A spider population in flux: selection and abandonment of artificial web-sites and the importance of intraspecific interactions in Lephthyphantes tenuis (Araneae: Linyphiidae) in wheat.

Lepthyphantes tenuis, a small sheet-webbuilding linyphiid spider is one of the most abundant spider species of cereal fields in Europe. In the present study we examined the process of web-site selection and web-site tenacity by adult females of this species in a winter wheat field. Spiders were selective in their choice of web-site. Different immigration rates into various manipulated web-sites, in field and laboratory, suggested that structural support and suitable micro-climate (high humidity) are the most important factors in the selection. Small holes dug in the ground were the most favoured web-sites. Web-site occupation was influenced by the presence of other conspecific spiders. Territorial contests occurred between spiders attempting to occupy the same web, these almost invariably led to the take-over of the web when the intruder was heavier. Interference, but also a certain level of tolerance, between spiders within the same web-site but in different webs was suggested by direct and indirect evidence. Many holes supported two or even three spiders in vertically stratified webs. Leaving probability of marked spiders was significantly higher in multiply occupied holes than in holes with a single web. Comparison with the results of a no-interference stochastic model showed that multiple occupancy in nature is less frequent than predicted by the model. There was further evidence for weak extra-web-interference between spiders in that multiple occupancy was even less frequent and overall occupancy was lower in web-sites which were packed close to each other. However, a level of tolerance for crowding is shown by the fact that closely packed hole colonies supported a spider density 13 times higher than in natural web-sites in the field. A marking experiment was carried out to gain information on web-site tenacity (i.e. the length of time a spider spends in a web-site) and abandonment. The average duration of tenacity was less than 2 days. A random loss function gave a good fit to the tenacity distribution and suggested that spiders abandoned web-sites randomly with a fixed leaving probability of c. 0.5. Individual webs were often used consecutively by more than one spider, and some spiders built more than one web in the same web-site. Calculations showed that abandonment is the most frequent leaving mode, while take-over by contest between spiders and disappearance due to destruction were some-what less frequent and equally likely modes of ending tenacity. It is suggested that the apparent contradiction between the selectiveness and competitiveness of spiders for web-sites and the relatively short tenacity observed can be resolved by hypothesising that spiders leave websites soon because they apply the strategy of spreading risk: spiders by frequently moving from one web-site to another distribute their reproductive efforts across several localities. This hypothesis is further supported by changes in web-site preference and ballooning behaviour at the onset of the reproductive stage in L. tenuis.