Behavioural Processes 53 (2001) 57 – 64 Transitive predatory relationships of spider species (Arachnida, Araneae) in laboratory tests B.A. Heuts a, *, T.M. Brunt b a Uniersity of Amsterdam, Swammerdam Institute of Lifescience, Kruislaan 320, 1098 SM Amsterdam, The Netherlands b Former Department of Animal Behaiour, Plantage Muidergracht 14, Amsterdam, The Netherlands Received 14 August 2000; received in revised form 12 October 2000; accepted 16 October 2000 Abstract Interspecific predation of spiders was studied in the laboratory in view of possible competition in the wild. Certain species killed other species even if handicapped by smaller size. Thirty eight spider species were involved in such a relationship and their predatory relationships were significantly reliable and transitive (‘linear’ or ‘non-triangular’). A theridiid species (Theridion tinctum) showed the highest rank in terms of killing seven ‘beta species’, i.e. species capable of killing at least one alien species of larger size than themselves. Another theridiid (Steatoda grossa ) obtained the second rank by killing five beta species. Experiments in both the wild and laboratory may, further, investigate other factors than body size that may be relevant to competition, such as behaviour-related characteristics (e.g. web structure and biting speed) and ecological factors (e.g. different susceptibilities of the species to parasite or predator attack). © 2001 Elsevier Science B.V. All rights reserved. Keywords: Spiders; Araneae; Interspecific predation; Transitive relationship www.elsevier.com/locate/behavproc 1. Introduction Spiders that can kill heterospecific spiders sur- passing them in size may be assumed to have an advantage in competition if the species occur in the same habitat and do not importantly differ from each other in other characteristics than size, such as susceptibility to parasite and predator attack, access to prey items, food conversion, and calories needed for reproduction. We investigate here the reliability and linearity of a species hierarchy based on individuals that kill (and usually eat) larger individuals of other spider species. We disregard predation of larger on smaller individuals (since all species are similar in their capacity of killing smaller individuals: pers. obs.). We define body size as a composite of larger cephalothorax and leg size (longer and thicker legs) while disregarding abdomen size, hair size on the legs, and relative size of mouth parts in the cephalothorax. Our laboratory tests can be regarded as ‘neu- tral’ tests in which several naturally varying fac- tors are held constant, such as the availability of * Corresponding author. Tel.: +31-20-5257731; fax: +31- 20-5257662. E-mail address: heuts@bio.uva.nl (B.A. Heuts). 0376-6357/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII:S0376-6357(00)00138-8