Search

Additional file 1: Figure S1. The results of the Bayesian clustering analysis with STRUCTURE. The plot shows the values of ΔK calculated according to Evanno et al. [47]. Figure S2. (a) Results of BIC value versus the number of clusters to determine the optimal clustering scheme in the DAPC analyses. (b) The Cross-Validation test results determining the optimal number of PCs retained. Figure S3. The species tree chronograms generated from the BEAST calibration dating analyses, A: mtDNA chronogram, B: mtDNA+nDNA chronogram. The red dots were the five constrained calibration points for calibration dating analyses. Figure S4. The macroevolution cohort matrix for the seven clades of the P. tentorius species complex. BAMM Bayesian diversification rate analysis based on the mean phylorate plot trees, are shown at the top and on the left side of the cohort matrix, for purposes of comparison. The matrix shows pairwise probabilities of two groups sharing the same evolutionary dynamics. The “warm” colours represent high cohort similarities (highest value “1” refers to 100% similarity), whilst, the “cool” colours represent low cohort similarities (lowest value “0” refers to 0% similarity). Figure S5. Visualizing the single-chain MCMC diagnostics for a CoMET analysis with empirically estimated diversification hyperpriors. Blue bars/dots represent passed tests and red bars/dots mean failed tests (failed convergence). Table S3. The node age (Ma) and 95% HPD at each node (see Fig. 5 and Fig. S3) generated from BEAST calibration dating analyses in both gene trees and species trees of the mtDNA and (mtDNA+nDNA) datasets, respectively. Table S4. The independent BioGeoBEARS model test results for the six habitat reconstruction models with consideration of the “founder effect” parameter “J” on geographic regions, biome and topographic barriers datasets. The selected best model of each analysis with its criteria is shown in bold. Table S5. The ANOVA based LRT test results retrieved from the likelihood function models of different scenarios investigated under character dependency analyses BiSSE and MuSSE with different models of regions, biomes and the two sides of the Orange River. “Minimal”: the null model assumed all parameters are equal between different characters states; “all different”: the full parameter model considered lambda, mu and q as different between character states; “free.lambda”: the model considered lambda as different only between different character states, “free.mu”: the model considered only mu as different between different character states, “free.lambda.mu”: the model considered only lambda and mu as differ between different character states, “free.q”: the model considered only q as different between different character states. The best model with its criteria from the LRT test in each analysis is shown in bold. Table S6. The average test AUC criterion under the ROC curve and its standard deviation (SD) generated by ENM analyses for each group [P. t. tentorius (Ptt-A – Ptt-D), P. t. trimeni (Ptr), P. t. verroxii (Pv-A and Pv-B), and the entire P. tentorius species complex] through the given timeline. LIG: Last Interglacial, LGM: Last Glacial Maximum, and MIDH: Middle Holocene. Table S7. The results of the optimization of parameters settings for all ENM analyses across groups (group abbreviations as in Table S6, with All denoting the entire P. tentorius species complex) and timelines. Features: Linear features (L), Quadratic features (Q), Product features (P) and Hinge features (H); RM = Regularization multiplier. LIG: Last Interglacial, LGM: Last Glacial Maximum, and MIDH: Middle Holocene. Table S8. The pairwise results of niche difference analysis between clades (taxon abbreviations the same as group abbreviations in Table S6). The pairwise comparisons showing significant niche differences are in bold, whilst, comparisons without significant niche differences are underlined. Note: If the point estimated value (the matrix at the bottom-left) was lower than the 5% threshold value (the matrix at the top-right), then the niches were significantly different. Table S9. The ENM analyses estimating the area of suitable habitats of the seven clades (group abbreviations as in Table S6), as well as changes in the area of suitable habitats against the timeline (for details about the timeline see the Materials and Methods). The arrow " ↑ " represents increasing area, whilst, " ↓ " indicates decreasing area. Groups currently showing a decrease in the area of suitable habitats or future habitats are given in bold. Table S13. Optimal partition scheme, substitution model, likelihood score (-InL), Gamma shape, proportion of estimated invariable sites for the mtDNA+nDNA based BEAST phylogenetic inference. Table S16. Optimal partition scheme, substitution model, likelihood score (-InL), Gamma shap, and proportion of estimated invariant sites for the BEAST calibration dating analyses. Table S17. The constraint of the five most recent common ancestors used in the BEAST calibration dating analyses for the mtDNA and mtDNA+nDNA datasets.

We investigated genetic variation within and among three populations of Chersina angulata; two in the southwestern Cape (West Coast National Park and Dassen Island) of South Africa and Kleinzee in the northwestern parts of the species’ range. Mitochondrial DNA sequences (274 base pairs) of the cytochrome b gene of 25 C. angulata identified 10 haplotypes in three lineages. Estimates of coalescence time for the three lineages ranged from approximately 1.3 to 3.8 million years, implying that the lineages’ present distribution may reflect radiation during the late Pliocene to early Pleistocene Epochs. The mtDNA sequences of Dassen Island and West Coast National Park populations did not differ significantly. The geographic heteroge neity in haplotype frequency distribution suggests that the Kleinzee population represents a distinct lineage from the two southwestern populations. The observed genetic structure of C. angulata has management implications for the conservation of this species. Until more information is available, we recommend that C. angulata should not be translocated between the northwestern and southwestern regions, to prevent disruption of local gene pools.Key words: tortoise, cytochrome b, mtDNA, population genetics.

Body condition and haematological indices provide powerful information when assessing wildlife health. Reference intervals for these indices can facilitate wildlife management, and would benefit initiatives to save the Critically Endangered geometric tortoise (Psammobates geometricus). We collected data from 126 geometric tortoises to establish baseline values reflecting variation over four seasons (spring 2000 to winter 2001) and among three groups (female, male and juvenile). We measured body condition index (BCI; mass to shell volume), blood urea nitrogen (BUN), plasma chloride, packed cell volume (PCV), haemoglobin concentration (Hb) and red blood cell count (RBC), and used PCV, Hb and RBC to calculate erythrocyte indices. BCI correlated poorly with haematological measures but had a strong inverse relationship with BUN. BCI did not vary among groups, but all groups had low condition indices in autumn, the end of the dry season. High BUN and chloride values in autumn indicate dehydration, particularly in males. Males had the highest PCV, Hb and RBC values, especially during summer and autumn when they moved long distances, likely pursuing mates. Female and juvenile indices were similar, except that female Hb was higher than juvenile Hb, possibly to meet female energy needs associated with their large size and reproductive metabolism. Low Hb levels in winter coincided with low temperatures and reduced movements. Our results illustrate how intrinsic and extrinsic factors influence the physiology of geometric tortoises and provide reference intervals to monitor their health.Keywords: blood urea nitrogen, erythrocyte indices, haemoglobin, packed cell volume, red blood cell count, reference intervals

Many terrestrial chelonians are parasitized by ticks, but we have a poor understanding of what determines tick infestations on chelonian hosts. We counted ticks on Homopus signatus signatus during each spring in the years 2001–2004, and evaluated tick frequencies in relation to tortoise anatomy, climate and microhabitat. Tortoises hosted Ornithodoros compactus and O. savignyi, and ticks seemed to prefer the hindlimbs, avoid the shell, and use the forelimbs and neck in intermediate frequencies. The number of ticks on the neck and forelimbs did not differ among males, females and juveniles, but for the hindlimbs and all body parts, males and females usually had similar tick numbers and adults often had more ticks than juveniles. The number of ticks on the hindlimbs correlated with the body size of male and female tortoises, whereas the number of ticks on the forelimbs correlated with juvenile body size. Males and females had similar incidences of ticks despite sexual dimorphism in body size (female size > male size), and this may be due to the relatively large shell openings (soft skin exposure) of males compared to females. Body condition and microhabitat had no effect on the number of ticks. It is unclear why tick numbers on H. s. signatus increased in dry years. The adverse effects of drought on mammals may have caused ticks to switch from mammals to tortoises when preferred hosts were no longer available.Keywords: host switching, morphology, Ornithodoros, parasite, sexual dimorphism, tortoise.

Behavioural frequencies and time budgets for male and female Chersina angulata were recorded in spring, September 2004. The daily activity of the population was 10.51 ± 0.42 h (mean ± CI), but individual males and females were in the open for 2.57 ± 1.12 h and 1.58 ± 1.44 h, respectively. Both sexes spent nearly 3.5 h per day basking with 90% of the basking time in the cover of sparse vegetation. Feeding and walking were the most frequent active behaviours. The activity pattern of the sexes differed between 10:00 and 12:00 in the morning when males spent more time out of cover on active behaviours, particularly feeding, than females did. The daily activity pattern of the population peaked between 10:00 and 12:00, reflecting the activity pattern of the more active sex, males. Females were out of cover, active, and feeding at higher temperatures than were males. The high temperatures probably allowed females to complete activities more efficiently, reducing their exposure time and predation risk. Although fighting and mating represented small fractions of male time budgets, the extended physical presence through walking and feeding may help males to establish dominance hierarchies and enhance mating success.Key words: activity pattern, behaviour, temperature, thread-trailing, time budget, tortoise.

Pelomedusa subrufa (Bonnaterre, 1789) sensu stricto 1789 Testudo subrufa Bonnaterre —Restricted type locality (Bour 1982): Taolañaro, Madagascar; holotype: Muséum National d’Histoire naturelle, Paris, MNHN 7970 (Fig. 2 in Fritz et al. 2014) 1798 Testudo badia Donndorff (nomen novum) 1798 Testudo rubicunda Suckow (nomen novum) 1935 Pelomedusa galeata damarensis Hewitt —Type locality: Quickborn near Okahandja, Namibia; lectotype (Fritz et al. 2014): Port Elizabeth Museum, PEM R14953 (Fig. 7 top in Fritz et al. 2014) 1937 Pelomedusa subrufa wettsteini Mertens —Type locality: Majunga (Mahajanga), western Madagascar; holotype: Senckenberg-Museum, Frankfurt am Main, SMF 7958 (Fig. 8 in Fritz et al. 2014) Diagnosis: Small to medium-sized helmeted terrapins with a known maximum straight carapacial length of 19.7 cm. Yet, the shell length of most adults is below 14 cm. Pectoral scutes in most specimens with broad contact at plastral midseam; however, terrapins with narrow contact or, exceptionally, with triangular pectorals without midseam contact occur. One large undivided temporal scale on each side of head. Two small barbels under chin. Carapace uniform light to dark brown. Plastron in larger individuals mostly light coloured, in younger specimens often with dark markings. Pelomedusa subrufa sensu stricto differs from all other Pelomedusa species by the presence of cytosine (C) instead of thymine (T) at position 117, by the presence of thymine (T) instead of adenine (A) at position 169, by the presence of thymine (T) instead of adenine (A) or cytosine (C) at position 191, by the presence of guanine (G) instead of adenine (A) at positions 226 and 296, by the presence of adenine (A) instead of cytosine (C) at position 280, and by the presence of cytosine (C) or thymine (T) instead of adenine (A) or guanine (G) at position 289 of the 360-bp-long reference alignment of the 12S rRNA gene (Supporting Information). In addition, P. subrufa sensu stricto differs from all other Pelomedusa species except P. galeata by the presence of cytosine (C) instead of adenine (A) or guanine (G) at position 148, by the presence of guanine (G) instead of adenine (A) at position 159, by the presence of cytosine (C) instead of thymine (T) at position 167, and by the presence of guanine (G) instead of adenine (A) or thymine (T) at position 343. Moreover, P. subrufa sensu stricto differs from all other Pelomedusa species except P. galeata and possibly P. gehafie, P. kobe and candidate species B (in which the respective character states are unknown) by the presence of thymine (T) instead of cytosine (C) at position 26 and by the presence of cytosine (C) instead of thymine (T) at position 38. Distribution: Southern Angola (this study), Botswana, southeastern Democratic Republic of the Congo, Malawi (Vargas-Ramírez et al. 2010), Namibia (Wong et al. 2010; this study), South Africa (Limpopo; this study), and Kilimanjaro region of Tanzania (this study). Introduced to Madagascar (Vargas-Ramírez et al. 2010; Wong et al. 2010). Remarks: Pelomedusa subrufa sensu stricto corresponds to mtDNA lineage VIII of Vargas-Ramírez et al. (2010). Besides P. galeata, it is one of the two southern species of Pelomedusa. In the course of the present study, we recorded P. subrufa sensu stricto for the first time for South Africa, based on one sample from the western border region of the Kruger Park (Hoedspruit, Limpopo, South Africa). To the west and south, the next genetically verified records are approximately 300 km distant and are P. galeata. It seems possible that the ranges of the two species abut or overlap in northeastern South Africa. In the Zoologische Staatssammlung München there are genetically verified specimens of P. subrufa sensu stricto collected on 3 May 1937 by Wolfgang Uthmöller (25 October 1904 – 25 August 1951) at the northern slope of the Kilimanjaro, Tanzania (ZSM 320/1937:1–4), and additional genetically verified specimens with the date 5 May 1937 originating from the same collector, but bearing only “Tanganyika” as locality data (ZSM 285/ 1937:1–3). The latter specimens represent another species (P. kobe, see above), suggesting that both taxa occur in close geographical proximity or even sympatry in Tanzania. This is also supported by further genetically verified specimens of P. kobe from the Arusha region, Tanzania, from where also the new species P. neumanni is recorded (see above). The identity of mtDNA sequences of P. subrufa sensu stricto from Tanzania and Madagascar points to Tanzania as geographical source region for the introduced Malagasy helmeted terrapins. Among 94 museum specimens of P. subrufa, two terrapins have triangular pectorals without midseam contact, and 19 terrapins have pectoral scutes with narrow plastral midseam contact, whilst the remaining specimens have pectoral scutes with broad contact at the midseam. Pelomedusa subrufa can cope with extremely arid conditions. For Namibia, there are records from the mouths of temporary streams (so-called riviers) in the Namib Desert (Branch 2008; Boycott & Bourquin 2008; vouchers in the Museum für Naturkunde, Berlin). It is well known that Namibian terrapins burrow underground during arid periods, and if there is no rain, they may evidently survive up to six years burrowed (A. Schleicher, pers. observ., Omaheke, Namibia). The monogenean flatworm Polystomoides chabaudi Euzet & Combes, 1965 was described from Pelomedusa subrufa sensu stricto (collected at Betioky, Madagascar). The same polystomatid species was also reported from helmeted terrapins from the Kampala area, Uganda (Tinsley 1973), that is, probably from P. neumanni., Published as part of Petzold, Alice, Vargas-Ramírez, Mario, Kehlmaier, Christian, Vamberger, Melita, Branch, William R., Preez, Louis Du, Hofmeyr, Margaretha D., Meyer, Leon, Schleicher, Alfred, Široký, Pavel & Fritz, Uwe, 2014, A revision of African helmeted terrapins (Testudines: Pelomedusidae: Pelomedusa), with descriptions of six new species, pp. 523-548 in Zootaxa 3795 (5) on pages 542-543, DOI: 10.11646/zootaxa.3795.5.2, http://zenodo.org/record/4915083, {"references":["Bonnaterre, P. J. (1789) Tableau encyclopedique et methodique des trois regnes de la nature. Erpetologie. Panckoucke, Hotel de Thou, Paris, xxviii + 70 + (1) pp., 35 plates. http: // dx. doi. org / 10.5962 / bhl. title. 59326","Fritz, U., Petzold, A., Kehlmaier, C., Kindler, C., Campbell, P., Hofmeyr, M. D. & Branch, W. R. (2014) Disentangling the Pelomedusa complex using type specimens and historical DNA. Zootaxa, 3795 (5), 501 - 522 http: // dx. doi. org / 10.11646 / zootaxa. 3795.5.1","Vargas-Ramirez, M., Vences, M., Branch, W. R., Daniels, S. R., Glaw, F., Hofmeyr, M. D., Kuchling, G., Maran, J., Papenfuss, T. J., Siroky, P., Vieites, D. R. & Fritz, U. (2010) Deep genealogical lineages in the widely distributed African helmeted terrapin: evidence from mitochondrial and nuclear DNA (Testudines: Pelomedusidae: Pelomedusa subrufa). Molecular Phylogenetics and Evolution, 56, 428 - 440. http: // dx. doi. org / 10.1016 / j. ympev. 2010.03.019","Wong, R. A., Fong, J. J. & Papenfuss, T. J. (2010) Phylogeography of the African helmeted terrapin, Pelomedusa subrufa: genetic structure, dispersal, and human introduction. Proceedings of the California Academy of Sciences, 61, 575 - 585.","Branch, B. (2008) Tortoises, Terrapins & Turtles of Africa. Struik Publishers, Cape Town, 128 pp.","Boycott, R. C. & Bourquin, O. (2008) Pelomedusa subrufa (Lacepede 1788), helmeted turtle, helmeted terrapin. In: Rhodin, A. G. J., Pritchard, P. C. H., van Dijk, P. P., Saumure, R. A., Buhlmann, K. A. & Iverson, J. B. (Eds.), Conservation Biology of Freshwater Turtles and Tortoises: A Compilation Project of the IUCN / SSC Tortoise and Freshwater Turtle Specialist Group. Chelonian Research Monographs, No. 5. Chelonian Research Foundation, Lunenburg, MA, pp. 007.1 - 007.6. http: // dx. doi. org / 10.3854 / crm. 5.007. subrufa. v 1.2008","Euzet, L. & Combes, C. (1965) Parasites des cheloniens malgaches. Polystomoides chabaudi n. sp. (Monogenea) chez la tortue d'eau douce Pelomedusa subrufa Lacepede 1788. Annales de Parasitologie Humaine et Comparee, 40, 455 - 450.","Tinsley, R. C. (1973) Observations on Polystomatidae (Monogenoidea) from East Africa with a description of Polystoma makereri n. sp. Zeitschrift fur Parasitenkunde, 42, 251 - 263. http: // dx. doi. org / 10.1007 / bf 00328887"]}

Pentonix [sic] americana Cornalia, 1849 Cornalia (1849: pp. 312–313) described a specimen allegedly from “Novaeboracum” [= Novum Eboracum, New York] under this name. After its description, Pentonix americana has never been regarded as a valid taxon. It is traditionally identified with the helmeted terrapin (e.g., Boulenger 1889: p. 198; Mertens 1937: p. 139; Loveridge 1941: p. 473; Wermuth & Mertens 1961: p. 284, 1977: p. 115; Fritz & Havaš 2007: p. 345; van Dijk et al. 2012: p. 000.295), even though Strauch (1865: p. 111) doubted that it could be reliably assigned to any species. The type specimen was in the collection of the natural history museum of Milan (Museo Civico di Storia Naturale di Milano) and was destroyed, together with most of the herpetological collection, by British air raids in 1943 (S. Scali, pers. comm.). Pentonix americana is best treated as nomen dubium, because there is no further information available. Mertens’ (1937: p. 139) identification of “ Pentonyx americana ” with South African helmeted terrapins (“offenbar ein südafrikanisches Stück”), and thus with the nominotypical subspecies sensu Mertens (1937), is pure speculation. Mertens’ suggestion was later repeated by Loveridge (1941: p. 473 “probably a South African specimen”)., Published as part of Fritz, Uwe, Petzold, Alice, Kehlmaier, Christian, Kindler, Carolin, Campbell, Patrick, Hofmeyr, Margaretha D. & Branch, William R., 2014, Disentangling the Pelomedusa complex using type specimens and historical DNA (Testudines: Pelomedusidae), pp. 501-522 in Zootaxa 3795 (5) on page 513, DOI: 10.11646/zootaxa.3795.5.1, http://zenodo.org/record/4915064, {"references":["Cornalia, E. (1849) Vertebratorum synopsis in museo mediolanense extantium quae per novam orbem cajetanus osculati collegit annis 1846 - 47 - 48. Speciebus novis vel minus cognitis adjectis nec non descriptionibus atque iconibus illustratis. Corbetta, Milan, pp. 301 - 315, 5 plates.","Boulenger, G. A. (1889) Catalogue of the Chelonians, Rhynchocephalians, and Crocodiles in the British Museum (Natural History). New Edition. British Museum (Natural History), London, x + 311 pp., 6 plates. http: // dx. doi. org / 10.5962 / bhl. title. 13541","Mertens, R. (1937) Bemerkungen uber die Rassen von Pelomedusa subrufa (La Cepede). Zoologischer Anzeiger, 117, 139 - 142.","Loveridge, A. (1941) Revision of the African terrapin [sic] of the family Pelomedusidae. Bulletin of the Museum of Comparative Zoology, 88, 467 - 524.","Wermuth, H. & Mertens, R. (1961) Schildkroten, Krokodile, Bruckenechsen. VEB Gustav Fischer, Jena, XXVI + (1) + 422 pp.","Fritz, U. & Havas, P. (2007) Checklist of chelonians of the world. Vertebrate Zoology, 57, 149 - 368.","van Dijk, P. P., Iverson, J. B., Shaffer, H. B., Bour, R. & Rhodin, A. G. J. (2012) Turtles of the world, 2012 update: annotated checklist of taxonomy, synonymy, distribution, and conservation status. In: Rhodin, A. G. J., Pritchard, P. C. H., van Dijk, P. P., Saumure, R. A., Buhlmann, K. A., Iverson, J. B. & Mittermeier, R. A. (Eds.), Conservation Biology of Freshwater Turtles and Tortoises: A Compilation Project of the IUCN / SSC Tortoise and Freshwater Turtle Specialist Group. Chelonian Research Monographs, No. 5. Chelonian Research Foundation, Lunenburg, MA, pp. 000.243 - 000.328. http: // dx. doi. org / 10.3854 / crm. 5.007. checklist. v 5.2012","Strauch, A. (1865) Die Vertheilung der Schildkroten uber den Erdball. Ein zoogeographischer Versuch. Memoires de l'Academie imperiale des Sciences de St. - Petersbourg, VII e Serie, VIII (13), 1 - 207."]}