[Ecological isolation and the importance of competition, exemplified by the distributional pattern of the lycosids of a coastal landscape].

In a coastal landscape at the Baltic Sea the distribution of the Lycosidae was analysed, mainly by pitfall traps. 11 dominant species occurred in the 17 different types of habitat-belonging to salt meadows, dry heath, and dunes-and were isolated with only a few exceptions. Isolation of the species was observed 1. in the macrospatial distribution, 2. in the microspatial horizontal distribution, 3. in the microspatial vertical distribution, 4. in fluctuations of population density during a period of 6 years, 5. in the seasonal periodicity, 6. in the diurnal activity.In some cases the distributional pattern overlapped in such a way that biotic relations between lycosid spiders of comparable dominance were possible. In a model study such pairs of species were observed in glass jars in the laboratory as mixed and unmixed populations with a high density (Pirata piraticus-Pardosa purbeckensis, Pardosa purbeckensis-Pardosa pullata, Alopecosa pulverulenta-Pardosa pullata). In comparison with the unmixed populations the death rate of one species in the mixed populations was higher, so that this species became extinct after a period shorter than 30 days. The factors of mortality were: 1. interference (spatial competition)-in some cases the rival was even killed (without being eaten); 2. competitive exploitation of food; 3. predator-prey-relations. These components were not clearly seperated. As this is also true for interference and predation, the predator-prey-relations should be part of the definition of "competition", at least for predatory groups. In the natural habitats only predator-prey-relations between species of lycosids not being isolated may have played a certain role.The quantitative spatial distribution of the lycosids in the 17 habitats was expressed as biovolume of the catches of one pitfall trap. It was clearly correlated with a "prey index" calculated from the biovolume of catches of collemboles, leafhoppers, and flies (r=+0.86). In contrast the Carabidae did not show any correlation with the prey index. There were "vacant niches" in the spatial and temporal distribution of predators with the same potential prey as the lycosids (in addition mainly carabids). Lycosids and carabids scarcely overlapped in their seasonal periodicity. When, in spring, both groups occurred together, obviously interference did not play an important role.Abiotic factors obviously were most important in limiting the abundance of those lycosid spiders that attained only low densities in some types of habitat. The possibility of competitive displacement of one lycosid by another was analysed on the basis of ecological data concerning the North German species. It was considered to be improbable in most cases.To sum up, the distributional pattern of the lycosids in the investigated area and their relation to other predators depend much more on endogenous factors (e.g. environmental requirements, periodicity) than on competition.