lable at ScienceDirect

Animal Behaviour 81 (2011) 359–360

Contents lists avai

Animal Behaviour

journal homepage: www.elsevier .com/locate/anbehav

In Focus

Featured Articles in This Month’s Animal Behaviour

Figure 1. Supplementary food in their territories leads male great tits to delay theirmorning singing bouts. Photo: Soenke Morsch.

Extra Rations Delay the Dawn Chorus

Almost half the households in the U.K. provide food for gardenbirds, and it has been calculated that the amount of wild bird foodsold per year could, hypothetically, feed 30 million great tits. Thesame is true for theU.S.We are, it seems, engaged in a giant ecologicalexperiment. It has been known that such food supplements canadvance egg-laying dates and also increase fledging success, butwhat about other effects on behaviour? In this month’s issue(pp. 361–365), Katja Saggese, Fränzi Korner-Nievergelt, Tore Slags-vold and Valentin Amrhein investigate how male singing behaviour,specifically the tendencyofmale great tits to singbefore dawnduringthe breeding season, is influenced by supplementary feeding.

Male great tits establish breeding territories in the spring, and singbefore dawndthe famous dawn chorusdduring this period. Giventhat singing is energetically expensive and that it has to be tradedoff against other activities, such as foraging, providing males withfood can be expected to influence their singing behaviour. Extra foodcould, forexample, attractothermales to the territory, sothat the terri-tory owner has to invest more energy in defence, whether by singingor actively chasing intruders away. In addition, theoretical modelssuggest thatextra foodshould leadmales to singearlier in themorningor increase the effort they put into their singing bouts. Studies sup-porting these predictions have been conducted before, but the periodof supplementary feeding has been very short, around 1–5 days, soSaggese and colleagues decided to investigate what would happen ifsupplementary feeding was extended. They provided supplementaryfood tomales for aperiod of 2weeks, so thatmales receivedextra fooduntil about 3 weeks before the start of the egg-laying period.

Males in territories with extra food were found to start singinglater relative to sunrise than males in unsupplemented territories(Fig. 1). In 36% of observation sessions, supplemented males startedto sing only after the sun had risen, skipping their regular dawnsinging session. Extra food did not, however, affect the male’ssong output. The late morning start of supplemented males alsocontinued for 2 weeks after the end of supplementary feeding.

Given that theoretical considerations suggest supplementedmales should start singing earlier, these results are both surprisingand intriguing. One suggestion is that males postpone singingbecause they can eat human-provided food in the dark, whichisn’t possible for their natural insect-hunting behaviour. As theresearchers never saw a single great tit male feeding before sunrise,however, this explanation seems unlikely.

Another possibility is that the extra food attracted other great titmales to the scene, and so requiredmales to invest in extra defencebut, again, Saggese and colleagues found no evidence of more greattit intruders in supplemented territories. A third possibility is that

0003-3472/$38.00 2010 The Association for the Study of Animal Behaviour. Published bdoi:10.1016/j.anbehav.2010.12.008

more predators would be attracted to territories with a reliablesupply of food in a particular location, and so males may haveavoided singing before dawn to reduce their predation risk. Thelikelihood of this as an explanation must be tempered, however,by the fact that males continued to delay singing for 2 weeks afterfeeding was discontinued; as Saggese and colleagues note, it wouldbe surprising if predation risk continued to be elevated for this long.

So why would males continue singing later once the extra foodwas removed? Saggese et al. suggest that supplemented malescould have acquired higher energy reserves or their perception ofterritory quality could have changed. It is also possible, however,that males could have continued to get food from other human-provided sources, or that the quality of the supplementary foodcould have changed the nature of their territorial behaviour.

At present, we don’t have any good answers to these questions,nor is it apparent why this longer-term study should show differentresults to short-term supplementation. Clearly, replication of thisstudy is needed to determine whether the effect is robust, and totease out why longer-term supplementation should have theseeffects. Indeed, Saggese and colleagues actively encourage furtherexperiments so that we can get a more precise handle on howandwhy supplementary feeding changes male territorial behaviourin these long-lasting ways, as well as understand how such feedingmight also influence female behaviour.

Louise BarrettExecutive Editor

y Elsevier Ltd.

Figure 2. A male Schizocosa ocreata wolf spider (centre), with the four natural substrates on which it courts (clockwise from upper left): leaf, wood, soil and rock.Photo: Shira D. Gordon and George W. Uetz.

In Focus / Animal Behaviour 81 (2011) 359–360360

Flexible Communication in Wolf Spiders

Some animals have the ability to vary their signalling behaviourin response to external circumstances, for example by adjusting theamplitude and frequency characteristics of vocalizations inresponse to noise. These kinds of adjustments are typically adap-tive, in that they enhance the efficiency of communication in theface of environmental change. Traditionally, such adaptive plas-ticity has been considered a hallmark of vertebrate behaviour,and especially of the behaviour of birds and mammals. Recently,however, it has become apparent that some invertebrates are alsoable to vary their signalling adaptively to fit environmental condi-tions. A study of wolf spiders in this issue (pp. 367–375), writtenby Shira D. Gordon and George W. Uetz of the University of Cincin-nati, provides a convincing new example of such adaptive plasticityin invertebrate communication.

Gordon and Uetz studied the courtship signals of brush-leggedwolf spiders, Schizocosa ocreata. Males of this species court femaleswith visual displays such as the ‘wave and arch’, in which a malespider raises and bends one of its forelegs, which is ornamentedwith tufts of bristles found only in males. Courting males alsoproduce seismic displays, vibrating the surface onwhich they standby tapping their forelegs or bouncing their whole bodies against it.Previous studies have shown that either visual or seismic displayalone can elicit mating behaviour from females, and that the twoacting together are even more effective.

In their natural habitat, male brush-legged spiders signal ona variety of surface types, including soil, rock, wood and leaf litter(Fig. 2). In one experiment, Gordon and Uetz induced captive malesto signal to females on each of these substrates. Males signalling onleaf litter weremore than twice as successful in inducing females to

mate as were males signalling on the other surfaces. Given thatsubstrate type is unlikely to affect transmission of visual signals,these results suggest that substrate type must affect seismic trans-mission.Totestwhether this is thecase,GordonandUetzuseda laservibrometer to measure the substrate vibrations produced by malespiders signalling on the various types of surfaces. Seismic signalstransmittedwell on leaf litter, and hardly at all on soil, rock orwood.

The authors gave captive wolf spiders a choice of occupying anyof the four substrate types. Males visited all four substrates equallyoften, but remained about twice as long per visit on leaf litter. Totaltime spent per substrate was significantly greater for leaf litter thanfor the other substrate types for both males and females. The mostintriguing results concerned the form of male courtship on the foursubstrates. On the one hand, males performed about as many of the‘body bounces’ and ‘leg taps’ that produce seismic signals when sig-nalling on rock, wood and soil, which do not transmit the vibra-tions, as when signalling on leaf litter, which does. On the otherhand, males produced significantly more of the primarily visualwave and arch display on the substrates that do not transmit vibra-tions as on the substrate that does.

The continued use of the movements that produce seismicsignals on substrates that do not convey those signals seemsnonadaptive. Perhaps the explanation is that the visual compo-nents of these movements are also important as signals. Bycontrast, the increase in purely visual displays when seismic signal-ling is unavailable provides an excellent illustration of adaptiveplasticity: when one channel is closed, the signaller puts moreeffort into another channel that remains open.

William A. SearcyExecutive Editor