A review of geological evidence bearing on proposed Cenozoic land connections between Madagascar and Africa and its relevance to biogeography
Introduction
Although Madagascar's iconic land-bound vertebrate assemblage comprises >900 species, this diversity has developed from no more than about thirty reptile, land-mammal, and amphibian clades (Crottini et al., 2012). For instance, the four extant land-mammal groups, (carnivorans, lemurs, rodents, and tenrecs), are today represented by between 10 and 100 species (Mittermeier et al., 2021). The marked difference between Madagascar's mammal suite and the one on Africa combined with its highly unbalanced/filtered nature led William D. Matthew (1915) to propose that the colonizations had resulted from over-water dispersal events (as opposed to range expansion via a former land-route) at various times in the early and middle Cenozoic, and possibly the Cretaceous. The issue was further considered by George G. Simpson (incidentally, Matthew's protégé), who developed the associated ‘Sweepstakes' colonization concept (1940). The land-mammal taxa that crossed from Africa were ‘drawn’ from a pool of small-bodied animals and transported passively to the island aboard vegetation rafts (see also O'Dea et al., 2016; Ali et al., 2021). Key to Simpson's argument was the idea that if a land-bridge had connected the two landmasses, then appreciably more colonizations should have taken place, as has been the case with the Panama Isthmus. In that example, the joining of South America with Central America (and North America) c. 3 Ma led to a large bi-directional transfer of land-vertebrate genera (e.g. Simpson, 1980; O'Dea et al., 2016). However, despite the Matthew-Simpson view having garnered wide acceptance (e.g. Krause, 2003; de Queiroz, 2005, de Queiroz, 2014; Vences, 2004; Rabinowitz and Woods, 2006; Yoder and Nowak, 2006; Van der Geer et al., 2010; Samonds et al., 2012), it has at various times been challenged (McCall, 1997; Stankiewicz et al., 2006; Masters et al., 2006; Mazza, 2014; Mazza et al., 2019). Resistance to the view is rooted in the belief that land vertebrates, especially the land-locked mammals, cannot survive such journeys, which would take a few to several weeks (Ali and Huber, 2010), due to a lack of food and an absence of freshwater. Furthermore, those waifs that endured the passages, as well as their near descendants, would have been susceptible to extirpation through selection pressures due to their restricted genetic diversity. Ali and Vences (2019) countered all three of these arguments and emphasized the fact that the sorts of land mammals that had established themselves on the remote islands and archipelagoes, for instance on the Canaries, Christmas Island, Galápagos Islands, and Madagascar, had one or more characteristics that facilitated the process: small body mass, low energy requirement, ability to go into torpor, and propensity to hibernate. When Mammalia are viewed as a whole, very few taxa emerge as victors in a ‘Simpson lottery.’
Unencumbered by these considerations, however, Masters et al., 2021, Masters et al., 2022 have recently proposed that Madagascar's land-vertebrates colonized the island via three temporally-separate causeways at 66–60 Ma, 36–30 Ma, and 12–5 Ma. It should be noted, though, that the notion of former terrestrial connections is not new. McCall (1997) surmised that a land-bridge existed 45–26 Ma based on a combination of biological and geological data. Much earlier, and well before the theories of continental drift and plate tectonics had been formulated, Wallace (1880) argued for a Cretaceous-age routeway. This was preceded by Sclater's (1864) proposal for Madagascar having once being connected to Africa, the Indian subcontinent and Southeast Asia via a now-sunken landmass that he labelled ‘Lemuria’.
As Madagascar's biota is the focus of an enormous amount of research activity, and where the interpretation of a clade's history is dependent upon knowing the time since it ancestral colonization event, that in turn reflects the mode of arrival, this review evaluates the geological evidence for former land-bridges. The proposals of Masters et al., 2021, Masters et al., 2022 are the primary focus, but it is straightforward to extend the analysis to the model of McCall (1997). Also, it is possible to consider other issues and to touch upon the palaeogeographical scenarios of Sclater (1864) and Wallace (1880).
Section snippets
Publications underpinning the Masters et al. ‘three-causeways' model
The palaeogeographical aspect of Masters et al., 2021, Masters et al., 2022 was constructed around three physiographical maps of the Mozambique Channel at the aforementioned time windows. To facilitate discussions, we label the scenarios/land-bridges ‘Early Paleocene’ for 66–60 Ma, ‘Eo-Oligocene’ for 36–30 Ma and ‘Late Miocene’ for 12–5 Ma. However, with the first and the third there are slight discrepancies with regards to the formally defined geological intervals (e.g. Walker et al., 2018);
Basis for McCall's 45–26 Ma land-bridge
McCall (1997) proposed that the Davie Ridge (Fig. 1) played a key role in the lemur, tenrec, carnivoran, and rodent colonizations. First, he presented biological arguments suggesting it was impossible for land-mammals to over-water disperse from Africa to Madagascar. He then made the case for a land-bridge, summarizing key geological data from areas along and adjacent to the Davie Ridge. Notably, the sedimentary sequence recovered at DSDP Site 242 (676-m-long) on the Macua Seamount's northeast
Davie Ridge
The Davie Ridge (Fig. 1) underpins all the land-bridge/stepping-stone routes between Africa and Madagascar. Although no parts of the high are presently exposed, the feature is now reasonably well known due to it having been studied for over half a century through geophysical surveys, dredging and drilling (e.g. Heirtzler and Burroughs, 1971; Simpson et al., 1974; Mougenot et al., 1986; Mascle et al., 1987; Coffin and Rabinowitz, 1987; Leclaire et al., 1989; Hall et al., 2017). In the last
Masters et al. land-bridge models
To support discussions of the proposals of Masters et al., renditions are presented of their three palaeogeographical maps (Fig. 11, Fig. 12, Fig. 13). Each shows land, seabed 0–100 m deep, seabed >100 m deep, and the inferred locations of volcanic centres. Notably, although they are less complex than the originals, they retain all of the information for making an assessment. Alongside each plot is a related figure from the PhD thesis of Antoine Delaunay (2018). Although Masters et al.
Evaluating the land-bridge models of Masters et al. and McCall
Due to the large amount of information that is contained within the various land-bridge scenarios, particularly those of Masters et al., it is a challenge to comprehend the various elements. It was thus decided to simplify matters by converting the four proposed pathways into topological models, and to then compare each with data-validated ones (see below). Here, the obvious physiographic elements have their own ‘sectors' for instance ‘Macua Seamount’, ‘Juan de Nova Island’, but so do the
Early Paleocene land-bridge proposal of Masters et al
The Masters et al. model for 66–60 Ma and the data-evaluated configuration are shown in Fig. 15. With the former, all sectors are assumed to been exposed. However, the analysis carried out here indicates that only the coast strip at Nacala was truly dry, whilst Juan de Nova Island was an atoll. All other elements were covered by water.
Eo-Oligocene land-bridge proposal of Masters et al
The Masters et al. model for 36–30 Ma and the data-evaluated configuration are shown in Fig. 16. With the former, the Juan de Nova-Melaky Province route was
Discussion
None of three land-bridge proposals of Masters et al., 2021, Masters et al., 2022, nor the one of McCall (1997), stands up to scrutiny. Africa and Madagascar were not connected by continuous causeways or stepping-stone chains at any time in the last 66 million years. As is the case today, the extent of dry land in the Mozambique Channel in the early, middle and late Cenozoic was negligible. A better explanation for Madagascar's land-fauna colonizations is centred around many over-water
Conclusions
Since the mid-19th century, biologists have proposed ancient land-bridges and stepping-stone paths linking Africa and Madagascar to explain Madagascar's unique biota. The hypothesized routeways are significant because they were thought to have enabled the mini-continent to be colonized by the ancestors of its extant and recently-extinct land-bound vertebrate clades (c. 30). As the Malagasy assemblage is scientifically important at a global level, the age and origin of its constituent lineages
Data availability
No primary data were generated for this contribution. At the time of writing, e-versions of the PhD theses of Delaunay (2018) and Ponte (2018) were available on-line, respectively at https://tel.archives-ouvertes.fr/tel-01865476 and https://tel.archives-ouvertes.fr/tel-01865479. If required, pdfs can be obtained from JRA.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
We thank Denis Mougenot for sharing information, and Wilfried Jokat and an anonymous reviewer for their constructive critiques of the manuscript. Sun-Yi Ali checked a late-stage draft.
References (98)
- et al.
Gondwana to Asia: Plate tectonics, paleogeography and the biological connectivity of the Indian sub-continent from the Middle Jurassic through latest Eocene (166–35 Ma)
Earth-Sci. Rev.
(2008) - et al.
Post-rift stratigraphic evolution of the Atlantic margin of Namibia and South Africa: implications for the vertical movements of the margin and the uplift history of the South African Plateau
Mar. Petrol. Geol.
(2018) - et al.
Madagascar volcanic provinces linked to the Gondwana break-up: geochemical and isotopic evidences for contrasting mantle sources
Gond. Res.
(2010) Petrological and geochemical investigation of rocks from the Davie fracture zone (Mozambique Channel) and some tectonic implications
J. Afr. Earth Sci.
(1992)- et al.
Anomalous bathymetry and palaeobathymetric models of the Mozambique Basin and Riiser Larsen Sea
Earth Planet. Sci. Lett.
(2016) The East African rift system
J. Afr. Earth Sci.
(2005)- et al.
Growth and demise of Cenozoic isolated carbonate platforms: new insights from the Mozambique Channel seamounts (SW Indian Ocean)
Mar. Geol.
(2016) - et al.
Impact of tectonic and volcanism on the Neogene evolution of isolated carbonate platforms (SW Indian Ocean)
Sed. Geol.
(2017) - et al.
Nature of the crust beneath the islands of the Mozambique Channel: constraints from receiver functions
J. Afr. Earth Sci.
(2021) - et al.
Modern sediments and Pleistocene reefs from isolated carbonate platforms (Iles Eparses, SW Indian Ocean): a preliminary study
Acta Oecol.
(2016)
The crustal structure of the Central Mozambique continental margin – wide angle seismic, gravity and magnetic study in the Mozambique Channel, eastern Africa
Tectonophys.
Evolution of the Glorieuses seamount in the SW Indian Ocean and surrounding deep Somali basin since the Cretaceous
Mar. Geol.
History of the development of the East African Rift System: a series of interpreted maps through time
J. Afr. Earth Sci.
Geophysical evidence for the crustal variation and distribution of magmatism along the central coast of Mozambique
Tectonophys.
The Zambezi delta (Mozambique Channel, East Africa): high resolution dating combining bio-orbital and seismic stratigraphies to determine climate (palaeoprecipitation) and tectonic controls on a passive margin
Mar. Petrol. Geol.
Shear velocity structure of the crust and upper mantle of Madagascar derived from surface wave tomography
Earth Planet. Sci. Lett.
The Africa–Madagascar connection and mammalian migrations
J. Afr. Earth Sci.
Development of the Mozambique and Ruvuma sedimentary basins, offshore Mozambique
Sed. Geol.
Disentangling interglacial sea level and global dynamic topography: Analysis of Madagascar
Earth Planet. Sci. Lett.
Late Cretaceous magmatism in Madagascar: palaeomagnetic evidence for a stationary Marion hotspot
Earth Planet. Sci. Lett.
The crustal structure of the southern Davie Ridge offshore northern Mozambique – a wideangle seismic and potential field study
Tectonophys.
Calcareous nannofossil biostratigraphy: historical background and application in Cenozoic chronostratigraphy
Lethaia
Mammalian biodiversity on Madagascar controlled by ocean currents
Nature
Mammals and long-distance over-water colonization: the case for rafting dispersal; the case against phantom causeways
J. Biogeogr.
Monkeys on a free-floating island in a Colombian river: further support for over-water colonization
Biogeographia
The Davie Ridge in the Mozambique Channel: crystalline basement and intraplate magmatism
N. Jb. Geol. Paläont. Mh. H.
An updated account of the vascular flora of the Iles Eparses (Southwest Indian Ocean)
Atoll Res. Bull.
A new CGMW map: the structural map of the Indian Ocean: an attempt at ocean cartography
Episodes
The African plate
South Afr. J. Geol.
The African Erosion Surface: a continental-scale synthesis of geomorphology, tectonics, and environmental change over the past 180 million years
Mem. Geol. Soc. Am.
Over-water dispersal of lizards due to hurricanes
Nature
Reconstruction of Madagascar and Africa: evidence from the Davie Fracture Zone and western Somali Basin
J. Geophys. Res.
Geodynamics of the SW Pacific: A brief review and relations with New Caledonian geology
The offshore East African rift system: new insights from the Sakalaves seamounts (Davie Ridge, SW Indian Ocean)
Terra Nova
Vertebrate time-tree elucidates the biogeographic pattern of a major biotic change around the K-T boundary in Madagascar
Proc. Natl. Acad. Sci. U. S. A.
The resurrection of oceanic dispersal in historical biogeography
Trends Ecol. Evol.
The monkey’s Voyage: How Improbable Journeys Shaped the History of Life
Madagascar: heads it’s a continent, tails it’s an island
Annu. Rev. Earth Planet. Sci.
Les mouvements verticaux de Madagascar (90–0 Ma): Une analyse couplée des formes du relief et de l'enregistrement sédimentaire des marges ouest Malgaches
The geology and geomorphology of Madagascar, and a comparison with eastern Africa
Q. J. Geol. Soc. Lond.
Tectonic development of the western branch of the East African rift system
Geol. Soc. Am. Bull.
Atoll vegetation and salinity
Pacific Sci.
The offshore East African Rift system: structural framework at the toe of a juvenile rift
Tectonics
Microcontinent formation around Australia
The vegetation of Aldabra Atoll: preliminary analysis and explanation of the vegetation map
Phil. Trans. Roy. Soc. Lond. Ser. B
Site U1476
Proc. Int. Ocean Disc. Prog.
Madagascar’s paleoposition: new data from the Mozambique Channel
Science
Revised lithostratigraphy of the Mesozoic-Cenozoic succession of the onshore Rovuma Basin, northern coastal Mozambique
South Afr. J. Geol.
Late Cretaceous vertebrates of Madagascar: A window into Gondwanan biogeography at the end of the age of dinosaurs
Cited by (10)
Slow and steady wins the race: Diversification rate is independent from body size and lifestyle in Malagasy skinks (Squamata: Scincidae: Scincinae)
2023, Molecular Phylogenetics and EvolutionCitation Excerpt :On the other hand, the presence of rich and structured biotas on continental islands may in turn limit the opportunities for the diversification of new colonisers (McPeek and Brown, 2007; Toussaint et al., 2015; Ali, 2017). Madagascar is the fourth largest island in the world (ca. 587,000 km2), initially part of the Gondwana supercontinent until it separated from Africa ca. 160–130 million years ago (Ma), it remained isolated from all other landmasses since ca. 80–65 Ma (Ali and Aitchison, 2008; Ali and Hedges, 2022). Madagascar’s Cretaceous vertebrate biota was likely almost entirely depleted during the mass extinction at the boundary between the Cretaceous and Paleogene (K-T boundary) (Samonds et al., 2013) and time-calibrated phylogenies and biogeographic reconstructions suggest that most present-day biodiversity results from subsequent colonisation events rather than vicariance (Vences et al., 2003; Yoder and Nowak, 2006; Vences et al., 2009; Crottini et al., 2012; Samonds et al., 2012; Welt and Raxworthy, 2022).
Drainage divide migration and implications for climate and biodiversity
2024, Nature Reviews Earth and EnvironmentBiogeography of Greater Antillean freshwater fishes, with a review of competing hypotheses
2024, Biological ReviewsThe macroevolutionary impact of recent and imminent mammal extinctions on Madagascar
2023, Nature CommunicationsThe colonisation of Madagascar by land-bound vertebrates
2023, Biological Reviews