ACTA ORNITHOLOGICA Vol. 38 (2003) No. 1 Clutch size ad breedig performace of Marsh Tits Parus palustris i relatio to hole size i a primeval forest Tomasz W e s o ł o w s k i D epartm et of Avia Ecology, W rocław Uiversity, Siekiewicza 21, 50-335 Wrocław, POLAND, e-mail: tomwes@ biol.ui.wroc.pl W esołowski T. 2003. Clutch size ad breedig performace of Marsh Tits Parus palustris i relatio to hole size i a primeval forest. Acta O m ithol. 38: 65-72. Abstract. I studied M arsh Tit, a secodary hole estig bird i a area with superabudat holes (primeval forest i the Białowieża N atioal Park, E Polad), expectig to fid o relatioship betw ee hole size ad, either clutch size, or breedig perform ace. Aalysis of about 350 est histories collected over 13 years revealed, as expected, o differece betw ee breedig i m edium -sized or large holes, but birds usig small holes (lowest 25th percetile) laid sm aller clutches, fledged m argially fewer youg ad lost m ore broods tha the birds usig the two larger size classes. This variatio was ot due to differeces i tim ig of layig or the age of females across the hole classes. It is proposed that the use of small holes persisted because of relatively low fitess costs of makig such a sub-optim al choice. Key words: M arsh Tit, Parus palustris, Białowieża Natioal Park, est success, tree holes Received March 2003, accepted May 2003 INTRODUCTION Secodary hole esters, species depedet o the existig supply of tree holes for estig, may be forced to accept est sites far from optimal, if these are i short supply. I m ay m aaged forests hole availability seems to be limitig (reviews i Newto 1995, 1998), ad it is usually assumed that it has always bee so, that hole estig birds have ivariably faced chroic hole shortage ad fierce competitio for holes (reviews i Newto 1995, 1998). Though ot corroborated by data from prim eval coditios (see below), this covictio led to a expectatio that clutch size of hole esters would ofte be costraied by hole size (bottom area). Thus, birds breedig i larger holes would lay more eggs ad rear more youg tha the oes forced to use smaller, less tha adequate holes. However, attempts to dem ostrate a positive relatioship betwee hole/est box size ad clutch size have give equivocal results. Sometimes the predicted relatioship was foud, sometimes ot (review i Wiebe & Swift 2001). These discrepacies could be partially attributable to desig problems (e.g. size of est boxes used outside the rage of atural holes used) but they also could reflect real differeces across the study sites (Fig. 1). U) o> c u> "O a> Z Hole size Fig. 1. A hypothetical relatioship betwee hole size (cross-sectio) ad breedig productivity i a hole estig bird: grey area holes too small to accom m odate a bird; 1 holes useable, clutch size ad productivity icrease with hole size; 2 optim um plateau, variatio i either clutch size, or productivity related to hole size; 3 holes too large, productivity, but ot clutch size, declies with hole size.
66 T. W esołowski O e could expect a optim um rage of hole sizes, w ithi w hich the breedig productivity w ould be highest (Fig. 1). Birds usig too small or too large holes w ould produce fewer youg, though for differet reasos. There is obviously som e m iim um size of hole below w hich birds caot physically squeeze ito. However, above this threshold, there should be a positive correla tio betw ee hole size ad productivity over the sm aller tha optim um p art of the size rage (Fig. 1, sectio 1), o liear relatioship w ithi the opti m um rage (sectio 2), ad a egative relatio ship w he holes becom e too large (sectio 3). There are several biological reasos behid these expectatios. The disadvatages of usig too arro w holes could be m aifold: brood hyper therm ia, high am m oia cocetratio, tram plig of youg, icreased disease trasm issio, iad equate est saitatio or icreased predatio risk (Lohrl 1973, Slagsvold 1989, Erbeldig-D ek & Trillmich 1990, Wiebe & Swift 2001, Wesołowski 2002). Breedig i too large holes, o the other had, could be costly due to problem s w ith heat loss d u ri g icubatio/broodig ad due to a icreased risk of hole loss to stroger com peti tors (Lóhrl 1973, L udescher 1973). W he forced to accept sm all holes, fem ales could adjust their broods to available space by layig fewer eggs ad avoid overcrow dig problem s. I too large holes this possibility does ot exist, layig fewer eggs could m ake situatio eve worse. As the relatioship betw ee hole size ad bird p roductivity could be either positive or egative (Fig. 1), the result of a field study w ould critically d eped o array of hole sizes available w ithi the study area, thus, differet studies o the sam e species m ay produce apparetly cotradictory results. H ere the relatioship betw ee the size of holes used, clutch size ad estig success of M arsh Tits breedig i Białowieża N atioal Park (E Polad) is exam ied. Holes i this old-grow th prim eval forest are sup erab u d a t ad secod ary hole-esters m ay choose from a large variety of holes (W alakiewicz 1991, W esołowski 1996, 2001a, 2002). M arsh Tits do select holes before settlig i cavity the females ofte try tw o or three holes i a row. Additioally, M arsh Tits are ot oly passive acceptors, they ca both elarge (by rem oval of m ould ad debris) ad reduce (by ad d i g m ore est m aterial) iteriors of their holes (Wesołowski 1996, 1998, 1999). W ith the ability to choose, ad behavioural m eas to adjust their hole dim esios, all females should be able to procure holes of adequate size. Therefore, I expected to fid o relatioship betw ee the hole size ad, either clutch size, or breedig perform ace of M arsh Tits estig i these prim eval coditios. I accordace w ith this expectatio, the ectopara site ifestatio rates of M arsh Tit broods w ere idepedet of hole size (W esołowski 2001a). METHODS S tudy area ad data collectio The Białowieża Forest complex is situated at the Polish-Belarussia border (52 41'N, 23 52'E). The forest represets a rem at of the vast low lad forests that oce covered large parts of tem perate Europe. Its preset uique features result from its cosiderable size ad a excep tioally good state of coservatio (Faliński 1986, Tomiałojć et al. 1984, Tomiałojć & W esołowski 1990, W esołowski & Tomiałojć 1995). A lthough traces of h u m a presece are kow from the N eolithic period, itese tim ber-cuttig did ot start before the 20th cetury. The m ajority of tree stads i the Polish p art are ow u d e r m aage m et, b u t a 47.5 km 2 block of the best preserved prim eval stads has bee strictly protected w ithi the Białowieża N atioal Park (BNP). The prim eval stads i the strictly protected p art of the Park are distiguished by a w hole array of features: they are multi-storey, m ixedspecies, ueve-aged, com posed of trees reachig uusual heights ad cotai a large am out of dead tim ber ad uprooted trees. D etailed descrip tios ad photographs are give i Tomiałojć et al. (1984), Faliński (1986), Tomiałojć & Wesołowski (1990, 1994), so oly a brief review of the study areas is give below. D ata o M arsh Tit estig a d dim esios of their holes were collected i 1987-1989 ad i 1992-2001. Itese searches for ests, durig w hich holes of all (except few) breedig pairs w ere foud, w ere m ade i four deciduous areas. These were: plot K, correspodig to plot K i Tomiałojć et al. (1984) 33 ha, m ostly riverie forest (Ash Fraxius excelsior, A lder Alus glutiosa, Spruce Picea abies), at forest edge; plot C, correspodig to plots CW ad CE i Tomiałojć et al. (1984) 48 ha, forest iterior, dom iated by Lime Tilia cordata, H orbeam Carpius betulus ad Spruce (formig respectively 46%, 24% ad 23% of stads Wesołowski 1996); plot M, correspodig to plots MS ad MN i Tomiałojć et al. (1984) 54 ha, forest iterior,
Hole p aram eters ad clutch size of M arsh Tit w ith H orbeam, Lime, ad Spruce as dom iat trees (form ig respectively 47%, 27% ad 13% of stads W esołowski 1996). plot W, correspodig to plots WE ad WI i Tomiałojć et al. (1984) 50.1 ha, at forest edge, H orbeam, Lime ad Spruce beig m ost u m er ous (respectively 34%, 28% ad 18% of stads Wesołowski 1996). The latter three areas w ere all situated i the oak-hom beam habitat. The plots w ere 1-2 km apart. The m axim um distace betw ee them am outed to 6 km. To docum et breedig ad the fate of ests, the holes were checked regularly (m ostly from the groud). Ispectios of the est cotets (usig a small bulb o a bedable w ire ad a sm all m irror) allow ed to estim ate fledgig dates. A roud this time, the holes w ere observed from a distace c. every 24 hrs, up to the day o w hich o parets were observed brigig food to the hole. If o the previous day youg w ere at least 18 days old (the yougest age of fledgig of u d istu rb ed broods Wesołowski 2000) ad there was o sigs of attem pted depredatio, the est was cosidered to be successful. If o feedig was observed at a hole cotaiig youg about to fledge (16-17 days old), w e searched for paret birds (m ost of them colour-riged) to check w hether they w ere collect ig food for prem aturely-fledged youg. If they did, the est was classified as successful. All other cases of p rem ature cessatio of paretal activity (o sigs of paret presece d urig a h o u r obser vatio sessio) were treated as est failure. U riged adults carryig food for estligs w ere caught i door-traps placed at the hole etrace, ad colour-riged. M ost of birds (up to 90% of birds i som e years) w ere riged by the ed of the seaso. H ole m easurem ets, w ere take d u ri g the estlig period, b u t som etim es after the youg had fledged. To m easure holes, a collapsible ruler ad a sm all bulb o a bedable wire, w ere used. Several m easurem ets w ere take at each hole (W esołowski 1996, 2002) b u t the oe relevat here is iterior diam eter of the cavity. It w as m easured at the est rim level. Two values least ad great est iterior diam eter w ere alw ays recorded. Takig accurate m easurem ets of hole diam eter i the pear-shaped holes (W esołowski 1996) was quite difficult, at tim es im possible. A alysis Cross-sectios of alm ost 84% M arsh Tit holes w ere approxim ately circular, ad m ost others were 67 regular, elogated (Wesołowski 1996). Hece, to calculate area of hole cross-sectio at the est level, the equatio for the area of a ellipse was used. Sam ple sizes differ am og idividual aaly ses, as gatherig of a full set of m easurem ets was ot possible i every case. Some holes w ere used m ore tha oce, i differet years of the study. This could lead to o-idepedece of m easure m ets. However, this problem was m ore ap p ar et tha real, as m ost holes w ere i livig trees (W esołowski 1996), their characteristics chaged w ith time; the "sam e" hole m easured i vari ous years could have quite differet dim esios. Moreover, if the sam e place was used, it was rarely used by the sam e birds (W esołowski 2002). First egg dates w ere calculated firstly as abso lute (caledar) dates expressed as a um ber of days passig from the begiig of April, w ith 1 A pril = day 1, 2 A pril = day 2 ad so o, ad sec odly as relative dates. The latter w ere expressed as deviatios of idividual layig dates (um ber of days) from m edia absolute layig dates (= day 0) of respective years. The relative dates were calculated oly for seasos w ith larger sam ples of ests; years 1979, 1984, 1986 ad 1991 w ith oly 1-2 dates w ere excluded. H atchig ad fledgig dates were usually established by direct observatios. The first egg dates, however, were rarely observed directly i the field because i spite of efforts to fid ests at buildig stage, some ests were foud oly at later stages. Moreover, eve i ests foud early eough, it was quite difficult to ascertai w hether layig already had commeced; est cups aroud the layig period were filled w ith wool, ad m ay have bee hidde. Therefore it was ecessary to back-calculate the first egg dates. It was assum ed that oe egg was laid per day, that icubatio lasts 15 days, ad that estlig age correspods roughly to stages give i Wikel (1970) for Blue Tits Parus caeruleus. W he clutch size ad/or estlig age were ukow (e.g. holes cotrolled oly from a distace or holes cotaiig oly a sigle estlig) 40 days were subtracted from the fledgig date. This correspoded to six days allowed for egg-lay ig (7 eggs clutch assum ed), 15 days for icubatio ad 19 for estlig stage (Wesołowski 2000). Idividual m arkig allow ed to observe behav iour of idividual birds i cosecutive seasos ad to com pare perform ace of first breeders to older birds. After their first breedig M arsh Tits are extrem ely sedetary, stayig as a rule i the sam e rage (T. W esołowski, upublished data). Thus, uriged birds settlig i areas previously
68 T. Wesołowski used by riged oes are mostly probably youg birds breedig for the first time. Hece, except i the iitial rigig seasos, it was justified to treat uriged birds were youg, first-time breeders. O lder birds which had escaped capture i previous years were treated as youg. So the youg category may cotai some adm ixture of older birds. The old category, o the other had, is ot biased that way, cotaiig oly colour-riged birds breedig for at least their secod time. All statistical procedures follow the formulae give i STATISTICA for W idows (Aoymous 1996), w ith tw o tailed probability values. 11 10 9 8 7 6 5 4 3 2 1 0 50 100 150 200 Hole size (cm2) 250 300 RESULTS The size of M arsh Tits holes i BNP varied from 19 to 314 cm2 w ith m edia value of 63.6 cm2 (mea = 74.6, = 415). Overall clutch size ad um ber of hatched eggs icreased slightly but sigificatly w ith hole size (Fig. 2 ad 3), whereas the um ber of fledgligs was ot related to it (Fig. 4). However, w he calculated separately for differet parts of the hole size rage (Table 1), values of all variables icreased sigificatly with hole size i the lower (rs = 0.2-0.3), but ot i the upper part of the rage (rs = -0.01-0.06, Table 1). Holes size differed sigificatly across years (Fig. 5; Kruskal-Wallis ANOVA H 9 386 = 24.27, p = 0.0039). This was related to the timig of breedig; the m edia hole size icreased with the m edia layig dates rs = 0.87, p = 0.001, = 10 (Wesołowski 1998, 2001b). Two groups of years were distiguishable: those years w ith the m edia Fig. 3. N um ber of eggs hatchig i relatio to hole size (rs = 0.20, p = 0.0003, = 324). Dot size frequecy of idividual values, from 1 (smallest dots) to > 10 (largest dots). 11 10» 9 8 7 # 03 6 O) 03 5 m Z 4 3 m 2 1 0 0 50 100 150 200 250 300 Hole size (cm2) Fig. 4. N um ber of youg fledgig i relatio to hole size (rs = 0.01, p = 0.85, = 224). Dot size reflects frequecy of idividual values, from 1 (smallest dots) to > 8 (largest dots). 11 10 9 8 250 200 150 7 6 T, 100 5 4 50 I I T T I 3 0 50 100 150 200 Hole size (cm2) 250 300 1993 '94 '95 '96 '97 Year '98 '99 '00 '01 Fig. 2. Clutch size i relatio to hole size (rs = 0.18, p = 0.0005, Fig. 5. Sizes of M arsh Tit holes i differet years (medias, = 366). Dot size frequecy of idividual values, from 1 25-75% values, ad the rages of values). Oly seasos with (smallest dots) to > 18 (largest dots). >10 holes are show.
69 H ole p aram eters ad clutch size of M arsh Tit 1998, 2000), this could cofoud the relatioship. However, the egg layig dates w ere ot related to hole size, the differece i the relative layig dates am og hole size classes beig less tha a day (Kruskall-Wallis ANOVA (H 2413 = 4.02, p = 0.13). Similarly, there was o cosistet differ ece i size of holes used by youger ad older females (Table 3). Thus, the effect of hole size o clutch size seem s geuie. H atchig success i successful ests did ot deped o the hole size ad the type of seaso, i every category all eggs hatched i 46-59% of cases (x2 = 0.15-0.74, df=2, p = 0.69-0.93). Equally, the fledgig success show ed o relatioship, all youg fledged from 67-85% of ests i idividual categories (x2 = 0.50-2.43, df=2, p = 0.30-0.77). Overall, the um ber of youg fledgig from small holes was oly m argially less (by 0.5 youg) tha fledgig from the m edium sized holes (Table 4). This differece could be totally explaied by the sm aller clutches laid i small holes (Table 2). Hole size, however, w as related to the total failure rate. Broods i sm all holes had about 50% higher chace of total failure tha ests i the m edium or large holes. This tedecy was detectable i both types of seasos, b u t the dif ferece w as sigificat oly i the pooled sam ple (Table 5). D ue to this ad their low er clutch size, birds breedig i sm all holes produced 20-25% hole size w ithi 62.8-63.6 cm2, ad those w ith hole size exceedig 75 cm2 (Fig. 5). To remove a possible cofoudig ifluece of this iter-year variatio, years were grouped ito two classes: 1) small" sea sos m edia hole size < 75 cm2 ad, 2) large" seasos m edia hole size > 75 cm2. All aalyses were ru both separately for each year class + the pooled data from all years. To check w hether breedig i very large or very small holes had ay effect o reproductio the data w ithi each group w ere divided ito three classes: 1) small (25th per cetile), 2) m edium 25th-7 5 th percetile, ad 3) large holes 75th percetile. I the large seasos these correspoded to hole size < 63.6, > 63.5 ad < 95.0, ad > 95.0 cm2, respectively. I the small seasos the border values am outed to < 50.0, > 50 ad < 79.0, ad > 79.0 cm2. All holes of idetical size w ere p u t ito a sigle category, this som etimes resulted i slightly sm aller or larger sam ples i idividual categories. The respective classes of holes from both types of seasos were the pooled to m ake totals for each class. The stadardised clutch size of M arsh Tit w as sigificatly smaller, by 0.7 eggs, i small holes th a i m edium or large oes, this differ ece beig less proouced i the "large" sea sos (Table 2). As youg M arsh Tit females lay sm aller clutches tha older oes ad their clutch size strogly declies w ith seaso (Wesołowski Table 1. C orrelatios (Spearm a's rak) betw ee hole size ad M arsh Tit reproductio. Variable Clutch size ( eggs) hatchig eggs fledgligs 199 168 120 < media r 0.19 0.30 0.22 Hole cross-sectio > media r P 0.006 > 0.001 0.015 167 156 104 0.04 0.06-0.01 P 0.612 0.491 0.907 366 324 224 All groups r 0.18 0.20 0.01 P <0.001 <0.001 0.853 Table 2. S tadardised M arsh Tit clutch size i relatio to hole size ad type of seaso (see the text for defiitios). Hole size classes: sm all 25th, m edium 25th-7 5 th, large 75th percetiles. Type of seaso Hole size class Small Medium Large Kruskal-Wallis ANOVA Small Mea (SD) -0.5 (1.04) 0.1 (1.0) 0.2 (0.84) H = 20.50, p <0.0001 61 137 58 Large Mea (SD) -0.3(1.40) 0.2 (1.14) 0.2 (0.99) H = 3.59, p = 0.166 25 58 26 All Mea (SD) -0.5(1.16) 0.1 (1.07) 0.2 (0.89) H = 23.82, p <0.0001 86 195 84
70 T. Wesołowski Table 3. Size of holes used by M arsh Tit females of differet age i relatio to type of seaso (see the text for defiitios). Age classes: 1 year of rigig, i m ost cases probably the first breedig seaso, 2 the followig year, > 3 com bied data, 4th-8 th seasos after rigig. Type of seaso Age class Small Large All Kruskal-Wallis ANOVA Mea (SD) Mea (SD) Mea (SD) 1 68.4 (30.98) 113 85.9 (37.78) 49 73.7 (34.04) 162 2 79.7 (54.06) 40 92.3 (39.05) 26 84.7 (48.77) 66 3 81.8 (53.32) 30 99.3 (46.34) 11 86.5 ( 51.58) 41 > 3 74.3 (27.37) 13 69.9 (28.62) 13 72.1 (27.53) 26 H = 1.63, p = 0.65 H = 4.48, p = 0.21 H = 2.64, p = 0.45 Table 4. Frequecy of M arsh Tit total est failure i relatio to hole type ad seaso (see the text for defiitios). Probability values refer to x2 test, 2df. Hole size classes: sm all 25th, m edium 25th-75th, large 75th percetiles. Hole size class Type of seaso Small Large All % % % Small 67 34.3 30 30.0 97 33.0 Medium 146 21.2 70 20.0 216 20.8 Large 61 24.6 32 12.5 93 20.4 Probability 0.12 0.23 0.046 fewer youg tha the oes usig the m edium ad large holes (ca 4.2 vs. 5.4 or 5.3 fledgligs/estig attem pt respectively). DISCUSSION Cotrary to expectatios, both clutch size ad productivity of M arsh Tits i BNP did deped o hole size, but this relatioship was visible oly i the lower part of the hole rage. Birds breedig i large holes did ot show ay sigs of lowered productivity. Overall there was a weak positive relatioship betwee clutch ad hole size (rs = 0.18). Clutch size of other secodary hole esters i BNP showed either o (r = 0.1, Pied Flycatcher Ficedula hypoleuca, Czeszczewik 2001) or a weak positive (r = 0.17, Collared Flycatcher Ficedula albicollis, Walakiewicz 1991) correlatio w ith the hole bottom area. O the other had, clutch size of M arsh Tits i other areas very strogly correlated with hole size: r = 0.95 i a Germ a study (SW Germay, recalculated from Fig. 12 i Ludescher 1973) ad r = 0.74 i a Russia study (Courish Spit, Baltic Coast, Markovets 2001). The strog relatioship foud i Ludescher's study could be due to shortage of adequate est sites, avoidace of strog competitors, ad clutch size limitatio i arrow holes M arsh Tits there used m uch sm aller holes (mea30 = 45 cm2) tha at BNP ad bred ofte i old Willow Tit Parus motaus holes. The Russia results are more difficult to explai. M arsh Tits o the Courish Spit had plety of choice (up to 57 useable holes i a sigle territory), bred i almost as large holes as the birds at BNP (mea hole size 69 cm2), ad laid similar um ber of eggs (6-10, m ea 7.7 ) recalculated from M arkovets & Visotsky (1993), M arkovets Table 5. Stadardised um ber of M arsh Tit fledgligs i relatio to hole size ad type of seaso (see the text for defiitios). Hole size classes: small 25th, m edium 25th-75th, large 75th percetiles. Type of seaso Hole size class Small Large All Mea (SD) Mea (SD) Mea (SD) Small -0.2 (1.34) 37-0.2 (1.14) 13-0.2 (1.28) 50 Medium 0.3 (1.54) 95 0.1 (1.05) 25 0.3 (1.45) 120 Large 0.2 (1.64) 36-0.2 (1.91) 13 0.1 (1.70) 49 Kruskal-Wallis ANOVA H = 4.26, P = 0.119 H = 0.84, p = 0.658 H = 5.07, p < 0.079
H ole param eters ad clutch size of M arsh Tit (2001). D espite these apparet sim ilarities, the relatioship betw ee clutch ad hole size was m uch stroger o the C ourish Spit tha at BNP. Why did birds at BNP choose to breed i too small holes? Were they just makig mistakes? Was hole supply really limited? Or do females kowig their evetual clutch size select the hole size best fittig the pre-selected clutch size? W hatever the mechaisms resposible for settig clutch size i M arsh Tits, the latter explaatio caot be true. Youg or late commecig birds, which were layig smaller clutches (Wesołowski 1998), did ot favour small holes. Additioally, the birds i small holes ot oly laid fewer eggs, but also lost broods more frequetly, so by choosig larger holes they would probably icrease their chace of est success. The idea that M arsh Tits m ay have bee forced to breed i very arrow holes to avoid iterfer ece from stro g er com petitors m ay be also refuted. The birds used relatively spacious holes at BNP (their bottom areas equalled those used by very u m ero u s C ollared Flycatchers there W alakiewicz 1991), despite this overlap, the iterspecific com petitio for holes at BNP was alm ost o-existet (W esołowski 1989, i press, W alakiewicz 1991). T hough M arsh Tit holes were geerally i excess, it is still possible that i som e cases local shortages occurred ad that som e pairs w ith access oly to small holes o their territories w ere forced to accept them. U fortuately it is im possible to address this issue directly, as fidig all possible holes i this prim eval forest has bee im possible. However, as M arsh Tits could elarge the existig holes (Ludescher 1973, W esołowski 1999) such o choice situatios w ere probably rare. The possibility that birds w ere m akig m is takes by choosig sub-optim al estig places should also be cosidered. N atural selectio for avoidig sm all holes m ay be weak, if costs of selectig them were ot very high. This could be the case w ith M arsh Tits at BNP, w here eight of each te birds breedig i large ad seve of te breedig i small holes produced youg. So oly every teth bird choosig sm all hole was really severely puished. Perhaps this has ot bee strog eough differece, to m ake M arsh Tits avoid sm all holes etirely. Similarly, Wiebe & Swift (2001) proposed the low fitess costs as a explaatio for the lack of relatioship betw ee the hole ad clutch size i a prim ary hole ester, N orther Flicker Colaptes auratus. Breedig i sm all holes w as disadvatageous for M arsh Tits at BNP, b u t breedig i large holes 71 had o detectable detrim etal effects. This ot w ithstadig, the birds clearly avoided estig i very large holes. N um erous em pty N uthatch Sitta europaea holes (m ea bottom area 325 cm 2, W esołowski & Rowiński i review) i the study areas w ere alm ost ever used by estig M arsh Tits. However, as the N uthatch holes usually sim ultaeously possessed several features (spa cious bottom s, large etraces, coical kotholes, high above the groud) m akig them uattractive for M arsh Tits (Wesołowski 1996, 2002), it is dif ficult to decide w hether the birds avoided them because of their size or for som e other reaso. ACKNOWLEDGEMENTS I heartily thak M. C zuchra, T. Kliś, B. O rłow ska, ad P. Rowiński for their participatio i the field work. The fiacial aid of the ALA ad Schweizerische Vogelwarte helped to partially cover costs of field w ork i various periods. REFERENCES A oym ous. 1996. Statistica for W idow s. StatSoft, Ic., Tulsa. Czeszczew ik D. 2001. [Breedig ecology of Pied Flycatcher Ficedula hypoleuca i atural ad m aaged w oods], PhD thesis, W roclaw Uiversity. Erbeldig-D ek C., Trillmich F. 1990. N estbox clim ate ad its effects o Starlig (Sturus vulgaris) estligs. J. O rithol. 131: 73-84. Falihski J. B. 1986. Vegetatio dyam ics i tem perate forests (Ecological studies i Białowieża Forest). D r W. Juk Publ., D ordrecht. Lohrl H. 1973. Ifluece of est-box size o clutch size of the G reat Tit (Parus major). J. O rithol. 114: 339-347. L udescher F. B. 1973. [The M arsh Tit (Parus palustris) ad the W illow Tit (Parus motaus) as sym patric siblig-species], J. O rithol. 114: 3-56. M arkovets M. Yu. 2001. [Populatio ecology of M arsh Tit (Parus palustris)]. PhD thesis, Zoological Istitute, Russia Acad. Sci., Sakt-Peterburg. M arkovets M. Yu., Visotsky V. G. 1993. B reedig biology of the m arsh tit (Parus palustris) o the C ourish Spit. Russia J. O rithol. 2: 61-69. N ew to I. 1995. The role of est sites i lim itig the um bers of hole-estig birds: a review. Biol. Coserv. 70: 265-276. N ew to I. 1998. Populatio lim itatio i birds. A cadem ic Press, Lodo. Slagsvold T. 1989. E xperim ets o clutch size ad est size i passerie birds. Oecologia 80: 297-302. Tomiałojć L., W esołow ski T. 1990. Bird com m uities of the p ri m aeval tem perate forest of Białowieża, Polad. I: Keast A. (ed.). B iogeography ad ecology of forest b ird com m u ities. SPB A cadem ic Publishig, H ague, pp. 141-165. Tomiałojć L., W esołow ski T. 1994. Die Stabilitat vo V ogelgem ieschafte i eiem U rw ald der gem assigte Zoe: Ergebisse eier 15 jahrige Studie aus dem
72 T. W esołowski N atioalpark vo Białowieża (Pole). O m ithol. Beob. 91: 73-110. Tomiałojć L., W esołowski T., W alakiewicz W. 1984. Breedig bird com m uity of a prim aeval tem perate forest (Białowieża N atioal Park, Polad). Acta Orithol. 20: 241-310. W alakiewicz W. 1991. Do secodary-cavity estig birds suf fer m ore from com petitio for cavities or from predatio i a prim aeval deciduous forest? Nat. A reas J. 11: 203-212. W esołowski T. 1989. N est-sites of hole-esters i a prim aeval tem perate forest (Białowieża N atioal Park, Polad). Acta O m ithol. 25: 321-351. W esołowski T. 1996. N atural est sites of M arsh Tit (Parus palustris) i a prim aeval forest (Białowieża N atioal Park, Polad). Vogelwarte 38: 235-249. W esołowski T. 1998. T im ig ad sychroisatio of breedig i a M arsh Tit Parus palustris p opulatio from a prim aeval forest. A rdea 86: 89-100. W esołowski T. 1999. M arsh Tits (Parus palustris) are ot exca vators. Ibis 141: 149. W esołowski T. 2000. Time savig m echaism s i the reproduc tio of M arsh Tits Parus palustris. J. O m ithol. 141: 309-318. W esołow ski T. 2001a. H o st-parasite iteractios i a tu ral holes: m arsh tits (Parus palustris) a d blow flies (Protocalliphora falcozi). J. Zool. 255: 495-503. W esołowski T. 2001b. G roud checks a efficiet ad reliable m ethod to m oitor holes' fate. O ris Fe. 78: 193-197. W esołow ski T. 2002. A tip red a to r a d ap tatio s i estig m arsh tits Parus palustris the role of est site security. Ibis 144: 593-601. W esołowski T. (i press). Bird com m uity dyam ics i a prim aeval forest is iterspecific com petitio im portat? O m is H ug. W esołow ski T., Rowiński P. (i press). The breedig behaviour of the N uthatch Sitta europaea i relatio to atural hole attributes i a prim eval forest. W esołow ski T., Tomiałojć L. 1995. O rithologische U tersu ch u g e im U rw ald vo Białow ieża eie Ubersicht. O m ithol. Beob. 92: 111-146. Wiebe K. L., Swift T. L. 2001. Clutch size relative to tree cavity size i N orther Flickers. J. Avia Biology 32: 167-173. W ikel W. 1970. H iw eise z u r Art- ud A ltersbestim m ug vo N estlige hohlebriiteder Vogelarte a had ihrer K órperetw icklug. Vogelwelt 91: 52-59. STRESZCZENIE [W ielkość ziesieia i udatość lęgów sikor ubogich w zależości od rozmiarów d ziup li w lesie pierwotym ] W ściśle chroioej części Białowieskiego Parku Narodowego dziuplaki wtóre dyspoują adm ia rem dziupli, m ogą więc wybierać do gieżdżeia dziuple optym alych rozmiarów. W tych w aru kach oczekiwao, że wielkość ziesieia, liczba wyklutych młodych, liczba podlotów i poziom strat w lęgach ie będą zależeć od wielkości dziupli (po wierzchi da, Fig. 1). To przypuszczeie spraw dzao a daych z poad 350 lęgów zebraych w ciągu 13 lat. Zgodie z oczekiwaiami, w dziu plach dużych i średiej wielkości ie obserwowa o zw iązku m iędzy powierzchią da a którąkol wiek z aalizowaych cech reprodukcji. Natom iast w dziuplach ajmiejszych (dole 25%) wielkość lęgu zwiększała się z powierzchią da (Fig. 2 4). W ajmiejszych dziuplach ptaki składały miej jaj (Tab. 2), w yprow adzały iezaczie miej m łodych (o 0.4 podlota, Tab. 5), a lęgi w ich częściej ulega ły ziszczeiu (Tab. 4). Poieważ ai wiek ptaków (Tab. 3), ai pora składaia jaj ie były zw iązae z wielkością używaej dziupli, miejsze ziesieia w małych dziuplach były praw dopodobie wyikiem dopasow yw aia przez ptaki liczby składaych jaj do wielkości dziupli. Przypuszczalie gieżdżeie w małych, wąskich dziuplach mogło się utrzym y wać w tej populacji ze w zględu a stosukowo iski koszt dostosowaia, stosukowo iewielką obiżkę dostosowaia ptaków dokoujących suboptymalego wyboru.