1993 Asiatic Research Vol. 5, 147-165 I December Herpetological pp. The Ecology of the Caucasian Salamander (Mertensiella caucasica Waga) in a Local Population DAVID N. TARKHNISHVILI 1 AND IRINA. A. SERBINOVA2 'Institute of Zoology, Georgian Academy of Sciences, Tbilisi, Georgia ^Moscow Zoo, 123242 Moscow, Russia Abstract. -The different aspects of ecology of Mertensiella caucasica Waga, 1876 were investigated in a local population from Borjomi Canyon (central Georgia) for five years (1985-1990). The aspects of the is about 16.9 female. species' life cycle were more precisely determined. The main fecundity eggs per There are about 2 years in a period from egg deposition (June to first half of July) to the end of metamorphosis in nature. Animals have spent most of the time in shelters after metamorphosis. They a appear on the ground surface at night during the breeding period. Commonly the adults don't retreat to small great distance from population localities. Localities are situated in comparatively plots (100-300m) 1189 along the streams. Estimation of adult animal number showed that the population consists of specimens (1989). Annual adult survival is higher than known values of most amphibians (approaches 0.77). Larval survival is 0.27-0.32 in the second year of life. The characteristics of demography in on subUe constitution of (especially, low renewal rates) and spatial restriction localities depends mostly of M. the species (which is a result of allometric growth specifics). The small recent geographical range caucasica is explained as a result of morphological and ecological peculiarities. General morphological salamander tribe. constitution limits adaptive possibilities of any particular representative of die European lusitanica. This is an explanation of quite high ecological similarity of M. caucasica and Chioglossa Key Words: Amphibia, Caudata, Salamandridae, Mertensiella caucasica, Caucasus Mountains, Georgia, population ecology. Introduction stationary investigations about the ecology of the rare or narrow-ranged species. Long the Natural populations are the single way of term research of such species enlarges of the of wide species existence. Autoecological research knowledge biology doesn't allow a complete understanding of taxonomic groups. Moreover, these useful to find out the life of a species in nature. That is why investigations may be there must be information of life cycles, ways of rare species preservation. geographical range, population size, number dynamics, etc. On the other hand it A local population of the endemic from is hard to explain ecological aspects of the salamander, (Mertensiella caucasica), Caucasus of has been species existence without any information the western Georgia a five of their habitat preferences, feeding habits, investigated for year period (1985- additional breeding sites, etc. 1990). This work gives information about the life history of this By analyzing connections between species. species population ecology and autecology, of the as well as morphology and geographical The geographical distribution distribution, the most complete notion can Caucasian Salamander was mainly of this be formed. Investigations on some established in the beginning century. summarized amphibian species biology have allowed Information was by Nikolsky scientists to elaborate complex works (1913). Later investigations commonly localities or in connected with different aspects of their life took place in earlier reported areas. Some new localities for history. A wonderful example is Bell's adjacent works on the Smooth Newt (Bell and salamanders were found by Bakradze and Lawton, 1975; Bell, 1977). Tartarashvili (pers. comm.). The real geographic range of M. caucasica was There aren't many data of regular established. The Caucasian Salamander is © 1993 by Asiatic Herpetological Research Vol. 5 p. 148 Asiatic Herpetological Research December 1993 FIG. 1. Distribution of Mertensiella caucasica. distributed in external spurs of the sibiricus, more restricted to water habitats Trialetian Mountain Range. Probably it is than M. caucasica, is geographical limited the result of historical changes in the Kura by coniferous forests like M. caucasica in River bed (Fig. 1). Populations are mainly eastern localities (Paraskiv, 1953). Local distributed in the forest belt, but in some populations, distributed along tributaries of places they can be found close to subalpine the Chorokh and Kura rivers (in upper meadows. Humidity in the species' flow), are formed by salamanders within its locations reaches 1000 mm or more per area. Width of streams in salamander plots year (another narrow-ranged representative is not more that 1-1.5 m in spring and of the salamander tribe, Chioglossa because of stepped disposition of streams, lusitanica, has similar requirements of they run slowly in some places. There are humidity). many slowly draining pools about 20-30 cm in depth with a lot of shelters. The In the most dry part of the range of M. bottoms of streams and pools are covered caucasica, the eastern one, salamanders live with stones, and there is a lot of non- only in coniferous forest. When humidity decayed organic matter. Stepped reaches 1200 mm/year in the middle area, disposition of streams is formed by stoned they can also be found in subalpine conglomerations and fallen logs. meadows. Salamanders are distributed in Apparently, mountain ranges between deciduous forest only close to the Black stream canyons don't allow wide Sea coast, where humidity is very high salamander migration and local populations (2000-2400 mm/year; Fig. 1). The high are comparatively isolated. There is no dependence of the animal on humidity does evidence that direct migrations of animals not itself limit the species distribution, but occurs during their life cycle. Individuals determines sensitivity of specimens to other are found a maximum of 200-300 m environmental factors. It is very interesting distance from streams. that the rheophilous species Ranodon December 1993 Asiatic Herpetological Research Vol. 5 p. 149 yyj— stream bed and branches -*S —fallen trees and piles of stones O — large pools where larvae were found \ — steep banks where shelters are located FIG. 2. Schematic diagram of the study site for Mertensiella caucaska. Tfie Study Area of each adult animal was mapped, substrate type and distance from stream bank (more The studied population inhabits or less than 50 cm) was recorded. Adult coniferous forest ecosystems along the animals were marked individually by toe- second range tributary of the Kura River in clipping. Combinations from clipped digits Borjomi Canyon (eastern part of the in hind-limbs (not more than 2 in 1 foot) species' range) (Fig. 1). The plant responds to individual number of animals association is formed by Taxus baccate, from 1 to 99. Zero-1 clipped digit in the Picea orientalis and deciduous spots. The front leg mean number of hundred. Marks size of the inhabited location is a bit more of salamanders recaptured in the next year than 200 m, and it is situated between 1000 were renewed. Data of capture-recapture and 1300 m altitude, about 2 km from the were statistically counted as in Kaughley stream mouth. Slopes are precipitous, built (1977). Substrates of animals caught were by corrosion of underground tree roots or subdivided in 6 types: shallow water; sand relatively gradual, partially covered by and pebbles above water shore; wet stones; pteridium, Matteuccia struthiopteris, from wet ground; moss or lichens; dry ground the adjoining stream banks. There are and stones. These types were ranked some stone conglomerations and fallen according to their humidity. Basic trees in the study area, shown on the map investigations were conducted on 8-10 and (Fig. 2). Air temperature is close to stream 21-23 June, 1986, 24-28 June, and 5-7. water temperature (13-15°C in summer) in August, 1987, 3-5 and 21-24 July, 1988, shelters formed by stones and logs. 16 June- 12 July, 1989, 2-9 July, 1990. Dynamics of air temperature in Borjomi Canyon in May to July, 1989 is shown in We had 337 contacts with males and 202 Fig. 3. Quiet pools and shelters are relative with females (including specimens found rare, slopes are steeper and stream flow is two or more time). Recording of larvae faster at upper and lower localities. Density was conducted during night excursions. of salamanders here falls rapidly as well as away from the stream banks in these We caught females from nature in the places. reproductive period and obtained eggs using a hormonal stimulation method Methods (Gontcharov et al., 1989) to study some ecological and morphological features of The main quantitative data were obtained early development. Eggs were incubated in during excursions with a lantern after Weiss bowls in dechloronated water at a sunset along the study area. The location temperature of 14°C as well as in aquaria at Vol. 5 p. 150 Asiatic Herpetological Research December 1993 Date FIG. 3. Air temperature at the Mertensiella caucasica study site during the period of reproductive activity. Stippled bars represent periods of rainfall. -i precision of 0.1 mm. The coloration 40 Males Females patterns of some animals was also i- 30- recorded. X> S 20- 3 Results Z 10- The niche oflarvae and adult specimens. 12 3 4 YV~ i-2 5 6 Salamanders don't have an even Substrate distribution within the study site. Preference to every substrate depends on FIG. 4. Location of salamanders by substrate type. the amount of moisture of each 1- shallow water; 2- wet sand and pebbles; 3- wet particular substrate. of decreases stones; 4- moist ground; 5- moss or lichens; 6- dry Frequency captures with distance from water or stones. Solid bars represent males. Stippled bars potential represent females. shelters. The number of animals captured out of shelters depends on time and season. a temperature varying from 5-22°C. Before Most adult specimens were recorded close completion of metamorphosis, larvae were to the stream (less than 50 cm from the kept in 20 liter aquaria, where water was water shore): 60±4% of males, 62±5% of changed every third day.