We investigated the mitochondrial phylogeography of spur-thighed tortoises (Testudo graeca) in the Western Mediterranean. In North Africa, four major lineages (A-D) occur that together constitute a well-supported clade corresponding to one of the six major clades within T. graeca; the North African clade is sister to a Caucasian clade representing the subspecies T. g. armeniaca. Phylogenetic relationships between the North African lineages are badly resolved. Lineage A is distributed in Tunisia and adjacent Algeria, lineage B in Algeria and northern Morocco, lineage C in the Libyan Cyrenaica Peninsula, and lineage D north of the High Atlas Mts. and in the Souss Valley (southern Morocco). Lineage B is subdivided into two subgroups, B1 (eastern Morocco and Algeria) and B2 (north-western Morocco). Italian tortoises harbour haplotypes of lineage A, Spanish tortoises of subgroup B1. Based on a relaxed molecular clock calibrated with fossil evidence, the six major mtDNA clades of T. graeca are estimated to have diverged approximately 4.2-1.8 Ma ago; the split between the clades representing the eastern subspecies T. g. ibera and T. g. terrestris is younger than the split between Western Mediterranean tortoises and T. g. armeniaca. The Western Mediterranean lineages A-D were dated to have diverged at least 1.4-1.1 Ma ago; B1 and B2 split approximately 0.7 Ma ago. Our results suggest that Italian and Spanish tortoises were either introduced or originated from trans-oceanic dispersal in historic or prehistoric times. Spur-thighed tortoises invaded North Africa probably across Near Eastern landbridges that emerged in the Late Tertiary. Their diversification in North Africa seems to be correlated with habitat aridization cycles during the Pleistocene. The ranges of the Western Mediterranean lineages largely correspond to the distribution of morphologically defined subspecies in North Africa, with exception of T. g. graeca and T. g. whitei, and of T. g. lamberti and T. g. marokkensis, which are not differentiated. We propose to lump the first two subspecies under the name of T. g. graeca and the latter under the name of T. g. marokkensis. The complex differentiation of spur-thighed tortoises in North Africa implies that the model of a bipartite east-west differentiation, as proposed for other Maghrebian amphibians and reptiles, may be too simplistic, reflecting incomplete locality sampling rather than actual phylogeographic differentiation.
20 February 2014 in Organisms Diversity & Evolution
Hyalomma aegyptium ticks were collected from tortoises, Testudo graeca, at localities in northern Africa, the Balkans, and the Near and Middle East. The intensity of infestation ranged from 1-37 ticks per tortoise. The sex ratio of feeding ticks was male-biased in all tested populations. Larger tortoises carried more ticks than did the smaller tortoises. The juveniles were either not infested, or carried only a poor tick load. Hyalomma aegyptium was absent in the western Souss Valley and Ourika Valley in Morocco, the Cyrenaica Peninsula in Libya, Jordan, and the Antilebanon Mountains in Syria. Hemolivia mauritanica, a heteroxenous apicomplexan cycling between T. graeca and H. aegyptium, was confirmed in Algeria, Romania, Turkey, Syria, Lebanon, and Iran. Its prevalence ranged from 84% in Romania (n = 45), 82% in eastern Turkey (n = 28), and 82% in the area of northwestern Syria with adjacent Turkish borderland (n = 90), to 38% in Lebanon (n = 8) and in only 1 of 16 sampled tortoises in Algeria. The intensity of parasitemia in the studied areas ranged from 0.01% up to 28.17%. The percentage of Hemolivia-infected erythrocytes was significantly higher in adults. All tortoises from Hyalomma-free areas were Hemolivia-negative. Remarkably, all 29 T. graeca from Jabal Duruz (southwestern Syria) and 36 T. graeca from the area north of Middle Atlas (Morocco) were Hemolivia-negative, despite the fact that ticks parasitized all adult tortoises in these localities. Identical host preferences of H. aegyptium and H. mauritanica suggest the occurrence of co-evolution within the Testudo-Hyalomma-Hemolivia host-parasite complex.
Despite being one of the most charismatic elements of the Mediterranean Basin fauna and its threatened status, the western Mediterranean range of Testudo graeca is at present very poorly known. The present work provides the most detailed geographical and ecological description for the North African clade of T. graeca so far. We gathered 283 occurrence data of T. graeca in North Africa and modelled the distribution by means of presence-only distribution modelling tools. The obtained model was then projected to southern Europe in order to explore whether the environmental characteristics of European populations fall into the predicted niche of the species in North Africa. T. graeca showed a wide environmental range in North Africa. Presence localities ranged from the sea level to 2090 m of altitude and from 116 to 1093 mm of annual precipitation. The presence-only model indicates that distribution in North Africa is mainly related to rainfall, specifically rainfall values in the wettest and coldest quarter of the year. The distribution model showed a range of ca. 1 000 000 km 2 . The projection of the model to southern Europe showed that the southern Iberian and Balkan Peninsulas, as well as most Mediterranean islands, present climatic conditions within those found in the range of the species in North Africa.
Demography and population biology aspects of the Moorish tortoise, Testudo graeca graeca were studied in the Parc National d'El Kala, north-eastern Algeria. Density, which was assessed by capture-mark-recapture procedures in a study plot of 30 ha surface, was 3.73 individuals × ha−1. Adult sex-ratio was nearly equal (1.06 : 1). Sexual maturity was reached at 7-8 years of age in the males and at 9-11 years in the females. The mean age of tortoises was 13.8 ± 6.4 years (n = 88). Maximum longevity was over 20 years in both sexes. The females reached larger body size than the males. The growth was rapid during the early years of age in both sexes, and tended to stop completely by the 9-12 years of life in the males and by the 8-10 years in the females. The instantaneous growth of adults which were recaptured in multiple years indicated an average growth rate of 11.34 mm/year in the males, and of 26.30 mm/year in the females. Mortality was estimated at 3.94%. The main mortality cause for the tortoises of this population was the occurrence of fires spreading throughout the study area. These data are of interest because demographic patterns of Testudo graeca seem to vary not only because of their relative genetical variability, but also because of their relative environmental stress.