.

The writers say MOST species emerged 100k to 200k years ago.

While you say that most species survived all the ice ages.

If you are right, it means the writers must, at the very least, revise their terminology to say “most NEW species” appeared 100k to 200k ago.

[Or it may be @swamidass’ position that the authors are just plain wrong about their findings!]

They are just plain wrong. Don’t give them so much credit. This study is going to get trashed be scientistis.

Also the first ice age was almost 2 million years ago (from memory).