Nine million years ago, elephants invaded South America

ResearchBlogging.org

For much of the past 130 million years South America was an island continent, and on it organisms evolved in "splendid isolation." Mammals, especially, evolved into forms not seen anywhere else, and while some mammalian immigrants made their way to South America during the past 30 million years it was not until about three million years ago, with the closing of the isthmus of Panama, that large animals from North and South America began to wander across the new landbridge and mix with the endemic faunas. This is why there were elephants in South America and giant ground sloths in North America, but a new study just published in the Journal of the South America Earth Sciences suggests that the timing of this "Great American Interchange" was a little different.

As stated by scientists Kenneth Campbell, Donald Prothero, Lidia Romero-Pittman, and Nadia Rivera, in order to properly understand earth history you need to know when events occurred, but for years the ages of recent geological deposits in the Amazon Basin have remained contentious. If paleontologists are going to understand when the earliest North American mammals arrived in South America, the ages of the deposits in Amazonia have to be ascertained. To this end they investigated the magnetostratigraphy of a site along the Madre de Dios River in Peru, a site of particular importance given a controversial fossil found there.

In 1996 Romero-Pittman described the prehistoric elephant Amahuacatherium peruvium from the bottom of the outcrop at the Madre de Dios River site, and later studies of the locality assigned it to a Late Miocene age (about 9.5 million years old). If correct, this would place it millions of years before the major pulse of the Great American Interchange, and this led other authorities to challenge the dating of the site and suggest that Amahuacatherium was really a specimen of the more recent genus Haplomastodon. The age of the deposits from which this animal was removed make all the difference in settling this debate.

After testing samples obtained from almost the entire Madre de Dios Formation, the scientists found that the results closely matched those reported from the same site nine years ago. Deposits at the bottom of the formation dated to about 9 million years ago, while those at the top were about 3 million years old. This confirmed the antiquity of the Amahuacatherium bones from the underlying Ipururo Formation, estimated to be about 9.5 million years old. The remains certainly did not come from the more recent Haplomastodon, especially since the minerals in the Ipururo Formation show a different paleomagnetic direction than those above. As the authors state, since the polarity of the earth's magnetic field has been normal for the past 800,000 years or so, this difference rules out the idea that these bones are from a more recent kind of fossil elephant.

What this means for the history of the Great American Interchange is that some of the larger animals began to make their way south much earlier than previously thought. Clearly the ancestors of Amahuacatherium did not wait for the Panama land bridge to close before shuffling across, nor did other prehistoric mammals such as tapirs and peccaries which are also found in these Late Miocene deposits. Together they may represent an initial pulse of dispersal from North America into South America millions of years before the continents became connected. The great exchange which occurred after the isthmus of Panama formed cannot be denied, but it appears that the pattern of dispersal and evolution of mammals in the western hemisphere during the last 10 million years was more complicated that previously understood.

For more on controversial fossils from South America, see my post on Ameghino's "elephants"

Campbell Jr., K., Prothero, D., Romero-Pittman, L., Hertel, F., & Rivera, N. (2010). Amazonian magnetostratigraphy: Dating the first pulse of the Great American Faunal Interchange Journal of South American Earth Sciences, 29 (3), 619-626 DOI: 10.1016/j.jsames.2009.11.007

More like this

The mylodontid ground sloth Thinobadistes also made a very early interchange into North America during the Miocene, something like 13 million years ago. I guess this is something we should expect, given that there was still an 'interchange' of primates and rodents between South America and Africa even after they had fully split.

... while some mammalian immigrants made their way to South America during the past 30 million years it was not about three million years ago, with the closing of the isthmus of Panama,

S/B "while some mammalian immigrants made their way to South America during the past 30 million years it was until not about three million years ago, with the closing of the isthmus of Panama,"

... especially since the minerals in the Ipururo Formation show a different paleomagnetic direction than those above. As the authors state, since the polarity of the earth's magnetic field has been normal for the past 800,000 years or so, this difference rules out the idea that these bones are from a more recent kind of fossil elephant.

This doesn't make sense. According to Wiki Haplomastodon was "endemic to South America during the Pleistocene from 1.810 Maâ11,000 years ago, living for approximately 1.789 million years.", which includes times prior to 800,000 years ago.

Although I can't get into the actual article, a quick look at the abstract explains the problem:

To refine the chronostratigraphy of the Madre de Dios Formation, we report here the magnetostratigraphy of an outcrop on the Madre de Dios River in southeastern Peru. A total of 18 polarity zones was obtained in the ~65-m-thick Cerro Colorado section, which we correlate to magnetozones Chrons C4Ar to C2An (9.5â3.0 Ma) based on the prior 40Ar/39Ar dates.

IOW it's not the fact that the magnetic polarity is reversed, but the exact number (18) of intervening reversals.

Interesting post, Brian. Perhaps there were islands close enough together so the elephants could âisland-hopâ and swim very short distances to reach S. America.

In your first paragraph did you mean to write: ââ¦it was not till about three million years ago, with the closing of the isthmusâ¦â. And in paragraph 3: ââ¦this would place it (not is) millions of yearsâ¦â.

By JJ Anderson (not verified) on 20 Apr 2010 #permalink

I'm sorry I didn't see that AK already noticed the missing word!

By JJ Anderson (not verified) on 20 Apr 2010 #permalink

I hear Indian elephants have been found swimming extremely far off shore recently. How far apart were those islands? It begins to make more sense.

Ancient xenarthrans included a semi-aquatic armored sloth-like creature whose name eludes me, 'least according to my error-riddled Dougal Dixon book. Maybe the sloths also went for a swim?

-

By CS Shelton (not verified) on 20 Apr 2010 #permalink

Elephants are actually very good swimmers. A group brought to the Andaman Islands for logging over a century ago, and later allowed to go wild, have been observed swimming distances of over thirty kilometers between the islands. They are distantly related to dugongs and manatees and may have gone through a semi-aquatic stage in their early evolution because they share several anatomical features in common with sea mammals, such as internal testes.

Did they use their swimming trunks?

@CS Shelton: You may be thinking of Thalassocnus.

By Andreas Johansson (not verified) on 20 Apr 2010 #permalink

Oh, any little pachyderm can swim, any little bird can fly ... Oh no, that doesn't sound right.

That's pretty amazing what facts can be established. It all looks so easy though - who imagines the years it takes to plan, gather evidence, and analyze it or the decades spent by other people in the past to establish various techniques? Time to revise the books again ... I wonder if future generations will be using 'ebooks'; even in my era some of my textbooks were marked up with so many new discoveries that some pages were close to illegible.

By MadScientist (not verified) on 21 Apr 2010 #permalink

@ak: It's not only the 18 magnetic reversals that rule out Anhuacatherium being a Pleistocene Haplomastodon, but even more the fact that there's a high-quality argon-argon date on the lowest part of the deposits (and above the mastodon, peccaries, and tapirs), which constrains all of their ages even without the magnetics. PLUS the reversed polarity of the Ipururo Fm. around the mastodon rules out it being younger than 800,000 years, which is most of its late Pleistocene history (and that's what the critics were claiming--that it was less than 800,000 years old).

The mylodontid ground sloth Thinobadistes also made a very early interchange into North America during the Miocene, something like 13 million years ago. I guess this is something we should expect, given that there was still an 'interchange' of primates and rodents between South America and Africa even after they had fully split.

Ancient xenarthrans included a semi-aquatic armored sloth-like creature whose name eludes me, 'least according to my error-riddled Dougal Dixon book. Maybe the sloths also went for a swim?

I hear Indian elephants have been found swimming extremely far off shore recently. How far apart were those islands? It begins to make more sense.

Ancient xenarthrans included a semi-aquatic armored sloth-like creature whose name eludes me, 'least according to my error-riddled Dougal Dixon book. Maybe the sloths also went for a swim?

This is why there were elephants in South America and giant ground sloths in North America, but a new study just published in the Journal of the South America Earth Sciences suggests that the timing of this "Great American Interchange" was a little different.