J. Avian Biol. 41: 501504, 2010 doi: 10.1111/j.1600-048X.2010.05103.x # 2010 The Authors. J. Compilation # 2010 J. Avian Biol. Received 23 December 2009, accepted 8 May 2010 Absence of egg discrimination in a suitable cuckoo Cuculus canorus host breeding away from trees Anton Antonov, Ba ˚rd G. Stokke, Peter S. Ranke, Frode Fossøy, Arne Moksnes and Eivin Røskaft A. Antonov (anton.antonov@bio.ntnu.no), Ba ˚rd A. Stokke, Frode Fossø´y, A. Moksnes and E. Rø´skaft, Dept of Biology, Norwegian Univ of Science and Technology (NTNU), Realfagbygget, NO-7491 Trondheim, Norway and Centre for Advanced Study at the Norwegian Academy of Sciences and Letters, Drammensveien 78, NO-0271, Oslo, Norway.  Peter S. Ranke, Dept of Biology, Norwegian Univ of Science and Technology (NTNU), Realfagbygget, NO-7491 Norway. There is at present considerable variation in the level of antiparasite defences among different host species of avian brood parasites, but in many potential hosts some individuals reject poorly matching parasite eggs. Here we present unique absence of egg discrimination behaviour backed up by a lack of egg recognition abilities in a suitable common cuckoo Cuculus canorus host, the skylark Alauda arvensis. Skylarks did not show any clear rejection response to experimentally added highly non-mimetic foreign eggs in any behavioural context, even before they had started laying or when the whole clutch was exchanged with foreign eggs. This absence of antiparasite defence can be explained by the breeding habitat of larks consisting of largely treeless open landscapes where cuckoos have little access to the nests, thereby eroding the possibility of coevolutionary interactions. Our results are strikingly consistent with the spatial habitat structure hypothesis proposed to explain the occurrence and extent of avian host-parasite co-adaptation. Avian obligate brood parasites typically impose severe fitness costs on the host species they exploit for reproduc- tion. Consequently, hosts and parasites are involved in intricate coevolutionary arms races of building up compet- ing reciprocal adaptations (Davies and Brooke 1989b). The most common type of antiparasite defence in hosts of avian brood parasites is the recognition and rejection of foreign eggs, which is often followed by evolution of egg mimicry in the parasite (Rothstein and Robinson 1998). Eggs of the common cuckoo Cuculus canorus (hereafter cuckoo) have been found in at least 125 passerine species in Europe but only about 16 of these are regular hosts (Moksnes and Røskaft 1995, Davies 2000). At present, there is consider- able variation among and within host species in parasitism rates, egg rejection rates and cuckoo egg mimicry (Brooke and Davies 1988, Moksnes and Røskaft 1995, Stokke et al. 2007, 2008), which seems to reflect differential history and dynamics of host-parasite interactions within the framework of the arms race model (Davies and Brooke 1989a; Davies and Brooke 1989b). Furthermore, host-parasite interactions may evolve in complex geographic mosaics if there is gene flow between populations and local selection pressures vary (Thompson 2005). Cuckoos need elevated perches such as trees to be able to find and monitor host nests but the different host species and their populations often vary extensively in their tendency to breed close to trees, providing ample potential for locally diverse selection pressures (Øien et al. 1996, Røskaft et al. 2002). Accordingly, Røskaft et al. (2002) proposed the spatial habitat structure hypothesis which explains the occurrence and extent of adaptation in host- brood parasite systems by the proportion of host popula- tions breeding in the vicinity of trees, thereby accessible to cuckoos. Host species always breeding near trees were predicted to be strong rejecters, those breeding both near and far from trees intermediate rejecters and species always breeding far away from trees should be acceptors. Com- parative analyses supported the spatial habitat structure hypothesis but information on any suitable host species habitually breeding away from trees was unavailable (Røskaft et al. 2002). However, knowing host antiparasite defences at this extreme of habitat ranges is crucial for assessing the general applicability and validity of spatial habitat structure hypothesis. In this study we tested experimentally for the first time egg discrimination abilities of skylarks Alauda arvensis,a typical member of the Alaudidae family, whose species are suitable as cuckoo hosts but most of them breed in very open landscapes, away from trees (Cramp 1998). We show unique absence of egg discrimination backed up by complete lack of egg recognition abilities consistent with the scenario that antiparasite defences never evolved in skylarks due to the spatial structure of their breeding habitat, limiting the accessibility for the brood parasite. 501