Diets of adult and nestling starlings (Sturnus vulgaris) in Hawke s Bay, New Zealand

Transcription

1 New Zealand Jurnal f Zlgy ISSN: (Print) (Online) Jurnal hmepage: Diets f adult and nestling starlings (Sturnus vulgaris) in Hawke s Bay, New Zealand Abdul Meed T cite this article: Abdul Meed (198) Diets f adult and nestling starlings (Sturnus vulgaris) in Hawke s Bay, New Zealand, New Zealand Jurnal f Zlgy, 7:2, , DOI: 1.18/ T link t this article: Published nline: 12 Jan 212. Submit yur article t this jurnal Article views: 154 View related articles Citing articles: 4 View citing articles Full Terms & Cnditins f access and use can be fund at

2 New Zealand Jurnal f Zlgy, 198, Vl. 7, Diets f adult and nestling starlings (Sturnus vulgaris) in Hawke's Bay, New Zealand clgy ABDUL MOD Divisin, Department f Scientific and Industrial Research, Private' Bag, Lwer Hutt, New Zealand Giards were examined frm 334 adult and 62 nestling starlings cllected in mixed farmland during The birds ate insects, spiders, earthwrms, snails, millipedes, centipedes, seeds, and fruits. Starlings ate fewer subterranean animals (7%) than thse usually living partially hidden (45%), n the grund (31%), r n vegetatin (18%). Abut half the invertebrates eaten were 2-5 mm lng and abut a quarter 6-1mm lng. Nestlings tend t be fed significantly larger items than were eaten by adults. The cmmnest items in adult starlings were Cleptera adults, Lepidptera larvae, Hemiptera, and fruits; in nestlings, Cleptera adults and Diptera adults and larvae were imprtant. arthwrms were fund in all the nestlings. Mre than 5% f adult giards cntained earthwrm chaetae in wetter mnths, but fewer in drier mnths. The diet f starlings, despite cnsiderable verlap with the fds eaten by mynas, rks, and magpies, included smewhat different cmpnents and prprtins f the fd supply. Fruits were fund in adults nly; mst were prbably eaten after harvest. Althugh predatin n tw majr pasture pests-cstelytra ealandica and Wiseana cervinata larvae-was insubstantial, nearly 4% f the ttal invertebrates eaten by adult starlings were insect pest species. Keywrds: Sturnus vulgaris; adult diet; nestling diet: fd availability; Hawke's Bay; invertebrate pests. INTRODUCTION The fd and feeding habits f starlings (Sturnus vulgaris) have been studied verseas by many wrkers. In New Zealand, apart frm brief investigatins by Caithness (1968) at Gisbrne and Wellingtn airprts, the diet f starlings has been studied nly in Canterbury (Lbb & Wd 1971, Meed 1976, Cleman 1977). Starlings frm large flcks n pastures, where they capture bth surface-dwelling and subterranean invertebrates. Since these feeding habits are f cnsiderable interest in the bilgical cntrl f insect pests f pastures, a study was made near Hastings, Hawke's Bay, t determine the diet f starlings with regard t the seasnal availability f fds. STUDY ARA The study area (Fig. 1) is a part f the Heretaunga Plains used etensively fr graing, crpping, market gardening, and fruitgrwing. The sil is a Recent alluvium depsited mainly by the Ngarurr, Tutaekuri, and Tukituki rivers. Rainfall and air temperature data fr :Iavelck Nrth ver the 12-mnth study perid are given in Fig. 2. // / Turame Ngarurr River Wellwds nglan 17S 36 Middle Rad HAWK BAY Fig. 1. The study area, shwing where starlings were cllected (e) and where invertebrates were sampled (.). km 5 Received 5 Nvember 1979 Published nline 12 Jan 212

3 248 N.Z. Jurnal f Zlgy, 198, Vl. 7 MTHODS STARLING SAMPLS Adult starlings (n = 334) were sht (frm July 1971 t June 1972) at varius times f day, buit mainly in the evening at pre-rsts and night-rsts at Tukituki, Middle Rad, Turame, and Wellwds (Fig. 1). The nestlings (n = 62) were cllected in Nvember-December 1971 frm nest bxes and frm natural nest sites at Lnglands. The invertebrates in the giards were examined as described earlier (Meed 1976), and the presence f fruits and seeds was recrded. INVRTBRAT SAMPLS Pasture invertebrates were sampled in the Lnglands area (Fig. 1); 36 spade-square samples f tpsil ttalling 1 m' and abut 75 mm deep were taken each mnth. ach sample was placed in a plastic bag and examined later in the labratry by first crumbling the sil, then washing it thrugh a set f sieves f 4., 2., and 1.3 mm mesh. The animals retained n the sieves were cllected, and thse remaining in the plant material in the largest-mesh sieve were extracted in Tullgren funnels. In additin t the sil samples, 3 pit traps (Meed 1976) were placed at grund level. The inner cllectr tube in these traps was clsed at the bttm, and Galt's slutin (see Barber 193) was used as a preservative. The traps wer e emptied mnthly. T shw the preference f adult starlings fr certain fds, in relatin t the abundance f the fds, I used Ivlev's index f preference {Cck 1978), (S/s-P/p)/(S/s+P/p), where Sand P are the numbers f each fd item fund mnthly in the starlings and pasture samples respetively, and sand p are averages f mnthly values fr the year. T determine the differences in prey eaten by the adult and nestling starlings, anther Ivlev preference rati was calculated by the frmula (N-A) I(N+A), where N is the average number f each fd item per nestling and A is the average number per adult in Nvember-December. eaten, the frmer in spring and summer (Fig. 3A) and the latter in all mnths except August. Many ther beetles were less frequently recrded (Table 1). Adult Listrderes delaiguei and Hyperdes bnariensis weevils were eaten in all mnths, but mstly in winter and spring (Fig. 3B,C), althugh mre appeared t be available in late spring and summer. Larvae f several species were als eaten (Table 1). The pit traps revealed that Cleptera wer e active n the surface thrughut the year. Cstelytra ealandica adults emerging in spring and early summer increased the numbers f trapped beetles slightly. Other species wer e mst abundant in winter and spring, especially Haplaner sp. and Aphdius sp., althugh Agrypnus variabilis, Listrderes delaiguei, Hyperdes bnariensis, and Graphgnathus /euclma were als imprtant. The ccurrence f Hap/aner sp. and Aphdius sp. in sil samples matched their OCCUffence in starlings (Fig. 3D,); they were eaten mre in winter and spring, when they were mre abundant, than frm Nvember nward. Mst Cleptera larvae, except Cste/ytra ealandica and Listrderes de{laiguei, were distributed incnsistently in the sil samples. C. ealandica and L. de/aiguei were at densities f 1-171m' and m' respectively in July-August and April June. The frmer was recrded in starlings in July 3 J Ftttt-f Tnll "-, «RSULTS The adult starlings cntained insects, spiders, earthwrms, snails, millipedes, centipedes, seeds, and fruits (Table 1). The incidence f Lepidptera, Hemiptera, Diptera, Dermaptera, Orthptera, and Hymenptera in the adults is apparent frm Table 1. Data n ther fd grups are enlarged n belw. COLOPTRA Clepteran adults were the mst cmmnly recrded insects. Beetles frequently eaten were Haplaner sp., Aphdius sp., Archecrypticus tpali, Hypharpax austmlis, and Aphdius granarius. Cstelytra ealandica and Agrypnus variabilis were als -1til i I I I I I I I 2 l.s 1 ' "r cr: J A 5 and 1971 Ttal rainfall 127 mm F M A M 1972 Fig. 2. Air temperature and rainfall, Havelck pitted abve and belw mnthly means. Nrth, July 1971 t June Range, mean maximum, and mean minimum temperatures

4 1 " S /"'- x\6\\ I x 6 f'j',,,, :-.,,,, ". A \ "'-6... " 1 'OJ Ii; S. Z -1 '" /- - \ / A _..." C / "_... / X\ : x / x/?--x--x,- y I I iii I iii +1 I \ JJ _1, "%", 'fi" _... Meed: Diet f Starlings l3. g 'a. SO g 1S..._"- " X <,- " x x/ \ / x,...-x-x-x O'i iii \'='1' i t I I i I! B,1 \/\ "'-_- '\ X\ x +1 -g,...,, "... '/ (J) Q.i _1 / ' A-A x 4 l3. "4 g iii 15. 'OJ j\, 2 g a;2 x x -"- -. Ii;. / I I i I I >< +1 " ", u " a; (J) -1 2, 4-R D j \,. " ylo -,-y y Y ( I JO <n. 6 'a. M , /X\ " x 2 '- / jlo1'/.1 ".!:... '" 4 :_- 4L N 2 8. O "- " JJ_ 1 6_/ "- / " 'OJ - _ F x\ //'\ """' ' ls<",ls 6 \ / \ """' x x / 'x_x...x / 'x Iii i Iii i -- iii.1 J A SON J F M A M J 1972, c _ \ ' \ /'.g \ \ ) }, (J) " _1 3 l3. 2:;;. M 1 8 " 4 'OJ 2 +' 6 _ I i I I I I 'r '( I 1 I JASONDJ 1971 F M A M J / \ G t4 /x X\\ all 2; x B / x_x, --/ '"'' yy1 6-6 _ - i \/\ ] J (J) /'_- w _1 -- Fig. 3. Numbers f (A) Cstelytra ealajldica. (B) Listrderes delaiguei. (C) Hyperdes bnariensis. (D) Hap/aner sp" () Aphdius sp., (F) Lycsidae, and (G) earthwrms in pasture (--) and in starling giards (x--x adults, nestlings). The preference rati varies frm -1 when nne f a fd was taken, thugh available, t +1 when it was taken but nt fund in the pasture (see text).

5 25 N.Z. Jurnal f Zlgy, 198, Vl. 7 nly, and the latter in July-Octber and April (Table 1). ARACHNIDA All arachnids in the diet were Lycsidae, except fr ne specimen f Ixdidae in September. They were eaten in all mnths, but mstly in July. Lycsids were trapped mre in Octber-April than in May September, but were nt eaten mre ften when mre abundant (Fig. 3F). LUMBRICIDA Mst earthwrm remains (ther than chaetae) were fund in July-Nvember, but wrms were als eaten in March and June. At the sampling site, Alllbphra spp. were present thrughut the year in the tp 75 mm f sil. They were numerus in winter and early spring but scarce in summer (Fig. 3G); their cntinued presence at lw density in the tpsil in summer was due t irrigatin. Despite the pr,esence f earthwrms at the sampling site in summer and autumn, their remains, ther than chaetae, were nt recrded in starlings during this perid except fr March. Since the starlings were sht mstly at pre-rsts r night-rsts, mst earthwrm remains wuld have passed thrugh the giard. Hwever, earthwrm chaetae remained lnger in the giards, and were present in all mnths (Table I). Mre than 5% f giards cntained chaetae in July-September and May-June, when earthwrms were numerus in the tpsil. OTHR INVRTBRATS Snails (Valnia exce'ntrica, Helicella separta, Ptampyrgus antipdarum) were eaten infrequently thrughut the year. In additin, a few unidentified slugs wer,e eaten in April. Millipedes (Diplpda) were eaten in all mnths except August and January February, and rather mre ften in July, December, and April-May. Fewer centipedes (Chilpda) than millipedes were eaten, and fragments were fund in July-August, Octber-Nvember, and May. Millipedes and centipedes ccurred in the sil samples irregularly, and nt nly in the perids when they were eaten. Wdlice (Prcelli scaber) were eaten in July. PLANT AND OTHR MATRIAL Fruits and/r seeds f asparagus (Asparagus fjicinalis), apple (Malus dmestica), grape (Vitis vinijera), pea (Pisum sativum), pear (Pyrus cmmunis), nightshade (Slanum nigrum), and tmat (Lycpersicum esculentum) were eaten (Table 1). Mst f these fd plants grew in hrticultural areas and rchards. Grass and clver leaves were the mst cnsistently eaten items, being present in giards in all mnths except March-May. They were prbably "I...,cn, - l...jig J,;.l --.::: 6:::i VJ _ :;, - < g l -rfj. "'11-."" r- c.. '!. 1<5:': "5 '" " - ;;- II ""i --::: ;>,[J'},j " :r. : J5 ] J5 5-4 _:: c "':;1'] i t.' "= ' ci1:. r.cr. 6!j 5:l: Jj 8ii_ 5 I' N c.. N;;::J' : 12 :c I -- i: OL"'rfJ. ;:,t'"l -g.2 e >,,.." g =.,3 ' 8 I..;.:,: '" < il 5 ri ""5l""', cr, r,-r-l e'-.:;; -ci - - -:;-;:; < < ««: :..: i.i.. Cl lL:.. " '1 " = ;:; ".g w] 2 88.g" :::1.1 = g.] "S;;' ' I ;"1" 'xv --, e, ;IV V r#... e, c 1(',1_ I...-.,...,"1"..::: 'T= -,- I ---ci-ci ««ci, < ';3 ::.:: ci,:::l ] s' lf ""':l" "."". i'. >...J U...l.. '1":=N i;:::j I ::l""i""n =r-l I I 1 ""r--n I _... r-- '(,I'>I;)='1"-. ("1--r--'C... fi :::-el (", I -('I... a e_ 1-, e, ''1" i'c I ("I r--is:;j I 1_;:::, I -- I 11t"1;:S:; I I I '1"OO 1 I I ("j I I t V..., 'T I! ("1... _1' X: '1" I I =::-] -t... ;::jl'"l I.,., I ::::; I I -e.,.,,,,,-e I I'... - cr. ""NO-. "1"N_"1'".e 1 -N...'- J'1"--J- I --..c r'-f"i-'c '1'"- -,...j I...-,... ('j I' r"t'1'" -::IO_:lCl"'-. e-.on-r-- J l>on... ;q'1" O' _ ""1- -1"11 ('I... 1_1 _ 1""'. -1"" '-1V'l Xl ;::("1 r-- r-_,...1 'C...-, _ :;: I I -.:: I : \C' :;...:; I ("I: -, I -,...l""'.,...iiir--i... r-'c- rl e... N - n i ij. "l:s.-::: _.-::.., I:S "la...;:] S.. ] '8... t.l}... C)..::: Q"""",, : _'.en 1- J I ("I; 1'1 rio' V).,., V'l N "'::i"'::i'ti -ci -ci. 'ci-ci 'ci. -ci 'ti - -d. ««-<:.J«<.J««g e g.sg. 2. ' s {;..::::..:::: ':: a :::"..::: :t'c ",",UrJ'l::t:: ", " :g c - "'" gg j.. "' <U Q

6 Clcridae Nel'Tbia ruficltis Ad. Cicind'didac Nccicinckla,ubf!rcula'tl Ad. 2 I 1 4 Chrysmelidac Parpsis charybdis Ad Staphylinidae Xanthfinus al1thrac;nu.'" Ad ) I 7 5 I 1 3 I I I 1 Curculinicac Ustrmh'rcs delaiguci Ad L I(vperdc... hvnariensis Ad ) Graphgnathus leuclruu Ad I I I Apinidtc Apin «/ids Ad. 19 I 4 I DJPTRA Sarcphagidac liybpykia,'uria Ad. 1 n, L ) Calliphridae Calliphra I'idna Ad. 4 L. 5 I 2 I 24 Muscidae (ind;;:i.) Ad L. ) Therevidae AllaharrIlYllchu.' sp. L. Sciaridac (indl) Ad. Drsphilidac Scaptf1l)'::.a sp. Ad. Ll'.PJOOPTRA Crambidac Orcramhus spp. Ad. 13 L Clephridae Chphra sp. Ad. L. 1) 4 6 I I liepialidac JViS('llflU ccrdnqia Ad. L. Nctuidac l'scudalctia separata Ad. 7 L ) A J.:r()ti. "psi/ml Ad. 2 1 L I I Pupa (indct.) 1 I HYMNOPTRA Frmieidae Chelcl1Icr (IIItarcticus Ad. ) I I Chneumnidac Ptercrmus prmis.wrius Ad I I 1 I 1 I Degithilla hersilia Ad. ) ) I I 6 I 7 ) 1 I 3 1 I I Vespidae Vespula germullica Ad Silky ccn (ind!.) 4 '- I ;:-... ARANIDA Lycsidac ( indt.) 41 7 I) I I, 2 ) I II 2 4 I rxdidae ( indclj I I OPILIONS CJ'l Phalangiidac 1)/ralaJlgiul1l pili l:: rriucola Lumbricidac All/(!hphra spp." 81 1 In 13 x5 12 II I I arthwrm chaetae (indt.) II II " " S OPRCULATA PtampyrRtlS untipdrllhi 15 ) I I I I ) II 4 ()tl PUI.MONATA l'hv.astra \ ariabilis 1 1 Helicel/a caperata I I '" Vallnia exccntrica Slug (indel) t:::l 'SOPOUA Oniscid<tC Prce/li.\"cahcr 21 DIPl.OPODA I ind!.) 15 1 (, H II 24 Cf-IILOPODA ( indet.) I I I I 4 PLANT MATRiAl Asparagus 3 I 16 4, Apple Grape I 4 Pea 4 4 1, Pear 1 I ) Nightshade 3 I I 4 2 II Tmat " Grass & clver 7 4 "Other than chaetae TOTAL tv... VI

7 252 N.Z. Jurnal f Zlgy, 198, Vl. 7 caten accidentally. Cleman (1974) reprted that harder fds are retained in the giard lnger than sfter nes. This is supprted by the mre frequent ccurrence f chaetae than f sfter parts f earthwrms in the starling giards. By analgy, the present data may underestimate the prprtin f fruit in the diet. Grit was present in 11 giards nly. FOOD OF NSTLINGS Alng with 334 adults, 27 nestlings aged 2-14 days were examined in Nvember and 35 in December In Nvember the nestlings were fed mre Agrypnus variabilis adults and larvae and Cstelytra ealandica adults (Fig. 3A) than any ther fd items, and were fed nly a few adults f Haplaner sp. and Aphdius sp. (Table I, Fig. 3D,). Like adults, nestlings cntained mre Cleptera than ther invertebrates. Three species (Scymnus ntescens, Graphgnathus leuclma, and Apin ulicis) fund in adults were absent in nestlings, but Pyrnta festiva was recrded nly in the nestlings. xcept fr a few adult Orcrambus spp. and Wiseana cervinata, th e Lepidptera fed t nestlings were larval Pseudaletia separata and Agrtis ipsiln. Amphipsalta sp. and Neara viridula, neither fund in adults, were recrded in nestlings. Hybpygia varia, Calliphra vicina, and muscid adults and/r larvae were present in mst nestlings. Lycsid spiders were fed mre in Nvember (Fig. 3F) than December, and a harvestman, Phalangium pili, absent frm the adult diet, was recrded in Nvember (Table 1). arthwrm remains were fund mre in Nvember than in December (Fig. 3G); chaetae were present in all giards. Ptampyrgus antipdarum, Physastra variabilis (nt recrded frm adults), and an undetermined slug were present in December. Millipedes and centipedes eaten by adults were nt present in the nestlings. The nly plant material in the nestlings was grass and clver; leaves were present in abut half the giards, and seeds in 13%. DISCUSSION In Hawke's Bay, starlings eat bth invert ebnhe and vegetable fds. Althugh predatin n tw majr pasture pests-cstelytra ealandica and Wiseana cervinata larvae-was insubstantial, starlings did eat ther pest species (Ferr 1976) f pastures (Hyperdes bnariensis, Graphgnathus leuclma, Aphdius spp., Pseudaletia separata, Agrtis ipsiln, Phaulacridium marginale), f crps (Agrypnus variabilis, Clephra sp., Dictytus caensus, Nysius spp.), and f dwellings (Calliphra vicina, unidel'ltifie Muscidae, Chelaner antarcticus). Nearly 4% ('f the ttal invertebrates eaten by adult starlings were msect pest species. Starlings als ate apples, grapes, pears, peas, and tmates, mst f them prbably after harvest r when left n the grund. Cllinge ( ) reprted that starlings in Great Britain wer e becming pests themselves, instead f predatrs f insect pests. Kalmbach (1922) cnsidered starlings beneficial in Nrth America, but Besser et al. (1968) later estimated that the birds caused cnsiderable lsses t cattle wners by eating high-prtein pellets at feedlts. Starlings take bth surface-dwelling and subterranean invertebrates. Table 2 shws that the starlings ate fewer subterranean animals (7%) than thse usually living partially hidden (45%), n the grund (31 %), r n vegetatin (18%). ast & Pttinger (1975) reprted a cnsiderable reductin in numbers.f the subterranean Cstelytra ealandica larvae in an irrigated pasture in Canterbury due t predatin by numerus starlings rsting nearby. The paucity f subterranean insects, especially C. ealandica, in the diet in the present study may be due either t the scarcity f C. ealandia n the feeding grunds r t the majrity f the pasture nt being irrigated, and thus harder than that f ast & Pttinger, and mre difficult t prbe. Kalmbach & Gabrielsn (1921) in Nrth America fund that Cleptera were eaten mre than any ther insect grup. In this study als, Cleptera were the mst frequently recrded grup in all mnths except February, when Lepidptera larvae dminated the diet. Hwever, cmpared t sftbdied insects, hard-bdied Cleptera adults wuld have remained lnger in the giards, and thus their imprtance wuld be exaggerated. Frequent amng Cleptera were the small (ca mm) beetles Haplaner sp. and Aphdius sp. usually fund in assciatin with dung. Aphdius sp. belngs t a grup f beetles which feed in dung and decaying vegetatin as well as scavenging; Haplaner sp., a carabid, prbably preys n fly larvae in the dung and n Aphdius sp. The dung prbably prvides a better habitat fr these beetles in winter and spring than in summer. It dries rapidly in the warm Hawke's Table 2. Percentage distributin, by seasn, f ttal prey items eaten by adult starlings frm different lcatins in pasture, Hawke's Bay (summarised frm Table 1). Summer Autumn Winter Spring (Dec- (Mar- (Jun- (Sep-- Prey lcatin Feb) May) Aug) Nv) Ttal Subterranean Partially hidden On grund Abve grund n (prey items)

8 Meed: Diet f Starlings 253 Bay summer, thereby restnctmg the habitat and cnsequently the nurnbers f beetles (Fig. 3D,). Purchas (1973) similarly explained the seasnal availability f fly larvae t rks (Crvus frugilegus) in the area. The lw densities f Hap/tcmer sp. and Aphdius sp. in early winter and late spring wuld cause starlings t spend much time searching, such that energy lsses may hav e been greater than gains. Hwever, in late winter and early spring (Fig. 3D,), when the beetles were abundant, it wuld have been mre ecnmical fr starlings t feed n them. Fewer Haplaner sp., Aphdius sp., and earthwrms were eaten in July 1971 and June 1972 than in ther mnths f cmparable availability (Fig. 3D,,G). There seems t be a delay befre starlings recgnised the availability f these as fds. This culd be accunted fr if the birds used specific search images (Gibb 1962), but it is als pssible that sme ther preferred fd item became scarce at the end f July. In the perid between the decline in the numbers f Haplaner sp. and Aphdius sp. and the availability f fruits, starlings subsisted n ther Cleptera, including Listrderes aelaiguei and Hyperdes bnariensis, and n Lepidptera, Hemiptera, Diptera, and Orthptera (see Table 1), the abundance f which was nt measured. The starlings tk fruit mainly in summer and autumn, when it was abundant. Damage by starlings t stne and pip fruits in New Zealand has been reprted by Dawsn & Bull (197). Such feeding n temprarily abundant fd has been reprted by Meed (1976) fr Lepidptera larvae and by ast & Pttinger (1975) fr Cleptera larvae. Cleman (1977) has reprted that starlings prefer certain fds, even t the extent f changing feeding sites and hunting behaviur t accmmdate such preferences. He als stated that ease f capture is a decisive factr in determining fd preference; his caged starlings fed selectively n many fds, including spiders, which they apparently preferred t adult beetles, flies, sme mths, and earthwrms. In Hawke's Bay starlings ate few spiders «1% f ttal invertebrate fd), despite their abundance (Fig. 3F); this suggests that ther fds, especially insects and fruits, were preferred. A pssible explanatin is that spiders have a lwer energy cntent than sme f the ther fds eaten (see Cleman (1977) fr energy values f sme starling fd items). A similar explanatin is pssible fr Listrderes delaiguei and Hyperdes bnariensis, which were als eaten less by adult starlings when mst abundant (Fig. 3B,C). Cleman fund that the energy value f digestible tissue is less in H. bnariensis than in spiders, and that f C. ealandica beetles is high. Adult starlings in Hawke's Bay selected these beetles in Octber 1971, but ate them less when they were mst abundant, in Nvember December (Fig. 3A). Nestling starlings had mre C. ealandica in their giards in Nvember-December than did the adults (Fig. 3A, Table 1); this suggests that the adult starlings were selecting the high-energy beetles fr their nestlings. T summarise, it appears that starlings in Hawke's Bay were shrt f fd in July-Octber, when they ate the lw-energy Listrd'eres delaiguei, Hyperdes bnariensis, and Lycsidae; they did nt eat them mre when mst abundant (Fig. 3B,C,F). Haplaner sp., Aphdius sp., and earthwrms were eaten mre in the same 4 mnths, when mst abundant, but nt at all when at lw density. This suggests that there is a threshld f ecnmic explitatin: at lw densities f these prey, the energy gained by eating them may have been less than the energy spent searching fr them. It is als likely that ther fds (e.g., Dictytus caensus, Agrypnus varil{lbilis, Graphgnathus leuclma, larval Lepidptera, and fruits) were preferred frm Nvember t June (Table 1), but their abundance was nt measured. Cccinellid beetles frmed less than 1% f the starlings' invertebrate diet in Hawke's Bay. Starlings are cnsidered unwilling t take cccinellids, r nt t eat them at all (Bgucki 1974). Cleman (1977) suggested that starlings ate few cccinellids because they are distasteflll. Mrever, cccinellids usually live n fliage, frm which starlings btained nly 18% f their invertebrate fd (Table 2). Althugh the adult and nestling samples did nt cme frm exactly the same lcality, it appears that the differences in their diet must be explained by fd selectin (Fig. 4, Table 1). Carabidae and Curculinidae were eaten mre by adults, whereas mre lateridae were recrded in nestlings. Mre Pseudaletia separata and Agrtis ipsiln larvae were fed t nestlings than were fund in adults. Similarly, mre Hybpygia varia larvae were fund in nestlings than in adults. Amphipsalta sp. and Neara viridula were recrded in nestlings nly. Tw larger Orthptera (Lcusta migratria and Cncephalus semivittatus) were eaten by adults but were nt fed t nestlings, whereas the smewhat smaller Phaulacridium marginale and Pternembius bigelwi were fund in bth adults and nestlings. The smaller Haplaner sp. and Aphdius sp. beetles, althugh present near the nests, were fed t nestlings less ften than they were eaten by adults (Table 1). Mre Lepidptera and Diptera adults and larvae were fed t nestlings than were eaten by adults, and mre Cleptera larvae were eaten by adults than by nestlings. The prprtins f adult Cleptera in the diets f adult and nestling starlings were similar. Cleman (1977) fund that newly hatched nestlings

9 254 N.Z. Jurnal f Zlgy, 198, Vl. 7 were ften fed n single sft-bdied invertebrates, but that lder nestlings were given bth sft- and hard-bdied fd. My samples were t small fr this questin t be examined. Because f differences in the species cmpsitin f adult and nestling diets, the sies f prey items were cmpared (Fig. 4). This revealed that nestlings tend t be fed significantly larger items (P<.5) than were eaten by adults. Perhaps it was nt wrthwhile fr adults t make flights t the nest with small fd items when they culd find larger nes in the same area. Hwever, the great variatin abut the regressin line (Fig. 4) suggests that ther factrs, such as the distance between the nest and the feeding area and the pssibility f nt finding larger fd items at all, may als influence the chice f prey. This intraspecific divisin f fd resurces pints t an interesting evlutinary pssibility. If starlings eat smaller fd items as encuntered but carry the larger nes t their nestlings, they may make very efficient use f the available fd supply. Starlings in Hawke's Bay share their fd resurces with mynas (Acridtheres tristis), rks (Crvus jrugilegus), and magpies (Gymnrhina tibicen). These fur species take a cnsiderable number f invertebrates frm n r near the grund surface, thugh starlings als prbe, and rks turn the turf ver t btain fd. Wilsn (1965) reprted that mynas in Hawke's Bay feed n insects as wen as fruit. Meed (1975) fund that althugh the diets f nestling starlings and mynas verlapped cnsiderably, Cleptera, Lepidptera, Dermaptera, and Ispda were eaten mre by starlings than by mynas, whereas Hemiptera, Diptera, Odnata, Hymenptera, spiders, and snails were eaten mre by mynas than by starlings. Prter (1979) reprted that rks in Hawke's Bay feed mainly n invertebrates (especially Diptera larvae and Cleptera adults) in warmer mnths and n walnuts (Juglans nigra) and acrns (Quercus spp.) in autumn and winter. Many f the invertebrate species (e.g., Hybpygia varia, Agrypnus variabilis, Cstelytra ealandica, Graphgn'athus leuclma, Listrderes delaiguei, Wiseana spp., and Pseudaletia spp.) eaten by rks were als pr esent in starlings, but mst f the smaner insects (ca mm) eaten by starlings in this study were nt fund in rks. McIlry (1968) reprted magpies eating many Cleptera, Lepidptera, Hymenptera, Diptera, Hemiptera, and Orthptera frm several lcalities in New Zealand. Thugh little is knwn f the fd f magpies in Hawke's Bay, Meed (1976) reprted that magpies near Christchurch included in their diet many invertebrate species (e.g., Cstelytra ealandica, Hyperdes bnariensis, Listrderes delaiguei, Agrtis Table 3. Percentage distributin, by seasn, f ttal prey items f different sie classes eaten by adult starlings, Hawke's Bay (summarised frm Table 1). Summer Autumn Winter Spring Length f (Dec- (Mar- (Jun- (Sepprey (mm) Feb) May) Aug) Nv) Ttal < n (prey items) J '';:::: c... ls C... I QJ QJ I / '+ll: a.. J -1 R=O.952 (I91OL)-O.87 Fig. 4. Relatinship between sie f fd items and prprtin f each given t nestling starlings r eaten by adults. The preference rati ranges frm -1 fr an item taken by adults nly t +1 fr an item taken by nestlings 'Only..- I I I iii I I 1 2/ Length (L). mm 3 4

10 Meed: Diet f Starlings 255 spp., Orcrambus spp., Phaulacridium marginale, Frficula auricularia, and Phalangium pili) that als ccur in starlings in Hawke's Bay. The diets f starlings, mynas, rks, and magpies shw that, despite cnsiderable verlap, these fur species use smewhat different cmpnents and prprtins f the fd supply, either because f differences in fraging behaviur r because their bdy sie restricts them t capturing efficiently nly a certain sie f prey. Kluijver (in Grmadki 1969) reprted that the prey f starlings ranged in bdy sie frm 1.5 mm t 5 mm. Dunnett (1955) fund that starlings did nt eat animals smaller than 1 mm, and that mst fd items were lnger than 5 mm and lighter than 4-5 g. In the present study the invertebrate fds fall clser t Dunnet's descri!'tin than t Kluijver's. A breakdwn f data n the sies f invertebrates eaten by adult starlings in Hawke's Bay (Table 3) shws that abut half the prey items were 2-5 mm lng and abut a quarter 6-1 mm lng, and that they were eaten mstly in winter and spring. In Canterbury, the insect prey brught t lder nestling starlings ranged in length frm 2 mm (Hemiptera nymphs) t 3mm (Lepidptera larvae) (Cleman 1977). Mst f the starling fd items reprted by Lbb & Wd (1971) and Meed (1976), als in Canterbury, were 2-2 mm lng, althugh many larger Lepidptera larvae and earthwrms were als eaten. Althugh Caithness (1968) and Lbb & Wd (1971) did nt reprt plant material in starlings, fruits and s,eeds are nevertheless cmmnly eaten. Meed (1976) fund cereals (Triticum sp. and A vena sp.), peas (Pisum sp.), and blackberries (Rubus sp.) in 4% f starling giards. Cleman (1977) reprted that cereals (Triticum sp. and Hrdeum sp.), clver (Triflium sp.), and weed seeds cnstituted 1% by vlume f the diet, and that fruits were rarely taken in his study area in Canterbury, where they were prbably scarce. In the present study, fruits and seeds were present in 44% f adult starlings (Table 2), prbably because f the pi'evalence f hrticultural areas and rchards in Hawke's Bay. Reginal differences in the diet f starlings in New Zealand reflect the available fd resurces f the habitats. ACKNOWLDGMNTS I am grateful t Drs D. G. Dawsn and P. e. Bull fr cnstructive criticism f the manuscript, and especially t Dr J. A. Gibb fr suggesting the frm f the discussin and fr ther helpful criticisms; t Mr R..R. Prter fr valuable assistance; t Mr G. Thelwell fr permissin t wrk n his farm; t Dr F. M. Clim fr identifying the mlluscs; and t the Systematics Sectin f ntmlgy Divisin, DSIR, fr identifying the insects. RFRNCS BARBR, H.S. 193: Traps fr cave-inhabiting insects. Jurnal f the lisha Mitchell Science Sciety 46: BSSR, J.F.; D GRAZIO, J.W.; GUARINO, J.L. 1968: Csts f wintering starlings and red-winged :->Iackbirds at feedlts. Jurnal f Wildlife Management 32: BOGUCKI. Z. 1974: A study f the diet f starling nestlings. Acta Zlgica Cracviensia 19: CAITHNSS, T.A. 1968: Insects and birds n tw New Zealand airfields. N.Z. ntmlgist 4: COCK, M.J.W. 1978: The assessment f preference. Jurnal f Animal clgy 47: COLMAN, J.D. 1974: Breakdwn rates f fds ingested by starlings. Jurnal f Wildlife Management 38: : The fds and feeding f starlings in Canterbury. Prceedings f the N.Z. clgical Sciety 24: COLLING, W : The Fd f Sme British Wild Birds. The authr, Yrk. DAWSON, D.G.; BULL, P.C. 197: A questinnaire survey f bird damage t fruit. N.Z. Jurnal f Agricultural Research 13: DUNNT, G.M. 1955: The breeding f the starling Sturnus vulgaris in relatin t its fd supply. Ibis 97: AST, R.; POTTINGR, R.P. 1975: Starling (Sturnus vulgaris L.) predatin n grass grub (Cstelytra ealandica (White), Mellnthinae) ppulatin in Canterbury. N.Z. Jurnal f Agricultural Research 18: FRRO, D.N. (d.) 1976: New Zea/and Insect P'ests. Lincln University Cllege f Agriculture, Canterbury. GIBB, J.A. 1962: L. Tinbergen's hypthesis f the rle f specific search images. Ibis 14: GROMADZKI, M. 1969: Cmpsitin f fd f the starling, Sturnus vulgaris L., in agrcenses. klgia Plska, Series A, 17: KALMBACH,.R. 1922: A cmparisn f the fd habits f British and American starlings. The Auk 39: KALMBACH,.R.; GABRILSON, I.N. 1921: cnmic value f the starling in United States. U.S. Department f Agriculture Bulletin N LOBB, W.R.; WOOD, J. 1971: Insects in the fd supply f starlings in mid-canterbury. N.Z. ntmlgist 5: McILROY, J.e. 1968: "The bilgy f magpies (Gymnrhina spp.) in New Zealand." Unpub!. M.Agr.Sc. thesis, Lincln Cllege, University f Canterbury. MOD, A. 1975: Diets f nestling starlings and mynas at Havelck Nrth, Hawke's Bay. Ntrnis 22: : Birds and their fd resurces at Christchurch Internatinal Airprt, New Zealand. N.Z. Jurnal f Zlgy 3: PORTR, R..R. 1979: Fd f the rk (Crvus frugilegus L.) in Hawke's Bay, New Zealand. N.Z. Jurnal f Zlgy 6:

11 256 N.Z. Jurnal at Zlgy, 198, Vl. 7 PURCHAS. T.P.G. 1973: "The feeding eclgy f the rk (Crvus frugilegus frugilegus L.) in the Heretaunga Plains, Hawke's Bay." UnpubI. Ph.D. thesis, Victria University f Wellingtn. WILSON, P.R. 1965: "Fd analysis f the gut cntents f the myna (Acridther& tristis), in Hawke's Bay." UnpubI. B.Sc. (Hns) prject. Victria University f Wellingtn.