828 SHORT COMMUNICATIONS

-t 0.4 daysbetween the third- and fourth-hatchedchicks. FUOLE,G. N., ANDS. I. ROTHSTEIN.1977. Clutch size If these limited data are representative (no other de- determination, egg size, and eggshellthickness in tailed information is available), all chicks are well sep- the Pied-billed Grebe. Auk 94~371-373. arated in the pattern of size differences initially pro- Got, M. 1986. Colonial versus territorial breeding duced within broods by hatching asynchrony. First- of the Great Crested Grebe Podicepscristatus on laid eggsaveraged 2.6% less than second-laid eggsin Lake Druzno. Acta Om. 22:95-145. volume, but it seems safe to infer that this difference HENRIKSEN,K. 1992. Nesting ecologyand production must be of little importancein reducingthe initial body- of young in the Great CrestedGrebe Podicepscris- mass difference between first- and second-hatched tatusin a hypereutrophicDanish lake. Dansk Om. chicks. Foren. Tidsskr. 86:163-168. In conclusion,egg volume seemsto remain constant LACK, D. 1954. The natural regulation of animal after the first-laid es.ain clutchesof Great CrestedGrebes numbers. Oxford Univ. Press, London. not adding to the ze hierarchy of chicks arising from LEBLANC,Y. 1987. Intraclutch variation in egg size asynchronoushatching. In addition, many authors ar- of Canada Geese. Can. J. Zool. 65:3044-3047. gue that the increase in size of the second-laid egg PARSONS,J. 1976. Factors determining the number relative to the first-laid found in clutchesof many bird and size of eggslaid by the Herring Gull. Condor speciesmay be related to morphological or physiolog- 78:481-492. ical constraints on the size of the first-laid egg (e.g., REYNOLDS,A. 1974. Measurements made on Great Parsons 1976, Leblanc 1987, Arnold 1991). Hence, no Crested Grebe eggsduring the incubation period. adaptive hypothesis seems necessaryto explain intra- Rep. of the Rye-Meads Ringing Group 7:28-34. clutch variation in eggvolume of Great CrestedGrebes. SIMMONS,K.E.L. 1970. Ecological determinants of breeding adaptationsand social behaviour in two LITERATURE CITED fish-eating birds, p. 37-77. In J. H. Crook [ed.],

ARNOLD, T. W. 1991. Intraclutch variation in egg Social behaviour in birds and mammals-essays size of American Coots. Condor 93: 19-27. on the social ethology of animals and man. Aca- BASECKE,K. 1953. Vom Haubentaucher. Vogelwelt demic Press, New York. 74:149-150. SLAGSVOLD,T., J. SANDvtK, G. ROFSTAD,b. LoR- &, S. AND K.E.L. SIMMONS.[EDS]. 1977. The ENTSEN,AND M. HUSBY 1984. On the adaptive birds of the Western Palearctic, Vol. 1. Oxford value of intraclutch egg-size variation in birds. Univ. Press, Oxford, U.K. Auk 101:685-697. FORBES,M.R.L., AND C. D. ANKNEY 1988. Intra- ZAR, J. H. 1984. Biostatisticalanalysis. Prentice-Hall, clutch variation in egg weights of Pied-billed Englewood Cliffs, J. Grebes. Condor 90:709-7 11.

The Condor97~828-831 0 The CooperOrnithological Society I995

PIED FLYCATCHERS PREFER TO NEST IN CLEAN NEST BOXES IN AN AREA WITH DETRIMENTAL NEST ECTOPARASITES ’

SANTIAGOMERINO AND JA~MEPOTTI Departamento Biologta’ Animal, Univ. Alcald de Henares, E-28871 Madrid, Spain

Key words: Ectoparasitism;hole-nesting birds; nest ing previously used cavities for nesting could be ex- box studies;nest site selection;Ficedula hypoleuca. posed to higher levels of infestation by nest ectopar- asites than birds choosing empty holes for breeding. Moller (1989, 1992) pointed out a number of possible Theory predicts that if reproductive costs due to nest artifacts in nest box studiesdue to the common practice ectoparasitesare severe,birds should selectempty cav- of removing old nests from nest boxes. There are ap- ities-orremove old nest material before breeding(Mell- parently few bird speciesthat remove old nest material er 1989. 1992: but seeThomoson and Neil1 1991. Da- before breeding(Moller 1989). Therefore, birds choos- vis et al. 1994). Several studies have tested this pre- diction and failed to demonstrate discrimination of dirty/parasitized boxes in areas where the nest ecto- parasiteshave few detrimental effectson birds’ fitness 1 Received 20 December 1994. Accepted 20 April (Thompson and Neil1 1991, Ore11et al. 1993, Davis et 1995. al. 1994, Mappes et al. 1994). Two studies were con- SHORT COMMUNICATIONS 829 ducted with Pied Flycatchers (Ficedula hypoleuca),a TABLE 1. Number (n) of experimental(provided with common hole-nesting passerinewith a wide distribu- an old nest) and control (empty) nest boxes in each tion range in the Palearctic(Cramp and Perrins 1993). study year. Also shown are numbers of boxes occupied After finding no effect of nest box condition (dirty vs. by tits (Pam spp.). clean) on nest box choice, or of ectoparasiteson nest- ling growth in a northern population of flycatchers, 1993 1994 Ore11et al. (1993) hypothesized that in southern lati- Used Used tudes, with presumably higher abundanceof ectopar- by Avail- by AVaiL asites, Pied Flycatcherscould behave differently in or- N&box n tits able n tits able der to avoid the detrimental effectsof ectoparasitism. Experimental 20 0 20 13 2 11 Here, we show that in the southernEurope, where the Control 39 6 33 17 0 17 premise of reproductive costs causedby ectoparasites is fulfilled (Potti and Merino 1994, Merino and Potti 1995), Pied Flycatchersprefer empty, clean nest boxes over nest boxes with old nest material. in another 20% of nest boxes, the nest material was partly removed. METHODS In 1994, Pied Flycatchers rejected using nest boxes The experimentswere conductedin centralSpain where with old nests (Fig. la), although the availability of Pied Flycatchers have been studied since 1984 (e.g., empty nest boxes was lower in the experimental area Potti and Montalvo 1991). There were 175 nest boxes (t = 2.95, 14 df, P = 0.01). Only two nest boxes con- (internal dimensions 11 x 11 x 17 cm, entrance hole taining old nest material were usedand both were from 3.2 cm) in approximately 100 ha, with a mean distance the late period of the laying season,one of them being between adjacent nest boxes of 25 m. Nests of Pied a replacement clutch of a female that first chose an Flycatchers from the 1992 breeding seasonwere de- empty neighbor nest box and abandoned after being faunated in Berlese funnels for 48 hr, and then dis- captured by us. No nest material was apparently re- mantled in search of other ectoparasitesnot obtained moved from experimental boxes in this year. by this method (see Merino and Potti 1995). After Overall, there was a preference of Pied Flycatchers this, nest material was placed, trying to reshapeit in a for empty nestboxes(log-linear analysis;log-likelihood form similar to natural nests, before the breeding sea- x2 = 10.03, 1 df, P = 0.0015) and this effect was in- son of 1993 inside 20 nest boxes, leaving two empty dependent of year (log-likelihood x2 = 2.43, 1 df, P = nest boxes between each manipulated nest box with 0.1189). old nest (see Table 1). The experiment was repeated As effectsof parasitesmay differ dependingon breed- in a different part of the study area in 1994, using 13 ing phenology (Merino and Potti 1995) we examined unmanipulated nests which had been left untouched if choice of nest boxes is affected by the date of laying. over the non-breeding season(Table l), in an attempt When nests are divided into early and late in relation to simulate conditions in nature (see Moller 1989). to standardizedlaying date (aboveand below the mean), Some of thesenests had been parasitizedby fleas,mites there was a clear preference for empty nest boxes in and blowflies in 1993, although we did not check for the late period (xZl = 10.32, P = O.OOl), while that their presencebefore the experiment. However, we may preferenceis not statisticallysignificant in the early half assumethat somelife stagesofthese ectoparasitesover- of the breeding season(xZl = 2.19, P = 0.14; Fig. lb). wintered in the nest material and were present in low numbers,as supportedby earlier studies(Merino 1993). In this year, only one empty nest box was available for DISCUSSION the birds between two nest boxes with old nests. Our data demonstrate that Pied Flycatchers prefer to Immediately after the arrival of males from spring nest in empty nest boxes. Although studies with this migration we checked nest boxes every other day to passerine species in northern Europe found the op- record any change in disposition of the nest material posite (Ore11et al. 1993, Mappes et al. 1994) in none and the progressof nest building. We considered as of them was there strongpressure by nest ectoparasites occupied nests those where laying began. Nest boxes to make adaptive the choice of clean nest boxes. The occupied by tits (Pam spp.) were excluded from all opposite is true in our population, where nest-dwelling computations.To control for between-yeardifferences, mites (Atari), blowflies and fleas have been shown to the laying date was standardizedas the deviation from decreasenestling growth (tarsuslength and mass) and the yearly mean. survival (Merino 1993, Potti and Merino 1994, Merino Statistical tests are two-tailed. The Yates correction and Potti 1995). Choice of a clean cavity for breeding was applied in chi-square tests. is obviously adaptive in these circumstances(e.g., Op- pliger et al. 1994, Richner et al. 1994). RESULTS Alternatively, birds seem able to ascertainwhether In 1993, Pied Flycatchersoccupied more often empty nest boxes, either as roosting cavities (Christe et al. nest boxes that boxes with old nests(Fig. 1a), although 1994, Me&i and Allander 1995) or breeding holes this preference was not statistically significant (t-test (Oppliger et al. 1994), are parasitized or not. This per- comparing an observed proportion versus that ex- ceptual factor may be involved in our study in the high pected by chance, t = 0.73,34 df, P = 0.47). However, rate of rejection of nest boxes with old nests in 1994, many (n = 11, 55%) of the nest boxes where old nests and is also suggestedby the fact that no nest material had been placed were almost completely emptied, and was removed by the birds that year. On the contrary, 830 SHORT COMMUNICATIONS

suspectPied Flycatcher males removed the nest ma- n 1993 terials, althoughwe did not witnessone instance.How- ever, we have strongevidence that male Pied Flycatch- I0 1994 I er regularly remove nest material as, contrary to fe- males for which we lack observations, we have seen 0.8 males removing nest materials in other years in nest I boxes owned by other Pied Flycatcher pairs who did not complete nest-buildingthere, and also in nestboxes disputed to Blue Tits (Parus caeruleus).Rejection of s 0.6- nest boxes with old nests in the second half of the seasonmay be due to a stronger selection on later ar- 9 20 rived females to chooseclean available options, as oc- ii cupation of dirty nest boxes prolongs the time spent s building and may therefore delay start of egg laying .e 0.4- (Mappes et al. 1994). Time constraintsare important : in the breeding biology of Pied Flycatchers due to a e P lower probability of recruitment with the advancing season(Potti and Montalvo 1991, Lundberg and Ala- 11 0.2 - tale 1992). The hazards of late breeding on old nests due to the increaseof some nest ectoparasitepopula- tions with the advancingseason (e.g., blowflies in some years, Merino and Potti 1995) may be another factor selectingfor choice of nest site. O- li If given a choice Pied Flycatchersexposed to repro- empty nest box old nest ductive costsdue to nest ectoparasitesin southernEu- treatment rope prefer to breed in unparasitized nest boxes. Fur- thermore, Pied Flycatchers are capable of readily re- move old nest material, at least when this is not par- asitized. Moller’s (1989) warnings urging ecologiststo take into accountnest ectoparasiteswhen studyingnest site selection,reproductive success and nestlinggrowth, among others, are substantiatedby our results. Also, our data lend support the suggestion(Ore11 et al. 1993) 0.8 that latitudinal differencesin the occurrenceof harmful ectoparasitesmay be associatedwith latitudinal dif- 21 ferencesin anti-parasitic behaviors. We wish to thank J. C. Merino for help with putting up experimental nests, to L. M. Carrascalfor being as helpful as ever, and to Juha Merila for his constructive criticisms on an earlier draft. Our work was supported by the Spanish CICYT (project PB91-0084-C03-03). LITERATURE CITED Crnusra, P., A. OPPLIGER, AND H. RICHNER. 1994. Ectoparasitesaffects choice and use of roost sites in the Great Tit, Parus major. Anim. Behav. 47: 895-898. CUP, S., AND C. M. Paaams. 1993. Handbook of the birds of the Western Palearctic. vol. VII. Ox- ford Univ. Press. Oxford. ’ 0 DAVIS, W. H., P. J. KALISZ,AND R. J. WELLS. 1994. early late EasternBluebirds prefer boxescontaining old nests. J. Field Omithol. 65:250-253. laying date LUNDBERG,A., AND R. V. ALATALO. 1992. The Pied FIGURE 1. Proportion of nest boxes occupied by Flycatcher. T. & D. Poyser, London. Pied Flycatchersin relation to (A) experimental treat- MAPPES,T., J. MAPPES,AND J. KOTIAHO. 1994. Ec- ment in two years;(B) lay date and experimental treat- toparasites,nest site choice and breeding success ment, data from 1993-l 994 combined. in the Pied Flvcatcher. Oecoloaia 98:147-149. Mtnuti, J., AND R. ALLANDER. 1995. Do Great Tits (Parz4.smajor) prefer ectoparasite-freeroost sites?- old nest material was removed in at least 7S”/aof nest an experiment. Ethology 99:53-60. boxes with old nestsin 1993, maybe becausethat year MERINO, S. 1993. Efectos de1 ectoparasitismoen el they were not exposedto the risk of being parasitized Papamoscas Cerrojillo Ficedula hypoleuca. (Thompson and Neil1 1991, Christe et al. 1994). We M.Sc.thesis, Univ. de Alcala de Henares, Madrid. SHORT COMMUNICATIONS 831

MERINO,S., ANDJ. Porr~. 1995. Mites and Blowflies PO-IX, J., AND S. MERINO. 1994. Heritability esti- decreasegrowth and survival in nestling Pied Fly- matesand maternal effectson tarsuslength in Pied catchers.Oikos 73:95-103. Flycatchers, Ficedulu hypoleucu.Oecologia 100: MBLLER, A. P. 1989. Parasites, predators, and nest 33 l-338. boxes: facts and artefacts in nest box studies of POTTI,J., AND S. MONTALVO. 1991. Return rate, age birds. Oikos 56:421423. at first breeding and natal dispersal in a Spanish MOLLER, A. P. 1992. Nest boxes and the scientific population of Pied Flycatchers,Ficedula hypoleu- rigour of experimental studies.Oikos 63:309-3 11. cu. Ardea 79~419-428. OPPLIGER, A., H. RICHNER, AND P. CHRISTE. 1994. R~CHNER,H., A. OPPLIGER,AND P. CHRISTE. 1993. Effectofan ectoparasiteon lay date, nest-sitechoice, Effect of an ectoparasiteon reproduction in Great desertion, and hatching successin the Great Tit Tits. J. Anim. Ecol. 62:703-710. (Purus major). Behav. Ecol. 5: 130-l 34. THOMPSON,C. F., ANLIA. J. NELL. 1991. HouseWrens ORELL, M., S. RYTK~NEN, AND K. Iu)M_&KI. 1993. Do do not prefer clean nestboxes.Anim. Behav. 42: Pied Flycatchers prefer nest boxes with old nest 1022-1024. material? Ann. Zool. Fenn. 30:3 13-3 16.

The Condor 97~8312334 0 The CooperOrnithological society 1995

MOBBING OF EASTERN SCREECH-OWLS: PREDATORY CUES, RISK TO MOBBERS AND DEGREE OF THREAT’

FREDERICKR. GEHLBACHAND JILL S. LEVERIT? Department of Biology, Baylor University,Waco, TX 76798

Key words: Cultural tutoring;Eastern Screech-Owl; owl nest and roost sites (McPherson and Brown 1981, kin selection;mobbing behavior;Otus asio; predatory Chandler and Rose 1988, Gehlbach 1994). Mobbing cues. is often seennear active nests,where the owls do most monotonic singingand hunting, so it may be keyed to Avian mobbing of raptorial birds is a common re- site or song. Despite suchfocus, fledgling screech-owls sponseto potential threat and may warn prey and teach in the nest area are seldom mobbed and then only naive individuals about the danger or deter predation mildly, presumably because they do not kill birds (Altmann 1956; Vieth et al. 1980; Frankenberg 1981; (Gehlbach 1994). Curio et al. 1978, 1983). Birds are important prey of If the mobbers of Eastern Screech-Owls correctly Eastern Screech-Owls(Otus asio) in the spring nesting assessrisk as some European birds do (Curio et al. season(Van Camp and Henny 1975, Turner and Dim- 1983), males and permanent residentsshould respond mick 1981, Gehlbach 1994); and frequent mobbers of to the owls’ monotonic trills more often, longer, or the owls are those songbirdsmost often eaten, includ- more intensely than its descendingtrills, especiallyin ing permanent residents and males (Gehlbach 1994). the springnear nests.Also, they should mob adult owls Studies of screech-owlssuggest that mobbing is most more than fledglings.These postulateshave not been intense in the spring-earlysummer nestingperiod (Alt- tested, and some experimental variables have been mann 1956;McPherson and Brown 198l:Shedd 1982, confounded in the past. Earlier investigators, for ex- 1983: Chandler and Rose 1988: Gehlbach 1994). ample, induced mobbing with tapesofcombined songs Eastern Screech-Owls sing two seasonally distinct (McPherson and Brown 1981) or did not mention the songs (Hough 1960, Ritchison et al. 1988). Their song(s) they used (Shedd 1982, 1983; Chandler and monotonic trill is a nest-siteadvertisement and family- Rose 1988). contact song, primarily while nesting in spring-early Becausewe conducted mobbing experiments in an summer, whereas the descending trill is a territorial area where Eastern Screech-Owlsand their avian prey defense signal largely in late summer-fall (Gehlbach had been studied (Gehlbach 1994), we devised very 1994). Mobbing songbirds orient to these songsand specifictests. We wanted to know if mobbers correctly assessrisks of predation by recognizingcertain visual (plumage, nest-site area) and auditory (song)cues with respect to the seasonalvariation in these cues. Previ- I Received 9 January1995. Accepted 21 March 1995. ously, this has not been possible with any raptor, nor 2 Present address:742 NE 15th St., Oklahoma City, has mobbing been tested as regardsthe relative jeop- OK 73104. ardy or reproductive investment of specific mobbers