Bulletin of the African Bird Club 2 8 MAR 2006 f - JEO! IA82D Vol 13 No 1 What is Pogoniulus makawaP White-winged Flufftail in Ethiopia Status of Shoebill in Malagarasi Sierra Leone Prinia in Fouta Djalon, Guinea Golden Nightjar vocalisations from central Mali Plain Swift breeding biology on Gran Canaria Ring Ouzels wintering in Atlas Mountains, Morocco Rose-coloured Starling in Ethiopia First record of Bridled Tern for The Gambia Grand-due du desert au Burkina Faso Blackcap in Mozambique Mottled and Alpine Swifts in Niger Picatharte du Cameroun au Congo-Brazzaville Pink-backed Pelican in Madagascar First record of Kermadec Petrel for Seychelles Lake Bedo— a little- known wetland hotspot in Madagascar ISSN 1 352-481 XISSN 1 352-481 X ^ African Bird Club The African Bird Club aims to: • provide a worldwide focus for African ornithology • encourage an interest in the conservation of the birds of the region • liaise with and promote the work of existing regional societies • publish a twice-yearly colour bulletin • encourage observers to visit lesser known areas of the region • encourage observers to actively search for globally threatened and near-threatened species • run the ABC Conservation Programme Registered Charity No 1053920 ABC particularly wishes to thank its Corporate Sponsors for their invaluable financial support in 2006: Avian Adventures, Avifauna, Birding Africa, Birdquest, Safariwise Namibia, Sunbird, Tropical Birding, WildSounds, Wildwings and Zeiss. ABC Council John Caddick (Treasurer), Julie Childs (co-opted), Elaine Cook, Moira Hargreaves, Bill Quantrill, Geoff Randall (Secretary, co-opted), Claire Spottiswoode, Neil Thomas (co-opted) and Richard Webb (Chairman). Bulletin Editorial Board Chairman of the Board: Richard Webb Managing Editor: Guy Kirwan Assistant Editor: Ron Demey David Allan, Chris Bowden, Callan Cohen, Lincoln Fishpool, Peter Lack, Pete Leonard (Graphics Editor), Jeremy Lindsell, Roger Safford, Steph Tyler and Richard Webb ABC Membership Membership is open to all. Annual subscription rates are: Individual Europe & Africa: UK£18 Rest of the World: UK£20 Family Europe & Africa: UK£21 Rest of the World: UK£23 Student Europe & Africa: UK£10 Rest of the World: UK£12 Supporting UK£30 minimum Life UK£350 To join or for further details please visit the ABC web site (where there are secure online payment facilities) or write to the Membership Secretary — see contact information below. ABC Website http : // www. africanbirdclub.org Contact ABC African Bird Club, c/o BirdLife International, Wellbrook Court, Girton Road, Cambridge CB3 0NA. E-mail: info@africanbirdclub.org • website: http://www.africanbirdclub.org Further information can be obtained directly from individual Council members by writing to them at the Club’s postal address, or by e-mail as follows: Chairman Richard Webb chairman@africanbirdclub.org Secretary Geoff Randall secretary@africanbirdclub.org Treasurer John Caddick treasurer@africanbirdclub . org Bulletin Editor Guy Kirwan editor@africanbirdclub.org Conservation Officer Steph Tyler conservation@africanbirdclub.org Information Officer Keith Betton info@africanbirdclub.org Membership Secretary Bill Quantrill membership@africanbirdclub.org Sales Officer Moira Hargreaves sales@africanbirdclub.org Rutland Officer Neil Thomas rutland@africanbirdclub.org Representatives Julie Childs reps@africanbirdclub.org The Bulletin of the African Bird Club The Bulletin of the ABC provides a forum for news, letters, notices, recent publications, expedition results, reviews and interim publication of studies on African birds by contributors from throughout the world. Publication of results in the Bulletin of the ABC does not preclude publication of final results as journal papers either by the ABC or elsewhere. No material should, however, be submitted simultaneously to the Bulletin of the ABC and to any other publication. Brief notes for contributors appear elsewhere in this Bulletin and further details are available from the Editor (editor@africanbirdclub.org). ©2006 Copyright African Bird Club and contributors. Quotations should carry a full acknowledgement. No part etc may be reproduced, copied or stored in a retrieval system without the prior written permission of the Club or authors. Contents Bull ABC Vol 13 No 1 News & Comment 5 Club News Compiled by Bill Quantrill 8 Requests 8 Corrigenda 9 Africa Round-up Compiled by Ron Demey and Guy Kirwan 96 Recent Reports Compiled by Ron Demey 110 Reviews 1 1 2 Notes for Contributors Front cover plate Egyptian Plover Pluvianus aegyptius by Nigel Blake Illustrations Claudia Donati, Pete Leonard Photographs Clive Barlow, John Caddick, Nigel Collar, Klaas-Douwe Dijkstra, Frangoise Dowsett-Lemaire, Cas Eikenaar, Lars Dinesen, Wouter Favyets, Armine Flitti, Alain Fosse, Eduardo Garcla-del-Rey, John High, Flemming Jensen, Frederic Jiguet, Tasso Leventis, Alistair Mclnnes, Martim Melo, Laura Payne, Yvan Perre, Bill Quantrill, Colin Ryall, Valery Schollaert, Per Smitterberg, Claire Spottiswoode, Kevin Vangs, Edwin Winkel 18 28 37 45 49 56 60 75 77 78 80 82 84 86 88 91 2 8 MAR 2036 PURCHASES Tf#-|3 USHASY Features What is Pogoniulus makawai ? N. J. Collar and L. D. C. Fishpool White-winged Flufftail Sarothrura ayresim Ethiopia: notes on habitat, densities, morphometries, nests and eggs, and associated waterbirds David G. Allan, Alistair M. Mclnnes and Mengistu Wondafrash Status of Shoebill Balaeniceps rex in Malagarasi, Tanzania Lars Dinesen and Marc Baker Sierra Leone Prinia Schistolais leontica in the Fouta Djalon of Guinea, its song, distribution and taxonomic status C. R. Barlow, R. B. Payne, L. L. Payne and M. D. Sorenson First reliable sound recording of Golden Nightjar Caprimulgus eximius, in the rocky hills of central Mali Frangoise Dowsett-Lemaire and Robert J. Dowsett Notes on the breeding biology of Plain Swift Apus unicolor on Gran Canaria, Canary Islands Eduardo Garcla-del-Rey Some factors affecting foraging and habitat of Ring Ouzels Turdus torquatus wintering in the Atlas Mountains of Morocco Colin Ryall and Kevin Briggs First record of Rose-coloured Starling Sturnus roseus for Ethiopia and sub-Saharan Africa Valery Schollaert First record of Bridled Tern Sterna anaethetus for The Gambia John High Premiere mention du Grand-due du desert Bubo (bubo) ascalaphus pour le Burkina Faso Guilhem LesaJJre et Yvan Perre First records of Blackcap Sylvia atricapilla for Mozambique Martim Melo, Rita Covas and Klaas-Douwe Dijkstra First records of Mottled Swift Tachymarptis aequatorialis and Alpine Swift T. melba for Niger Kim Diget Christensen, Anders P. Tottrup and Flemming Pagh Jensen Premiere observation du Picatharte du Cameroun Picathartes oreas au Congo-Brazzaville Victor Mamonekene et Frederic Lambert Bokandza-Paco First Pink-backed Pelican Pelecanus rufescens sightings in Madagascar since 1960 Martin Mwema and Felix Razafindrajao First record of Kermadec Petrel Pterodroma neglecta for Seychelles Cas Eikenaar and Adrian Skerrett Lake Bedo— a little-known wetland hotspot in Madagascar H. Glyn Young and Felix Razafindrajao Bull ABC Vol 13 No 1 (2006)- 1 WildSounds eGuides Mobile versions of the most popular field guides. They are ideal companions to the book and have electronic features that will add to your enjoyment of birding. Unless otherwise stated, the entire collection of pictures, sound & text is stored on a fast high-quality 256Mb Memory Card. No PC necessary. Just insert the card, run the install files and go birding. None of the PDA ’s internal memory is used for storage. eGuides require a PDA running Windows Mobile (or Pocket PC) 2003 or 2005 (we recommend HP iPAQ hx2190 Pocket PC). eGuides can be installed on one PDA only. postfree Wildlife Boohs, Audio & Multimedia Guides Southern African Bird Sounds Gibbon 900 announced species; almost all the species occurring in the region. On the CD version the sounds are each announced but are in groups of 10 species per indexed track. 7.5 hours □ 6 Cassette Set £55^5 £49.95 (£42.‘ □ 6 CD Set £57.09 £49.99 (£42.54) eBirds: Birds of Southern Africa for Pocket PC v2 A mobile version of the best-selling Sasol Birds of Southern Africa field guide. Features: Pictures of all 954 bird species found in Southern Africa. Distribution maps and text. Over 540 sounds. A logging facility that allows you to easily keep records of your bird sightings. □ Software on 256Mb SD Card £99r95 £89.95 (£76.55) 4 Kingdon eGuide to African Mammals ' y{ Based on Kingdon Pocket Guide to African Mammals by Jonathan Kingdon. An essential electronic field guide for those visiting and living in Africa with an interest in wildlife. The coverage in this guide is the most comprehensive currently available. Every illustration and all the text from the book has been included in the program. Easy-to-use interface and fast searches allow you to sort through the 460+ species. □ eGuide on 256MB SD Card £99.95 £79.95 (£68.04) Collins Bird eGuide for Pocket PC A mobile electronic version , of the best-selling field guide i Collins Bird Guide by Mullarney, Svensson et al. Features: Pictures of over 750 birds found in Britain & Europe. Distribution maps and text. Over 780 sounds of 473 species. Includes a logging facility. □ eGuide on 256MB CF Card £105.95 £93.95 (£79.96) □ eGuide on 256MB SD Card £99:95 £89.95 (£76.55) eSnakes of Southern Africa Comprehensive electronic guide to the 151 snakes indigenous to southern Africa. □ eGuide on 256MB SD Card £99:95 £79.95 (£68.04) PDA for illustration purposes only eWildlife of Southern Africa Pocket PC version of the popular Wildlife of Southern Africa field guide by Vincent Carruthers. Information on over 2000 species with more than 1600 images. □ eGuide on 256MB SD Card £99:95 £79.95 (£68.04) Books Audio & Multimedia Bird Sounds of Europe & North-west Africa Roche & Chevereau Songs and calls of 483 species and sub-species. Species are in systematic order and are indexed by track number only and not interrupted by announcements. Plays for almost 12 hours! Birds of Africa south the Sahara Sinclair & Ryan (with Chrii & Hockey) More than 3000 images i some 340 plates, of nea 2100 species. Includes Socotra and islands in I i Gulf of Guinea, but excl Madagascar, Mauritius, | Reunion and Seychelles. 712 pages. □ Softback £29.99 7 have been itching to get my hands on this new 10-CD set.’ - Birdwatch □ Boxed 1 0 CD Set £58t95 £49.95 (£42.51 ) DVD-Video Guide to the Birds of Britain & Europe J ~~~ — (including N Africa and Middle East) Doherty A stunning set, at a very reasonable price. Covers 650 species with high quality moving footage and expert commentary. More than 13 hours of playing time! . . certainly good value for money’ - Steve Hay, Birdwatch □ 6 DVD-Video Set £44.95 (£38.26) Bird Recordings from Ethiopia Smith 72 announced species. Home-made quality. □ Cassette £9.50 (£8.09) Field Guide to the Birds of the Atlantic Islands Clarke (May 2006) Covers all resident, migrant and vagrant species found in Macaronesia (Azores, Cape Verde, Canaries & Madeira) . Over 450 species are illustrated with full details of all plumages and major races likely to be encountered. 320 pages. □ Softback £29.99 £25.99 Field Guide to Birds of E; Africa Stevenson & Fanshaws Helm Field Guide covering Ke Tanzania, Uganda, Rwanda a Burundi. 1388 species illustra! on 287 superb colour plates. 1 text plus distribution map and! illustrations for each species are on fac pages. 632 pages. □ Softback £24.90 £22.99 Bird Song of The Gambia & Senegal Barlow, Hammick & Sellar Vocalisations of 265 species and subspecies. Indexed but not announced. □ 3 CD Set £27.49 (£23.40) Bird Sounds of Madagascar, Mayotte, Comoros, Seychelles, Reunion, Mauritius Huguet & Chappuis Voices of 327 bird species. All recordings are extensively documented in the 115-page booklet (in French and English). □ 4 CD set £60:09 £46.95 (£39.96) Pocket Guide to the Reptiles & Amphibians of East Africa Spawls, Howell & Drewes (April 2006) This new portable guide ha been partly adapted from the popular and highly acclaimed Field Guide to tf Reptiles of East Africa. It covers the most prominent 150 reptiles & 80 amphibians found in the region. 144 pa □ Softback £14.99 Use this web-sili and do your conservl http:/ / wm.wildsound^muk/ index.htm'trtf An appeal from WildSfl Do not use the playback of songs or calls to attract a rare or endangered bird. It could feel severely threatened & desert its nest or tel of A Field Guide to Birds of Gambia & Senegal Barlow, Wacher & Disley Almost 570 species are illustrated on 48 colour plates. 408 pages. □ Softback £24:99 £22.99 d Birds of ian Ocean z Langrand s all the 359 encountered of Madagascar, Reunion, js, the Seychelles Comoros. 184 ock £17:99 £15.99 iraphic Guide to »f the Indian Ocean Langrand & Andriamialisoa 06) A selection of the most jly encountered bird species of scar, the Seychelles, the Comoros, Vlascarenes. 128 pages, ack £14.99 £12.99 >f Madagascar f Hawkins >ur photos, several lates and numerous 28 pages, jack £29:99 £26.99 jn Pocket Guide to African als Pocket ID guide to all the African land mammals by one of the world’s leading naturalists. The illustrations are in an easy-to-use plate format and are placed opposite the text. 108 colour plates and over 1 250 maps. 272 pages. fn Softback £16.99 A Birdwatchers’ Guide to Morocco Patrick & Fedora Bergier 2004. Revised, expanded 2nd edition of this essential site guide. 172 pages. □ Softback £16.75 Raptors of the World: A Field Guide Ferguson-Lees & Christie Uses all of the plates from the Helm ID Guide Raptors of the World, with concise, revised text on facing pages, to create a conveniently-sized, lightweight field reference covering all 330 raptor species. 320 pages. □ Softback £19.99 Sasol Birds of Southern Africa Sinclair, Hockey, Tarboton 3rd edition. Over 200 colour plates and distribution maps. 445 pages. □ Softback £19:99 £17.99 WTWB in Uganda Rossouw & Sacchi 1 998. Still one of the best where to watch books yet published for a single country. Colour photos & maps throughout. 114 pages. □ Softback £44:99 £9.99 Field Guide to the Birds of Western Africa Borrow & Demey Covers all 1285 species found in the region with distribution maps. 512 pages. □ Softback £20.00 £25.99 WTWB World Cities Milne (June 2006) A new site guide to urban areas and prime sites within easy reach of 60 major cities. 480 pages. □ Softback £16.99 Important Bird Areas in Zambia Leonard 220 pages. □ Softback £17.00 The Complete Guide to Antarctic Wildlife Shirihai & Jarrett Every bird & mammal species occurring in the area is described in detailed species accounts illustrated with plates, maps & colour photos. 512 pages. □ Hardback £32^€< ROBERTS birds jthern African Birdfinder: ere to find 1400 bird species in southern Africa an Cohen & Claire Spottiswoode, assisted by Jonathan Rossouw rch 2006) The ideal companion to all the local bird field guides. Covers over 200 top ing sites and associated birds across South Africa, Namibia, Botswana, Zimbabwe, Indian Ocean islands and the little-documented, but increasingly popular areas of lola, Mozambique, Zambia & Malawi. A guide to finding the region’s top 100 birds an annotated checklist are included. 448 pages. ioftback £49 .99c Roberts Birds of Southern Africa VIIth Edition Hockey, Dean & Ryan (July 2006) This amazing tome was so popular in South Africa that it is being reprinted. We should have stock in late July 2006. 1296 pages. □ Hardback £11 0.00 Delivery Charges United Kingdom All Mainland UK orders (excluding parcels over 2 Kg to Scottish Highlands & Islands, Channel Isles & N Ireland, etc) are postfree, dispatched within 5 working days of receipt, by 2nd Class mail or courier. If you prefer your order to be sent 1st Class please add 10% of order value (min £3.00, max £9.00). Deliveries outside the UK Please visit our web-site: www.wildsounds.com for a precise delivery charge. WildSounds, dept A1 1 , ] CROSS STREET, SALTHOUSE, NORFOLK, NR25 7XH, UK CREDIT CARD HOTLINE 1 call (uk) +44 (0)1263 741100 fax (uk) +44 (0) 1263 741838 web-site: www.wildsounds.com e-mail: sales@wildsounds.com CO "0 I! 3 hotographed on last year’s Birding Africa tour to Gabon Advertise in the Bulletin of the ABC All advertisements must be sent prepaid (cheques made payable to the African Bird Club) as high-resolution hard copy or on floppy disk/CD to: The Advertising Officer, African Bird Club, c/o BirdLife International, Wellbrook Court, Girton Road, Cambridge CB3 ONA, UK. If adverts are sent on floppy disk/CD we can accept QuarkXpress 6.5, Pagemaker 7, CorelDrawlO files, PDF or unformatted ASCII text files and uncompressed TIF graphics files. The current rates are as follows and are based on a print run of 1,400 copies. These rates are guaranteed for the September 2006 Bull ABC. Please address all queries to the Advertising Officer at the above address. African Bird Club Advertising Rates Black & white Full-page £95 (210 x 145mm) Half-page £60 (100 x 145mm) Quarter-page £40 (100 x 70mm) Eighth-page £25 (50 x 70mm) Colour Please contact the Club's Advertising Officer at the address given above left. Copy deadlines Spring Bulletin 1 5 January Autumn Bulletin 05 June Advertisement Bull ABC Vol 13 No 1 (2006) - 17 What is Pogoniulus makawai ? N. J. Collar a and L. D. C. Fishpooft Qu’est-ce Pogoniulus makawaP. Le Barbion a poitrine blanche Pogoniulus makawai a ete decrit en 1965 sur la base d’un seul specimen collecte dans la foret a Cryptosepalum au nord-ouest de la Zambie. A cause de l’absence d’observations ulterieures, la validite de l’espece a ete mise en ques- tion. Les arguments suivants on ete avances: (1) il pourrait s’agir d’un individu aberrant du Barbion a croupion jaune P bilineatus , (2) bien que la localite type appartienne a un milieu tres particulier, elle n’apparait pas comme un centre d’endemisme, (3) de nombreux observateurs ont cherche l’espece en vain. Les auteurs presentent toutefois treize caracteristiques par lesquelles P makawai differe de P bilineatus , un degre de difference qui, selon eux, ne peut etre attribue a une aberration. Du reste, le structure du milieu n’est pas du tout uniforme et les efforts pour redecou- vrir P. makawai a la localite type et dans ses environs, bien que considerables, ne peuvent etre con- siders comme exhaustifs. De vastes etendues de foret a Cryptosepalum , un habitat dans lequel il est difficile de travailler, n’ont jamais ete prospectees. Les auteurs estiment done qu’il est pre- mature de traiter P. makawai comme un synonyme. Des inventaires systematiques et complets sont necessaires, s’etendant peut-etre jusqu’aux regions limitrophes d’ Angola ou en Republique Democratique du Congo. What is Pogoniulus makawaP. The White-chested Tinkerbird Pogoniulus makawai was described in 1965 from a single specimen collected in Cryptosepalum forest in north-west Zambia. Lack of subsequent records has led to it being increasingly discounted as a valid species, because: (1) it could be an aberrant Yellow-rumped Tinkerbird P bilineatus , (2) although Cryptosepalum forest is a very distinctive habitat, the type locality does not appear to be in a centre of endemism, and (3) many people have since searched for it without success. However, we find 13 separate char- acters by which it diverges from P. bilineatus , a degree of difference which we feel cannot be ascribed to aberration. Moreover, the habitat of the area is by no means uniform; and the efforts to rediscover P. makawai in and around its type locality, while considerable, cannot be regarded as exhaustive, particularly since large areas of Cryptosepalum forest, extremely difficult habitat in which to work, have never been visited. Assigning P. makawai to synonymy is, we feel, prema- ture; systematic and comprehensive surveys, perhaps into adjacent Angola or DR Congo, are needed. From the very moment of its naming, the White-chested Tinkerbird Pogoniulus makawai has been haunted by doubt over its taxonomic sta- tus. The paper in which Benson & Irwin (1965a) described the species — taken by their remarkable collector Jali Makawa, in whose honour it was named, in an area of Cryptosepalum forest north of ‘Mayau’ in north-west Zambia — was immediately followed by a comment (Goodwin 1965) which postulated the notion that it might, in fact, be an aberrant Yellow-rumped Tinkerbird 7? bilineatus. Owing to the subsequent inability of anyone to confirm the existence of P makawai — a comment about ‘fresh material’ in Fjeldsa (2003) proving to have been unfounded (J. Fjeldsa in litt. 2005) — this possibility has been entertained with increas- ing conviction by two pairs of authorities, Dowsett & Dowsett-Lemaire (1980, 1993) and Short & Horne (1985, 1988, 2001, 2002). As a result, the species was not recognised by Sibley & Monroe (1990), Dowsett & Forbes-Watson (1993), Aspinwall & Beel (1998) or Dickinson (2003). In the face of this substantial scepticism on the part of two highly authoritative world lists, one equally authoritative African list, and the Handbook of the Birds of the World , BirdLife International, having treated makawai as a threat- ened species (Collar & Stuart 1985, Collar & Andrew 1988, Collar et al. 1994), has since 2000 opted to regard it as Data Deficient (BirdLife International 2000) — meaning that its taxonomic status is unclear — although the species’ was still 18 -Bull ABC Vol 13 No 1 (2006) What is Pogoniulus makawaP: Collar & Fishpool used to help define an Important Bird Area in Zambia (Leonard 2001, 2005). Short & Horne (1988) retained it as a species (with considerable reluctance), as did Sinclair & Ryan (2003) — whose report that the voice of makawai is ‘subtly different from Yellow-rumped Tinkerbird’ is presumably based on the comment in Aspinwall & Beel (1998) that the voices of the two taxa ‘may differ’ — but the trend in general suggests that because of this taxonomic uncertain- ty P. makawai will steadily disappear as a target of ornithological interest and investigation, and hence of conservation activity. This is in spite of two rather strong declarations in favour of makawai as a good species, one by the late C. W. Benson in a personal communication to NJC in Collar & Stuart (1985: 355), in which its ‘validity as a species has been most emphatically reasserted’, and the other by G. R. Graves, also to NJC, reported in Collar & Rudyanto (2003: 107-108), in which ‘following a preliminary (two-hour) inspection of the type, the White-chested Tinkerbird seems likely to prove a good species’. Neither of these judgements was published in a place where much notice would be taken of it — nor perhaps was the plea by Irwin (2003) — and 40 years after the species was named we feel the time has come to examine the evidence afresh, and to weigh more carefully the case for and against makawai as a taxonomic entity. The case for The case for has hardly been made since the first description. The only subsequent arguments in favour are the two personal judgements just quot- ed, neither of which comes with any detail to sup- port the conviction. The first thing to be done, therefore, is simply to list out, as clearly as possi- ble, the diagnostic features of P makawai as they emerge in Benson & Irwin’s (1965a) comparison with P bilineatus (not all of which are explicitly indicated as distinguishing marks, but which our comparison of text and specimens suggests was their intention): (1) white supraorbital stripe lack- ing; (2) white line below the ear-coverts only com- mencing behind the gape, not running below the eye in a continuous band from over the bill; (3) yellow fringes to the secondaries and wing-coverts paler, possibly narrower; (4) chin black, flecked centrally with white (chin whitish in bilineatus ); (5) throat and upper breast creamy white, fading to pale yellow on the lower chest (throat to belly pale whitish grey in bilineatus , belly with a slight greenish tinge); (6) lower breast to belly lacking greenish tinge; (7) central belly black (no such mark in bilineatus ); (8) entire underparts below breast with pale blackish ‘shadow-barring’ (absent in bilineatus ); (9) underside of the bend of wing black, not white; (10) tibial feathering more suf- fused black; (11) bases of feathers on mantle and underparts pale (dark in bilineatus ); (12) bill heav- ier, more arched and less conical, with cutting edges of the upper mandible flared around the gape; (13) black bill whitish basally and from the nostrils to halfway along the cutting edges (all black in bilineatus ); (14) rictal bristles at the level of greatest development found in any individuals of bilineatus ; (15) legs and feet markedly paler; (16) toes and claws ‘equally pallid’ as the legs and feet; and (17) legs slightly longer and more robust. From our own examination we would make the following comments and qualifications on these numbered characters: the wing fringes (3) are barely perceptibly paler but unquestionably narrower than in bilineatus ; in contrast to the glossy, inky black of the rest of the plumage, the chin (4) is a matt greyish black; the central belly patch (7) is a rather irregular smudge; the shadow- barring (8) is actually throughout the underparts, even on the creamy-white throat, but so slight as to be virtually invisible on the specimen when held at arm’s length; the gape-flange swelling (12) may not be greater than in some bilineatus , and as Goodwin (1965) pointed out, some allopatric bilineatus have bills that match makawai in size; the rictal bristles (14) are likewise barely different from those on bilineatus ; the continuous col- oration of the legs, feet, toes and claws (15, 16) form a single character; the legs are not longer than bilineatus and if they are more robust (17) this is too slight and unquantifiable a feature to allow. Thus we would say that the diagnosis of makawai rests on characters 1-11, 13 and 15 above, making 13 features in all. There is one possible further feature in Benson & Irwin (1965a): the white streak below the ear- coverts is ambiguously described as ‘joining with the pale under parts’ but then ‘from which it is separated by a black malar streak’. On balance, we interpret this to mean that the streak is continuous with the white neck, but the illustration accompa- nying the description clearly shows the opposite, What is Pogoniulus makawai ?: Collar & Fishpool Bull ABC Vol 13 No 1 (2006) -19 with both bilineatus and makawai having this lower facial stripe entirely enclosed by black. In partial contrast, the illustrations of the two taxa in Short & Horne (2001) and in Sinclair & Ryan (2003) show this streak meeting the pale under- parts in bilineatus but being enclosed by black in makawai , as if this is a distinct character differ- ence; and the illustrations in Short & Horne (2002), which omit makawai , again depict bilin- eatus with a streak continuous with the pale collar (although the position of the painted birds makes this very easy to miss). However, careful inspection of the type of makawai reveals that there is no essential difference between it and bilineatus in this regard, both taxa having a narrow line of black that in some positions appears to isolate the white cheek-stripe and in others is broken by it. Photographs in Short & Horne (2002: 139) and in Ginn et al. (1989: 398) show both conditions in bilineatus. The case against Of what, then, does the case against consist? As noted, there are three independent sources of doubt: (a) Goodwin (1965), (b) Dowsett & Dowsett-Lemaire (1980, 1993), and (c) Short & Horne (1985, 1988, 2001, 2002). In reality, how- ever, Goodwin (1965) only very tentatively sug- gested that ‘the possibility of its being an aberrant individual of P bilineatus cannot be entirely excluded’, and most of his commentary was weighted against this notion. He pointed out that makawai shows greater melanism on the head, underwing and central belly than bilineatus , and less melanism on the remaining belly area and breast, admitting that ‘it would be most unusual, but not unprecedented, for an aberrant individual to have more melanin than normal in some areas and less elsewhere’. He also pointed out that the greater curvature of the culmen and width of the bill of makawai are ‘not in themselves of great sig- nificance’, given that some forms of bilineatus have bills that approach and even match it in these characters; but he acknowledged that specimens of bilineatus (race mfumbiri) from near the type locality of makawai all have more slender, conical bills, suggesting some ecological separation, and again admitted that ‘it would certainly be surpris- ing if an aberrantly coloured individual happened also to have a slightly aberrant bill’. He then observed that makawai and bilineatus differ more strikingly in colour pattern than do bilineatus and Yellow-throated Tinkerbird P. subsulphureus , and pointed to the sharp difference in facial pattern of makawai and bilineatus when viewed front-on (a feature illustrated by Benson & Irwin), remarking that ‘this difference could function as an isolating mechanism as there is abundant circumstantial evidence that the coloration of the head and upper breast of birds is often of primary significance in this respect’. Goodwin thus concluded ‘that makawai is best considered as a new species, at least provisionally’, and both Mayr (1971) and Snow (1978) followed this judgement, the former adding a plea for comparative studies of the calls, the latter mistakenly referring to Yellow-throated Tinkerbird P. subsulphureus instead of Yellow- fronted Tinkerbird P chrysoconus as the third Pogoniulus in the area. Dowsett & Dowsett-Lemaire (1980), in their first of two brief comments on makawai , took much the same line as Goodwin, but, writing 15 years later, pointed out that ‘several visits to the type locality have failed to produce any further evidence, and in particular no unusual Pogoniulus vocalisations have been heard.’ After a further 13 years their patience had worn thinner: ‘investiga- tions by a number of observers in north-western Zambia have failed to rediscover it. . . As anticipat- ed by Dowsett & Dowsett-Lemaire (1980), we now believe it is no longer justified to recognise makawai as other than an aberrant P. bilineatus (Goodwin 1965)’ (Dowsett & Dowsett-Lemaire 1993). Short & Horne were always unconvinced. ‘Despite intensive searches in western Zambia, they wrote, makawai ‘remains known from but one specimen’; and because that specimen ‘comes from no distinctive habitat or area of endemism, and rather closely resembles P. bilineatus , we are inclined to regard it as a very aberrant specimen of bilineatus (Short & Horne 1985). Three years later, in The Birds of Africa (Short & Horne 1988) they allowed the species an entry but were pro- foundly sceptical: Status highly uncertain. Only 1 bird found, despite repeated searches... No ‘odd’ tinker- bird calls have been heard or unusual indi- viduals seen... at type locality... P makawai could prove to be an aberrant Yellow-rumped Tinkerbird, if its distinctive features are sim- ple melanism. 20 - Bull ABC Vol 13 No 1 (2006) What is Pogoniulus makawai?: Collar & Fishpool By the start of this century, in their book on barbets, their view had hardened further: they could now point to the failure of ‘three decades of searching by various ornithologists and bird- watchers’, and indeed they did so twice, mention- ing again the ‘numerous searches’ in the ‘relatively non-distinctive habitat in which it was found’ (Short & Horne 2001). Thus makawai is ‘almost certainly a very aberrant individual’ of bilineatus , although in the caption to their illustration of it they described it as a ‘morph’ (which is a very dif- ferent biological category). Even so, they gave it a separate account ‘because there seems to be no simple genetic explanation for all of its distinct features, e.g. melanism would account for some features, but not the lack of yellow and grey below, nor the heavy bill found in this male’ (Short & Horne 2001). On the other hand, only a year later they remarked that makawai ‘is now generally accepted as representing an odd variant of the Yellow-rumped Tinkerbird’ (Short & Horne 2002: 143), with a similar comment under the lat- ter species (Short & Horne 2002: 184). The case against makawai therefore depends on the following points: (1) that it could be an aberrant bilineatus', (2) that the type specimen was obtained in an area believed to be undifferentiated by habitat or by endemism; and (3) that searches have failed to find it or even to detect any unknown Pogoniulus calls, with emphasis variably placed on the number of searches — ‘intensive’, ‘repeated’, etc. — and simply the length of time — ‘three decades’ — without renewed contact (Snow [1978] stated, for example: ‘All attempts to obtain further specimens have so far proved unavailing’). These three objections need to be examined in turn. The case against examined 1. Could it be an aberrant Yellow-rumped Tinkerbird? — The possibility of makawai being an aberrant bilineatus seems to us to have been fairly well undermined by Goodwin (1963) even as he raised it. He admitted that aberrant specimens that are both more and less melanistic than typical birds are highly unusual; and he further admitted that for any such specimen also to be aberrant in bill morphology would compound the degree of anomaly. Apart from this, we regard the evidence in Benson & Irwin’s description — amounting in our judgement to 13 points of divergence — as simply too much to be ascribed to aberration. In particular, the redistribution of colour pattern — the black chin, the missing white supraorbital and supraloral bands (but the retained white cheek- stripe), the black belly patch, the part-pale bill and all-pale legs, the whitish dorsal underfeathering — is entirely uncharacteristic of aberrant individuals (although it is of course somewhat problematic to speak of what is typical of atypicality); certainly nothing in the entry ‘Plumage, abnormal’ in Campbell & Lack (1983) indicates otherwise, and we can think of no comparable case where so dis- tinctive a specimen has been disallowed taxonom- ic validity. Moreover, since Goodwin’s time of writing very considerable advances have been made in understanding the genetic basis of black plumage in birds (reviewed by Mundy 2005). In the light of these, the probability of a melanin-related muta- tion accounting for this divergence deserves recon- sideration. Across a wide range of taxa intraspecif- ic polymorphisms in melanin-based colours have repeatedly been found to be associated with varia- tion in a single gene (MC1R), but these typically involve a consistent increase in the extent of melanised feathers, rather than the simultaneous darkening and lightening seen in P makawai (as anticipated by Goodwin). Moreover, such muta- tions are typically not associated with simultane- ous side-effects on other traits, such as other com- ponents of morphology. Both of these points would tend to imply that the morphological dif- ferences between makawai and bilineatus are high- ly unlikely to have arisen as a consequence of a one-off mutation generating a single aberrant individual. 2. Is the type locality undifferentiated by habitat or endemism? — That the type of makawai comes from an area of low endemism and from a wide- spread habitat is not in serious dispute. Nevertheless, Benson & Irwin (1965a) pointed out that Cryptosepalum forest in this area could be seen as ‘an evolutionary centre’ given the presence there of ‘such distinctive forms’ as the red-necked race of Crested Guineafowl Guttera edouardi kath- leenae (although this is synonymised in Crowe et al. 1986), plus Margaret’s Batis Batis margaritae kathleenae and, ‘in this part of its range’, Gorgeous Bush-shrike Telephorus viridis. Benson & Irwin (1965b) and Benson et al. (1971) expanded on What is Pogoniulus makawai ?: Collar & Fishpool Bull ABC Vol 13 No 1 (2006) -21 this, indicating that the first and third of these taxa are known in Zambia only from this area of Cryptosepalum , which Irwin (2003) stressed ‘can hardly be described as “non-distinctive”’. T. B. Oatley (in lift. 2003) agrees: ‘One needs to look at the region, not just the Cryptosepalum forest, and Macronyx grimwoodi [Grimwood’s Longclaw] (and, if I remember rightly, some butterflies) can then be added to the list of local endemics.’ Benson & Irwin (1963b) offered the following scenario: Pogoniulus makawai is according to present knowledge endemic to Cryptosepalum... The ancestral population may have been wide- spread and plentiful in a former more exten- sive area of Cryptosepalum. Thereafter, as recently as 12,000 years ago, according to Moreau, a drier period ensued, during which this population may have become isolated, and speciated into makawai. Subsequently, under a moister, modern regime, bilineatus has perhaps reinvaded the Cryptosepalum. It may be in active and successful competition with makawai, which may before long become extinct. Whether or not it is plausible that makawai speciated as recently as 12,000 years ago, this explanation for its rarity makes considerable sense. It is not necessarily the case, however, that there is direct competition between makawai on the one side and bilineatus and, indeed, the syntopic Yellow-fronted Tinkerbird P. chrysoconus on the other; rather, one might expect makawai with its more voluminous bill to occupy a feeding niche that allows its co-existence with these closely relat- ed species. Benson et al. (1971) made this point, and F. Dowsett-Lemaire {in lift. 2005) has point- ed out that some species of Pogoniulus — including indeed chrysoconus — are specialists on mistletoe berries (Loranthaceae and Viscaceae) (see Dowsett-Lemaire 1988), so some kind of special- ism in makawai would seem likely to explain the co-occurrence of three congeners. 3. Has it been exhaustively searched for ? — Finally, there is the number of times that the type locality and nearby areas of Cryptosepalum have been visit- ed with no evidence of makawai being found. From some of the language used by those consid- ering the issue (reference to ‘intensive’ surveys and ‘three decades’ of searching), it is easy to assume that very considerable endeavours have gone into the quest for makawai. But what is the truth of this? Benson & Irwin (1965b) were the first to report on a new search. The specimen was collect- ed on 6 September 1964 during a four-day prospection of the area (3-7 September), and the area was revisited for five days, 8-12 November 1964, when ‘every effort was made to find the species again, but completely without success’. The following year, in May 1965, Oatley (1969) spent three weeks in north-west Zambia, explicit- ly in order to document the avifauna better and to discover more about makawai', he camped near ‘Mayau’ on 2-8 May, at a time when tinkerbirds were breeding, but heard no unusual calls, observ- ing that ‘possibly the single known specimen was a vagrant from some other locality, perhaps farther west in eastern Angola (and speculating that the heavier bill might be less important in feeding ecology than in hewing nesting cavities in harder timber than that encountered by bilineatus). Fifteen years after Oatley’s endeavour, Dowsett & Dowsett-Lemaire (1980) reported that ‘several vis- its’ to the type locality had drawn blank. The most important of these was by R. J. Dowsett with J. Makawa himself in 1973, when they spent 9-20 August at Mayau and collected ten bilineatus (Dowsett 1973, R. J. Dowsett in litt. 2005). The only other publication on Cryptosepalum birds in north-west Zambia, by Bowen (1980), was a study which did not involve ‘Mayau’, although one of the two sites surveyed was deemed very like ‘Mayau’, since it produced a bird list very similar to that in Benson & Irwin (1965b). Bowen, like Oatley, was constantly alert for novel tinkerbird calls but heard none, and speculated ‘that makawai does not extend as far north as the areas I covered’. Altogether, therefore, the evidence appears to be that relatively little work has been done in the area of the kind that would be appro- priate to a serious endeavour to rediscover the species. Thus A. J. Scott, director of the Wildlife Conservation Society of Zambia, wrote to J. FT Fanshawe at BirdLife International in 1989 to report that ‘no-one has made any attempt to find [P. makawai\ since Bob Dowsett... in the early 1970s’. From 1989 things evidently changed somewhat. R. J. Dowsett (in litt. 2005) himself has very helpfully enumerated the observers who have been to the type locality, and he suggests that, 22 -Bull ABC Vol 13 No 1(2006) What is Pogoniulus makawai ?: Collar & Fishpool in all, more than two man-months have been spent there by competent field workers: Dylan Aspinwall (several times); Carl Beel (incl. 1-3 Sep 1993); Clide Carter (several visits), with Nigel Hunter (5-6 Oct. 1989); Pete Leonard (incl. Aug. 1996); Jorg Mellenthin (Feb. & Aug. 2000, also 2002); Bob Payne; Bob Stjernstedt (incl. 14 Apr. 1974); Paul van Daele (incl. Sep. 1998). Details of other visits are unknown, but everyone who has lived in Zambia for any length of time has visited the area (and some visiting birders), hoping to become famous. I have also passed by Mayau briefly on other occasions. In addition, Fran^oise [Dowsett- Lemaire] and I spent 12 days camped a bit further south in the neighbouring West Lunga National Park in Nov. 1978 (Dowsett & Dowsett-Lemaire 1978), with no sign of any unusual tinkerbird. C. Beel (in litt. 2002) informed us that ‘Clide Carter, Dylan Aspinwall and Bob Stjernstedt tried’ to find it at some stage, although the sadly late Aspinwall (in litt. 1981 and 1994) did not men- tion such attempts and indeed in 1994 reported that he had no immediate plans to investigate the species; by that stage he was inclining to the view that makawai was most probably an aberrant bilineatus , which is, as already noted, how it was treated in Aspinwall & Beel (1998). Meanwhile Beel himself believed that only he, P. M. Leonard and P. van Daele had visited the area in recent years for any time, although Ryan & Cassidy (2003) paid a brief visit, and bird tours stop there to find Gorgeous Bush-shrike. Beel (in litt.) continued: Nearly all of us have spent time along the main Mwinilunga-Kabompo road. It is not easy to enter the forest away from this road. There is (was?) a narrow track to a game camp on the edge of West Lunga National Park which starts at Mayau. I tried to follow it, lost my car antenna and nearly my mir- rors, then turned back. This track crosses a vast block of Cryptosepalum where no-one has tried to look. Cryptosepalum itself is very hard to get around in; it is very dense and tangled. This makes it very difficult to get a good look at a tinkerbird even a short dis- tance into the forest. I think none of us spent more than a day, maybe two in the area. Very few tinkerbirds can be seen. Those seen have always been Golden-rumped and Yellow- fronted. No-one has been trying to call up each and every tinkerbird or concentrate fully on them. It is definitely true that no hard effort was made to prove or disprove the existence of P. makawai. Nearly all efforts were near the type locality. It might just be possible that the bird is not in prime habitat (anymore?) at that spot, especially as more people have moved into the area and cut lots of trees and burnt undergrowth. There are large tracts of Cryptosepalum remaining, but difficult of access and unvisited. Slightly more recently, P. M. Leonard (in litt. 2004) has commented as follows: Very few serious expeditions have taken place. Perhaps Terry Oatley’s in the 60s was the most serious, but I know of very few who have stayed more than a couple of nights before being driven away by sweat-bees and truly impenetrable jungle. I don’t think the searching has been thorough at all. Everybody tags it onto a Mwinilunga trip as a token gesture, but it needs a month of hardcore canopy scanning. This testimony is important: it suggests that a relatively rare bird of the canopy could easily go undetected in such dense and unwelcoming habi- tat (Beel stressed the need to provide plenty of water on a visit, and to prepare for the sweat-bees). If in fact there is some subtle habitat selection by makawai within the Cryptosepalum — for example, perhaps in taller growth along watercourses — then the lack of records since 1964 might be all the more understandable. T. B. Oatley (in litt. 2005) is surprised at Beel and Leonard’s comments about the impenetrabili- ty of the habitat, and offers the following reflexions: In 1965, the mavunda was not Very dense and tangled’ or ‘hard to get around in’ and it certainly was not ‘truly impenetrable jungle’. In fact, once one got through the dense edge- effect road fringe, one could walk around over a fairly open forest floor with scattered thickets of undergrowth [R. J. Dowsett in litt. 2005 reports similar conditions in 1973]. It sounds to me as though there has been heavy exploitation of the taller timber (resulting in dense secondary growth) in the area, which lies distant from district adminis- trative centres and is probably seldom What is Pogoniulus makawai ?: Collar & Fishpool Bull ABC Vol 1 3 No 1 (2006) -23 policed by forestry officers. I note that Beel explicitly refers to lots of tree cutting... Thinking over the possibilities, we concen- trated most of our searching in the Cryptosepalum forest patches, assuming that makawai was a forest canopy bird. But it need not have been. The area where it was collected was a mosaic of miombo woodland and mavunda forest, and bird parties from the woodland regularly entered the canopies of the forest patches that were in their path. Another point that arises is that September, when Jali collected the bird, is the early spring month when birds (especially floaters) move around a lot looking for mates or vacancies in the breeding population. Sept/Oct is also the time when many altitu- dinal and other Afrotropical migrants are on the move, the time when one can find odd birds briefly sojourning in atypical habitats! To this M. P. S. Irwin [in litt. 2005) adds some further thoughts: With long experience of collecting in the field, it is extremely difficult to say that something is not there or has been missed. Move camp a mile or two into a slightly dif- ferent habitat and there will be a complete change in the species one is likely to collect. All collecting [at Mayau] was done back from the river of that name where the forest was densest. But what about somewhere miles away? And look at the Cryptosepalum in Angola, which seems particularly dark and dense and quite unlike that at Mayau. Conclusion All of these factors lead us to the conclusion that there is no firm basis yet to discount Pogoniulus makawai as a good species. From this it must fol- low that either (a) it is very uncommon through- out Cryptosepalum forest in north-west Zambia, or (b) it is restricted to a relatively uncommon habi- tat within or adjacent to Cryptosepalum (if the type was taken on the Mayau River, then perhaps it was in or near riverine forest, not Cryptosepalum ), or (c) , as Oatley (1969) speculated, it is a straggler in this area from somewhat different habitats further west in near-adjacent Angola, a very little explored area, to which we would add the southern Democratic Republic of Congo, close to the bor- der of which the type locality of makawai also lies. With respect to this last point, it is worth noting that the type locality is also right on the edge of the range of bilineatus in Central Africa (see the maps in Snow 1978, Short & Horne 1988), so it would seem plausible that makawai may prove to be a replacement of bilineatus in slightly different habitats in this area. Certainly a concerted, systematic endeavour over several months is likely to be required either to relocate makawai within the general area of the type locality or to discount it properly from the local avifauna — but even then we could not coun- tenance abandoning the species to synonymy without much more work further west or north. (A hybrid origin of makawai seems, incidentally, never to have been mooted, but if a hybrid shows a degree of character intermediacy and bilineatus is one parent — both of which are reasonable assumptions — no plausible candidate for the other parent presents itself; on this point the explanation at once founders. Another consideration is that the type specimen of makawai might yet possess sufficient DNA to test against bilineatus , which possibly would resolve the debate at a stroke — but we can do no more than encourage the exploration of this notion.) Meanwhile, from our study of the type speci- men in the Natural History Museum, Tring, UK, we can confirm 13 points of distinction from P. bilineatus mfumbiri indicated by Benson & Irwin (1965a) as enumerated above (some of them evi- dent on the accompanying plates). The specimen (BMNH 1964.33.1) was an adult breeding male, testes 9.0 x 6.5 and 7x6 mm, wing 56 mm, tail 32 mm, tarsus 15 mm, culmen from base of skull 13 mm (Benson & Irwin 1965a); its skull was fully ossified (BMNH label data). It was collected ‘high up’ in the canopy of Cryptosepalum forest on Kalahari sand on 6 September 1964, four miles north of Mayau, Kabompo District, Zambia, at roughly 12°42’S 24°16’E (Benson & Irwin 1965a). The geographic component of this infor- mation was adjusted by Dowsett (1980): ...because of the importance of discovering further specimens of this species it is desir- able to publish as accurate a type-locality as possible. There is no locality named Mayau, but there is a river of that name (or Mayowo) and a plain. The specimen was collected about 6 km north of where the track from 24 -Bull ABC Vol 13 No 1 (2006) What is Pogoniulus makawai ?: Collar & Fishpool Credit for Figure 1: N. J. Collar (© Natural History Museum) ' Credit for Figures 2-3: C. N. Spottiswoode (© Natural 1 History Museum) Mwinilunga to Kabompo crosses the Mayau; this is at an altitude of 1,150 m at about 12°42’S 24°15’E. As precisely as possible, it is at this site that a systematic long-term search for the species should commence, whether the habitat has been modified 3 or not, spreading out methodically into the sur- rounding areas, with a particular interest in sam- pling any slight modifications of Cryptosepalum forest caused by water or other features, taking special note of mistletoe taxa and abundance, and with an eye on the fact that at this locality bilinea- Figures 1-3. The type and only specimen of Pogoniulus makawai (BMNH 1964.33.1, above), an adult male, taken near the Mayau River in north-west Zambia, September 1964, with an adult male P. bilineatus (BMNH 1964.33.2, below) taken at the same site three days earlier. Characters of makawai visible here are: no white supraorbital stripe; white line below ear-coverts commencing only behind gape; yellow edges to second- aries and wing-coverts narrower; chin black, flecked cen- trally white; throat and upper chest creamy-white, fading to pale yellow on lower chest; lower chest to belly with- out greenish tinge; central belly black; whole underparts below chest with blackish ‘shadow-barring’; tibia! feath- ering more suffused black; black bill whitish basally and from nostrils to halfway along cutting edges; legs and feet markedly pale. Other characters include underside of bend of wing black, not white; bases of feathers on man- tle and underparts light, not dark. L’ unique specimen (le specimen type) de Pogoniulus makawai (BMNH 1964.33. 1 , en haut), un male adulte, collecte pres de la Mayau, Zambie du nord-ouest, sep- tembre 1 964, avec un male adulte P. bilineatus (BMNH 1964.33.2, en bas) collecte a la meme localite trois jours auparavant. Les caracteristiques suivantes de makawai peuvent etre notees: absence de trait supraorbital blanc; trait blanc sous la joue commen^ant seulement en arriere de la commissure; remiges secondaires et couvertures alaires a liseres jaunes plus etroits; menton noir, tachete de blanc au centre; gorge et haut de la poitrine blanc- creme, devenant jaune pale sur le bas de la poitrine; absence de teinte verdatre sur le bas de la poitrine et le ventre; milieu du ventre noir; ventre barre de sombre; plumes du tibia teintees davantage de noir; bee noir, avec la base des mandibules blanchatre depuis les commissures jusqu’a la moitie du bee; pattes nettement pales. D’autres caracteres comprennent: dessous de la courbe de l’aile noir, non pas blanc; base des plumes du manteau et des parties inferieures claire, pas sombre. What is Pogoniulus makawai ?: Collar & Fishpool Bull ABC Vol 13 No 1 (2006) -25 tus is, apparently, at the southern edge of its range in the country. There was a time when the failure of ornithol- ogists to locate the Red-tailed Newtonia Newtonia fanovanae was attributed to the type and only specimen being an aberrant Red-tailed Vanga Calicalicus madagascariensis (see Collar & Stuart 1985). This view was, incidentally, strongly opposed by none other than C. W. Benson (Benson et al. 1977), but it took years before the increased ornithological interest in Madagascar from the early 1980s yielded a conclusive result — indeed, two virtually simultaneous results (Goodman & Schulenberg 1991; also Evans 1991). The rediscoveries of the Yellow-throated Serin Serinus flavigula and Sao Tome Grosbeak Neospiza concolor, each after over 100 years of absence, were likewise fundamentally a matter of scrutinising the evidence (see Collar & Stuart 1985) and patiently covering the ground in the most likely places (Ash & Gullick 1990, Sergeant et al. 1992). These three rediscoveries all occurred after considerable prior ornithological activity in the vicinity of the respective type localities, and they speak to us with the same single message: it is still too soon to place Pogoniulus makawai in syn- onymy with P. bilineatus , and if we look long and hard enough we may yet be pleasantly surprised. Acknowledgements Mark Adams of the Natural History Museum, Tring, UK, answered several queries following our examina- tion of the type specimen of P. makawai , to which Robert Prys-Jones allowed us access. Claire Spottiswoode commented on several drafts of this paper, took two of the photographs, and suggested the comment concerning melanism, whilst Nick Mundy checked and approved that comment. Subsequent drafts were read and commented on by John Colebrook-Robjent, Tim Dodman, Bob Dowsett, Fran^oise Dowsett-Lemaire, Pete Leonard, Terry Oatley and Michael Stuart Irwin. We are most grateful for all this help. References Ash, J. S. 6c Gullick, T. M. 1990. Serinus flavigula redis- covered. Bull. Hr. Omithol. Cl.\ 10: 81-83. Aspinwall, I). 6c Bed, C. 1998. A Field Guide to Zambian Birds not found in Southern Africa. Lusaka: Zambian Ornithological Society. Benson, C. W. 6c Irwin, M. P. S. 1965a. A new species of tinker-barbet from Northern Rhodesia. Bull. Br. Omithol. Cl 85: 5-9. Benson, C. W. 6c Irwin, M. P. S. 1965b. The birds of Cryptosepalum forests, Zambia. Arnoldia Rhod. 1(28): 1-12. Benson, C. W., Brooke, R. K., Dowsett, R. J. & Irwin, M. P. S. 1971. The Birds of Zambia. London, UK: Collins. Benson, C. W., Colebrook-Robjent, J. F. R. 6c Williams, A. 1977. Contribution a fornithologie de Madagascar. Oiseau & R.F.O. 47: 41-64. BirdLife International. 2000. Threatened Birds of the World. Cambridge, UK: BirdLife International. Bowen, P. St.J. 1980. The birds of Cryptosepalum forest near Mwinilunga. Bull Zambian Omithol. Soc. 12: 48-54. Campbell, B. 6c Lack, E. (eds.) 1985. A Dictionary of Birds. Cal ton, UK: T. & A. D. Poyser. Collar, N. J. 6>C Andrew, P. 1988. Birds to Watch: The ICBP World List of Threatened Birds. Cambridge, UK: International Council for Bird Preservation. Collar, N. J. 6c Stuart, S. N. 1985. Threatened Birds of Africa and Related Islands: the ICBP HU CN Red Data Book. Cambridge, UK: International Council for Bird Preservation 6c International Union for Conservation of Nature and Natural Resources. Collar, N. J., Crosby, M. J. 6c Stattersfield, A. J. 1994. Birds to Watch 2: The World List of Threatened Birds. Cambridge, UK: BirdLife International. Collar, N. J. 6c Rudyanto 2003. The archive and the ark: bird specimen data in conservation status assessment. Bull Br. Omithol. Cl 123A: 95-113. Crowe, T. M., Keith, S. 6c Brown, L. H. 1986. Phasianidae, guineafowl, Congo Peacock, quail, partridges and francolins. In Urban, E. K., Fry, C. H. 6c Keith, S. (eds.) The Birds of Africa. Vol. 2. London, UK: Academic Press. Dickinson, E. C. (ed.) 2003. The Howard & Moore Complete Checklist of the Birds of the World. Third edn. London, UK: Christopher Helm. Dowsett, R. J. 1973. New distributional records from North-western Province. Bull. Zambian Omithol. Soc. 5: 66-69. Dowsett; R. J. 1980. Comments on some ornithologi- cal type-localities in Zambia. Zambia Mus. J. 5: 7-16. Dowsett, R. J. 6c Dowsett-Lemaire, F. 1980. The sys- tematic status of some Zambian birds. Gerfaut 70: 151-199. 26 - Bull ABC Vol 13 No 1(2006) What is Pogoniulus makawai ?: Collar & Fishpool Dowsett, R. J. & Dowsett-Lemaire, R 1993. Comments on the taxonomy of some Afrotropical bird species. Tauraco Res. Rep. 5: 323-389. Dowsett, R. J. & Forbes- Watson, A. D. 1993. Checklist of Birds of the Afrotropical and Malagasy Regions, 1; Species Limits and Distribution. Liege: Tauraco Press. Dowsett-Lemaire, F. 1988. Fruit choice and seed dis- semination by birds and mammals in the evergreen forests of upland Malawi. Rev. Ecol. (Terre et Vie) 43: 231-285. Evans, M. I. 1991. The Red-tailed Newtonia Newtonia fanovanae in the Ambatovaky Reserve, north-east Madagascar. Bird Conserv. Intern. 1: 47-52. Fjeldsa, J. 2003. Patterns of endemism in African birds: how much does taxonomy matter? Ostrich 74: 30-38. Ginn, P., Mcllleron, W. G. & Milstein, P. le S. 1989. The Complete Book of Southern African Birds. Cape Town: Struik Winchester. Goodman, S. M. & Schulenberg, T. S. 1991. The redis- covery of the Red-tailed Newtonia Newtonia fanovanae in south-eastern Madagascar with notes on the natural history of the genus Newtonia. Bird Conserv. Intern. 1: 33-45. Goodwin, D. 1965. Some remarks on the new barbet. Bull. Br. Ornithol. Cl. 85: 9-10. Irwin, M. P. S. 2003. Review of Short & Horne (2001). Honeyguide 49: 237-238. Leonard, P. M. 2001. Zambia. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, UK: BirdLife International. Leonard, P. M. 2005. Important Bird Areas in Zambia. Lusaka: Zambian Ornithological Society. Mayr, E. 1971. New species of birds described from 1956 to 1965./. Ornithol. 112: 302-316. Mundy, N. I. 2005. A window on the genetics of evo- lution: MC1R and plumage colouration in birds. Proc. Roy. Soc. Lond. B272: 1633-1640. Oatley, T. B. 1969. Bird ecology in the evergreen forests of north western Zambia. Puku 5: 141-180. Ryan, P. & Cassidy, R. 2003. The ghost in the mavun- da. Africa — Birds & Birdingl (6): 37-41. Sergeant, D. E., Gullick, T., Turner, D. A. & Sinclair, J. C. 1992. The rediscovery of the Sao Tome Grosbeak Neospiza concolor in south-western Sao Tome. Bird Conserv. Intern. 2: 157-159. Short, L. L. & Horne, J. F. M. 1985. Social behavior and systematics of African barbets (Aves: Capitonidae). Pp. 255-278 in Proc. Intern. Symp. Afr. Vertebr. Bonn: Museum A. Koenig. Short, L. L. & Horne, J. F. M. 1988. Capitonidae, bar- bets and tinkerbirds. In Fry, C. H., Keith, S. & Urban, E. K. (eds.) The Birds of Africa. Vol. 3. London, UK: Academic Press. Short, L. L. & Horne, J. F. M. 2001. Toucans, Barbets and Honeyguides. Oxford: Oxford University Press. Short, L. L. & Horne, J. F. M. 2002. Family Capitonidae (barbets). In del Hoyo, J., Elliott, A. & Sargatal, J. (eds.) Handbook of the Birds of the World. Vol. 7. Barcelona: Lynx Edicions. Sibley, C. G. & Monroe, B. L. 1990. Distribution and Taxonomy of Birds of the World. New Haven & London, UK: Yale University Press. Sinclair, I. & Ryan, P. 2003. Birds of Africa South of the Sahara. Cape Town: Struik Publishers. a Conservation Biology Group, Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, UK, and BirdLife Lnternational, Wellbrook Court, Girton Road, Cambridge CB3 ONA, UK E- mail: nigel. collar@birdlife. org b BirdLife Lnternational, Wellbrook Court, Girton Road, Cambridge CB3 ONA, UK. E-mail: lincoln.fishpool@birdlife. org Received 7 June 2005; revision accepted 1 1 November 2005 What is Pogoniulus makawai?: Collar & Fishpool Bull ABC Vol 1 3 No 1 (2006) -27 White-winged Flufftail Sarothrura ayresi in Ethiopia: notes on habitat, densities, morphometries, nests and eggs, and associated waterbirds David G. Allan* , Alistair M. Mclnnes a and Mengistu Wondafrashb Le Rale a miroir Sarothrura ayresi en Ethiopie: notes sur l’habitat, la densite, les mensurations, le nid et les oeufs, et les oiseaux d’eau associes. Du 24 juillet au 2 aout 2003, les auteurs ont mene des etudes a Weserbi (pres de Sululta) et Berga, les deux seuls sites connus du Rale a miroir Sarothrura ayresi , espece globalement menacee, en Ethiopie. Aucune vocalisation des rales n’a ete entendue, malgre le fait que les oiseaux etaient en train de nicher et que les auteurs y ont prete particulierement attention avant l’aube et apres le coucher du soleil. Un male adulte a ete capture a Weserbi et ses mensurations ainsi que des details sur sa mue sont presentes. Des Rales a miroir ont ete leves environ 20 fois pendant chacun des deux jours passes a Weserbi. A Berga, dix rales furem leves en 433 minutes en ratissant avec une corde l’ensemble des zones apparemment favo- rables, et beaucoup d’autres en dehors de cet exercice. Des details sur les oiseaux d’eau rencontres a Berga sont presentes et compares a des donnees similaires de six des neuf zones humides prin- cipales en Afrique du Sud oil 1’espece a ete notee depuis les annees 1980. L’oiseau d’eau le plus commun a Berga et dans les zones humides sud-africaines est la Becassine africaine Gallinago nigripennis. Des donnees sont presentees sur 12 nids de becassines trouves a Weserbi et Berga. La taille des pontes de ces becassines (quatre oeufs) est exceptionellement elevee pour la Becassine africaine. Des details sont egalement presentes sur l’habitat du Rale a miroir a Berga, y compris l’habitat utilise pour nicher. Des sept nids de Rale a miroir trouves sur ce site, un contenait qua- tre oeufs, tandis que les six autres etaient vides. Des donnees quantifies sur les nids et les oeufs sont presentees. Deux jours consacres a la recherche de nouveaux sites de nidification potentiels du Rale a miroir au nord d’Addis-Abeba sont restes sans resultat. L’identification erronnee d’estomacs supposes avoir appartenu a des rales (Allan 2004) est corrigee. Summary. We visited the only two known sites, Weserbi (near Sululta) and Berga, of the global- ly threatened White-winged Flufftail Sarothrura ayresi in the highlands of Ethiopia on 24 July-2 August 2003. No vocalisations of the flufftails were heard despite the birds breeding and our lis- tening for them before and after sunrise and sunset. An adult male was captured at Weserbi and details of its measurements and moult are presented. White-winged Flufftails were flushed on c.20 occasions on each of two day-trips to Weserbi. At Berga, ten flufftails were flushed during 433 minutes of rope-dragging covering all apparently suitable habitat and many more were flushed while at this wetland engaged in other activities. Details of other large waterbirds encoun- tered at Berga are presented and compared with similar data from six of the nine main wetlands in South Africa where the species has been recorded since the 1980s. African Snipe Gallinago nigripennis was the most common large waterbird at both Berga and the South African wetlands. Details are presented of 12 snipe nests found at Weserbi and Berga. Clutch size (typically four eggs) was unusually large for African Snipe. Details are also presented of White-winged Flufftail habitat, including breeding habitat, at Berga. Of the seven White-winged Flufftail nests found at this site, one contained four eggs, whilst the others were empty. Quantified details of the nests and eggs are presented. Two days were spent searching unsuccessfully for potential new White- winged Flufftail breeding sites north of Addis Ababa. An error presented in Allan (2004) relating to the misidentification of alleged flufftail stomachs is corrected. 28 -Bull ABC Vol 13 No 1 (2006) White-winged Flufftail in Ethiopia: Allan et al. White-winged Flufftail’s Sarothrura ayresi global conservation status is ‘Endangered’ and its total population size is estimated at c.7 00 individuals (BirdLife International 2000, 2005). The presence of this Afrotropical endemic has been confirmed in only three countries: Ethiopia, South Africa and Zimbabwe; records from Rwanda and Zambia being unconfirmed (BirdLife International 2000). In Ethiopia it currently is known from two sites, Weserbi, in the Sululta area, and Berga, and it has been proved to breed at both (Atkinson et al. 1996, Taylor 1997, 1999, BirdLife International 2000). Both sites are Important Bird Areas (Fishpool & Evans 2001). In South Africa, where there is no confirmed evi- dence of breeding, the species has been recorded from nine main sites (all of which are within Important Bird Areas) since the 1980s and the total estimated population is 235 birds (BirdLife International 2000, Taylor 2000). In Zimbabwe there are two records from the 1970s and evidence for possible breeding in the 1950s (Hopkinson & Masterson 1984, Taylor 1994, BirdLife International 2000), although the first confirmed Figure 1. Head of adult male White-winged Flufftail Sarothrura ayresi , Weserbi, Ethiopia, 24 July 2004 (Alistair M. Mclnnes) Rale a miroir Sarothrura ayresi , tete de male adulte, Weserbi, Ethiopie, 24 juillet 2004 (Alistair M. Mclnnes) Figure 2. Upperwing pattern of adult male White- winged Flufftail Sarothrura ayresi, Weserbi, Ethiopia, 24 July 2004 (Alistair M. Mclnnes) Rale a miroir Sarothrura ayresi, male adulte, pattern du dessus de l’aile. Weserbi, Ethiopie, 24 juillet 2004 (Alistair M. Mclnnes) Figure 3. Open-wing outline (right wing; taken with bird lying on its back) of adult male White-winged Flufftail Sarothrura ayresi, Weserbi, Ethiopia, 24 July 2004 Contour de l’aile ouverte d’un Rale a miroir Sarothrura ayresi, male adulte (aile droite; prise avec l’oiseau couche sur le dos), Weserbi, Ethiopie, 24 juillet 2004 Figure 4. Active White-winged Flufftail Sarothrura ayresi nest, Berga, Ethiopia, 31 July 2004 (David G. Allan) Nid occupe du Rale a miroir Sarothrura ayresi, Berga, Ethiopie, 31 juillet 2004 (David G. Allan) White-winged Flufftail in Ethiopia: Allan et al. Bull ABC Vol 13 No 1 (2006) -29 Table 1. Results of rope-dragging at Berga 26-27 July 2003. Tableau 1. Resultats du ratissage avec corde du milieu a Berga, 26-27 juillet 2003. Section/Date southeast ■ -26 July northeast • -27 July northwest — 27 , July Transect no. 1 2 CO 3 4 LO CXI 5 LO 6 LO CO 1 LO LO 2 CO CXI 3 CXI 4 LO LO 5 CO 6 LO CXI 1 CO CXI 2 3 LO CO LO CO 4 CO m cz Time Mins ■'xj- CO CO _c: LO co _c= LO T CXI _cz LO CO cn co S CO T LO co CO LO -£= CO CXI -g V LO LO g CO CM CO CO H cl OO 03 CVJ LO co 2 s E <=> % s 123 20 20 24 co 25 S 30 46 13 18 12 28 27 31 16 433 in Long-tailed Cormorant Phalacrocorax africanus 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0.14 Black-headed Heron Ardea melanocephela 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0.14 Blue-winged Goose Cyanochen cyanopterus 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0.28 Yellow-billed Duck Anas undulata 0 0 1 0 0 2 1 0 1 0 0 2 0 1 0 0 8 1.12 African Water Rail Rallus caerulescens 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 2 0.28 Rouget's Rail Rougetius rougetti 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0.14 White-winged Flufftail Sarothrura ayresi 1 0 3 0 0 0 1 2 0 0 0 0 1 2 0 0 10 1.40 African Snipe Gallinago nigripennis 11 19 6 11 16 10 4 7 5 3 3 4 4 6 5 9 123 17.22 description of eggs from Ethiopia (Taylor 1999, Taylor et al. 2004) further calls into question the early putative Zimbabwean breeding evidence (Taylor et al. 200 4). It is uncertain if the species migrates between Ethiopia and South Africa or if each country sup- ports its own isolated population (BirdLife International 2000). Enigmatically, this flufftail’s vocalisations have been reported only from South Africa, where the purported main call has been likened to, and apparently is easily confused with, the unison call of the Grey Crowned Crane Balearica regulorum (Taylor 1998, BirdLife International 2000). The habitat of the species comprises seasonally flooded upland marshes and the major threat faced by this flufftail is the loss and degradation of these wetlands (BirdLife International 2000). Here we present results of field work on White-winged Flufftails at Weserbi (09°12’N 38°43’E) and Berga (09°16’N 38°23’E) marshes in Ethiopia, conducted from 24 July to 2 August 2003. We visited Weserbi marsh twice (on 24 July and 2 August) and Berga marsh on five days (23-28 and 31 July). A comparison of the water- bird populations present at Berga and at South African wetlands where White-winged Flufftails have been reported is presented. In addition, we undertook a two-day vehicle trip north of Addis Ababa to search for additional White-winged Flufftail breeding sites. A popular account cover- ing aspects of this work was presented by Allan (2004). Information presented in Allan (2004) is not repeated here, but additional details of the results are given, especially as related to quantified data. An error in the earlier account is corrected. Aims of the study Our major aims during this Ethiopian study were the following: 1) To collect the first descriptions and tape- recordings of the calls of White-winged Flufftail on their confirmed breeding grounds. 2) To capture the flufftails wherever possible, to gather morphometric and moult information. 3) To attempt quantified censuses of the flufftails at their breeding wetlands by plotting calling males and using playback and rope-dragging, and to record details of other large waterbird species encountered during these censuses. 4) To note habitat details at the areas in the wet- lands where the flufftails occur. 5) To search for flufftail nests and other evidence of breeding. 6) To. search for new breeding sites of flufftails in the Ethiopian highlands. 7) To collect feather samples of resident Ethiopian highland birds, especially large waterbirds, present at the flufftail breeding wetlands, and from the flufftails themselves, for potential trace-element analysis relevant to investigating the potential migration of the flufftails between Ethiopia and South Africa. 30 - Bull ABC Vol 13 No 1 (2006) White-winged Flufftail in Ethiopia: Allan et al. Table 2. Comparison of results of rope-dragging at South African wetlands (Wakkerstroom, Vanger, Bedford-Chatsworth, Murphy’s Rust, Franklin and Hebron) in 2001-2004 (covering a total of 27 hours and 8 minutes) and at Berga, Ethiopia, July 2003 (covering a total of 7 hours and 13 minutes). Tableau 2. Comparaison des resultats du ratissage avec corde des zones humides sud-africaines (Wakkerstroom, Vanger, Bedford-Chatsworth, Murphy’s Rust, Franklin et Hebron) en 2001-2004 (27 heures et 8 minutes au total) et a Berga, Ethiopie, juillet 2003 (7 heures et 13 minutes au total). Wakkerstroom Vanger Bedford / Chatsworth Murphy’s Rust Franklin Hebron South Africa Totals Berga Dec. 2002 Sep. 2001 Sep. & Dec. 2001 Dec. 2002 Dec. 2001 Jan. 2002 Feb. 2004 c /> July 2003 c o c n d d £ CO d d £ d £ d '£ d d *£ d d £ d d £ d £ d '£ CO d d £ CO Ik CO CO c/5 “O 15 ’£ cz CO % "O 15 CO CM 05 kk CO “O 15 OO kk CO “O 15 c\l lo CM kk S -C3 Id CO "77 c; CO CO “O 15 CO CM CO % “O 15 CO kk % “O 15 Long-tailed Cormorant Phalacrocorax carbo 0 0 0 0 1 0.2 0 1 0.04 1 0.14 Black-headed Heron Ardea melanocephala 0 0 3 0.2 0 4 1 5 2.7 12 0.44 1 0.14 Purple Heron A. purpurea 0 0 1 0.1 0 1 0.2 0 2 0.07 Little Egret Egretta garzetta 0 0 0 0 2 0.5 0 2 0.07 Yellow-billed Egret E. intermedia 0 0 0 0 1 0.2 0 1 0.04 Hadeda Ibis Bostrychia hagedash 0 0 6 0.4 0 0 0 6 0.22 Blue-winged Goose Cyanochen cyanopterus 0 0 0 0 0 0 2 0.28 Yellow-billed Duck Anas undulata 4 1.6 0 78 5.5 11 4.5 40 9.6 4 2.2 137 5.05 8 1.12 African Black Duck A. sparsa 0 0 2 0.1 0 0 0 2 0.07 Hottentot Teal A. hottentota 0 0 0 0 2 0.5 0 2 0.07 Spur-winged Goose Plectropterus gambensis 0 0 5 0.4 0 5 1.2 20 10.9 30 1.11 African Marsh Harrier Circus ranivorus 2 0.8 0 8 0.6 0 2 0.5 2 1.1 14 0.52 Wattled Crane Bugeranus carunculatus 0 0 4 0.3 0 0 3 1.6 7 0.26 Grey Crowned Crane Balearica regulorum 1 0.4 0 10 0.7 1 0.4 12 2.9 2 1.1 26 0.96 African Water Rail Rallus caerulescens 4 1.6 0 2 0.1 2 0.8 5 1.2 0 13 0.48 2 0.28 Rouget’s Rail Rougetius rougetti 0 0 0 0 0 0 1 0.14 Corncrake Crexcrex 0 0 1 0.1 0 0 1 0.6 2 0.07 Black Crake Amaurornis flavirostris 0 0 0 0 1 0.2 0 1 0.04 Baillon’s Crake Porzana pusilla 0 0 3 0.2 0 4 1 0 7 0.26 Red-chested Flufftail Sarothrura rufa 0 0 1 0.1 1 0.4 0 0 2 0.07 White-winged Flufftail S. ayresi 0 0 0 0 0 0 10 1.4 Purple Swamphen Porphyrio porphrio 0 0 0 0 1 0.2 0 1 0.04 Common Moorhen Gallinula chloropus 0 0 0 1 0.4 7 1.7 0 8 0.3 Three-banded Plover Charadrius tricollaris 0 0 0 0 0 1 0.6 1 0.04 Blacksmith Lapwing Vanellus armatus 5 2.1 0 2 0.1 0 0 3 1.6 10 0.37 African Wattled Lapwing V. senegallus 4 1.6 0 3 0.2 0 0 0 7 0.26 Wood Sandpiper Tringa glareola 0 0 0 0 2 0.5 0 2 0.07 Common Greenshank T nebularia 0 0 0 0 1 0.2 0 1 0.04 African Snipe Gallinago nigripennis 24 9.8 8 10.9 178 12.5 20 8.2 41 9.8 13 7.1 284 10.47 123 17.22 African Grass Owl Tyto capensis 4 1.6 0 4 0.3 0 1 0.2 1 0.6 10 0.37 Marsh Owl Asio capensis 1 0.4 0 31 2.2 0 0 0 32 1.18 Malachite Kingfisher Alcedo cristata 0 0 1 0.1 0 1 0.2 0 2 0.07 Cape Wagtail Motacilla capensis 0 0 6 0.4 1 0.4 0 0 7 0.26 TOTALS 49 8 349 37 134 61 632 23.30 148 20.50 White-winged Fluffiail in Ethiopia: Allan et al. Bull ABC Vol 13 No 1 (2006) -31 These samples are still being analysed and this aspect of the study will not be reported on fur- ther here. Results and discussion White-winged Flufftail vocalisations No flufftail vocalisations were heard at any stage. Weserbi was visited only during daylight hours, but at Berga we listened for the birds before and after sunset, and before and after sunrise (in addi- tion to the many daylight hours we were present). The periods devoted to this at Berga were as fol- lows: 25 July: I6h30 until c.l9h30 (late afternoon until well after dark); 26 July: 05h30-07h00 (dark until full light); 27 July: 05h45-07h00 and 18h30-19h45; and 28 July: 05h30-09h00. No calling was noted, despite the birds being clearly in the early stages of breeding at Berga (see below) and calling from other rallids present was com- mon, e.g. African Water Rail Rallus caerulescens and Rouget’s Rail Rougetius rougetti. A single Red- chested Flufftail Sarothrura rufa was tape-recorded at Berga on our last ‘listening stint’. This species has been noted previously at Berga (Taylor 1997). We can only speculate as to our failure to hear any White-winged Flufftails. Possibly they call at their breeding sites only prior to initiating nesting and were already silent by the time we arrived. Perhaps they only call very late at night, when we were absent. Possibly their calls are inconspicuous and we overlooked them. Conceivably we were unlucky and they did not call (at least within earshot) during our five prime listening opportu- nities. Or maybe White- winged Flufftails are totally silent throughout their sojourn on their breeding grounds. This, however, would be remarkable considering the vocal nature of the Rallidae, at least during the breeding season (Taylor & van Perlo 1998). Taylor et al. (2004) also report the species to be ostensibly silent on the breeding grounds. Taylor & van Perlo (1998) point out that the only other members of the Rallidae that have white secondaries are the three species of Coturnicops. They describe the main call of one (Speckled Rail C. notatus of southern South America) as ‘unobtrusive and easily masked by other marsh sounds’, another (Yellow Rail C. noveboracensis of North America) as ‘a series of metallic clicks’ and the third (Swinhoe’s Rail C. exquisitus of the Far East) as ‘not recorded’. Sarothrura and Coturnicops may be closely related (Taylor & van Perlo 1998). White-winged Flufftail morphometries and moult An adult male White-winged Flufftail was caught by hand, and subsequently released unharmed, at Weserbi on 24 July (Figs. 1-2). The following measurements were taken: wing-length (flattened chord) 72 mm, tarsus-length (notch on rear of tibio-tarsal joint to point of divergence of upper surface of inner toe) 23.8 mm, culmen-length (to feathering) 10.6 mm. In addition, its open-wing outline was traced onto paper (Fig. 3). There was no sign of moult in the remiges and its entire plumage appeared very fresh. Taylor & van Perlo (1998) present the following relevant measurements for White- winged Flufftail males: mean wing-length (flattened chord) 76.3 mm (range=73. 0-80.0; n= 14), mean tarsus-length (inter-tarsal joint to distal end of last undivided scale before toes diverge) 18.5 mm (ranges 17.0-1 9. 5; »=14) and mean culmen-length (to base) 12.4 mm (range= 12.0-13.5; n- 13). Our wing-length, using the same method as Taylor & van Perlo (1998), was 1 mm shorter than their presented range. Our tarsus- and culmen-lengths were measured differently to Taylor & van Perlo (1998) and are not directly comparable. Taylor & van Perlo (1998) also state that Ethiopian birds examined in late July were in very fresh plumage. Large waterbird censuses The absence of calling ruled out any census of White-winged Flufftail based on plotting calling males or on playback efforts. White-winged Flufftails were flushed on c. 20 occasions on each of the two day-trips to Weserbi. On both days, many of these instances probably represented the same individuals flushed more than once. No rope-dragging was performed at Weserbi. Atkinson et al. (1996) estimated that two pairs were present at Weserbi in 1995. Taylor (1997) estimated the site to support 10-15 pairs in 1996 and 1997, and we would subjectively agree with this estimate. A major activity at Berga comprised rope- dragging, a widely accepted method of assessing cryptic-waterbird abundances (see e.g. Green 1985). This method was used at Berga previously 32 - Bull ABC Vol 13 No 1 (2006) White-winged Flufftail in Ethiopia: Allan et al. by Taylor (1999), and we employed it on 26-27 July. We covered all of the apparently suitable habitat, subjectively estimated at c.50-100 ha. Taylor (1997) estimated the suitable habitat at c.200 ha in 1997. The number of large waterbirds counted during these efforts is presented in Table 1 . Ten White-winged Flufftails were flushed during 433 minutes of rope-dragging (and many others were flushed while we were at the wetland engaged in other activities). The flufftail was the second-most common of the eight large waterbird species encountered while rope-dragging, after the ubiquitous African Snipe Gallinago nigripennis. Taylor (1997) estimated the flufftail population at Berga at c.200 pairs. This is hard to reconcile with our flushing of only ten individuals whilst rope- dragging the entire area. However, the number of flufftails that failed to flush, and therefore remained undetected, during the rope-dragging is unknown. In addition, the relatively high number of nests found (see below), especially on 31 July, and the close spacing of some of these, suggests that flufftail numbers were much higher than indi- cated by the rope-dragging efforts. It was our impression that most, if not all, White-winged Flufftails flushed at both Weserbi and Berga were males, based on the bright chest- nut appearance of their foreparts. However, as sex- ual dimorphism in plumage features is reduced in this flufftail (Taylor et al. 2004), with females also showing some chestnut-brown on the foreparts, this conclusion is tentative. Many of the birds flushed were not seen well but those that were all seemed to be males. No flufftails were seen on the ground. Table 2 compares the results of our rope- dragging with comparable rope-dragging efforts made in South Africa in 2001-04 (all during the austral spring-summer periods September- February) at six of the nine major wetlands where the species has been recorded since the 1980s (although our attempts to locate White-winged Flufftails in South Africa have been singularly fruitless). African Snipe also was easily the most commonly encountered species in the South African wetlands, although its abundance was higher at Berga (10.3 snipe/60 minutes of rope- dragging vs. 17.2 snipe/60 minutes). Interestingly, Yellow-billed Duck Anas undulata was the second- most commonly encountered species in South Africa and the third at Berga. Overall, the South African wetlands boasted a much higher diversity of large waterbird species than Berga (30 vs. 8). This is unsurprising, as more rope-dragging was performed in South Africa (1,628 mins vs. 433 mins), over a longer time period, during periods when Palearctic migrants are present (as opposed to absent in the case of the Berga survey; although only three species at the South African wetlands were Palearctic migrants: Corncrake Crex crex, Wood Sandpiper Tringa glareola and Common Greenshank T. nebularia ), and at more (six vs. one) and widely spread localities. The number of large waterbird individuals counted per 60 min- utes of rope-dragging was similar between the South African wetlands and Berga (23.3 vs. 20.5). Additional large waterbirds of conservation interest noted at Berga at times other than while rope-dragging were a pair of Wattled Cranes Bugeranus carunculatus (Vulnerable), apparently only recorded there once previously in August 1998 (Fishpool & Evans 2001), and a single Black Crowned Crane Balearica pavonina (Near Threatened). African Snipe nests We located 12 snipe nests, one at Weserbi and 11 at Berga, on 24-31 July. Nine of these contained four eggs, two had three eggs and one contained two eggs. One of the nests with three eggs obvious- ly had been damaged by a flash-flood on the previ- ous evening; one of the eggs had been partially dis- placed from the nest cup and another had been washed out of the nest and was lying adjacent to it; additional eggs may have been washed away com- pletely. The nest with two eggs and the other with three eggs may have held incomplete clutches (no nests were subjected to follow-up visits). Taylor (1999) found at least 20 nests’ of snipe at Berga in 1999 most with a full clutch of four eggs’. Clutch size of African Snipe in southern Africa is typically two eggs, occasionally one or three (Maclean 1993); for the entire continent it is given as 2-3 eggs (Urban et al. 1986). Clutch size in Ethiopia — typically four eggs — therefore is remarkably large. Common Snipe G. gallinago, known only as a non- breeding migrant to Ethiopia (Urban & Brown 1971) and Africa (Urban et al. 1986), has a typical clutch size of four eggs (Cramp & Simmons 1982), raising the interesting, but admittedly unlikely, possibility that the snipe breeding in the Ethiopian highlands may represent this species. An examina- White-winged Flufftail in Ethiopia: Allan et al. Bull ABC Vol 13 No 1 (2006) -33 tion of any specimens collected from Ethiopian highland wetlands during the boreal summer might cast further light on this issue. The Ethiopian highlands support isolated breeding populations of three other bird species characteris- tic of the Palearctic region: Golden Eagle Aquila etoSy discovered only in 1993 (Clouet & Barrau 1993), Ruddy Shelduck Tadorna ferruginea (Ash 1977) and Red-billed Chough Pyrrhocorax pyrrhocorax (Urban & Brown 1971). The snipe nests we found were located in short, dense aquatic vegetation comprising sedges, grasses and flowers (Asteraceae), including four nests in sedge tufts (one an Eleocharis sp.), one in a tuft of the aquatic grass Odontelytrum abbysinicum and one situated in the base of an aquatic flower. The nests were well-concealed pads of grasses, sedges and flower stems. Their bases were c. 1 cm above water level. In the areas where the nests were located mean water depth was 6 cm (SD=4 cm, range 1-13 cm, n= 12) and mean veg- etation height 42 cm (SD=20 cm, range 30-100 cm, «=12). Mean nest diameter was 11 cm (SD=1 cm, range 10-13 cm, 72=10) and mean egg dimensions were 40.7 x 29.3 (37.4-43.1 x 28.0-30.6, 72=44). White-winged Flufftail habitat Atkinson et al. (1996), Tilahun et al. (1996), Taylor & van Perlo (1998), Fishpool & Evans (2001) and Taylor et al. (2004) provide general descriptions of White-winged Flufftail habitat at Weserbi and Berga marshes, including botanical details, and we have little to add to these accounts. We recorded the habitat at those places where we flushed the ten flufftails while rope-dragging at Berga. Mean water depth was 4.9 cm (SD=2.2 cm, range 2-8 cm, 72=9) and mean vegetation height 44.4 cm (SD= 16.1 cm, range 25-80 cm, 72=9). As mentioned above, the wetlands comprise a mix- ture of aquatic grasses, sedges and Asteraceae. At the ten spots from which flufftails were flushed, one comprised largely sedges, one largely Asteraceae, four a mixture of sedges and Asteraceae, and four a mix of grasses, sedges and Asteraceae. Particularly prominent grasses includ- ed Odontelytrum abbysinicum and a possible Leersia sp., and amongst prominent sedges an Eleocharis sp. It was our impression that an abun- dance of a large, leafy, yellow-flowered Asteraceae ( Trifolium / Haplocarpha/ Ranunculus sp.?), amongst a mixture of aquatic grasses and, especial- ly, sedges, was the most characteristic botanical feature of places where flufftails were flushed both at Berga and at Weserbi. At Berga most flufftails were flushed in fairly deep water close to the main watercourse in the central half of the wetland, in areas characterised by a particular abundance of Asteraceae, relative to the shallower outer half of the wetland, characterised by an apparently greater preponderance of sedges. White-winged Flufftail nests Seven White-winged Flufftail nests were located at Berga, one each on 25 and 26 July and five on 31 July. The first nest was shown to us by local peo- ple, the second was found while rope-dragging and the five nests located on 3 1 July were all found by local people who joined us in a casual search for nests on that day covering only a relatively small section of the marsh. All but one of the nests was empty and apparently still under construction. The active nest (Fig. 4), located on 31 July, was c. 1 00 m from the main watercourse, in a water- logged area covered by dense aquatic vegetation c.40 cm high, comprising aquatic grasses, sedges and Asteraceae. Water depth was less than 1 cm. The nest was built in a sedge ( Cyperus sp.) tuft 40 cm high. It was a ball-shaped structure and its base was set c. 1 cm above ground level with a side entrance. Live plant stems had been pulled over the top and woven together to form a dome. Nest dimensions were: width 12 cm, height 17.5 cm, entrance width 5.5 cm, entrance height 6.5 cm, interior width 8.5 cm, interior height 8.5 cm. The entrance faced south. The nest contained four unmarked, ivory-white eggs. Egg measurements and weights were: 29.3 x 19.8 (5.5 g), 28.9 x 20.2 (5.8 g), 28.8 x 20.1 (5.8 g) and 28.4 x 19.9 (3.3 g). The six empty nests were all found in the same general area and habitat, and were similar in struc- ture. Additional details were noted for three of these. Two were in sedge tufts and a third was interwoven between sedge and aquatic grass tufts. Tuft height was 40-45 cm (»=3). The bases of the nests were 1-4 cm above ground level (t2=3). Nest dimensions were: width 12-14 cm, height 14-19 cm, entrance width 5. 0-8. 5 cm, entrance height 6. 0-9.0 cm, interior width 8. 9-9.0 cm, interior height 9.5-1 1.0 cm (t2=3). Entrance aspects were south (72= 1), north-west (72= 1) and 34 - Bull ABC Vol 13 No 1 (2006) White-winged Flufftail in Ethiopia: Allan et al. north-east {n= 1). One nest was only c. 5 m from another and a third was c. 30 m from these two. No flufftails were seen at any of these nests, although a few were flushed in the general area. The nests were situated in the shallower outer half of the wetland (water depth less than 1 cm), large- ly away from the deeper, more central parts of the marsh (water depth 2-8 cm, mean 4.9 cm) where we flushed most flufftails. Taylor (1999) and Allan (2004) present brief popular descriptions of the nests and eggs of White-winged Flufftails from Berga, both accom- panied by photographs. Taylor et al. (2004) pro- vide a more formal description of a single nest, the same as that covered by Taylor (1999). The nest was c.60 m from the main watercourse in aquatic vegetation 30-45 cm tall and dominated by the sedge Eleocharis marginulata , a grass species and a flower Ranunculus sp. Tuft height was 40-45 cm. The nest was 1 cm above ground in an area with water depth 1-5 cm. It was a ball-shaped structure constructed of live Eleocharis , grass and Ranunculus leaves. Nest dimensions were: width 15 cm, height 17.5 cm, entrance width 4.8 cm, entrance height 5 cm, interior width 9.8 cm, inte- rior height 9.4 cm. The complete clutch consisted of six unmarked white eggs, suggesting that the clutch we found may have been incomplete, and the eggs were laid in mid-August 1999. Dimensions and weights of three eggs were: 27.2 x 19.8 (5.47 g), 27.3 x 20.0 (5.65 g) and 27.8 x 19.9 (5.78 g). In all respects, the breeding habitat, nest and eggs described by Taylor et al. (2004) are similar to those found by us. Searches for potential new White-winged Flujftail breeding sites Our relatively brief search for potential new White-winged Flufftail breeding sites comprised a two-day trip starting 29 July We travelled from Addis Ababa north-east to Debre Birhan, where we over-nighted, and then west to the Fitche area, before returning to Addis Ababa late on 30 July. The main, and apparently only, roads between these three urban centres were followed. The total distance covered was 435 km. We stopped the vehicle and walked potentially suitable wetland areas visible from the roads. Six sites were exam- ined at the following localities: 09°15’N 39°12’E, 09°16’N 39°14’E, 09°30’N 39°27’E (Hadewa Shet River), 09°36’N 39°30’E (International Livestock Research Institute, Debre Birhan Station); 09°49’N 38°35’E (Arkiso River) and 09°34’N 38°51’E. No 'White-winged Flufftails were found at any of these sites. The first two localities were extremely small and unlikely candi- dates for the species. The other four were larger and parts of extensive wetland systems. All but one of these, however, were heavily grazed by livestock and did not present suitable vegetation height and density. The exception, the wetland at the International Livestock Research Institute 9 km south of Debre Birhan, presented an extensive wetland area with some potentially suitable patch- es of habitat. Snipe were drumming at this site, the only place where we found this away from Weserbi and Berga. An intensive walk around the wetland over several hours, however, failed to produce any flufftails and the vegetation was perhaps too short, sparse and patchy overall, and the wetland too deeply flooded and channeled in many areas. This wetland was entirely fenced, with surrounding livestock excluded, hence its relatively tall aquatic vegetation. This route had already been covered by Dr Barry Taylor accompanied by one of us (MW) in earlier searches for the flufftail, although the Arkiso River site apparently was the only locality where our actual searching efforts overlapped. White-winged Flujftail stomachs and predation — a correction Allan (2004) reported on observations made dur- ing our visit to Weserbi, on 24 July, of an Augur Buzzard Buteo augur capturing a flushed White- winged Flufftail, and on the collection of three stomachs of flufftails, one apparently from the captured flufftail, all apparently discarded by the buzzard while feeding on the birds. A subsequent detailed examination revealed that the stomachs were of rodents. The exact sequence of events was that one of our party (Deon Coetzee) observed the Augur Buzzard capturing a flushed flufftail as it landed in a sparse wheat field adjacent to the wet- land. He flushed the buzzard, and picked up sev- eral flufftail feathers displaced by it when the fluff- tail was caught. The buzzard flew off with the flufftail to a low bush in the wetland. A local herdsman then ran to the bush, again flushing the buzzard which flew off with the flufftail. The herdsman picked up several more flufftail feathers from below the bush, as well as a stomach. We White-winged Flufftail in Ethiopia: Allan et al. Bull ABC Vol 1 3 No 1 (2006) -35 thus assumed the stomach to have come from the flufffail and ascribed the other two lone stomachs found elsewhere in the wetland as also coming from flufftails. It seems probable that the buzzard was, in fact, feeding primarily on rodents and that the capture of the flufftail was an isolated and unnatural’ incident related to our disturbance of the birds. The comments in Allan (2004) as to the over-grazed nature of the wetland having resulted in this predation of the flufftails by the buzzard therefore are groundless. Acknowledgements We thank the Middelpunt Wetland Trust (MWT), especially its Chairman Deon Coetzee, for providing the major funding that made field work in Ethiopia (and South Africa) possible. Deons companionship, focus, energy and drive while accompanying us in Ethiopia were also greatly valued. DGA and AMM offer heartfelt thanks to the Ethiopian Wildlife & Natural History Society and its staff for their guidance, assistance, companionship and logistical support throughout our stay. Steven Evans, at the time employed by BirdLife South Africa, accompanied us in part at Weserbi and Berga, and we are grate- ful for his involvement. Malcolm Drummond, an MWT trustee, provided invaluable background logistical assistance. Thanks also to Warwick Tarboton for pointing out the significance of the snipe clutch sizes. Additional funding was provid- ed by the Claude Leon Harris Foundation Fund of the Durban Natural Science Museum, from monies channeled through the Percy FitzPatrick Institute of African Ornithology. References Allan, D. G. 2004. Ethiopia — enigmas and endemics. Africa — Birds & Birding 9(2): 28-34. Ash, J. S. 1977. First known breeding of the Ruddy Shelduck Tadorna ferruginea south of the Sahara. Bull. Br. Ornithol. Cl. 97: 56-59. Atkinson, R, Robertson, R, Dellelegn, Y, Wondafrash, M. &c Atkins, J. 1996. The recent rediscovery of White-winged Flufftails in Ethiopia. Bull. ABC 3: 34-36. BirdLife International. 2000. Threatened Birds of the World. Barcelona: Lynx Edicions & Cambridge, UK: BirdLife International. BirdLife International. 2005. Species fact sheet: Sarothrura ayresi. Downloaded from http://www.birdlife.org on 2 August 2005. Fishpool, L. D. C. & Evans, M. I. (eds.) 2001. Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, UK: BirdLife International. Green, R. E. 1985. The Management of Lowland Wet Grasslands for Breeding Waders. Sandy, UK: Royal Society for the Protection of Birds. Hopkinson, G. & Masterson, A. N. B. 1984. The occurrence and ecological preferences of certain Rallidae near Salisbury, Zimbabwe. Proc. V Pan-Afr. Orn. Congrr. 425-440. Maclean, G. L. 1993. Roberts Birds of Southern Africa. Sixth edn. Cape Town: Trustees of the John Voelcker Bird Book Fund. Taylor, B. 1997. Hope for the White-winged Flufftail. Africa — Birds & Birding If))'. 14-15. Taylor, B. 1999. White-winged Flufftails in Ethiopia - another mystery solved. Africa — Birds & Birding 4(5): 23-25. Taylor, B. & van Perlo, B. 1998. Rails: A Guide to the Rails, Crakes, Gallinules and Coots of the World. Robertsbridge: Pica Press. Taylor, B., Wondafrash, M. & Demeke, Y. 2004. The nest, eggs and chicks of the White-winged Flufftail Sarothrura ayresi. Bull. Br. Ornithol. Cl. 124: 233-239. Taylor, P. B. 1994. The biology, ecology and conserva- tion of four flufftail species, Sarothrura (A ves: Rallidae). Ph.D. thesis. Pietermaritzburg: University of Natal. Taylor, P. B. 2000. White-winged Flufftail Sarothrura ayresi. In Barnes, K. N. (ed.) The Eskom Red Data Book of South Africa, Lesotho and Swaziland. Johannesburg: BirdLife South Africa. Tilahun, S., Edwards, S. & Egziabher, T. B. G. (eds.) 1996. Important Bird Areas of Ethiopia: A First Inventory. Addis Ababa: Ethiopian Wildlife & Natural History Society. Urban, E. K. & Brown, L. H. 1971. Checklist of the Birds of Ethiopia. Addis Abba: Addis Ababa University Press. Urban, E. K., Fry, C. H. & Keith, S. (eds.) 1986. The Birds of Africa. Vol. 2. London, UK: Academic Press. a Durban Natural Science Museum, PO Box 4085, Durban 4000, South Africa. E-mail: AllanD @du rban.gov. za b Ethiopian Wildlife and Natural History Society, PO Box 60074, Addis Ababa, Ethiopia. 36 -Bull ABC Vol 13 No 1 (2006) White-winged Flufftail in Ethiopia: Allan et al. Status of Shoebill Balaeniceps rex in Malagarasi, Tanzania Lars Dinesena and Marc Bakerb Statut du Bec-en-sabot du Nil Balaeniceps rex a Malagarasi, Tanzanie. En Tanzanie, le site de Malagarasi est le plus important pour la survie du Bec-en-sabot du Nil Balaeniceps rex et est sup- pose abriter la plus grosse population de Fespece en dehors du Sudd au Soudan. Les tentatives pour estimer la taille de cette population semblent cependant avoir resulte en de considerables surestimations. L’examen de la litterature et les comptages aeriens realises en novembre 2001 par les auteurs suggerent que la population de Malagarasi ne comprendrait que quelques centaines d’oiseaux seulement. Les comptages aeriens de 200 1 ont egalement montre que l’etendue de cer- taines zones humides a Malagarasi avait ete serieusement surestimee. Les auteurs considerent done que le changement du statut de conservation du Bec-en-sabot de ‘Quasi-menace’ en ‘Vulnerable’ (espece dont la population entiere compte moins de 10.000 individus) est justifie. Le Bec-en- sabot du Nil a une repartition restreinte, s’etendant du sud du Soudan au nord de la Zambie et de la Republique Democratique du Congo orientale a la Tanzanie occidental, et Fespece est menacee dans tous les sites importants, comme les marais du Sudd, Malagarasi, Bangweulu et en Ouganda. Un protocole est propose pour les futurs comptages aeriens du Bec-en-sabot, afin de definir les zones principales occupees par Fespece et d’effectuer des comptages complets. Summary. The Malagarasi ecosystem is the stronghold for Shoebill Balaeniceps rex in Tanzania and is believed to host the largest population of the species outside the Sudd in Sudan. However, attempts to establish population numbers in Malagarasi seem to have resulted in considerable over-estimations. A review of the literature and aerial counts in November 200 1 by the authors suggest that the Malagarasi population might comprise 100 or a few hundred birds only. The aer- ial survey in 2001 also revealed that previous estimates of the extent of certain wetland habitats in Malagarasi have been significantly overestimated. The authors therefore support the upgrade of the conservation status of Shoebill from Near Threatened to Vulnerable, i.e. with a total pop- ulation of less than 10,000 individuals. Shoebill has a limited range from southern Sudan to northern Zambia and from eastern DR Congo to western Tanzania, and the species is threatened at all its main localities e.g. the Sudd swamps, Malagarasi, Bangweulu and in Uganda. A design for future aerial surveys of Shoebill is proposed focusing on identifying core areas for the species and comprehensive counts. Shoebill Balaeniceps rex has a discontinuous dis- tribution, restricted mainly to a few large per- manent wetlands from southern Sudan, Uganda, eastern DR Congo and northern Zambia to west- ern Tanzania. Its world population was estimated at only c. 1,500 individuals two decades ago (Brown etal. 1982), and at c.l 1,000 (Elliott 1992) or 12,000-15,000 in 1986 (Collar 1994, Rose & Scott 1994, BirdLife International 2000). This apparent increase was due to the use of new cen- sus methods, including aerial surveys. In 2002, however, the population was estimated at 5,000-8,000 birds and declining, based on new information (Delany & Scott 2002). The Tanzanian population was estimated at maximum 2,500 individuals, based on aerial sur- veys by the Tanzania Wildlife Conservation Monitoring (TWCM, now Conservation Information Centre under Tanzania Wildlife Research Institute (TAWIRI) between 1990 and 2000 (Tanzania Wildlife Conservation Monitoring 1991, 1998, 1999 and unpubl., Jones & Hill 1994a, Collar 1994, Baker 1996, BirdLife International 2000). The prime locality for Shoebill in Tanzania are the permanent swamps and inundated floodplains along the Moyowosi, Nikonga, Kigosi, Igombe, Ugalla and Malagarasi rivers (hereafter: the Malagarasi) forming a mosaic of wetlands in western Tanzania (Fig. 1). The Malagarasi was listed as a Ramsar site in August 2000, when Tanzania ratified the Convention on Wetlands (Ramsar 1971), and the Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Bull ABC Vol 13 No 1 (2006) -37 Figure 1. Map of Tanzania showing protected areas and the Malagarasi-Muyovozi Ramsar sites. Carte de la Tanzanie indiquant les aires protegees et le Site Ramsar Malagarasi-Muyovozi. designated area covers c.32,500 km2. The main rationale was that it supports appreciable numbers of rare, vulnerable or endangered species’ such as Shoebill, Wattled Crane Bugeranus carunculatus, Sitatunga Tragelaphus spekei and Slender-snouted Crocodile Crocodylus cataphractus (Anon. 1994a,b, Jones & Hill 1994a). However, it is recognised that additional research is needed to establish more precise population numbers and distribution patterns of these species. This paper focuses on Shoebill and is based on field work on the ground in 1995, aerial transect counts in November 2001, subsequent visits in 2002-03 and a review of the existing literature. Study area A large part of the Malagarasi ecosystem was included in the Malagarasi-Muyovozi Ramsar sites with the notable exception of the outlet of the Malagarasi River into Lake Tanganyika. The site includes a number of permanent rivers, such as the Moyowosi, Nikonga, Kigosi and Igombe in the north-east, the Malagarasi in the north and the Ugalla in the south-east, and adjacent floodplains. The rivers meet in a central area, where the lakes Sagara (328 km2) and Nyamagoma (53 km2) are located (Tanzania Wildlife Research Institute 2002). From the lakes the water flows west for c. 1 00 km as the Malagarasi River, until it forms a delta when flowing into Lake Tanganyika. The 38 -Bull ABC Vol 13 No 1 (2006) Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Malagarasi watershed forms c.30% of the Lake Tanganyika catchment (Tanzania Wildlife Research Institute 2002) and is the second-largest river basin in Tanzania after the Rufiji. Three game reserves lie within the Ramsar site: Ugalla (3,000 km2), Kigosi (9,000 km2) and Moyowosi (11,000 km2). There are no legal settlements in the game reserves and access is limited to trophy hunting by foreign tourists. Seasonal fishing and beekeeping by local communities are undertaken in Ugalla via a dispensation from the Wildlife Division. During an aerial survey in 2001, in which the authors participated, it was estimated that only 13% of the Ramsar site constitutes wet- land habitats: open water (lakes and rivers) 0.6%, permanent swamp 3.2% and seasonally inundated grassland 9.2% (Tanzania Wildlife Research Institute 2002). Thus the site comprises mainly Miombo woodland {Julbernardia-Brachystegia : 63%), with pockets of riverine forests and Combretum and Acacia bushland. The extent of papyrus swamps was estimated at 480-900 km2 (Tanzania Wildlife Research Institute 2002), which is far less than the previously proposed fig- ure of 2,000 km2 (Anon. 1994b). Permanent swamps were estimated at 1,335 km2 (based on an aerial transect survey) or 1,625 km2 (based on 1:250,000 land cover maps prepared from satellite images) (Tanzania Wildlife Research Institute 2002). A previous estimate of 3,200 km2 for the northern Moyowosi and Kigosi Game Reserves alone again appears to be a considerable overestimate. Material and methods Ground survey in 1995 The major wetlands in Tanzania were surveyed by 17 teams in 1995 in the first comprehensive waterbird count in the country (Baker 1996), which included the first ground-based waterbird survey of the eastern part of the Malagarasi ecosystem. Aerial surveys in 1971 and 1990s The population of Shoebill in Malagarasi was first estimated by an aerial survey in 1971 (Parker 1984). In the 1990s it was estimated by aerial sur- veys using Systematic Reconnaissance Flights (Tanzania Wildlife Conservation Monitoring 1991, 1998, 1999 and unpubl., Jones & Hill 1994a) and the methodology described in Norton-Griffith (1978). These surveys were undertaken by TWCM in the Moyowosi and Kigosi Game Reserves during counts of large mammals and were not specifically designed for Shoebill. However, specific attention was paid to Shoebill and Wattled Crane during surveys in September 1990, November 1992 and June 1993, and a helicopter was used in the 1992 counts in some of the areas (observers unknown, Jones & Hill 1994a). The surveys in the 1990s are difficult to compare with each other and impossible to compare with other surveys because of the wide variation in methodology, observer expertise, areas covered and areas used for extrapolation (Jones & Hill 1994a). Aerial survey in 2001 The Ramsar site, covering the vast majority of the Malagarasi ecosystem, was surveyed on 17-28 November 2001, at the end of the dry season, when water levels were at their lowest. A Systematic Reconnaissance Flight (SRF) was con- ducted with two Cessna 182 aircraft flying at a nominal altitude of c. 100 m (actual range c.60-300 m) and a mean speed of 200 km/h. Norton-Griffiths’ (1978) methodology was used. A total 76 transects with a total length of c.8,000 km were flown. Transects were spaced 5 km apart. Both aircraft flew c.6-7 hrs per day, including an average of 1. 5-2.0 hours to and from base. One skilled birder operated as rear-seat observer in each of the two planes (M. Baker and J. Anderson); the other rear-seat observer was a member of the TWCM unit specialised in count- ing mammals. Observations were limited to the strip of land visible between two markers attached to the wing struts. Each transect was divided in sub-units of 30 seconds. The rear-seat observers recorded all wildlife, selected bird species and human activities in a small tape-recorder. These data were subsequently transcribed onto data sheets and analysed using Herd Count 2000 soft- ware developed for SRF surveys. However, the combined total area of the observation strips of the two aircraft for all sampled transects covered only c.6-7% of the Ramsar site. The counts were extrapolated to cover the whole survey area using Jolly’s Method 2 of Unequal Sized Units (Jolly 1969) including the 95% confidence intervals. Incomplete transects were not included in the final data analysis. Surveys in the 1990s used a Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Bull ABC Vol 13 No 1 (2006) -39 similar census and data analysis methodology. Additionally, L. Dinesen undertook an oppor- tunistic morning and afternoon count at the end of the survey in some selected core areas for Shoebill, and had spent 14 days counting water- birds in Malagarasi in January 1995. Results The surveys in the Malagarasi ecosystem between 1972 and 2001 involving Shoebill counts can be grouped into six categories. All Shoebill counts are summarised in Table 1 . 1) An aerial count in 1971 (Parker 1984), which estimated the Shoebill population at ‘more than 300 birds’ within an estimated 200 km2 of suitable habitat. 2) Six aerial wildlife counts (SRF) undertaken by TWCM between 1990 and 2000 aimed at estimating large-mammal populations in Moyowosi and Kigosi Game Reserves. All Shoebills or cranes were also counted. Population estimates vary from 2,260 Shoebills in 1990 to 235 in 1998 following massive extrapolations. However, the areas covered by the surveys varied from 3,771 km2 to 21,870 km2 and so does the extent of Shoebill habitat used as the basis for the extrapolations. 3) The survey in 2001 of the 32,500 km2 Ramsar site, using similar methods as under 2 and resulting in an estimate of 134 Shoebills. 4) A total count in 1992 carried out by helicop- ter with specific focus on Shoebill and Wattled Crane. The presumed core areas were covered and 578 birds counted (Jones & Hill 1994a). Extrapolation resulted in an estimated popula- tion of 2,489 birds, the highest estimate for the Malagarasi. 5) A ground count of accessible swamps in the eastern part around Lakes Sagara, Nyamagoma and Masimba as part of the nationwide wet- lands survey in 1995 (Baker 1996). This pro- duced 44 Shoebills, with concentrations in Lumbe and Masimba. 6) A rapid one-day aerial count in November 2001 in perhaps less than half of the expected Shoebill core areas produced 56 birds. The surveys in 1971, 1992 and 2001 (Parker 1984, Jones & Hill 1994a, this study) and ground surveys in 1995 (Baker 1996) revealed, unsurpris- ingly, a clumped distribution of Shoebills, often located in ‘bays’ of permanent grassland swamps (not papyrus) fringed by Miombo woodland. Based on these surveys the key areas for Shoebill in Malagarasi are largely known. Table 1. Shoebill numbers recorded and estimated during aerial surveys in the Malagarasi ecosystem 1971-2001. Tableau 1. Nombre de Becs-en-sabot du Nil Balaeniceps rexcomptes et estimes pendant les comptages aeriens dans I’ecosysteme de Malagarasi en 1971-2001. 1971 1990 1992 1993 1994 1995 1998 2000 2001 2001 Estimate 300+ 2,258 2,489 1,028 689 723 235 997 134 none SE ±899 ±642 ±443 ±241 ±102 ±296 ±46 Actual count 78 578 35 45 61 15 136 9 56 Time of year Aug Sep Nov June Wet Dry May Dry Nov Nov Areas covered* SA GRs SA GRs GRs GRs GRs GRs RS SA Method** TC SRF TC SRF SRF SRF SRF SRF SRF TC Area covered (km=) 20,183 2,125 4,188 21,870 3,771 21,870 21,666 Sources: 1971: Parker (1984); 1990: Tanzania Wildlife Conservation Monitoring (1991); 1992: Jones & Hill (1 994a,b); 1993: Jones & Hill (1994a, b); 1994: Tanzania Wildlife Conservation Monitoring unpubl. via Baker (1996); 1995: Tanzania Wildlife Conservation Monitoring unpubl. via Baker (1996); 1998: Tanzania Wildlife Conservation Monitoring (1999); 2000: unpubl.; 2001 Tanzania Wildlife Research Institute (2002). * GRs = Moyowosi/Kigosi Game Reserves; RS = Malagarasi-Muyovozi Ramsar Site; SA = Specific Areas, considered to be core areas for Shoebill. ** SRF = Transect count; TC = Total count. 40 - Bull ABC Vol 13 No 1 (2006) Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Figure 1. Fisherman in Lumbe swamp, Malagarasi; the number of users in the Malagarasi floodplains has increased dramatically in recent decades placing natural resources under pressure (Lars Dinesen) Un pecheur dans le marais de Lumbe, Malagarasi; l'aug- mentation importante du nombre d'utilisateurs des zones humides de Malagarasi pendant les dernieres decennies consume une menace pour les ressources naturelles (Lars Dinesen) Figure 2. The survival of Shoebill Balaeniceps rex in Malagarasi requires 'realistic' conservation planning, edu- cation and awareness-raising, and a portion of luck (Lars Dinesen) La survie du Bec-en-sabot Balaeniceps rex a Malagarasi depend de la planification realiste des actions de conservation, d'education et de sensibilisation, et une part de chance (Lars Dinesen) Discussion The Malagarasi Shoebill population The variation in estimated numbers (ranging from 134 in 2001 to 2,489 in 1994) prompts two ques- tions: (1) What is the actual population size of Shoebill in the Malagarasi? and (2) What is the population trend since the first survey in 1971. Moreover, it seems crucial to know whether sea- : Figure 3. Shoebill Balaeniceps rex has been brought to the brink of extinction within c. 150 years since explorers reached the interior of the continent (Lars Dinesen) Environ 150 ans apres l'arrivee des explorateurs dans l'interieur du continent, le Bec-en-sabot Balaeniceps rex est au bord de l'extinction (Lars Dinesen) sonal movements to areas outside the Malagarasi ecosystem could explain the variation, as suggest- ed by Jones & Hill (1994a). There are no indications of migration out of Malagarasi. Although there are sightings of Shoebills outside the Malagarasi ecosystem (Baker 1996, Tanzania Wildlife Research Institute 2002) and outwith other possible breeding grounds in Tanzania, e.g. the Kagera and Mara swamps, there is no evidence for regular seasonal migration, nei- ther from this nor from other localities (Guillet 1978, Brown et al. 1982, Elliott 1992). However, it is expected that birds make local movements in relation to seasonal flooding regimes, food avail- ability and disturbance. A study in southern Sudan by Guillet (1978) found that birds move seasonally between nesting and fishing sites, according to the flood regime. Odd records in Tanzania are more likely caused by adverse situa- tions, e.g. birds leaving their home range in dry years due to extensive fires. Under the reasonable assumption that the bulk of Shoebills is confined to the Malagarasi ecosystem we think that the extrapolated numbers (under points 2 and 3) are crude at best. The basis for extrapolation over such large areas is thin because the distribution of Shoebill is discontinuous in space and time. The actual numbers counted are small (9-136 in six surveys) and the surface area used for extrapola- tion follows protected area boundaries rather than the extent of suitable habitat. Moreover, there are Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Bull ABC Vo! 13 No 1 (2006) -41 major differences in observer skills and most counts were undertaken as part of large-mammal surveys. A total count attempted partly by helicopter in 1992 resulted in 578 birds counted (Jones & Hill 1 994a) and densities reported of 5 birds/km2 in an area of 65 km2 and a estimated total of 2,489 indi- viduals. These figures appear crude because Shoebill is considered to be a solitary bird (Brown et al. 1982, Elliott 1992, Collar 1994). Parker (1984) estimated 0.64 Shoeb ills/km2 in suitable habitat in Moyowosi in 1971. Howard & Aspinwall (1984) counted 23 singles and five pairs in the Bangweulu swamps in Zambia during a Lechwe Kobus leche survey. The sightings by the authors in 1995 (44 birds) and 2001 (65 birds) in Malagarasi were of singles or pairs, although a clumped distribution was observed. The densities of other large swamp birds such as Wattled Crane and Saddle-billed Stork Ephippiorhynchus sene- galensis in prime habitat in the Okavango, Botswana, were 0.12 and 0.10 per km2 respective- ly (Craig & Gibson 2001). We therefore suggest a survey designed specifically for Shoebill based pri- marily on total counts and conducted by experi- enced birdwatchers (see Appendix). There has been a huge expansion of human activities in Malagarasi in recent decades (Mutch 1977, 1980, Anon. 1994a,b, Jones & Hill 1994a,b, Tanzania Wildlife Conservation Monitoring 1998, Tanzania Wildlife Research Institute 2002) and a decline in Shoebill popula- tion is to be expected. The species is specialised in both its feeding and nesting habits, and vulnerable to disturbance. It requires special fishing sites with solid platforms, standing water, plenty of fish and long periods of undisturbed fishing in order to secure adequate quantities of food (Guillet 1978). Information from the Sudd indicates that the bird is intolerant of even a low level of human distur- bance (Guillet 1978). In Malagarasi large areas of Miombo woodland adjacent to the swamps are being cleared for tobacco farming and agriculture, and the human population, consisting of farmers, fishermen, refugees and semi-nomadic Watutsie pastoralists, has increased very rapidly in recent decades. Hunting companies reported that the annual dry-season burnings and cattle grazing in the core area of the Shoebills (southern Moyowosi and Kigozi) are severe and expanding. The first rice paddies were seen to appear at the edge of key Shoebill swamps. However, quantitative land use data are lacking. Guillet (1978) considered fire and cattle to be the main threats in the Sudd. There are no data on breeding success from Malagarasi, but it is of concern that the long-liv- ing Shoebill has a low reproduction rate and non- productive populations may exist for decades in formerly productive areas. Global conservation status The Shoebills main global stronghold is the Sudd swamps in southern Sudan, where a total of 6,407 birds was counted in 1979-82 (Robertson 2001). Guillet (1978) warned that the rapid growth of the human population jeopardised the survival of Shoebill here. A serious additional threat is the planned Jonglei canal, which will drain a large part of the swamp. Uganda has c.400-600 birds, con- fined mainly to papyrus swamps (Elliott 1992). A cautious estimate of the populations in western Tanzania (Malagarasi) and northern Zambia (Bangweulu) gives not more than a few hundred birds for each site (Howard & Aspinwall 1984, Collar 1994, Leonard 2001, this study) and both populations are under increasing pressure. The populations in Mara in Tanzania, Akagera in Rwanda/Tanzania, and localities in eastern DR Congo are much smaller and, in the latter sites, under significant pressure due to the civil unrest (Kanyamibwa 2001). In sum, the handful of core Shoebill populations are under increasing threat due to disturbance, conversion of swamp habitat and locally from collectors (see also Elliott 1992 for a review) and nowhere is its survival secure. We therefore strongly support the upgrade of its glob- al conservation status from Near Threatened (BirdLife International 2000) to Vulnerable (VU Cl) (IUCN 2001, BirdLife International 2004) i.e. a population of fewer than 10,000 mature individuals and a decline of more than 10% in three generations. Al though there is no hard evi- dence for a 10% decline, empiric reports from the core localities indicate a serious and steady decline in all sub-populations. More importantly, conser- vation actions should target the few key wetlands in Zambia, Tanzania, Uganda and Sudan where the species has a reasonable chance to survive. 42 -Bull ABC Vol 13 No 1(2006) Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Acknowledgements The Danish International Development Agency DANIDA) supports the Sustainable Integrated Management of Malagarasi-Muyovozi Ramsar Site 'SIMMORS) project and the aerial surveys. The project, Isabell von Oertzen and Charles Mulokosi ire warmly thanked for their assistance. The field ;urvey in 1995 was supported by BirdLife Denmark md Tanzania, and was undertaken with H. Meltofte, 1C. Rahbek and M. Rahner. We thank Honorary Maliti and the pilots Vilfred Minja and Michael Masota for their cooperation in the field, and Clive [ones for information on earlier surveys. David Moyer, Geoffrey Howard and Julius Arinaitwe made constructive comments on drafts of this paper. References Anon. 1994a. The Malagarasi/Moyowosi/Kigosi/Ugalla Riverine Wetland Ecosystem. A Wetland of the United Republic of Tanzania to be included in the List of Wetlands of International Importance as the first Ramsar Site. Dar es Salaam: Ministry of Natural Resources and Environment, Tanzania. Anon. 1994b. The MalagarasilMoyowosil Kigosi/Ugalla Riverine Wetland Ecosystem. Required Information for the Ramsar Database. Dar es Salaam: Ministry of Natural Resources and Environment, Tanzania. Baker, N. E. 1996. Tanzania Waterbird Count. The First Coordinated Count on the Major Wetlands of Tanzania, January 1995. Cambridge, UK: Wildlife Conservation Society of Tanzania and BirdLife International. BirdLife International. 2000. Threatened Birds of the World. Barcelona: Lynx Edicions & Cambridge, UK: BirdLife International. BirdLife International. 2004. Threatened Birds of the World 2004. CD-ROM. Cambridge, UK: BirdLife International. Brown, L., Urban, E. K. & Newman, K. 1982. The Birds of Africa. Vol. 1. London, UK: Academic Press. Craig, C. G. & Gibson, S. 2001. Aerial Survey of Wattled Cranes in the Okavango Delta, Botswana, August 2001. Botswana: Unpubl. report of DG Ecological Consulting to BirdLife Botswana Crane Working Group. Collar, N. J. 1994. The Shoebill. Bull. ABC 1: 19-20. Delany, S & Scott, D. 2002. Waterbird Population Estimates. Third edn. Wageningen: Wetlands International. ■ Guillet, A. 1978. Distribution and conservation of the Shoebill (Balaeniceps rex) in the southern Sudan. Biol. Conserv. 13: 39-49. Elliott, A. 1992. Family Balaenicipitidae (Shoebill). In del Hoyo, J., Elliott, A. & Sargatal, J. (eds.) Handbook of the Birds of the World. Vol. 1. Barcelona: Lynx Edicions. Howard, G. W. & Aspinwall, D. R. 1984. Aerial cen- suses of Shoebills, Saddlebilled Storks and Wattled Cranes at the Bangweulu Swamps and Kafue Flats, Zambia. Ostrich 55: 207-212. IUCN. 2001. IUCN Red List Categories and Criteria. Version 3.1. Gland: International Union for Conservation of Nature and Natural Resources. Jolly, G. M. 1969. Sampling methods for aerial census- es of wildlife populations. E. Afr. Agricultural & Forestry J. 34: 46-49. Jones, C. & Hill, J. 1994a. The Malagarasi/Moyowosi/ Kigosi/Ugalla Riverine Wetland Ecosystem. Arusha, Tanzania: Frankfurt Zoological Society. Unpubl. report. Jones, C. & Hill, J. 1994b. The Malagarasi/Moyowosi/ Kigosi/Ugalla Riverine Wetland Ecosystem. Proposed Boundary Extension to Kigosi Game Reserve. Unpubl. report for Frankfurt Zoological Society. Kanyamibwa, S. 2001. Rwanda. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, U.K: BirdLife International. Leonard, P. M. 2001. Zambia. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, UK: BirdLife International. Mutch, G. R. P. 1977. The flora of the Moyowosi Game Controlled Area. Tanzania Notes & Records Nos. 81 & 82. Mutch, G. R. P. 1980. The larger mammals of the Moyowosi. Tanzania Notes & Records No. 84. Norton-Griffiths, M. 1978. Counting Animals. Handbook No. 1. Revised second edn. Nairobi: African Wildlife Foundation. Parker, I. S. C. 1984. Shoebill Balaeniceps rex and Wattled Crane Grus carunculatus in the Moyowosi swamp, Tanzania. Scopus 8: 24-25. Robertsen, P. 2001. Sudan. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Bull ABC Vol 13 No 1 (2006) -43 Newbury: Pisces Publications & Cambridge, UK: BirdLife International. Rose, P. M. & Scott, D. A. 1994. Waterfowl Population Estimates. Slimbridge: International Waterfowl and Wetlands Research Bureau. Tanzania Wildlife Research Institute. 2002. Aerial cen- sus in the Malagarasi-Moyowosi Ramsar Site, dry season 2001. Arusha: Tanzania Wildlife Research Institute & SIMMORS. Unpubl. report. Tanzania Wildlife Conservation Monitoring. 1991. Wildlife Census: Moyowosi-Kigosi. Arusha: Frankfurt Zoological Society. Unpubl. report. Tanzania Wildlife Conservation Monitoring. 1998. Aerial Wildlife Census: Moyowosi-Kigosi Game Reserves, May 1998. Arusha: Tanzania Wildlife Conservation Monitoring / Frankfurt Zoological Society / EU Wildlife Survey. Unpubl. report.. Tanzania Wildlife Conservation Monitoring. 1999. Wildlife Census Moyowosi-Kigosi. Arusha: Frankfurt Zoological Society. Unpubl. report. aAavej 1, 4470 Svebolle, Denmark. E-mail: dinesen- lars@hotmail. com bpO Box 14268, Arusha, Tanzania. E-mail: ruppells@hotmail. com Received 14 March 2005; revision accepted 25 October 2005 Appendix 1 . Proposed survey design Annexe 1. Protocole pour les comptages aeriens futurs All Shoebill observations by the authors (AM 09) in the Malagarasi ecosystem have been in waterlogged grassland. The stronghold is the permanent swamp areas formed by the Moyowosi, Nikonga and Gombe Rivers in and south of the Kigosi and Moyowosi Game Reserves (Jones & Hill 1994a,b). We recom- mend aerial surveys using a stratified approach with a high focus on total counts in the permanent swamps and floodplains of the ecosystem. This should be possible within one or two weeks, depend- ing on the season and keeping in mind that the max- imum extent of Shoebill habitat is <7.4,225 km2 and the area of permanent swamp c. 1,040 km2. Because Shoebill is dependent on water for both feeding and nesting, only the waterlogged plains need to be sur- veyed. At the end of the wet season, in April-May, a large area is flooded and this coincides with the expected breeding time. Shoebill has been recorded breeding in Malagarasi on a few occasions (Jones & Hill 1994a,b). Pre-count reconnaissance flights using a few days should be undertaken to potential Shoebill areas before a final decision on specific count areas is taken. Flight transects should be c.500 m apart (Craig & Gibson 2001) and flights should be at <7.250 m altitude in a high-winged air- craft. In Malagarasi individual count areas should follow the different floodplains. Count areas in key areas should be covered 1 00% and, depending on the resources, count categories could perhaps be planned with 50% (each second line) and as a minimum of 25% (each fourth line) coverage respectively. However, it is recommended as a minimum that half the survey area is covered 100%. Two skilled bird- watchers should be rear-seat observers and record Shoebills on a dictaphone, while a front-seat observ- er would record the count area codes and coordinates of all end points. Attempts to count other waterbirds or mammals should be avoided. In the data analysis of the Wattled Crane population in the Okavango, Jolly’s (1969) method for sample units of unequal size was used to calculate estimates of density and variance (Craig & Gibson 2001), and a similar method should be used in the analysis of the 50% and 25% sampled count areas. Surveys should be undertaken in both the dry and wet seasons, and indication of breeding should be noted (expected to be in the peak wet season, in April/May onwards). 44 - Bull ABC Vol 13 No 1 (2006) Status of Shoebill in Malagarasi, Tanzania: Dinesen & Baker Sierra Leone Prinia Schistolais leontica in the Fouta Djalon of Guinea, its song, distribution and taxonomic status C. R. Barlow* , R. B. Payneb, L. L. Payneb and M. D. Sorenson c Prinia de Sierra Leone Schistolais leontica dans le Fouta Djalon, Guinee, son chant, sa distri- bution et son statut taxonomique. Un couple de Prinias de Sierra Leone Schistolais leontica a ete observe le 10 octobre 1999 pres de Dalaba, dans le Fouta Djalon, Guinee (11°40’N 12°17’W). Un des oiseaux, qui a pu etre capture, avait une becquee d’insectes et presentait une plaque incu- batrice. Ceci constitue une extension considerable vers l’ouest de l’aire de repartition de cette espece menacee, ainsi que la premiere donnee de nidification. Un sonogramme du chant est presente et compare a celui de la Prinia a gorge blanche S. leucopogon , espece presentant un com- portement et un chant en duo semblables. Les differences vocales et genetiques prouvent toute- fois qu’il s’agit bien de deux especes distinctes. Figures 2-3. Sierra Leone Prinia Schistolais leontica near Dalaba, Guinea, 10 October 1999 (Laura Payne) Prinia de Sierra Leone Schistolais leontica pres de Dalaba, Guinee, 10 octobre 1999 (Laura Payne) /^Vn 10 October 1999 we observed a pair of V>/ small grey warblers at the edge of a village c. 1 km north-west of Dalaba, Guinea, at 1,160 m (1 1°40’N 12°17’W), which we identified as Sierra Leone (White-eyed) Prinias Schistolais leontica. The birds had a long tail and rounded wings, a whitish belly, distinctly buff flanks, a thin black bill, relatively long pink legs, and a cream-white iris (Figs. 2-3). They occurred together in thickets near a trail and stream by the edge of forest; they did not enter the forest (Fig. 1). One bird was car- rying insects and, when captured, was found to have an active brood patch. Its tail was bent to one side, indicating it was attending a covered nest which was not found. When moving through the vegetation, the birds held the tail low, not cocked over the back; the two species of Schistolais war- Figure 1. Habitat within 10 m of Sierra Leone Prinia Schistolais leontica near Dalaba, Guinea, with dense thick- ets, rocky outcrops and disturbed forest (Laura Payne) Habitat situe a 10 m de la Prinia de Sierra Leone Schistolais leontica pres de Dalaba, Guinee, avec des bosquets denses, des affleurements rocheux et de la foret degradee (Laura Payne) Sierra Leone Prinia in the Fouta-Djalon, Guinea: Barlow et al. Bull ABC Vol 13 No 1 (2006) -45 1 b I I fi' A' ^ A A A' A A? 8 6 N 54 2 0 1.0 2.0 s Figure 4. Sonograms of Sierra Leone Prinia Schistolais leontica and White-throated Prinia S. leucopogon. A: song of Sierra Leone Prinia at Dalaba, Guinea (original recording by CB on Sony TC-D5M recorder with Beyer M69 micro- phone in 33 cm parabola); B: song of Sierra Leone Prinia at Mt Nimba, Liberia (Chappuis 2000); C: song of White- throated Prinia at Garoua Boulai, Cameroon (Chappuis 2000). Sonograms generated with Kay Electrics DSP Sona- Graph 3500 and Printer 5509, at 234 Hz. Sonogrammes de la Prinia de Sierra Leone Schistolais leontica et la Prinia a gorge blanche S. leucopogon. A: chant de la Prinia de Sierra Leone a Dalaba, Guinee (enregistrement original par CB avec un enregistreur Sony TC-D5M et un microphone Beyer M69 dans une parabole de 33 cm); B: chant de la Prinia de Sierra Leone au Mont Nimba, Liberia (Chappuis 2000); C: chant de la Prinia a gorge blanche a Garoua Boulai, Cameroun (Chappuis 2000). Sonogrammes realises avec Kay Electrics DSP Sona-Graph 5500 et imprimante 5509, a 234 Hz. v' ' \is\ v, * vn \ •'MA '' MAA' ^ biers, S. leontica and White-chinned Prinia S. leu- copogon, differ in this behaviour from warblers in the genus Prinia (Irwin 1997). Although Bates (1931) described S. leontica cocking their tails’, Irwin (1997) suggests Schistolais simply raise the tail over the back. The birds sang both singly and together. The vocalisations were recorded and sonograms are presented here for the first time (Fig. 4a). When duetting, they uttered their notes at different rhythms, and the notes of the two birds were not closely synchronised. The songs recorded at Mt Nimba, Liberia, by Stuart Keith, were described as a tuneless, unstructured, unsynchronized duet; one bird gives rapid, high-pitched ‘sipsipsipsip...’ and the second bird gives a lower, nasal, measured ‘bur-bur-bur-bur...’.’ (Irwin 1997). Two of the three songs on Chappuis (2000) show the same phrasing and sequence of notes, so there is some repeated structure in the song (Fig. 4b). Songs at Mt Nimba and Dalaba were similar, and it appears these songs are species-typical. Song duets of Sierra Leone Prinia differ from those of its con- gener, the Central African White-chinned Prinia, whose notes are lower pitched (the short notes 5 vs. 7 kHz; the long notes 3-4 kHz vs. 4-5 kHz), and have a narrower frequency envelope (the short notes 1.0-1. 5 kHz vs. 3-4 kHz; the long notes 1 kHz vs. w kHz) (Fig. 4c). The irregular sequence of the higher pitched notes in Fig. 4 show that two birds independently gave the notes each at a differ- ent rhythm, and neither bird was in tight syn- chrony with the rhythm of a third bird with the lower, longer notes. Another recording from Foumban, Cameroon (Chappuis 2000), indicates two birds giving the lower notes at different rhythms, and a third bird giving the high notes at a third rhythm. 46 - Bull ABC Vol 13 No 1 (2006) Sierra Leone Prinia in the Fouta-Djalon, Guinea : Barlow et al. White (1962) considered Sierra Leone and White-chinned Prinias to be the same species, as did Morel & Morel (1988) when listing ‘leuco- pogon for Guinea. However, the two differ in size (leontica being smaller) and appearance (Irwin 1997), and their songs are distinct. In molecular genetic analyses, the two Schistolais are more close- ly related to the genera Camaroptera and Apalis than to Prinia subflava and Cisticola spp. (Sefc et al. 2003). The genetic sequences of their mito- chondrial ND2 gene are c.4% divergent (Sefc et al. 2003), reflecting a historical divergence of perhaps two million years for the two species. The warbler lineages Schistolais and Apalis , Prinia and Cisticola are included in the family Cisticolidae in Sibley & Monroe (1990), Sefc et al. (2003) and Dickinson (2004). Although Urban et al. (1997) included all within the family Sylviidae, recent genetic analyses suggest this assemblage, were it monophyletic, would include certain babblers and white-eyes Zosterops as well (Cibois et al. 1999, Sefc et al. 2003, Barker et al. 2004). Schistolais and many Apalis sing in duet. In contrast, most warblers of the genus Prinia , such as P. subflava, sing alone rather than in duet (Irwin 1997), although Banded Prinia P. bairdii regularly duets (Brosset & Erard 1986). Duetting behaviour in songbirds is involved in maintaining an exclusive pair-bond, and the distribution of duetting song tends to fol- low systematic relatedness rather than certain habitats (Payne 1971, Farabaugh 1982, Smith 1994, Langmore 1998, Slater et al. 2002). The observation at Dalaba constitutes a con- siderable range extension and also the first breed- ing record of this globally threatened species, which is categorised as Vulnerable (Irwin 1997, BirdLife International 2004). Considered Very local and uncommon’ (Irwin 1997), its previously known range included north-east Sierra Leone, south-east Guinea (Pic de Fon and Mt Nimba), northern Liberia (Mt Nimba and other ranges in northern Nimba county) and western Cote d’Ivoire (Bates 1930, 1931, Colston & Curry- Lindahl 1986, Gatter 1997, Irwin 1997, Demey & Rainey 2004). Pic de Fon (08°31’N 08°34’W) is c.500 km from Dalaba; Birwa Peak, in the Tingi Mountains, Sierra Leone (08°34’N 10°48’W), from where the species first was described (Bates 1930), is 339 km distant. Bates (1931) also col- lected it near Saiama, in southern Guinea just across the Sierra Leone border. The species’ range coincides with the inland massifs and an annual rainfall of 2,000-2,600 mm (Gwynne-Jones et al. 1978, Gatter 1997). Dalaba lies on the rugged, rocky massif of the Fouta Djalon, which extends as an inland plateau through western Guinea south into north-east Sierra Leone. The habitat of wet thickets and patches of secondary forest among granitic outcrops, ravines and ridges, of which we only briefly sampled a small area around Dalaba during our visit in October 1999, occurs through- out the Fouta Djalon and the species may well occur elsewhere in the region. References Bates, G. L. 1930. [New birds from West Africa.] Bull. Br. Ornithol. Cl. 51: 47-54. Bates, G. L. 1931. Account of an expedition to Sierra Leone and French Guinea on behalf of the British Museum. Ibis 1931: 446-466. BirdLife International. 2004. Threatened Birds of the World 2004. CD-ROM. Cambridge, UK: BirdLife International. Brosset, A. & Erard, C. 1986. Les oiseaux des regions forestieres du nord-est du Gabon. Vol. 1. Paris: Societe Nationale de Protection de la Nature. Chappuis, C. 2000. African Bird Sounds: Birds of North, West and Central Africa and Neighbouring Atlantic Islands. 15 CDs. Paris: Societe d’Etudes Ornithologiques de France & London, UK: British Library National Sound Archive. Cibois, A., Pasquet, E. & Schulenberg, T. S. 1999. Molecular systematics of the Malagasy babblers (Passeriformes: Timaliidae) and warblers (Passeriformes: Sylviidae), based on cytochrome b and 16S rRNA sequences. Mol. Phyl. & Evol. 13: 581-595. Colston, P. R. & Curry-Lindahl, K. 1986. The Birds of Mount Nimba, Liberia. London, UK: Br. Mus. (Nat. Hist.). Demey, R. & Rainey, H. J. 2004. The birds of Pic de Fon Forest Reserve, Guinea: a preliminary survey. Bull. ABC 11: 126-138. Dickinson, E. C. (ed.) 2004. The Howard & Moore Complete Checklist of the Birds of the World. Third edn. London, UK: Christopher Helm. Farabaugh, S. M. 1982. The ecological and social sig- nificance of duetting. In Kroodsma, D. E. & Miller, E. H. (eds.) Acoustic Communication in Birds. Vol. 2. New York: Academic Press. Gatter, W. 1997. Birds of Liberia. Robertsbridge: Pica Press. Sierra Leone Prinia in the Fouta-Djalon, Guinea: Barlow et al. Bull ABC Vol 1 3 No 1 (2006) -47 Gwynne-Jones, D. R. G., Mitchell, P. K., Harvey, M. E. & Swindell, K. 1978. A New Geography of Sierra Leone. Singapore: Longman. Irwin, M. P. S. 1997. Genera Prinia , Schistolais and Apalis. In Urban, E. K., Fry, C. H. & Keith, S. (eds.) The Birds of Africa. Vol. 5. London, UK: Academic Press. Langmore, N. E. 1998. Functions of duet and solo songs of female birds. Trends Ecol. & Evol. Biol. 13: 136-140. Morel, G. J. & Morel, M.-Y. 1988. Liste des oiseaux de Guinee. Malimbus 10: 143-176. Payne, R. B. 1971. Duetting and chorus singing in African birds. Ostrich Suppl. 9: 125-145. Sefc, K. M., Payne, R. B. & Sorenson, M. D. 2003. Phylogenetic relationships of African sunbird-like warblers: Moho ( Hypergerus atriceps ), Green Hylia (Hylia prasina) and Tit-hylia (. Pholidornis rushiae). Ostrich 74: 8-17. Sibley, C. G. & Ahlquist, J. E. 1990. Phytogeny and Classification of Birds: A Study in Molecular Evolution. New Haven & London, UK: Yale University Press. Slater, P. J. B., Gil, D., Barlow, C. R. & Graves, J. A. 2002. Male-led duets in the Moho, Hypergerus atri- ceps, and Yellow-crowned Gonolek, Laniarius bar- barus. Ostrich 73: 49-51. Smith, W. J. 1994. Animal duets: forcing a mate to be attentive./. Theor. Biol. 1 66: 221-223. Thiollay, J.-M. 1985. The birds of Ivory Coast. Malimbus 7: 1-59. Urban, E. K., Fry, C. H. & Keith, S. (eds.) 1997. The Birds of Africa. Vol. 5. London, UK: Academic Press. White, C. M. N. 1962. A check list of the Ethiopian Muscicapidae (Sylviinae). Part II. Occ. Pap. Natl. Mus. Southern Rhodesia 26: 653-694. a Birds of the Gambia, POB 279, Banjul, The Gambia, b Museum of Zoology and Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, Michigan 48109, USA. c Department of Biology, Boston University, 5 Cummington Street, Boston, Massachusetts 02215, USA. Received 19 October 2004; revision accepted 19 July 2005 Rockjumper Birding Tours Best of African, Asian, & Malagasy Birding Contact us for the most comprehensive selection of African birding tours Tour Destinations: Angola; Botswana; Cameroon; Canary Islands; Cape Verde; Egypt & Petra; Ethiopia; Gabon; Ghana; Kenya; Madagascar; Malawi; Mauritius; Morocco; Namibia; Reunion; Rwanda; Sao Tome & Principe; Seychelles; Sierra Leone; South Africa; Tanzania; Tunisia; Uganda; Zambia. Our Asian destinations include: Bhutan; China; India; Malaysia & Borneo; Papua New Guinea; Philippines; Sri Lanka; Thailand. E-mail: info@rockjumper.co.za Website: http://www.rockjumper.co.za PO Box 13972, Cascades 3202, South Africa, Tel/fa x: +27 333 471 234 48 -Bull ABC Vol 13 No 1 (2006) Sierra Leone Prinia in the Fouta-Djalon, Guinea: Barlow et al. First reliable sound recording of Golden Nightjar Caprimulgus eximius, in the rocky hills of central Mali Fmngoise Dowsett-Lemaire and Robert J. Dowsett Premier enregistrement certifie du chant de 1’Engoulevent dore Caprimulgus eximius , obtenu dans les collines rocheuses du Sahel malien. Le premier enregistrement certifie du chant roule de l’Engoulevent dore Caprimulgus eximius a ete obtenu pres de Gao, dans le Sahel malien, en juin 2004. La repasse du chant a permis d’observer les chanteurs de pres en fin de nuit et jusqu’a l’aube. Le chant est un ronronnement prolonge de 23 notes/s et apparait quasi identique a celui de l’Engoulevent terne C. inornatus (enregistre au Yemen et au sud-est du Nigeria, d’oii C. eximius est absent). Au Mali (Douentza a Gao) et ailleurs, C. eximius semble infeode aux mileux rocheux ou rocailleux pour y chanter et nicher, mais peut se nourrir dans les plaines arides environnantes. C. inornatus est egalement associe aux milieux rocheux (du moins au sud et a Test de la zone de sympatrie), ou aux prairies caillouteuses (parfois meme en milieu montagneux). II reste a etudier comment les deux especes, au chant si semblable, se separent ecologiquement dans leur zone de sympatrie. Summary. We obtained the first definite tape-recording of Golden Nightjar Caprimulgus eximius near Gao in the Sahel of Mali, in June 2004. Using playback, the singing birds were examined at close range, just before and until dawn. The song consists of a prolonged churr, of 23 notes/second, and appears virtually identical to that of Plain Nightjar C. inornatus (tape-recorded in Yemen and Nigeria outside the area of sympatry). In Mali (Douentza to Gao) and elsewhere C. eximius is clearly associated with rocky hills and stony tracts, whilst feeding can occur in the surrounding arid plains. Outside the area of sympatry, to the south and east of the range of C. eximius , C. inornatus is also associated with rocky hills, or at least pebbly grassland (sometimes in montane areas). It remains to be seen how these two similarly churring nightjars are segregated ecologically in the area of sympatry. Jackson (2003) summarised our knowledge of the voice of the Golden Nightjar Caprimulgus eximius , stressing the lack of reliable tape- recordings. Since Rothschild & Wollaston (1902) it has been known that this species has a churring song: they collected 16 specimens in the Sudan and described the song as a ‘churr’, uttered on the ground for several minutes. Fry (1988) published a sonogram, attributed to C. eximius , of a song tape-recorded by S. Keith in Kenya, but as explained by Jackson (2003) there was a typo- graphical error (an omission of some text) and the sonogram in question was in fact of Dusky Nightjar C. fraenatus (published in Chappuis 1981). C. eximius is indeed unknown in Kenya. Chappuis (1981) attributed a churring song he tape-recorded in Niger, 100 km north of Tillabery, to C. eximius , but the bird was not captured nor observed in good light (C. Chappuis pers. comm.). Therefore, after FD-L obtained a record- ing of Plain Nightjar C. inornatus from the Mambilla Plateau in eastern Nigeria which proved to be identical to the Tillabery recording, Chappuis (2000) changed his attribution of the Tillabery bird to C. inornatus. The Mambilla nightjar was tape-recorded at dusk, on 26 March 1988, singing from gravelly short grassland beside a large patch of montane rainforest at Leinde Fadali, at 1,680 m (cf. Dowsett-Lemaire 1989). The bird was not cap- tured but was seen moderately well, showing a tail of normal length (thus excluding Long-tailed Nightjar C. climacurus) and a small white wing patch. The bird was attributed to C. inornatus , in part because of the habitat and southern location, which seemed to exclude C. eximius , which is vir- tually endemic to the Sahel. A few years later a very good recording of C. inornatus from Yemen was obtained by P. Davidson. The author of the song was not very well seen (P. Davidson in litt .), Sound recording of Golden Nightjar, in Mali: Dowsett-Lemaire & Dowsett Bull ABC Vol 1 3 No 1 (2006) -49 but one C. inornatus was mist-netted in the same general area (Dymond (1996) and the song described as a prolonged churr. D. J. Pearson (in litt .), who heard C. inornatus in Ethiopia, also described the song as a churr, of moderate speed. The tape-recorded songs of C. inornatus from Mambilla and Yemen are virtually identical: the beats cover the same frequency range and are given at a rate of 22/second (Yemen) and 23/second (Nigeria) respectively. The Tillabery bird (better referred to as Caprimulgus sp.) has a churr of the same pitch, and of 23 notes/second. C. inornatus is silent in eastern Africa, being a non-breeding migrant there (Chapin 1939, Zimmerman et al. 1996). Two C. fraenatus tape-recorded in Kenya (pub- lished by Chappuis 1981 and Ranft & Cleere 1998) produce a faster churr, of 33 and 29 notes/second respectively. The recording pub- lished in 1981 is by Mrs Keith and was identified by M. E. W. North (Chappuis 1981, Jackson 2003), that published in 1998 is by D. J. Pearson, who has long field experience of East African nightjars. Trained ears can distinguish the churr of C. inornatus from the faster churr of C. fraenatus without great difficulty (D. J. Pearson in litt., FD- L pers. obs.). Tape-recording of C. eximius in Mali In June 2004 we spent five days in the Sahel of Mali, between Douentza and Gao. One of our main objectives was to obtain a tape-recording of the song of C. eximius. Robertson (2001) had list- ed C. eximius for the Ag Oua-Ag Arbech Important Bird Area, which covers an area imme- diately east and north-east of Gao. P. Robertson (in litt.) informed us that he and two colleagues had found one bird day-roosting amongst rocks, in April 1 989. This was the site of a former French fort, on a small rocky escarpment a few kilometres east of Gao; a Desert Eagle Owl Bubo bubo ascala- phus was discovered roosting on the same rock face. Rocky hills are rare around Gao. We first searched for this habitat along the road to the north, towards Kidal, but failed to find any. When scanning the surroundings from atop a sand dune, we realised that the only likely area was some way to the south-east. Thus we took the road from Gao east towards Djebok, and c.G km from Gao we were within a short distance of the escarpment and the ruins of the French fort. The escarpment is several hundred metres long but very low: the maximum height of the rock face near its centre is c. 20 m. At both sides, the escarpment slopes down in the form of broken rocks of volcanic appearance (Fig. 2). On 16 June, we established camp at 16°16’N 00°03’E, c. 50 m from the small cliff, adjacent to the only large tree in the area (a Balanites aegyptiaca the sparse vegetation consist- ed mainly of scattered small bushes and the spiky grass Schoenefeldia gracilis. A sandstorm started at 17.00 hrs, and although that did not prevent a pair of Bubo b. ascalaphus from emerging onto the cliff and bravely facing the wind, everything else was sheltering. When we awoke next day, at 04.30 hrs, the wind had calmed to occasional bursts and we heard a churring song nearby. It appeared that two pairs of nightjar were present, with both males singing. The nearer male, which was tape-recorded, sang from atop nearby rocks and bushes, including the Balanites by our tent. It responded to playback on several occasions, flying over and around us while giving a double wing- clap. The large, broad white wing patch was clear- ly visible, as were the white tail-corners; both fea- tures were also obvious when the bird was perched. In addition to the churring song, a call- note was uttered a few times (not tape-recorded) consisting of 3-4 koro, koro, koro, given at the rate of three notes per second. All four nightjars were active until almost full daylight, when the mottled golden plumage became very obvious. On the evening of 18 June we camped near Bota village, 5 km north of Douentza on the road toTombouctou (13°05’N 02°58W). Bota is at the foot of Mt Gandamia, a huge rocky mountain ris- ing several hundred metres above the surrounding plain (Fig. 3). We briefly heard a churring nightjar at dusk, but playback of the Gao tape failed to interest it, and it flew towards the plain. Next day we were woken at 04.30 hrs by a churring song coming from broken rocks in the foothills. This time playback of the Gao tape attracted a nightjar within seconds and it flew around wing-clapping or churred from a rock close to the ground. We obtained good views of the bird, especially as it remained active until almost full daylight. Eventually it landed on the ground in front of a small golden rock 6 m away from us and apparent- 50 - Bull ABC Vol 13 No 1(2006) Sound recording of Golden Nightjar, in Mali: Dowsett-Lemaire & Dowsett 0.5 1.0 1.5 s Figure 1. Sonograms of churring songs of nightjars: A: Golden Nightjar Caprimulgus eximius tape-recorded at Gao, Mali (original recording by FD-L on Panasonic RQ-L335 recorder with Hama microphone); B: Plain Nightjar C. inor- natus tape-recorded in Yemen (by P. Davidson); C: presumed C. inornatus tape-recorded in south-east Nigeria (original recording by FD-L on Sony TCM-150 recorder, modified by Mineroff, with Beyer Dynamic microphone). Sonograms generated with Avisoft Professional programme. Sonogrammes de chants roules d’engoulevents: A: Engoulevent dore Caprimulgus eximius enregistre a Gao, Mali (enregistrement original par FD-L avec un enregistreur Panasonic RQ-L333 et un microphone Hama); B: Engoulevent terne C. inornatus enregistre au Yemen (par P. Davidson); C: C. inornatus presume enregistre au sud-est du Nigeria (enregistrement original par FD-L avec un enregistreur Sony TCM- 130, modifie par Mineroff, avec microphone Beyer Dynamic) . Sonogrammes realises avec le programme Avisoft Professional. Sound recording of Golden Nightjar, in Mali: Dowsett-Lemaire & Dowsett Bull ABC Vol 13 No 1 (2006) -51 ly fell asleep until full daylight. It was unquestion- ably C. eximius. At Hombori, nearly halfway between Douentza and Gao, we had camped at the foot of another huge rocky mountain, the ‘Main de Fatma’ (15°14’N 01°48’W; Fig. 4), on 15-16 June, and heard brief snatches of a churring night- jar at dusk and dawn; we saw one bird flying off towards the surrounding plain at dusk. It could not be tape-recorded as it sang for too short a peri- od, but we believe it too was C. eximius , as the habitat was identical to that at Bota and the loca- tion intermediate between Bota and Gao. When we subsequently analysed the song on the Gao tape, we found that it had a tempo and frequency so similar to those of C. inornatus of Nigeria and Yemen (and to that of Caprimulgus sp. from Tillabery, Niger) as to be indistinguishable. It had 23 beats per second, and was of similar pitch. Fig. 1 presents sonograms of the two species’ songs. The songs share features of tempi, general frequency and shape of beats. It is possible that the song of C. eximius is a little louder at the level of 1 kHz, but more recordings would be necessary to test whether this is a permanent feature of the species’ voice. Our recording of C. inornatus from Mambilla (south-east Nigeria) is only slightly lower pitched than that from Yemen. Habitat of C. eximius in Mali At the Gao and Douentza sites, C. eximius was found to be associated with rocky hills, whether low or very tall, although some birds were seen, at both sites and at Hombori, flying towards the sur- rounding plain to feed. The habitat could be described succinctly as a more arid version of the rocky places favoured by Freckled Nightjar C. tristigma. The latter was not found at any of the three rocky hills explored in June 2004, but, on a previous visit, on 2 March 2002, a C. tristigma was clearly heard at Gono, c. 25 km east of Douentza (15°05’N 02°44’W), at the foot of a large amphitheatre which is part of the same rocky mas- sif as Mt Gandamia, and stretches almost all the way to Boni (15°04’N 02°13’W). The cliff at Gono faces south, the ‘cirque’ is more sheltered and the vegetation somewhat more developed than at Bota, with more large Acacia (especially A. albida, whereas the dominant tree at Bota is A. tor - tilis). Birds associated with rocks at Gono includ- Figure 2. Low rocky escarpment near Gao, Mali, where Golden Nightjar Caprimulgus eximius was tape-recorded (Franqoise Dowsett-Lemaire) Escarpement rocheux de faible hauteur pres de Gao, Mali, ou l’Engoulevent dore Caprimulgus eximius a ete enregistre (Fran^oise Dowsett-Lemaire) Figure 3. Tall rocky mountain near Douentza, Mali; Golden Nightjars Caprimulgus eximius occupy the lower levels with broken rocks (Franchise Dowsett-Lemaire) Haute montagne rocheuse pres de Douentza, Mali; les Engqulevents dores Caprimulgus eximius occupent les niveaux inferieurs (Franqoise Dowsett-Lemaire) ed Fox Kestrel Falco alopex , Stone Partridge Ptilopachus petrosus , Rock Pigeon Columba livia , Barn Owl Tyto alba , Spotted Eagle Owl Bubo africanus , Rock Martin Hirundo fuligula, Cliff Chat Myrmecocichla cinnamomeiventris, Familiar Chat Cercomela familiaris , Neumann’s Starling Onychognathus neumanni and House Bunting Emberiza striolata. Neither Cercomela familiaris nor Onychognathus neumanni (nor, as already mentioned, C. tristigma) were found at Bota, where the main cliff is more exposed as it faces 52 - Bull ABC Vol 13 No 1 (2006) Sound recording of Golden Nightjar, in Mali: Dowsett-Lemaire & Dowsett west, and the area appears too sparsely vegetated for these birds. Other species associated with rocks or cliffs at Bota were, in addition to various rap- tors, mainly Ptilopachus petrosus, Tyto alba , Bubo b. ascalaphus (calling from the tall cliff), B. africanus (heard and seen among the broken rocks in the foothills), Columba livid. Mottled Swift Apus aequatorialis , Pallid Swift A. pallidus , Hirundo fuligula , Myrmecocichla cinnamomeiventris, and both Emberiza striolata and Cinnamon-breasted Rock Bunting E. tahapisi. At the Main de Fatma at Hombori the vegetation is thinner and the main rock-associated bird species include Ptilopachus petrosus, Columba livia, Tyto alba, Bubo b. ascala- phus, B. africanus (also in the foothills, as at Bota), Apus aequatorialis, A. pallidus, Hirundo fuligula, Myrmecocichla cinnamomeiventris, Emberiza strio- lata and E. tahapisi. Finally, the small rocky escarpment near Gao is in very much more arid country, and bird species recorded adjacent to the rocks were very few: no diurnal raptors nor Ptilopachus petrosus, no swifts nor martins, but only one pair of Bubo b. ascala- phus, one pair of Tyto alba (seen at dawn returning to a rock crack only metres away from that inhab- ited by the larger Bubo), a pair of Desert Larks Ammomanes deserti with young (among the bro- ken rocks), one Blackstart Cercomela melanura and a few Emberiza striolata. Discussion That C. eximius may be associated with rocky hills and escarpments in the same way as C. tristigma is not clearly mentioned in the literature, but at least an association with stony ground is well estab- lished. Indeed, in an arid environment there may be little else but rocky hills or at least broken rocks to provide the necessary shelter for day-roosting and nesting. Rothschild & Wollaston (1902: 20-21) described the habitat in Sudan as ‘general- ly sloping ..., coarse and gravelly, often with a good many scattered stones and tufts of grass’. The Figure 4. ‘Main de Fatma , Hombori, Mali. A charring nightjar song, almost certainly belonging to Golden Nightjar Caprimulgus eximius, was heard here (Fran^oise Dowsett-Lemaire) ‘Main de Fatma , Hombori, Mali, ou un chant roule a ete entendu, presque certainement celui de l’Engoulevent dore Caprimulgus eximius (Fran^oise Dowsett-Lemaire) Sound recording of Golden Nightjar, in Mali: Dowsett-Lemaire & Dowsett Bull ABC Vol 13 No 1 (2006) -53 main habitat description in Fry & Harwin (1988: 183) is evidently based on the same reference: ‘typ- ically on sloping, coarse gravelly and stony ground with scattered tufts of grass and dwarf scrub’. Even though Rothschild & Wollaston (1902) never saw C. eximius perched on a tree (a detail repeated by Fry & Harwin 1988), we can vouch that this does indeed occur: the bird tape-recorded near Gao fre- quently perched in trees while singing, as well as on rocks. Few other data on habitat have been published, C. eximius having remained one of the least-studied African nightjars. Lynes (1923: 370), however, added a few details of interest in his notes from the Sudan: ‘We found this lovely nightjar here and there ... on yellow to reddish- yellow gravelly or stony tracts ... Along our route across the flat plains such ground is more or less confined to isolated patches’. On the basis of these publications and our observations in Mali we conclude that the species is associated with stones and rocks, including bro- ken rocks on small escarpments or situated in the foothills of large mountains. The similarity between the songs of C. eximius and C. inornatus is a problem. Normally, churring nightjars that are largely sympatric chur at differ- ent speeds: see for example C. inornatus and C. fraenatus. The latter produces a chur that is 30-50% faster than that of its congener, with 29-33 notes/second versus 22-23, and there is no doubt that birds, which have a better appreciation of time-resolution than humans, can make the dis- tinction. Another striking example is that of C. cli- macurus , which produces the fastest of all African nightjars’ churrs, at 42 beats/second (measured on a sonogram made from the recording published by Chappuis 1981). This is so fast as to sound almost like a purr. C. eximius is near-endemic to the Sahel and unknown east of Sudan (Fry & Harwin 1988). The range of C. inornatus extends further south and south-east (mainly as a non-breeding migrant; e.g. Chapin 1939, Zimmerman et al. 1996), and east to Yemen, but the two species overlap in a large part of the Sahel. C. inornatus appears to occupy a very broad range of habitats, from very dry to reasonably green, having even been found breeding in montane grassland on Mt Nimba, Liberia (based on a nestling collected by A. D. Forbes- Watson, in Colston & Curry-Lindahl 54 -Bull ABC Vol 13 No 1 (2006) 1986, Fry & Harwin 1988); this appears rather similar to the habitat where FD-L tape-recorded it on the Mambilla Plateau. In Bui National Park in western Ghana in March 2005, some (presumed) C. inornatus were singing in two places in recently burnt woodland with pebbly or rocky substrate (pers. obs.). In the more rocky of the two sites there were also several C. tristigma. The churring song sounded similar to that taped in Nigeria and Yemen, i.e. clearly slower than that of sympatric C. climacurus, and one C. inornatus was seen very well by day in the same area. In Faro National Park in Cameroon one presumed C. inornatus (i.e. with a similar slowish churr) was heard on a stony hill next to C. tristigma (pers. obs. March 1999). It was not seen but this was in Sudanian woodland way south of the range of C. eximius. Possibly C. inornatus is ecologically well segre- gated from C. eximius within the Sahel, but that remains to be proven. Any ornithologist coming across a churring nightjar in West Africa should be equipped with at least a copy of a tape-recording (of either C. eximius , C. inornatus or Caprimulgus sp. from Tillabery in Chappuis 2000, all similar), and play it in the early morning to discover which species reacts. From our experience in Mali, and even more so in the forests of northern Congo (where we undertook extensive playback experi- ments with Brown Nightjar C. binotatus and sus- pected Itombwe Nightjar C. prigoginev. Dowsett- Lemaire & Dowsett 1998), nightjars respond far better in the hour just preceding dawn than in the early part of the night. They then return to their territory on a full stomach and are ready to chal- lenge a competitor. Moreover, good views of the singer can be obtained as dawn approaches. Acknowledgements We thank P. Davidson, N. Dymond and G. M. Kirwan for their useful comments. Thanks are due to Joanne Nicholson at the British Library (Sound Archive, London) for producing the sonograms. References Chapin, J. P. 1939. Birds of the Belgian Congo. II. Bull. Amer. Mus. Nat. Hist. 75: 1-632. Chappuis, C. 1981. Illustration sonore de problemes bioacoustiques poses par les oiseaux de la zone ethiopienne. Alauda 49: 35-58. (With disc no. 12.) Chappuis, C. 2000. Oiseaux dAfrique (African Bird Sounds), 2. West and Central Africa. 1 1 CDs. Paris: Sound recording of Golden Nightjar, in Mali: Dowsett-Lemaire & Dowsett Societe d’Etudes Ornithologiques de France & London, UK: British Library National Sound Archive. Colston P. R. & Curry-Lindahl K. 1986. The Birds of Mount Nimba, Liberia. London, UK: Br. Mus. (Nat. Hist.). Dowsett-Lemaire, F. 1989. Physiography and vegeta- tion of the highland forests of eastern Nigeria. Tauraco Res. Rep. 1: 6-12. Dowsett-Lemaire, F. & Dowsett, R. J. 1998. Vocal and other peculiarities of Brown Nightjar Caprimulgus binotatus. Bull. ABC 5: 35—38. Dymond, J. N. 1996. The Plain Nightjar Caprimulgus inornatus in Yemen. Sandgrouse 17: 132-133. Fry, C. H. 1988. Skulls, songs and systematics of African nightjars. Proc. Sixth Pan-Afr. Orn. Congr:. 105-131. Fry, C. H. & Harwin, R. M. 1988. Order Caprimulgiformes. In Fry, C. H., Keith, S. & Urban, E. K. (eds.) The Birds of Africa. Vol. 3. London, UK: Academic Press. Jackson, H. D. 2003. Song of the Golden Nightjar, Caprimulgus eximius. Ostrich 74: 241-242. Lynes, H. 1925. On the birds of North and Central Darfur, with notes on the West-Central Kordofan and North Nuba Provinces of British Sudan. Part 4. Ibis (12) 1: 344-416. Ranft, R. & Cleere, N. 1998. A Sound Guide to Nightjars and Related Nightbirds. CD. Robertsbridge: Pica Press & London, UK: British Library National Sound Archive. Robertson, P. 2001. Mali. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas of Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, UK: BirdLife International. Rothschild, N. C. & Wollaston, A. F. R. 1902. On a collection of birds from Shendi, Sudan. Ibis (8) 2: 1-33. Zimmerman, D. A., Turner, D. A. & Pearson, D. J. 1996. Birds of Kenya and Northern Tanzania. London, UK: Christopher Helm. Le Pouget, 30440 Sumene, France. E-mail: Dowsett@aol.com Received 30 November 2004; revision accepted 6 July 2005 Sound recording of Golden Nightjar, in Mali: Dowsett-Lemaire & Dowsett Bull ABC Vol 13 No 1 (2006) - 55 Notes on the breeding biology of Plain Swift Apus unicolor on Gran Canaria, Canary Islands Eduardo Garcia-del-Rey Notes sur la biologie de reproduction du Martinet unicolore Apus unicolor a Gran Canaria, lies Canaries. La biologie de reproduction du Martinet unicolore Apus unicolor a ete etudiee de fa$on quantitative a Gran Canaria. En moyenne, la premiere ponte est deposee le 2 1 avril et la seconde le 24 juin (une troisieme ponte a egalement ete notee), prolongeant la periode de nidification sus- pectee auparavant d’environ 1,5 mois. Les tailles moyennes des premieres et deuxiemes pontes sont tres semblables (1,97 oeufs ±0.06 vs. 1,95 oeufs). Le pourcentage des deuxiemes pontes est de 70% et la reussite moyenne des premieres et deuxiemes pontes de 74,44±6,77 et 64,29±5,05 respectivement. Ces caracteristiques refletent la position geographique et l’aspect oceanique des lies Canaries par rapport a la region temperee septentrionale et l’Afrique tropicale. Summary. The breeding biology of Plain Swift Apus unicolor was quantitatively studied on Gran Canaria. Mean laying date of the first clutch was 21 April and that for the second clutch 24 June (a third clutch was also recorded), extending the previously suspected breeding period by c. 1.5 months. The mean sizes of the first and second clutches were very similar (1.97 eggs ±0.06 vs. 1.95 eggs). The percentage of second broods was 70% and the mean breeding success of first and second broods was 74.44±6.77 and 64.29±5.05 respectively. These intermediate life-history traits reflect the geographical location and oceanic aspect of the Canaries compared to the northern temperate region and tropical Africa. Although Plain Swift Apus unicolor breeds reg- ularly in the Canary Islands, its reproductive biology has never been documented quantitatively (Garcia del Rey 2001, Martin & Lorenzo 2001). As the Canaries lie midway between the tropics and the temperate zone, it might be expected that Plain Swifts show life-history traits intermediate between those of swifts of the northern temperate Figure 1. Study site at Puente Silva, Gran Canaria (Eduardo Garcia-del-Rey) Site de 1’etude a Puente Silva, Gran Canaria (Eduardo Garcia-del-Rey) region and those of tropical Africa. The abun- dance of insect life is thought to have a significant impact on the breeding biology of swifts, and may explain why breeding coincides with the wet sea- son in the tropics and summer in the temperate zone (Chantler 1999). Weather, temperature, sun- shine, wind velocity and precipitation all influence feeding conditions and prey abundance, hence affecting the breeding biology of many swifts (Lack 1973, Chantler 1999). The avian breeding season on the Canaries has been variably stated to extend from March to August (Bannerman 1963, Bannerman & Bannerman 1965, Cramp 1985, Chantler 1999) or April-September (Martin & Lorenzo 2001). I present here information on some aspects of the breeding biology of Plain Swift, in particular laying dates, clutch size, num- ber of clutches and breeding success. Methods This study was undertaken during 2003 on Gran Canaria, Canary Islands (28°00’N 15°30’W). The climate on this oceanic archipelago is Mediterranean with cool, wet winters and hot, dry summers, and is influenced by the local north-east trade winds, the proximity of the Sahara on the 56 - Bull ABC Vol 13 No 1 (2006) Notes on the breeding biology of Plain Swift: Garcia-del-Rey African continent and the high altitude of the cen- tral and westerly islands (Marzol-Jaen 1984). The study site was Puente Silva (Fig. 1.), a 400-m-long bridge with an internal cavity 2 m high and 10m wide, near Agaete, in the north-west of the island. Swifts could access the interior of the bridge through narrow tubes and nested on the concrete floor, which was very convenient for nest inspec- tion. Data were collected once a month from April to September during afternoon visits. The follow- ing assumptions were made in order to reconstruct the breeding phenology: • Mean incubation period (measured from the laying to the hatching of the last egg) was assumed to be 20 days. Most swift species incubate for this length of time and incubation starts when the clutch is complete (Lack 1973). • Laying was assumed to occur at an interval of two days (Lack 1973). • Minimum nestling period was assumed to be 37 days (Cramp 1983). Nestling swifts are known to be able to slow down their growth in bad weather when food is scarce, thus saving energy for vital functions but thereby prolong- ing the nestling period considerably. Hence the nestling period can vary by up to three weeks (Lack 1973). It is assumed that the weight curve for a young swift does not vary greatly with season in the stable weather of Gran Canaria (see Marzol-Jaen 1984 for details of climate). All eggs found were touched to check if incu- bation had started. Nests in which eggs were laid but not incubated were excluded from the study. In order to estimate the age of the nestlings, each chick was assigned to one of four categories: 1 = 1- 7 days old (pink/naked chick); 2 = 8-15 days old (dark chick with very tiny or no pin feathers on wings); 3 = 16-30 days old (dark chick with pin feathers with brush-tipped feathers of several sizes); 4 = 31-37 or more days old (full-feathered chick with short to long tail ready to fledge). The assignment to categories was aided with photo- graphs taken during the course of the study (Figs. 2- 3). When a brood with chicks of different sizes was found, the age category assigned was based on the largest chick. Once the age of a nestling was estimated, the date on which the first egg in the clutch was laid was extrapolated. Figure 2. Nestling of category 1 = 1-7 days old, naked and pink (Eduardo Garcia-del-Rey) Oisillon de la categorie 1 = age de 1-7 jours, nu et rose (Eduardo Garcia-del-Rey) Figure 3. Nestling of category 2 = 8-15 days old, dark with very tiny or no pin feathers on wings (Eduardo Garcia-del-Rey) Oisillon de la categorie 2 = age de 8-15 jours, fonce avec de tres petites plumes ou sans plumes aux ailes (Eduardo Garcia-del-Rey) For each nest the following parameters were determined: laying date (date of first egg), clutch size, breeding success, and occurrence of a second clutch (laid in same nest from which a first brood had fledged successfully). The percentage of second clutches was also calculated. As no birds were ringed, it has been assumed that all second (and third) clutches were laid in the same nest as the first. All statistical analyses were performed using SPSS 11.0 and results presented as mean ± standard error. Notes on the breeding biology of Plain Swift: Garcia-del-Rey Bull ABC Vol 13 No 1 (2006) -57 Figure 4. Nestling of category 3 = 16-30 days old, dark with brush-tipped feathers of several sizes (Eduardo Garcia-del-Rey) Oisillon de la categorie 3 = age de 16-30 jours, fonce avec des plumes de plusieurs tailles (Eduardo Garcia-del- Rey) Figure 3. Nestling of category 4 = 31-37 or more days old, fully feathered with short to long tail, ready to fledge (Eduardo Garcia-del-Rey) Oisillon de la categorie 4 = age de 31-37 jours ou plus, ayant routes ses plumes avec queue courte ou longue, pret a quitter le nid (Eduardo Garcia-del-Rey) Results The mean laying date of first clutches was 21 April (51.57±5.22; 1 = 1 March) and that for second clutches 24 June (1 1 3.91 ±4.32). One pair laid a third clutch on 21 July (see Table 1). Four pairs laid their first egg on the inspection day or the day before (assuming a laying interval of two days). This allowed checking the assumptions 1 and 3 (see Methods), and both were found to be correct. The mean sizes of first and second clutches were very similar (1.97±0.06 for first clutches and Table 1. Mean laying date (1=1 March), average clutch size and mean breeding success (±SE) of first, second and third clutches/broods. Sample size in parentheses. Tableau 1. Date moyenne de ponte (1=1 mars), taille moyenne des pontes et reussite moyenne (±SE) des pre- mieres, deuxiemes et troisiemes pontes / nichees. Taille de I’echantillon entre parentheses. First Second Third clutches/broods clutches/broods clutches/broods (n=30) (n=21-22) M) Laying 21 April 24 June 21 July date 51.57±5.22 (n= 22) 11 5.91 ±4.32 143 Clutch 1.97±0.06 1.95±0.05 1 size Mi) Breeding 74.44±6.77 64.29±5.05 100 success % Mi) 1.95±0.05 for second clutches) (Table 1), and no statistically significant differences were found between these (Student t-test: t=0.18, df=49, T-^0.05). The number of second clutches was 70%. Of the 30 pairs that laid a first clutch, the majority (19 pairs) raised all young (i.e. 100% breeding success), six pairs lost half the chicks (50% breeding success), one pair had a 33.3% breeding success and four pairs did not produce any offspring. Thus, mean breeding success of first broods was 74.44±6.77 (Table 1). Of the 21 pairs laying a second clutch, only six raised all young, whereas the majority (15 pairs) raised 50% of the chicks. The mean breeding success for second broods dropped to 64.29±5.05 but was not statis- tically significant (Mann-Whitney U-test: U=235.5, P>0.05). The causes of failure in the first half of the breeding season are unknown. Discussion The present study suggests that the breeding sea- son of the Plain Swift in Gran Canaria commences in early March and ends in mid September, extending the previously suspected breeding peri- od by c.1.5 months. However, mean laying dates presented here (Table 1) should be viewed with caution, as the precise date on which each pair in the colony laid their first egg is unknown. Any variation at the different stages of breeding will also affect the laying date, and Apus are known to have variable incubation and nestling periods 58 - Bull ABC Vol 13 No 1 (2006) Notes on the breeding biology of Plain Swift: Garcia-del-Rey (Lack 1973). Daily visits to the colony during sev- eral years are needed to improve the data present- ed here (i.e. to establish the precise laying date and incubation and nestling periods of each nest, and to confirm the two-day laying interval). Most of the second clutches were laid immediately after the fledging of first clutches (i.e. c. 1 day later). The nestling period of first broods could therefore not exceed c. 37 days, which facilitated more accurate estimation of the laying date of the second clutch’s first egg. Only three broods of second clutches had nestling periods longer than 37 days (43, 48 and 49 days, respectively). However, as the mean incu- bation period (20 days) and the minimum nestling period (37 days) were used to calculate the mean laying date in the colony, the extension of the breeding phenology found is this study is justified. My results on clutch size (first clutch size = 1.97±0.06 eggs) agree with those of other authors (e.g. Cramp 1985, Chantler 1999), who also sug- gested that double broods (70% in this study) are frequent in Plain Swift. Clutch size in the genus Apus varies between one to three eggs (Lack 1973). Mean clutch for Common Swift Apus apus in northern Africa (Cramp 1985) is the same as that of Plain Swift in Gran Canaria. However, breeding success of first clutches in Plain Swift is higher than the mean reported for Common Swift (74.4% vs. 58-65%) (Chantler 1999). Life- history traits associated with breeding are assumed to be determined by natural selection to maximise the production of offspring (Baker 1938, Lack 1954). The pioneering work of Lack (1973) point- ed out the highly adaptable breeding seasons in swifts. For example, Common Swift raises only one brood per year at Oxford, England, whereas Pallid Swift raises two successive broods in the Mediterranean region, and the Afrotropical White-rumped Swift A. cajfer raises three broods each year (Lack 1973). The length of the warm- weather period has been suggested by Lack (1973) to explain this difference. Plain Swift in Gran Canaria seems to have adapted to the stable weath- er on the island, resulting in a high number of sec- ond clutches (occasionally even a third), low chick mortality and high breeding success. Both clutch size and number of breeding attempts per season seem to reflect the geographical location and oceanic aspect of these islands: clutch size is indeed lower than in the northern temperate region, whilst the number of breeding attempts is higher than in the north but lower than in Africa. Acknowledgements My study was funded by Sociedad Ornitologica Canaria (SOC). I thank Guillermo Delgado for helping during the study and Charlie Collins for his constructive comments on a draft of this paper. References Baker J. R. 1938. The evolution of breeding seasons. In de Beer, G. R. (ed.) Evolution: Essays on Aspects of Evolutionary Biology Presented to Professor E. S. Goodrich on his Seventieth Birthday. Oxford: Clarendon Press. Bannerman, D. A. 1963. Birds of the Atlantic Islands. Vol. 1 . Edinburgh & London, UK: Oliver & Boyd. Bannerman, D. A. & Bannerman, W. M. 1965. Birds of the Atlantic Islands. Vol. 2. Edinburgh & London, UK: Oliver & Boyd. Cramp, S. (ed.) 1985. The Birds of the Western Palearctic. Vol. 4. Oxford: Oxford University Press. Chantler, P. 1999. Family Apodidae (swifts). In del Hoyo, J., Elliott, A. & Sargatal, J. (eds.) Handbook of the Birds of the World. Vol. 5. Barcelona: Lynx Edicions. Garcia del Rey, E. 2001. Checklist of the Birds of the Canary Islands. Santa Cruz de Tenerife: Publicaciones Turquesa. Lack, D. 1954. The Natural Regulation of Animal Numbers. London, UK: Oxford University Press. Lack, D. 1973. Swifts in a Tower. London, UK: Chapman & Hall. Martin, A. & Lorenzo, J. A. 2001 . Aves del Archipielago Canario. La Laguna: Francisco Lemus. Marzol-Jaen, M. V. 1984. El clima. In Afonso, J. (ed.) Geografia de Canarias. Vol. 1. Santa Cruz de Tenerife: Editorial Interinsular Canaria. Departamento de Ecologla, Facultad de Biologla, Universidad de La Laguna, E-38260 La Laguna, Tenerife, Canary Islands, Spain. E-mail: avesecot@redkbs. com Received 24 September 2004; revision accepted 24 May 2005 Notes on the breeding biology of Plain Swift: Garcla-del-Rey Bull ABC Vol 13 No 1 (2006) -59 Some factors affecting foraging and habitat of Ring Ouzels Turdus torquatus wintering in the Atlas Mountains of Morocco Colin Ryalla and Kevin Briggs b Quelques facteurs affectant la recherche de nourriture et l’habitat du Merle a plastron Turdus torquatus hivernant dans l’Atlas marocain. L’etude presentee avait pour objet d’examiner le role des baies de genevrier comme source principale de nourriture des Merles a plastron Turdus torqua- tus hivernant dans F Atlas marocain. Des deux especes de genevrier examinees, l’Oxycedre Juniperus oxycedrus est la plus repandue, mais se trouve en faible densite dans d’autres milieux boises. Dans les 43 sites etudies dans le Moyen et Haut-Atlas, les Merles a plastron ont ete observes uniquement a l'interieur ou pres des bois de Genevrier de Phenicie J. phoenicea. Malgre le fait que le nombre de baies de genevrier mures dans de tels bois varie beaucoup (de 2,7 x 104 a 2,6 x 10(1 par ha), il apparait que le niveau des degats causes aux arbres par des coupes, et le niveau general de perturbation, sont des facteurs plus determinants plus importants pour la presence des Merles a plastron que le nombre de baies. Les bois de genevriers rencontres pendant cette etude sont degrades et vieillissants, et il n’y a pas de regeneration, ce qui suggere un declin a long terme avec des implications potentielles pour la future disponibilite de genevriers et la survie des Merles a plastron. Summary. This study aimed to shed light on the role of juniper berries as the principal food source for Ring Ouzels Turdus torquatus wintering in Morocco’s Atlas Mountains. Of the two juniper species surveyed, Prickly Juniper Juniperus oxycedrus was the most widespread, but occurred at low densities in other types of woodland. Of 43 sites surveyed in the Middle and High Atlas, Ring Ouzels were only seen in or close to Phoenician Juniper J. phoenicea woodland. Although the number of ripe juniper berries in such woodland ranged from 2.7 x 104to 2.6 x 1 O'’ per ha, the degree of damage to the trees from cutting, indicative of general levels of distur- bance, appeared to be a stronger determinant of Ring Ouzel presence than did the number of berries. Juniper woodland encountered in this study was in a degraded and ageing state with no recruitment by younger trees, suggesting a long-term decline with potential implications for juniper availability and Ring Ouzel survival in the future. Ring Ouzel Turdus torquatus breeds in upland areas of Europe and Fennoscandia, and win- ters around the Mediterranean, North Africa and the Middle East (Snow & Perrins 1998). Nominate T. t. torquatus , which breeds in Britain and Fennoscandia, winters in southern Spain and north-west Africa, predominantly in the Atlas mountains, from Morocco to Tunisia (Wernham et al. 2002). Although populations in continental Europe ( 7. t. alpestris) appear largely stable, there has been a decrease in numbers and a contraction of range in Spain and Britain (Heath et al. 2000, Wotton et al. 2002) and the species is now included on the UK Red List (Gregory et al. 2002). The decline has been attributed to a range of factors including habitat change, disturbance, global climate change, predation, pollution, increased competi- tion from Blackbirds T. merula , and problems on the migration routes or wintering grounds (Tyler & Green 1994, Murray et al. 1998, Stott et al. 2002), but the exact reason remains unclear (Burfield 2002). Burfield (2002) suggested that as UK birds appear to share wintering grounds with birds from stable continental populations, factors causing this decline are most likely to be acting in the breeding grounds or migration routes. Nevertheless, the ecology of Ring Ouzels during the 3-6 months when they are migrating and win- tering outside their breeding area is poorly understood. The species’ winter food requirements are not well understood, although they have been report- ed mainly eating juniper berries in Morocco and Algeria (Heim de Balsac 1931, Heim de Balsac & Mayaud 1962, Arthur et al. 2000). In the Sierra 60 -Bui ABC W 13 No 1 (2006) Ring Ouzels wintering in Morocco: Ryall & Briggs Nevada of Spain, Zamora (1990) found juniper berries, supplemented by arthropods, to constitute the Ring Ouzel’s main diet in winter. Zamora also suggested that this restricted diet reflected the lim- ited choice of food in this area, as Ring Ouzels wintering elsewhere in Spain ate other berry species where available. Four species of juniper occur in Morocco: Prickly Juniper Juniperus oxycedrus , Phoenician Juniper J. phoenicea , Spanish Juniper J. thurifera and Common Juniper J. communis. The first three are common at moderate to high altitudes in Morocco, but Common Juniper is rare, growing only on high mountains (Jahandiez & Maire 1931). In Morocco, Ring Ouzels are common to locally abundant winter visitors, especially in the Central and Eastern High Atlas (Thevenot et al. 2003). They are rare and irregular in the western Middle Atlas, but more regular in the eastern part. Thevenot et al. (2003) state that they occur in open coniferous woodland on stone slopes from 1,000 m to 2,700 m, especially among J. phoenicea and J. thurifera at 1,800-2,200 m or in mixed J. oxycedrus / Quercus ilex woodland, often near rivers or waterholes. The present study of Ring Ouzels overwinter- ing in the Atlas Mountains focused primarily on their relationship with juniper and factors which may influence the availability of juniper berries, and hence the survival of Ring Ouzels in winter. Methods During two visits to Morocco in the winters of 1993 and 2000, a total of 43 sites was examined along two transects through the Middle and High Atlas, respectively, plus an additional site (3.1) in the central High Atlas (Fig. 1). Sites representative of a range of habitats were assessed for the pres- ence of Ring Ouzel, juniper and other berry- bearing species, potential food for Ring Ouzels, as they were encountered along the transects. Food availability and pressures on food sources were evaluated. Figure 1. Map of the Middle Atlas and High Atlas Mountains of Morocco showing the positions of Transects 1 and 2 and Site 3.1. Carte du Moyen et Haut-Atlas marocain, indicant la localisation des Transects 1 et 2 et du Site 3.1. Ring Ouzels wintering in Morocco: Ryall & Briggs Bull ABC Vol 13 No 1 (2006) -61 Transects Transect 1 ran c.110 km, north to south from Azrou, across the Middle Atlas, to Tounfite and Midelt on the north-east slopes of the High Atlas (Fig. 1). The preliminary visit, in December 1993, aimed to locate Ring Ouzels and potential berry sources, focusing on juniper. Qualitative observa- tions were made at nine sites (1.1-1. 9). Juniper- rich sites were located with the aid of information from Eaux et Forets (Moroccan Water and Forestry Department) and from Berber farmers and shepherds. Transect 2, in January 2000, ran primarily east to northeast along the northern slopes of the High Atlas from Demnate in the west to Midelt, c.230 km to the east, and crossed Transect 1 at Tounfite; so sites 1.7 and 2.19 are the same. On the second transect, 33 locations (2.1-2.33) representing a series of distinct habitats, at least 1 ha in extent, ranging from ploughed arable land to forest, were assessed as they were encountered. However, greater emphasis was given to areas containing juniper and other berry-bearing species as these were likely to be more important to Ring Ouzels. An additional juniper-rich site (3.1) c.20 km south of Taddert was surveyed (Appendix 3). Ring Ouzel density At locations where Ring Ouzels were observed, on Transect 2 and site 3.1, they were counted along one 100 x 30 m transect through each site. Surveys were carried out in the morning only, between 09.00 and 12.00 hrs. Presence or absence of other Turdus spp. and their scats on and around juniper trees was noted. Juniper tree parameters Quantitative data on juniper trees were collected in Transects 2 and site 3.1, i.e. canopy diameter, tree height, number of trees per ha in 10 w 10 m2 quadrats chosen at random (see Table 2 and Appendix 3). The degree of cutting damage to trees was scored on a scale of 0-5 (0 = no visible damage; 5 = only stump remaining). Stands of juniper were classified as ‘undegraded’ if the mean scored was <2.5 and ‘degraded’ if they scored >2.6. At some sites (sites 2.1-2.19, 2.29 and 3.1) the number of juniper stumps per ha was counted and at Tounfite the diameter of juniper trunks at 0.25 m above ground was also recorded. Juniper berry parameters Ripe and unripe berries were counted per 0.25 m2 of tree canopy surface on the north, south, east, west aspects and at the top of ten randomly select- ed Phoenician and Prickly Juniper trees at selected sites. The number of ripe and unripe juniper berries per ha was extrapolated by calculating the surface area of juniper trees. The proportion of aborted or insect-parasitised berries was calculated at several sites and the berry diameter of Phoenician Juniper and Prickly Juniper berries was measured with a micrometer. Casual observations and coordinates Signs of pressures on juniper tree density, survival and hence habitat quality were noted, as well as other factors that might effect survival of Ring Ouzels, e.g. habitat degradation, disturbance and proximity of sites to water. Other berry-bearing species were recorded along with topography, alti- tude and compass orientation of escarpments. For Transect 1 bearings or altitude were approximated using maps, but for Transects 2 and site 3.1, a Magellan GPS 2000XL global positioning system was employed. Results In total, 43 sites were investigated in the two phas- es of this study over an altitude range of 1,003-2,208 m (Appendices 1-2), and 11 distinct habitat types were discerned (Table 1). Occurrence of juniper Three species of juniper were encountered: Prickly, Phoenician and Spanish Juniper. Prickly Juniper formed bushes or small trees averaging 2.8 m in height and, being recorded at 21 of the 43 sites at 1,231-2,208 m, was the most wide- spread species. It grew mainly as a secondary species in stands of Phoenician Juniper as 5-20% (mean 9%) of the trees present or in Holm Oak Quercus ilex woodland as 5-30% (mean 21%). At two sites (1.9 and 2.2), Prickly Juniper was the dominant tree species. Phoenician Juniper woodland was encoun- tered mostly at the eastern end of the High Atlas, around Tounfite and south of Midelt, but also in the Central High Atlas, south of Marrakech at 1,918 m (3.1), at 1,003 m near Demnate (2.1) and in lower numbers in the Middle Atlas at 1,800 and 2,000 m (1.4 and 1.6). It grew as rounded 62 - Bull ABC Vol 13 No 1 (2006) Ring Ouzels wintering in Morocco: Ryall & Briggs Table 1. Occurrence of Ring Ouzels Turdus torquatus and other thrushes Turdus spp. in 1 1 different habitats surveyed in 1993 and 2001, and the presence of berry bearing species. Tableau 1. Presence du Merle a plastron Turdus torquatus et d’autres grives Turdus spp. dans 1 1 habitats differents exam- ines en 1993 et 2001 , et la presence d’especes produisant des baies. Vegetation type and condition* Berry-bearing species present** Number of sites surveyed Number of sites with Ring Ouzels Other thrushes present Holm Oak woodland Prickly Juniper J. oxycedrus 7 0 Blackbird T. merula Undegraded Phoenician Juniper woodland Phoenician Juniper J. phoenicea Prickly Juniper J. oxycedrus (Spanish Juniper J. thurifera) (Hawthorn Crataegus monogyna) (Mistletoe Viscum cruciatum) (Dog Rose Rosa canina) 7 6 Blackbird T. merula Mistle Thrush T. viscivorus Song Thrush T. philomelos Redwing T. iliacus Degraded Phoenician Juniper woodland Phoenician Juniper J. phoenicea Prickly Juniper J. oxycedrus (Hawthorn C. monogyna ) 5 0 Blackbird T. merula Mistle Thrush T. viscivorus Redwing T. iliacus Undegraded/degraded Prickly Juniper woodland Prickly Juniper J. oxycedrus (Lentisc Pistacia lentiscus) 2 0 Blackbird T. merula Mistle Thrush T. viscivorus Redwing T. iliacus Degraded Spanish Juniper woodland Spanish Juniper J. thurifera (Prickly Juniper J. oxycedrus ) 1 0 None Scrub woodland Hawthorn C. monogyna Mistletoe V. cruciatum Dog Rose Rosa canina (Prickly Juniper J. oxycedrus) 6 1 Blackbird T. merula Mistle Thrush T. viscivorus Song Thrush T. philomelos Redwing T. iliacus Cedar woodland None 5 0 Blackbird T. merula Plantation— pine or olive None 3 0 None Low herb and tussock grassland None 2 0 None Arable land None 4 0 None Bare stony ground None 1 0 None * Condition of vegetation: Undegraded = mean score for tree damage of 0-2.5; Degraded = mean score for tree damage of 2.6-5.0 **Scarce species in parentheses conical trees on bare stony ground, averaging 3.5 m in height, at 1,780-1,928 m at 15 of the 43 sites studied, at densities of 7.6-84 trees per ha and was invariably the dominant species. At Tounfite, some individuals reached >8 m with trunk diameters up to 0.6 m (Fig. 2). Table 2 pres- ents parameters for juniper trees at seven sites con- taining Phoenician Juniper, and further details are provided in Appendix 3. Spanish Juniper, albeit severely degraded by cutting, was encountered at Inifif (1.5), near Col du Zad, in the Middle Atlas. In the High Atlas, it was found as an occasional secondary species to Phoenician Juniper at sites 2.19 and 3.1 at alti- tudes of c. 1,920 m (Appendices 1-2). The trees were all c.3 m in height. Condition of juniper trees All juniper-rich sites had damaged trees, ranging in severity from removal of a few lateral branches to total destruction leaving splintered stumps (Figs. 3-4). On a scale of 0-5, all Phoenician Juniper sites contained damaged trees with nearly half rated at above 3 (moderate damage to only stump remaining) (Table 2). The situation was similar for Holm Oak and Prickly Juniper (Appendix 2). Local people with donkeys carrying bundles of juniper wood were frequently encountered. Fig. 5d shows that damage to Phoenician Juniper correlated positively with the number of juniper stumps present (r = 0.788) and negatively with presence of Ring Ouzels (Fig 5b: r= -0.733), but not with number of ripe berries (Fig 5a: r = -0.300) or altitude (Fig 5f: r = 0.047). Ring Ouzels wintering in Morocco: Ryall & Briggs Bull ABC Vol 13 No 1 (2006) -63 Figure 2. Large Phoenician Juniper Juniperus phoenicea tree of c.8 m in near-perfect condition, near Tounfite (Site 2.19) in the eastern High Atlas, Morocco. Denuded ground from overgrazing can be seen (Colin Ryall) Grand Genevrier de Phenicie Juniperus phoenicea d’envi- ron 8 m en etat quasi parfait, pres de Tounfite (Site 2.19) dans le Haut-Atlas marocain oriental. On peut voir le sol denude par le surpaturage (Colin Ryall) Figure 3. Large juniper tree in advanced stage of progres- sive destruction from firewood collection near Aguelmame de Sidi Ali (Site 1.4) in the Middle Atlas, Morocco (Colin Ryall) Grand genevrier en etat avance de destruction progressive par la coupe de bois de chauffe, pres d’Aguelmame de Sidi Ali (Site 1 .4) dans le Moyen Atlas marocain (Colin Ryall) Stumps of younger trees usually showed re- sprouting from the base but this was absent in older stumps. All juniper woodland was heavily grazed by sheep and browsed by goats, with the ground totally denuded of vegetation (Fig. 3) and with copious animal droppings. No juniper seedlings or young trees were found. It was common to encounter juniper woodland in the process of being cleared for agriculture, e.g. at Demnate and Figure 4. Stump of a large juniper destroyed from fire- wood collection near Tounfite (Site 2.16) in the eastern High Atlas, Morocco (Colin Ryall) Souche d’un grand genevrier detruit par la coupe de bois de chauffe, pres deTounfite (Site 2.16) dans le Haut- Atlas marocain oriental (Colin Ryall) Tounfite, and even lone junipers in ploughed fields, e.g. west of Boumia and south of Azrou. Juniper berries The number of berries on juniper and the ratio of ripe to unripe were very erratic, between sites and between trees at each site (Table 2). Mean ripe berry densities for Phoenician Juniper trees ranged from 191-1,309 per m2of canopy and in Prickly Juniper from 0-239 per m2* with many trees devoid of berries. Where present Phoenician Juniper always contributed the majority of berries. The number of ripe berries in stands of Phoenician Juniper ranged from 2.7 x 1 04— 1 .2 x 105per ha (mean = 0.7 x 1 05) for degraded sites and 1.1 x 105-2.6 x 106 per ha (mean = 0.9 x 106) for undegraded sites (Table 2). The mean ripe berry crop for all juniper species at all sites was 6.9 x 105per ha. At most sites, some juniper berries were abort- ed (shrivelled and lacking pulp) or parasitised by insects (a small exit hole, or white scale and lack- ing pulp). At site 2.2, Prickly Juniper dominated, 9.3% of berries were aborted or parasitised, for Phoenician Juniper it was 18% at site 2.16, 6% at 2.17, 0.33% at 2.18 and 0.2% at 2.19 (overall mean = 6.1%). The mean berry diameters for ten trees at site 2.15 were 10.2 mm for Prickly Juniper and 10 mm for Phoenician Juniper and in both species, ripe berries were red-brown and sweet when ripe. 64 -Bull ABC Vol 13 No 1 (2006) Ring Ouzels wintering in Morocco: Ryall & Briggs Table 2. Juniper tree parameters and fruit crop at seven Phoenician Juniper Juniperus phoenicea rich sites on Transect 2 and site 3.1, and occurrence of Ring Ouzels Turdus torquatus Tableau 2. Parametres des genevriers et production de baies dans sept localites riches en Genevrier de Phenicie Juniperus phoenicea le long du Transect 2 et au Site 3.1 , et presence de Merles a plastron Turdus torquatus Site No. Tree spp. Mean Mean number Mean tree Juniper Mean berry count Berries per ha Ring Ouzels present* canopy of each condition stumps /m2 canopy (SE) per diameter Juniper sp. rating per ha 100 '50 m (m)(S£) per ha (SE) (0-5)** (SE) ripe unripe ripe unripe transect (SE) 2.15 Jp 3.75 33.9 1.5 0 436 712 3.3 x 105 5.3 x 105 14 (0.34) (4.3) (106) (180) (4.04) Jo 2.17 1.5 2 0 0 0 0 (0.30) (0.52) 2.16 Jp 3.14 7.6 3.5 9.2 1033 158 1.2 x 105 1.9 x 104 0 (0.22) (0.76) (0.55) (189) (63) Jo 2.08 2.2 3.5 0 0 0 0 (0.24) (0.73) 2.17 Jp 3.39 18.4 2.4 7 448 68 1.5 x 105 2.2 x 104 1 (0.30) (1.9) (1.8) (137) (14) (0.70) Jo 2.4 12.8 3.7 239 0 2.8 x 104 0 (0.86) (4.2) (276) 2.18 Jp 3.82 83.6 2.5 11.9 759 58 1.5 x 106 1.5 x 105 3.4 (0.43) (10.8) (3.17) (168) (23) (0.45) Jo 2.64 3.6 3 0 0 0 0 (0.31) (3.5) 2.19 Jp 5.62 40 2 5.2 1309 92 2.6 x 106 1.8 x 105 7.8- (0.37) (2.23) (1.82) (224) (80) (3.45) Jo 2.9 16 2 0 0.8 0 0 (0.78) (4.66) (0.9) Jt 3.15 0.4 1 - - - (0.10) (0.45) 2.29 Jp 2.12 20 4 30.8 191 157 2.7 x 104 2.2 x 104 0 (0.55) (3.16) (5.13) (191) (157) 3.1 Jp 3.63 24.8 2.5 2 214 63 1.1 x 105 3.2 x 104 3.4 (0.30) (2.41) (1.22) (77) (45) (0.97) Jo 3.63 19.2 2.5 170 0 6.7 x 104 0 (0.36) (2.79) (162) Jt 3.1 0.8 2.5 319 480 3.9 x 103 5.8 x 103 (0.33) (0.55) (20) (76) 'Jp = J. phoenicea; Jo = J. oxycedrus, Jt = J. thurifera "Condition of vegetation: Undegraded = mean score for tree damage of 0-2.5; Degraded = mean score for tree damage of 2.6-5.0 (see Methods, Juniper tree parameters for further details) Other potential food sources Table 1 shows that other berry species, namely Hawthorn Crataegus spp., Dog Rose Rosa canina , Red-berried Mistletoe Viscum cruciatum and Lentisc Pistacia lends cus, were present in damper sites, such as valley bottoms, northern slopes and where overgrazing was less severe. Presence of Ring Ouzels Ring Ouzels, seen at close quarters, appeared to be nominate T. t. torquatus. Although other species of thrush were recorded in all six types of berry rich habitat (Table 1), Ring Ouzels were only encoun- tered in undegraded Phoenician Juniper wood- land, except once where one was seen feeding on Ring Ouzels wintering in Morocco: Ryall & Briggs Bull ABC Vol 13 No 1 (2006) -65 Figure 5a Phoenician Juniper tree condition vs. log berry bount {r = 4).3fl©} Figure 5d Phoenician Juniper tree condition vs. number tree stumps (r =0.703} I 8 I 6 - 40 i 30 I 20 B 10 * 0 * ♦ \ ▼ 0 1 2 3 4 5 T ree condition score T ree condition score Figure 5b Ring Ouzels vs. Phoenician Juniper condition (r = -0.733} T ree condition score Figure 5c Ring Ouzels vs. altitude (r = -0.085) n 15 JZ iq 1 10 © N o 5 1,800 1,850 1,900 Altitude (m) 1,950 Figure 5f Phoenician Juniper tree condition vs. altitude ( r = 0.047) 1,820 1,840 1,860 1,880 1,900 1,920 1,940 Altitude (m) Figures 5a-f. Correlation coefficients ( r ) for a series of parameters relating to Ring Ouzels Turdus torquatus and juniper in the High Atlas. Coefficients de correlation ( r ) pour une serie de parametres concernant le Merle a plastron Turdus torquatus et le genevrier dans le Haut-Atlas marocain. small rosehips in scrub close to sparse Phoenician Juniper (site 1.8). It is noteworthy that all sites where Ring Ouzels were seen or indicated by plen- tiful scats were less than 500 m from a source of water. Ring Ouzels were mainly in fast-moving flocks of c.4-30 birds accompanied by small num- bers of other Turdus spp. (Blackbird T. merula , Mistle Thrush 77 viscivorus, Song Thrush 77 philomelos and Redwing 77 iliacus). There was no correlation between altitude and number of Ring Ouzels seen (Fig. 5c). Near Tounfite, Ring Ouzels’ preference for Phoenician Juniper was attested by frequent sight- ings and copious droppings, on and around the trees. Such dense scat deposits were absent in other types of woodland. Ring Ouzels were not evident at sites containing Prickly Juniper without Phoenician Juniper, whether Prickly Juniper dom- inated (sites 1.9 and 2.2 only) or was secondary to Holm Oak (e.g. sites 2.3, 2.4 and 2.14). As shown in Table 2, no Ring Ouzels were recorded where Phoenician Juniper was seriously degraded (mean tree damage score >2.6), and Fig. 5b shows a strong negative correlation between condition of Phoenician Juniper trees and number of Ring Ouzels present (r= -0.733). However, the number 66 -Bull ABC Vol 13 No 1 (2006) Ring Ouzels wintering in Morocco: Ryall & Briggs of ripe berries in Phoenician Juniper correlated weakly with number of Ring Ouzels seen (r = 0.355, Fig. 5e). No Ring Ouzels were seen in the Middle Atlas during the 1993 visit (Transect 1) although at A'in Nokra, Berber shepherds said they sometimes saw them in small groups in winter. Discussion This study, albeit small in scale, is a first attempt to quantify factors that may influence the survival of Ring Ouzels wintering in the Atlas Mountains and indicates areas for further elucidation. Juniper as a food source for Ring Ouzels All Ring Ouzels were seen in or close to Phoenician Juniper, often accompanied by smaller numbers of other Turdus spp., closely paralleling the observations of Snow (1952), Heim de Balsac (1931) and de Juana & Santos (1981). Heim de Balsac (1931), Blondel (1962) andThevenot et al. (2003) considered both Prickly Juniper and Phoenician Juniper berries a major part of their winter diet. In the Spanish Sierra Nevada too, Ring Ouzels feed almost exclusively on juniper during the winter (Zamora 1990; pers. obs.) and their occurrence is strongly correlated with berry availability (Jordano 1993). This apparent speci- ficity for juniper may apply only during midwin- ter because, as reported by other workers, Ring Ouzels occur in varied habitats during migration and then exploit other species of berry, in addition to juniper. In this study, Ring Ouzels were only seen feed- ing in Prickly Juniper where it was with Phoenician Juniper, but not where it grew among Holm Oak. This may reflect overall berry avail- ability. The berries of the two species are so simi- lar, physical and to taste, that selection of one over the other seems unlikely. The mean ripe berry crop for all juniper sites (both juniper species) of 6.9 x 105 per ha closely matches that for Common Juniper (7 x 105per ha) recorded in the Sierra Nevada (Garcia et al. 1996) but, like Jordano (1993), we found that the crop varied widely between areas. As a monoecious species (bearing male and female flowers on the same plant), all Phoenician Juniper trees can potentially bear berries, whereas in Prickly Juniper and Spanish Juniper, being dioecious (bearing male and female flowers on separate plants), only female trees (c.50%) can do so. The zero berry counts in Prickly Juniper at several sites may be due to a concentration of male plants or to a local berry failure in female trees. Jordano (1991) pointed out that the monoecious state in Phoenician Juniper is variable but is more than 90% in Morocco. Thus Prickly Juniper must usually contribute a small proportion of the total berry crop where both species occur. As wide-ranging, opportunistic feeders, Ring Ouzels may be attracted primarily to extensive areas of juniper, e.g. Phoenician Juniper wood- land, with high berry densities and low levels of disturbance, thus maximising foraging success, rather than to isolated patches or individual fruit- ing trees, e.g. Prickly Juniper, outnumbered amongst non-berry bearing species, e.g. by Holm Oak, though these may serve as stop-off points for migrants. The condition of Phoenician Juniper trees, as well as number of berries, seems to be a key deter- minant of the presence of Ring Ouzel. We only found Ring Ouzel at five sites where cutting dam- age to Phoenician Juniper was low, but not in degraded stands, even with good berry crops. At an intermediate level of cutting, where trees still produce a good berry crop, disturbance from the frequent visits by small-scale wood collectors and livestock may keep Ring Ouzels away much of the time — a factor which also operates in parts of their breeding range (Burfield 2002). However, not all berries on a juniper tree are edible. They take two years to ripen so, in autumn, trees contain both ripe and unripe berries, the ratio being variable (Table 2). In addition, a vari- able proportion of berries are either aborted or parasitised by insects, resulting in berries lacking pulp and with reduced nutritional value (Ionesco & Sauvage 1969). Garcia et al. (1999) found that Ring Ouzels rejected aborted and parasitised Common Juniper berries in the Sierra Nevada, thus necessitating greater foraging effort. The pro- portion of unpalatable berries can be substantial; we found levels of aborted or parasitised ranged from 0.2-18% (mean 6.1%), butTraveset & Sans (1994) recorded moth infestation levels to range from 3-50% of the crop in Phoenician Juniper in the Balearic Islands. A further key factor in determining Ring Ouzels’ choice of feeding site appears to be prox- imity to water. In this study, locations where the Ring Ouzels wintering in Morocco: Ryall & Briggs Bull ABC Vol 1 3 No 1 (2006) -67 birds were found were all within a few hundred metres of a water source. Ring Ouzels need to drink regularly whilst feeding on juniper berries. Heim de Balsac (1931), Arthur et al. (2000) and Thevenot et al. (2003) have commonly observed them drinking at rivers and waterholes in Morocco, as they do also in the Sierra Nevada when feeding on juniper berries (pers. obs.). This issue clearly needs further investigation. Occurrence of juniper Prickly Juniper was widespread, and like Quezel (1980), we found it associated with either Holm Oak or Phoenician Juniper woodland as a second- ary or subdominant species. Phoenician Juniper was less widespread but where present was the domi- nant tree, forming open woodland interspersed with smaller numbers of Prickly Juniper and Holm Oak. Trees averaged 3.3 m in height, although near Tounfite some reached 8 m, the maximum for this species (Maire et al. 1932), with some trunk diam- eters of 0.6 m, indicating trees more than 500 years old, based on annular rings of stumps. In Emberger’s (1938) day, Phoenician Juniper was very widespread in both the Middle and High Atlas up to 2,200 m, but we found it to be frequent if patchily distributed at 1,780-2,208 m in the High Atlas, and scarce in the Middle Atlas. Spanish Juniper has an altitudinal range of 1,800-3,150 m (Emberger 1938). Its apparent rarity in our study reflects the limits of altitudes visited. Its presence at 2,000 m near Inifif in the Middle Atlas (1.5) was also noted by Sauvage (1956). Juniper damage and decline Most of our sites had trees showing moderate to severe damage, which concurs with Quezel & Barbero (1981), who described stands of Phoenician Juniper as very degraded by human influence and livestock. We commonly saw local Berbers with donkeys, and even trucks, laden with juniper and Holm Oak wood. Of course, damage may be less severe at sites more remote than those in our study. All three juniper species are used for burning for cooking, heating and construction by the Berbers (Auclair 1996), and Phoenician and Spanish Junipers are used for livestock feed in droughts (Ionesco & Sauvage 1969). We noted that stumps of smaller junipers often re-sprouted but, in older trees, some more than 500 years old, this was absent. This loss of regen- erative ability with age was recognised by Emberger (1938) and Metro & Sauvage (1955). This habitat degradation is further exacerbated by the lack of recruitment of young trees due proba- bly to a combination of overgrazing and drought, seen also in the Sierra Nevada (R. Zamora pers. comm.), and Quezel & Barbero (1981) noted a complete lack of juniper regeneration in the region of the Atlas Mountains. Despite a statement to the contrary by Arthur et al. (2000), juniper is no longer being used sus- tainably in the Atlas. Auclair (1996) pointed out that traditional controls used to work well to pre- serve mountain forest but population increase in parts of the High Atlas is resulting in permanent forest loss; indeed, wood removal is twice the rate of production and stocking levels twice the sus- tainable level. Lone junipers in vast areas of ploughed arable land on the lower mountain slopes and plains tes- tify to large-scale clearance of juniper woodland. We are late in a long-term process of deforestation. In Roman times more than half of North Africa was densely forested (Blondel & Aronson 1999), whereas 17% remains in Morocco, including oak, cedar and juniper forest. Conacher & Sala (1998) observed that agricultural clearance and deforesta- tion in the mountains of North Africa intensified from the late- 18th century due to excessive wood- cutting and overgrazing by sheep and goats. This scenario involves the long-term fragmentation and destruction of a key resource for Ring Ouzels. As Zamora (1990) and Jordano (1993) found in Spain, the Ring Ouzel is most probably the main dispersion vector for Juniper species in North Africa. Conclusions We focused on the occurrence of Ring Ouzels in relation to species of juniper, berry crop and degree of damage to trees. It must be recognised that this was a small-scale study, with Ring Ouzels only being detected at seven sites, and so the fol- lowing conclusions must therefore be considered provisional. The well-established link between wintering Ring Ouzels and juniper is confirmed, but this link is primarily with Phoenician Juniper, which contributes far more berries than Prickly Juniper, and therefore offers the most productive foraging option. 68 - Bull ABC Vol 13 No 1 (2006) Ring Ouzels wintering in Morocco: Ryall & Briggs The presence of Ring Ouzels correlated with the condition of Phoenician Juniper. Where trees were severely damaged, indicative of a high level of human disturbance, there was no evidence of Ring Ouzel visits, even where ripe berries were plentiful. Several factors are resulting in a long-term process of juniper woodland decline: • unsustainable harvesting for firewood and forage • loss of regenerative ability in the ageing stock of juniper trees • an ageing population of juniper due to lack of recruitment of young trees from overgrazing, agricultural clearance and drought. Burfield (2002) points out that, if UK birds share their wintering grounds with those from sta- ble continental populations, the factors causing their decline are most likely to be acting in the breeding grounds or migration routes. However, at present, little is known of the ecology or move- ments of the two races, T t. torquatus and T. t. alpestris, during the 5-6 months they are migrat- ing and wintering in the mountains of North Africa. Neither is it known how much juniper is needed to support this Ring Ouzel population, particularly in a year with a poor berry crop and/or water is scarce near berry sources. The total acreage of juniper woodland is unknown, as is its condition and the rates of frag- mentation and loss. Juniper berry availability may not yet be a limiting factor for the species but as destruction continues this point must eventually come. Poor foraging in winter, for whatever rea- son, means poorer condition for migration and breeding. Our study covered a small number of more accessible sites, during a short part of the winter period and did not include Spanish Juniper sites, which occur at higher altitudes. Nevertheless, our findings indicate aspects that future, more exten- sive studies could focus on: • how much juniper is required to support the current Ring Ouzel population, in view of the variability of berry production • the extent and status of juniper-rich habitat in Algeria and less accessible parts of the Moroccan Atlas • the condition and rate of loss of the juniper woodland that remains • the status of other berry species and their importance for migrating and wintering Ring Ouzels • the importance and availability of water as a factor limiting the Ring Ouzel’s ability to exploit available berry supplies. Acknowledgements We thank Farnborough College for financial support for this work throughout, along with the British Trust for Ornithology and Bird Exploration Fund for their assistance with the 1993 phase of the study. Thanks also to Prof. Jacques Franchimont, Zin Labaadin Afhaz and Mick Green for their assistance on the 1 993 research trip, to Jeanne Capey for help with translation and to the late Rae Vernon for his helpful information. Patrick Bergier commented on a draft of this paper. References Arthur, D. S. C., Ellis, P. R., Lawrie, R. & Nicoll, M. 2000. Observations of wintering Ring Ouzel and their habitat in the High Atlas Mountains, Morocco. Scottish Birds 21: 109-115. Auclair, L. 1996. L’Appropiation communautaire des forets dans le Haut Atlas marocain. Cahiers des Sciences Humaines 32: 177-194. Blondel, J. 1962. Migration prenuptiale dans les Monts des Ksours (Sahara septentrional). Alauda 30: 1-29. Blondel, J. & Aronson, J. 1999. Biology and Wildlife of the Mediterranean Region. Oxford: Oxford University Press. Burfield, I. J. 2002. The breeding ecology and conser- vation of the Ring Ouzel Turdus torquatus in Britain. Ph.D. thesis. Queens’ College, University of Cambridge. Conacher, A. J. & Sala, M. 1998. Land Degradation in Mediterranean Environments of the World: Nature and Extent, Causes and Solutions. Chichester: John Wiley & Sons. Emberger, L. 1938. Les Arbres du Maroc et Comment les Reconnaitre. Paris: Lacrosse. Garcia, D., Gomez, J. M., Hodar, J. A. & Zamora, R. 1996. Ecologia reproductiva del enebro en Sierra Nevada: factores que determinan la regeneration natural de las poblaciones. In Chacon, J. & Rosua, J. L. (eds.) First Conf. Lnt. Sierra Nevada: Conservacion y Desarrollo Sostenible, Univ. of Granada, vol. 2: 441-453. Garcia, D., Zamora, R., Gomez, J. M. & Hodar, J. A. 1999. Bird rejection of unhealthy fruits reinforces the mutualism between juniper and its avian dis- persers. Oikos 85: 536-544. Ring Ouzels wintering in Morocco: Ryall & Briggs Bull ABC Vol 13 No 1 (2006) -69 Gregory, R. D., Wilkinson, N. I., Robinson, J. A., Brown, A. E, Hughes, J., Procter, D. A., Gibbons, D. W. & Galbraith, C. A. 2002. The status of birds in the United Kingdom, Channel Islands and Isle of Man: an analysis of conservation concern 2002-7. Br. Birds 95: 410-450. Heath, M., Borgreve, C. & Peet, N. 2000. European Bird Populations: Estimates and Trends. Cambridge, UK: BirdLife International. Heim de Balsac, H. 1931. Les lieux d’hivernage du Merle a plastron ( Turdus torquatus) en Algerie. Alauda 3: 250-256. Heim de Balsac, H. & Mayaud, N. 1962. Les Oiseaux du Nord-Ouest de lAfrique. Paris: Lechevalier. Ionesco, T. & Sauvage, C. 1969. Fichier des especes cli- max. AlAwamia 32: 105-124. Jahandiez, E. & Maire, R. 1931. Catalogue des Plantes du Maroc: Spermatophytes et Pteridophytes. Vol. 1. Paris: Lechevalier. Jordano, P. 1991. Gender variation and expression of monoecy in Juniperus phoenicea (L.) (Cupressaceae). Bot. Gaz. 152: 476-485. Jordano, P. 1993. Geographical ecology and variation of plant-seed disperser interactions: southern Spanish junipers and frugivorous thrushes. Vegetation 107-108: 85-104. de Juana, E. & Santos, T. 1981. Observations sur l’hivernage des oiseaux dans le Haut Atlas (Maroc). Alauda 49: 1-12. Maire, D. R., Guinochet, M. & Faurel, L. 1952. Flore de LAfrique du Nord. Paris: Lechevalier. Metro, A. & Sauvage, C. 1955. La Nature au Maroc 1: Flore des vegetaux ligneux de la Mamora. Rabat: Societe des Sciences Naturelles et Physiques de Maroc. Murray, R. D., Holling, M., Dott, H. E. M. & Vandome, P. 1998. The Breeding Birds of South-east Scotland: A Tetrad Atlas 1988-1994. Edinburgh: Scottish Ornithologists’ Club. Quezel, P. 1980. Biogeogaphie et ecologie des coniferes sur le portour mediterraneen. In Pesson, P. (ed.) Actualites dEcologie Forestiere: Sol, Flore, Faune. Paris: Gauthier-Villars. Quezel, P. & Barbero, M. 1981. Contribution a l’etude des formations pre-steppiques a genevriers au Maroc. Bol. Soc. Brot., Ser. 2 53: 1 137-1160. Ryall, C. & Green, M. 1994. Le Merle a Plastron (Turdus torquatus)-, espece europeenne hivernante peu connue au Maroc. Porphyrio 6: 3-6. Sauvage, C. 1956. Compte rendu floristique de l’excur- sion marocaine du VUIeme Congres International de Botanique (14 juin-24 juin 1954). Travaux de ILnstitut Scientifiques Cherifien Serie Botanique No. 8. Snow, D. W. & Perrins, C. M. (eds.) 1998. The Birds of the Western Palearctic. Concise Edition. Oxford: Oxford University Press. Stott, M., Callion, J., Kinley, I., Raven, C. & Roberts, J. 2002. The Breeding Birds of Cumbria: A Tetrad Atlas 1997-2001. Keswick: Cumbria Bird Club. Thevenot, M., Vernon, R. & Bergier, P. 2003. The Birds of Morocco: An Annotated Checklist. BOU Checklist No. 20. Tring: British Ornithologists’ Union & British Ornithologists’ Club. Traveset, A. & Sans, A. 1994. Insect frugivory in Juniperus phoenicea ( Cupressaceae ) in Cabrera Island (Balearic Archipelago). Boll. Soc. Hist. Nat. Baleares 37: 141-149. Tyler, S. J. & Green, M. 1994. The status and breeding ecology of the Ring Ouzel Turdus torquatus in Wales, with reference to soil acidity. Welsh Bird Rep. 7: 78-89. Wernham, C., Toms, M., Marchant, J., Clark, J., Siriwardena, G. & Baillie, S. 2002. The Migration Atlas: Movements of the Birds of Britain and Ireland. London, UK: T. & A. D. Poyser. Wotton, S. R., Langston, R. H. W. & Gregory, R. D. 2002. The breeding status of the Ring Ouzel Turdus torquatus in the UK in 1999. Bird Study 49: 26-34. Zamora, R. 1990. The fruit diet of Ring-Ouzels ( Turdus torquatus) wintering in the Sierra Nevada. Alauda 58: 67-70. a School of Earth Sciences & Geography, Kingston University, Penrhyn Road, Kingston upon Thames, Surrey KT1 2EE, UK. E-mail: colin. ryall@ntlworld. com ^2 Osborne Road, Farnborough, Hants GUI 4 6PT, UK Received 24 December 2004; revision accepted 5 July 2005 70 - Bull ABC Vol 13 No 1 (2006) Ring Ouzels wintering in Morocco: Ryall & Briggs Appendix 1. Summary of the sites in Transect 1 . Annexe 1. Apergu des sites le long du Transect 1. Site no. Site name Location Approx. Site description Juniper Other berry Ring altitude (m) spp. condition species Ouzels frequency (score) 1.1 Aguelmame SW of Azrou 1,800 Scrub None N/A Hawthorn . Affenourir (MA) lake, grazed grassland Mistletoe, Rose 1.2 Aguelmame SW of Azrou 1,800 Cedar forest None N/A None . Affenourir (MA) little ground cover 1.3 Aguelmame de S of Azrou 1,800 Scrub None N/A Hawthorn . Sidi Ali (MA) adjacent arable land, lake, river Mistletoe, Rose 1.4 Aguelmame de S of Azrou 1,800 Cedar forest J p scarce 4 None . Sidi Ali (MA) (moribund) J o scarce / 1.5 Inifif (MA) S of Azrou 2,000 Jt woodland forestry exclosure J r moderate 3 None - 1.6 Ain Nokra (MA) S of Azrou 2,000 Q ilex woodland J o common 3 Hawthorn . river valley Jp scarce / Mistletoe, Rose 1.7 Tounfite (HA) WSW of 1,928 J p woodland J p common 2 Hawthorn + (2.19) Midelt adjacent arable J o moderate 2 Mistletoe, Rose J t rare / 1.8 Asselim (HA) S of Midelt 1,600 Scrub J p moderate 3 Rose + adjacent juniper arable J o rare 1.9 S of Khenifra S of Khenifra / J o scrub Jo dominant 1 Rose Key: MA = Middle Atlas; HA = High Atlas J p = Juniperus phoenicea; J o = J. oxycedrus; J t = J. thurifera / = Data not collected Condition of vegetation: Undegraded = mean score for tree damage of 0-2.5; Degraded = mean score for tree damage of 2.6-5.0 (see Methods, Juniper tree parameters for further details) + = Ring Ouzels present; - = Ring Ouzels absent Appendix 2. Summary of the sites in Transect 2 and of Site 3.1 . Annexe 2. Apergu des sites le long du Transect 2 et du Site 3.1 . Site no. Location Coordinates Alt itude (m) Site description Juniper spp. condition frequency (score) Other berry species Ring Ouzels 2.1 SW of Demnate 31°38’N 07°14’W 1,003 J p woodland clearance for arable Jp dominant 3.5 Pistacia sp. - 2.2 W of Azilal 31°54’N 06°42’W 1,247 J o woodland dense ground cover Jo dominant 1.5 Pistacia sp. - 2.3 SE of Azilal 3T51’N 06°25’W 1,865 Q ilex woodland adjacent arable J o moderate 4 None - 2.4 W of El Kebab 32°43’N 05°34’W 1,231 O ilex woodland adjacent arable J o scarce 4 None - 2.5 S of El Kebab 32°4TN 05°34’W 1,295 Q ilex woodland adjacent arable J o moderate 3 None Ring Ouzels wintering in Morocco: Ryall & Briggs Bull ABC Vol 13 No 1 (2006) -71 2.6 SE of El Kebab 32°4TN 05°30’W 1,665 Bare ground near valley None N/A None ."v 2.7 Tanout-Ou-Filali 32°40’N 05°29’W 1,859 Q ilex woodland bare ground J o scarce 3 None - 2.8 Sidi Tiar 32°40’N 05°27’W 2,070 Plantation (pine) adjacent arable J o scarce 4 Hawthorn Mistletoe, Rose - 2.9 W of Boumia 32°39’N 05°24’W 1,842 Arable land adjacent tussock herbage None N/A None ■ 2.10 N of Tounfite 32°40’N 05°17’W 1,637 Arable land Totally denuded None N/A None - 2.11 N of Tounfite 32°34’N 05°16’W 1,813 Arable land adjacent tussock herbage None N/A None - 2.12 N of Tounfite 32°32’N 05°16’W 1,870 Scrub adjacent tussock herbage None N/A None - 2.13 N of Tounfite 32°31’N 05°16’W 1,895 Q ilex woodland adjacent arable J o scarce 3 None - 2.14 N of Tounfite 32°30’N 05°15’W 1,874 Q ilex woodland bare stony ground J o common 3 Hawthorn, Mistletoe - 2.15 N of Tounfite 32°29’N 05°14’W 1,847 J p woodland near river, near farm Jp dominant J o moderate 1.5 2 Hawthorn, Mistletoe, Rose + 2.16 N of Tounfite 32°3TN 05°12’W 1,902 J p woodland ploughed below Jp dominant J o moderate 3.5 3.5 None - 2.17 N of Tounfite 32°3TN 05°12’W 1,855 J p woodland mixed vegetation Jp dominant J o moderate 2.4 3.7 Rose + 2.18 N of Tounfite 32°30’N 05°13’W 1,843 J p woodland near river Jp dominant J o scarce 2.5 3 None + 2.19(1.7) E of Tounfite 32°29’N 05°1 0’W 1,928 J p woodland bare stony ground, adjacent arable Jp dominant J o scrace J t rare 2 2 / None + 2.20 SE of Tounfite 32°27’N 05°09’W 1,850 Jp woodland near river Jp dominant J o scarce 3.5 4 Hawthorn Mistletoe - 2.21 S of Tounfite 32°26’N 05°09’W 1,896 Scrub river gorge None N/A Rose - 2.22 S of Tounfite 32°25’N 05°09’W 1,866 Scrub river gorge None N/A Rose - 2.23 W of Tounfite 32°38’N 05°17’W 2,030 Cedar forest adjacent arable Jp scarce / Hawthorn Mistletoe - 2.24 W of Tounfite 32°28’N 05°18’W 2,208 Cedar forest adjacent arable J p scarce / Hawthorn Mistletoe - 2.25 W of Tounfite 32°27’N 05°20’W 2,126 Cedar forest adjacent arable J p scarce J o rare / Hawthorn Mistletoe - 2.26 N of Boumia 32°34’N 05°11’W 1,810 Arable land adjacent tussock None N/A None - 2.27 S of Midelt 32°37’N 04°32’W 1,780 Jp woodland boulders, tussock J p dominant 4 None - 2.28 S of Midelt 32°36’N 04°3TW 1,840 J p woodland boulders, tussock J p dominant 2 None - 2.29 S of Midelt 32°36’N 04°3TW 1,863 J p woodland boulders, tussock Jp dominant 3 None ■ 2.30 S of Midelt 32°35’N 04°32’W 1,990 Plantation (pine) bare stony ground None N/A None - 72 -Bull ABC Vol 13 No 1(2006) Ring Ouzels wintering in Morocco: Ryall & Briggs 2.31 S of Midelt 32°34’N 04°29’W 1,722 Tussock grassland low herbage None N/A None - 2.32 S of Midelt 32°34’N 04°29’W 1,466 Tussock grassland low herbage None N/A None - 2.33 Er Rich 32°34’N 04°29’W 1,308 Plantation (olive) adjacent arable None N/A Rose - 3.1 Road to Telouet 31°15’N 07°2TW 1,918 J p woodland J p dominant 2.5 Rose + Wooded valley J o common 2.5 some Q ilex J t rare 0.25 Key: Jp = Juniperus phoenicea; Jo = J. oxycedrus; Jt=J. thurifera I = Data not collected Condition of vegetation: Undegraded = mean score for tree damage of 0-2.5; Degraded = mean score for tree damage of 2.6-5.0 (see Methods, Juniper tree parameters for further details) + = Ring Ouzels present; - = Ring Ouzels absent Appendix 3. Juniper density, condition and fruit crop at juniper-rich sites on Transect 2 and Site 3.1 , and occurrence of Ring Ouzels Turdus torquatus. Annexe 3. Densite et condition des genevriers et production de baies aux sites riches en genevriers le long du Transect 2 et au Site 3.1, et presence de Merles a plastron Turdus torquatus. Site no. Altitude (m) Tree Mean Mean canopy % of total Mean Mean berry Mean ROs / spp. height diameter trees condition count /m2 (SE) 100 “50 m present (m) (SE) (m) (SE) present rating (0-5) ripe unripe transect 2.2 1,247 Jo 3.1 2.25 95 1.5 298 0 0 (0.40) (0.47) (64) Non J 5 2.3 1,865 Jo 2.0 2.2 20 4 0 0 0 (0.20) (0.27) Qi 80 2.5 1,295 Jo 1.8 2.1 30 3 4.5 (0.5) 0 (0.16) (1.5) (4.8) 0.5 Qi 3.6 4.0 70 (0.45) (0.24) 2.7 1,859 Jo 5 3 4.5 0 0 (4.7) Qi 6.2 4.6 (0.54) (0.26) 95 2.8 1,945 Jo 10 4 0 0 0 Qi 4.9 5.5 (0.37) (0.35) 90 2.13 1,895 Jo 15 3 2.8 0 0 Qi 2.4 2.15 (0.21) (0.31) 85 (3.6) 2.14 1,874 Jo 2.0 1.9 32 3 9.6 0 0 (0.25) (0.42) (10.7) Qi 2.5 3.25 68 (0.25) (0.8) Ring Ouzels wintering in Morocco: Ryall & Briggs Bull ABC Vol 13 No 1 (2006) -73 2.15 1,847 2.16 1,902 2.17 1,855 2.18 1,843 2.19 1,928 2.29 1,863 3.1 1,918 Jp 2.7 3.75 95 1.5 436 712 14 (0.14) (0.34) (106) (180) (4.04) Jo 2.9 2.17 5 2 0 0 (0.17) (0.30) Jp 3.5 3.14 82 3.5 1033 158 0 (0.23) (0.22) (189) (63) Jo 2.78 2.08 24 3.5 0 0 (0.22) (0.24) Cyp 9 Jp 3.8 3.39 12 2.4 448 68 1 (0.40) (0.30) (137) (14) (0.70) Jo 3.4 2.4 9 3.7 239 0 (0.91) (0.86) (276) Jp 2.9 3.82 90 2.5 759 58 3.4 (0.35) (0.43) (168) (23) (0.45) Jo 2.5 2.64 4 3 0 0 (0.21) (0.31) Jp 5.6 5.6 40 2 1309 92 7.8 (0.49) (0.37) (224) (80) (3.45) Jo 2.9 2.9 15 2 0 0.8 (0.40) (0.78) (0.9) Jt 3.35 3.15 0.4 1 - - (0.10) (0.10) Jp 2.1 2.12 60 4 191 157 0 (2.1) (0.55) (191) (157) Jp 3.8 3.63 28 2.5 214 63 3.4 (0.53) (0.30) (77) (45) Jo 4.4 3.63 13 2.5 170 0 (0.37) (0.36) (162) Jt 3.1 3.1 0.8 0.25 319 480 (0.24) (0.33) (20) (76) Key: J p = Juniperus phoenicea\ J o = J. oxycedrus; Jt=J. thurifera; Qi = Quercus ilex;-, Cyp = Cyperus sp.; ROs = Ring Ouzels 74 - Bull ABC Vol 13 No 1 (2006) Ring Ouzels wintering in Morocco: Ryall & Briggs First record of Rose-coloured Starling Sturnus roseus for Ethiopia and sub-Saharan Africa Valery Schollaert Premiere mention de FEtourneau roselin Sturnus roseus pour l’Ethiopie et l’Afrique sub- saharienne. Un Etourneau roselin Sturnus roseus a ete observe en compagnie d’Etourneaux caron- cules Creatophora cinerea et photographic a environ 50 km a l’ouest de Yavello, le long de la route vers Arba Minch (05°06’N 37°53’E), Ethiopie, le 23 mars 2005. Ceci constitue la premiere don- nee pour l’Afrique sub-saharienne de cette espece palearctique, connue pour son nomadisme et ses mouvements erratiques. On 23 March 2005, whilst leading a birding trip in Ethiopia, I decided to stop for break- fast c. 50 km west of Yavello, on the road to Arba Minch (05°06’N 37°53’E). While the guides and drivers were preparing the food, we birded in the area. At the edge of a small field that made a clear- ing in the dense bush, we found some perching Magpie Starlings Speculipastor bicolor. On the ground was a group of Wattled Starlings Creatophora cinerea with an odd individual. Compared to the Wattled Starlings, the other bird was very dark, almost black, with a very pale breast and belly and a yellowish bill. I immediately iden- tified it as a Rose-coloured Starling Stunus roseus , a species with which I had previous experience in Bulgaria, a country where I lead groups regularly. Henry Brousmiche, a member of the group, quickly took some photographs. Thereafter we were able to get closer, observe all the plumage details and obtain better photographs (Fig. 1, p. 76). After some 15 minutes, the bird flew away to the west with some of the group of Wattled Starlings. Although broadly similar to the Wattled Starlings in size and shape, the bird appeared slightly smaller and slimmer, and the bill was slightly thinner. The head, neck and throat were glossy blackish, except the lores which were solid black. The blackish colour of the hindneck also reached the upper mantle, the rest of the mantle and the scapulars being paler, somewhat dirty whitish. The wings were relatively long and black- ish, with slightly paler coverts. The medium-long tail was also blackish. The breast, flanks and upper belly were dirty whitish, with a pinkish (or salmon) tinge. The lower belly and vent were dirty blackish and the undertail-coverts pale brown dot- ted black. The bill was yellowish orange, the legs pink. The bird foraged in a similar manner to the Wattled Starlings. This description matches an adult Rose-coloured Starling in non-breeding plumage (Cramp & Perrins 1994, Feare & Craig 1998). This appears to be the first record south of the Sahara of this Palearctic species (Fry et al. 2000), which is nomadic and often erratic in its move- ments. Its main breeding range extends from the Balkans to Central Asia, and it migrates to the Indian subcontinent for the winter. Vagrants have been recorded in Western Europe, North Africa and Seychelles (Fry et al. 2000). Acknowledgements I thank Henry Brousmiche for the photograph. Ron Demey and Nik Borrow commented on a draft of this note. References Cramp, S. & Perrins, C. M. (eds.) 1994. The Birds of the Western Palearctic. Vol. 8. Oxford: Oxford University Press. Feare, C. & Craig, A. 1998. Starlings and Mynas. London, UK: Christopher Helm. Fry, C. H., Keith, S. & Urban, E. K. (eds.) 2000. The Birds of Africa. Vol. 6. London, UK: Academic Press. Rue Lambiotte 121, 1030 Brussels, Belgium. E-mail: valeryschollaert@scarlet. be Received 30 August 2005; revision accepted 23 October First record of Rose-coloured Starling in Ethiopia: Schollaert Bull ABC Vol 13 No 1 (2006) -75 Figure 1 . Rose-coloured Starling Sturnus roseus , c.50 km west of Yavello, Ethiopia, 23 March 2005 (Henry Brousmiche) Etourneau roselin Sturnus roseus , environ 50 km a l’ouest de Yavello, Ethiopie, 23 mars 2005 (Henry Brousmiche) Figure 2. Unidentified egg in Royal Tern Sterna maxima colony, Bijol Islands, The Gambia, 25 April 2005 (John High) CEuf non identifie dans la colonie de Sternes royales Sterna maxima , lies Bijol, Gambie, 25 avril 2005 (John High) Figure 3. Birdled Tern Sterna anaethetus , Bijol Islands, The Gambia, 25 April 2005 (John High) Sterne bridee Sterna anaethetus , lies Bijol, Gambie, 25 avril 2005 (John High) Figure 4. Habitat du Grand-due du desert / habitat of Desert Eagle Owl Bubo {bubo) ascalaphus, Markoye, Burkina Faso, 23 mars 2005 (Yvan Perre) Figures 5-6. Grand-due du desert / Desert Eagle Owl Bubo {bubo) ascalaphus , Markoye, Burkina Faso, 21 novembre 2005 (Yvan Perre) 76 -Bull ABC Vol 13 No 1 (2006) First records of Rose-coloured Starling, Bridled Tern, Desert Eagle Owl First record of Bridled Tern Sterna anaethetus for The Gambia John High Premiere mention d’une Sterne bridee Sterna anaethetus pour la Gambie. En 2005 une Sterne bridee Sterna anaethetus adulte a sejourne sur les lies Bijol, dans la Reserve d’oiseaux de Tanji, des le 24 mars. Le 25 avril, l’oiseau etait observe dans la colonie de Sternes royales S. maxima. Un ceuf blanc-sale uni fut trouve, dont la taille (43.9 w 33.5 mm) correspondait a celle notee pour la Sterne bridee, mais la coloration et la forme ne correspondaient pas et l’ceuf n’a pas pu etre iden- tifie. La Sterne, dont pas plus d’un individu ne fut observe, etait presente au moins jusqu’au 24 aout. Ceci constitue la premiere donnee pour la Gambie. On 24 March 2005, whilst assisting the Gambian Department of Parks and Wildlife Management (DPWM) staff with their monthly census work on the Bijol Islands, which is part of Tanji (Karanti) Bird Reserve, Western Division, The Gambia (13°23’N 16°44’W), I noticed a dark-winged tern resting on the ground. It was smaller than nearby Grey-headed Gulls Larus cir- rocephalus but much larger than the Little Terns Sterna albifrons. I watched the bird in good light at c. 30 m with 8 x 42 binoculars. The dark grey colour of the upperparts was similar to that of Lesser Black-backed Gull Larus fuscus and paler than that of nearby Kelp Gulls L. dominicanus. The underparts were white. The jet black cap extended to the nape, contrasting with the white forehead, which extended as a supercilium slightly beyond the eye. The wings were long and extend- ed almost to the tip of the long tail. The bill was slender and black, the legs blackish. I did not flush the bird to see its flight features. I had not seen this species before and referred to A Field to the Birds of The Gambia and Senegal (Barlow et al. 1997) on my return home, but failed to find it illustrated. I then consulted Birds of Western Africa (Borrow & Demey 2001) and immediately identified the bird as an adult Bridled Tern Sterna anaethetus. On my next visit to the Bijol Islands, on 3 April, I did not see the bird, but during the following monthly census, on 25 April, it was present again (Fig. 3, p. 76). In the Royal Tern S. maxima colony we found a different, plain off-white egg measuring 43.9 x 33.5 mm (Fig. 2, p. 76), the only egg of this size and colour of a total of 10,500 eggs counted that day. Although its size is consistent with the known range for Bridled Tern (40-46 x 28.5-33.2 mm: Cramp 1985, Urban et al. 1986), its shape and coloration are not (Cramp 1985, D. Russell in litt.). Royal Tern eggs are noticeably larger (67-69 x 46 mm: Urban et al. 1986) and very heavily blotched or spotted. Although it can- not be wholly eliminated that the egg in Fig. 2 was an abnormal Bridled Tern egg, this seems a rather remote possibility (D. Russell in litt .); the egg, therefore, has to remain unidentified. The Bridled Tern was aloft close to the egg location throughout the period we were surveying the tern colony and often chased Grey-headed Gulls in the vicinity. Additional features observed in flight included the white underparts, off-white under- wing becoming dark grey on the flight-feathers, a deeply forked tail with at least the outer web of the very long outer tail-feathers white. No more than one Bridled Tern was seen at any one time over several visits. The bird was still in attendance during subsequent visits on 24 July and 24 August, but was not seen thereafter. This appears to be the first record of Bridled Tern for The Gambia: no records are mentioned in Barlow et al. (1997), nor have any been reported since (C. Barlow pers. comm.). However, its pres- ence was to be expected, as the species has been recorded in Senegal, where a few pairs breed in the lies de la Madeleine and Langue de Barbarie National Parks, and Guinea-Bissau (Morel & Morel 1990, Dowsett 1993, Barlow et al. 1997). Elsewhere in West Africa, it breeds on the Banc d’Arguin in Mauritania and on the Gulf of Guinea islands Sao Tome (at the islets of Sete Pedras) and Annobon. It disperses offshore after breeding and has been sparsely recorded along the coast, from First record of Bridled Tern for The Gambia: Ftigh Bull ABC Vol 13 No 1 (2006) -77 Guinea-Bissau to Equatorial Guinea and (proba- bly) Gabon (Borrow & Demey 2001). Acknowledgements DPWM rangers assisted in the monthly census counts, Douglas Russell, of the Natural History Museum, Tring, examined the photograph of the egg, and Clive Barlow and Ron Demey made useful comments on a draft and assisted in finalising this note. References Barlow, C., Wacher, T. & Disley, T. 1997. A Field Guide to Birds of The Gambia and Senegal. Robertsbridge: Pica Press. Borrow, N. & Demey, R. 2001. Birds of Western Africa. London, UK: Christopher Helm. Cramp, S. (ed.) 1985. The Birds of the Western Palearctic. Vol. 4. Oxford: Oxford University Press. Dowsett, R. J. 1993. Afrotropical avifaunas: annotated country lists. Guinea-Bissau. Tauraco Res. Rep. 5: 19-23. Morel, G. J. & Morel, M.-Y. 1990. Les Oiseaux de Senegambie. Paris: ORSTOM. Urban, E. K., Fry, C. H. & Keith, S. (eds.) 1986. The Birds of Africa. Vol. 2. London, UK: Academic Press. PO Box 2175, Serrekunda, The Gambia. E-mail: johnhigh33 @hotmail. com Received 7 April 2005; revision accepted 25 October 2005 Premiere mention du Grand-due du desert Bubo (bubo) ascalaphus pour le Burkina Faso Guilhem Lesctjfrea et Yvan Perreb First record of Desert Eagle Owl Bubo (bubo) ascalaphus for Burkina Faso. On 21 February 2005 we flushed two Desert Eagle Owls Bubo (bubo) ascalaphus from a small bushy tree in the Markoye area of northernmost Burkina Faso (14°43’N 00°00). The birds were seen again, and one of them photographed, on 23 March and 16 and 21 November 2005. This is the first record of this species for the country. The nearest known sites, both in Mali, are Hombori-Douentza and Gao, 200-300 km west-northwest and 180 km north of Markoye, respectively. Le 21 fevrier 2005, a la recherche de FAmmomane isabelline Amommanes deserti , nous prospectons des buttes d’eboulis granitiques qui se dressent dans la region de Markoye, province de l’Oudalan, a F extreme nord du Burkina Faso (14°43’N 00°00), en zone saheli- enne. Alors que nous nous dirigeons vers un petit arbre touffu isole — un Maerua crassifolia de 3,5 m environ — , situe dans la partie inferieure d’une butte, un grand rapace nocturne s’envole du coeur du feuillage et nous survole, montrant un dessous d’aile presque blanc avec une nette tache sombre au poignet. L’oiseau poursuit son vol et disparait derriere un relief proche. Nous approchons de l’arbre pour verifier la presence de pelotes de rejection lorsqu’un deux- ieme oiseau s’envole de la meme maniere que le premier. II se pose a une centaine de metres, dans les blocs rocheux en contrebas, se laissant observer durant plusieurs minutes. L’oiseau, de bonne taille, presente un aspect general tres pale. Le dessus (manteau, scapulaires et ailes) est beige soutenu crible de taches brun sombre et blanchatres. Le dessous est beige pale, bien marque de taches noiratres alignees en rayures au niveau de la poitrine. Les disques faciaux sont beige pale, bordes et soulignes d’une bande noiratre. Les aigrettes sont bien visibles. L’iris est orange. 78 - Bull ABC Vol 13 No 1 (2006) First record of Bridled Tern for The Gambia: High L’ ensemble de ces caracteres correspond a ceux presentes par le Grand-due du desert Bubo {bubo) ascalaphus (Borrow & Demey 2004). Le 23 mars, ainsi que le 16 et le 21 novembre, YP se rend sur le site, revoit les oiseaux et parvient a photographier (digiscopie) l’un d'eux (Fig. 3, p. 76). Cette donnee constitue la premiere mention de Bubo (bubo) ascalaphus pour le Burkina. Elle s’inscrit comme un jalon supplementaire de la Fin- ite sud de l’espece, que la carte figurant dans Borrow & Demey (2004) situe a hauteur de Test du lac Tchad, du sud du Niger et du centre du Mali, soit aux abords de 1 6°N. Les sites connus les plus proches, tous deux au Mali, sont Hombori- Douentza (Dowsett-Lemaire & Dowsett 2006) et Gao (Robertson 2001), respectivement a 200-300 km ouest-nord-ouest et 180 km au nord de Markoye. Douentza est a 15°00’N exactement. La butte concernee, haute de 20 m et longue de 250 m environ, fait partie d’une serie de six eboulis rocheux allonges et quasi contigus, s’eten- dant au total sur environ 2 km. Le plus petit, mesurant une centaine de metres de long, est a peine sureleve par rapport a la plaine environ- nante, tandis que le plus haut la domine de quelque 40 m. II existe d’autres zones d’ eboulis analogues dans la region, dont la butte de la mine de Tambao, situee a une dizaine de kilometres du site de l’observation, ou la grande ligne d’eboulis couvrant 2 ou 3 km le long de la piste entre Markoye et Gorom. Une prospection de l’ensem- ble de ces milieux favorables pourrait permettre de preciser le statut du Grand-due du desert dans le nord du Burkina. Remerciements Nous remercions Jean-Marc Thiollay pour nous avoir encourages a publier cette note, Ron Demey pour l’accueil qu’il nous a reserve et Fran<;oise Dowsett-Lemaire pour les remarques constructives qui ont permis de situer la mention dans son con- texte regional. Bibliographie Borrow, N. & Demey, R. 2004. Field Guide to the Birds of Western Africa. London, UK: Christopher Helm. Dowsett-Lemaire, F. & Dowsett, R. J. 2006. First reli- able sound recording of Golden Nightjar Caprimulgus eximius, in the rocky hills of central Mali. Bull. ABC 13: 49-55. Robertson, P. 2001. Mali. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas of Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, UK: BirdLife International. a Centre ornithologique d’lle-de-France, 9 rue Pergolese, 75116 Paris, France. E-mail: guilhem. lesaffe@wanadoo.fr b Institut de Recherche pour le Developpement, 956 rue Agostino Neto, Ouagadougou 01, Burkina Faso. E- mail: yvan.perre@ird. bf Re$u le 18 juin 2005; revision acceptee le 2 novembre 2005 Premiere mention du Grand-due du desert pour le Burkina: Lesaffe & Perre Bull ABC Vol 13 No 1 (2006) -79 First records of Blackcap Sylvia atricapilla for Mozambique Martim Meloa, Rita Covasa and Klaas-Douwe Dijkstra ^ Premieres mentions de la Fauvette a tete noire Sylvia atricapilla pour le Mozambique. Le 6 decembre 2001 un male de la Fauvette a tete noire Sylvia atricapilla a ete capture au filet japon- ais sur le Plateau de Muretha, qui fait partie du Massif de Namuli, au nord du Mozambique. Plusieurs cris et un chant ont ete entendus, permettant d’estimer qu’au moins cinq individus etaient presents. Ces donnees sont les premieres pour le Mozambique. Les auteurs suggerent que le Plateau de Muretha pourrait constituer un aire d’hivernage pour cette espece. On 1-6 December 2001, we visited the Namuli massif in northern Mozambique (15°12’S 36°52’E), spending most of our time in the Ukalini forest and on the Muretha Plateau. The objective of our visit was to gather informa- tion on the birds, through observation and mist- netting (a report on the visit can be obtained from MM). Palearctic migrants we observed included Garden Warbler Sylvia borin and Willow Warbler Phylloscopus trochilus (common), and Tree Pipit Anthus trivialis (one). On 3-6 December we visited the Muretha Plateau. The habitat is high-altitude grassland with scattered small patches of montane forest. Upon arrival we heard several times a metallic tacc , repeated regularly, apparently the contact call of a Blackcap Sylvia atricapilla , but we could not con- firm this as we did not see the callers and were unfamiliar with the vocalisations of the birds of the area. After we had set up a 6-m mist-net in one Figure 1 . Blackcap Sylvia atricapilla trapped on the Muretha Plateau, Mozambique, December 2001 (Klaas- Douwe Dijkstra) Fauvette a tete noire Sylvia atricapilla capture sur le Plateau de Muretha, Mozambique, decembre 2001 (Klaas-Douwe Dijkstra) of the forest patches where Garden and Willow Warblers were common, we immediately caught a male Blackcap (wing: 73.5 mm, tarsus: 23.1 mm, mass: 18.4 g; Fig. 1). Subsequently we continued hearing the contact calls and KDD once heard a song. At least five individuals were present. Two Garden Warblers were also captured. To our knowledge these are the first records of Blackcap for Mozambique. Status and distribution Blackcap is one of the best-studied Palearctic migrants. Its breeding population spans the Western Palearctic, reaching south-west Siberia in the east, western Norway in the north, and north- west Africa and Iran in the south (Urban et al. 1997, Shirihai et al. 2001). The species has differ- ent migratory strategies corresponding to breeding area: northern breeding populations are wholly migratory, south-western populations are partial Figure 2. The Muretha Plateau, looking north-east, with the Namuli Massif in the background, 6 December 2001 (M. Melo) Le Plateau de Muretha, vu vers le nord-est, avec le Massif de Namuli en arriere-plan, 6 decembre 2001 (M. Melo) 80 - Bull ABC Vol 13 No 1 (2006) First records of Blackcap for Mozambique: Melo et al. migrants, and Atlantic and Mediterranean popula- tions are largely resident. Most migrants winter in Africa, some in south-western Europe and a win- tering population has recently become established in Great Britain. In Africa, three wintering areas can be defined: north of the Sahara, in the savan- nas of West Africa and in the highlands of north- east and East Africa. The southernmost wintering population is found in the mountains of Malawi (c. 14°30’S; Urban et al. 1997). There are several recent records from the eastern highlands of Zimbabwe, very close to the Mozambican border (Cohen 1997). Vagrants have been reported in South Africa since 1985 (Sinclair et al. 1987), sug- gesting that the species might be a more regular visitor than currently thought. It may have been overlooked in the past due to its inconspicuous behaviour or the records could indicate an expan- sion of its wintering range (Cohen 1997). The Namuli Massif The Namuli Massif contains one of the most over- looked remnants of Eastern Arc montane forests. The ornithological importance of the area was demonstrated by an expedition in 1998, nearly 70 years after it was last visited by a naturalist (Ryan et al. 1999a, b). Namuli Apalis Apalis \thomcica ] lynesi, Mozambique’s only endemic bird, is restricted to this area, and two other restricted- range species, Cholo Alethe Alethe choloensis and Spot-throat Modulatrix orosthrutus , occur. All three species are of global conservation concern (Parker 2001). Our Blackcap records suggest that the Muretha Plateau (and probably other areas of the Namuli Massif as well) might constitute a win- tering area for the species, as it offers a habitat typ- ical of the winter quarters in East Africa and there is a wintering area further south, in Malawi. Acknowledgements Ron Demey encouraged us to submit this note. Peter Ryan, Claire Spottiswoode and Callan Cohen helped in the organisation of the trip. RD, CS and CC pro- vided helpful comments on earlier drafts. References Cohen, C. 1997. European Blackcap Sylvia atricapilla. In Harrison, J. A., Allan, D. G., Underhill, L. G., Herremans, M., Tree, A. J., Parker, V. & Brown, C. J. (eds.) The Atlas of Southern African Birds. Vol. 2. Johannesburg: BirdLife South Africa. Parker, V. 2001. Mozambique. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, UK: BirdLife International. Ryan, P. G., Bento, C., Cohen, C., Graham, J., Parker, V. & Spottiswoode, C. 1999a. The avifauna and conservation status of the Namuli Massif, northern Mozambique. Bird Conserv. Intern. 9: 315-331. Ryan, R, Spottiswoode, C., Parker, V., Graham, J., Cohen, C. & Bento, C. 1999b. The birds of Namuli, northern Mozambique: retracing Vincent’s footsteps. Bull. ABC 6: 138-143. Shirihai, H., Gargallo, G. & Helbig, A. J. 2001. Sylvia Warblers: Identification, Taxonomy and Phylogeny of the genus Sylvia. London, UK: Christopher Helm. Sinclair, J. A. & the Rarities Committee 1987. S.A.O.S. Rarities Committee’s annual report. Bokmakierie 39: 12-14. Urban, E. K., Fry, C. H. & Keith, S. (eds.) 1997. Birds of Africa. Vol. 5. London, UK: Academic Press. a Institute of Evolutionary Biology, University of Edinburgh, Kings Buildings, EH9 3JT Edinburgh, UK. E-mail: martim.melo@ed.ac.uk, rita.covas@ed.ac.uk b Gortestraat 11, 2311 MS Leiden, Netherlands. E- mail: dijkstra@nnm.nl Received 26 May 2005; revision accepted 5 November 2005 First records of Blackcap for Mozambique: Melo et al. Bull ABC Vol 13 No 1 (2006) -81 First records of Mottled Swift Tachymarptis aequatorialis and Alpine Swift T. melba for Niger Kim Diget Christensena , Anders P. Tottrup ^ and Flemming Pagh Jensenc Premieres mentions pour le Niger du Martinet marbre Tachymarptis aequatorialis et du Martinet alpin T. melba. Les auteurs documentent la decouverte de deux especes nouvelles pour le Niger. Un Martinet marbre Tachymarptis aequatorialis a ete observe dans un groupe mixte de martinets pres du Grand Hotel a Niamey, le 27 aout 2004. Six Martinets alpins T. melba ont ete vus en compagnie de 14 Martinets noirs Apus apus a environ 65 km a l’ouest de Zinder (13°69’N 09°57’E), le 31 aout 2004. In 2004 we made bird observations in Niger as part of the Project Regional de Lutte Integree contre les Sauteriaux au Sahel (PReLISS), which was initiated in 2002 and funded by the Danish International Development Agency (DANIDA). During this project intensive ornithological research was conducted and in 2003 four species that were new to Niger were recorded (Christensen et al. 2005). We report two other additions to the country’s avifauna, recorded dur- ing the survey in 2004. Mottled Swift On 27 August 2004, at 08.30 hrs, a large dark swift with a paler belly was discovered in a mixed flock of Common Swifts Apus apus , Little Swifts A. ajfinis and African Palm Swifts Cypsiurus parvus near the Grand Hotel, Niamey (zone 2 in Giraudoux et al. 1988). It was observed for c.10 minutes until the flock disappeared to the other side of the river. A few minutes later the large swift returned with the flock and flew above us for another five minutes. It was clearly larger than the other species in the flock and its wingbeats were noticeably slower. We estimated that its wingspan was c.30% larger than that of the Common Swifts. The plumage was dark brown overall, but the belly was mottled paler brown to grey and the throat was grey. The tail was forked. These are diagnostic features of Mottled Swift Tachymarptis aequatori- alis (Borrow & Demey 2001). The only other swift of that size is Alpine Swift 77 melba , which has a white belly and throat, and is well known to us. The distribution of Mottled Swift in West Africa is patchy and inadequately known. It has been recorded from Guinea to eastern Chad and north-east Central African Republic (Fry et al. 1988, Borrow & Demey 2001). The nearest records are c.400 km to the north-west in Mali and c.500 km to the west in Burkina Faso (Balan^a & de Visscher 1993, Dowsett & Dowsett-Lemaire 2005), only a short distance for a large swift. This is the first documented record of Mottled Swift for Niger: the species is not included in Dowsett’s (1993) checklist for the country and is not mapped for Niger in Borrow & Demey (2001). Alpine Swift On 31 August 2004, at 07.30 hrs, a flock of swifts, containing 14 Common Swifts and six Alpine Swifts, was found c. 65 km west of Zinder in west- ern Niger (13°69’N 09°57’E, zone 3 in Giraudoux et al. 1988). The birds were foraging fairly low at 15-20 m above the ground in excellent light, making viewing conditions ideal. They were observed for ten minutes before they disappeared to the south. The Alpine Swifts were c.20% larger than the Common Swifts and had distinctively slower wingbeats. The upperparts were uniformly brown and contrasted with the pure white belly and throat, which were separated by a dark brown- ish breast-band. The dark brownish tail was shal- lowly forked. The underwing-coverts appeared brownish with slightly paler flight-feathers. Mottled Swift, the only other large swift in Africa, has very differently coloured and patterned underparts. Alpine Swift is a Palearctic passage migrant and winter visitor to West Africa, from Mauritania to Liberia east to Cameroon, with large flocks being frequently recorded in Ghana, Togo and 82 -Bull ABC Vol 13 No 1(2006) First records of Mottled Swift and Alpine Swift for Niger. Christensen et al. Nigeria (Fry et al. 1988, Borrow & Demey 2001). Although it has been reported to breed on the Bandiagara escarpment, central Mali, where com- mon (Thiollay 1974), this has recently been ques- tioned by Dowsett & Dowsett-Lemaire (2005). The observed birds are most likely Palearctic visi- tors from north-west Africa or Europe. Migration through the southern Palearctic occurs mainly in September-October, though there is some south- ward movement in August by juveniles at least (Cramp 1985). In Mali, birds are said to arrive in September-October (Lamarche 1980). This appears to be the first documented record of Alpine Swift for Niger. Although Fry et al. (1988) mention records of 2-200 Alpine Swifts in Niger and Nigeria, we have not been able to track the origin of these records and the species is not included in Dowsett’s (1993) checklist for the country, nor is it mapped or mentioned for Niger in Borrow & Demey (2001). Acknowledgements The Danish International Development Agency (DANIDA) is thanked for financing the PReLISS project. Ron Demey and Franchise Dowsett-Lemaire are thanked for useful comments on a draft of this note. References Balan^a, G. & de Visscher, M. N. 1993. Nouvelles don- nees de distribution pour deux especes d’oiseaux au Burkina Faso. Malimbus 15: 89-90. Borrow, N. & Demey, R. 2001. Birds of Western Africa. London, UK: Christopher tielm. Christensen, K. D., Tottrup, A. P., Rahner, M. C. & Brouwer, J. 2005. First records for Niger of Red- chested Cuckoo Cuculus solitarius, Grassland Pipit Anthus cinnamomeus , Buff-bellied Warbler Phyllolais pulchella and Isabelline Shrike Lanius isabellinus. Bull. ABC 12: 162-164. Cramp, S. (ed.) 1985. The Birds of the Western Palearctic. Vol. 4. Oxford: Oxford University Press. Dowsett, R. J. 1993. Afrotropical avifaunas: annotated country checklists. Niger. Tauraco Res. Rep. 5: 97-102. Dowsett, R. J. & Dowsett-Lemaire, F. (2005) On the apparent status of Mottled Swift Apus ( Tachymarptis ) aequatorialis and Alpine Swift A. ( T) melba in Mali, West Africa. Bull. Br. Ornithol. Cl. 125: 296-298. Fry, C. H., Keith, S., & Urban, E. K.(eds.) 1988. The Birds of Africa. Vol. 3. London, UK: Academic Press. Giraudoux, P., Degauquier. R., Jones, P. J., Weigel, J. & Isenmann, P. 1988. Avifaune du Niger: etat des connaissances en 1986. Malimbus 10: 1-140. Lamarche, B. 1980. Liste commentee des oiseaux du Mali, lere partie. Malimbus 2: 121-158. Thiollay, J.-M. 1974. Nidification du Martinet pale Apus pallidus et du Martinet alpin Apus melba en Afrique occidentale. Alauda 42: 223-225. aTraegaarden 2. 2. tv., DK-2300 Copenhagen S, Denmark. E-mail: kim@diget.dk ^ Zoological Museum, University of Copenhagen, Universitetsparken 15, DK-2100 Copenhagen 0, Denmark. E-mail: aptottrup@zmuc.ku.dk CDDH Consulting A/S, Ringstedvej 20, DK-4000 Roskilde, Denmark, E-mail: fpj@Hedeselskabet.dk Received 14 February 2005; revision accepted 5 July 2005 First records of Mottled Swifi and Alpine Swift for Niger. Christensen et al. Bui! ABC Vol 13 No 1 (2006) -83 Premiere observation du Picatharte du Cameroun Picathartes oreas au Congo-Brazzaville Victor Mamonekenea et Frederic Lambert Bokandza-Paco ^ First record of Grey-necked Picathartes Picathartes oreas for Congo-Brazzaville. We report the sighting of Grey-necked Picathartes Picathartes oreas in Mayombe forest, Congo-Brazzaville, on 22 December 1994. This record constitutes the first for the country and is the southernmost locality for the species, formerly known only from eastern Nigeria to Gabon. Le 22 decembre 1994, au cours d’une mission dans la foret du Mayombe, au sud-ouest du Congo-Brazzaville, nous avons observe un Picatharte du Cameroun Picathartes oreas le long de la Route Nationale n°l, a environ 20 km a l’ouest de Les Saras (04°22’S 12°22’E). L’oiseau etait au sol dans un bas-fond pres d’un ruisseau et sautillait entre les pieds d 'Aframomum giganteum (Zingiberacees) sans s’envoler. II presentait indiscutablement la taille, la forme et les couleurs caracteristiques du Picatharte du Cameroun et nous avons pu noter le rouge du dessus de la tete et de la nuque, la longue queue foncee et les longues pattes gris-fonce. Le biotope etait constitue de foret vierge avec un sous-bois oil dominait Aframomun giganteum et, au bord de la route, des especes de foret secondaire ayant pousse a la faveur du nettoyage de la voie. Cette partie du Mayombe est peu perturbee en comparaison avec des zones adjacentes; elle fait partie de la Reserve de Biosphere de Dimonika (non fonctionnelle). L’observation a ete faite au debut de la saison des pluies dans cette partie du Congo. Ceci constitue la premiere observation au Congo-Brazzaville du Picatharte du Cameroun, qui jusqu’alors n’etait connu que du Nigeria, du Cameroun, de la Guinee Equatoriale et du Gabon (Dowsett-Lemaire & Dowsett 1991, Dowsett 1993, Fry 2000, Borrow & Demey 2001). La localite la plus proche est Mouila, au Gabon, environ 300 km au nord-ouest (Collar & Stuart 1983). L’espece est consideree comme menacee et sa population est estimee a seulement 10.000 individus (BirdLife International 2000). Bien que les travaux de Dowsett-Lemaire &c Dowsett (1989a, b, 1991) aient considerablement augmente les connaissances sur l’avifaune du Mayombe, la rencontre fortuite de cette espece jamais signalee auparavant sur ce site indique que la foret cache encore une diversite qui meriterait des recherches plus poussees. Remerciements Nous remercions Michel Louette du Musee Royal de l’Afrique Centrale, Tervuren, Belgique, pour nous avoir encourages a partager cette information et ouvert sa bibliotheque. Framboise Dowsett-Lemaire et Ron Demey sont remercies pour leurs commen- taires sur le manuscrit. Bibliographie BirdLife International. 2000. Threatened Birds of the World. Barcelona: Lynx Edicions & Cambridge, UK: BirdLife International. Borrow, N. & Demey, R. 2001. Birds of Western Africa. London, UK: Christopher Helm. Collar, N. J. & Stuart, S. N. 1983. Threatened Birds of Africa and Related Islands. Cambridge, UK: International Council for Bird Conservation & Gland: International Union for Conservation of Nature and Natural Resources. Dowsett, R. J. 1993. Afrotropical avifaunas: annotated country checklists. Congo. Tauraco Res. Rep. 5: 189-194. Dowsett-Lemaire, F. & Dowsett, R. J. 1989a. Enquete faunistique dans la foret du Mayombe (Dimonika): itineraire, resultats et recommandations. Tauraco Res. Rep. 2: 1—4. Dowsett-Lemaire, F. & Dowsett, R. J. 1989b. Liste commentee des oiseaux de la foret du Mayombe (Congo). Tauraco Res. Rep. 2: 5-16. Dowsett-Lemaire, F. & Dowsett, R. J. 1991. The avi- fauna of the Kouilou Basin in Congo. Tauraco Res. Rep. 4: 189-239. Fry, C. H. 2000. Grey-necked Picathartes Picathartes oreas. In Fry, C. H., Keith, S. & Urban, E. K. (eds.) The Birds of Africa. Vol. 6. London, UK: Academic Press. 84 -Bull ABC Vol 13 No 1 (2006) Picatharte du Cameroun au Congo-Brazzaville: Mamonekene & Bokandza-Paco a Institut de Developpement Rural, Universite Marien Ngouabi, Brazzaville, Congo. Email: mamonekene@hotmail. com b Direction de la Faune et des Aires Protegees, Ministere des Eaux et Forets et de I’Environnement, Brazzaville, Congo. Re$u le 3 fevrier 2003; revision acceptee le 25 mai 2005 Note de la redaction. Bien que R. C. Fotso (1993. Contribution a l’etude de la biologie du Picatharte chauve du Cameroun Picathartes oreas. Proc. VIII Pan-Afr. Orn. Congr. 431-437) affirme que le domaine de l’espece s’etend jusqu’au nord-est du Congo, a la frontiere avec le Cameroun, aucune observation nest connue de cette zone jusqu’a present. L’observation dans le Mayombe, presentee ci-dessus, constitue done bien la premiere pour le Congo-Brazzaville. Tropical Birding J Africa and Asia Madagascar, Namibia & Botswana, South Africa, Zambia, Cameroon, Ethiopia, Tanzania, Gabon, Sao Tome & Principe, Uganda, India, Taiwan and Borneo. Americas 8c Australia Guapi Assu bird lodge -Tropical | SEHT ■- JM Birding's new lodge in the midst I Tv fll of the Atlantic rainforests of SE plP; j ; '*7- • 0 Brazil, only 70km from Rio. 1 » • .**.»>. / SE Atlantic endemics, Pantanal, \ * V Soa Nova, Ecuador, NW Ecuador, 1 Andes East Slope, Southern r A Ecuador, Cuba, Venezuela, Peru & h~ — ^ i Australia. Set-departure and customised trips for all destinations For more information explore our catalogue on-line Call us toll free at 1-800-348-5941 info@tropicalbirding.com www.tropicalbirding.com Fax : +593-2-290-9753Tel: +593-9-923-1314 Picatharte du Cameroun au Congo-Brazzaville: Mamonekene & Bokandza-Paco Bull ABC Vol 13 No 1 (2006) - 85 First Pink-backed Pelican Pelecanus rufescens sightings in Madagascar since 1960 Martin Mwemaa and Felix Razafindrajao^ Premieres observations du Pelican gris Pelicanus rufescens a Madagascar depuis 1960. Trois observations du Pelican gris Pelecanus rufescens au Lac Bedo, dans le sud-ouest de Madagascar, sont rapportees: deux individus ont ete vus le 21 septembre 2003 et un individu le 18 novembre et le 12 decembre 2004. L’espece n’avait plus ete vue depuis I960. On the morning of 18 November 2004, a pel- ican was sighted at Lake Bedo (19°33’S 44°32’E), south-west Madagascar, by a group of eight Tropical Biology Association course partici- pants (for a description of the site, see Young & Razafindrajao 2006). The bird flew out of the veg- etation on a small island near the middle of the lake, c.500 m from the shore. It circled around the island for cPJ minutes before disappearing again into the vegetation, from which it did not emerge. At one point the pelican came as close as c. 1 00 m from the shore. It thus gave the observers ample opportunity to note its features and consult an identification guide (Langrand 1990) while it was still in view. The bird was largely whitish with a pale grey cast. The bill and pouch appeared pale grey. It had a black patch in front of the eye resembling a teardrop from a distance. In flight, there was no conspicuous contrast between the dark flight- feathers and the pale grey wing-coverts. MM, who first noticed the bird, identified it as a Pink-backed Pelican Pelecanus rufescens , based on the above features. Great White Pelican P. onocro- talus is larger and whiter and has darker, blackish, flight-feathers. He pointed it out to Dr Julia Jones, the group leader, who agreed with his identifica- tion. Other group members also positively identi- fied the pelican from the guidebook. MM returned to the site with another group of students in the afternoon, but the pelican was not seen again. FR, having been informed by Dr Jones of the observation, visited the site on 12 December 2004 and found the pelican still present. FR had seen two pelicans at Lake Bedo previously, on 21 September 2003, but had been unable to positive- ly identify them to species. Having subsequently observed the two pelican species in Kenya, he is now confident that these birds were also P. rufescens , based on the above features. These are the first pelican sightings in Madagascar since 1960 (Langrand 1990, Morris & Hawkins 1998). Pink-backed Pelican was first mentioned for Madagascar by Verreaux (1865) and the species was subsequently known as an accidental visitor (Delacour 1932). Paulian (1959), citing P. Griveaud, reports the discovery of a small breeding colony south-west of Antsalova, c.150 km north of Lake Bedo, in 1958. A group was seen by Y. Therezien and R. Legendre in August 1959 north of Belo-sur-Tsiribihina, c.75 km north of Lake Bedo (Milon et al. 1973). The colony in the Antsalova Lake region was still present in I960 (Paulian 1961), but as there were no subsequent sightings it is thought that the colony was exterminated by local villagers (Langrand 1990, Morris & Hawkins 1998). Although the possibility of a colony existing some- where in the country cannot be entirely eliminat- ed, most observers consider the species to be a vagrant. Pink-backed Pelican is the only pelican observed in Madagascar to date. Great White Pelican, a common migratory species in East Africa that could reach Madagascar as a vagrant, has never been recorded from the Malagasy region (Elliott 1992). Acknowledgements We thank Frank Hawkins at Conservation International and Alain Crivelli at Wetlands International for providing some hard-to-find refer- ences. Dr Julia Jones of the Tropical Biology Association assisted in many ways. Glyn Young and Roger Safford commented on the draft. Pink-backed Pelican sightings in Madagascar: Mwema & Razafindrajao 86 -Bull ABC Vol 13 No 1(2006) References Delacour, J. 1932. Les oiseaux de la Mission Franco- Anglo-Americaine a Madagascar. Oiseau & R.F.O. 2: 1-96. Elliott, A. 1992. Family Pelecanidae (pelicans). In del Hoyo, J., Elliott, A. &t Sargatal, J. (eds.) Handbook of Birds of the World. Vol. 1. Barcelona: Lynx Edicions. Langrand, O. 1990. Guide to the Birds of Madagascar. New Plaven & London: Yale University Press. Milon, P., Petter, J.-.J. & Randrianasolo, G. 1973. Faune de Madagascar XXXV, Oiseaux. Tananarive & Paris: ORSTOM & CNRS. Morris, P. & Hawkins, F. 1998. Birds of Madagascar: A Photographic Guide. Robertsbridge: Pica Press. Paulian, R. 1959. Notules ornithologiques. Le Naturalist e Malgache X, fascicule 1-2: 173-174. Paulian, R. 1961. La Zoogeographie de Madagascar et des ties voisines. Faune de Madagascar XIII. Tananarive-Tsimbazaza: Institut de Recherche Scientifique. Verreaux, J. 1865. Catalogue des oiseaux de Madagascar connus jusqu’a ce jour. In Vinson, A. Voyage de Madagascar au couronnement de Radama II Annexe B: 1-6. Paris: Librarie Encyclopedique de Roret. Young, G. & Razafindrajao, F. 2006. Lake Bedo — a little-known wetland hotspot in Madagascar. Bull. ABC 13:91-95. a Ornithology Department, National Museums of Kenya PO Box 40658,00100, GPO, Nairobi, Kenya. E-mail: kbirds@africaonline. co. ke ^ Durr ell Wildlife Conservation Trust, BP8511, Antananarivo 101, Madagascar. E-mail: dw. madagascar@durrell. org Received 18 August 2005; revision accepted 18 October 2005 www.birdquest.co.uk Birdquest - the natural choice in Africa with unrivalled experience, small group sizes and remarkable leaders! Morocco • Egypt • Gambia • Senegal • Ivory Coast • Cameroon • Gabon Sao Tome & Principe • Ethiopia • Kenya • Uganda • Tanzania • Rwanda • Malawi Zambia • Angola • Namibia • Botswana • Zimbabwe • South Africa • Madagascar All included in our extensive programme of more than 150 tours worldwide! To learn more about / The Ultimate in Birding', please visit our v website. Alternatively call us on +44 (0)1254 826317 or E-mail: birders@birdquest.co.uk ~ an0^eque^^\^^hure Pink-backed Pelican sightings in Madagascar: Mwema & Razafindrajao Bull ABC Vol 13 No 1 (2006) -87 First record of Kermadec Petrel Pterodroma neglecta for Seychelles Cas Eikenaara and Adrian SkerretP Premiere mention du Petrel des Kermadec Pterodroma neglecta pour les Seychelles. Un Petrel des Kermadec Pterodroma neglecta etait present a Cousin le 29 aout 2003 et (probablement le meme individu) le 29 juin 2004. Cette mention a ete acceptee par le Comite d’ Homologation Seychellois comme la premiere pour le pays. n 29 August 2003 Cas Eikenaar (CE) noted a medium-sized seabird on the hill of Cousin Island Nature Reserve within an open patch near dense vegetation. It was very obvious- ly different to the Audubon’s Shearwaters Puffinus Iherminieri and Wedge-tailed Shearwaters P. pacificus , common in the vicinity. Having taken two photographs (Fig. 1), CE approached the bird in order to examine it in the hand and take more photographs. The bird moved into the vegetation where it was caught next to a Wedge-tailed Shearwater, with which it showed no aggressive interactions. In the hand, the bird proved to be very strong and could only be controlled by holding the tail, wings and legs (Fig. 2). As a result, with no one to assist, the underwing was not properly examined and not photographed. Following release, the bird settled next to the Wedge-tailed Shearwater, again show- ing no aggressive interactions. The bird could not be relocated the next day. The photographs and description were submitted unidentified to the Seychelles Bird Records Committee (SBRC). CE departed Cousin Island on 12 September 2003, returning on 20 May 2004. On 29 June 2004, what appeared to be a similar or the same bird was located at exactly the same spot as the previous year. The bird was caught and examined (Figs. 3-4). With the assistance of Lyanne Brouwer, it was possible to examine the under- wing and to take measurements and a blood sam- ple. Again, details were sent to SBRC. Description of first bird Pale grey-brown to white underparts contrasting with darker upperparts and wings. Back dark; upper back to neck gradually becoming greyish brown. Paler head also contrasted with darker upperparts. Whitish patch on lores extending to above and below striking dark eye. Sturdy neck and short, thick dark bill. Area below gape and chin darker, greyish brown. Crown, hindneck and face-sides darker greyish brown. Forecrown slight- ly darker (especially feather centres). Dark tail, tip shorter or equal to tip of wings. Did not appear weak (as far as that can be judged without previ- ous experience of the species) and resisted with great strength upon capture. Description of second bird Appeared identical to first bird. In addition, examination of underwing revealed this was mainly dark with white patch at base of primaries and white primary shafts. White primary shafts noted on upperwing. Measurements: weight 410 g, tarsus 41.8 mm, tarsus-toe 103 mm, folded wing 30.7 cm, stretched wing 46.7 cm, undertail 120 mm, uppertail 105 mm, head + bill 80.6 mm, bill from tip of nail in straight line to furthest point of gape 33.3 mm, bill depth measured just before nasal tube 30.1 mm. Analysis by SBRC Following circulation around the committee, examination of museum specimens at The Natural History Museum (Tring, UK) and consultation with contacts having experience of Pterodroma breeding at Mauritius and on islands in the Pacific, the SBRC concluded that the notes and photographs submitted for the first observation were insufficient to conclusively distinguish between Herald Petrel Pterodroma arminjoniana and Kermadec Petrel P. neglecta. The second obser- vation, however, was accepted as Kermadec Petrel. Given the absence of previous records from Seychelles, the extreme similarity in overall appearance of the birds photographed in August 2003 and June 2004 and the fact that the observa- 88 -Bull ABC Vol 13 No 1(2006) First record of Kermadec Petrel for Seychelles: Eikenaar & Skerrett Figures 1-2. Kermadec Petrel / Petrel des Kermadec Pterodroma neglecta , Cousin, Seychelles, 29 August 2003 (Cas Eikenaar) tions were made at the same locality, it appears very probable they refer to the same individual. In summary, the SBRC accepted the record on the basis of the following: • White primary shafts, diagnostic of Kermadec (Herald has black shafts); • Underwing mainly dark; white confined to basal two-thirds of primaries’ inner webs (underwing whiter in Herald, including a pale strip inwards, slightly rear of centre); • Rounded tail, characteristic of Kermadec (Herald has wedge-shaped tail); • Mass within range for Kermadec (exceptional for Herald); • Tarsus near average for Kermadec (too great for Herald); Figure 3. Kermadec Petrel Pterodroma neglecta , Cousin, Seychelles, 29 June 2004 (Cas Eikenaar). Note white patch on basal two-thirds of inner primary webs. Petrel des Kermadec Pterodroma neglecta , Cousin, Seychelles, 29 juin 2004 (Cas Eikenaar). Notez la tache blanche sur les deux-tiers proximaux des vexilles internes des remiges primaires. Figure 4. Kermadec Petrel Pterodroma neglecta , Cousin, Seychelles, 29 June 2004 (Cas Eikenaar). Note white primary shafts in upperwing. Petrel des Kermadec Pterodroma neglecta , Cousin, Seychelles, 29 juin 2004 (Cas Eikenaar). Notez les rachis blancs des remiges primaires vues du dessus. • Folded wing within range for Kermadec (too great for Herald); • Tail near average for Kermadec (too short for Herald; Kermadec is a short-tailed species despite being generally larger); • Tarsus/proximal feet grey with bluish tinge, normal for intermediate-morph Kermadec (unknown in Herald). First record of Kermadec Petrel for Seychelles : Eikenaar & Skerrett Bull ABC Vol 13 No 1 (2006) -89 Status and distribution This is the first accepted record for Seychelles. Kermadec Petrel was once believed to breed exclu- sively in the South Pacific from Lord Howe Island east to Easter Island and some islands west of Chile (Carboneras 1992). More recently it was discovered breeding alongside Herald Petrel at Round Island, Mauritius (Brooke et al. 2000, Brooke 2004) and at Ilha da Trindade, Brazil (Imber 2004). Acknowledgements Thor Veen is thanked for help and advice to CS with the submission made to the SBRC and the prepara- tion of this note. The SBRC are grateful for the assis- tance of Vincent Bretagnolle, Alan Burger, Carl Jones, Mike Imber, Tony Palliser, Peter Ryan, Ian Sinclair and Vikash Tatayah, all of whom acted as consultants to the committee to assist in the confir- mation of this record. The members of the SBRC who assessed this record included Michael Betts, Ian Bullock, David Fisher, Ron Gerlach, John Phillips, Rob Lucking, Bob Scott and Adrian Skerrett. Mike Imber commented on a draft of this note. References Brooke, M. 2004. Albatrosses and Petrels across the World. Oxford: Oxford University Press. Brooke, M. de L., Imber, M. J. & Rowe, G. 2000. Occurrence of two surface-breeding species of Pterodroma on Round Island, Indian Ocean. Ibis 142: 154-158. Carboneras, C. 1992. Family Procellariidae (petrels and shearwaters). In del Hoyo, J., Elliott, A. & Sargatal, J. (eds.) Handbook of the Birds of the World. Vol. 1. Barcelona: Lynx Edicions. Imber, M. J. 2004. Kermadec petrels (. Pterodroma neglecta ) at Ilha da Trindade, South Atlantic Ocean and in the North Atlantic. Notornis 5 1 : 33-40. a Animal Ecology Group, Centre for Ecological and Evolutionary Studies, University of Groningen, PO Box 14, 9750 AA Haren, Netherlands. E-mail: c. eikenaar@biol. rug. nl b Seychelles Bird Records Committee, PO Box 336, Victoria, Seychelles, or Hazeley Brook, Keele Road, Keele, Staffs, ST 5 5AL, UK. E-mail: adrian @skerrett. fsnet. co. uk Received 26 May 2005; revision accepted 20 July 2005 90 - Bull ABC Vol 13 No 1 (2006) First record of Kermadec Petrel for Seychelles : Eikenaar & Skerrett Lake Bedo— a little-known wetland hotspot in Madagascar H. Glyn Younga and Felix Razafindrajao ^ Lac Bedo — un point chaud peu connu a Madagascar. Le Lac Bedo (19°55’S 44°32’E), situe au centre-ouest de Madagascar, entre Belo-sur-Tsiribihina et Morondava, est peu connu par les ornithologues etrangers visitant le pays. Ce lac, qui s’etend sur environ 400 ha, est pourtant un site important pour des oiseaux d’eau, tels que cigognes, ibis, spatules, flamants, herons, canards (la Sarcelle de Bernier Anas bernieri y est souvent presente) et limicoles. Les auteurs esperent que l’interet ornithologique et faeces facile au site attireront des visiteurs, qui sont invites a contribuer a une meilleure connaissance de l’avifaune en envoyant leurs observations a Madagascar@durrell . org. Waterbirds are widespread in Madagascar and include representatives of 20 families of true waterbirds with 36 endemic taxa recognised (Young 2003). A typical itinerary of visiting bird- ers includes some, usually small, wetlands, often adjacent to forest reserves. These sites are normal- ly well known, frequently visited and may hold one or more endemics such as Madagascar Teal Anas bernieri , Meller’s Duck A. melleri, Madagascar Plover Charadrius thoracicus or Slender-billed Flufftail Sarothrura watersi. Several larger wetlands, such as Lake Alaotra, are well known but rarely visited as they are difficult to reach and too large for easy searching. The major- ity of Madagascar’s other large wetlands, however, are rarely visited by ornithologists and their water- birds remain poorly known. Recent field work by local conservationists in western Madagascar has begun to show the importance of a series of saline lakes for a wide variety of waterbirds, in particular during the dry season. Of these lakes, those in Kirindy Mitea National Park, south of Belo sur Mer, and Lake Bedo, north of Morondava (Fig. 1), are of particular note. Whereas Kirindy Mitea remains very hard to get to, Lake Bedo (19°53’S 44°32’E) is very accessible and, being close to, but unseen from, the main Morondava to Belo-sur- Tsiribihina road, is on many visitors’ routes — a fact that makes the poor knowledge of this wet- land all the more surprising. With this brief summary we hope that more birders will visit Lake Bedo and may in some way contribute to the economy of nearby villages, thus increasing local peoples’ respect for the site and aiding researchers by sending in their records (to the authors at Madagascar@durrell.org). Lac Bedo hotspot, Madagascar: Young & Razafindrajao Figure 1 . Location of Lake Bedo in western Madagascar. Situation du Lac Bedo au centre-ouest de Madagascar. Description of site Lake Bedo is north-west of the village of Beroboka Nord at the southern end of an extensive area of tanne (the bare, salty, open ground on the land- ward side of coastal mangroves) that may be wide- ly flooded during the wet season (January-March), and that stretches many miles Bull ABC Vol 13 No 1 (2006) -91 to the north. Areas within the tanne may be hyper-saline and salt-encrusted pools may develop (northern parts of this tanne have been extensive- ly modified as a prawn farm). Lake Bedo, howev- er, is filled partly by fresh water from the rains and is only slightly saline. This salinity affects the wet- land’s flora and rush {Juncus)- like vegetation pre- dominates, and there are no water lilies Nymphea or beds of reed Phragmites or papyrus Cyperus. Typha marks freshwater channels and there are some areas of Typha marsh at the lake’s ends. It is, however, this general salinity that prevents the conversion of the wetland to the rizculture that typifies so many of Madagascar’s wetlands. Historically, these southern limits of the tanne have not been fully mapped or adequately described. However, it is likely that they have changed extensively even in recent years: Otto Appert visited in June-July 1974 and described the area as the ‘flooded area west of Beroboka (Appert 1996) and Roger Safford in August 1993 found ‘three small, seasonal pools (combined area of less than 1 ha)’, and local guides were certain that no larger water areas existed at that time (Safford 1993 and in litt. 2003). Today the shal- low lake occupies an area of r.400 ha, but the extent of open water may still be highly variable: in the height of the dry season, in September-November, it may occupy less than 1 00 ha in most years and in some it may dry out completely. The changeable nature of Bedo means that visitors in the future may, however, find a very different, but still highly important, wetland. Extensive areas of flat, sparsely vegetated sand and dried mud border the open water for much of the year. Dry forest at the lake’s western edge even- tually joins the coastal mangrove. Baobab trees are common in the forest, as are lemurs, notably Verreaux’s Sifaka Propithecus verreauxi. The area was previously well known to duck hunters and, although not within a reserve, is included in the Important Bird Area ‘Wetlands of the Tsiribihina delta and upper river’ (Projet ZICOMA 1999, 2001). Access Lake Bedo is approached by driving off the main Morondava to Belo-sur-Tsiribihina road, the road famous at its southern end as the ‘Allee des Baobabs’. A rough track to the lake goes through agricultural land, mostly grazing for cattle, and scrub before entering the dry forest bordering the eastern lake shore. The point where the track leaves the road is marked with a sign (which fur- ther informs visitors that hunting at the lake is for- bidden), but the sign is not obvious and visitors are advised to stop in the roadside village of Beroboka Nord to ask for directions and, possibly, for a guide. The lake can be reached by foot through interesting scrub and forest from Beroboka, but it may be preferable to drive as the vehicle may make a good hide. The short drive along the forested track offers opportunities for watching scrubland and forest birds, particularly at dawn and dusk (see below). The lake can be watched from any one point on the shore. Drinks etc. can be readily purchased at several small local stores or restaurants (hotelys) in Beroboka Nord village. Birds Storks, ibises, flamingos and herons The track comes out near the northern edge of the lake, where flocks of up to 200 African Spoonbills Platalea alba feed and may be joined by flamingos. The latter range widely in western Madagascar and numbers at the lake vary on an almost hourly basis: flocks of several hundred Greater Flamingos Phoenicopterus roseus in the morning may be replaced by equally large numbers of Lesser Flamingos Phoeniconaias minor by evening. Flocks may include both species and, in September and October 2004, recently fledged young whose ori- gin is unknown (did they hatch in Madagascar or Captions to photos on opposite page: Figure 2. Lake Bedo turn-off, 16 October 2004 (H. Glyn Young) Embranchement vers le Lac Bedo, 16 octobre 2004 (H. Glyn Young) Figures 3-7. Lake Bedo, September/October 2004 (H. Glyn Young) Lac Bedo, septembre/octobre 2004 (H. Glyn Young) Figures 8-9. Lake Bedo from the air, June 2005 (Durrell Wildlife Conservation Trust) Vue aerienne du Lac Bedo, juin 2005 (Durrell Wildlife Conservation Trust) Lac Bedo hotspot, Madagascar: Young & Razafindrajao 92 -Bull ABC Vol 13 No 1 (2006) Lac Bedo hotspot, Madagascar: Young & Razafindrajao Bull ABC Vol 13 No 1 (2006) -93 migrate from East Africa? P. minor is not known to breed in Madagascar). Large numbers of Glossy Ibises Plegadis fal- cinellus forage in the lake and roost at dusk after feeding in rice fields to the south. Yellow-billed Storks Mycteria ibis are often present and flocks of up to 50 have been seen, sometimes joined by African Openbill Storks Anastomus lamelligerus. Herons and egrets may be abundant, including flocks of several hundred Black Egrets Egretta arde- siaca and large numbers of Great Egrets E. alba. Up to 25 Madagascar Herons Ardea humbloti may be present throughout the lake. Black-crowned Night Heron Nycticorax nycticorax , Grey Ardea cinerea, Purple A. purpurea , Squacco Ardeola ral- loides and Green-backed Herons Butorides striatus are common. Dimorphic Egret Egretta dimorpha is rarer but usually seen, and singles of Little Bittern Ixobrychus minutus and the migratory Madagascar Pond Heron Ardeola idae are occasionally present. Madagascar White Ibis Threskiornis (< aethiopicus ) bernieri has been recorded at Lake Bedo in the past (Appert 1996) and, whilst not recorded here in recent visits, is still a possibility. Wildfowl Wildfowl numbers at Bedo can be exceptional at some times of year, particularly in the dry season as lakes elsewhere dry out, with Dendrocygna vid- uata and Fulvous Whistling Ducks D. bicolor, Red-billed Anas erythrorhyncha and Hottentot Teals A. hottentota the most common. Knob-billed (Comb) Duck Sarkidiornis melanotos too is often seen, as is, strangely (as there are no water lilies) African Pygmy Goose Nettapus auritus , although the latter may simply be moving between more typical habitats. Madagascar Teal nests in the near- by mangroves and can be found at the lake edges, usually shunning the flocks of other wildfowl and normally remaining in pairs: up to 17 have been counted recently. Red-knobbed Coots Fulica cristata may be common in the dry season (flocks of over 200) and may occasionally be joined by both Little Tachybaptus ruficollis and Madagascar Grebes T. pelzelnii. White-backed Duck Thalassornis leuconotus has not yet been recorded at Lake Bedo but has been seen nearby in the Andranomena Special Reserve, and although a lily specialist, like the pygmy geese, this highly disper- sive bird may too stop off at Bedo as it moves between more suitable sites. Rails Rails at Bedo are usually shy and rarely observed. The exception is the noisy and conspicuous Purple Swamphen Porphyrio porphyrio whose elephantine trumpetings can be heard even as the lake is approached. Common Moorhen Gallinula chloro- pus is widespread and common and White- throated Rail Dryolimnas cuvieri and Baillon’s Crake Porzana pusilla are relatively easy to find but their status at the lake is unclear — check the Typha beds. Sakalava Rail Amaurornis olivieri, as yet unrecorded at Lake Bedo, is a possibility as it has been found at other west coast lakes. Shore birds Black-winged Stilt Himantopus himantopus (96 were counted in December) often nests quite openly. The shores are frequented by large num- bers of small plovers. Kittlitz’s Charadrius pecuar- ius, White-fronted G marginatus and the rare endemic Madagascar Plover C. thoracicus occupy the drier areas and nest at the lake, whereas Three- banded C. tricollaris and wintering Greater Ringed Plover C. hiaticula can be found in the wetter areas. Madagascar Pratincole Glareola occularis too is easy to see when migrants from East Africa arrive in September and October. Greater Painted-snipe Rostratula benghalensis is common in the more vegetated areas of the lake shore, but may prove hard to find until dusk. In the northern winter large numbers of Common Greenshank Tringa nebularia, Curlew Sandpiper Calidris ferruginea (1,000+) and Common Sandpiper Actitis hypoleucos are present, and Grey Plover Pluvialis apricaria and Little Stint C. minu- ta have also been recorded. Other birds Whiskered Tern Childonias hybrida may be very common and be joined by smaller numbers of Caspian Tern Sterna caspia. Pink-backed Pelican Pelecanus rufescens has been seen three times recently (21 September 2003 and 18 November and 12 December 2004: Mwema & Razafindrajao 2006). Raptors, notably Yellow-billed Kite Milvus migrans aegyptius, are common and the migratory Eleanoras Falco eleonorae and Sooty Falcons E con- Lac Bedo hotspot, Madagascar: Young & Razafindrajao 94 -Bull ABC Vol 13 No 1 (2006) color feed at the lake. Several passerines are found around the lake amongst which Madagascar Swamp Warbler Acrocephalus newtoni is perhaps the most notable. Forest and scrubland birds The drive to the lake and the forest surrounding it contain several endemics. The kestrels on the dead palms should be checked, as at least one pair of Banded Kestrels Falco zoniventris occurs. Care must be taken not to run over coveys of Madagascar Buttonquails Turnix nigricollis that seem to like the track better than the surrounding land. Madagascar Nightjar Caprimulgus madagas- cariensis and Sickle-billed Vanga Falculea palliata are easily seen from the track. Acknowledgements We are indebted to Durrell Wildlife Conservation Trust for their commitment to the conservation of wetlands and their bird fauna in western Madagascar. Thanks are due in particular to Joanna Durbin, Richard Lewis, Bin Aboudou Abdalah Iahia and our driver Bruno Razanadraibe. Roger Safford has been a constant source of support and advice throughout work at Lake Bedo, where HGY abandoned him in 1993, and Ron Demey helped with the preparation of this article. Above all, however, we wish to recog- nise the support and assistance of Department des Eaux et Forets, especially the Chef Circonscription des Eaux et Forets, Morondava, and the people of the region, notably in Beroboka Nord. References Appert, O. 1996. A contribution to the ornithology of the region of Morondava, western Madagascar. Working Group on Birds in the Madagascar Region Newsletter 6 ( 1 ) : 18-54. Mwema, M. & Razafindrajao, F. 2006. First Pink- backed Pelican Pelecanus rufescens sightings in Madagascar since I960. Bull. ABC 13: 86-87. Projet ZICOMA. 1999. Les Zones dlmportance pour la Conservation des Oiseaux a Madagascar. Antananarivo: Projet ZICOMA. Projet ZICOMA. 2001. Madagascar. In Fishpool, L. D. C. & Evans, M. I. (eds.) Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Newbury: Pisces Publications & Cambridge, UK: BirdLife International. Safford, R. 1993. The Madagascar Teal Anas hernieri. A preliminary survey from Antsalova to Morondava. Dodo, J. Wildlife Preservation Trusts 29: 95-102. Young, H. G. 2003. Freshwater birds. In Goodman, S. M. & Benstead, J. P. (eds.) The Natural History of Madagascar. Chicago: University of Chicago Press. aDurrell Wildlife Conservation Trust, Les Augres Manor, Trinity, Jersey JE3 5BP, UK. E-mail: Glyn. Young@durrell.org b Durrell Wildlife Conservation Trust, BP8J11, Antananarivo 101, Madagascar. Received: 18 May 2005; revision accepted 22 July 2005 Lac Bedo hotspot, Madagascar: Young & Razafindrajao Bull ABC Vol 13 No 1 (2006) -95 96 - Bull ABC Vol 13 No 1 (2006) Recent Reports Recent Reports These are largely unconfirmed records published for interest only; records are mostly from 2005, with a few from earlier dates. We thank all birders who have sent in their records and urge them to submit frill details to the relevant national or regional organisations. It is sug- gested that observations of each species be compared with relevant literature to set new data in context and that observers who are unfamil- iar with the status of birds in a par- ticular country refer to R. J. Dowsett’s (1993) Afrotropical avi- faunas: annotated country checklists (in: R. J. Dowsett & F. Dowsett- Lemaire. A Contribution to the Distribution and Taxonomy of Afrotropical and Malagasy Birds. Tauraco Res. Rep. 5. Liege: Tauraco Press) or more recent or appropriate sources before submitting records. Les observations ci-apres sont en majeure partie non confirmees et sont publiees uniquement dans le but d’informer. La plupart des don- nees sont de 2005; quelques-unes sont plus anciennes. Nous remer- cions tous les ornithologues qui ont pris la peine de nous faire parvenir leurs donnees et nous recomman- dons de les envoyer, dument docu- mentees, aux organisations nationales ou regionales concernees. II est conseille de verifier le statut des especes observees dans la littera- ture appropriee, afm de mettre les nouvelles donnees en perspective, et de consulter notamment R. J. Dowsett (1993) Afrotropical avifau- nas: annotated country checklists (in: R. J. Dowsett & F. Dowsett- Lemaire. A Contribution to the Distribution and Taxonomy of Afrotropical and Malagasy Birds. Tauraco Res. Rep. 5. Liege: Tauraco Press) ou des sources plus recentes ou appropriees. Angola In October 2005, the first ABC con- servation tour recorded 14 Angolan endemics and several range exten- sions. For a succinct report, see Club News pp.6-7. Azores The following records are from April-August 2005. On the Salvages, up to four Red-billed Captions to figures on opposite page Tropicbirds Phaethon aethereus remained from 25 May until at least 1 5 June, and a pale-morph Eleonora’s Falcon Falco eleonorae was seen on 10 June. An American Coot Fulica americana was at Furnas Lagoon, Sao Miguel, on 8-16 April at least. A Bar-tailed Godwit Limosa lapponica and a Black-tailed Godwit L. limosa were found at Cabo da Praia, Terceira, on 1 1 June. On Praia islet, Graciosa, an adult Bridled Tern Sterna anaethetus remained from mid May to early August. Other August reports included a Lesser Yellowlegs Tringa flavipes also there on 4th. A Semipalmated Plover Charadrius semipalmatus was observed at Cabo da Praia, Terceira, on 3 August. Least Sandpipers Calidris minutilla were at Capelinhos lighthouse, Faial, from 30 July to 2 August and at Cabo da Praia, Terceira, on 3 August. An adult and a pair of Sooty Terns Sterna fuscata were present on Ilheu da Vila this year (reportedly, one chick hatched). On 4 August, a Bridled Tern was recorded at Ilheu da Praia, Graciosa. Three Common House Martins Delichon urbicum were at Povocao, Sao Miguel, on 13 April, and a male Black-eared Wheatear Oenanthe his- panica was reported from the Salvages on 1 1 June. An exceptional number of Nearctic vagrants was reported in September-November 2005, with more than ten potential first records, mostly on the westernmost islands of Corvo and Flores. A Pied-billed Grebe Podilymbus podiceps was at Figure 1. Chimney Swift / Martinet ramoneur Chaetura pelagica , Sao Miguel, Azores, 28 October 2005 (Frederic Jiguet) Figure 2. Leach’s Storm-petrel / Oceanite culblanc Oceanodroma leucorhoa, off La Palma, Canary Islands, 6 October 2005 (Edwin Winkel) Figures 3-4. Presumed South Polar Skua / presume Labbe de McCormick Stercorarius ( Catharacta ) maccormicki , off La Palma, Canary Islands, 6 October 2005 (Edwin Winkel) Figure 5. River Prinia / Prinia aquatique Prinia fluviatilis , Komadougou Yobe River, Diffa, Niger, 23 September 2005 (Flemming Jensen) Figure 6. American Golden Plover / Pluvier bronze Pluvialis dominica, Technopole, Dakar, Senegal, 29 October 2005 (Wouter Favyets) Figure 7. American Wigeon / Canard d’Amerique Anas americana , Djoudj National Park, Senegal, 20 December 2005 (Amine Flitti) Figure 8. Laughing Gull / Mouette atricille Larus atricilla , Sine Saloum, Senegal, 28 December 2005 (Amine Flitti) Recent Reports Bull ABC Vol 1 3 No 1 (2006) -97 Lagoa Azul, Sao Miguel, in the first half of November (presumably first seen at Lagoa Furnas on 27 October), with a second at Lagoa da Funda, Flores, on 9 November, and a report from Terceira, on 1 3 November. Blue-winged Teals Anas discors were reported from Cabo da Praia, Terceira (one), Lagoa Azul, Sao Miguel (two) and Flores (four). Other vagrant waterfowl included up to seven Ring- necked Ducks Aythya collaris and up to three Lesser Scaup A. affinis at Lagoa Azul, Sao Miguel, in the first half of November and, on Flores, a female American Wigeon Anas americana on 2 November, three American Black Ducks A. rubripes on 8 November, with one there on 18 th, and three Ring-necked Ducks also on 18th. American Bitterns Botaurus lentiginosus were found at Faja dos Cub res, Sao Jorge, on 21 September, on Sao Miguel on 27 October, and on Flores on 17 November. Up to four Great White Egrets Egretta alba of the race egretta were identified on Sao Miguel from 1 November and one was seen on Terceira on 4 November. A Great Blue Heron Ardea herodias stayed on Flores from 22 September to at least 21 November. An immature Hen Harrier Circus cyaneus hudsonius flew over Lagoa Rasa, Flores, on 8 November. A crake, at Praia da Vitoria, Terceira, from 1 9 October to 2 November, was not a Sora Rail Porzana Carolina but proved to be the first Spotted Crake P. parva for the Azores since 1 966. A Killdeer Charadrius vociferus was on Flores on 9-19 November at least. One or two Least Sandpipers were seen at Cabo da Praia, Terceira, on 20-22 September and from 23 October to at least 1 3 November. On Flores, a Wilson’s Snipe Gallinago gallinago delicata was reported at Lagoa Rasa on 8 November. Three Short-billed Dowitchers Limnodromus griseus were seen at Cabo da Praia on 1 3-22 September and one was at Fajo dos Cubres, Sao Jorge, on 21 September. A (Hudsonian) Whimbrel Numenius phaeopus hudsonicus was reported from Flores and Corvo. An Upland Sandpiper Bartramia longicauda stayed on Flores from 30 October to at least 1 1 November and a Solitary Sandpiper Tringa solitaria was on Terceira on 16-22 September. A Greater Yellowlegs T. melanoleuca was observed on Flores on 30 October. Other waders in autumn 2005 included up to seven Semipalmated Plovers Charadrius semipalmatus , a few American Golden Plovers Pluvialis dominica, several Semipalmated Sandpipers Calidris pusilla , small groups of White- rumped Sandpipers C. fuscicollis, several Pectoral Sandpipers C. melanotos, a few Buff-breasted Sandpipers Tryngites subruficollis, three Lesser Yellowlegs Tringa flavipes , one Solitary Sandpiper T. solitaria and several Spotted Sandpipers Actitis macularius. Up to four Laughing Gulls Lams atricilla were seen on Flores from 1 November, an adult was on Terceira on 4 November, and up to nine were counted at Ponta Delgada har- bour on Sao Miguel from 7 November. First-winter Franklin’s Gulls L. pipixcan were present on Terceira on 4-12 November at least, and on Sao Miguel on 13-18 November. This influx of Laughing and Franklin’s Gulls was also record- ed in Morocco and south-western Europe, with large numbers of the first-named species also reaching the British Isles. Two American Herring Gulls L. argentatus smithsonianus were identified on Flores on 17 November. A first-winter Forster’s Tern Sterna forsteri was observed on Terceira on 30 October and 10 November. An American Mourning Dove Zenaida macroura on CQrvo on 2 November was the first for the Azores. A Yellow-billed Cuckoo Coccyzus americanus was also seen on Corvo, on 19 October. A Common Nighthawk Chordeiles minor was reported from Faial on 6 November, with another on Terceira on 1 1 November. From 28 October to 7 November, c. 150 Chimney Swifts Chaetura pelagica were recorded, from Corvo (up to 27) and Flores (up to 21) to Faial (12), Terceira (up to 14) and Sao Miguel (up to 30); see Fig. 1 . Tree Swallows Tachycineta bicolor were seen at Sete Cidades, Sao Miguel, on 26-27 October (one) and 2 November (two), on Corvo on 5-6 November (one) and at Punta Lopo Vaz, Flores, on 2 November (two). The latter were accompanied by an American Barn Swallow Hirundo rustica erythrogaster. The first American Barn Swallow for the Azores was a juvenile at Ponta Delgada, Flores, on 30-31 October. An American Cliff Swallow Hirundo pyrrhonota on Corvo on 5 November was the second for the islands. The first and second American Buff- bellied Pipits Anthus rubescens rubescens for the Azores were on Corvo on 29 October and at Cabo da Praia, Terceira, on 2-3 November. A Grey-cheeked Thrush Catharus minimus was found on Corvo on 1 November. A White-eyed Vireo Vireo griseus on Corvo on 22 October was another first, not only for the archi- pelago but for the Western Palearctic. Also on Corvo, a Red-eyed Vireo V. olivaceus was present on 26-28 October and a Philadelphia Vireo V philadelphicus on 26 October. If accepted, a male Common Crossbill Loxia curvirostra at Pico da Vara, Sao Miguel, on 18 September will also be a first record for the Azores. A Myrtle (Yellow-rumped) Warbler Dendroica coronata on Sao Miguel on 2 1 November was the second for the islands. More firsts for the Azores, all from Corvo, included a first-winter Black-throated Blue Warbler D. caerulescens from 24 October to 1 November at least, an adult male Hooded Warbler Wilsonia citrina on 26-27 October, an Arctic Redpoll Carduelis hornemanni on 20 October, a first-winter Scarlet Tanager Piranga olivacea on 29 October and 5 November, a Lapland Longspur Calcarius lapponicus on 28 October, and up to seven Indigo Buntings Passerina cyanea on 24-27 October, with three on Flores on 1 November and one on 3-9 November. Other Nearctic vagrants on Corvo included an Ovenbird Seiurus aurocapilla on 1 November, a White-crowned Sparrow Zonotrichia leucophrys on 25-27 October, a first- winter Rose- 98 - Bull ABC Vol 13 No 1(2006) Recent Reports breasted Grosbeak Pheucticus ludovi- cianus from 24 October to 9 November at least (with another at Lagoa Lomba, Flores, on 1 9 November), two Bobolinks Dolichonyx oryzivorus on 1 9-24 October, and an immature Baltimore Oriole Icterus galbula on 6 November (per Birding World 18: 200, 240, 324 & 377, 434, 455^56 & 465-478; per Dutch Birding 27 : 349-350 & 413-425). Benin Belated reports from August-October 2003 include the following. A Dwarf Bittern Ixobrychus sturmii flew over the university at Calavi on 4 August, one was at the Kota waterfalls on 9 October, and several were near Malanville on the shores of the River Niger on 11-13 October. An Ahanta Francolin Francolinus ahantensis was seen between the villages of Hlagba Denou and Deme on 16 September. Lesser Moorhen Gallinula angulata was seen a few times near Malanville on 1 1-13 October. A few Caspian Terns Sterna caspia were at Djacquot beach on 27 August and two were at Ouidah on 24 October. Rosy Bee- eater Merops malimbicus was found to be common near the village of Lokoli. Two Speckled Tinkerbirds Pogoniulus scolopaceus were observed in the southern part of the forest of Lokoli on 8 September, and one in the northern part on 30 September. Other records from Lokoli include a Red-winged Warbler Heliolais ery- thropterus, a pair of Buff-throated Apalises Apalis rufogularis on 2 October, a Red-bellied Paradise Flycatcher Terpsiphone rufiventer on 26 September, and a male Buff- throated Sunbird Chalcomitra adel- berti seen well on 1 September. A male Marsh Tchagra Antichromus minutus with a begging juvenile Black Cuckoo Cuculus clamosus was observed near Lokoli on 3 September. Grosbeak Weavers Amblyospiza albifrons were seen a few times at Lokoli and a small flock was at the edge of the forest near Deme. Two Grey-headed Negrofinches Nigrita canicapillus were near Lokoli on 25 September (SF). More belated reports include African Openbill Storks Anastomus lamelligerus , found to be common in and around Cotonou and Ganvie in October 2004, a Lesser Kestrel Falco naumanni seen in Cotonou on 20 March 1999, a Purple Swamphen Porphyrio porphyrio at Bymin, near Porto Novo, also in March 1999, and a Fire-crested Alethe Alethe diadema- ta , apparently the first for Benin, recorded on 27 March 1999 {JG). Botswana Records from the period January 2005-January 2006 include the fol- lowing. There was a high count of 166 Great Crested Grebes Podiceps cristatus at Bokaa Dam, in the south- east, on 24 July. At Letsibogo Dam, in eastern Botswana, there was an exceptional count of 450 Great Cormorants Phalacrocorax carbo on 1 July. Counts at just eight roosts in the Okavango Delta in July-August produced over 1 ,200 Slaty Egrets Egretta vinaceigula or dark egrets, pre- sumed to be Slaty Egrets (and not Black Herons E. ardesiaca ), the majority of these at just four roosts. On 1-2 July, 23 Pink-backed Pelicans Pelecanus rufescens were seen at Letsibogo Dam; six nests were occupied. At a ’heronry’ on islands in Lediba la Dinonyane near Kananain in the Okavango Delta, 33 Pink- backed Pelican nests were counted on 13 September, with adults incu- bating on most. There were also 20 nests of Marabou Storks Leptoptilos crumeniferus and 592 nests of African Openbill Storks Anastomosus lamel- ligerus. One Black Stork Ciconia nigra was seen on the Thamalakane River in Maun on 23 April and three along the Tati River at Francistown on 11 February. An estimated 10,000-12,000 Lesser Flamingos Phoeniconaias minor were nesting on Sua Pan in late January 2005. On 2 March, some 5,000-10,000 birds were still there, seen from Thlapama Hill near Mea Pan. Apart from small numbers of Lesser Flamingos at dams and sewage ponds in east and south-east Botswana, there were c.300 at Bokaa Dam on 3 December and c.400 at Lake Ngami in mid December, plus 330 Greater Flamingos Phoenicopterus {ruber) roseus. A young Cape Vulture Gyps coprotheres , fitted with a satellite- tracking device at Waterberg, Namibia, spent two weeks in the western part of the Okavango Delta in mid 2005 (per Raptors Namibia Newsletter 4 June 2005). Lake Ngami, in north-west Botswana, flooded for the second consecutive year. On 26 July and 7 August there were 33 and 46 White Storks Ciconia ciconia , respectively. These were presumably European birds staying on. In September there was a single Whimbrel Numenius phaeopus. In mid December, when the lake was drying, there were 750 Great White Pelicans Pelecanus onocrotalus , 67 Pink-backed Pelicans, flamingos (see above), many water- fowl including 6,400 Red-billed Teal Anas erythrorhyncha, an Osprey Pandion haliaetus, c. 1 5 Lesser Kestrels Falco naumanni , 1,000 Black-winged Pratincoles Glareola nordmanni , 30 Common Ringed Plovers Charadrius hiaticula , 60 Caspian Plovers C. asiaticus, five Grey Plovers Pluvialis squatarola and 1 2 Black-tailed Godwits Limosa limosa. Two Corncrakes Crex crex were ringed in Maun at the start of December. Two Grey Crowned Cranes Balearica pavonina were near Maya Pan, in Moremi Game Reserve, on 2 1 May and one was south-east of Mmatshumo, in 2125B4, in September 2005. About 300 Chestnut-banded Plovers Charadrius pallidus , two Greater Sand Plovers C. leschenaultii , a Grey Plover, four Sanderling Calidris alba and a Eurasian Curlew Numenius arquata were at the drying edge of Sua Pan and in dying pools in the River Nata, in the Nata Delta, on 12 August. At Bokaa Dam there was a Black-tailed Godwit on 25 September and three Grey Plovers and two Black-winged Pratincoles on 3 December {ST; HB, NBo, CBr, UF, PH, RH, GM, ZM, MM per ST). Recent Reports Bull ABC Vol 13 No 1 (2006) -99 Cameroon Twenty-six White-faced Whistling Ducks Dendrocygna viduata were counted in Bamenda town on 1 2 November 2005 (JvdW). A surprise find was a Secretary Bird Sagittarius serpentarius at the Tiko Golf Club, between Douala and Limbe, on 12 August and 27 September (. HvdL ). Flocks of Grey-headed Gulls Lams cirrocephalus were seen at Lake Lagdo, near Garoua in the north, in the first half of December 2005 (JG). Canary Islands Records from pelagic trips off Puerto Rico, Gran Canaria, on 10-14 May 2005 included ten Bulwer’s Petrels Bulweria bulwerii , three Wilson’s Storm-petrels Oceanites oceanicus, 1 1 White-faced Storm-petrels Pelagodroma marina , four Madeiran Storm-petrels Oceanodroma castro , a Red-billed Tropicbird Phaethon aethereus and six Roseate Terns Sterna dougallii (per Birding World 18: 200; per Dutch Birdingll : 274). Noteworthy records from July-December 2005 include the fol- lowing. An adult Red-billed Tropicbird was seen close inshore off Roque Grande, El Hierro, on 9 July, and an adult Roseate Tern was off Tazacorte, La Palma, on 17 July (per Birding World 18: 281; per Dutch BirdingU: 346). A Leach’s Storm- petrel Oceanodroma leucorhoa was photographed 1 mile off Tazacorte, La Palma, on 6 October (EW; Fig. 2). A few Common Shelduck Tadorna tadorna , Tufted Duck Aythya fuligula and a female Pintail Anas acuta were on a pond by La Lajita, Fuerteventura, on 28 August; although they appeared to be wild these records are rather early in the year and the birds perhaps had escaped from a nearby zoo (PL & DR). A male Ring-necked Duck Aythya collaris was at Catalina Garda, Fuerteventura, on 10 October (per Birding World 18: 434). Two female Greater Scaup A. marila were at Salinas del Janubio lagoon on 16 November (NS). A female Common Scoter Melanitta nigra was at Valle Molina Reservoir, Tenerife, on 26 December (KT). On Fuerteventura, an Eleonora’s Falcon Falco eleonorae flew over Corralejo on 27 August, and a Peregrine Falcon F. peregrinus was seen near the lighthouse at El Cotillo on 30 August (PL & DR). A White-rumped Sandpiper Calidris fuscicollis was reported from Salinas del Carmen, Fuerteventura, on 20-22 October (per Birding World 18: 434). A Spotted Sandpiper Actitis macularius was near Puerto de la Cruz, Tenerife, on 30 October. A presumed South Polar Skua Stercorarius (Catharacta) maccormicki photographed 2 miles off La Palma on 6 October may be the first for the Canaries if accepted; the skuas outer primary was in growth, indicating moult, which in late summer is strong indication of Southern Hemisphere origin (EW; per Dutch BirdingU: 404-424; Figs. 3-4). Two Chimney Swifts Chaetura pelagica were on Tenerife on 29-3 1 October and five on Lanzarote on 1 November (per Dutch BirdingU: 404-424). A small ’fall’ of migrants on 4 September on the Jandia Peninsula, Fuerteventura, included several Olivaceous Warblers Hippolais pallida and two Melodious Warblers H. polyglotta; on Lanzarote several Melodious Warblers were also reported next day (PL & DR). The Northern Mockingbird Mimus poly- glottos at Arguineguin, Gran Canaria, was still present on 3 October; it was first reported in November 2004 (cf. Bull. ABC 12: 179). A juvenile Common Rosefinch Carpodacus ery- thrinus was claimed from Alegranza, Lanzarote, on 1 October; if accepted, it will be the first for the Canary Islands (per Dutch Birding 27 : 404^24). Cape Verde Islands Records from October-November 2005 include the following. The first Snowy Egret Egretta thula for the archipelago was at the sewage ponds near Mindelo, Sao Vicente, on 1-3 November. On 31 October, 46 Cape Verde Purple Herons Ardea (purpurea) bournei, including 26 juveniles, were counted at the species’ only nesting tree, at Liberoa, Santiago. Also on Santiago, a Glossy Ibis Plegadis falcinellus was seen on 30 October. A first-winter Semipalmated Plover Charadrius semipalmatus was found at Tarrafal, Santiago, on 30 October. Other vagrant waders at the sewage ponds of Mindelo, Sao Vicente, on 1-3 November, included a first-winter Semipalmated Plover, an American Golden Plover Pluvialis dominica, three Semipalmated Sandpipers Calidris pusilla , three White-rumped Sandpipers C. fuscicollis , a Green Sandpiper Tringa ochropus and a Spotted Sandpiper Actitis macularius (per Dutch Birding 27: 404-408). At Praia, Santiago, a White-rumped Sandpiper and a Spotted Sandpiper were seen on 31 October, with another White-rumped Sandpiper at Santa Maria, Sal, on 5 November (per Birding World 18: 456-457)- The captive Barn Owl Tyto alba taken as a chick on Maio in 2004 was not the first for the island, as erroneously stated in a previous Recent Reports (Bull. ABC 12: 180): the first published record was indeed made in October 2000, but breeding had already been recorded in March that year by another observer and was eventually published (RB). Two Greater Hoopoe Larks Alaemon alaudipes were discovered on Sal on 5 November; the species is only known as a breeder on Boavista and Maio (per Dutch BirdingU: 404-408). Greater Hoopoe Lark Alaemon alaudipes (Pete Leonard) 100 - Bull ABC Vol 13 No 1 (2006) Recent Reports Central African Republic The following records, from the peri- od May 2005-January 2006, are from around the village of Djoubissi, c.70 km north of Bambari, Ouaka prefecture, in the centre of the coun- try. Additions to the country list are Alpine Swift Tachymarptis melba, observed on 15 May, and Isabelline Wheatear Oenanthe isabellina, seen on 13-14 November, whereas Great Snipe Gallinago media , found on 14 September, and White Wagtail Motacilla alba , remaining during 4-14 November and 29 November-9 January at least, are rarely recorded vagrants. Other interesting records, com- pared to the known distribution of these species ( cf maps in Borrow & Demey, 2004, Field Guide to the Birds of Western Africa) include the follow- ing (records are of single birds, unless otherwise indicated): Saddle-billed Stork Ephippiorhynchus senegalensis (15 July), Common Quail Coturnix coturnix (8 December), Spotted Thick-knee Burhinus capensis (7 May), Ross’s Turaco Musophaga rossae (28 December and 6 January), Pallid Swift Apus pallidus (20 on 22 November and 150 on 29 November), Black-headed Bee-eater Merops breweri (two on 23 December and 2 January), Speckled Tinkerbird Pogoniulus scolopaceus ( 1 1 November), Speckle-breasted Woodpecker Dendropicos poecilolae- mus (two on 8 May, one on 1 2 November), Grey-rumped Swallow Pseudhirundo griseopyga (multiple records), White-throated Blue Swallow Hirundo nigrita (two on 29 July), Ansorge’s Greenbul Andropadus ansorgei (11 November, at least four on 1 December), Common Redstart Phoenicurus phoenicurus (multiple records, includ- ing five on 31 October), Common Rock Thrush Monticola saxatilis (2 January), Foxy Cisticola Cisticola troglodytes (2 January), Buff-throated Apalis Apalis rufogularis (two on 1 1 November, one on 1 December), Shrike Flycatcher Megabyas flammu- latus (two on 1 1 November and 1 December), Brown-crowned Tchagra Tchagra australis (5 May), Purple Glossy Starling Lamprotornis purpureus (multiple records), Heuglin’s Masked Weaver Ploceus heuglini (multiple records, including a breeding pair), Compact Weaver Pachyphantes superciliosus (two on 22 May) and Brown-rumped Bunting Emberiza ajfinis (2 January) (NV). Congo-Brazzaville Records from April-October 2005 include the following. A female Red- footed Falcon Falco vespertinus was seen in Lac Tele Community Reserve in early October (HR). A Helmeted Guineafowl Numida meleagris was killed at Tchissinga, near Diosso north of Pointe-Noire, in April; apparently the species is quite regu- larly caught in semi-forested hills along the coast (TD). In Brazzaville, Common Kestrel Falco tinnunculus was recorded breeding in August and Ring-necked Dove Streptopelia capi- cola was calling in April. In Lac Tele Community Reserve, an African Grass Owl Tyto capensis was seen in July and more than 850 Rosy Bee- eaters Merops malimbicus were found breeding in at least two colonies. Four Palearctic migrants, Eurasian Marsh Harrier Circus aeruginosus, Great Snipe Gallinago media , Green Sandpiper Tringa ochropus and Barn Swallow Hirundo rustica , were observed in the savanna on 5 September, which is rather early. Weyn’s Weaver Ploceus weynsi was seen again on a few occasions (HR) . Egypt Records from the period September 2005-January 2006 include the fol- lowing. An adult Goliath Heron Ardea goliath and a first-winter Lesser Sand Plover Charadrius mongolus were at Wadi Lahami on 6 November (per Birding World 18: 457). Twenty Red-breasted Geese Branta ruficollis were observed from a cruise boat on the Nile at Esna, in late November (AM). At Ain Sukhna, an adult Greater Spotted Eagle Aquila clanga was seen on 30 December and 2-4 January, and an immature and four adult Steppe Eagles A. nipalensis on 2 January (DRo). At least 29 Lappet- faced Vultures Torgos tracheliotus were Red-breasted Geese Branta ruficollis (Pete Leonard) counted at Ber Shalatan on 1 8 September. A Demoiselle Crane Anthropoides virgo was seen in a flock of 130 Common Cranes Grus grus at Hurghada on 1 0 October. A female Namaqua Dove Oena capensis was photographed at Wadi Gama on 1 9 October (per Dutch Birding 27 : 404-408). The latter species was also reported from Hurghada airport dur- ing October (per Birding World 18: 434). A Red-tailed Wheatear Oenanthe xanthoprymna was observed at Giza pyramids on 16 November (per Birding World 18: 457). Ethiopia In December 2005 the following species were reported. A pair of White-backed Duck Thalassornis leu- conotus with seven young was observed at Lake Chelekleka on 2nd. A Western Banded Snake Eagle Circaetus cinerascens was at Lake Langano on 8th and two were at Wadera Forest, near Kibre Mengist, on 16th. A juvenile Eastern Imperial Eagle Aquila heliaca was at Lake Langano on 9th. Three Degodi Larks Mirafra degodiensis were claimed from Bogol Mayo on 18th, and six Sidamo Larks Heteromirafra sidamoensis from the Liben Plains, Negele, on 16th. A Golden Pipit Tmetothylacus tenellus was seen near Negele on 1 6th and a Philippas Crombec Sylvietta philip- pae near Bogol Mayo, on 18th (EE). Two female or first-year Menetries Warblers Sylvia mystacea were found at Lalibella on 18 November 2004 (BP). Recent Reports Bull ABC Vol 13 No 1 (2006) -101 Gabon A White Stork Ciconia ciconia was seen near Port Gentil in November 2005 (EdB) . A remarkable record is that of two males and a female Pennant-winged Nightjar Macrodipteryx vexillarius captured on video as they were flying around an oil platform, 1 0 miles west of Pointe Iguele, on the Gabon coast, at 02°35’S, in mid-July 2005 (PM). The Gambia Records from July-December 2005 include the following. A group of 25 Wilson’s Storm-petrels Oceanites oceanicus , encountered c. 25 miles off the Gambian coast at 13°30’N 17°02’W on 2 July, is the largest group recorded in Gambian waters in recent years (CB & MGr). A juvenile Dwarf Bittern Ixobrychus sturmii at a seasonal pond in Sao Forest Park, on the north bank in Central River Division (CRD), was well watched and photographed on 7 October; this is a rare bird in The Gambia (PF & CB). A young Helmeted Guineafowl Numida meleagris killed on the road and collected 1 0 km west of N’Jau, CRD, on 8 October, is the first proof of breeding in that Division. At Kaur, CRD, a record 107 White-headed Lapwings Vanellus albiceps were counted on 8 October. At the same site, c.2,000 Collared Pratincoles Glareola pratin- cola roosted on 5 October (CB). In December, a Greater Sand Plover Charadrius leschenaultii was found at Palm Grove lagoon, Banjul, on 12th and an American Golden Plover Pluvialis dominica at Cape Creek, near Banjul, on 4th (DL per CB). Several calling Adamawa Turtle Doves Streptopelia hypopyrrha were at Kunkilling and Tankandama Eco- Trails, CRD, on 15 December (CB). Ghana Two American Golden Plovers Pluvialis dominica were in rice fields c.50 km east of Accra on 12-13 November 2005 (AH, HF & CP). A Red-necked Phalarope Phalaropus lobatus was at the Panbros Salt Works, outside Accra, on 1 5 January 2006 (AH, HF & PS). At Sakumo Lagoon, a Black Noddy Anous minu- tus was discovered in a roost of a few thousand terns Sterna spp. and three African Skimmers Rynchops flavirostris on 9 July 2005, and was still there on 11th (AH). A Common Chiffchaff Phylloscopus collybita was seen in Mole National Park on 16 January 2006 and three Common Waxbills Estrilda astrild were observed in Accra on 9th (RCr). A belated and intriguing report was received of an adult male Wattled Starling Creatophora cinerea in non-breeding plumage, seen at Dansoman, Accra, from a distance of 15-20 m, around 1 July 2003 (SHo). Guinea A Yellow-casqued Hornbill Ceratogymna elata was seen in Conakry’s botanic garden on 1 5 March 2005; a rather surprise find of this species in a very small patch of forest in the centre of a busy city (BP). Kenya The following records are from May-October 2005, unless otherwise stated. A Little Bittern Ixobrychus minutus was at Nguuni Wildlife Sanctuary, Mombasa, on 28 July. Nine Madagascar Pond Herons Ardeola idae were observed in the Mara on 4 July and one near the Carnivore Restaurant, Nairobi, on 3 August. Two White Storks Ciconia ciconia were near Nakuru on 2 July. A group of 75 African Open-billed Storks Anastomus lamelligerus in the Mara on 4 July is a large concentra- tion for this site. A Eurasian Honey Buzzard Pernis apivorus was at Kakamega on 19 July; sightings out- side October-April are rare. Montagu’s Harriers Circus pygargus were recorded at Kitengela, near Nairobi, on 23 June (a subadult male), at Amboseli on 24 June (an adult male and female) and at Naro Moru on 30 June (a female). A young female Eurasian Marsh Harrier C. aeruginosus was in Amboseli on 24 June and another in the Mara on 5 July. A Lizard Buzzard Kaupifalco monogrammicus was in Nairobi, where it is uncommon, on 20 July. A Long-legged Buzzard Buteo rufinus was reported from Mundui, Naivasha, around 20 October, and a Grey Kestrel Falco ardosiaceus from the Marigat area on 8 July. An exhausted Corncrake Crex crex was rescued from a cat at Watamu on 21 September, but later died; this is the first — and an early — record of southward migration on the coast. Possibly the first Shelley’s Francolin Francolinus shelleyi for the Mara was noted on 5 July. A Stone Partridge Ptilopachus petrosus between Kalacha and Loiyangalani, on 25 April, was at the north-eastern edge of this uncommon species’ range. Four Lesser Jacanas Microparra capensis were found on a pond near Kipsaos shopping centre, Eldama Ravine-Eldoret road, on 20 October. About 20 Black-tailed Godwits Limosa limosa were at Lake Naivasha on 25 June; a high number for this late date. A Marsh Sandpiper Tringa stagnatilis and a Wood Sandpiper T. glareola at Nakuru on 2 July were possibly early-returning migrants. A Common Sandpiper Actitis hypoleu- cos at Amboseli on 24 June represents an unusual date. A dark-morph Pomarine Skua Stercorarius pomari- nus was reported from Lake Naivasha on 20 September; there are fewer than 20 records of this species for Kenya. About 1,100 pairs of Roseate Terns Sterna dougallii bred on Whale Island, Watamu, in July-September, together with c.20-30 pairs of Sooty Terns S. fuscata (one juvenile Sooty was the first confirmed breeding suc- cess for this species here); 66 Roseate pulli and one adult were ringed. On the Sabaki River estuary, 44 Brown Noddies Anous stolidus were counted on 4 June and 49 on 16 September; this species is rare inshore and these numbers are particularly unusual; a pair possibly bred on Whale Island, Watamu, in July-September. In Kakamega Forest, 23 Grey Parrots Psittacus erithacus were recorded on 8 October; ten years ago this species was thought to be extinct at the locality. On 17 September, a Great Spotted Cuckoo Clamator glandarius was reported near Mt Suswa, Rift 102 - Bull ABC Vol 13 No 1 (2006) Recent Reports Grey-crested Helmet-shrike Prionops poliolophus (Pete Leonard) Valley, whilst European Bee-eaters Merops apiaster were seen in Nairobi. A Grey-throated Barbet Gymnobucco bonapartei at Molo on 3 July was far to the east for this species. A pair of Zanzibar Sombre Greenbuls Andropadus importunis was found breeding at Karura Forest, on the edge of Nairobi, on 18 May; this species is very uncommon around the capital. A female Northern Wheatear Oenanthe oenanthe at Eremito Gate, Amboseli, on 25 June is an unusual date for this species and a Pied Wheatear Oe. pleschanka in Ngong Hills on 9 September is an early record. Two Spotted Ground Thrushes Zoothera guttata were ringed at Gede Ruins on 6 October; a further two were found at a new site on the northern edge of Arabuko-Sokoke Forest on 29 October. A pair of Karamoja Apalises Apalis karamojae sighted in the Mara on 6 July was probably the same pair as previously observed (still to be accepted as the first for Kenya by the East African Rarities Committee). Six Hinde’s Babblers Turdoides hindei occurred by the Nyeri road, at the Tana River bridge, on 23 July; the species was also recorded breeding at Wajee Camp, Mukurweini, on 16 August. At the Oserian Wildlife Sanctuary, 16 Grey-crested Helmet- shrikes Prionops poliolophus were seen on 1 August. An Olive Sunbird Cyanomitra olivacea ringed on 1 0 September at A Rocha Kenya proper- ty, Karen, is the first record for Nairobi (per CJ). Three Abbott’s Starlings Pholia femoralis , together with Sharpe’s Ph. sharpii, Kenrick’s Poeoptera kenricki and Waller’s Starlings Onychognathus walleri, were observed at Mountain Lodge on 28 June (per CJ), with two also there in October and two more in Kieni Forest in the same month {BAP); an adult male seen feeding young at a nest in Gatamaiyu Forest on 20 and 22 October is the first confirmed breeding record in Kenya (per CJ; BP, CK). Liberia During field work in three National Forests, North Lorma in the north- west, Gola in the west and Grebo in the east, from 19 November to 1 1 December 2005, some noteworthy records were made. Olive Ibis Bostrychia olivacea was found in North Lorma; the species is not men- tioned for the Wologizi area by Gatter (1997, Birds of Liberia). A Spot-breasted Ibis Bostrychia rara flew over the village of Jalipo, Grebo, on 7 and 1 1 December. A melanistic Black Sparrowhawk Accipiter melanoleucus was seen in Gola; although this is a widespread forest resident in Liberia, it had not previ- ously been recorded at this site. A Lanner Falcon Falco biarmicus was seen at Monrovia on 14 December; this species is mapped only for the north of the country by Gatter (1997). Two groups of White- breasted Guineafowl Agelastes melea- grides were encountered in Grebo Forest. Also in Grebo, Black-collared Lovebird Agapornis swindernianus was observed at two locations; Gatter (1997) mentions that it is a very rare or extinct resident in Liberia. Gola and Grebo constituted new localities for Cassin’s Honeybird Prodotiscus insignis, whereas Yellow-footed Honeyguide Melignomon eisentrauti was recorded at North Lorma. A male Cardinal Woodpecker Dendropicos fuscescens was seen well at Voinjama on 26 November; only one previous record, from 1984 near Bawomai, is mentioned by Gatter (1997). Up to 25 Fanti Saw- wings Psalidoprocne obscura, four Lesser Striped Swallows Hirundo abyssinica, a pair of Plain-backed Pipits Anthus leucophrys and a Black-winged Oriole Oriolus nigripennis were observed at Fishtown, near Grebo, on 11-12 December; these species were not previously mapped for the area. A colony of Preuss’s Cliff Swallows Hirundo preussi numbering c. 1 00 nests was found at Voinjama; this species is said to be a rare (dry- season?) visitor by Gatter (1997). A Western Wattled Cuckoo-shrike Lobotos lobatus was seen within a mixed-species flock in Grebo. A singing Blue-shouldered Robin Chat Cossypha cyanocampter observed near Jalipo, Grebo, on 1 1 December, with another at Fishtown the next day, were not previously mapped for the area. A pair of Black-headed Rufous Warblers Bathmocercus cerviniventris was found near a small stream at Luyema, North Lorma. Up to three Western Olivaceous Warblers Hippolais {pallida) opaca were observed in detail in a clearing at Gola, on 27 November-3 December; this is a new locality for this Palearctic migrant, for which there are apparently only two previous records. In the same clearing, a pair of Short-winged Cisticolas Cisticola brachypterus was singing; this species was apparently known only from coastal and northern savannas. Black- capped Apalis Apalis nigriceps was common in Grebo, which constitutes a new locality, and five singing Nimba Flycatchers Melaenornis annamarulae were also found there. A singing Grey-throated Flycatcher Myioparus griseigularis was seen at Gola and another at Jalipo, Grebo; this species was previously recorded only from Yekapa/Nimba. Lead- coloured Flycatcher M. plumbeus was found singing at Luyema and in a nearby clearing, North Lorma, on 19 November, and in a clearing in Gola on 27 November-3 December; the only records mentioned by Gatter (1997) are two collected in 1891 near Monrovia. In Gola, a pair of Bioko Recent Reports Bull ABC Vol 13 No 1 (2006) -103 Batises Batis poensis was observed at their nest, which contained two feathered nestlings on 3 December; this is a new locality for this species, of which very few nests have ever been found. A large rock with 20 nests of Yellow-headed Picathartes Picathartes gymnocephalus was located inside Grebo Forest. A single and a pair of Lagden’s Bush-shrikes Malaconotus lagdeni were observed in mixed bird parties in Grebo. Two pairs of the Endangered Gola Malimbe Malimbus ballmanni, each with a juvenile, were foraging with mixed-species flocks in Gola Forest (RD, FM, ES&KD). Libya During a visit to Libya from 1 to 29 April 2003 the following notable records were logged. In Tripolitania, a flock of 12 Glossy Ibis Plegadis fal- cinellus flew over Gharyan at dusk on 2nd. A strong northerly passage of harriers Circus spp. was observed north of the Gebel Nafusa both at the start and the end of the month, with Montagu’s Harrier C. pygargus seeming to predominate. During a trip offshore from Bukamash to Farwa Island on 28th, seven Common Cranes Grus grus were seen wading in the sandy shallows at the western end of the island, close to the Tunisian border. A Wryneck Jynx torquilla was at Abugrin on 20th. In the Fezzan, in the south-west, a Common Kestrel Falco tinnunculus was seen at Mavo Lake on 14th. Laughing Doves Streptopelia sene- galensis were common in Ghat, Germa and at Ubari lakes on 10-1 6th. A Tristram’s Warbler Sylvia deserticola was around the camel trough in the Akakus on 1 1th. A Woodchat Shrike Lanius senator was observed in Ghat on 12th and a Masked Shrike L. nubicus in Wadi Mathkendoosh next day. Seven Fulvous Babblers Turdoides fulvus frequented the grounds of Germa Hotel on 14th. Desert Sparrows Passer simplex were nesting at Ghat campsite on 10th and were present in Akakus and at the Ubari lakes. Spanish Sparrows P. hispaniolensis were nesting at Germa camp on 1 4th and were present at Umm al Maa and Mandala lakes. In Cyrenaica, a pair of Egyptian Vultures Neophron percnopterus was seen in Wadi al-Khuff on 26th, along with Crag Martin Hirundo rupestris , Blue Tit Pams caeruleus of the sub- species cyrenaica and Common Raven Corvus corax. Common Chaffinch Fringilla coelebs africana was abundant alongside smaller num- bers of European Serin Serinus seri- nus and Common Linnet Carduelis cannabina at the ancient site of Cyrene on 22nd-24th. A European Goldfinch C. carduelis was at Qasr Libya on 22nd (SH). Madagascar An adult Sooty Gull Lams hemprichii observed at the island of Nosy Ve, south-western Madagascar, on 1 1 November 2003 constitutes the first for the country (ZXS) . A pair of Madagascar Red Owls Tyto soumag- nei was found at c. 1,000 m elevation in Montagne d’Ambre National Park, in the far north of the island, on 1 0 June (VD). Madeira Records from July 2005 at Selvagem Pequena included a Purple Heron Ardea purpurea on 8th, a pale-morph Eleonora’s Falcon Falco eleonorae on 7th, and a Sooty Tern Sterna fuscata on 7-8th, following a pair there briefly a month earlier. An adult Red- billed Tropicbird Phaethon aethereus was seen north of the Salvages on 9 th (per Birding World 18: 281). A first- Eleonora’s Falcon Falco eleonorae (Pete Leonard) summer Ring-billed Gull Larus delawarensis was in Funchal harbour on 17 July (per Birding World 18: 281). The 2005 breeding season was very encouraging for the Critically Endangered Zino’s Petrel Pterodroma madeira : six new nests were found, bringing the total number known to 75; of these, 68 were active in 2005 and 39 young fledged. The popula- tion is now estimated at 70-85 pairs (per Parque Natural da Madeira ). Mauritania A research cruise to the upwelling zone off Banc d’Arguin in July 2005 produced an estimated total of c. 10,000 Wilson’s Storm-petrels Oceanites oceanicus on 15- 17th, including feeding flocks of up to 700 birds. These numbers are unprece- dented here, although it seems likely that large congregations of this species regularly visit the area in mid- summer. Also recorded in the area during this period were c. 10,000 Black Terns Chlidonias niger , c.700 Royal Terns Sterna maxima and c.350 Cory’s Shearwaters Calonectris diomedea, all of the form borealis where this could be determined (per Birding World 18: 324; per Dutch Birding 27 ': 346). Records from November 2005 include the following. On the Banc d’Arguin, 21 Common Shelduck Tadorna tadorna were found close to Nair Island, on 27th. In Diawling National Park, two Black Storks Ciconia nigra , at least 100 White Storks C. ciconia and three Dark Chanting Goshawks Melierax metabates were seen on 24th. Also there, a perched Lizard Buzzard Kaupifalco monogrammicus was dis- covered on 29th; this is c.7 0 km north of the Senegal border and some 400 km north of its usual range. A Brown Snake Eagle Circaetus cinereus was in the Marigot Three area on 30th. Four Desert Sparrows Passer simplex were observed at km 253 on the Nouadhibou-Nouakchott road on 25th and four more on the Ten- Alloul track, the turning from km 253, on the 28th, with a pair nest building in an Acacia tree ( RCr ). 104 - Bull ABC Vol 13 No 1 (2006) Recent Reports Morocco The first Greater Yellowlegs Tringa melanoleuca for Morocco was at Oued Massa on 16 November. Morocco’s second and third Franklin’s Gulls Larus pipixcan were a first-winter at Oued Massa on 3 November and an adult at Oued Souss, Agadir, on 4 November. The third Yellow-browed Warbler Phylloscopus inornatus was observed north-east of Agadir on 3 November (per Dutch Birding 27 : 408-421; per Birding World 18: 437). A. Light-mantled Sooty Albatross Phoebetria palpebrata (Claudia Donati) Mozambique A Light-mantled Sooty Albatross Phoebetria palpebrata was found exhausted on the beach at Ponto Malangane in late June 2005. It was taken into care in South Africa and released off Cape Town when it had recovered (BF t 1 al. per TH). A flock of 30 Greater Frigatebirds Fregata minor with one Lesser Frigatebird F ariel was at Pomene on 1 9-20 September (AV). Another Lesser Frigatebird was photographed at Ponta D’Oura on 13 January 2006 (EEk). Two pairs of Locust Finches Paludipasser locustella were seen in grasslands between Beira and Savane on 13 November 2005 (BP). Namibia In April-December 2005 the follow- ing were reported. A Little Penguin Eudyptula minor was photographed ashore on Ichaboe Island in mid- April 2005 (. NU & TDe per PRy ). This puzzling record is the first of this Australasian species for Africa and raises the question how the bird got here; the bird was captured and examined but showed no signs of prior captivity. Two Wandering Albatrosses Diomedea exulans were spotted c.60 nautical miles west of Walvis Bay on 22 July, with a Grey- headed Albatross Thalassarche chrysostoma in the same area next day (. BR ). A European Honey Buzzard Pernis apivorus was seen at Kalizo Lodge, east of Katima Mulilo, in the extreme north-east, on 6 December (DR). Single Booted Eagles Hieraaetus pennatus were at Walvis Bay Sewage works on 22 October (TO) and 1 November (BM & FT). Eurasian Oystercatchers Haematopus ostralegus were seen at Walvis Bay on 22 October 2005 (TO) and 1 November (BM & TT). Also at Walvis Bay, an American Golden Plover Pluvialis dominica was found on 22 October (TO). Single Common Redshanks Tringa totanus were seen throughout the period at Walvis Bay and/or Swakopmund (MB, BM, TO & TT; KW per TH). A Terek Sandpiper Xenus cinereus was at Walvis Bay on 30 October (MB, TO, TT). Five Red-necked Phalaropes Phalaropus lobatus were at Walvis Bay on 3 June (AW per TH), with four also there on 14 November, and one at Mile 4, Swakopmund, on 4-5 November (BM & TT). Namibia’s first Lesser Crested Tern Sterna bengalensis was found at Mile 4 Salt Works, Swakopmund, on 5 May (MB)-, it was last reported on 1 1 November (MB, TT). Also at Swakopmund, 20,000+ Common Terns S. hirundo were counted on 14 November, c.2,000 Black Terns Chlidonias niger on 24 October, with 4,000+ there on 14 November, and one White-winged Tern Ch. leu- copterus on 5-6 and 1 3 November, with 40+ there on 14 November (BM & TT). A Grey Wagtail Motacilla cinerea was found in Halali Camp, Etosha National Park, on 19 October; it remained there for some time and was photographed on 2 November (PLw). Also in Etosha, an Olive-tree Warbler Hippolais olivetorum was dis- covered near Namutoni on 24 November (DR). Shelley’s Sunbird Cinnyris shelleyi was regularly observed and also photographed at Kalizo Lodge, near Katima Mulilo, in September-December (EE, ED, DH-, per TH). Niger The following species, for which there are few published records in Niger, were observed in the south- east in September-October 2005. A Black Stork Ciconia nigra was c. 20 km north of Diffa on 4 October. A Lesser Spotted Eagle Aquila pomari- na flew east of Maine-Soroa on 7 October. An African Hobby Falco cuvierii was chasing grasshoppers near Maine-Soroa on 24 September. A White-rumped Swift Apus caffer was noted north of Diffa on 27 September and another, with Little Swifts A. ajfinis, at Diffa on 6 October; these are the first records from eastern Niger, all previous records being from the south-west. A pair of River Prinias Prinia fluviatilis was nest building at the Komadougou Yobe River, Diffa, on 23 and 29 September (Fig. 5), and another was by the River Niger at Tillabery on 1 1 October; these would constitute the first definite records for the country, if accepted. A Yellow-spotted Petronia Petronia pyr- gita was found west of Diffa on 26 September (FPJ & JFR). Following the first observations of Mottled Swift Tachymarptis aequato- rialis and Alpine Swift T. melba in 2004 (see elsewhere in this issue), Mottled and possibly also Alpine Swifts were seen on 25 October 2005 just east of Arlit, in a mixed flock with Common Swifts Apus apus. Pallid Swifts A. pallidus and a Little Swift A. affinis (WM & RQ. A pho- tograph of a possible African Black Swift Apus barbatus in the same flock is still being examined. The latter species has not yet been confirmed for Niger, although there are more than ten probable observations dur- ing the rainy season, including possi- ble breeding in caves in the Dallol Bosso, 100 km east of Niamey (JB). Recent Reports Bull ABC Vol 13 No 1 (2006) - 105 Senegal In October 2005, an American Golden Plover Pluvialis dominica was present at Technopole, Dakar, on 29th (WF; Fig. 6). In December 2005, two more Nearctic vagrants were found: an American Wigeon Anas americana at Djoudj National Park on 20th, probably a young male (Fig. 7), and a Laughing Gull Larus atricilla at Sine Saloum on 28th (AF; Fig. 8). Seychelles Reports from May-December 2005 received by Seychelles Bird Records Committee (SBRC) include the first Sooty Gull Larus hempricbii for Seychelles, an adult in breeding plumage at Aride Island on 29 August. Nine Flesh-footed Shearwaters Puffinus carneipes were reported from five locations, on 25-27 October, off the north and north-east edge of the Seychelles Plateau (north of Bird and Denis islands); four previous records have been accepted by SBRC to date. The fifth Wilson’s Storm-petrel Oceanites oceanicus for the archipelago was also reported in the same area on 27 October. A Red-billed Tropicbird Phaethon aethereus stayed at Bird Island from 6 May until 28 August. The fifth Great Bittern Botaurus stellaris for Seychelles was also found there on 5-13 November, as well as the third Madagascar Pond Heron Ardeola idae east of the Aldabra group (where it breeds on Aldabra Atoll) on 14 September. A Great White Egret Egretta alba was found at an unnamed islet off Benjamin, St Joseph Atoll, on 21 July, and a Purple Heron Ardea purpurea at North Point, Mahe, on 23 October. A White-faced Whistling Duck Dendrocygna viduata was at Cinq Cases beach, Aldabra, on 20 August. Eleven Amur Falcons Falco amurensis were on North Island on 30 November, with four still present on 23 December. A Eurasian Hobby F. subbuteo was seen at Mahe on 8 November and an Allen’s Gallinule Porphyrio alleni at Picard, Aldabra, on 6 June. The second Stone-curlew Burhinus oedicnemus for Seychelles Madagascar Pond Heron Ardeola idae (Pete Leonard) was at Bird Island on 1 November (interestingly, the first record, now accepted by SBRC, was at this same location the previous season), with the fifth Collared Pratincole Glareola pratincola also there on 29 October, and two from 3 1 October to 23 December. A Grey-tailed Tattler Heteroscelus brevipes at Glacis, Mahe, on 30 September, was the third for Seychelles. A European Turtle Dove Streptopelia turtur stayed at Bird Island on 4-9 November and a European Roller Coracias garrulus at Cousine on 13-29 November. The third Greater Short-toed Lark Calandrella brachydactyla remained at Bird Island from mid October to 12 November. A Mascarene Martin Phedina borbonica was at L’Union Estate, La Digue, on 4 July. Single White Wagtails Motacilla alba were observed at Bird Island on 5 May and at Cousine Island on 24 November. A Red: throated Pipit Anthus cervinus was at Bird Island on 8-12 November. A Common Redstart Phoenicurus phoenicurus was ship-assisted from east of La Digue on 29 October to La Digue on 30 October, and anoth- er was at Bird Island on 3-5 November. An Isabelline Wheatear Oenanthe isabellina on North Island on 12 October constitutes the fourth report for Seychelles. Second reports were received for Sedge Warbler Acrocephalus schoenobaenus at Bel Air, Mahe, on 9 November, and Icterine Warbler Hippolais icterina at Bird Island on 1 0 November. Also at Bird Island were a Spotted Flycatcher Muscicapa striata on 7-9 November and a European Golden Oriole Oriolus oriolus on 5-9 November ( all per AS). Sierra Leone During a bird census carried out in Sierra Leone’s coastal wetlands in January-February 2005, four species were found that had apparently not been reported previously: five Great White Pelicans Pelecanus onocrotalus were counted at the Searcies estuary and 253 at Yawri Bay, 12 Eurasian Spoonbills Platalea leucorodia at Searcies, 18 Northern Shovelers Anas clypeata at Yawri, and a Terek Sandpiper Xenus cinereus , also at Yawri (www.projects.wiwo.org/). Among the more interesting records made during a birding trip to this little-visited country in the sec- ond half of November 2005 were a Least Honeyguide Indicator exilis , claimed from Tiwai Island, two Yellow-throated Cuckoos Chrysococcyx flavigularis in Gola Forest and, intriguingly, Chattering Cisticola Cisticola anonymus in a flooded area south of Gola; the latter would confirm the existence of an isolated Upper Guinea population of this species (AH). South Africa Records from April-December 2005 include the following. On pelagic trips out of Cape Town, at least 27 Wandering Albatrosses Diomedea exulans were recorded in August-November, ten Southern Royal Albatrosses D. (e.) epomophora between 30 July and 10 September, and 1 3 Northern Royal Albatrosses D. (e.) sanfordi between 28 May and early October (AG, JGr ). A juvenile Grey-headed Albatross Thalassarche chrysostoma was seen c.60 nautical miles west- northwest of Cape Columbine, Western Cape, on 3-4 August (B W), and an adult off Cape Point in mid September (TH). The Light-mantled Sooty Albatross Phoebetria palpebrata picked up in Mozambique in late June (see above) 106 - Bull ABC Vol 13 No 1 (2006) Recent Reports was taken out to sea and released off Cape Town when it had recovered (per TH, CD). A very late Great- winged Petrel Pterodroma macroptera was observed on 1 1 June {JGr ). Single Spectacled Petrels Procellaria (aequinoctialis) conspicillata were seen in mid September (two), in early October (per TH) and on 27 December (RW). A Grey Petrel P. cinerea was found on 3 September ( AG) and 5 November (JGr); this species is very rare in southern Africa. An unexpected Cory’s Shearwater Calonectris diomedea was seen on 1 7 June. A Manx Shearwater Puffinus puffinus on 21 December was the first for quite some time (RW). A Little Shearwater P. assimilis was found rather close to shore on 7 May and another in mid September. Single Grey Phalaropes Phalaropus fulicarius were spotted in early October and on 3 November (JGr). The Bulwer’s Petrel Bulweria bul- werii seen at 35°24’S 1 1°08’E on 1 1 December 2004 (Bull. ABC 12: 71) is not the most southerly record yet, as suggested; the southernmost was one seen 3 km south-east of Cape Nelson, Victoria, Australia, at 38°31’S 141°34’E on 14 September 1986 (cf. Handbook of Australian, New Zealand and Antarctic Birds 1 : 336) (SH&NQ. A King Penguin Aptenodytes patagonicus found at Camps Bay beach, outside Cape Town, on 1 1 August, was taken into care but died the next day; there are few records of this species from southern Africa (per TH). A White-tailed Tropicbird Phaethon lepturus reportedly flew over Plettenberg Bay, Eastern Cape, on 7 October (RG). An Australian Gannet Sula senator , discovered in the Cape Gannet S. capensis colony on Bird Island, Western Cape, on 7 August (TH, MG, DN, GK), was present until at least 11 September (per TH). A female Greater Frigatebird Fregata minor was at Pennington Beach, on the KwaZulu-Natal south coast, on 14 July (IRa) . The two Slaty Egrets Egretta vina- ceigula first reported from Marievale Bird Sanctuary, Gauteng, on 23 January, were still present on 1 5 May, with one still there on 1 September (RM per TH), and up to three were at this site on 9 October (per TH) . Another was at Grootvaly Wetland Reserve, Gauteng, on 26 July (SM) and remained until at least 21 August (per TH). One was at Ndumo Game Reserve, KwaZulu-Natal, on 2 1 October (TM). Southern Africa’s third Little Blue Heron E. caerulea was still at the River Olifants estuary, north of Lambert’s Bay, Western Cape, on 5 September (per TH); this bird has now been present in this area four years. During a pelagic trip out of Simonstown on 22 October, a flock of at least 150 White Storks Ciconia ciconia was seen c.45 km south of Cape Point (JGr). A European Honey Buzzard Pernis apivorus flew over Pietermaritzburg, KwaZulu-Natal, on 8 September (MBr); probably the same bird was also seen on 12th (per TH). Another flew over Wierda Park, Gauteng, on 14 October (PW). An adult Egyptian Vulture Neophron percnopterus was with a group of Cape Vultures G. coprotheres attending a carcass along- side the road to Coffee Bay in the Transkei on 1 November (PC per TH). A Ruppell’s Griffon Vulture Gyps rueppellii was seen at Crook’s Corner near Pafuri, in the north of Kruger National Park, on 1 3 August (AO); the long-staying bird at the Cape Vulture colony in Blouberg Nature Reserve was still present on 24 September (per TH). A Striped Crake Aenigmatolimnas marginalis was reported from the Lake Panic bird hide near Skukuza, Kruger National Park, on 2 July (KJ). An American Purple Gallinule Porphyrio martinica was seen wandering along the N3 just outside Johannesburg on 3 May (RM per TH). The regularly returning Eurasian Oystercatcher Haematopus ostralegus at Elands Bay, Western Cape, was again present on 6 September (VW); another was at St Lucia, KwaZulu- Natal, on 10 September (PLa), two were at the Umfolozi River mouth, KwaZulu-Natal, on 13 September (AHa,JS), and three on the beach at Cape Point, Western Cape, on 6 November (GKi). Southern Africa’s 13th White-rumped Sandpiper Calidris fuscicollis was found at Marievale Bird Sanctuary, Gauteng, on 17 September (RM & DD); it was still present on 13 October (per TH). A Pectoral Sandpiper C. melanotos was at Twee Rivieren in Kgalagadi Transfrontier Park, Northern Cape, on 18 August (BV), and one was at Vaalkop Dam on 1 October (RM). The two Black-tailed Godwits Limosa limosa found at De Plaat on the Berg Rivier, Western Cape, on 22 March remained until the end of August at least (per TH), with one still there on 16 October (VW). Three were seen near Welkom, Free State, on 20 May (BC & JL per TH), with one there on 1 October (GR). A flock of 28 at Spitskop Dam, near Kimberley, Northern Cape, on 1 1 August is an interesting record, both for the number of birds and for the date (MA). One was at Marievale Bird Sanctuary, Gauteng, on 8 October (DD). A Common Redshank Tringa totanus was at Malandeni, Ladysmith, KwaZulu- Natal, on 20 August (KG) and anoth- er at Marievale Bird Sanctuary, Gauteng, from 24 September (RM & DD) until at least 19 October (per TH). In Western Cape, a Red-necked Phalarope Phalaropus lobatus was at Still Bay Bird Sanctuary from 9 December until at least 13th (HE), and eight were at their usual site at Velddrift on 28 December (PC). A Yellow-billed (Greater) Sheathbill Chionis alba was in Cape Town harbour on 24 June, with a second individual there three days later; both remained until at least 1 July (EFo) and one was still present on 13 September, with the other being relocated at Kommetjie on 13 August (CH & NH), where it was last seen the next day (per TH). Records of Franklin’s Gulls Larus pip- ixcan include one at the Umgeni River mouth, Durban, KwaZulu- Natal, on 6 June (MO), one just off Dyer Island, near Gansbaai, Western Cape, on 21 June (RCo), one in full breeding plumage at Lambert’s Bay, Western Cape, on 2 August (RMu) and still present on 6 September (per TH), and one at St Helena Bay, Recent Reports Bull ABC Vol 13 No 1 (2006) - 107 Western Cape, on 13 September ( FG ). A Common Black-headed Gull L. ridibundus was at Durban Bay, KwaZulu-Natal, on 27 April (per TH) with one in breeding plumage on 19 October {DA & AMc). One returned to Driftsands Reclamation Works in Port Elizabeth, Eastern Cape, on 11 November (AT); this bird has now spent the past few seasons there (per TH). A Heuglin’s Gull L. { fuscus ) heuglini , first located in Durban Bay, KwaZulu-Natal on 26 April, remained in the area until at least 28 May {AMc per TH). The long-staying Gull-billed Tern Gelochelidon nilotica at Kromme Rivier estuary, St Francis Bay, Eastern Cape, first reported on 25 February, remained there until 29 May at least ( JBr per TH). Perhaps the same indi- vidual was reported at Cape Recife, Port Elizabeth, Eastern Cape, on 3 September {CL). A Lesser Crested Tern Sterna bengalensis was claimed from Tsitsikamma lagoon, Eastern Cape, on 29 October {RA per CQ. The Bridled Tern S. anaethetus that returned for its fifth successive year to Cape Recife, Port Elizabeth, Eastern Cape, on 27 May, was last reported on 20 August (per TH) . An African Skimmer Rynchops flavirostris was still present at Roodekoppies Dam, Gauteng, on 17 July {RAT). On 16 December, a Garden Warbler Sylvia borin was mist-netted at Tygerberg Nature Reserve, Western Cape {PN). The first Rose-coloured Starling Sturnus roseus for southern Africa was photographed in Kalahari Gemsbok National Park / Kgalagadi Transfrontier Park, Northern Cape, on 15 July {JSa ). Although this species is known to wander widely, the only previous record south of the Sahara of this Palearctic species is of one seen in southern Ethiopia on 23 March 2005 (see elsewhere in this issue) and the possibility of an escape needs to be eliminated before the present record can be accepted. On 27 August, a Black-eared Seedeater Serinus mennelli was found near Punda Maria, in the north of Kruger National Park {AB); this apparently constitutes the first record for the park, with only a few others claimed in the country. Sudan During a visit to Rumbek, a small town surrounded by wooded savanna and a few cultivated areas in southern Sudan (06°50’N 29°42°E), on 21-28 October 2005, several species were recorded whose occurrence is not indicated in the relevant one- degree square of their distribution map in Nikolaus (1987, Distribution Atlas of Sudan’s Birds). These include the following: Little Sparrowhawk Accipiter minullus (two on 27th), Wahlberg’s Eagle Aquila wahlbergi (one on 26th), Tawny Eagle A. rapax (one on 27th), Booted Eagle Hieraaetus pennatus (a pale morph on 28th), Grey Kestrel Falco ardosiaceus (one on 22nd; two togeth- er on 27th), Bruce’s Green Pigeon Treron waalia (singles on 22nd-25th), African Mourning Dove Streptopelia decipiens (regularly 2-3 in town), Red-eyed Dove S. semitorquata (one on 22nd), Rose- ringed Parakeet Psittacula krameri (seen daily, up to five togeth- er), Woodland Kingfisher Halcyon senegalensis (one on 26th), European Roller Coracias garrulus (singles on 22nd and 25th), Black-and-white- casqued Hornbill Bycanistes subcylin- dricus (two on 21st), Yellow-fronted Tinkerbird Pogoniulus chrysoconus (singles on 22nd and 26th), Common Bulbul Pycnonotus barbatus (seen daily in small num- bers), Spotted Palm Thrush Cichladus a guttata (one on 24-25th), Eastern Olivaceous Warbler Hippolais pallida (two on 23rd), Common Fiscal Lanius collaris (one on 24th), Black-headed Gonolek Laniarius erythrogaster (one on 22nd), Brubru Nilaus afer (one singing on 22nd; two immatures on 25 th), Speckle-fronted Weaver Sporopipes frontalis (5-10 on 22nd; singles on 24-26th), Northern Masked Weaver Ploceus tae- niopterus (a male on 22nd) and Vitelline Masked Weaver P. velatus (at least three near the Barnaam (or Naam) River, c.20 km south-east of Rumbek, on 26th) {BP). Tanzania Records from 2005 include the fol- lowing. A small drying pond near the Speke Bay Lodge turn-off had a female Striped Crake Aenigmatolimnas marginalis on 4 August {AK). An African Finfoot Podica senegalensis was found on the stream between Kisima Ngeda Camp and Lake Eyasi in August {KM). At least 20 Broad-billed Sandpipers Limicola falcinellus were at the lake in the Ngorongoro crater on 1 1-12 July {AD). A juvenile Great Spotted Cuckoo Clamator glandarius begging aggressively for food from its host, a Superb Starling Lamprotornis superbus, was photographed on 10 July along the road from Ndutu Lodge to the main road through the Serengeti {GG per KM). Exceptionally large concentrations of Thrush Nightingales Luscinia lus- cinia were reported in parts of south- ern Tanzania in December 2005-January 2006, as well as large numbers of Grasshopper Buzzards Butastur rufipennis and Irania Irania gutturalis-, possibly, the dry conditions further north are the cause of this influx (. NBa ). Kungwe Apalis Apalis argentea was found east of Mahale National Park, western Tanzania, in almost every patch of riverine forest visited; the only two previous records away from Mahale are two specimens collected in the late 1930s {DM per NBa). A sighting of Bertram’s Weaver Ploceus bertrandi at Irente Farm, Lushoto, in the West Usambaras, in early September, is a significant record; this montane forest-edge species survives in degrad- ed habitat at low density {JD per NBa). Togo White-browed Forest Flycatcher Fraseria cinerascens , observed in Fazao-Malfakassa National Park on 1 1 November 2005, constitutes an addition to the Togo list {PR). Tunisia Records from April-June 2005 include the following. An adult Yellow-billed Stork Mycteria ibis was with Eurasian Spoonbills Platalea leu- 108 - Bull ABC Vol 13 No 1 (2006) Recent Reports corodia at Korba Lagoons, Cap Bon, from 27 June until at least 10 July. Five family parties of Marbled Ducks Marmaronetta angustirostris were also there on 27 June (per Birding World 18: 281; per Dutch Birding 27: 346). Several Steppe Eagles Aquila nipalen- sis, including a group of five, were reported from Cap Bon in April-May (per Dutch Birding 27: 277). Several Egyptian Nightjars Caprimulgus aegyptius were in song at Ghidma, south of Douz, in early May ( AvdB ). In June, up to 14 Cream-coloured Coursers Cursorius cursor and c. 30 Desert Sparrows Passer simplex were observed at Bir Soultane on 24th, two male and one female Egyptian Nightjar Caprimulgus aegyptius at Ghidma, with another male at El Matrouha, on 23th, and four Levaillant’s Woodpeckers Picus vaillantii at Ain Draham on 23rd (per Birding World 18: 281; per Dutch BirdingT]'. 346). On 9 October 2005, the Korba and Soliman Lagoons held 4,300 Greater Flamingos Phoenicopterus {ruber) roseus, including 87 ringed, and also Marbled Ducks, Ferruginous Ducks Aythya nyroca and Purple Swamphens Porphyrio porphyrio {HA). Uganda For July-August 2005 the following records were received. A Broad-billed Sandpiper Limicola falcinellus was found at Lake Katwe, Queen Elizabeth National Park, on 29 August; there are few records for this Palearctic migrant in Uganda. The previous day, a Ruddy Turnstone Arenaria interpres was seen on Kazinga Channel in the park. A pair of Green-breasted Pittas Pitta reichenowi was nest building in Kibale National Park on 22 July; what was almost certainly the same pair was seen again on 7 August. At Mubwindi Swamp, Bwindi Impenetrable National Park, a pair of Green Broadbills Pseudocalyptomena graueri was observed at its nest on 1 5 July, and a pair with a juvenile was seen on 14 August {DH). Two Fischer’s Lovebirds Agapornis fischeri were seen in Entebbe’s Botanical Garden on 14 January 2006; they had apparently been recorded before at this site and would constitute an addition to the Ugandan list, if accepted as wild birds (JK). Zimbabwe Contrary to the statement in a previ- ous Recent Reports {Bull. ABC 12: 191), the Basra Reed Warbler Acrocephalus griseldis claimed from Victoria Falls on 2 January 2005 would constitute the first (not the second) record for Zimbabwe, if accepted. There have been two sub- missions so far, from 2 March 1998, at Victoria Falls, and 4 November 1998, at Chikwenya, but neither has been accepted {IR). Records were collated by Ron Demey from contributions supplied by David Allan (DA), Mark Anderson (MA), Ray Archer (RA), Hichem Azafzaf (HA), Neil Baker (NBa), Clive Barlow ( CB), Ruben Barone (RB), Errol de Beer (EdB), Hannelore Bendsen (HB), Arnoud van den Berg (AvdB), Mark Boorman (MB), Nik Borrow (NB), Andre Botha (AB), Nicky Bousfield (NBo), John Bradshaw (JBr), Chris Brewster ( CBr), Joost Brouwer (JB), Mark Brown (MBr), Robert Cheke (RC), Neil Cheshire (NC), Philip Coetzee (PC), Callan Cohen ( CC), Brian Colahan (BC), Robert Cope (RCo), Richard Cruse (RCr), Alexis DallAsta (AD), Dave Deighton (DD), Tony Delport (TDe), Eckart Demasius (ED) , Ron Demey (RD), Klaas-Douwe Dijkstra (KD), Vladimir Dinets (VD), John Dixon (JD), Tim Dodman (TD), Cliff Dorse ( CD), Eric Ehlers (EE), Ehren Eksteen (EEk), Wouter Favyets (WE), Berrie Ferreira (BF), Pete Ferrera (PE), Simon Feys (SF), Hamish Fletcher (HE), Amine Flitti (AF), Erik Forsyth / Rockjumper Birding Tours (EE) , Eugene Fourie (EFo), Ursula Franke (UF), George Gerdts ( GG), Margaret Gibbs (MG), Ray Goodwin (RG), Jan Goossens (JG), Ken Gordon (KG), John Graham (JGr), Anne Gray (AG), Malcom Greene (MGr), Felicity Grundlingh (FG), Adrian Haagner (AHa), Pete Hancock (PH), Trevor Hardaker (TH), Richard Hearn (RH), Andrew Hester (AH), Carol Hewitt ( CH), Nigel Hewitt (NH), Sveinung Hobberstad (SHo), David Hoddinott / Rockjumper Birding Tours (DH), Stan Howe (SH), Colin Jackson (CJ), Flemming Pagh Jensen (FPJ), Kevin Joliffe (KJ), Johnnie Kamugisha (JK), Chege Kariuki ( CK), Geoff Kieswetter ( GKi), Alastair Kilpin (AK), Gordon King (GK), Peter Lack (PL), Phil Langston (PLa), Peter Lawson (PLw), Janine Lieffrig (JL), Hans Linde (HL), Derek Lister (DL), Hans van der List (HvdL), Chris Lotz (CL), Stan Madden (SM), Graham McCulloch ( GM), Alistair McLnnes (AMc), Kevin Mlay (KM), Flomo Molubah (FM), Richard Montinaro (RM), Patrick Morton (PM), David Moyer (DM), Zee Mpofu (ZM), Temba Mthembu (TM), Ben Mugambi (BM), Mark Muller (MM), Wim Mullie (WM), Ray Murray (RMu), Andy Musgrove (AM), Doug Newman (DN), Peter Nupen (PN), Anton Odendal (AO), Michael ODonaghue (MO), Tim Osborne (TO), Craig Pearman ( CP), Bram Piot (BP), Paul Radley (PR), Hugo Rainey (HR), Lan Ralph (LRa), Jan Fischer Rasmussen (JFR), Gert Rautenbach ( GR), Lan Riddell (LR), Diane Ridgley (DR), DetlefRobel (DRo), Barrie Rose (BR), Peter Ryan (PRy), Jorgen Sagvik (JSa), Peter Samuels (PS), David Shackelford (DS), James Sibiya (JS), Adrian Skerrett (AS), Neville Skinner (NS), Evangeline Swope (ES), Alf Taylor (AT), Keith Temple (KT), Tony Tree (TT), Steph Tyler (ST), Neville-Nash Uhongora (NU) , Brian Vanderwalt (BV), Andrew Viljoen (AV), Nigel Voaden (NV) , Jaap van der Waarde (JvdW), Ross Wanless (RW), Vincent Ward (VW), Barry Watkins (BW), Keith Wearne (KW), Peter Wilgenbus (PW), Edwin Winkel (EW) and from Birding World, capebirdnet, Dutch Birding, Parque Natural da Madeira, SARareBirdAlert, www.projects.wiwo.org/ and www.zestforbirds.co.za. Contributions for Recent Reports can be sent to Ron Demey, Van der Heimstraat 52, 2582 SB Den Haag, Netherlands and also by e-mail: rondemey@compuserve. com. Recent Reports Bull ABC Vol 13 No 1 (2006) -109 Reviews Oiseaux de Tunisie / Birds of Tunisia Paul Isenmann, Thierry Gaultier, Ali El Hill, Hichem Azafzaf, Habib Dlensi & Michael Smart. 2005. Paris: Societe d’ Etudes Ornithologiques de France (SEOF), e-mail: seof@mnhn.fr, website www. mnhn. fr/assoc/seof/accueil.htm. Paperback, 432 pp, 432 pages, 130 plates, 150 maps. ISBN 2-9506548-9-4. €38. Tunisia is a relatively small country (164,500 km2) in north-west Africa. The climate is typically Mediterranean in the north, becom- ing more arid toward the south and Saharan in the southern half. A wide range of habitats is present: cork oak forests, mountain ranges (highest peak 1,544 m), steppe plains, an immense depression with salt lakes, 1,300 km of Mediterranean coastline with steep cliffs, coastal lagoons and a considerable area of intertidal flats, oases and impressive sand dunes. This variety in habitats, the rich avi- fauna, good infrastructure and touristic facilities (including cheap flights) make the country a popular destination for visiting birders. In addition there is a growing local birdwatching and conservation com- munity. The three Tunisian co- authors are members (El Hili also founder and President) of ‘Les Amis des Oiseaux, a succesful organisation established in 1975. Ail 46 Important Bird Areas (IBAs) have been protected since 2001! Following the synthesis of the birds of north-west Africa (Les Oiseaux du Nord-Ouest de TAfrique), which included Tunisia, written by H. Heim de Balsac and N. Mayaud in 1962, the only recent attempt to produce a modern annotated list of the country was The Birds of Tunisia by Peter Thomsen and Peder Jacobsen (Copenhagen, 1979). Although the latter book was rather incomplete, based on limited sources and is now out of print, it was my main source of information during a few visits to Tunisia. However, dur- ing my next visits I will definitely be carrying another book. Carrying in the most literal sense of the word, since the new Birds of Tunisia has become an impressive 432-page vol- ume on good-quality paper. The size is due to the fact that the book is entirely bilingual (English and French). Short, informative introductory chapters cover geography, history of Tunisian ornithology, a compact checklist of all 395 species recorded (of which 193 breed), biogeographi- cal analysis, and short comparisons with the avifauna of Algeria and Libya, as well as describing changes to the breeding avifauna in the 20th century, and the Mediterranean and trans-Saharan migration systems. The main part of the book is formed by the annotated checklist, sum- marising available information on taxonomy, status, distribution, habi- tat and phenology of each species, together with data on nesting and ringing recoveries. The information presented is rather anecdotal, with many dates, localities and observers (mostly visiting birders) mentioned. The 150 distribution maps, given for most of the regular breeding species show the ‘potential breeding range’. Knowing that one of the authors (Gaultier) has been working on a breeding atlas based on half-degree squares, I would have preffered to see (in addition) the actual known breeding distribution by using sym- bols for possible, probable and defi- nite breeding. English species accounts within the same light purple shading as used for the English versions of the introductory chapters would have made reading easier. The book is lav- ishly illustrated with over 130 high- quality colour photographs (many by Gaultier and Azafzaf). Dates and localities are not mentioned, and a photo captioned Botaurus stellaris shows an immature Little Bittern Ixobrychus minutus. Nonetheless, despite these few minor remarks, it is an excellent book, and the authors are to be congratulated on their achievement. Peter L. Meininger Waterbird Monitoring in the Bijagos Archipelago, Guinea- Bissau / Monitorizagao de Aves Aquaticas no Arquipelago dos Bijagos, Guine-Bissau Tim Dodman and Joaozinho Sa. 2005. Dakar: Wetlands International & Bissau: Gabinete de Planificapao Costeira / Organizagao para a Defense e o Desenvolvimento das Zonas Humidas na Guine-Bissau. 157 pp, line drawings, maps. Softback. Distributed by NHBS. Price unknown. The Bijagos Archipelago is the second-most important site in West Africa for migratory waterbirds using the East Atlantic Flyway, after Mauritania’s Banc d’Arguin, and reg- ularly supports 600,000-900,000 waders. Its relative importance in the subregion may even be increasing, at least for the most numerous species, perhaps due to changing conditions at other West African key sites. It also supports significant colonies of resident waterbirds, notably herons, gulls and terns. This bilingual, English / Portuguese, publication presents an overview of the impor- tance of Guinea-Bissau’s coastal zone for waterbirds and analyses the data of waterbird surveys conducted between the 1980s and 2001. The 1 10 - Bull ABC Vol 13 No 1 (2006) Reviews most recent survey, undertaken in January-February 200 1 , produced an estimate of 871,750 migratory water- birds, including 505,000 Curlew Sandpipers Calidris ferruginea, 133.000 Red Knot C. canutus , 97.000 Bar-tailed Godwits Limosa lapponica, 24,500 Little Stints Calidris minuta, 28,000 Common Redshanks Tringa totanus, 23,500 Grey Plovers Pluvialis squatarola and 13.000 Whimbrels Numenius phaeo- pus. The publication also provides recommendations for future surveys and monitoring in the Bijagos, and includes a training manual for the execution of wetland surveys and waterbird monitoring programmes. Ron Demey Soutpansberg-Limpopo Birding Sarah Venter. Third edn. 2005. 40 pp, colour photographs, map. Softback. ISBN 0-620-33623-4. Distributed by Soutpansberg-Limpopo Birding Route non-profit organisation. E-mail: contactus@limpopobirding.com. R25 plus postage. The stated aim of this colourful and informative booklet is to introduce the Soutpansberg and Limpopo Valley areas to the birding world, and to encourage ecotourism, stimulate rural economic development and help create awareness of the need for bird conservation. The region covered lies in north-east South Africa, on the border with Botswana and Zimbabwe (the Limpopo River forming the bor- der), with the northern part of Kruger National Park in the east. This relatively little-known area, which has some spectacular scenery, boasts over 540 bird species, includ- ing 29 southern African endemics and 32 near-endemics. The booklet contains general information about the region and suggestions on self- drive routes, information on tour operators, a brief presentation of 40 birding sites and an annotated birdlist. Trip repoits and site lists can be accessed on the Soutpansberg- Limpopo Birding Route website: www.limpopobirding.com. Ron Demey Reviews Bull ABC Vol 13 No 1 (2006) -111 01384 373013 The very best of Worldwide Birding 76 tours - our African venues include: Botswana Gambia Ghana Kenya Namibia South Africa Uganda Friendly Professional Leaders Small Groups Worldwide Birding ... with & 49 Sandy Road • Norton • Stourbridge • DY8 3AJ www.avianadventures.co.uk avianadventures@btinternet.com The ABC welcomes original contributions on all aspects of the birds of Africa, here defined as the area covered by Collar, N.J. and Stuart, S.N. 1985. Threatened Birds of Africa and Related Islands: The ICBP/IUCN Red Data Book. Cambridge, UK: International Council for Bird Preservation, namely continental Africa, Indian Ocean islands west of 80"E, e.g. Madagascar, the Mascarene Islands and Socotra; Atlantic Ocean islands on or east of the mid-Atlantic ridge, e.g. the Tristan da Cunha group, the Azores and the Canaries. Contributions will be accepted subject to editing and refereeing by independent reviewers, where appropriate. The Editorial Team will be happy to advise authors on the acceptability of material at draft stage if desired. Submissions Two hard (printed) copies should be sent unless submitting by e-mail (preferred) to the editor’s address on the inside front cover. Typewritten manuscripts should be double- spaced, on one side of the paper only, with wide margins all round. All submissions are acknowledged. Contributions are accepted in English or French: French summaries are required for all Notes for Contributors papers published in English, and vice versa. Those submitting papers should supply a summary for translation into English, or French, as appropriate. If you submit your contribution on CD or floppy disk, please state computer (e.g. IBM compatible PC, Macintosh) and word- processing package (e.g. Word, WordPerfect) used. When sending your contribution on disk, please do not key anything in ALL CAPS (i.e. with the CAPS LOCK key depressed) unless the combination always occurs in that form (e.g. ‘USA’). Do not use the carriage return key at the end of lines, and do not right justify the margins. When for- matting tables use one tab, and not spaces, between each column. Unless a sketch map is provided as part of the article, the names of places should follow those on standard or readily available maps (preferably a recent edi- tion of The Times Atlas of the World). Preferred names Given the current instability over worldwide lists of bird names, authors are requested to follow those used in The Birds of Africa Vols. 1-7. The African Bird Club has recently pub- lished (www.africanbirdclub.org/resources/ checklist.html) a checklist of birds in its region. This is based on Birds of Africa but incorporates more recent revisions where appropriate. It includes preferred scientific, English and French names, as well as races and alternatives used by publications widely used in Africa. For bird names this list should be used or at least the preferred name used there should be given as an alternative. For non -Birds of Africa species (e.g. from the Malagasy region) use Dowsett & Forbes- Watson (1993). Deviation from such works should be noted and the reasons given. The Editorial Team will keep abreast of changes in nomenclature and when an agreed list of African names is available, will consider switching to follow it. Style Authors are requested to follow conventions used in The Bulletin of the African Bird Club and to refer to a recent issue for guidance. A detailed style guide can be obtained, either electronically or as a hard copy, on request from the Managing Editor. Design & t initial layout by Alcedo Publishing, Pennsylvania, USA • email engli@mailsnare.net Colour repro & print production by Crowes of Norwich, UK • tel +44 (0)1603 403-349 • email graphics@crowes.co.uk Angola: Pedro de Franca Doria vaz Pinto, Rua Helder Neto 12, 7°A; Luanda. E-mail: pvpinto@clix.pt. Australia: K. David Bishop, PO Box 1234, Armidale, NSW 2330. E-mail: kdbishop@ozemail.com.au. Austria: Graham Tebb, Graf Starhemberggasse 20/14, 1040 Vienna. E-mail: tebb@fwf.ac.at. Belgium: Jan Goossens, Vruntebaan 18, 2320 Emblem. Tel/fax: +32 3 488 13 71. E-mail: jan.goossens5@pandora.be. Botswana: Chris Brewster, PO Box 26292, Gaborone. E-mail: cbrewster@botsnet.bw. Canada: Antonio Salvadori, 17 Colborn Street, Guelph, Ontario. NIG 2M4. E-mail: rosella@snowhite.cis.uoguelph.ca. Canary Islands/Spain: Tony Clarke Cl Republica Dominicana No. 61, Barrio de Fatima, 38500 Gtiimar, Tenerife. E-mail: clark@arrakis.es. Denmark: Uffe Gjol Sorensen, Ovengaden Oven Vandet 68,2, 1415 Copenhagen. E-mail: ugs@post7.tele.dk. Democratic Republic of Congo: Byamana Robert Kizungu, Head of Ornithology Laboratory, CRSN- Lwiro, DRC, BP02 Cyangugu, Rwanda. E-mail: kbyamana@yahoo.com Egypt: Sherif & Mindy Baha El Din, 2 Abdalla El Katib St. Apt. 3, Dokki, Cairo. Tel/Fax: 3608160. E- mail: baha@internetegypt.com. France: Bob & Framboise Dowsett, Le Pouget, Sumene, F30440. E-mail: Dowsett@aol.com. Finland: Annika Forsten, Tornvalksv. 2 bst 15, 02620 Esbo. E-mail: annika.forsten@alula.fi. The Gambia: Solomon Jallow, do WABSA, Department of Parks & Wildlife, Management HQ, Abuko Nature Reserve, Abuko, PMB 676 S/K. E- mail: habitatafrica@hotmail.com. Ghana: Samuel Kofi Nyame, PO Box KIA 30284, Airport, Accra. E-mail: samknyame02@yahoo.com. Italy: Giuseppe Micali, Via Volterra 3, Milano, MI 1- 20146. E-mail: xeaym@tin.it. Kenya: Colin Jackson, PO Box 383, Watamu. E-mail: colin.jackson@bigfoot.com. Liberia: Moses A. Massah, Society for the Conservation of Nature of Liberia, Monrovia Zoo, Supported and Affiliated Membership The Supporting Members scheme is a key part of the Clubs strategy of encouraging the spread of knowledge and understanding of birds as widely as possible throughout Africa. The scheme enables Africans who would not other- wise have the resources to join, to become mem- bers of the Club. The scheme is funded by Supporting Members who pay a minimum of UK£30 to cover their own membership and the subscription of at least one African member. The money they contribute over and above their own subscription is placed in a special fund that is used to cover the membership expenses of African members whom they may have nomi- nated, or who have been nominated by other Club members. Although we have suggested a minimum of UK£30 to become a Supporting Member, any contribution is welcome. All members of the Club, even if they do not feel able to become Supporting Members themselves, are invited to nominate candidates for supported member- ships. Candidates should be nationals of an African country, with a genuine interest in wild birds but without the resources to become mem- bers in their own right. Africans who think they may qualify are very welcome to put their own ABC Representatives PO Box 2628, Monrovia. E-mail: mosesmassah@yahoo.com. Madagascar: Julien Ramanampamonjy, Section Oiseaux, PBZT, BP 4096, 101 Antananarivo. E- mail: julien_asity@mel.wanadoo.mg (mark FAO: Julien Ramanampamonjy). Malawi: Lawrence Luhanga, Malawi Ornithological Society, do Dept of Ornithology, Museums of Malawi, PO Box 30360, Chichiri, Blantgre 3. E-mail: info@malawibirds.org Morocco: Jacques Franchimont, Dept. Biologie Faculte des Sciences de Meknes, B P 4010, Beni M’Hamed 50003, Meknes. E-mail: j.franchimont@iam.net.ma. Namibia: Tim Osborne, PO Box 22, Okaukuejo, vis Outjo 9000. E-mail: kori@iway.na. Netherlands: Ron Demey, Van der Heimstraat 52, 2582 SB Den Haag, Netherlands. E-mail: rondemey@compuserve.com. Nigeria: Vincent Chikwendu Ejere, Dept, of Zoology, University of Nigeria, Nsukka. E-mail: misunn@aol.com. Sao Tome & Principe: Angus Gascoigne, CP 289, Sao Tome. E-mail: agascoigne@eits.st. Seychelles: Adrian Skerrett, Shipping House, PO Box 336, Victoria, Mahe. Fax: 380538. E-mail: maheship@seychelles.net or adrian@skerrett.fsnet.co.uk South Africa: Steven Evans, PO Box 1994, Cresta 2118. E-mail: stevene@cwt.org.za. Swaziland: Dr Ara Monadjem, UNISWA, P/Bag 4, Kwaluseni. E-mail: ara@uniswacc.uniswa.sz. Tanzania: Maurus Msuha, Tanzania Wildlife Research Institute, PO Box 661, Arusha. E-mail: carnivores@habari.co.tz. Tunisia: Hichem Azafeaf, 11 rue Abou el alia al maari, cite el houda, 2080 Ariana. E-mail: azafzaf@gnet.tn Uganda: Prof. Derek Pomeroy, Makerere University Institute of the Environment and Natural Resources, PO Box 7298, Kampala. E-mail: derek@imul.com. Zimbabwe: The Executive Officer, BirdLife Zimbabwe, PO Box RV100, Runiville, Harare. E- mail: birds@zol.co.zw. names forward, supported by a letter of recom- mendation from someone such as their employ- er, teacher or an officeholder in a local wildlife organisation. The scheme now also includes clubs who wish to be affiliated with the African Bird Club in African countries where it is difficult for local individuals to become members in their own right. Clubs accepted for membership under the scheme receive up to six copies of each issue of the bulletin for circulation among their mem- bers. Instead of paying a membership fee, Clubs are asked to provide a short annual report on their activities that may be published in the bul- letin. Clubs interested in becoming Affiliated Member Clubs are invited to apply to the ABC Secretary giving details of their membership, their constitution or a statement of their objec- tives and conditions of their membership, and their activities to date. ABC Information Service ABC offers a service to help members with information requests. Perhaps you are planning a trip to Africa and need local advice, or maybe you are in search of an obscure fact about an African species. The Club does not guarantee to find all the answers but will try to help. The service is free to ABC members. Contact: Keith The ABC Representatives scheme aims to support existing members by providing a local point of contact in their region, for example, to answer queries to the Club, to solicit submissions for the bulletin, and possibly to arrange local meetings for members. Existing ABC members can contact their local Representative in the first instance with queries relating to the Club. ABC Representatives help to recruit new members in their region, for example, by distributing posters and arranging local advertising. In Africa, ABC Representatives help to identify opportunities to invest the ABC Conservation Fund and candidates for the Supported Membership scheme. The Club aims to appoint many further ABC Representatives. If you are interested in supporting and promoting the Club in your region, have any queries, or require further information relating to the ABC Representatives scheme please do not hesitate to contact the Membership Secretary at the Club address, e-mail membership@afficanbirdclub.org. ABC is seeking Country Representatives in the following countries, principally within the Club’s region: Ageria, Azores, Benin, Burkina Faso, Burundi, Cameroon, Cape Verde Islands, Chad, Comoros & Mayotte, Cote d’Ivoire, Djibouti, Equatorial Guinea, Ethiopia, Gabon, Guinea-Bissau, Guinea Conakry, Libya, Madeira, Mali, Mauritania, Mauritius, Mozambique, Namibia, Niger, Reunion, Rodriguez, Rwanda, Senegal, Sierra Leone, Socotra, Somalia, St Helena, Sudan, Togo, Tristan da Cunha and USA. Betton, who is also custodian of ABC’s journal library, at 8 Dukes Close, Folly Hill, Farnham, Surrey, GU9 ODR, UK. Tel: +44 1252 724068. Fax: +44 171 637 5626. E-mail: info@african- birdclub.org. AfricanBirding e-mail discussion list Launched, in October 2000, by the ABC and the Pan-African Ornithological Congress, AfricanBirding or AB, as it is known, has become a useful forum for those interested in African birds. To join the discussion, which averages 1-2 messages a day, send a blank e-mail to AfricanBirding-subscribe@egroups.com. You will then receive an email instructing you how to join. The Club also maintains a list of members’ e-mail addresses. This list is confidential and used only for Club purposes, e.g. for informing members of upcoming events and news concern- ing the Club. It is not divulged to anybody out- side the Club or used for commercial advertising. At present it includes addresses for about 50% of the membership. Please send any additions or amendments to the membership secretary: mem- bership@africanbirdclub.org. Male Chestnut-bellied Sandgrouse Pterocles exustus, Marsabit, Kenya, 2005 (Kevin Vang) Female Chestnut-bellied Sandgrouse Pterocles exustus, Marsabit, Kenya, 2005 (Kevin Vang)