SOCIAL ORGANIZAnON IN A POPULAnON OF THE HOODED CROW JON LOMAN

SOCAL ORGANZAnON N A POPULAnON OF THE HOODED CROW JON LOMAN Dept. f Animal Eclgy, Eclgy building, S 223 62 Lund, Sweden Received 16 February 1982, revised 8 May 1984 CONTENTS 1. ntrductin... 61 2. Study area... 61 3. Methds... 62 3.1. Trapping and tagging... 62 3.2. Recrding... 62 3.3. Radi-tracking... 62 4. Results... 62 4.1. Territrial and flck crws in spring... 62 4.2. Acquisitin f territries and pair bnd... 64 4.3. Attachment t the territry... 65 4.4. Seasnal variatin in flck size and feeding statins 66 4.5. Distributin f crws during winter.. 68 4.6. Migratin... 70 4.7. Survival... 70 5. Discussin... 71 5.1. Segregatin f territries and flck area... 71 5.2. Pair frmatin... 71 5.3. Territry establishment... 71 5.4. Occupancy f territries during winter... 72 5.5. Juveniles staying in their parents' territries... 72 5.6. Distributin f feeding sites during winter 73 5.7. Functin f rsting behaviur 73 6. Acknwledgements... 74 7. Summary... 74 8. References... 74 9. Samenvatting... 74 1. NTRODUCTON Data presented in this paper will be used fr tw kinds f cmparisn that give infrmatin n the evlutin f scial traits in the Hded Crw Crvus crnix and ther species. By cmparing infrmatin frm studies made at different lcalities, it is smetimes pssible t evaluate the imprtance f envirnmental variables fr certain scial traits. Other studies n aspects discussed in this paper are, e.g. thse by Tmpa (1975) in Switzerland, Kalchreuter (1971a) and Wittenberg (1968) in West Germany and Charles (1972) in Sctland. These studies refer t the Carrin Crw C. carne. Althugh prefer t cnsider it a species different frm the Hded Crw, its eclgical niche is sufficiently similar t justify cmparisns. t is als pssible t cmpare the scial rganisatin f crws with that f ther, systematically related, species. Crvid scial rganizatin can rughly be classified as territrial, clnial, r cmmunal. Territrial systems are thse where the living space is split up int exclusive territries, at least during the breeding seasn. Clnial species have their nests cncentrated and feed in a cmmn area arund the nesting clny but each nest is tended by a single pair. Cmmunal scieties are thse where a family grup tends ne nest in a shared territry. The Hded Crw mainly belngs t the territrial categry. Abshagen (1963) has described hw it can adpt a clnial nesting strategy and here will demnstrate that Hded Crws shw sme traits f cmmunal nesting t. Sme further data are included, but nt directly used fr cmparative purpses. This is t present a picture as cmplete as pssible f the crw ppulatin studied. believe this facilitates understanding f phenmena t be discussed. have previusly reprted n the breeding bilgy f the crw ppulatin studied (Lman 1977, 1980) and sme data frm these studies are given as a backgrund. Eggs are laid in early April and mean clutch size is 4.3 eggs. Abut 10% f the clutches are lst r deserted. Sme f these are replaced. Eggs hatch after an incubatin perid f abut 18 days. Hatching is asynchrnus, spanning ver 2-3 days. Twenty-five per cent f the brds are lst t predatrs. A mean f 2.8 yung fledge frm the remaining brds. The yungest ne in a brd ften starves t death. The number fbreeding pairs in the study area varied between 39 and 52 (1.9-2.5 pairs per km 2 ) during 1972-1979. 2. STUDY AREA The study was cnducted in suthern Sweden (55 0 40' N, 13 0 30' E). An "intensive study area" cvered abut 20 km 2 f the Revinge area, a military training field. t was used fr military training abut 5 weeks per year and fr the rest f the year much f it was grazed by cattle. There were sme Ardea 73 (1985): 61-75

62 SOCAL ORGANZATON HOODED CROW [Ardea 73 marsh-areas and a eutrphic lake, Krankesjn. All trapping and tagging was dne in the intensive study area, where attempted t find all nests. As all large winter rsts and sme ther places f imprtance fr tagged crws were situated utside the intensive study area, als made bservatins in adjacent areas. Agriculturalland dminated utside the Revinge area. Sme places in the study area prvided a large supply f fd fr crws thrughut the year r during certain seasns. These places will hencefrth be referred t as "cncentrated feeding places", CFP. CFP in the study area were: 1. A small fenced area where pigs were kept in the pen all year. The crws had pprtunities t take pig's fd. 2. Tw centres fr cattle raising in the Revinge area. Cattle were always in the vicinity f these centres during winter and were fed with hay and supplementary fd in the pen. The crws were ften seen t feed amng the hay. All calves were brn in the pen at these centres in spring. Dead calves and afterbirths prvided imprtant fd fr the crws. 3. Municipal garbage tips. 4. A dung heap that was particularly favured by crws during winter. 5. Winter ptat stres. These prvided fd fr varius mnths during winter. The breeding crws were prtected in the intensive study area. was always able t detect whether r nt nests belnged t tagged crws in this area. Nesting by tagged individuals utside the intensive study area culd smetimes be recrded. Other cmmn crvids in the study area were Rk C. frugilegus, Jackdaws C. mnedula, Magpie Pica pica, and Jay Garrulus glandularius. The frmer tw frequented the same winter rsts as the Hded Crws. The density f Rks increased cnsiderably in the curse f the study but this did nt seem t influence the density r distributin f the Hded Crw ppulatin. A pair f Hded Crws that nested in a cpse retained their nestsite despite the fact that Rks started nesting in the same cpse. 3. METHODS 3.1. TRAPPNG AND TAGGNG Mst results f this wrk are based n spt bservatins f tagged crws. Sme supplementary infrmatin was gained frm cntinuus bservatin f radi-tagged individuals. Mst crws trapped were taken in small, tw-cmpartment traps in their territries during the perid March t June. Trapping in the flck area was als dne mainly during this seasn. Altgether 60 territrial and.25 flck crws were trapped during the springs f 1974-1979. A Nrwegian crw trap (Kalchreuter 1971b) was tried during sme winters. This was nly successful fr ne shrt perid in December 1975 when 19 crws, mst f them winter migrants, were trapped. Mst crws were tagged as nestlings,at the age f abut 25 days. Altgether 230 nestlings were tagged during the springs.f 1975, 1976, and 1977. Of these, 136 were still in the study area in July r later in the year f hatching and 45 were present in July f next year. All crws trapped received wing tags (Piczzi 1971). The size f the tags was 33 x 70 mm. T facilitate identificatin and t allw calculatin f the rate f tag lss, identical tags were used n bth wings. Fr crws tagged as adults, the prprtin f individuals lsing bth tags was calculated t be 0% after ne year, 3% after tw and 30% after three years. The tags culd be identified at a distance f up t 300 m, using a 25x telescpe. All crws received fficial, numbered leg rings in additin t the wing tags. This permitted identificatin f crws fund dead r killed by peple utside the study area. 3.2. RECORDNG Mst bservatins f crws were made frm a car. The data in the results sectin are based n all bservatins f tagged crws but the majrity was made during tw standard car rutes in rder t reduce bias. One, cmpletely inside the intensive study area, cmprised 40 km and was cmpleted nce every week frm 1 April 1975 t 30 June 1978. The secnd rute ran up t 5 km frm the intensive study area, encircling it. This rute cmprised 70 km and was cmpleted nce each week frm 1 September 1976 t 30 June 1978. Althugh the standard rutes were n lnger cmpleted after this perid, additinal bservatins were made during the perid 1 April t 30 June 1979 in rder t determine the breeding status f tagged individuals. Each time a tagged crw was bserved recrded, amng ther things the size f the flck in which the tagged bird was seen. Aggregatin f crws were cnsidered as flcks, if the distance between individual birds was less than 50 m. 3.3. RADO-TRACKNG Sme bservatins were btained frm radi-tracking seven crws. The transmitters used perated n the 27 MHz band. They had a range f 400-800 m, a life f abut 10 days and weighed, abut 30 g. Each crw was tracked fr up t three full days. 4. RESULTS 4.1. TERRTORAL AND FLOCK CROWS N SPRNG 4.1.1. The territrial system and territryfidelity Almst all crws culd easily be classified as either territrials r flck crws during the mnths April t June. A crw that was always bserved singly r in a pair within a restricted area during this perid,was cnsidered a territrial crw; therwise it was cnsidered a flck crw. These "labels" remained until the next breeding seasn. During spring mnths, territrial pairs ccupied hme-ranges that did nt verlap but were prbably mre r less adjacent. The hmeranges were therefre cnsidered territries during the perid April t June. Mst f the territry was cvered every day in search f fd and territrials usually rsted within their territries. Territrials were ccasinally bserved utside their territries during the breeding seasn: 1. Neighburing pairs culd smetimes be seen feeding peacefully at the brder between

1985] SOCAL ORGANZATON HOODED CROW 63 their territries. 2. Smetimes crws were seen making shrt excursins utside their territries. 3. f a territrial crw was trapped, neighburs culd be seen in its territry within a shrt perid f time. A territrial crw was nce slightly wunded in a trap. After release it was attacked and prbably killed by a neighburing pair. 4. Territrial crws usually rsted within their territry. Hwever, fur times tagged r raditagged territrials were bserved t participate in evening gatherings either n the grund r in trees in their wn f in neighburing territries. These gatherings prbably invlved mainly territrials. After these assemblies, which lasted 15 t 30 min, the crws flew ff single r in pairs. * Rst 8h Hur in area 50 Flck size 1km 4.1.2. Distributinand hme-ranges f flck crws The hme-ranges f flck crws were larger than thse f territrials during the perid April t June and they verlapped widely. Flck crws were mainly restricted t CFP and their surrundings. Areas much used by flck crws apart frm CFP were prbably situated at the intersectin f several territries, far frm the nests (Fig. 1). These areas were utilized by the same flck crws as thse usually present at the clsest CFP. Nests were in sme cases fund Fig. 2. Daily activity pattern f tw radi-tracked flck crws. A was tracked n 18, 20 and 22 March 1977; n 18 March (brken line) during the afternn nly. B was track~ ed n 5 April 1978. Duratin f stay is nt given when entire day spent in ne place. clse t CFP and, at least in ne case, the hmerange f a nesting pair was regularly used by flck crws t. This pair was smetimes seen feeding tgether with flck crws but as the pair Tagged flck crw O Nest A territrial female, alne -~-,in flck fz22 CFP 0 0 Fig. 1. Distributin f flck crws during spring in relatin t a CFP and t nests. Sites f bservatin f a female that nested clse t the CFP are als given. e. 0 0 0 0 00 1km

64 SOCAL ORGANZATON HOODED CROW [Ardea73 was usually separate frm thers and bred, it was classified as a pair f territrial crws (Fig. 1). During spring and summer, flck crws usually rsted within ne r tw kilmetres frm their daytime feeding areas. Observatins f tw radi-tracked flck crws suggest that changes f rsting place were cmmn (Figs. 2A, 3A). Mst f the day was spent feeding at ne r tw CFP but sme flck crws did range further afield (Fig. 2B). Such excursins were made by single birds r small flcks. 4.2. ACQUSTON OF TERRTORES AND PAR BOND 4.2.1. Territry acquisitin by flck crws Flck crws were smetimes bserved alne r in pairs. During winter these bservatins were mre r less randmly scattered amng all places where flck crws were bserved. Hwever, during spring single r paired flck crws were ften bserved utside the area usually frequented by flcks. Flck crws that were bserved in such situatins will be called "prspectrs". This pattern is reflected by the relative distance f crws in flcks and prspectrs t the nearest CFP during winter January March) and spring (April-June). Cnsidering all bservatins f tagged individuals, these distances were 770 m (n = 180) and 1100 m (n = 10) during winter (nt significant, Median test,x 2 = 0.11) and 520 m (n = 155) and 1300 m (n = 25) during spring (significant, Median test, X 2 = 5.75, P < 0.05). The flck area frequented by prspectrs was the ne clsest t the area f prspecting (13 cases). Furthermre, if a crw had been bserved in a flck area the year befre it became territrial, this had been in the area that was clsest t its newly acquired territry (5 cases). Prspecting was ften fllwed by the ccupancy f a territry at the site f prspecting, r clse t it, in the fllwing year (4 cases; territries were 100, 200, 200 and 400 m frm the site f prspecting). Prspecting crws were single birds in 15 ut f 24 cases (62%). This was similar t the fractin f territrial crws that were bserved singly during spring (Fig. 8). Hwever the tw frac- tins may nt be quite cmparable as the territrial female spends much f the time n.the nest r in its immediate vicinity and nt with the male. thus think that sme f the 15 single prspecters really were unmated birds. Recrds f tw radi-tracked male crws further illustrate these pints. One single prspecting male was tracked fr three days in March 1975. He spent abut half f each day alne within a restricted area, the ther half in a flck at ne r tw places abut ne km away (Fig. 3A). Frm 1976 nwards he ccupied a territry clse t the area where he had been prspectc ing. n April 1975 the area where the single male had been prspecting was ccupied by anther radi-tracked crw that, with its mate, spent the whle day there and was thus territrial. This pair did nt breed that year. t rsted 5h * 1km Fig. 3. A. Activity range f a prspecting male, raditracked n 24 March 1975. t had asimilar range n 23 and 27 March. B. Activity range f a mated male, presumably a first-year territrial, radi-tracked n 3 March 1975. t had a similar range n 27 April. Symbls as in Fig. 4

1985] SOCAL ORGANZATON HOODED CROW 65 with flck crws abut tw km frm the territry (Fig. 3B). 4.2.2. Age f newly established territrials Nne f the sixty crws that had been tagged as nestlings and that were still in the study area with identifiable tags, had established a territry when abut ne year ld. At an age f abut tw years, 2 ut f 30 and at abut three years, 5 ut f 13 did pssess a territry. Of the frmer tw, ne was breeding and f the latter five, three were breeding in their first year as territrials. 4.2.3. The start f breeding Of 12 newly established territrial pairs, seven did nt breed in their first year f territrial life; five f these seven did s in their secnd year but tw did nt. Thse that had started breeding cntinued t d s in succeeding years. n eleven instances ne bird was recrded breeding fr the first time while its partner had already bred in the territry previusly. n nine ut f these eleven cases did the newly frmed pair cnsist f tw first time breeders. Hwever, mst f these pairs had been present in the territry fr nearly ne year while pairs f inexperienced birds smetimes had been present fr less than a mnth nly befre the start f breeding. 4.2.4. Mate fidelity Of 16 different pairs, bth partners were tagged and these were bserved fr a ttal f 34 pair-years. n these years, nly ne pair split up. This pair had established a territry but n breeding tk place. As bth had been tagged in that territr, it is uncertain whether they were first-year territrials but this seems likely as they did nt breed (sectin 4.2.3). The tw crws f this pair were mated with untagged partners next year. The male was still in its riginal territry, the female in an adjacent ne. 4.2.5. Pair frmatin Sme pairs prbably frmed between tw flck crws. This was mst likely the case with tw crws that prspected tgether at several places in ne spring and were als seen in the flck area at this time; Other pairs, prbably the majrity, frmed in the territries. All territrials that were knwn t have lst their mate (eight males and ne female) kept their territries in the fllwing spring. f the lss ccurred during the breeding seasn (seven cases), the widwed crw smetimes remated in the same seasn. This was bserved in fur cases, including in the female that lst her partner. n tw cases n remating was bserved until the fllwing breeding seasn, but such rematings may have gne unnticed. Fr ne case n pertinent infrmatin is available as the territry was difficult t bserve. 4.3. ATTACHMENTTOTHETERRTORY 4.3.1. Territrials Practically all territrial crws were in their territries during April t June (Fig. 4). By defi nitin all shuld be, but there are a few bservatins that cnsider exceptinal (sectin 4.1.1). n years with a late start f breeding sme territrial crws were seen in flcks away frm their territries during the first week f April; July and August were transitin mnths with an increasing number f bservatins f territrials frm utside the territry, thugh mstly in its vicinity (Fig. 4). Abut 25% f all bserved territrials were in their territry during the winter mnths. There was variatin between pairs with respect t the tendency t stay in their territry during winter. fund n crrelatin between experience as territrial (first-year territrials % 100 50 ---_...-----. JASONDJ FMAMJ 73 72 83 115 121 95 120 133 222 433 255 211 Fig. 4. Seasnal variatin in the distance at which territrials were bserved frm their territry. The drawn line gives the fractin (%) f all bservatins within the territry, the brken line the fractin f bservatins within 1 km f the territry brder. The number f bservatins is given belw the x-axis.

66 SOCAL ORGANZATON HOODED CROW [Ardea 73 vs. mre experienced nes) r bdy weight f the crw and the prprtin f all bservatins made within the territry frm Octber t March. Hwever, there was a tendency fr territries that were likely t yield an adequate fd supply during winter t be mre adhered t than thers. Such territries, e.g. thse in the vicinity f places where cattle were fed during winter, thse cntaining small streams, ne clse t a small village and even ne including the eclgical field statin, were significantly mre frequented during winter. (Mann-Whitney V-test, with territries ranked accrding t the prprtin f all bservatins f territries made within their territry during winter, V = 235, P < 0.01). 4.3.2. Juveniles Juveniles were virtually cnfined t their parents' territries in June (Fig. 5). Hwever, cmparatively few juveniles were bserved tgether with their parents in this mnth (Table 1). This was because mst juveniles left the nest in the beginning f June and spent their first weeks in its vicinity, waiting fr their parents t feed them. Later, in July, they fllwed their parents but fed mainly by themselves. During July and August, increasing numbers f juveniles left their native territries and almst nne were bserved there frm September nwards (Fig. 5). Juveniles were smetimes bserved tgether % 100 50-49 00 -..._""' \ \ \ \ \ "\ '...---------......_-------------- JJASONDJ FMAMJ 90+2 69+1 95+3 59+1 45+4 53+4 30+3 50+3 44+7 57+3 90+3 56+9 40+6 (68) (59) (59) (45) (30) (32) (24) (32) (27) (31) (43) (31) (19) Fig. 5. Seasnal variatin in the distance at which juveniles were bserved frm their natal territry. The drawn line refers t bservatins within the territry, the brken line t bservatins within ne kilmetre away frm it. The circles refer t tw juveniles that were regularly bserved in their parents' territry during the spring when they were abut ne year ld. Belw the x-axis the number f bservatins is given, excluding the tw atypical juveniles plus the number f bservatins f these latter; the number f individual juveniles invlved is given in parentheses. Table 1. Seasnal variatn in the assciatin f juveniles with their parents. N refers t the number f bservatins f tagged juveniles f which the mther r father was als tagged. "% with mther/father" gives the prprtin f these bservatins when a parent was recrded in the vicinity f the juvenile Mnth N % with N % with mther father hne 13 31 22 9 July 13 38 19 37 August 16 0 28 21 September 13 15 15 20 Octber 13 46 15 7 Nvember 11 9 18 0 December 78 0 86 3 with their parents in flcks during autumn but there was almst n assciatin after Nvember (Table 1). Tw juveniles, f different parents, were atypical. During their secnd and third calendar years they were bth still assciated with their parents' territries during the breeding seasn (Figs. 5, 6). At least ne f these tw crws seemed less restricted t the parents' territry when tw years than when ne year ld (Fig. 6B), still it was mre assciated with it than were "typical" juvenile crws (Fig. 5). Bth these crws were heavier at fledging than average (480 and 510 g, average 460 g) suggesting they were males. 4.4. SEASONAL VARATON N FLOCK SZE AND FEEDNG STATONS Flcks were largest in winter and smallest during the breeding seasn (Fig. 7). Thugh territrials and flck crws were ften seen in the same flcks there was a tendency fr territrials t be in smaller flcks. Juveniles were mainly tgether with their parents in July and August and the size f the flcks in which the tw categries were bserved was similar. Later, juveniles behaved mre like lder flck crws with respect t flck size. n late winter, juveniles were even fund in larger flcks than lder flck crws. The fractin f crws seen in pairs r alne during different seasns shwed a similar pattern (Fig. 8). Older flck crws very rarely were alne r in pairs during the perid Octber t December and thus shwed n tendency t prspect at this time. Juveniles bserved alne

1985] SOCAL ORGANZATON HOODED CROW 67 cp @ @Nest 00 Alne r with parents n a flck Fig. 6. Spring (April t June) distributin f tw crws (square and circular symbls) that remained assciated with their natal territries. The situatin in their secnd (A) and third (B) calendar-year is shwn. The father's (drawn plygn) and mther's (brken plygn) hme-ranges are given. Only ne parent was tagged in each pair and the tagged father had lst its tag in the third calendar-year.. in Nvember (Fig. 9D). Rabbits that had been killed by cars were used as a fd surce in December and January, usually when snw cvered the grund (Fig. 9C). There were differences in these respects between the three categries f crws. Cmparing the tw mst imprtant feeding site categries, CFP and fields withut cncentratins, flck crws frequented CFP mre than did territrial crws (36%, N = 615 and 4%, N = 1761 respectively). There was n bvius seasnal variatin in this respect; the prprtin f territrial crws at CFP varied between 0 and 18% and that f flck crws between 19 and 44%. n bth categries, a maximum ccurred in the perid Nvember-December. Juveniles behaved similar t territrial crws during their first mnths (13%, N = 139, at CFP in July-August) but mre like lder flck crws later n (40%, N = 484, during the rest f their first year). During midwinter and early spring they were even mre assciated t CFP than were lder flck crws (56%, N = 191 during January t April). These tendencies shuld be cmpared t anther juvenile tendency, viz., t stay with the parents in July and August (Table 1) and t the difference in size between flcks jined by juveniles and lder flck crws during the perid January t March (Fig. 7). r in pairs (the partner may well have been a sibling r a parent) were usually still in their native territry during late summer r autumn. Territrial crws were usually alne r in pairs when bserved in their territries (73%, N = 1193). This was especially prnunced during the breeding seasn (88%, N = 846). When far frm their territries (> 1000 m) they were nearly always in flcks (nly 5% single r in pairs, N = 322). This fractin was similar t that fr flck crws (Fig. 8). Mst crws fed in places where n bvius cncentratin f fd was fund (Fig. 9E). A' cnsiderable prprtin was seen at CFP during winter (Fig. 9A). n sme mnths, many crws utilized fields where, fr a shrt perid f time, fd was plentiful as well as well dispersed.ver an extensive area, e.g. newly swn fields in April and field where manure had been spread nd 30 20 10 J A S 0 N D J F M A M J T 20 38 51 85 83 62 100 95 161 67 14 20... J 41 80 50 49 45 30 43 41 40 87 62 41 -F 25 29 37 68 76 45 47 35 87 78 46 21 --- C 605 1017 1037 1594 1858 2774-2537 2553 2143 1078 202 229 Fig. 7. Mean size f flcks. T: Territrials. J: Juveniles, F: Flck crws, C: All crws bserved during weekly censuses, regardless f whether tagged r nt. Crws bserved singly r in pairs excluded. Number f crws given belw x-axis.

68 SOCAL ORGANZATON HOODED CROW [Ardea 73 % 100 N 280 50 C F A T 81 67 ------- J 83 113 -F 28 28 ---- C 748 1083 Fig. 8. Fractins (%) f crws bserved alne r with family (mate, parents, yung f the year). Letters as in Fig. 7. Number f bservatins given belw x-axis. ~ t, ", " 240 200 160 120 80 40 20,2.1975/ l "i,--j. i 30.1.1979/ t \... -~' / \ \ 16" 17",..----<0,,/ \ 5.3.1975 18" Time 4.5. DSTRBUTON OF CROWS DURNG WNTER 4.5.1. Size, distributin, and characteristics f rsting places Mst crws were in large rsts during winter and these rsts remained in the same place frm year t year. Crws smetimes als rsted in smaller grups during winter as exemplified by ne radi-tracked crw (Fig. 13B). All three large winter rsts in the study area were in small plts f spruce frest as were tw ut f three ther knwn rsts in the vicinity f the study area. % Fig. 10. Number f crws (N) at different days n evening gathering places, frm arrival f the first crws t departure fr the rsts. 4.5.2. Flights t and frm rsts The crws usually arrived at the rsts abut 30 min after sunset. The final flight t the rst usually started frm ne r a few gathering places in the neighburhd (Fig. 11). n winter several hundred crws, and als Jackdaws and Rks, left the gathering places tgether. Rsting places were traditinal nes, changing psitin a few hundred f metres at mst within r between winters. The number f crws at a gathering place usually increased gradually during the hur befre rsting. Departure fr the 50 0-- _ JASONDJ FMAMJ 748 1083 1124 1778 2037 29502677 2717 2342 1304 499 409 Fig. 9. Fractins (%) f all crws using distinct types f feeding site in the curse f a year. A: Places with cncentrated fd f permanent r lng-term availability (CFP). B: Places with cncentrated fd available fr mre than a few days. C: Places with cncentrated fd available fr ne r a few days. D: Extensive but temprary feeding sites with plenty f fd. E: Other sites, mainly fields with n apparent cncentratins f fd. F: Fd. Number f crws bserved given belw x-axis. * Rst * Small rst Fig. 11. Lcatin f large, permanent winter rsts and f sme smaller rsts, used irregularly during winter.

1985] SOCAL ORGANZATON HOODED CROW 69 Fig. 12. Rst-recruitment areas. Daytime feeding sites are indicated fr all tagged crws identified at each f fur gathering places (large symbls) during Octber-March. ncluded are all sites where these birds were bserved within ne mnth frm the date when first seen at a gathering place. Use f a lnger time interval wuld have increased the pssibility f cnfusin due t ccasinal changes f gathering place within ne seasn. Rst-recruitment areas are separated by thick lines, based n the distributin f the bservatin sites. Crws flew t the same rst frm tw f the gathering places. 1km..._--""', e. rst, n the ther hand, tk nly abut ten min fr the whle flck (Fig. 10). Crws arriving at the gathering places smetimes frm traditinal sub-gathering places, usually came in flcks that were larger than thse bserved during daytime feeding. Gathering places, and als traditinal sub-gathering places, were situated 100--300 m frm the CFP (Fig. 11). The flight pattern in the mrning was apprximately the reverse f that in the evening. The same gathering places and sub-gathering places were used. Hwever, less time was spent befre reaching the feeding places in the mrning than n the evening flights away frm these. 4.5.3. Rst-recruitment and winter activity range Crws utilizing gathering places in winter were bserved feeding in different areas during daytime. These areas are here termed rst-re e cruitment areas. These were rughly adjining (Fig. 12). The hme-range f an individual crw usually cvered nly a part f ne rst-recruitment area. Sme crws changed recruitment Table 2. Frequency f rst-recruitment area change (see Fig. 12) fr different crw categries during different seasns. A: Percentage f bservatins in the area where each individual crw was bserved mst frequently. Band C: Percentage in the secnd and third mst frequented area fr each crw. N: Ttal number f bservatins Octber-March April-June July-September N A B C N A B C N A B C Juveniles 297 95 4 1 198 89 8 3 233 96 4 0 Flck crws 423 95 4 0 213 95 5 0 106 97 3 0 Territrial crws 829 99 1 0 903 100 0 0 230 100 0 0

70 SOCAL ORGANZATON HOODED CROW [Ardea 73 Table 3. Frequency f bservatins fr varius categries f crws during winter. The expected mdel distributin is als given in parentheses tgether with x2-value fr a test f fit between bserved (zer-truncated) distributin and the mdel distributin. Fr juveniles and flck crws, the Pissn distributin served as a mdel; fr territrial crws the distributin btained by taking the mean value f the Pissn distributin and the Gemetric distributin. These mdels gave the best fit t the data Number f bservatins during winter (Octber-March) 1-2 3-4 5--6?:.7 X 2 d.f. F Number f individuals bserved the given number f times Juveniles Flck crws Territrials 26 ( 1.3) 11 ( 8.6) 10 (11.3) 3 ( 5.5) 3 ( 1.7) 2.95 3 0.5-D.3 12 ( 0.5) 8 ( 5.9) 8 (10.7) 8 ( 7.2) 2 ( 3.3) 2.03 3 0.7-D.5 11 (14.8) 27 (28.7) 32 (28.3) 13 (17.8) 16 (11.7) 3.46 3 O. 5---'-D. 3 area in the curse f ne winter. This was mst crnman fr juveniles, less crnman fr lder flck crws and least fr territbrials (Table 2). Juveniles and lder flck crws smetimes changed rst-recruitment area during spring and summer but territrials never did during these seasns. During winter, the crws usually spent the whle day at ne r tw CFP, usually in large flcks (Fig. 13). They smetimes mved mre extensively in smaller flcks within the rst-recruitment area (Fig. 13B). These data frm radi-trackings are crrbrated by infrmatin n the size f flcks at different types f..ljl.------------- 1/2h 5 ~ /----,- ~1/2h-""" "... ---------~-..._---,) Fig. 13. Daily actvty ranges f tw crws radi-tracked during winter: A was tracked n 26 and 27 December 1975 and B n 2 and 7 January 1977. Symbls as in Fig. 4. Duratin f stay is nt given when entire day was spent in ne place. \, ~ feeding sites. Mean flck size at a CFP in December t February was 27, at ther sites 6.3 birds. Furthermre, mst winter hrne-ranges cmprised at least ne CFP. 4.6. MGRATON A large fractin f the juveniles and lder flck crws that were tagged befre winter and bserved later ne were never bserved during the winter mnths (Octber t March) (Table 3). This suggests they had left the study area. Practically all territrials were bserved in the study area during winter and cnclude that mst, prbably all, stayed. Twelve ringed crws were recvered dead r sht during winter utside the study area. Five f these were fund in Denmark (islands f Sjaelland, Llland and Fyn), the ther seven in the Swedish prvince Sk{me, where the study area is situated. All seven summer recveries utside the study area carne frm Skane. The latter prbably represent permanent emigratin while the winter recveries may have resulted frm bth emigratin and temprary migratin. Out f 19 crws trapped and ringed during winter, three were later sht in Finland. 4.7. SURVVAL Survival was calculated as the prprtin f birds bserved in the study area during ne breeding seasn (April t June) and bserved again during the next seasn. Only crws whse tags were nt lder than tw years were included as tag lss was neglegible during this perid f time. These values thus represent minimum survival as missing crws may have emigrated.

1985] SOCAL ORGANZATON HOODED CROW 71 Survival was 92% fr territrial crws (based n 57 birds and 132 bird-years); sme birds were re-tagged in the curse f the study. Fr flck crws (exluding the first year f life) survival was 73% (bases n 54 birds and 62 bird-years). Tw f the territrials that disappeared and ne flck crw are knwn t have been sht during winter, while the fate f the thers remains unknwn. 5. DSCUSSON 5.1. SEGREGATON OF TERRTORES AND FLOCK AREA n spring mst flcks were fund at r arund a CFP. This phenmenn was als bserved by Charles (1972) and Bhmer (1976). Territrial pairs prbably have n influence n the distributin f flcks (Bhmer 1976). Flcks were als fund elsewhere, at places away frm crws' nests. This agrees with Charles' (1972) bservatin that "grup intrusins" mainly take place at the brderline between territries. The fact that territrial crws restrict the mvements f flck crws utside the flck area was als evident frm an experiment; after territrial females were remved in May, flck crws invaded in the territries (Granssn & Lman 1982). Charles (1972) regularly bserved "grup intrusins at nests" t. This was nt recrded in my study area, which may explain the lw rate f predatin n eggs and chicks (Lman 1980) as cmpared t that recrded by Charles (1972) and Wittenberg (1968). They suggested that flck crws were respnsible fr this predatin. The fact that pairs nesting in the flck area raised yung t fledging suggests that predatin frm flck crws was f minr imprtance in my study area. The reasns fr these differences are nt clear. The functin f evening gatherings f territrial crws remains bscure; the phenmenn may cnstitute sme kind f defence against predatin, similar t the functin suggested fr pre-rst gatherings (sectin 5.7). 5.2. PAR FORMATON Mst previus investigatrs have stated that pairs are frmed in flcks (Wittenberg 1968, Kalchreuter 1971a, Charles 1972). This was als true in my study area, at least in sme cases. Hwever, it seems likely that single birds prspect t as was als suggested by Kalchreuter (1971a). f these single birds are successful in establishing a territry, they prbably find a mate in this territry. Hwever, successful territry-establishment by single birds was never b" served and it is difficult t see hw this wuld be pssible in cmpetitin with prspecting pairs. t may be significant that the single male that was radi-tracked while prspecting lst his ptential territry t a pair. t is pssible that prspecting by single birds is beneficial t them by becming acquainted t the area, smething that may be useful when prspecting with a mate later n. All widwed males (and ne such female) in my study area were able t maintain their teritries and frm new pairs. Pair-frmatin mst likely tk place within the territry. This is in cntrast t Charles'(1972) bservatin that a single crw cannt defend its territry. f the female died during the breeding seasn, flck crws appeared in the territry (Granssn & Lman 1982). This culd actuallypenefit the single male, enabling him t find a' new mate quickly withut having t leave its territry pen t cmpeting pairs. 5.3. TERRTORY ESTABLSHMENT Evidently, the ccupancy f a territry is a prerequisite fr breeding. Besides, territry wnership might increase survival. Hwever, the figures shwing higher survival rate fr territrial crws than fr flck crws are biassed by differences between bth grups in the tendency t emigrate. Als, the difference culd simply be due t higher survival rate f individual crws that are particularly fit and therefre are the ne's mst likely t establish territries. A yung nn-territrial crw has tw ptins: t remain in the natal territry mst f the time, using it as a base fr acquisitin f its wn territry, r t jin a flck and spend mst f the time in the flck area. The advantages and disadvantages f the tw ptins are discussed in sectin 5.5. D flck crws select a preferred territry r d they settle mre r less at randm? The situatin in my study area suggests that there prbably is little pprtunity fr birds t chse at

72 SOCAL ORGANZATON HOODED CROW [Ardea 73 all. Cmpetitin fr a territry was intense as few teritrials died, leaving vacancies t the number f flck crws present. The availability f territries may have been even mre limited, since flck crws usually restricted prspecting t ne rst-recruitment area. Within the study area crw territries are situated in different habitat types. n dry areas clutch size has been shwn t be relatively small (Lman 1977). Since n differences have been fund in breeding experience f crws inhabiting different habitats (dry vs ther), clutch size differences can be regarded t reflect differences in habitat quality. t is striking that territrial crws were nt fund t attempt exchanging their territry fr a better ne. This culd be because the establishment f a new territry implies less attentin fr the ne already ccupied, with the risk f lsing bth territries. These bservatins stress the difficulties crws have t btain territries. Despite the cmparatively high survival rate (fr passerines) f flck crws, it is likely that many never succeed in acquiring a territry. 5.4. OCCUPANCY OF TERRTORES DURNG WNTER The tendency fr territrial crws t stay in the territry during winter differs between the varius ppulatins studied. n Sctland (Charles 1972, Spray 1978), Switzerland (Tmpa 1975), and nrthern West Germany (Wittenberg 1968) territrials stay and defend their territries thrughut the winter. n suthern West Germany the crws remain in their territries mst f the winter but depart under snw cnditins (Kalchreuter 1971a). n my study area mst crws were cnfined t their territries nly in the perid April t June. There was little snw but it is pssible that during winter there were less pprtunities fr feeding than in the three study areas mentined abve as these were situated in farming areas. Lack f fd cannt explain the crws' absence frm their territries during summer in my study, but it is pssible that new territries are seldm established during this seasn and thus there is n need t defend a territry. This is supprted by Wittenberg's (1968) finding that territrial crws (in his study ppulatin) are absent frm their territries during summer but return later in autumn. f the territrial area that studied really ffered relatively little fd, especially as cmpared t the flck area, this culd explain three ther differences between my study and thers': 1. Kalchreuter (1971a) and fund that juveniles left their natal territries in late summer while Charles (1972) and Tmpa (1975) fund that they stayed in the teritry fr a lnger perid f time, t leave it nly gradually during their first winter. A difference between my findings and thse f Kalchreuter (1971a) is that he fund that juveniles frmed exclusive flcks in summer. fund n indicatin f this; juveniles were fund in flcks with lder flck crws and teritrials. 2. n Charles' study, flck crws established small and temprary territries during winter and early spring. Later n, during the breeding seasn, they were usually back in the flcks. did nt bserve this. 3. Charles (1972) als fund that persistent intruders were tlerated during winter as third birds in the territries but they were evicted befre the fllwing breeding seasn. This was nt bserved in my study area. 5.5. JUVENLES STAYNG N THER PARENTS' TERRTORES Only Tmpa (1975) has previusly bserved crws, apart frm the pair, that stayed permanently in a territry during the breeding seasn. Tw ut f 36 territries bserved by him held three crws. He had n further bservatins cncerning rigin r nature f these birds. The third birds that Charles (1972) bserved were bviusly f different status. They were recruited frm the flck and evicted by the start f the breeding seasn. bserved juveniles that has sme cnnectin t their natal territry thrughut the winter and, like their parents, became a permanent resident there during the breeding seasn. The functin f this behaviur fr juveniles and parents is nt clear but will discuss sme pssibilities. 1. The juvenile that remains in the parents' territry culd serve as a helper (Skutch 1935), defending its parents' territry and assisting with feeding the new brd. N bservatins were made specifically t determine this, but at least in ne f the territries

1985] SOCAL ORGANZATON HOODED CROW 73 under bservatin, the juvenile was never bserved in the vicinity f the nest, making it unlikely that it was feeding its yunger siblings. 2. The juvenile may be better ff with respect t fd in the parents' territry than having t cmpete with ther crws in the flck area. By staying within the territry, the juvenile crw may even be able t ccupy part f it as a new territry, especially if the parents' territry is large. This was bserved fr Scrub Jays Aphecma ceruiescens by Wlfenden & Fitzpatrick (1978). By having a stable base in the territrial area, establishment f a separate territry culd be facilitated. Charles (1972) reprts that during winter, juveniles remaining in the natal territry were better tlerated by neighburing territrials than were unfamiliar crws. f this applies t my study area in spring, prspecting by these juveniles wuld be facilitated. This hypthesis is supprted by the fact that ne f the juveniles bserved t remain in the natal territry established a territry f its wn abut 500 m frm its parents' territry when three years ld. 5.6. DSTRBUTON OF FEEDNG STES DURNG WNTER n general, juveniles feed mainly at CFP and in large flcks during winter while territrials visit CFP less frequently and feed in smaller flcks. Older flck crws appear t be intermediate in these respects. Flck size is prbably strngly influenced by the nature f the feeding site. Territrials and lder flck crws, which are dminant and therefre are able t cmpete successfully fr fd, culd be expected t be mre ften present at a CFP. Hwever, the reverse is true. suggest this is because territrials need t guard their territry, whereas lder flck crws which are capable t defend a vacant territry shuld be infrmed abut the situatin in the territrial area. 5.7. FUNCTON OF ROOSTNG BEHAVOUR Several psitive effects n survival f rsting behaviur can be envisaged: 1. A rst utilized during winter may be a place with a particularly favurable micrclimate, thereby reducing energy lss during night (Swingland 1977). This cannt be the nly functin as mre energy is lst while flying t the rst than can be saved during night (Gyllin et al. 1977). 2. Cmmunal rsting culd serve as a defence against predatrs. This culd be accmplished in tw ways: a) Detectin f predatrs is prbably facilitated where a large number f crws has gathered. b) f the predatrs are mre r less territrial, cmmunal rsting may result in a reductin f the number f predatrs present relative t the number f crws (Bertram 1978). Gathering places are always situated in pen landscapes suggesting defence against predatrs t be an imprtant factr. Rsts, n the ther hand,are situated in cpses where ptential predatrs like the Gshawk Accipiter gentilis and the Great Hrned Owl Bub bub will have gd pprtunities t strike unnticed. Gathering places culd be imprtant because they allw crws t gather at a cmparatively safe place and reach the rsting place simultaneusly. f they arrived singly at the rst, thse arriving first wuld face the risk f being taken by predatrs. nterrupting their flight t the rst at a safe place, and nly cntinuing when sufficiently many ther crws are likely t fllw, may have started the evlutin f a gathering place system. think, the pattern f arrival t and departure frm the gathering places supprts this (Fig. 10). 3. ndividual crws which have failed in finding fd at a particular day, may fllw ther crws when departing frm the rst next mrning. n this way rsts may functin as an infrmatin centre fr fd-finding (Ward & Zahavi 1973). t has been previusly suggested that this is nt likely t play an imprtant rle in my study area, except perhaps under very severe weather cnditins (Lman & Tamm 1980). There was n cngruence between rst-recruitment areas and winter hme-ranges, thus they cannt be cnsidered grup territries. This is hardly surprising as bth areas are likely t be determined by different factrs. Hmeranges can be expected t be primarily determined by fd-availability, whereas rst-recruitment areas seem t depend n the distributin f suitable rsting places, which in turn depends n the distributin f predatrs. Mst territrial crws in my study area had their winter hme-range restricted t part f a rst-recruitment area, whereas part f the flck crws,

74 SOCAL ORGANZATON HOODED CROW [Ardea 73 especially the juveniles, visited several rst recruitment areas in ne winter. The latter prbably were lw-ranking birds, which were frced t mve frm ne place t anther under the pressure f fd cmpetitn. 6. ACKNOWLEDGEMENTS Sam.Erlinge, Hans Kristianssn and Olf Liberg suggested mprvements t the manuscript. Assistance in the field was given by many peple, especially Thmas Madsen and Niklas Trnlund. The illustratins were prepared by Steffi Duwes. Jnathan Thrntn and R. van Halewijn helped me with the English language. thank them fr their assistance. Grants were received frm the Swedish Natural Science Research Cuncil (t Sam Erlinge). 7. SUMMARY A ppulatin f wing-tagged Hded Crws Crvus crnix was studied in suthern Sweden during the years 1974 79. nterest was fcused n aspects f their scial rganizatin.and behaviural eclgy. Only part f the ppulatin studed were breeding birds. Breeding pairs were territrial during spring when nn-breeding crws lived in flcks mainly in rather restricted areas. f a territrial crw died' its mate kept the territry and btained a new partner: usually within a mnth. Pairs which had recently established a territry had usually spent the previus breeding seasn in the flck-area that was clsest t their territry. t was als cmmn fr a pair t visit ccasinally an area in the year befre they established a territry there. Sme crws became territrial at the age f tw years but the majrity had nt acquired a territry at the age fthree. Sn:e pairs were usually fund in their territries thrughut wmter whereas thers were less restricted t it. This difference was prbably related t differences between territries in pssibilities t find fd in winter. Mst juveniles had ceased altgether t visit their natal territries by September in the year f hatching. bserved tw juveniles which t sme extent remained assciated with their natal territries thrughut their first winter. Juveniles fed in large flcks in places with cncentratins f fd during winter. Territrial crws were usually bserved in smaller flcks and fed in areas withut bvius cncentratins f fd. Older flck crws behaved intermediately in these respects. Mst crws frequented large ~mter rsts stuated up t several km frm the crws' daytime feedmg ranges. The crws assembled at gathering places befre rsting. They left these mre r less simultaneusly fr the rst. Smaller rsts, within tw km frm feeding places, were used in summer. Sme flck crws but n territrials migrated during winter. Mst migrants apparently wintered in Denmark. Annual survival was at least 92% fr territrial crws and 73% fr flck crws mre than tw years ld. 8. REFERENCES Abshagen, K. 1963, Uber die Nester der Nebelkrahen Crvus crne crnix. Beitr. Vgelk. 8: 325-338.' Bertram, B. 1978. Living in grups: Predatrs and prey. n: J. Krebs & N. Davies (eds.). Behaviural Eclgy. Blackwell, Oxfrd. Bhmer, A. 1976. Zur struktur der schweizerischen Rabenkrahenppulatin Crvus crne crne. Om. Beb. 73: 109-136. Brwn, J. L. 1974. Alternate rl,ltes t sciality in Jays with a thery fr the evlutin f altruism and cmmunal breeding. Am. Zl. 14: 63-80. Charles, J. 1972. Territrial behaviur and the limitatin f ppulatin size in the Crw, Crvus crne and Crvus crnix. Ph. D. thesis Aberdeen University, Granssn, G. & J. Lman. 1982. Des shting f breeding Crws influence Pheasant prductin? - An experiment -. Trans. ntern. Cngr. Crame Bi. 14: 331-334. Gyllin, R., H. Kallander & M. Sylven. 1977. The micrclimate explanatin f twn centre rsts f Jackdaws Crvus mnedula. bis 119: 358-361. Kalchreuter, H. 1971a. Untersuchungen an Ppulatinen der Rabenkrahe (Crvus c. crne). Jb. Ges. Naturkde. Wiirtemberg 126: 284-339. Kalchreuter, H. 1971b. Untersuchungen an der Krahenmassenfalle. Z. Jagdwiss. 17: 13-19. Lman, J. 1977. Factrs affecting clutch and brd size in the Crw, Crvus crnix. Oiks 29: 294-301. Lman, J. 1980. Reprductin in a ppulatin f the Hded Crw Crvus crnix. Hlarc. Ecl. 3: 26--35. Lman, J. & S. Tamm. 1980. D rsts serve as "infrmatin centers" fr Crws and Ravens? Am. Nat. 115: 285-289. Piczzi, N. 1971. Wing tags fr raptrs. Ring 68/69: 169 170. Skutch, A. F. 1935. Helpers at the nest. Auk 52: 257-273. Spray, C. J. 1978. Territrial behaviur f the Carrin Crw, Crvus crne L., in relatin t fd supply: An experimental study. Ph. D. thesis. Aberdeen University. Swingland,. R. 1977. The scial and spatial rganizatin f winter cmmunal rsting in Rks (Crvus frugilegus). J. Zl. Lnd. 182: 509-528. Tmpa, F. S.1975. A preliminary investigatin f the Carrin Crw Crvus crne prblem in Switzerland. Orn. Beb. 72: 181-198. Ward, P. & A. Zahavi. 1973. The imprtance f certain assemblages f birds as "infrmatin-centres" ft fd-finding. bis 115: 517-534. Wittenberg, J. 1968. Freilanduntersuchungen zur Brutbilgie und Verhalten der Rabenkrahe (Crvus c. crne). Zl. Jb. Syst. 95: 16--146. Wlfenden, G. E. & J. W. Fitzpatrick. 1978. The inheritance f territry in grupbreeding birds. Bi Science 28: 104-108. 9. SAMENVATTNG n the prvincie Skane, zuidelijk Zweden, werd van 1974 tt 19790nderzek gedaan in een gebied van ruim 20 km 2 aan de 10kale Bnte Kraai ppulatie. Vr lover vgels knden wrden gevangen werden ze vrzien van ringen en vleugelmerken. n enkele gevallen kn gedurende enige dagen het bewegingspatrn van individuele vgels geregistreerd wrden m.b.v. zendertjes. Het veldwerk bestnd vral uit het dcumenteren van het verspreidingspatrn en het gedrag van gemerkte vgelsgedurende een aantal peenvlgende jaren, waarbij gebruik werd gemaakt van twee vaste aut-rutes in en rndm het studiegebied.

1985] SOCAL ORGANZATON HOODED CROW 75 Onderscheid wrdt gemaakt tussen territriale- en grepsvgels: territrialen wrden, tenminste in de peride april-juni, praktisch uitsluitend gezien binnen een gebied van beperkte mvang, en slitair f met hguit een andere vgel. Aile vgels welke (in april-juni) niet aan dit criterium videden werden beschuwd als grepskraaien; deze kunnen wrden aangetrffen in een grter areaal dan territriale vgels, maar tch was hun areaal tamelijk beperkt, althans in vrjaar en zmer, en wei tt plaatsen met gecncentreerde vedselbrnnen ("CFP" in de tekst; afvalhpen, mestvaalten, aardappel wintervrraden, pen varkens- en keiestallen). Arealen van territriale- en grepskraaien 6verlappen 's zmers sms enigszins, terwijl arealen van individuele grepskraaien nderling dan in hge mate verlappen. Grepskraaien resten in vrjaar en zmer p krte afstand van hun ferageergebied en deze vgels gebruiken in de lp van een seizen meestal meerdere restplaatsen. n vrjaar en zmer werden grepskraaien sms enigszins buiten het grepsareaal aangetrffen, en waren dan slitair f met een partner. Deze vgels, die meestal uder dan drie jaar zijn, wrden beschuwd als verkenners: een jaar later breden zulke vgels veelal in het eerder verkende gebied, maar niet aile bredvgels hebben een lange verkennersperide achter de rug. Paarvrming kan plaatsvinden tussen twee grepskraaien, maar treedt veelvuldiger p binnen een territrium. Bnte Kraaien blijken in hge mate truw te zijn aan hun territrium zwel als aan hun partner. Een kwart van aile gemerkte territriale kraaien verbleef het gehele jaar binnen het territrium; er zijn aanwijzingen dat dit vral z is bij vgels met territria met een relatief ruim, betruwbaar vedselaanbd. lange vgels verliezen in de peride september-nvember de binding met de uders en het uderlijk territrium. n twee gevallen werd echter geregistreerd dat jnge vgels geasscieerd bleven met het territrium van hun uder, in hun tweede en derde kalenderjaar. n de winter wrden territriale kraaien - vr zver ze de binding met hun territrium tijdelijk hebben pgegeven - vral in kleinere grepen aangetrffen, jnge vgels en grepskraaien in grtere grepen. Beide laatstgenemde categrieen vgels werden in de winter naar verhuding k veel meer p CFP waargenmen dan territriale kraaien. 's Winters frequenteren de meeste kraaien in het studiegebied enkele grte, traditinele restplaatsen gelegen in stukjes sparrenbs. Vgels arriveren hier massaal een half um na znsndergang, kmend vanaf een f meer vrverzamelplaatsen. Deze laatste zijn veelal eveneens traditineel, en liggen p krte afstand van CFP, en tt p enige kilmeters afstand van de grte restplaatsen. Bij vrverzamelplaatsen behren bepaalde ferageerarealen welke nderling nauwelijks verlappen. Vral jnge vgels wisselen in de lp van de tijd ngal eens van vrverzamelplaats en het daarbij behrende ferageergebied. De meeste jnge vgels en grepskraaien waren niet in het studiegebied aanwezig tussen ktber en maart. Ringvndsten suggereren dat deze vgels elders in zuid-zweden en in Denemarken verwinteren, en dat Finse vgels 's winters in het studiegebied vrkmen. De cijfers berekend vr verleving liggen p minimaal 92% vr territriale kraaien, en p minimaal 75% vr grepskraaien. n de discussie wrden de Zweedse gegevens vergeleken met resultaten van srtgelijk nderzek in West-Duitsland, Schtland en Zwitserland en wrden deze geinterpreteerdin een algemener, theretisch kader. - R. v. H.