PALEVO 1631-0683 Elsevier S1631-0683(03)00094-0 10.1016/j.crpv.2003.07.002 Research article Evolution (General Phylogenetics and systematic Theory) Molecular evidence for species level divergence in African Nile Crocodiles Crocodylus niloticus (Laurenti, 1786) Schmitz Andreas a.schmitz.zfmk@t-online.de a Mansfeld Patrick a Hekkala Evon b d Shine Tara c Nickel Hemmo a Amato George d Böhme Wolfgang a Zoologisches Forschungsinstitut und Museum Alexander Koenig, Adenauerallee 160, 53113 Bonn, Germany American Museum of Natural History, Division of Invertebrate Zoology, 79th street Central park West, New York, NY 10024, USA 42 Meadow Way, Kilkenny, Ireland Wildlife Conservation Society, 2300 Southern Boulevard, Bronx, NY 10460, USA 2 8 S1631-0683(00)X0015-2 703 712 © 2003 Académie des sciences. Published by Elsevier B.V. All rights reserved. 2003 Académie des sciences Full (PDF)

Relationships of the newly discovered dwarf crocodiles from Mauritania were inferred from mitochondrial 12S sequences. Specimens from 13 different Crocodylus niloticus populations (from East Africa, West Africa and Madagascar) were compared. Additional representatives of the genus Crocodylus (one from Africa and one from Australia), the African genus Osteolaemus and the South American alligatorid Paleosuchus palpebrosus (as outgroup) were included in the analysis. Maximum-likelihood and Bayesian analyses yielded relationships that were strikingly different from currently prevailing phylogenetic hypotheses. Both analyses consistently revealed two groups, one consisting of the monophyletic West- and Central African populations and the other of a paraphyletic group containing the East African and Madagascan populations. High genetic divergence between those groups indicates separation on the species level. Furthermore ‘C’ cataphractus is clearly shown not to be a member of the genus Crocodylus. The resulting nomenclatural changes are discussed.

Preuve moléculaire de divergence au niveau spécifique chez le crocodile du Nil : Crocodylus niloticus (Laurenti 1786). Les relations de parenté de deux crocodiles nains de Mauritanie ont été déduites de l’analyse de séquences mitochondriales 12S. Des spécimens de 13 populations différentes de Crocodylus niloticus (d’Afrique de l’Est, de l’Ouest et de Madagascar) ont été comparés. Des spécimens supplémentaires du genre Crocodylus (un d’Afrique et un d’Australie), du genre africain Osteolamus et de l’alligatoridé Paleosuchus palpebrosus (comme extra-groupe) ont été inclus dans l’analyse. La probabilité maximale et les analyses bayésiennes ont livré des relations de parenté qui sont remarquablement différentes des hypothèses phylogénétiques classiques. Les deux analyses ont révélé logiquement deux groupes : l’un comprenant les populations monophylétiques d’Afrique de l’Ouest et centrale et l’autre un groupe paraphylétique comprenant des populations d’Afrique de l’Est et de Madagascar. Une forte divergence génétique entre ces groupes indique une séparation au niveau spécifique. En outre, il est clair que « C ». cataphractus n’appartient pas au genre Crocodylus. Les changements de nomenclature qui en résultent sont discutés.

Crocodylus, C. niloticus, C. suchus, Mecistops, 12S mtDNA, Maximum-likelihood, Bayesian analysis, Taxonomy Crocodylus, C. niloticus, C. suchus, Mecistops, 12S mtDNA, Probabilité maximale, Analyse bayésienne, Taxonomie presented Presented by Philippe Taquet Introduction

There are three extant species of crocodilian currently known to exist in Africa. These are comprised of the endemic and monotypic dwarf crocodile (Osteolaemus tetraspis) narrowly confined to forests of West and Central Africa, the slender-snouted crocodile (Crocodylus cataphractus) of West and Central Africa and the Nile Crocodile (C. niloticus) which has an extensive distribution from Senegambia in the west to Egypt in the east, and southwards to South Africa and Madagascar [35] and [36]. Historically, several authors have proposed subspecies based on geographically correlated morphological differences. Many of these characters have been used in the reptile skin trade [4]. Nominal subspecies previously recognised [5], [34] and [36] are: C. n. niloticus Laurenti, 1768 (restricted type locality: Egypt); C. n. africanus Laurenti, 1768 (restricted type locality: Tanzania); C. n. chamses Bory, 1824 (restricted type locality: Congo River); C. n. cowiei Smith and Hewitt, 1937 (type locality: South Africa); C. n. madagascariensis Grandidier, 1872 (type locality: Madagascar); C. n. pauciscutatus Deraniyagala, 1948 (type locality: Lake Rudolph); and C. n. suchus Geoffroy, 1807 (type locality: Niger River).

Nile crocodiles have disappeared from many parts of their historic range, particularly in northern Africa and in the Sahara desert [2] and [14]. However, in one Central Saharan site, viz. the Ennedi Mts. in Chad, and in some places in southern Mauritania, these desert crocodiles have survived up to the present [30] and [33]. Individuals from these desert populations have been described as significantly smaller than those found in other populations (not exceeding a total length of 2.3 m) raising the possibility of long-term isolates with distinct evolutionary history [14] and [30].

Presented here are the first mtDNA sequence data comparing the relict populations to other extant Nile crocodile populations. Initial results suggest significant genetic differentiation between relict and Eastern Nile crocodile populations. However, data from additional West African populations suggest an even more complex evolutionary history for C. niloticus, as very marked genetic differences could be found between samples from West- and East Africa. We therefore examined additional specimens from several allopatric populations throughout the complete distribution area of C. niloticus, and, as a result, propose some taxonomic changes.

 Material and methods

To examine the genetic variation within Crocodylus niloticus, tissue samples of three Mauritanian, three more West African, and eight East African and one Malagasy niloticus population(s) were analyzed (Table 1; Fig. 2 ). To assess genetic differentiation between different species of the genus Crocodylus, we included C. cataphractus and C. johnsoni in the analysis. We further included the second African crocodile genus Osteolaemus. For outgroup comparison we further included the South American alligatorid, Paleosuchus palpebrosus. The voucher specimens with their collection numbers, localities and Genbank accession numbers are given in Table 1.

 Genetic analysis

DNA was extracted from liver or muscle tissue (either fresh, preserved in 98% ethanol, or dried for museum specimens) using QuiAmp tissue extraction kits (Qiagen). We used the primers 12SA-L (light chain; 5′ - AAA CTG GGA TTA GAT ACC CCA CTA T - 3′) and 12SB-H (heavy chain; 5′ - GAG GGT GAC GGG CGG TGT GT - 3′) [15] and [25] to amplify a section of the mitochondrial 12S ribosomal RNA gene. Cycling procedure was as follows: 35 cycles: denaturation 45 s at 94 °C, primer annealing for 60 s at 50 °C, extension for 120 s at 74 °C.

PCR products were purified using Qiaquick purification kits (Qiagen). Sequences were obtained using an automatic sequencer (ABI 377) and manually corrected using the computer program Sequence Navigator (Applied Biosystems).

Sequences were aligned using the computer program ClustalX [32] (default parameters). Alignment was subsequently adjusted manually using the computer program BioEdit [8]. We explored the quality of our alignment by varying alignment gap opening cost (6, 9, 12) and comparing alignments.

Prior to phylogenetic reconstructions, we tested for homogeneity of base frequencies among taxa using the χ2 test as implemented in PAUP*4.0b10 (which ignores correlation due to phylogenetic structure): (1) over all sites; (2) over parsimony-informative sites only; (3) without constant sites (parsimony-uninformative and constant sites will mislead the χ2) test [23]. We performed maximum likelihood (ML) and Bayesian reconstructions. All maximum likelihood analyses [3] were performed with PAUP*4.0b10 [31]. In order to compare the results obtained via maximum likelihood analysis and Bayesian inference, the hierarchical likelihood-ratio test was carried out using MRMODELTEST 1.1b [24], a simplified version of MODELTEST [26] and [27], selecting the best-fit model of nucleotide substitution for our data set. Parameters of the model (substitution parameters, shape of gamma distribution, proportion of invariable sites) were estimated from the data set. The ML tree was calculated with the parameter estimates obtained under the best-fit model. A heuristic search was made with 10 replicates of random stepwise addition and tree bisection-reconnection (TBR) branch-swapping. Because of the extensive computation time used in ML bootstrap calculations, the relative branch support in phylogenetic analysis was evaluated with 100 bootstrap pseudoreplicates (heuristic search, random addition of taxa with 10 replicates, TBR branch-swapping). We considered bootstrap values of 80 as giving strong support to the respective node, since this is a more conservative approach than the value suggested in [10].

Confidence in the phylogenetic signal for this molecular data set was assessed by calculating the skewness, or g1 statistic (implemented in PAUP*), which provides a measure of phylogenetic information content [11]. We produced 1000 randomly generated trees for ML (with outgroup excluded in both approaches; settings for ML are identical to the ones described above).

A matrix of pairwise sequence differences for the 12S rRNA genes was calculated using the p-distance (Table 3).

All Bayesian [13], [16], [17], [21] and [29] analyses were performed with MRBAYES, version 3.0b3 [12], which approximates the posterior probabilities (PP) of trees. The program uses a variant of Markov chain Monte Carlo (MCMC), Metropolis-coupled MCMC [6], which is less prone to entrapment in local optima than is normal MCMC [7], [9] and [22].

To select the best-fit model of nucleotide substitution for our data set, the hierarchical likelihood-ratio test was carried out using MRMODELTEST 1.1b [24]. Consequently, the settings of MRBAYES were specified according to the results of MRMODELTEST. Besides the specific parameters calculated by MRMODELTEST, the default settings of MRBAYES were used. We ran two MCMC analyses for 106 generations each. Each chain consisted of one cold and three heated chains and the Markov chains were started from a random tree. The Markov chains were sampled every 100th generation, resulting in 10 000 sampled trees from each chain. The initial 1000 (10%) trees were disregarded as ‘burn-in’ (the portion of the chain that was sampled before stationarity was reached). Inferences, then, were based on the 9000 trees samples from each chain. The topologies were used to generate a strict-consensus tree, with the percentage of samples recovering any particular clade representing that clade’s posterior probability [12]. Unlike the non-parametric bootstrap values of the ML analysis, these are the true probabilities of the clades under the assumed model [29]. Consequently, we consider probabilities of 95% or greater to be significantly supported.

 Results

The obtained 12S sequences (lengths referring to the aligned sequences including gaps) comprised 433 bp, with the exception of the tissue samples taken from populations from Senegal and Chad, which had strongly degenerated, making it impossible to get complete sequences from these samples. Nonetheless, we managed to get shorter fragments of the populations sequenced (195 bp for the Chad sample and 78 bp for the Senegal sample) (Table 1). As the latter two sequences are much shorter than the other obtained sequences they were excluded from the calculations. Nonetheless, as these two sequences are part of a hypervariable region of the 12S gene, they could be used to classify the respective specimens (see below). In our 12S data set no ambiguous sites could be detected. The complete alignment is available from the authors on request. The matrix for the uncorrected p-distances for all nucleotide sites is presented in Table 3.

In the data set a phylogenetic signal is clearly present (g1 = -1.8185, p = 0.01). When all characters were included, we found no significant deviation from the homogeneity of base frequencies among taxa (χ2 = 11.5660, p = 1.0000, df = 51). The same was true for the parsimony-informative sites only (χ2 = 46.8861, p = 0.6377, df = 51) and without constant sites (χ2 = 32.2954, p = 0.9739, df = 51).

Likelihood scores for models examined under MRMODELTEST showed that the GTR + G model [37] was determined to be the appropriate model for our data set. This model incorporates unequal base frequencies [π(A) = 0.3138, π(T) = 0.2117, π(C) = 0.2721, π(G) = 0.2023], and a gamma distribution shape parameter (α = 0.2931).

Both the ML and the Bayesian approaches produced identical topologies. Fig. 1 shows the ML tree (with lnL = -1292.25), with the posterior probabilities (if not identical, for the first and the second run) above the nodes and the ML bootstrap values below the nodes. Table 2 provides the estimates of the substitution parameters calculated by MRBAYES. The two independent MCMC runs converged on similar log-likelihood scores and reached stationarity no later than 100 000 generations. The posterior probability (PP) values supporting congruent nodes between the two runs were highly correlated (Fig. 1), further indicating that the analyses converged.

Three major clades are evident. The first clade comprises the two included Osteolaemus tetraspis and Crocodylus cataphractus. This clade, placing tetraspis and cataphractus as sister species, is only very weakly supported by a ML bootstrap value of 51 and posterior probability values of 0.55 and 0.57 for the first and for the second run, respectively. Nonetheless, that both species are very clearly placed outside the actual Crocodylus clade receives very high support in both the ML (87) and the Bayesian trees (0.98 for both runs). The second clade comprises all other Crocodylus species, with all East African Crocodylus specimens forming a large polytomy. This second clade is strongly supported by both the ML and the Bayesian analysis (ML: 87 / PP: 0.98). Within this monophyletic Crocodylus group our analyses revealed a group consisting of Crocodylus johnsoni and a maximally supported monophyletic group (ML: 100 / PP: 1.00), comprising all included West African Crocodylus niloticus specimens. However, the sister relationship of C. johnsoni towards the monophyletic West African Crocodylus group does not receive any bootstrap support.

On the genus level we find genetic differentiations (Table 3) between Crocodylus and Osteolaemus of at least 7.8% between C. cataphractus and Osteolaemus, between Crocodylus and Paleosuchus of at least 18.4%, between C. cataphractus and Paleosuchus, and between Osteolaemus and Paleosuchus of at least 20.0%. Genetic variation within Crocodylus ranged from 0.0% to 9.7%. The West African Crocodylus specimens show a genetic variation of 0.0%. The East African Crocodylus populations show a genetic variation of 0.0–0.7%. The West African Crocodylus specimens differ from C. johnsoni in 7.1–7.3%, and from their East African congeners in at least 4.3%. Crocodylus johnsoni shows a genetic differentiation of at least 4.5% from the East African Crocodylus specimens. C. cataphractus differs from the East African C. niloticus specimens in at least 7.4%. It differs from the West African specimens in 9.4–9.5%, and differs from Osteolaemus in at least 7.8%.

 Discussion

Our results demonstrate that the African crocodiles are genetically much more diverse than previously thought. Both the ML and Bayesian analyses revealed the existence of two independent clades of C. niloticus: One monophyletic lineage comprising all West African samples studied, the genetic variation within this clade being 0.0% (Table 3); and a second independent clade containing all East African samples studied including the Madagascan one, but showing, nonetheless, only a very slight genetic variation throughout the entire range. This result does not rule out further genetic subdivision within these lineages, as the 12S gene is rather conservative and may not resolve smaller differences to be found between conspecific populations. Therefore we will not try to address the question of intraspecific evolution in regard to the nominal subspecies described (see Introduction) without additional data. Such a study, employing broader regional sampling and additional mitochondrial and nuclear gene regions is underway (Hekkala et al., unpublished data).

The inclusion of C. johnsoni within the general Crocodylus clade shows this Australian species to be congeneric with the African Crocodylus species. However, the sister relationship of C. johnsoni towards the monophyletic West African Crocodylus group does not receive any bootstrap support, and together with the high genetic differences of johnsoni towards either niloticus clade (4.5–7.3) clearly shows that C. johnsoni can be regarded as a full independent species within the genus Crocodylus.

The very low internal genetic variation of 0.0–0.7% found within the eastern niloticus clade contrasts strongly with the high genetic divergence towards the western niloticus clade (at least 4.5%). As described in the results section above, the two short sequence fragments obtained from the Chad and the Senegal specimens are part of a hypervariable region within the 12S gene. Therefore, even these short fragments have accumulated a rather large number of substitutions and can consequently be safely assigned to belong to either of the described clades within Crocodylus. It becomes clear that the described western niloticus clade also includes the Central African sample (Chad) (only one substitution against the other members of the western clade, but 12 against the east African populations), and that the same is true for the sample from Senegal. Here is clear evidence that the West African Crocodylus clade reaches at least to Central Africa, even though due to the short sequence lengths the Chad and Senegal samples were excluded from the general phylogenetic analyses.

The level of distinctness indicated by these results suggests that taxonomic revision of C. niloticus is warranted. This includes resurrection of Crocodylus suchus Geoffroy, 1807, as the type locality of Crocodylus niloticus Laurenti, 1768 has been restricted to Egypt [see 35, 36].

The geographic split between the lineages runs roughly through an area that was strongly influenced by the repeated extensions and regressions of the Sahara desert, particularly during the Pliocene and Pleistocene [1], [18], [19] and [20]. Geographic barriers formed by rain forests may have repeatedly been effective enough to further an allopatric speciation process. The eastern species (C. niloticus s. str.) had always a hydrographic connection along the Nile from Egypt down to Ethiopia, Kenya and southwards, and the so-called ‘arid corridor’ plays an important biogeographic role as a link for many savannah-adapted plant and animal species [e.g. 28]. The Saharan relict crocodile populations, however, seem to be remnants of just the last humid phase of the desert, thus being isolated from the range of their big-growing southern conspecifics only for a few thousand years at best, by far not sufficiently long for establishing taxonomically relevant genetic differences. Clearly more work is needed to properly reconstruct the evolutionary history of Africa’s largest predators. What was once thought of as the common Nile crocodile, is perhaps not so common after all.

A final aspect important to note is that both ML and Bayesian inference consistently placed cataphractus outside the monophyletic Crocodylus clade. Confidence that cataphractus may not be congeneric with niloticus s.l. and johnsoni is indicated by the strong support of the corresponding node (ML: 87; PP: 0.98), which places cataphractus well outside the combined Crocodylus clade.

“C.” cataphractus clustered with Osteolaemus, but with only low support (ML: 51; PP: 0.55, 0.57), showing that it is also not congeneric with the O. tetraspis. High genetic differentiation to Osteolaemus (> 7.8%) shows that “C.” cataphractus is not a member of the genus Osteolaemus but probably deserves generic rank. Mecistops Gray 1844 would be available. Still, since we only included a single specimen and only one mitochondrial gene, it is clear that additional specimens and further genetic data (e.g. from nuclear genes) are required to clarify the taxonomic situation of cataphractus. This will be done in a subsequent study (Hekkala, unpublished data).

 Acknowledgements

We thank Nicolà Lutzmann (ZFMK), Hans-Werner Hermann (Zoological Garden Cologne), Thomas Wilms (SMN, Karlsruhe), Rudolf Wicker (Zoological Garden Frankfurt), Frank Glaw (ZSM, Munich), Miguel Vences (University of Amsterdam), the “Ferme aux crocodiles” (Pierrelatte), Michael Klemens (Wildlife Conservation Society), The American Museum of Natural History (AMNH), and The Field Museum of Natural History (FMNH), who provided samples for this study. The second author is indebted to John P. Huelsenbeck, Fredrik Ronquist, Jonathan P. Bollback and Johan Nylander for their introduction to the MrBayes program.

 Böhme W. Die wechselvolle Geschichte der Sahara: Untersuchung von Reptilienzönosen entlang eines westsaharischen Transektes, mit einem überaschenden Fund im Südosten Mauretaniens Tier u. Mus. 7 1 2000 11–21 de Smet K. Status of the Nile crocodile in the Sahara desert Hydrobiologia 391 1999 81–86 Felsenstein J. Evolutionary trees from DNA sequences: a maximum likelihood approach J. Mol. Evol. 17 1981 368–376 Fuchs K. Die Krokodilhaut 1977 E. Roether Darmstadt 183 p Fuchs K. Mertens R. Wermuth H. Die Unterarten des Nilkrokodils, Crocodylus niloticus Salamandra 10 3/4 1974 107–114 Geyer C.J. Markov chain Monte Carlo maximum likelihood Keramidas E.M. Computing Science and Statistics: Proc. 23rd Symposium on the Interface 1991 Interface Foundation Fairfax Station 156–163 P.J. Green, Reversible jump Markov chain Monte Carlo computation and Bayesian model determination, Biometrika 82 (1995) 711–732 Hall T.A. BIOEDIT: A user-friendly biological sequence alignment editor and analysis program for WindowS95/98/NT Nucl. Acids. Symp. Ser. 41 1999 95–98 Hastings W.K. Monte Carlo sampling methods using Markov chains and their applications Biometrika 57 1970 97–109 Hillis D.M. Bull J.J. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis Syst. Biol. 42 1993 182–192 Hillis D.M. Huelsenbeck J.P. Signal, noise, and reliability in molecular phylogenetic analyses J. Hered. 83 1992 189–195 Huelsenbeck J.P. Ronquist F. MRBAYES: Bayesian inference of phylogeny Bioinformatics 17 2001 754–755 Huelsenbeck J.P. Ronquist F. Nielsen R. Bollback J.P. Bayesian inference of phylogeny and its impact on evolutionary biology Science 294 2001 2310–2314 Joleaud L. Études de Géographie zoologiques sur la Berbérie Bull. Soc. Zool. France 58 1933 397–404 Kocher T.D. Thomas W.K. Meyer A. Edwards S.V. Pääbo S. Villablanca F.X. Wilson A.C. Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers Proc. Natl Acad. Sci. USA 86 1989 6196–6200 Larget B. Simon D. Markov chain Monte Carlo for the Bayesian analysis of phylogenetic trees Mol. Biol. Evol. 16 1999 750–759 Li S. Pearl D.K. Doss H. Phylogenetic tree reconstruction using Markov chain Monte Carlo J. Am. Stat. Assoc. 95 2000 493–508 Littmann K. War die Sahara in der Vorzeit grün? Klimaschwankungen und ökologische Veränderungen im größten Trockengebiet der Erde – ein Überblick Natur u. Mus. 117 1987 386–394 Maley J. The African rain forest – main characteristics of changes in vegetation and climate from the Upper Cretaceous to the Quaternary Proc. R. Soc. Edinb. 104B 1996 31–73 Martin C. Die Regenwälder Westafrikas. Ökologie, Bedrohung, Schutz 1989, 235 p Birkhäuser Basel Boston Berlin Mau B. Newton M. Larget B. Bayesian phylogenetic inference via Markov chain Monte Carlo methods Biometrics 55 1999 1–12 Metropolis N. Rosenbluth A.W. Rosenbluth M.N. Teller A.H. Teller E. Equations of state calculations by fast computing machines J. Chem. Phys. 21 1953 1087–1091 Misof B. Rickert A.M. Buckley T.R. Fleck G. Sauer K.P. Phylogenetic signal and its decay in mitochondrial SSU and LSU rRNA gene fragments of Anisoptera Mol. Biol. Evol. 18 2001 27–37 Nylander J.A.A. MrModeltest 1.1b 2002 Uppsala Sweden S.R. Palumbi, A. Martin, S. Romano, W.O. McMillan, L. Stice, G. Grabowski, The Simple Fool's Guide to PCR, Version 2.0, Department of Zoology, University of Hawaii, Honolulu, HI, 96822, 1991, Privately published document compiled by S. Palumbi 47 p Posada D. Crandall K.A. MODELTEST: testing the model of DNA substitution Bioinformatics 14 1998 817–818 Posada D. Crandall K.A. Selecting the best-fit model of nucleotide substitution Syst. Biol. 50 2001 580–601 Poynton J.C. The ‘arid corridor’ distribution in Africa: a search for instances among amphibians Madoqua 19 1 1995 45–48 Rannala B. Yang Z. Probability distribution of evolutionary trees: a new method of phylogenetic inference J. Mol. Evol. 43 1996 304–311 Shine T. Böhme W. Nickel H. Thies D.F.G. Wilms T. Rediscovery of relict populations of the Nile crocodile Crocodylus niloticus in south-east Mauritania, with observations on their natural history Oryx 35 3 2001 260–262 Swofford D.L. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Ver. 4 2002 Sinauer Associates Sunderland, Massachusetts Thompson J.D. Gibson T.J. Plewniak F. Jeanmougin F. Higgins D.G. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools Nucleic Acids Res. 24 1997 4876–4882 Tubiana J. Quelques aberrations sahariennes: les crocodiles d'Archaï Courr. Nat. 153 1995 27–29 Wermuth H. Fuchs K. Bestimmen von Krokodilen und ihrer Häute 1978 G. Fischer Stuttgart New York 100 pp Wermuth H. Mertens R. Schildkröten, Krokodile, Brückenechsen 1961 G. Fischer (reprinted 1996) Jena 506 pp Wermuth H. Mertens R. Liste der rezenten Amphibien und Reptilien Testudines, Crocodylia, Rhynchocephalia, Das Tierreich 100 1977 1–174 Yang Z. Maximum likelihood phylogenetic estimation from DNA sequences with variable rates over sites: approximate methods J. Mol. Evol. 39 1994 306–314

Map of sample localities for specimens used in this study. Locality names are as follows: (1) Senegal, (2) Gambia, (3) Aioun el-Atrouss, Mauritania, (4) Liberia, (5) Edéa, Cameroon, (6) Lambaréné, Gabon, (7) Ennedi Mts., Chad, (8) Lake Nasser, Egypt, (9) Chor Melk en-Nasir, Sudan, (10) Kenya, (11) Kariba Dam, Zimbabwe, (12) Natal, South Africa, (13) Madagascar. Where the exact locality of the voucher is not known, we placed the corresponding number at a place where crocodiles are known to occur in the given country. The three desert localities are indicated by black numbers.

Fig. 2. Carte des localités ayant fourni des spécimens utilisés dans cette étude. Les noms des localités sont les suivantes : (1) Sénégal, (2) Gambie, (3) Aïoun el-Atrouss, Mauritanie, (4) Libéria, (5) Edéa, Cameroun, (6) Lambaréné, Gabon, (7) massif de l'Ennedi, Tchad, (8) Lac Nasser, (9) Chor Melk en-Nasir, Soudan, (10) Kenya, (11) Barrage de Kariba, Zimbabwe, (12) Natal, Afrique du Sud, (13) Madagascar. Lorsque la localité exacte de la pièce n’est pas connue, nous avons placé le numéro correspondant à l’endroit où les crocodiles sont connus dans la région. Les trois localités du désert sont indiquées par des numéros en noir.

Phylogram of the maximum likelihood tree (ML) and Bayesian analysis tree obtained from PAUP* and MrBayes searches using Paleosuchus palpebrosus as outgroup. Numbers above nodes represent posterior probabilities (PP), numbers below nodes represent bootstrap proportions for 100 pseudoreplicates for the likelihood analysis. Bootstrap proportions less than 50% are not shown.

Fig. 1. Phylogramme de l’arbre de probabilité maximale (ML) et arbre de l’analyse bayésienne obtenus avec PAUP* et avec les recherches de MrBayes en utilisant Paleosuchus palpebrosus comme extra-groupe. Les numéros au-dessus des nœuds représentent les probabilités postérieures (PP), les nombres au-dessous des nœuds représentent les proportions des bootstraps pour 100 pseudo-répliques de l’analyse de probabilité. Les proportions des bootstraps inférieures à 50% ne sont pas indiquées.

List of voucher specimens for each species included in the present study, with their respective localities, collection numbers and accession numbers (12S)

Liste des spécimens de chaque espèce inclus dans la présente étude, avec leurs localités respectives, leurs numéros de collection et leurs numéros d’accession (12S)

Species Locality Collection number Accession number Paleosuchus palpebrosus South America ZFMK 73079 AY195960 Osteolaemus tetraspis 1 Edéa, Cameroon ZFMK 74854 AY195958 Osteolaemus tetraspis 2 Liberia ZFMK 50692 AY195959 Crocodylus cataphractus Lambaréné, Gabon ZFMK 73109 AY195941 Crocodylus johnsoni Australia ZFMK 73662 AY195942 Crocodylus niloticus Lake Nasser, Egypt ZFMK, uncatalogued AY195943 Crocodylus niloticus Gambia voucher not collected AY195944 Crocodylus niloticus Kenya voucher not collected AY195945 Crocodylus niloticus Madagascar voucher not collected AY195946 Crocodylus niloticus 1 Aioun el-Atrouss, Mauritania ZFMK, uncatalogued AY195947 Crocodylus niloticus 2 Aioun el-Atrouss, Mauritania ZFMK, uncatalogued AY195948 Crocodylus niloticus 3 Aioun el-Atrouss, Mauritania ZFMK, uncatalogued AY195949 Crocodylus niloticus 1 Natal, South Africa voucher not collected AY195950 Crocodylus niloticus 2 Natal, South Africa voucher not collected AY195951 Crocodylus niloticus 3 Natal, South Africa voucher not collected AY195952 Crocodylus niloticus Chor Melk en-Nasir, Sudan ZFMK 50489 AY195953 Crocodylus niloticus 1 Kariba Dam, Zimbabwe voucher not collected AY195954 Crocodylus niloticus 2 Kariba Dam, Zimbabwe voucher not collected AY195955 Crocodylus niloticus Senegal voucher not collected AY195957 Crocodylus niloticus Ennedi Mts., Chad voucher not collected AY195956

Summary of the uncorrected p-distances (sequences excluded from the tree are marked)

Sommaire des distances p, avant correction (des séquences exclues de l'arbre sont indiquées)

Species 1 2 3 4 5 6 7 8 9 10 11 12 1 Crocodylus cataphractus – 2 Crocodylus johnsoni 0.097 – 3 C. niloticus Egypt 0.074 0.045 – 4 C. niloticus Gambia 0.094 0.073 0.046 – 5 C. niloticus Kenya 0.077 0.048 0.002 0.044 – 6 C. niloticus Madagascar 0.076 0.045 0.000 0.046 0.002 – 7 C. niloticus 1 Mauritania 0.095 0.071 0.045 0.000 0.043 0.045 – 8 C. niloticus 2 Mauritania 0.095 0.071 0.045 0.000 0.043 0.045 0.000 – 9 C. niloticus 3 Mauritania 0.095 0.071 0.045 0.000 0.043 0.045 0.000 0.000 – 10 C. niloticus 1 Natal, S. Africa 0.076 0.045 0.000 0.046 0.002 0.000 0.045 0.045 0.045 – 11 C. niloticus 2 Natal, S. Africa 0.076 0.047 0.002 0.043 0.000 0.002 0.043 0.043 0.043 0.002 – 12 C. niloticus 3 Natal, S. Africa 0.076 0.047 0.002 0.043 0.000 0.002 0.043 0.043 0.043 0.002 0.000 – 13 C. niloticus Sudan 0.083 0.053 0.007 0.054 0.010 0.007 0.054 0.054 0.054 0.007 0.010 0.010 14 C. niloticus 1 Zimbabwe 0.076 0.045 0.000 0.046 0.002 0.000 0.045 0.045 0.045 0.000 0.002 0.002 15 C. niloticus 2 Zimbabwe 0.076 0.045 0.000 0.046 0.002 0.000 0.045 0.045 0.045 0.000 0.002 0.002 16 Osteolaemus tetraspis Cameroon 0.078 0.140 0.113 0.131 0.110 0.113 0.133 0.133 0.133 0.113 0.110 0.110 17 O. tetraspis Liberia 0.085 0.144 0.117 0.132 0.114 0.117 0.137 0.137 0.137 0.117 0.114 0.114 18 Paleosuchus palpebrosus 0.184 0.200 0.186 0.205 0.190 0.188 0.204 0.204 0.204 0.188 0.188 0.188 19 Crocodylus niloticus Chad 0.137 0.093 0.055 0.005 0.060 0.055 0.005 0.005 0.005 0.055 0.060 0.060 20 C. niloticus Senegal 0.118 0.106 0.044 0.000 0.044 0.044 0.000 0.000 0.000 0.044 0.044 0.044 Species 13 14 15 16 17 18 19 20 13 C. niloticus Sudan – 14 C. niloticus 1 Zimbabwe 0.007 – 15 C. niloticus 2 Zimbabwe 0.007 0.000 – 16 Osteolaemus tetraspis Cameroon 0.120 0.113 0.113 – 17 O. tetraspis Liberia 0.121 0.117 0.117 0.000 – 18 Paleosuchus palpebrosus 0.195 0.188 0.188 0.200 0.209 – 19 Crocodylus niloticus Chad 0.055 0.055 0.055 0.189 0.189 0.263 – 20 C. niloticus Senegal 0.044 0.044 0.044 0.148 0.148 0.211 0.000 –

Parameter estimates of the substitution model (GTR + G), sampled after the burn-in phase of the chain. The columns indicate the parameter, mean and 95% credible interval for the parameter. The parameters are TL, the tree length; rij , rate of substitution between nucleotides i and j measured relative to the rate between G and T (r GT = 1); π i, base frequencies; and α, gamma shape parameter for among-site variation. Upper values in each pair correspond to the run No. 1; lower values correspond to the run No. 2

Paramètre estimé du modèle de substitution (GTR+G), échantillonné après la phase de combustion de la chaîne. Les colonnes indiquent le paramètre, la moyenne et l’intervalle de 95% de crédibilité pour le paramètre. Les paramètres sont TL, la longueur de l’arbre ; rij , rapport de substitution entre les nucléotides i et j mesurés selon le rapport entre G et T (r GT = 1) ; π i, fréquences des bases ; α, forme gamma du paramètre pour la variation de chaque site. Les valeurs supérieures de chaque paire correspondent à l’essai 1 ; les valeurs inférieures correspondent à l’essai 2

Parameter Mean 95% Credity Interval TL 0.94 (0.67, 1.37) 0.94 (0.65, 1.39) r GT 1.00 1.00 r CT 36.43 (15.09, 49.64) 30.66 (13.59, 49.12) r CG 1.05 (0.04, 3.26) 1.03 (0.03, 4.22) r AT 6.33 (1.86, 12.82) 4.98 (1.53, 11.17) r AG 19.75 (9.36, 35.49) 19.15 (7.30, 32.72) r AC 8.93 (3.27, 16.88) 7.79 (2.80, 15.84) π A 0.313 (0.274, 0.351) 0.313 (0.275, 0.353) π C 0.274 (0.238, 0.312) 0.274 (0.239, 0.311) π G 0.196 (0.164, 0.231) 0.194 (0.162, 0.229) π T 0.217 (0.185, 0.252) 0.218 (0.185, 0.254) α 0.296 (0.194, 0.450) 0.295 (0.191, 0.454)