Crested Honey Buzzard Pernis ptilorhynchus orientalis is a long-distance migrant that breeds across southern Siberia to Sakhalin and Japan, and winters mainly in South-East Asia, Indonesia and the Philippines (Higuchi et al. 2005, Wells 2010, Orta et al. 2018). In southern Siberia, at its western limit, the breeding range partially overlaps with that of European Honey Buzzard P. apivorus (Stepanyan 1983, Ferguson-Lees & Christie 2001). Six subspecies of Crested Honey Buzzard are recognised, but only orientalis is a migrant (Orta et al. 2018). The species was recorded for the first time in the Western Palearctic at Borçka, north-east Turkey, in September 1979 (Laine 1996) and then at Eilat, Israel, in May 1994 (Shirihai 1994). P. ptilorhynchus is now considered regular in small numbers on passage through Israel, mainly at Eilat, which is a major passage bottleneck for European Honey Buzzards (Shirihai 1994). Fifteen to 20 individuals are recorded every spring, mainly in May, with 5–12 in autumn, mainly in mid September (Babbington & Campbell 2016). There have also been many recent records of P. ptilorhynchus at Batumi (Georgia) where the first to be officially accepted was in autumn 2007 (Abuladze 2013). Since then, the species has been identified annually at Batumi, with a total of 163 records until 2018 and a max. 51 birds in 2013 ( https://www.batumiraptorcount.org/migration-count-data#annual-totals).
It is presumed that those Crested Honey Buzzards recorded in Israel, and elsewhere in the Middle East in spring, joined flocks of P. apivorus wintering in Africa (Ferguson-Lees & Christie 2001). Many records have been suspected to be potential hybrids with P. apivorus (Babbington & Campbell 2016).
In Europe, Crested Honey Buzzard has been fully documented just twice: on Cyprus in October 2012 (Harrison 2014) and in Italy on 18 May 2011, at the Strait of Messina between Sicily and the mainland (Scuderi & Corso 2011).
On 2 May 2018, near the village of Alyfanta (39°06′04″N, 26°31′45″E) on Lesvos, 4 km from the largest urban centre on the island (Mytilini), we observed an adult male Crested (or Crested-type) Honey Buzzard. It was watched as it soared, gradually gaining height, for c.3 minutes at a distance of c.150 m from the observers (SPZ, YZ). SPZ managed to take 12 photographs of the bird (Figs. 1–3). Subsequently, it headed south-west and was not seen again. Identification was made by the authors, following the observation, based on the photographs. This is the first documented record for Greece and the third to be accepted for Europe.
P. ptilorhynchus can be easily confused with P. apivorus. In this case, the bird's structure was obviously different: heavier bodied, slightly larger and bulkier (more eagle-like) compared with P. apivorus. Furthermore, its wings appeared broader and the tail shorter than that of European Honey Buzzard. The absence of the diagnostic carpal patch of P. apivorus, the six clearly fingered primaries protruding from the trailing edge of the wing, and the dark tail with a broad white bar in the centre of the undertail, are diagnostic features of P. ptilorhynchus (Ferguson-Lees & Christie 2001, Svensson et al. 2009, Forsman 2016). The inner secondaries show two well-defined bars while a third bar is visible on the inner primaries and outer secondaries. The head is grey with a pale throat, bordered by a dark ‘gorget’ that contrasts with the pale sandy-ochre underparts and underwings. This plumage is commonest in adult males according to Forsman (2016).
Alternatively, the possibility of hybridisation between P. apivorus and P. ptilorhynchus (Faveyts 2011, Forsman 2016) and some structural and plumage features that do not match perfectly with P. ptilorhynchus made us consider the possibility that the bird was a potential hybrid. Specifically, the bird shows a quite rounded wingtip, rather than the blunt tip of Crested Honey Buzzard (p5 is not clearly longer). Furthermore, the wing is typically more rectangular in Crested Honey Buzzard, but in the Greek bird appears broadest at the carpal joint, tapering towards the body and tip. The intermediate underwing and tail barring, and possible hint of a darker carpal area, typical of supposed hybrids, reinforce this hypothesis (D. Forsman in litt. 2018).
However, the lack of genetic research into the hybridisation question, in parallel with the fact that P. apivorus and P. ptilorhynchus are (a) not known to form mixed pairs in the region of overlap (Mosquitin 1973, Kislenko 1974, Stepanyan 1983), and (b) are not even each other's closest relatives (Gamauf & Haring 2004), raises doubts as to whether it is justifiable to discuss hybrids between the two species. The unquestionable similarity of P. ptilorhynchus to P. apivorus and, for most European observers, the lack of understanding of their distinguishing features and especially their morphological variability, lead us to suspect that P. ptilorhynchus may be a more frequent vagrant to parts of south-east Europe than is currently perceived. More attention should be paid along the major raptor passage flyways in the Western Palearctic to better understand the western limit of Crested Honey Buzzard's migration route. Furthermore, genetic analysis is critical to provide a more solid basis for discussing hybridisation between these two Pernis species.
We thank Killian Mullarney and Dick Forsman for their help in identifying the bird, Rob Bijlsma and Andrea Corso for their very useful suggestions and comments on the submitted draft, and Apostolis Christopoulos, Eleni Galinou, Fanis Theophanopoulos and the members of the Hellenic Rarities Committee, headed by George Handrinos, for accepting the record.