Land bridges to Madagascar

 

Sweepstakes distribution as envisaged by G. G.Simpson

One of the first posts on this blog (here) was about sweepstakes distribution as proposed by George Gaylord Simpson. Simply put it envisaged that animals endemic to Madagascar, such as lemurs and tenrecs, as well as now extinct hippopotami, crossed the Mozambique Channel on rafts of vegetation. 

This has been widely accepted and bolstered by evidence that ocean currents in the Cenozoic might have been more favourable to rafting than they would be today (here).

Now a cross-disciplinary group has taken a fresh look at the question and reached quite a different conclusion (here). They begin by questioning whether small primates or tenrecs could survive on the crossing which they estimate could take several months. Further they dismiss the idea that a hippo could swim to Madagascar; their arguments on this point are convincing. Could a raft be strong enough to support a hippo during the crossing (as pictured in the cartoon above)?

Verreaux's Sifaka (Propithecus verreauxi)

They suggest that there has been too much focus on the width of the Mozambique Channel and not enough on its depth. They present a mass of data, bewildering to me, but supporting the possibility of land bridges from Africa to Madagascar. The first of these at the Cretaceous-Palaeocene boundary might have been used by the ancestors of lemurs and tenrecs (their African cousins are lorises and otter shrews).

A land bridge at the Eocene-Oligocene boundary could have been used by the ancestors of Malagasy rodents and Malagasy carnivores such as the fossa. Finally there was an incomplete land bridge at the Miocene-Pliocene boundary that might be a more plausible route for a hippopotamus to cross.

The obvious question is why the Malagasy fauna is so depauperate. Why are there no monkeys, large cats or ungulates? The authors suggest they may have made it to Madagascar but gone extinct. Fortunately for this argument, the fossil record on Madagascar is very poor.

And what about that other discussion of land bridges versus rafting - how did hystricognath rodents and the ancestors of Neotropical primates reach South America (see previous post)?

Aye-aye captain of its own raft

An aye-aye (Daubentonia madagascariensis) foraging
Joseph Wolf c. 1863 Wikipedia Commons (public domain)

We are used to molecular data shaping our view of evolution, so it is all the more delightful when a morphological study shakes the mammalian tree.  The pleasure is no less for it involving that strange Malagasy primate the aye-aye (Daubentonia madagascariensis).

It is widely accepted that Madagascar was colonized by mammals rafting across the Mozambique Strait (see previous post). The lemurs (Lemuriformes) and the aye-aye (within its own Infraorder Chiromyiformes) have hitherto been thought to be descended from a single founder.

New tree for strepsirrhine primates from Gunnel et al. 
Nature Communications 2018 (here) CC

Now a study in Nature Communications paves the way to a new scenario. The historical background is interesting. George Gaylord Simpson (a proponent of the rafting hypothesis) once proposed a new fossil species of strepsirrhine primate Propotto leakeyi from the early Miocene of Kenya. Subsequently this was reinterpreted as a fruit bat. Now Gunnell and co-authors have unearthed the specimens from the National Museums of Kenya and compared the morphology (especially the dentition) both to the aye-aye and to Plesiopithecus teras from a late Eocene site in Egypt.

Firstly, they show Simpson was right about Propotto. Secondly, they construct a tree (combining morphological and molecular data) showing the split between Lemuriformes and Chiromyiformes occurred in the Eocene. Thirdly, it is most parsimonious to assume two separate rafting events with the ancestor of the aye-aye drifting to Madagascar on its own raft. Finally, both rafting events are likely to have occurred in the Miocene, which explains the lack of fossils of earlier date.

Placentation in the aye-aye

Allantochorion of the aye-aye. From Hill & Burne 1922 (here)

Milne-Edwards briefly described the placenta of the aye-aye (C R Acad Sci 1884), but the first complete account is by Hill and Burne (here). They described the villous nature of the allantochorion and the presence of chorionic vesicles. This contributed to Hill's later characterisation of the "lemuroid stage" of placentation where the placenta is diffuse, non-deciduate and epitheliochorial (reviewed here).

Plesiorycteropus was a giant tenrec

Innomate (pelvic bone) of Plesiorycteropus an extinct mammal
from Madagascar (Drawing by Forsyth Major 1908)

The recent fauna of Madagascar included species that were much larger than those found today an example being the gorilla-sized lemur Megaladapis. One of these fossils Plesiorycteropus has defied interpretation. It has been compared to the aardvark and other myrmecophagous (termite-eating) mammals such as pangolins and anteaters.

The most thorough study of the subfossil material, by McPhee (here), defined two species, P.  madagascariensis and P. germainepetterae, and erected a new Order Bibymalagasia to accomodate them. More recently Asher (here) assigned Plesiorycteropus to the superordinal clade Afrotheria.

Phylogenetic analysis based on collagen sequencing has
Plesiorycteropus as sister to the tenrecs from Buckley 2013 (here)

A recent study (here) relied on molecular sequencing of collagen from a fossil Plesiorycteropus and a range of living mammals. The technique, which uses soft-ionization mass spectroscopy, is a promising alternative to sequencing of ancient DNA. The resultant tree confirmed that Plesiorycteropus belonged to Afrotheria but related to tenrecs rather than the aardvark. Therefore it is part of the radiation from a single colonisation event (Sweepstakes Distribution).

Tailless tenrec (Tenrec ecaudatus) the largest Malagasy tenrec
weighs about 1 kg. Photo by Markus Fink (GNU Free Documentation Licence)

Estimated size of Plesiorycteropus was 6-18 kg so it certainly was a giant compared with living Malagasy tenrecs and the giant otter shrew which weigh about 1 kg.

Evolution of Neotropical primates

Geoffroy's spider monkey (Ateles geoffroyi)
Wikimedia Commons

Derek Wildman's Group at Wayne State University have re-examined the evolution of Neotropical primates (here). Using genomic data from 36 species they derive a tree with few surprises but better support than in other recent analyses.

What I found interesting is their attempt to place platyrrhine evolution in a biogeographical context. They suggest that the most recent common ancestor of extant species lived in what today is the Amazon rain forest, Guiana Shield and Northern Andes. Then, however, this was largely an area of lowlands and mountains. In support of their interpretation, known fossil sites are within this region. They further suggest that diversification of platyrrhines occurred with the establishment and development of the Amazon rain forest. 


Placenta of the white fronted capuchin (Cebus albifrons)
From Carter and Mess (here)

The founder of the Neotropical primates arrived in South America during the Oligocene when the continent was completely isolated from other land masses. Personally I favour the view (espoused here) that they came by a transatlantic route. However, Jameson Kiesling et al. suggest they may have come directly from Asia. In support they cite the occurrence of all the early fossils on the west side of the continent and recent evidence that platyrrhines may have emerged in Asia rather than Africa. In either case the means of dispersal was most likely by rafting (see previous post on sweepstakes distribution).