No abstract

Migratory movements of the endangered stag beetles Lucanus cervus (18 males, 38 females) were monitored radio-telemetrically for three reproductive periods (2003)(2004)(2005). The aim of the study was to estimate the migratory range of freeranging individuals as a measure of connectivity among neighbouring populations for future conservation measures. Miniature transmitters of c. 350 mg (battery life: 10-15 days) were attached externally to the pronotum. Transmitter/beetle mass ratio was 12.8% on average (7.1-28.0%). Male dispersal behaviour consisted of frequent flights directed to sites with reproductive females and rarely of on-ground movement. Total displacement distance recorded was up to 2065 m, the maximum distance of a single flight being 1720 m. Flights always began at elevated structures such as trees and shrubs and took place in an air temperature range of 11-27 1C. Within this range, temperature did not influence flight distance. Female dispersal behaviour consisted mostly of a single flight, followed by mating and consequent ground movements towards oviposition sites. Total displacement distance recorded was up to 762.6 m, the maximum distance of a single flight being 701 m. Climatic constraints of flights were the same as in males. Modelling the dispersal behaviour suggests that about 1% of males are capable of maintaining gene flux among nest sites within a radius of about 3 km. However, the colonization of new nest sites depends on the dispersal ability of females and amounts to less than 1 km. Thus, isolated populations (distance to the next population greater than 3 km) have an increased probability of local extinction.

The migratory and reproductive behaviour ofBufo calamita was studied at four neighbouring breeding sites in the northern Rhineland, Germany, from 1986 to 1991. Radio telemetry and marking systems based on toe-clipping and on microchips were used to follow the tracks of toads and for individual recognition. Emphasis lay on estimates of (1) the exchange of reproductive individuals between neighbouring sites, and (2) the reproductive success at each site. Allozyme electrophoresis served to assess the genetic diversity of local populations. More than 90% of all reproductive males showed a lifelong fidelity to the site of first breeding, whereas females did not prefer certain breeding sites. Due to the female-biased exchange of individuals among neighbouring sites the genetic distance between local populations was generally low but increased with geographical distance. This pattern of spatial relations is consistent with the structure of a metapopulation. Morever, up to three mass immigrations of males per breeding period, replacing previously reproductive individuals, suggested the existence of temporal populations successively reproducing at the same locality. Genetic distances were considerably greater between temporal populations than between local ones, indicating partial reproductive isolation. In fact, an exchange of reproductive individuals between the temporal populations at each site was not detected, but gene flow due to the recruitment of first-breeders originating from offspring other than their own seems probable. Thus, natterjack metapopulations consist of interacting local and temporal populations. The reproductive success differed considerably among the four sites and also between the temporal populations. Three out of four local populations had low reproductive success as well as the latest temporal population. The persistence of these populations depended entirely on the recruitment of juveniles from the only self-sustaining local population. This "rescue-effect" impeded local extinction. The discussion focuses on the modifications required to fit the classical metapopulation concept to the empirical findings and their consequences for the dynamics of amphibian metapopulations.

The migratory behaviour of the toad Bufo bufo was studied from February 1985 to April 1986 in the submontane region of Bavaria, West Germany. Toads were fitted with a mechanical tracking device to record individual paths of migration. Three aspects of migratory behaviour were quantified: orientation in relation to the breeding site, straightness of path, and locomotory activity. The annual activity period began with migration from the hibernation sites to the breeding pond in April. The paths went straight towards the breeding pond independent of the distance (70-420 m). During the period of oviposition the preference for the breeding site direction vanished and toads moved away from the breeding pond, but in less straight paths than before. In summer migratory activity decreased considerably and was restricted to small areas, the home ranges, at distances of 55-1600 m from the natal breeding pond. The straightness of path was rather low, because toads often returned to their starting points. During rainy nights toads occasionally left their home ranges for extensive excursions. In autumn most toads again migrated towards the breeding pond, but paths were significantly less straight and direct than in spring. However, toads stopped before reaching the breeding pond and hibernated in holes or under the leaf layer. The mortality rate of tracked toads was about 45%. The relative influence of 17 environmental variables on locomotory activity was evaluated by principal component analysis and stepwise multiple regression. Temperature at night and rainfall variables accounted for significant amounts of variance, whereas temperature by day, air humidity, and atmospheric pressure showed no correlation. Activity decreased if temperature approached 0° C or after long periods without rainfall. Within a certain range of tolerance, however, the locomotory activity of the toads was widely independent of environmental factors, indicating that endogeneous factors are more important sources of variation in the migratory behaviour of these toads than commonly assumed.

Diploid and tetraploid populations of the leptodactylid frog Odontophrynus americanus were studied in the Cordoba province (central Argentine) to evaluate whether they represent a cryptic species pair rather than forms of the same species as they are considered at present. We examined three independent character complexes (external morphology, advertisement calls, allozymes) and quanti®ed the character-speci®c state of differentiation between diploids and tetraploids in comparison to that of the congeneric and widely sympatric O. occidentalis. Multivariate analysis of 15 morphometric characters indicated a remarkable degree of differention between both forms, but did not permit an unequivocal identi®cation of individuals (rate of erroneous classi®cation: 25%), whereas all O. occidentalis were classi®ed correctly. In contrast, the temporal structure of the advertisement call permitted a distinction of diploids, tetraploids and O. occidentalis even based on a single call. The latter differed from the other by producing a call consisting of several pulse trains. The pulse rate of the O. americanus call, which consists of a single pulse train, was signi®cantly lower in tetraploids than in diploids. The allozyme pattern and the corresponding allele frequencies indicated a close relationship among the three Odontophrynus taxa, but genetic distances obtained suggested a differentiation of gene pools at the level of species. Considering the degree of differentiation demonstrated in this study, the ploidy and the different geographical distribution, we conclude that diploids and tetraploids represent two species, similar to its North American counterpart Hyla chrysoscelis/H. versicolor. As the name O. americanus refers to tetraploids collected near Buenos Aires (Argentina), we describe the diploids as the new species Odontophrynus cordobae sp. nov.