New communal roosting tradition established through experimental translocation in a Neotropical harvestman

Abstract

‘Tradition’ has been invoked to explain instances in which animals aggregate repeatedly in the same locations for no apparent reason, but alternative explanations, such as cryptic habitat selection, are difficult to rule out. Distinguishing among these hypotheses requires field experiments. We studied a species of harvestman ( Prionostemma sp.) that forages solitarily at night and forms roosting aggregations in spiny palms by dawn. Aggregations have formed repeatedly in the same sites for over a decade, and yet the aggregation sites do not appear to differ from unused sites in tree characteristics or microclimate. Previous research suggested that the harvestmen find aggregation sites by detecting chemicals left behind by conspecifics. If so, it should be possible to establish new aggregation sites simply by moving harvestmen to new locations. We carried out such an experiment and, as predicted, release sites attracted harvestmen at much higher rates than did matched control sites. Most individuals that came to release sites were new recruits, and one site continued to attract several harvestmen for at least 7 days, suggesting that a new roosting tradition was established. We also revisited the question of whether any other factors, besides past usage, predict site use. Among sites used by the harvestmen, the size of the aggregations was strongly predicted by microclimate and spiny palm characteristics. Together these findings support the hypothesis that attraction to conspecific cues is the primary mechanism responsible for the long-term stability of aggregation sites, while also suggesting that site characteristics play a previously unrecognized role. Understanding how site traditions develop may have direct applications in conservation. ► We found that harvestmen have roosted communally in the same sites for 10 years. ► Site characteristics predicted the size, but not the location, of the aggregations. ► These appear to be traditional roosts maintained by chemical conspecific attraction. ► We established a new communal roost simply by translocating animals to new sites. ► Our findings may be relevant for conserving species that roost communally.