Transition between the landscapes and speciation in the Near Eastern brown frog, Rana macrocnemis

My publications on this subject

In the south-western part of the Caucasus Isthmus, the mountains of the Lesser Caucasus separate humid mountain forests of Colchis (somewhat similar to North American "temperate rainforests") from treeless volcanic uplands of the southern Georgia, Armenia and eastern Turkey. The border between these two landscapes is very sharp, and climatic conditions in some areas of central Georgia dramatically change at the distance of 10-20 km. Interestingly, the level of humidity here decreases and not increases with the elevation, in difference from most of mounainous regions of the western Eurasia.

There is a species, or a group of species of brown frogs (subgenus Rana ), related to European common or moor frog or north American wood frog. This group was a puzzle for taxonomists since over 100 years. At the end of XIX century, Boulenger described from the area two frog species: similar to common frog Rana macrocnemis and similar to moor frog Rana camerani . Common frog and moor frog are "good", distinct, reproductively isolated species, but their Caucasian equivalents are not. There are many populations where specimens have intermediate characters. Some taxonomists suggested that macrocnemis and camerani are rather subspecies than species. Others did not recognise taxonomic value of either, suggesting that there is only one monotypic, although highly variable species Rana macrocnemis .

We have studied the distribution of morphological and genetic traits of brown frogs throughout their range (the Caucasus, Turkey and the northern Iran), with the particular attention to the south-western Caucasus. In the western Caucasus, the forms are morphologically and genetically distinct. Nominal macrocnemis from humid forests of Colchis differ from nominal camerani from Javakheti plateau in southern Georgia in states of many morphological and genetic characters - including body shape, coloration, skin structure, the distribution of alleles at allozyme and microsatellite DNA loci. Actually, the forms are separated by multiple stepped clines going across the mountains of the Lesser Caucasus. However, frogs from the rest of the Near East, at least morphologically, are rather intermediate between these adjacent, but distinct geographic populations from the Caucasus. If the "sharp transition" region is ignored, one may decide that there is no geographically distinct pattern of variation in the group, and morphological and genetic differences between populations can be explained just by isolation by distance or adaptation to local environments! Moreover, even within the western Caucasus, the differences are strongest and clearest between populations of opposite forms from the transition (or hybrid) zone, and away from this zone the genetic and morphological differences become vague.

The analysis of the mitochondrial DNA of frogs from central Turkey to Daghestan showed, that the frogs from humid forests of Colchis with explicitly expressed traits of the forest form macrocnemis are marked by a monophyletic group of haplotypes, differing them from animals collected in other parts of the range. This means, macrocnemis and camerani from the western Caucasus represent different eviolutionary lineages and not just spontaneously developed morphotypes fixed by selection. However, in areas surrounding Colchis forests, haplotype of macrocnemis is found in admixture with another group of haplotypes, which reflects long-term hybridisation and introgression.

There are some evidences that the differences between macrocnemis and camerani at both sides of the Lesser Caucasus are maintained by differential landscape-dependent selection. In the same ecological conditions, camerani grow more rapidly than macrocnemis and produces more offspring per unit time. This provides the form a selective advantage in severe climatic conditions of Javakheti plateau. In other parts of the range, gradient of ecological conditions is not so sharp and no sharp limit between the forms can be maintained.

It is, however, another mechanism maintaining the differences between the forms in the transition zone. Microsatellite genotype distribution in populations from the centre of the cline significantly deviates from Hardy-Weinberg equilibrium and linkage equilibrium. There is a significante deficite of heterozygotes. In fact, in the center of the cline the two races coexist in same locations, and hybridization between them is limited.

The divergence within Rana macrocnemis has started in Ice Age, when populations of frogs isolated in humid forests of Colchis gained traits differing them from frogs living in surrounding areas. After Ice Age was over 10-13 thousands of years ago, features of the forest race spread over the entire Caucasus and the Asia Minor. However, the forms remained separated at both sides of the Lesser Caucasus, where the landscape-dependent selection maintained the differences. High specialisation of both forms across the hybrid zone triggered development of internal mechanisms limiting hybridisation.

In the result, we have two clearly differing subspecies or semispecies in very limited area at the eastern coast of the Black Sea, but neither of the two forms can be delimited in geographic space or in space of genetic and morphological traits. There are species-level differences but there are no species. Somewhat similar situation is known for so called "species rings". However, in our case the "extreme" populations of the group have been in contact at all stages of the divergence, whereas in typical "ring" species extreme populations usually have long history of independent evolution.

Publications:

Up to the title

1.      Tarkhnishvili, D. N. and Gokhelashvili, R. K., 1996. A contribution to the ecological genetics of frogs: age structure and frequency of striped specimens in some Caucasian populations of the Rana macrocnemis complex. Alytes (Paris), 14(1): 27-41.

2.      Tarkhnishvili, D. N., 1996. Genetic relationships in local populations of brown frogs - analysis of distribution of a character under selection. In: Population Genetic Group, 30th annual meeting, University of Edinburgh, 17-20 Dec. 1996, Paper Abstr., p.42.

3.      Tarkhnishvili, D. N., 1999. Phylogenetic reconstruction on the local population level: use of characters under selection. -Advances in Amphibian Research in the Former Soviet Union, 3: 35-42.

4.       Tarkhnishvili DN & Gokhelashvili RK, 1999. The Amphibians of the Caucasus. Pensoft publications, Sofia, 233p.

5.      Tarkhnishvili, D. N., Hille, A. A. and Böhme, W., 1999. Allozyme variations in brown frogs from Georgia: extensive gene flow between nominal taxa. In: Societas Europaea Herpetologica, 10 th Ordinary General Meeting, Book of Abstracts. Irakleio: 244-245.

6.       Tarkhnishvili, D. N., Arntzen, J. W. and Thorpe, R. S. , 1999. Morphological variability in brown frogs from the Caucasus and the taxonomy of the Rana macrocnemis group. Herpetologica, 55 (3): 406-417.

7.       Tarkhnishvili, D. N., Hille, A. A. and Böhme, W., 2001. Humid forest refugia, speciation and secondary introgression between two evolutionary lineages: differentiation in a near eastern brown frog, Rana macrocnemis . Biological Journal of the Linnean Society, 74: 141-156.

8.       Tarkhnishvili, D. N., 2001. The evolution of the External Morphology of Brown Frogs. - Journal of Morphology, 248 (3): 290 (Abstr. 6 th Congr. Vertebr. Morphol.)