A Mammalian Lost World in Southwest Europe During the Late Pliocene – PLoS ONE

There’s a very interesting new paper, through which prospective readers are free to roam and explore at will, by Alfonso Arribas et al, in which the site of Fonelas, Granada in southern Spain is described, where excavations have revealed that around 1.8 million yearsfp Figure 3. Plan view of part of Fonelas P-1 site (Trench B) with skulls of Gazellospira and ago, a vast suite of mammalian fauna from Asia, Europe and Africa congregated, affording us a unique glimpse into a previously hidden corner of the world at a time when the first hominids are thought to have made their appearance on the European stage. Moreover, these findings will prompt a great deal of thought as to how and why large mammals from such discrete and distant geographical locations came to occupy an area known today as the Guadix Basin, part of the Betic Cordillera. By way of a more formal introduction, here’s the abstract:


Over the last decades, there has been an increasing interest on the chronology, distribution and mammal taxonomy (including hominins) related with the faunal turnovers that took place around the Pliocene-Pleistocene transition [ca. 1.8 mega-annum (Ma)] in Europe. However, these turnovers are not fully understood due to: the precarious nature of the period’s fossil record; the “non-coexistence” in this record of many of the species involved; and the enormous geographical area encompassed. This palaeontological information gap can now be in part bridged with data from the Fonelas P-1 site (Granada, Spain), whose faunal composition and late Upper Pliocene date shed light on some of the problems concerning the timing and geography of the dispersals.

Methodology/Principal Findings

This rich fossil site yielded 32 species of mammals, among which autochthonous species of the European Upper Villafranchian coexist with canids (Canis), ovibovines (Praeovibos) and giraffids (Mitilanotherium) from Asia. Typical African species, such as the brown hyena (Hyaena brunnea) and the bush pig (Potamochoerus) are also present.


This assemblage is taxonomically and palaeobiogeographically unique, and suggests that fewer dispersal events than was previously thought (possibly only one close to 2.0 Ma) are responsible for the changes seen around 1.9–1.7 Ma ago in the fauna of the two continents.

The site at Fonelas itself has been described previously in 2006, (PDF), and this paper follows on from research published in 2007, reported at this very site, the BBC and elsewhere, whilst there is even a dedicated website, in Spanish, and the sheer number and density of the fossils found are in large part due to the scavenging activities of ancient hyena. However to bring us up to date, here’s another brief snippet from the opening paragraphs:

Lying within the western extreme of the Palaearctic, the Iberian Peninsula is known for palaeoenvironmental sites with evolutionary implications of paramount importance. Over long periods of geological time, this has been a land of transitions and physiographical heterogeneity, including the possible existence of islands in the Straits of Gibraltar (enabling exchanges with the African continent). Conceivably, throughout the Cenozoic, climatically influenced species turnover, invasions, and competitive exclusion combined with species survival produced unique associations of plant and animal species. Here, we report on the chronology and composition of the late Upper Pliocene Fonelas P-1 fossil assemblage. Analogous assemblages have not been documented in Eurasia and no other findings have been recovered in the Quaternary. This truly is a large mammal “Lost World”.

The finds seem to mostly date from 1.8 million years, a date traditionally ascribed to the Pliocene/Pleistocene boundary, also known as the Quaternary, (preceded by the Neogene) but as was reported recently, this boundary which describes the onset of global cooling which ushered in a series of intense glaciation episodes, has recently been officially re-dated some 800,000 years beforehand to 2.6 mya. This is because it was felt that the earlier date more accurately reflects the time when global cooling commenced, and moreover is opined to be of greater significance in the geological record.

The paper itself offers a graphic description of the site itself, of which only a tiny portion has thus far been excavated, and includes details of the various mammals which made their way to the Iberian peninsular,  in what the researchers believe was a single migration event. Of particular interest is how some of these mammals made their way from Africa, and owing to a lack of finds along the Levantine, they propose that the Strait of Gibraltar may have been a crossing point between the African and European continents.

Although there is speculation that island once existed there that would have allowed for such a crossing, there is no evidence that I could find in the geological record that any such islands existed  – presumably the mammals, including early Homo, would have swum between each of these putative islands, some heading north into Europe, others heading south into Africa, in a faunal (and floral) exchange.

It occurred to me that there may have been episodic freezing of the Strait, which would allow for a much neater and tidier fit than transient islands, or even a land bridge which would have had to be in existence long after the Messinian salinity crisis around 5.9 million years ago, the last time that the flow of water into and out of the Med from the Atlantic had been blocked. The result was that the Med evaporated in around 1,000 years, with the possible exception of a few briny lakes scattered across the abyssal floor.

However, the sheer volume of water flowing through the Strait in both directions would seem to dictate that the Strait would not have frozen, even on a more contrasted seasonal basis,  around the time of the glaciation in Europe around 1.8 million years ago – something would have needed to temporarily switch off the Atlantic current, whose reduced salinity would have been more prone to freezing. But these strong currents would also have been a major problem for terrestrial mammals attempting the crossing, with or without islands dotted here and there, and would therefore have presented a formidable obstacle.

However, I can find no evidence that even hints at the Strait freezing over at that time, or indeed ever, and although it might be possible that a very brief episode of (seasonal) freezing lasting only decades occurred around 1.8 mya, there might be no clues in the geologic records that could confirm or deny this. The southerly latitude of the Strait is far from the Alpine glaciation which occurred further north in Europe, and even in the recent glaciation, it is thought that icebergs were found no further south than the Bay of Biscay, which again would seem to argue against the southern Iberian peninsular being cold enough to freeze its coastal waters.

If there had been islands or a land bridge at the time, the flow of water from the Atlantic could have been severely restricted, possibly to the extent that sea levels in the Med would have fallen, as the outflow of rivers around the Med Basin isn’t enough to keep it topped up. But a thin ice sheet across the Strait of Gibraltar may have allowed enough water from the Atlantic to continue flowing into the Med without compromising the depth therein – allowing for this mysterious exchange of flora and fauna which later inhabited the ‘lost world’ of the Guadix Basin.

There’s plenty more in this paper worth checking out, most notably about how the dating was established, and the details of the impressive number of remains discovered, and it seems likely that future seasons of digging  and analysis will offer an even greater volume of data and further establish the Iberian peninsular as a unique location for establishing its importance in the search for ever clearer insights into the first archaic humans residents of Europe and the vast suite of mammalian fauna with whom they co-habited.

References: Arribas A, Garrido G, Viseras C, Soria JM, Pla S, et al. (2009) A Mammalian Lost World in Southwest Europe during the Late Pliocene. PLoS ONE 4(9): e7127. doi:10.1371/journal.pone.0007127

César Viserasa, Jesús M. Soriab, Juan J. Duránc, Sila Plaa, c, Guiomar Garridoc, Fernando García-Garcíad and Alfonso Arribasc (2006) A large-mammal site in a meandering fluvial context (Fonelas P-1, Late Pliocene, Guadix Basin, Spain) Sedimentological keys for its paleoenvironmental reconstruction (PDF)

Palaeogeography, Palaeoclimatology, Palaeoecology
Volume 242, Issues 3-4, 8 December 2006, Pages 139-168


15 thoughts on “A Mammalian Lost World in Southwest Europe During the Late Pliocene – PLoS ONE

  1. I find terribly unbelievable that the strait could have frozen at any point. We’re talking of sea water (hard to freeze) and of a region that even in the worst of the last glaciation only saw its highest mountains perpetually under ice. It’s a climatological nonsense. Plus the “brine lake” idea is even weirdest because in fact the Mediterranean is fresher than the Ocean precisely because it is rather closed and it seems that when the Black Sea was a lake (if it was), it was also a freshwater lake.

    Now, the strait was narrower in the Ice Ages because of lower sea levels and, if today is only 13 km wide there were surely moments when it was just like 5 or 6 km wide. Additionally the area is geologically active enough as to allow for islands to have appeared and vanished, I guess, even if that’s not proven. Today a healthy sportman can swim it with the help of a surfboard on a clear (I saw some guy doing it at TV), so it’s not like any totally impassable barrier, even today. Surely in the Ice Ages it would have been easier.

    There is some evidence suggesting that some 900,000 years ago, Homo erectus or habilis could have crossed it before a second wave arrived by the mainland: all oldest known sites in Iberia for Choppers and Acheulean industries are in the south of the Peninsula and the known datations suggest a south to north diffusion, rather than the opposite. This is not irrefutable but is at least very suggestive.

    Now, Homo spp. could have engineered some sort of raft to help them cross, something that animals could not. But it seems to me that the data of hominins supports that on other animals and vice versa, right?

  2. Luis – thanks for your comment and apologies for this delayed response.

    Re: your rather dismissive ‘climatological nonsense’ I would humbly submit you have no hard data to prove or disprove the Strait of Gib ever froze at c. 1.8 mya – you don’t need a huge ice sheet covering the region, just cold enough weather to temporarily freeze the surface water in the Strait, which as you say would be shallower and less saline in an ice age.

    Recall that in around 405/406 AD, the Rhine froze over, allowing the enemies of the western Roman empire free access across what had previously been an impenetrable riverine barrier, fortified on the opposite banks by the Roman military.

    Granted the Strait was saline, and a lot wider, but that doesn’t mean it couldn’t have frozen at the surface given the right local temperatures.

    Regarding the salinity of the Med vs. the Atlantic, I found this snippet at Answers.com (Strait of Gibraltar)

    “The sill of the Strait of Gibraltar acts to limit mixing between the cold, less saline Atlantic water and the warm Mediterranean waters. The latter are so much saltier that they sink below the constantly incoming Atlantic water and form a highly saline (thermohaline, both warm and salty) bottom water, called the Mediterranean outflow. ”

    So I’m not going to accept your assurances on that either.

    You may be right about islands forming and disappearing in the Strait, but you have absolutely no evidence for this, and in any case I don’t like the hop, swim and jump scenario that a chain of islands migration would prompt – African bush pigs and hyena swimming in the sea? I somehow doubt that to be true, unless of course you have evidence to the contrary.

    Why you bring surfboarding sportsmen on TV into the equation when discussing how fauna other than humans made the crossing at 1.8 million yrs bp, I have no idea.

  3. You may be right, Tim, on the salinity issue because, while the Black Sea is fed by very large and wide rivers (Danube, Dniepr, Don..), the much larger Mediterranean is not (with the notable exception of the Nile).

    But for the freezing of the strait, I still think you are most probably wrong. For example, nearby Sierra Nevada has now (in spite of its name) no perpetual snows, while in the Upper Paleolithic Ice Age it did have them but only from 2400 m. up in the northern side (2600 m. in the southern one). I have shown you that graph once and I’ll show it to you again whenever you wish. The perpetual snow limit for the last Ice Age at Gibraltar is estimated to be 3000 m. above sea level, what makes freezing of the strait a most unlikely event.

    Now, I was thinking that there could be another more logical possibility: a gigantic iceberg drifting so far south that could physically connect Africa and Europe for some time (maybe several years in very cold conditions). Just an idea but to me a much more reasonable one.

    Why you bring surfboarding sportsmen on TV into the equation when discussing how fauna other than humans made the crossing at 1.8 million yrs bp, I have no idea.

    Because it is a very good example that the Strait is not impassable at all for humans in modern conditions, much less in the past. And it may apply to some animals as well. Even today the opposite shore is visible in good weather: Gibraltar is not impassable today and that means it was not any absolute barrier in the remote past either, when it was much narrower.

  4. i still have to read the paper, yet what i immediatly thought of is that any freezing of the strait would actually by chance probably leave a geological trace, since it would only be caused by the kind of cooling from eg. huge vulcanic events, or meterorites. so perhaps we will still find an indication. otoh to drive diverse (remnant?) populations to the iberian peninsula for once suggests it was not frozen, and migration from africa northward would be counterintuitive. also an event of only a year or a few years might not serve to create stable populations. the sealevel was a lot lower at that time (well i don’t know but people here say so and it was lower a bit later), and i assume they swam.

  5. oh yes i think these saline lakes have been proven through research , since they naturally show up as salt formations during a certain era in the geological record in the mediteranean seabottom.

    not 100% it may have been a suggestion to proof it that way, and it’s just random info i picked up on the way. i actually thought it was later then 5m years that the mediteranean flooded again and in that i was wrong apparently. i will surely take more notice of the implications for that migration from now on. btw. the miocene iberian landscape must also have been very rich, to give rise to (some) great apes. one thing about great apes is that they live in abundant environs. it’s possible that the rich natural resources persisted in the pleistocene wich could partly be an explanation for the many species assembling.

  6. Found it, Onyx: the Messinian salinity crisis, when the early Mediterranean was closed for some time, c. 6 million years ago. Quite interesting.

    However this has little to do with what happened 1.8 million or 900,000 years ago, when the continents were already as they are now more or less. And it was a geological closure, not caused by any freeze. Furthermore, freezing would not shut down oceanic circulation (as only the top layer of the sea does freeze) and would also imply that high evaporation (the direct cause of the salinity crisis of the Miocene) would not be an active factor anymore (can’t freeze and evaporate at the same time, right?).

    And, Tim, I can imagine hyenas swimming. I have also watched them taking baths at Namibia’s coast. It should not be so surprising.

  7. no i was just protesting that the dry mediteranean would not have happened, i don’t think it has much ado with the events 1.8 m years back. It is an interesting link however, thx., made me curious what influx it actually had for the miocene apes of iberia, they would have had to cross back to africa before it flooded again.

  8. Luis and Onyx –

    Thanks for the additional comments – I too wondered briefly if an iceberg could have lodged itself in the Strait, and of course of that happened it would suffice just as well as a stable if temporary platform across which some large mammals could have crossed opportunistically, rather than a migration event to the north, which in an Ice Age I’d agree would otherwise be counter – intuitive, and therefore unlikely to have taken place across a chain of putative islands – it just sounds too complicated a model, but who knows.

    I mentioned the Med salinity crisis some 4 million years earlier as a point of interest and context, but of course it would have no bearing on events at 1.8 mya.

    I’m sure hyena do swim in the sea, but I can’t imagine they or other mammals swimming such a long distance with the sole intention of crossing from one land mass to another.

    On the other hand, due to the mountainous terrain either side of the Strait, both coastlines are clearly visible from either side, and I can imagine humans like erectus at around 1mya having the wherewithal to make a concerted attempt at a crossing, and quite possibly by means of a primitive raft – which might also explain the erectus presence on Flores around 840 kya.

    I guess more finds at places like Atapuerca and further south will determine the direction from which Iberia was initially populated, but I get the impression that it could well have been north from Africa and south-west from Asia, much in the same way that Fonelas was populated by mammals from both locations.

  9. I’m sure hyena do swim in the sea, but I can’t imagine they or other mammals swimming such a long distance with the sole intention of crossing from one land mass to another.

    Sounds quite silly, I agree. But remember that the region is seismically active and that tsunamis have happened before, even in historical times ar the Azores-Gibraltar fault (Lisbon earthquake of the 18th century). In such circumstances animals (and people) would just try to survive no matter what and some do, becoming stranded. If this happens to a single individual then it’s doomed but if it’s a number of them, then a whole new population may arise at their place of arrival. I understand that this sort of mechanism has worked in the past to transfer animal populations between not too distant islands (in Wallacea for example) and could have worked for Gibraltar too.

    Sincerely, to me it sounds more likely than hyenas crossing over a mountain of ice (even if it was my suggestion).

    Alternatively they might have decided to cross willingly (or pushed by hard to imagine circumstances), when the strait was narrow enough: 5 or 6 km is not such a huge distance for a dog or similar animal. Or they could have even arrived via Asia (other fauna mentioned is Asiatic).

    On the other hand, due to the mountainous terrain either side of the Strait, both coastlines are clearly visible from either side, and I can imagine humans like erectus at around 1mya having the wherewithal to make a concerted attempt at a crossing, and quite possibly by means of a primitive raft – which might also explain the erectus presence on Flores around 840 kya.

    I fully agree with that, because for humans swimming such long distances unaided is an extreme challenge. But there were also “dwarf” elephants in Flores (got extinct at the same time that H. floresiensis), which should have arrived by some other means (swimming by grade or force).

    1. nono, the dwarf elephants arrived there when much more of indonesia (strait sunda etc) was dry, and they could walk all the way from singapore to almost ambon. so even if they swam, probably not much, same for erectus.. at least that was the theory so far.

      1. Onyx: the Wallace line, that cuts between Bali and Lombok for the purposes of our discussion, implies that there was always a sea barrier at that point, because the species at both sides essentially belong to either Eurasian or Australasian continental macro-groups. There are exceptions, notably flying species and a few very mobile land mammals, of course, but the pattern is there very clearly and it means that the Minor Sunda chain was never a continuous part of Asia.

        Either humans or elephants had needed to cross the Lombok Strait in order to reach to Flores, which was never closed. And that means swimming or boating (in the human case). Today the Lombok strait it is almost three times the size of Gibraltar or Dover straits, with 35 km. (though should have been somewhat narrower in the Ice Age).

        1. okay, thx i guess, (ambon probs a bit of an exageration indeed i admit) however i am quite sure, when stegodonts were there the distance they had to swim would have been rather minimal. also i think it has been hypothised before me such would be the case. uhm i think it was eugene dubois himself who did that research, and i also saw a mention somewhere in the flores material a few years ago. (that they did perhaps only had to swim 6 or 7 miles.)

  10. i think anything within like 10, 15 miles animals would smell rotting meat, especially in some quantity. also afaik distances less then 10-20 miles do not appear a real problem for many animals,
    altho i think (at least for deer, bear?) it has been proven they need to smell the destination.
    the salinity crisis would have some effect on the fauna probably,nore eg. if there were north-south seasonal migrations. there is also a chance some faunal elements would be remnants of that event, however i think that is not what follows from the research/geology on the fossil sequences?

    it is so that smaller elements of the fauna tend to swim (not allways) smaller distances, so if the faunal elements from africa were mostly bigger species it follows they swam, perhaps, (but why) it is more special the asian (/south european) fauna went all the way, in wich case a geological proof would actually need be searched north and east of the pyrennees. i think it is not even improbable that a big influx of the north and east would have invited some of the african species to arrive at the same time. (more animals escape a forest fire then will logically immediatly establish new territory’s.)

Comments are closed.

A WordPress.com Website.

Up ↑

%d bloggers like this: