Over 3600 nest cards of raptors were submitted in 2004 (Appendix 1), covering 10 species. The weather in 2004 was on average warm, but rather wet (especially in June-July), the summer index being 67.8 (a warm summer, based on temperatures in May-August). The frost index of the preceding winter was 7.8 (mild winter, based on temperatures in November-March 2003/2004). Common voles Microtus arvalis on average were recovering from a low in 2003 (Fig. 1), but numbers were still poor in some regions; wood mouse Apodemus sylvaticus numbers were improving as well (Fig. 2). Rabbits Oryctolagus cuniculus were still well below numbers in the early 1990 s (i.e. before the outbreak of Rabbit Viral Haemorrhagic Disease; Bijlsma 2004 a), with some encouraging signs of recovery. Bird numbers probably did not change much as compared with the past few years. Social hymenoptera had an average year (index 3 on a scale of I -5), with a slow start in May (prospecting queens encountered adverse weather conditions) and an early demise of states during late July (Vespula germanica) and in the course of August ( V. vulgaris). The sightly improved prey base, as compared with 2003, was reflected in higher numbers of nestling raptors ringed (Table 1). European Honey-buzzard Pernis apivorus: clutch size was 2x 1 and 24x 2, brood size 14x 1 and 20x 2. Mean onset of laying was 29 May, however with a very large interval between first laying (12 May, never before clutch initiation in The Netherlands was found to be earlier than 15 May, Fig. 3) and last laying (19 June). Normally, pairs start breeding synchronously within a period of 10-20 days. In 2004, the last two starters may have been repeat layings as the interval between them and the rest amounted to 12 and 14 days. Interestingly, laying in the period 1997-2004 (n=193) was consistently earlier than during 1971-2004 (n=236, mostly before collected 1990). This was visible in start, peak and termination of laying (Fig. 3), presumably an effect of climate change. Of 36 nests checked, 24 were built by Honey-buzzards, 5 by Buzzards, 3 by Goshawks, 3 by Carrion Crows and 1 by Sparrowhawk. Failures were mostly due to predation of nestlings (3x) and/or parents (3x), mainly by Goshawks. Eurasian Marsh Harrier Circus aeruginosus: mean onset of laying was 22 April (SD=9.70, n=93, range 5 April-18 May; Appendix 2, Table 3). Clutch size averaged 4.46 (n=99, SD=1.02; Appendix 3), mean number of fledglings/successful pair 3.04 (n=103, SD=1.20; Appendix 4). Presumably, local variations in breeding parameters were mainly caused by variations in food supply (notably common voles, an important prey species; Table 5). All surviving nestlings in 74 nests were sexed, showing an almost equal sex ratio with 49.8% males (Table 4). Over the years, a biased sex ratio has been prevalent (Table 4: 52.5% males in 771 nests with 2522 nestlings). Apart from common voles, also rabbits and pheasant chicks were particularly important as prey species (Table 5), but biased sampling (mostly late nestling stage) may have under- or overestimated the relative importance of these and other prey species. Hen Harrier Circus cyaneus: the steady decline of the Dutch population resulted in the initiation of a special study into their breeding and feeding ecology on the Wadden Sea Island in 2004, the only region where the species is still surviving in some numbers (between 60 and 80 pairs). The data are summarised for the three most important islands, i.e. Texel, Terschelling and Ameland (Table 6). On average, laying started on 2 May (range 21 April-14 May, Appendix 2). Mean clutch size was 4.7 eggs (SD=1.0, n=31; Appendix 3), mean number of fledglings/successful pair 2.4 (SD=1.3, n=28; Appendix 4). On Texel and Terschelling, many eggs did not hatch. The sex ratio on 29 nests was 37 males and 30 females. On Ameland two males were polygynous: one male attended two females (resp. 3 and 2 chicks fledging), another three females (resp. 5, 5 and 1 chicks). Individual recognition of males was based on plumage characteristics (Fig. 4) in combination with behaviour and other typical features (carrying a ring or not, on which leg). On Texel, at least 4 out of 16 males were bigamous. Many breeding birds on the Wadden Sea Islands carry a ring. Ten out of eleven recoveries of nestling-ringed Hen Harriers on Texel turned out to have returned to within 2 km of their natal site, indicating poor dispersive qualities. This may be one of the bottlenecks experienced by Hen Harriers breeding on the Wadden Sea Islands: a closed population (little exchange with other populations) with poor reproductive output. The latter is probably linked with declining food supply (rabbits and pheasants, both important prey species, have declined precipitously). The impact of local increases in the numbers of Marsh Harrier, Common Buzzard and Northern Goshawk on Hen Harriers is not yet known, but may be another factor negatively influencing Hen Harriers. Montagu’s Harrier Circus pygargus: 39 pairs were recorded for The Netherlands, of which the majority nested in the province of Groningen (29), raising 85 fledglings. Breeding is still clustered in few regions. A detailed account is given by Koks et al. elsewhere in this Takkeling. See also: www.grauwekiekendief.nl Northern Goshawk Accipiter gentilis: with the colonisation of Ameland in 2004 (one pair, both immature birds, I fledgling), the colonisation of the Wadden Sea Islands – started in 1997 on Texel – is now completed (Table 7). A similar process is taking place in Zeeland (southwestern Netherlands), till recently without breeding Goshawks. The average start of laying was 2 April (SD=7,84, n=277, range 13 March-28 April; Appendix 2), with clear regional variations (Table 9). On average, breeding started earlier in the southern provinces, and later in recently colonised regions and in the north. On average, mean clutch size was 3.30 (SD=0.75, n=242; Appendix 3), mean number of fledglings/successful pair 2.78 (SD=0.78, n=341; Appendix 4). Among 277 nests, sex ratio at ringing was in favour of males (54.6%; Table 10) Among 39 identified nest failures, 23 were deliberately caused by humans. Local declines since at least the early 1990s stabilised at a lower density in recent years. In the well-studied population of western Drenthe (Table 8), 2004 showed by far the poorest breeding performance since 1990, with completed clutches as small as 1 (2x) and 2 eggs (5x), and a reproductive output of 6 fledglings by 10 egg-laying pairs. This trend is caused by a serious decline in food supply, and – among many other impacts – results in an increasing predation pressure on other raptor species. Elsewhere in The Netherlands, prey collection in the breeding season showed that in most areas few prey species are responsible for the majority of prey consumed (Appendix 9), i.e. pigeons, corvids and starlings, indicating better feeding conditions. Sparrowhawk Accipiter nisus: mean onset of laying was 2 May (19 April-23 May, SD=6.98, n= 183; Appendix 2), with distinct regional variations (Table 11). Mean clutch size was 4.47 (SD=4.86, n=152. Appendix 3), mean number of fledglings/successful pair 3.97 (SD=1.22, n=231. Appendix 4). The secondary sex ratio was malebiased (Table 12: 345 males, 315 females, 167 nests). Common Buzzard Buteo buteo: on average, egg laying started on 6 April (SD=7.47, n=675; Appendix 2). Clutch size averaged 2.50 eggs (SD=0.66, n=447; Appendix 3), brood size 2.02 fledglings/successful pair (including 2 with 5 and 1 with 6 chicks, the latter presumably by two females; SD=0.78, n=840; Appendix 4). This was slightly better than during the vole-crash in 2003, and was recorded in most regions except a few on sandy soil in the central and northern Netherlands (Table 13). Secondary sex ratio was female-biased: 188 males and 219 females on 266 nests (Table 14). Among prey remains found on nests, 52 species of bird , 17 species of mammals, 5 species of reptiles, 4 species of amphibians and 4 species of fish were recorded. Important prey species were voles (mainly common vole), moles, rabbits (locally only) and hares, starlings, corvids, thrushes, pigeons and pheasants (locally) (Appendix 10). Stints of adverse weather in June and July caused high mortality among nestlings, and negatively impacted the condition of surviving chicks (example in Fig. 5). Osprey Pandion haliaetus: summering Ospreys were recorded in the Oostvaar- dersplassen. Randmeren and in the southwestern Netherlands, sometimes accompanied with short bouts of nest-building. However, active nests were not recorded, despite specific searches in potential breeding areas. Eurasian Kestrel Falco tinnunculus: onset of laying averaged 18 April, with a wide variation from 25 March through 22 June (SD=12.16, n=407; Appendix 2) and between regions (Table 15). Mean clutch size was 4.91 (SD=5.50, n=491; Appendix 3), mean number of fledglings/successful nest 4.70 (SD=4.70, n=570, Appendix 4). Most Kestrels recorded were breeding in nest boxes (94%), of which 92.7% fledged at least one chick. Nesting success of 21 Kestrels breeding on open nests of corvids was slightly lower (i.e. 85.7%, n=21). Eurasian Hobby Falco subbuteo: mean start of laying was 8 June (range 25 May-223 June, SD=6.27, n=40; Appendix 2). Mean clutch size was 3.00 (SD=0.58, n=12, only 3 started in May, Appendix 3), mean number of fledglings/successful nest 2.34 (SD=0.74, n=83; Appendix 4). Presently, most pairs are recorded in open farmland where many pairs nest on electricity pylons and in poplars (Table 17, see relatively high numbers in Friesland, Noord-Brabant and Limburg). Secondary sex ratio in 10 nests was in favour of females: 9 males and 14 females (sexes identified by body mass and vocalisations) (Table 18). Food remains found at nests show a preponderance of swifts, hirundines and house sparrows (Table 19). As compared with prey choice in 1970s, several aspects of the feeding ecology have changed drastically. First of all, skylarks and tree sparrows are rarely caught nowadays; both were common prey species in the 1970s. Secondly, the ratio swift versus barn swallow switched from 27:73 in the 1970s to 63:37 in 2004. Both changes reflect the serious declines of skylark, barn swallow and tree sparrow in Dutch farmland; densities are very low, and in many regions these species have virtually disappeared. Peregrine Falcon Falco peregrinus: in 2004, the Dutch population increased again, from 17 territorial pairs in 2003 to 27 pairs in 2004 (of which 20 pairs laid eggs. Overall, 29 nestlings were raised to fledging (15 males and 11 females in 9 fully sexed nests; Table 20). Most nestlings were also colour-ringed. Mean onset of laying was 18 March, ranging from 2 March through 4 April. Mean clutch size was 3.3 (N=7; Table 20), mean number of fledglings/successful pair 2.9 (N=10; Table 20). Despite the ongoing increase in population size, fledgling production has been levelling off since about 2000 (Table 21), most likely a density-dependent effect. Newcomers mostly settle withing short distances of local pairs (400-1700 m), resulting in persistent territorial conflicts, delays in egg-laying and failed breeding attempts.