Skip to main content

Crocodiles of Africa, crocodiles of the Mediterranean, crocodiles of the Atlantic (crocodiles part VI)

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


The Tet Zoo crocodile series is not yet finished, and here we embark on part VI in the series (see below for links to previous parts). This time, we come to the Nile crocodile lineage, and I refer here to a ‘lineage’ rather than to a species since there’s now good evidence that C. niloticus of tradition is not a single taxon [adjacent image by Graeme Churchard]. Actually, however, it now seems that the crocodiles we're talking about here aren’t even members of the same single lineage, so even that term is probably incorrect. If you’ve read the previous articles you’ll know that Crocodylus crocodiles are found in some phylogenetic studies to consist of three main lineages: an Indopacific assemblage, a New World assemblage, and an African Nile crocodile lineage. Most studies find the Nile crocodile to be closer to the New World assemblage than to the Indopacific one.

Because numerous African Cenozoic fossils have been identified as archaic members of Crocodylus, it has generally been assumed that the members of the Nile crocodile lineage evolved in situ, and hence that early members of the New World assemblage crossed the Atlantic from east to west after originating from an African ancestor.

However, those ‘archaic Crocodylus’ crocodiles now turn out to be no such thing. They are, in fact, mostly or entirely members of the Osteolaemus group. As Brochu et al. (2010) hinted in their description of the Plio-Pleistocene C. anthropophagus from Olduvai Gorge (and see Brochu 2000), the possibility now exists that the members of the Nile crocodile lineage are relative newcomers to Africa that moved in from elsewhere. Further study is required to test this idea. Incidentally, C. anthropophagus is not a member of the Nile crocodile lineage – it was found instead to be in a polytomy with a Nile crocodile + New World assemblage clade, the Indopacific assemblage, and the fossil species C. palaeindicus (Brochu et al. 2010).


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Crocodylus niloticus’ is not monophyletic

The substantial variation seen within C. niloticus across its enormous range has resulted in the naming of seven subspecies (C. n. niloticus Laurenti, 1768, C. n. africanus Laurenti, 1768, C. n. chamses Bory de Saint-Vincent, 1824, C. n. cowiei Smith in Hewitt, 1937, C. n. madagascariensis Grandidier, 1872, C. n. pauciscutatus Deraniyagala, 1948 and C. n. suchus Geoffroy Saint-Hilaire, 1807). Furthermore, the subspecies originally named for populations in Egypt, Sudan, Nigeria, Madagascar, South Africa and elsewhere were originally described as distinct species [adjacent image by Marco Schmidt].

While the phylogenetic reality of some of these taxa has yet to be confirmed, molecular results published since 2003 have shown that crocodiles typically classified together in C. niloticus are genetically diverse, with initial studies showing that ‘west African’ crocodiles (occurring as far east as Chad) are notably distinct from east African ones (Schmitz et al. 2003). Given that the name C. niloticus has been specifically associated with crocodiles from Egypt, it’s the west African ones that need a ‘new’ name, and C. suchus Geoffroy, 1807 was resurrected for this purpose by Schmitz et al. (2003, 2004). As discussed in part V of this series, one especially problematic researcher has recently suggested that this species deserves its own ‘genus’, Oxycrocodylus. This is an unfounded suggestion that we should ignore.

Schmitz et al. (2003, 2004) found that west African crocodiles from Mauritania grouped together with Chadian animals, and they also regarded the ‘Nile crocodiles’ of Gambia and Senegal as part of the same taxon (C. suchus) as well. A ‘tidy’ picture of C. suchus in the mostly arid west and north, and C. niloticus in the east and south therefore emerged. However, complication # 1 is that crocodile fossils, archaeological specimens and rock art featuring crocodiles are present right across the whole northern half of Africa. Complication # 2 is that things ain’t so tidy after all, with C. suchus being far more widespread than thought in 2003 or 2004. Read on.

‘Sacred crocodiles’ of the Sahara, Sahel and Nile

These west African, C. suchus-type crocodiles were first mentioned by French explorer Henri Duveyrier in 1858 following his discovery of crocodile tracks in the Sahara, and live specimens were reportedly captured in 1910 and 1924 (Guggisberg 1972, de Smet 1999). Various explorers, naturalists and biologists reported small, Saharan or Sahelian crocodiles from the Hoggar Mountains of Algeria, Matmata in Tunisia, Tibesti Mountains of Chad, the Tagant and Ennedi plateaus of Mauritania and other west and north African localities during the following decades, so a general pattern of crocodile presence across this region gradually emerged (Guggisberg 1972, de Smet 1999).

It’s generally thought that these C. suchus populations are, or were, relicts, hanging on in small habitat pockets while the previously green environment around them degraded and disappeared. This fits, of course, with the whole ‘Green Sahara’ model supported by palynological, sedimentological and archaeological data: the Sahara was big and arid during the Last Glacial Maximum of the Pleistocene, but it was a green, wooded, well-watered place about 7000 years ago. The Sahel then developed about 5500 years ago, creating an arid belt that cut northern African off from the greener south. The arid belt expanded and moved over the next few thousand years.

These Saharan and Sahelian animals are small (always less than 2.3 m long* in total). A few morphological differences might allow them to be distinguished from C. niloticus, including robust snout proportions and a low number of belly scales.

* This apparent maximum length is based on a specimen collected by Paul Spatz from Lake Galula (“a desert pool at the foot at an impressive cliff”), Mauritania (Guggisberg 1972).

Their restriction to small habitat pockets makes them vulnerable to extinction and, sadly, they are often killed by people. de Smet (1999) drew attention to the fact that they had seemingly been extirpated from many areas where they had persisted until recently. They seem to have disappeared from Morocco by the 1960s, Tunisian crocodiles apparently persisted to as recently as the 1970s, those from the Tagant Hills in Mauritania had apparently gone extinct by 1996, and attempts to find them in some of their last strongholds in the Ennedi plateau of Chad were unsuccessful during the 1970s and 80s. de Smet (1999) concluded that “on balance one may conclude that the population of Saharan Nile crocodile are virtually extinct” (p. 84).

However, these western and northern crocodiles are secretive, poorly known, and typically occur in remote places rarely visited by people. Accordingly, it’s possible that they still inhabit places that haven’t been that well studied. This is exactly what Brito et al. (2011) found. In Mauritania alone, they reported crocodiles in 78 localities, increasing the number of known crocodile localities by 35%. However, while we can say that crocodiles persist in Mauritania, Chad and also Egypt, the numbers concerned are small: at most places only one or two animals are ever recorded. Tracks and dead crocodiles found away from water indicate that the animals can disperse between at least some of their strongholds, perhaps suggesting that these many small groups can be considered part of the same breeding population (Brito et al. 2011). Observations show that these animals are secretive and hide as soon as they see humans. Nevertheless, continuing human persecution occurs across much of this range. People deliberately kill the animals on sight, pulling them from the water and beating them to death.

Here’s a good question: are all of these crocodiles all members of C. suchus? Brito et al. (2011) didn’t even mention this name and referred to all of their crocodiles as C. niloticus. Furthermore, if crocodiles were formerly ubiquitous and continuously distributed across north-western, northern, and north-eastern Africa, where did the ranges of C. suchus and C. niloticus end and begin? Did these species occur in sympatry? Has C. suchus always been restricted to the north and west, or was it once more widespread? Did C. niloticus previously occur in the north and west as well, and are all recent and modern northern and western crocodiles definitely C. suchus and not C. niloticus?

At this stage we can’t answer all of these questions. However, Hekkala et al. (2011) looked at the DNA of mummified crocodiles from Egypt with these issues in mind and showed that individuals of C. niloticus and C. suchus were both present in the sample. They concluded that both species occurred sympatrically in the Nile, and noted Geoffroy Saint-Hilaire’s opinion of 1807 that the distribution of C. suchus “likely extended into the western Sahara” (Hekkala et al. 2011, p. 12). It seems that the ancient Egyptians knew of these two distinct kinds of crocodiles. C. suchus – referred to as the ‘Sacred crocodile’ by Geoffroy Saint-Hilaire – was regarded as a smaller, more docile animal, and nowhere near as formidable as C. niloticus.

In and around the Mediterranean, and the Tarasque as a French crocodile

Our questions about the species-level identification of north African crocs also affects the animals that inhabited the Mediterranean. Waitaminute… crocodiles in the Mediterranean? Believe it or don’t, historical records refer to the presence of crocodiles on Sicily (Anderson 1898, de Smet 1999) and also in Syria, Jordan and Israel (Anderson 1898, Guggisberg 1972, de Smet 1999). In fact, they were still present in the Kebara swamps in Israel until the early part of the 20th century.

If all of these records are real (it mostly depends on whether you trust Anderson as a source or not), they could show that crocodiles lived around the southern and eastern coasts of the Mediterranean until surprisingly recent times.

de Smet (1999) went even further with this and wondered whether crocodiles might have been present in the river deltas and swamps of France, Spain and the Balearic islands until the “the first centuries of our era” (p. 84). A particularly weird mythological entity – the Tarasque – has been suggested on occasion to be based on garbled, distorted accounts of crocodiles encountered in the French wilderness (Lavauden 1926). That might seem hard to accept if you know what the Tarasque is meant to look like, since it’s described as having a giant spiked carapace, six limbs, a golden flowing mane, and a long, snake-shaped tail, but the problem with the Tarasque is that it’s so weird that it’s hard (if not impossible) to match it with any known creature. I’m certainly not convinced that it’s based on sightings of crocodiles, but it’s an interesting idea.

If these alleged Mediterranean crocodiles really were living in coastal places and making sea crossings, it might seem more likely that they were C. niloticus rather than C. suchus. However, Hekkala et al. (2011) showed that C. suchus most likely underwent at least some marine dispersal during its history, since they found good evidence linking Sahelian populations to coastal populations of Senegal and Gambia as well as to those of the Upper Guinea Forest Basin countries (Nigeria, Ghana and so on… more on this view of a super-widespread C. suchus in a moment).

As for whether crocodiles were really present in France, Spain, Sicily and elsewhere in and around the Mediterranean but a few hundred years ago, bring on the discoveries of bones and such required to support this view. Old eyewitness accounts are suggestive, but not good enough to be convincing.

Crossing oceans and how it gets more complicated

Until very recently it was assumed that C. niloticus (in the strictest, most restrictive sense) and C. suchus are sister-taxa, in which case their recognition as distinct ‘species’ is subjective and depends on how your ‘species-o-meter’ is calibrated [Adjacent C. niloticus image credited to MathKnight and Zachi Evenor].

Hekkala et al. (2011) found crocodiles from across western Africa (Republic of Congo, Uganda, Gambia, Senegal, Mauritania, Burkina Faso, Ivory Coast, Ghana, Democratic Republic of Congo and Nigeria) to group together in a so-called ‘western clade’. Given that this clade includes crocodiles from Mauritania and is stated several times in their study to include the relict populations of the Sahara, this clade is either synonymous with, or includes, C. suchus.

The presence of these sorts of crocodiles in the Kidepo Valley of Uganda means that thinking of them as Saharan relicts, or as western specialities, is perhaps erroneous. Even more surprising is that Hekkala et al. (2011) referred to the possible presence of ‘western clade’ crocodiles in Ethiopia, potentially meaning that “the western clade is still distributed in this region though it may be restricted to marginal habitats” (p. 4212).

The biggest surprise of Hekkala et al. (2011), however, is their finding that this ‘western clade’ does not group together with C. niloticus. Instead, the ‘western clade’ was recovered as the sister-group to the clade that includes both the New World Crocodylus assemblage and C. niloticus. The same pattern of relationships had previously been recovered by Meredith et al. (2011); Oaks (2011) also published evidence in support of it. This means that the New World assemblage is bracketed on both sides by African taxa: a topology which means that an African ancestry for the New World assemblage is arguably better supported than it was before (the idea that a trans-Atlantic crossing occurred in crocodile evolution is pretty well established). However, things are now more interesting.

As mentioned above, Hekkala et al. (2011) found most west African Crocodylus populations – not just the Saharan and Sahelian relicts – to belong to the ‘western clade’, with the mostly eastern populations of Egypt, Uganda, Kenya, South Africa, Malawi, Zimbabwe, Tanzania and Madagascar forming a different clade. I say “mostly eastern” because populations from Gabon were also included in the ‘eastern’ clade. Note also that Uganda seemingly included both members of the ‘western’ and ‘eastern’ clades (Hekkala et al. 2011). If you recall that the New World assemblage forms the sister-group to the ‘eastern’ clade, one possibility is that the African crocodiles that crossed the Atlantic to give rise to the New World assemblage actually left Africa from the east (= Indian Ocean) side, not from the ‘logical’ west (= Atlantic) side. And there’s no indication that the ‘eastern’ crocodiles in the west (like those of Gabon) have a special relationship with the New World assemblage.

An idea put forward in interviews and news articles featuring Ekon Hekkala (like this one from September 2011 at Ed Yong's NERS) is that east African crocodiles used one of the currents that flows through the Mozambique Channel and around South Africa in a south-westerly direction. From here, they got into the Atlantic and obviously dispersed east to west. Incidentally, all of this is thought to have happened within the last 6 million years: it would mean that the New World assemblage is geologically young and would be in keeping with other work which shows that modern crocodiles are recently evolved animals (Oaks 2011), not old ones.

With C. suchus (and/or all ‘western clade’ crocodiles) being ancestrally African and ‘eastern’ C. niloticus being African too, the idea that the New World assemblage came from Africa is unavoidably obvious and parsimonious. However, parsimony isn’t always the same thing as reality when it comes to phylogeny. It’s (err, arguably) at least worth considering the possibility that things happened the other way round: that is, with C. niloticus being the descendant of an immigrant lineage that comes from the Americas. The fact that C. niloticus now seems to be mostly restricted to the eastern side of Africa counts against this idea. The presence of other modern crocodile groups in Asia and Australasia also suggests that crocodiles might have spread across the eastern side of Africa before moving across the western side.

The problem, though, is that we’re basing all of this on modern populations. That C. niloticus population in Gabon (Hekkala et al. 2011) hints at the possible former presence of C. niloticus throughout coastal western Africa. If that’s true we might need to reconsider the idea that the New World assemblage represents dispersal from a population that originated on the Indian Ocean coast of Africa.

An increasingly complex picture of crocodile evolution, history and dispersal is emerging the more we learn, and it’s really exciting stuff.

For the previous parts in this series, see...

And for other Tet Zoo articles on crocodyliforms (and/or crocodylomorphs), see...

Refs - -

Anderson, J. 1898. Zoology of Egypt, Volume 1, Reptilia and Batrachia. Bernard Quaritch, London.

Brito, J. C., Martínez-Freiría, F., Sierra, P., Sillero, N. & Tarroso, P. 2011. Crocodiles in the Sahara Desert: an update of distribution, habitats and population status for conservation planning in Mauritania. PLoS ONE 6(2): e14734. doi:10.1371/journal.pone.0014734

Brochu, C. A. 2000. Congruence between physiology, phylogenetics and the fossil record on crocodylian historical biogeography. In Grigg, G. C., Seebacher, F. & Franklin, C. E. (eds) Crocodilian Biology and Evolution. Surry Beatty & Sons (Chipping Norton, Aus.), pp. 9-28.

- ., Njau, J., Blumenschine, R. J., & Densmore, L. D. 2010. A new horned crocodile from the Plio-Pleistocene hominid sites at Olduvai Gorge, Tanzania. PLoS ONE 5(2): e9333. doi:10.1371/journal.pone.0009333

de Smet, K. 1999. Status of the Nile crocodile in the Sahara desert. Hydrobiologia 391, 81-86.

Guggisberg, C. A. W. 1972. Crocodiles: Their Natural History, Folklore and Conservation. David & Charles, Newton Abbot.

Hekkala, E., Shirley, M. H., Amato, G., Austin, J. D., Charter, S., Thorbjarnason, J., Vliet, K. A., Houck, M. L., Desalle, R. & Blum, M. J. 2011. An ancient icon reveals new mysteries: Mummy DNA resurrects a cryptic species within the Nile crocodile. Molecular Ecology 20, 4199-4215.

Lavauden, L. 1926. Les Vertébrés du Sahara. Albert Guénard, Tunis.

Meredith, R. W., Hekkala, E., Amato, G. & Gatesy, J. 2011. A phylogenetic hypothesis for Crocodylus (Crocodylia) based on mitochondrial DNA: evidence for a trans-Atlantic voyage from Africa to the New World. Molecular Phylogenetics and Evolution 60, 183-191.

Oaks, J. R. 2011. A time-calibrated species tree of Crocodylia reveals a recent radiation of the true crocodiles. Evolution 65, 3285-3297.

Schmitz, A., Mansfeld [sic], P., Hekkala, E., Shine, T., Nickel, H., Amato, G. & Böhme, W. 2003. Molecular evidence for species level divergence in African Nile crocodiles Crocodylus niloticus (Laurenti, 1786). C. R. Palevol 2, 703-712.

- ., Mausfeld, P., Hekkala, E., Shine, T., Nickel, H., Amato, G. & Böhme, W. 2004. Erratum to the article Molecular evidence for species level divergence in African Nile crocodiles Crocodylus niloticus (Laurenti, 1786). C. R. Palevol 3, 177.

Darren Naish is a science writer, technical editor and palaeozoologist (affiliated with the University of Southampton, UK). He mostly works on Cretaceous dinosaurs and pterosaurs but has an avid interest in all things tetrapod. His publications can be downloaded at darrennaish.wordpress.com. He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at tetzoo.com!

More by Darren Naish