er ay ed re Oe tet ORO R NT ery Cot ent Pesta ee aE ANDI MeN eoPR, Lat Peery re rere) eel eer a oe er ee eer eter. yy Pe Sh rma Die OS eae ea IMAL Ok Smee Ne eats Apa Tectia Tha RP Aer SAR ae EN Oe aa gel ae SAS _ Aghoeh etm ah. hatqune TAA ARED be RY RSS CASS RCs ASR AN N Sec Mate Ratha pian athe 8giea nts Mae OMB ANM ARAM ones 8 ate Aaead u Casa Sonera ye Sareea tin nas DYE ee Ceo LL oak eae eect Neth Atta Sega ene a be helen GR mma EN AOS Nee Mees Ia ON ay ame NNN Sty Metin Weoraagen aes tm panera tehnns Mama Mem vb sAnuls Ys stutinfin sane sn ni hi Prana Uy ANNE RONDA Re Re te See We tethereal ee ee tem ey® Weer evs AF NEN omen Re Ah de MONae M MEA AME RIA, ee ee tO mre ne ne as ce PR me ee eee Bren emer ie Wer Wi ony Ne et ee ee Paw 0 OP OO © OP. OR Hen» ee eee Pe wee ee ee ete en a QAM Healt: WME Malet yea aatly mana kasannen tien, Sof OR geen a THEN TE Elke Ot Nr enon ea Heads arhen Se tet PDE OMT ee a re ee en eee eee te . en eee eee ee ea Aa Rent en atn rte Re enn ele rte ace eva e Ry Aree Reine the Bese Rel ea aS Ke Vee tetonte te tet . Sienecaeneertctong ae AANA AN HINg HAAS Ho tter amine Ot oath Meher ee bet nen net ha 97.48 tae a tee dal WAIN SNE Yee te Pts Motmaernae em RS ASAE SN CERIO tek om Rate Be Ome Miah fe Nut Bawtig Segre beara ener ere ererer srr rire Peer Trac ko Anke MaDe Matte Cw amy as te aA TENT ee eh ere ene Co eam ihe Mon Ne MMA ON LN Ry eT ae a NFA Ta Peete et Prrurn rm r Un a es Po ROTOR LLnL to MLA Rous © pa ernetieaee® Seger gnegemanynedahasnsac snes 2a : er GL Lads a PENN ote Nan wane te rey See tn ee he pre ator ee ee a ad gots cy *. eee ee et Creer: oie ark Roe ten Na hae ete Veh ORs NR REM Rg MAUD ar . ee ee ee re ee ey me BTN en Cee ee ee ee Aa ee " ONE NM Yr kaen e WML Oh ORale ane & ene ANNI foe ie eae ye hee A Ne Liatyatenct ah Peereter trot ererirn an ' ake ermal et De he DER ee te get OFA De Ame ha Mates aa NANan ofa Saleen AE No DA atecsns attic (DRAIN: _ va aneunana na btnahenatiorite™® ining tae rarer amanem ur Rses Raine (an rasta on . 2 ON em, Rg OR art tgs cre NM Ra Nt et ae een DEN aha NAjRATS 4A NAMA NA rin earhatag Ria a The dy, Hassan hen NALS TAS NUN Geter gee Rite wee ee tae Ape a mata ag NY TM Dod Ree N inate tack NUR Nae Maal nee a Oren RL ag Worthe tat tte De Sor hs a FO abe NARS /TA TARE serra tap hee amet ATARN.AN wits Nh xs HaThenneriaing 1A wnt bene Aaraanene” Meriter ooh a ee Sg SAAN NAN A eee toe mee ne ter ua tation ANIME Sat Lethe Rte NOR Ne IRS NT Sammy 9 eget Pine "ta Mowe pt a AA Be 8a RAE AR aD Vr Vath Daly My Vv d Ra waite aw med whe AA APLO TAMA mDA LORAIN RNAS Tet hae gm hd tat a SA RIN NS Rete NN A Gat Bike Be tata merte Nes war Doel ag REN AIAN AS MRO he iaan tev af Tas ne pei mine niin Revi at PAIN ONS ARPA RaW Rameng | Le WANE MC Em aa TH m OT ee eal ae one Meet ket Prete tere Ome Pei Neath! Noon ENO Nr ee ee \etiet oe fess son ee et et Ph Uae ans NARS Na mee CU i ha Leo heey See AND Om ee ee ee Ban eee tO hte Hany Pee hh atime ee an a r a Ne MENS Malte he rk ee Nay ett en Dae Ne Ree Prine nee na Nt ee ae Se ate AV ene Ae een ern Pen ihe tener Sane haem Maan aNa esa Meek phensPeherasmetetes tet tos elena Ns a Ph i ie err Sor pate ee eet Rate ae see’ Samy ena tee Nowa ha re tenner ® Nene y ACen RWSL Hear na ten MAL Daten eaten beer nat is ba hee ee Rete etey Aan atte Nate Pin RP a. Peo, eemenieip iain fe ten ote cathe le ae Demian whe ten A NM Reta tert Nar BEM ee eee Na Bea om tin Ma rane Ree tee Nett ern erent ron. thon Line ea te ar 9m i NTN ASTI Ot Pete OM AFR RNR Nem Ratan sna TE Im ee kms Pee ee Oe Cay Seer ot ete er CA Nar eaten a be Me ke Me a Neda tetany Sh ate ereay ete Peete ee ae a ae aon! ve et rae LRG RA ETL GM a9 Vee ANS (he ee be ea et ye apne nite re nee hepa witha Yat set een tet anim stan ae nara rhe ee hae wate ta tee ney Dee te eee hes =Ae gabe pt aE tet erence thar wawanee nate ee he i SO re i pe ete sete te 8 ee > ah ee eh i oe => vy painted pee Stee s spe Or “4 me ip ast t oar a sepaney

= R> | SS \\; es a ee — 3 ITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS S31UYVYUSIT LI NVINOSHLIWS SMITHSONIAN NVYINOSHLIWS SW YN \ SMITHSONIAN . . LIBRARIES SMITHSONIAN INSTITUTION NOILN > “fy SJIYVYGIT_LIBRARIES SMITHSONIAN INSTITUTION ee \S NVINOSHLINS SA3IYVYSIT LIBRARIES SMITHSONIAN = uu =, wo es — = aa) te ad =e fed ye to 4 < ie at = = = 3 = é = — 7) res ” m = 7) = ” ” 7) = = 4 “4 i pa 7) 7) oO e) z z2 > > ra Z ~ SMITHSONIAN THSONIAN _INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVHSIT NOILNLILSNI LIBRARIES igo x SNOSHLINS S3IYVUSIT LIBRARIES SMITHSONIAN _INSTITUTION INSTITUTION NOILNLILSNI S31YVNAII LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI INSTITUTION S3iuvygi INSTITUTION NOILALILSNI NVINOSHLINS S3Ztuyvudit_ LIBRAR ° ? Ve lbs, MY z : . SMITHSONIAN INSTITUTION NOILNL CORN = W Yjiby AM) x Up” fi 7 . % ¢ SMITHSONIAN NVINOSHLIWS NVINOSHLIWS SMITHSONIAN BOSHLINS S31YVYSIT LIBRARIES tity yn ae 1 1ES SMITHSONIAN ARIES 'SON 7a JLILSNI TP ry ARIES ILILSNI uv le) m_ Pa a N ‘* . & “WN se. ZL. za THSONIA h WY f, % 7 NVINOSHLIWS SMITHSONIAN NVINOSHLIWS SMITHSONIAN SMI , w” — Ww _ = z= wW uJ Wg a 7) YW Ww feed Cc Me A: m = mM. — -_ ; 2 | cms oO a | = oad = IAN INSTITUTION NOILOLILSNI NVINOSHLINS S31uvualy s3iuvudii INSTITUTION INSTITUTION HLINS S31NVYSII SMITHSONIAN _ INSTITUTION NVINOSHLINS SJ3IYVUdIT LIBRARIES NVINOSHLINS S31Y¥Vudlt no rs N= < < 4 = > S = : > P = > ; me Ww) : ” a, ONIAN INSTITUTION NOILNLILSNI_NVINOSHLINS | S3 iyvusgII_ ie = WT 2 . ai i = “S o ei XY - Ty — Lr ASS tf = < a7 ~ < = = od a SS 0 of i ES am ms fea) HEL N m = ae a = = SHLINS SaIuvVuaIT LIBRARIES NOILNLI y = is = wi z i, = re) >a 2) = on = es a aS a a _ a a 2 a = .F E ms a _ = o = o 2 = w S Ww == a) eee | a ~S 2, “~ onl a ne | 4 LIBRARIES SMITHSONIAN INSTITUTION NOILALIL e ve ain’ S ow fe sd haat 7 > Ng > > = pan & 2 = 2 : O z a z INSTITUTION. NOILNLILSNI_NVINOSHLINS S3IMVYEIT_LIBRAR| SA1UVUGIT LIBRARIES SMITHSONIAN INSTITUTION | NOILNLI LIBRAR. NOILNLI INSTITUTION NOILNLILSNI NVINOSHLINS S3I Yvudil ats = * w & w” e x = . < = = & = YI z \ 5 Ys 7 Gy 2 i Uys N “” wm, SY - Wi vi, D 2 po! ee << SW: 2) ‘ Mh, a= oO Ya 7* he YQ = » SS = % ye = z — NY = ~S >” = > = S ae z ov = ? NOILNL 3 i¥vud ski BRARI ES SMITHSONIAN _INSTITUTION i w Kae = res 5) wl Vy 2 é SON uw sai z KAM) = KAM & Vile, = [5 vA oe | * et 9 2 =I 5 ve / / OT 80°W 70°W Fig. 2. Variation in amount of white colour on the undertail coverts of Andean Coots (circles) and Colombian Coots (squares). Five feathers with corresponding symbols are depicted, and geographic range of recognized taxa indicated. 21 [Bull Brit.Orn.Cl.19 83 103(1)] value 9.42, 3 individuals having white only distally on some inner webs. Andean Coots usually have small white spots, but there is considerable individual variation all the way from Narifo to the Titicaca area, while white spots are almost absent further south. The average score for Andean Coots was 2.33 in Narifio, 2.37 in Ecuador, 2.93 in Central Peru, 1.80 in the Peruvian/Bolivian altiplano and 0.35 in northern Chile and Argentina. Six birds which overlapped with the range of variation among Colombian Coots were from Narifio (1), central Ecuador (1), central Peru (3) and Titicaca (1). The variability may thus occur throughout the zone with polymorphic variation in soft part colours, although the amount of white colour on the secondaries was not associated with any particular soft part colour. Fig. 2 shows the variation in amount of white colour on the undertail coverts. These are completely white and form 2 broad white bands under the tail in most coots and gallinules, including Nearctic and Colombian Coots. This possibly primitive character is found in juvenile Andean Coots and in adults from the southern parts of the range. However, adult Andean Coots from coastal Peru northwards from Lima and from the paramos of Ecuador, and especially Narifio, have black inner webs on the undertail coverts and the outer webs more or less streaked or freckled black, in a few individuals virtually lacking any white colour. As seen from Fig. 2, very few birds from the highlands of central Peru and further south have dusky freckles on the undertail coverts, and no bird even approached the condition seen in the north of the range. Conclusion The variation in colours of bill and frontal shield and in wing-pattern give no basis for taxonomic subdivision, but suggest a genetic instability through- out much of the range of the Andean Coot. This could be a result of hybridization of 2 main morph types of Andean Coots (previously known as Fulica ardesiaca and F. americana peruviana), which possibly once were allopatric (Fjeldsa 1982). Although a considerable variability is found in the north of the range, no specimens appeared to be hybrids between any of these and Colombian Coots, and there was no clinal change towards the Colombian Coot. On the contrary, the northernmost Andean Coots decidedly diverged from Colombian Coots in one character, the pattern of the undertail coverts. As this pattern is exposed in certain displays (Fjeldsa 1983), one could suspect that the geographic trend was due to selection against hybridization in a period of sympatry. Although direct evidence as to how Colombian and Andean Coots would interact in sympatry is lacking, one can conclude that the geographic variation gives no direct evidence of past interbreeding, but instead suggests a possible divergence in a plumage display signal. This evidence supports the conclusion (Fjeldsa 1982) that the Andean Coot should be maintained as a separate species, Fusica ardesiaca. The variation in tail-pattern was overlooked by previous students of Andean Coots. As my data show no overlap in this character between birds from the Andean puna zone and birds from the paramos of Narifio and Ecuador and from coastal Peru, a recognition of subspecies appears to be justified on the basis of this character. Since the species as such was described first from the puna zone (Junin, 11°S, 76°W, 1843), the puna zone populations should be ranked as nominate subspecies. As also Morrison’s (1939) name [ Bull. Brit.Orn.Cl.19 83 103(1)] 22 peruviana refers to birds from Junin, a new name must be proposed for the aberrant populations found further north: Fulica ardesiaca atrura subsp. nov. Diagnosis. Differs from the nominate subspecies as the undertail covert feathers are not purely white, but have black inner webs and more or less extensive black streaking and freckling also on the outer webs; in some individuals, in fact, the undertail coverts are nearly completely black. Distribution. In paramos and some lowland swamps from Narifio in southern Colombia through Ecuador and coastal Peru south to Lima. As Andean Coots from paramos in northern Peru were not represented in the present data, their racial attachment remains unknown. Type specimen. Zool.Mus.Univ.Copenhagen 37.891, Ecuador, 7 October 1909. Material examined. 77 ad. Andean Coots, including 25 of this taxon. Acknowledgements: My travels were supported by grants Nos. 511-8136 and 11-2250 of the Danish Natural Science Research Council and by Queen Margrethe’s and Prince Henrik’s Foundation, Frank M. Chapman Foundation, G.E.C. Gads Foundation and Knud Hojgaards Foundation. References: Crawford, R. D. 1978. Tarsal color of American coots in relation to age. Wilson Bull. 90: 536-543. Fjeldsa, J. 1982. Biology and systematic relations of the Andean coot “‘Fulica americana ardesiaca”’ (Aves, Rallidae). Szeenstrupia 8: 1-21. Fjeldsa, J. 1983. Systematic and biological notes on the Colombian Coot Fulica americana columbiana (Aves, Rallidae). Steenstrupia, in press. Gill, F. B. 1964. The shield colour and relationships of certain Andean Coots. Condor 66: Hurtubia, J. 1973. Trophic diversity measurement in sympatric predatory species. Ecology 54: 885-890. Morrison, A. 1939. A new coot from Peru. Bull. Brit. Orn. Cl. 59: 56-57. Ripley, S. D. 1977. Rails of the World. David R. Godine: Boston. Address: Dr. Jon Fjeldsa, Zoologisk Museum, Universitetsparken 15, DK 2100 Kobenhayn, Denmark. ©British Ornithologists’ Club 1983. A new species of Thicket Warbler Cich/ornis (Sylviinae) from Bougainville Island, North Solomons Province, Papua New Guinea by Don Hadden Received 13 July 1982 The little known and elusive genus of thicket-warblers, Czch/ornis, was first described by Mayr (1933) from a specimen taken in 1926 by R. H. Beck, who collected one bird from mountain forest (2500 ft) on Espiritu Santo in Vanuatu. Mayr named this unique specimen C. whitney?. Another male and 3 females were also taken from Espiritu Santo between 1933 and 1935 by A. J. Marshall and T. Harrisson (Cain & Galbraith 1955). A new subspecies, C. w. turipavae, was also described by Cain & Galbraith in the same paper from a single specimen taken by native hunters from Turipava (4100 ft) on Guadalcanal. Two other. Cich/ornis specimens were collected in December 23 (Bull. Brit.Orn.Cl.1983 103(2)| 1958 at 5200 ft in the Whiteman Mountains, central New Britain by E. T. Gilliard (1960). These proved to be of a distinct species and were named Cichlornis grosvenori. My discovery of the Bougainville population of Cichlornis came about because I was trying to track down a rumour of nesting shearwaters. During my years (1976-1980) on Bougainville I had constantly asked local villagers if they knew the whereabouts of the nesting sites of birds that had webbed feet and lived in holes in the ground in high mountains. Eventually, I was informed by Tony Anung from a village behind Arawa, that he had found the nest of a bird in a hole in a bank near the top of the Crown Prince Range. Accordingly I arranged to spend a long weekend camping at that site to check on this nest and to search for additional shearwater sites. After about 7 hours walking we had progressed from sea level to 5000 ft and were at the nest site. The nest was not that of a shearwater, but obviously of a passerine of some sort. We camped over a ridge close to the nest site and while my guides spent the following day in the forest searching for shearwater nests, I erected mist nets on the ridge above camp. It was in one of these nets that a Cichlornis sp. was taken. By the time we had struck camp and returned to Arawa the Cich/ornis had started to decompose slightly, but the skin was saved and is now lodged in the American Museum of Natural History (AMNB). It proved to be a new species. Cichlornis llaneae sp. nov. Holotype: AMNH No. 824713, sex ?, apparently adult, from Crown Prince Range sooo ft (1550 m), central Bougainville Island, North Solomons Province, Papua New Guinea, approximately 6°19’S, 155°30’E; collected by Don Hadden, 17 June 1979. Distribution: Known only from the type locality. Description of holotype: Head, wings, back and rump sooty olive; feathers of rump not especially elongated or fluffy. Supra-orbital line rich cinnamon- rufous and a distinct black area before, behind and beneath the eye forming a small mask. The lesser wing coverts black, tipped with brownish olive, giving a scalloped appearance. Throat and upper breast cinnamon-rufous. Lower breast and abdomen cinnamon-rufous, shading into dull brown. Flanks and lower abdomen dull brown. The rectrices were in sheath, but black and acuminate, not spine-tipped. The shafts were not stiffened. Legs and bill dark. I have much pleasure in naming this new thicket warbler for my wife Llane Hadden. Measurements of type: Wing 73.5 mm, tail in sheath, bill from base 21.0 mm, tarsus 26.0 mm. Additional remarks: Comparison of C. /aneae with types of C. grosvenori, C. w. whitneyi and C. w. turipavae give the following distinct differences :- Throat and upper breast of C. //aneae cinnamon-rufous, whereas the other 3 populations are tawny buff, C. w. whitneyi being the lightest of the three. Lower breast and abdomen of C. //aneae cinnamon shading into dull brown, whereas the other 2 species are lighter tawny buff, except that in C. grosvenori there is a lighter central area with some feathers mottled buff and brown. The head and back of the other 2 species are brownish olive, whereas they are [ Bull. Brit.Orn.Cl.19 83 103(2)] 24 TABLE 1 Measurements of known specimens of Cichlornis Bill from Sex Wing Tail base Tarsus C. w. whitneyi (Type) (1) 3 72.0 70.0 + 21.0 28.0 (1) 3 68.5 65.0 20.0 pay ee (2) 2 63.0 57.0 20.0 25.5 (2) g 60.0 56.5 + 19.5 26.0 (2) 2 64.5 68.0 18.5 25.0 C. w. turipavae (Type) (2) 3 65.5 76.5 — 27.5 C. grosvenori (Type) (3) 2 72.0 59.0 20.0 31.0 (3) 3 71.0 65.0 19.5 31.0 C. Haneae (Type) ? Bits a 21.0 26.0 (1) Measurements from Mayr, 1933, p. 4 (2) Measurements from Cain & Galbraith, 1955, p. 91 (3) Measurements from Gilliard, 1960, p. 3 sooty olive in C. “aneae. The supraorbital line is rich cinnamon-rufous in C. Haneae, but the other two species have tawny buff lines. The black mask of C. /aneae is midway in size between the small mask of C. whitneyi and the large mask of C. grosvenori. The lesser wing coverts on C. //aneae are markedly scalloped, whereas there is only slight scalloping on C. w. whitneyi and none at all on C. w. turipavae or C. grosvenori. The most striking difference is that the rectrices of C. Maneae are not disintegrated at the tips nor the shafts stiffened as in the other 2 species. In addition the tail of C. //aneae is black, not brown as in the other 2 species. The fact that the tail feathers of the type of C. /aneae are in sheath precludes further comparison, but there is no indication that this specimen is immature. It seems more likely that the simultaneous regrowth of the rectrices is due to the bird having suffered an accident. The wing of C. /aneae is much more rounded than the wings of C. whitneyi and C. grosvenort. The outermost primary (No. 1) is not much shorter than No. 2. Primaries 2, 3, and 4 are equal to or slightly longer than No. 5. In C. whitneyi and C. grosvenori, primary No. 1 is much shorter than No. 2, and No. 2 is significantly shorter than No. 3, while Nos. 3 and 4 are usually slightly longer than No. 5. The tarsus is shorter and the legs and feet appear weaker in C. Haneae. In some respects the new bird is intermediate between C7ch/ornis and Ortygocichla (including Trichocich/a). A case might be made for describing the Bougainville form as a new genus but very little information is available on these genera, and C. /aneae is definitely closest to C7ch/ornis. R. Orenstein suggests (77 /itt.) that these 2 genera should be included in a larger genus, Megalurulus, to include M. mariae of New Caledonia, Trichocichla rufa of Fiji and Buettikoferella bivittata of Timor. However with so little information available, it seems preferable at present not to speculate further on generic limits. Obviously a complete revision is much needed. Description of nest and egg: 'The passerine nest mentioned above had been placed in a niche in a vertical wall of a creek. About 1 m downstream from the nest the creek disappeared underground and so the nest site was surrounded by walls on three sides. It was about 2 m above the bed of the creek, and the width between the creek walls was 2-3 m. Very little water was trickling down, the weather having been fairly dry the previous week. The nest was made of dark vegetation with a lining of lighter, finer fibres contrasting with the dark outer parts of the nest. A lip of dark vegetation 25 [ Bull. Brit.Orn.Cl.19 83 103(2)] hung down the wall from the nest, which contained 1 egg 25 x 18 mm and was obviously deserted. The oval egg was a light cream colour and was entirely covered with small brown spots, more heavily at the thicker end where they formed a brownish cap. The egg and nest are in the AMNH. This nest and egg, found in June 1979, turned out to be those of Czch/ornis Maneae, but this was not known until one year later when I again camped in the area in June and an identical nest was found in the very same niche as the one I had first seen. For further information on this nest and 2 other specimens of C. //aneae and photographs of the type and the nest, see Hadden (1981). Acknowledgements: My wife Llane Hadden has spent many days alone while I have pursued my interests in photography and ornithology. It is as a token of gratitude for her support and interest that I have named this new species for her. I am also grateful to Elliot Harding, Francis Munau and Tony Anung and other village men without whose help I would not have found the thicket-warbler area. Considerable help has also been given by Jared Diamond and R. Orenstein, and in particular I am much indebted to Mary LeCroy and Ian Galbraith who gave invaluable assistance with comparison of specimens from the AMNH and BMNH respectively, as well as their time and expertise, especially that of Mary LeCroy in helping to draft this note. References: Mayr, E. 1933. Birds collected during the Whitney South Sea Expedition. XXII. Three new genera from Polynesia and Melanesia. Amer. Mus. Novit. 590: 1-6. Cain, A. J. & Galbraith, I. C. J. 1955. Five new subspecies from the mountains of Guadalcanal (British Solomon Islands). Bu//. Brit. Orn. Cl. 75 : 90-93. Gilliard, E. T. 1960. Results of the 1958-1959 Gilliard New Britain Expedition, 2. A new species of thicket warbler (Aves, Cichlornis) from New Britain. Amer. Mus. Novit. 2008: 1-6. Hadden, Don. 1981. Birds of the North Solomons. Wau Ecology Institute Handbook No. 8. Wau, Papua New Guinea. 107 pp. Address: Don Hadden, 288 Yaldhurst Road, Christchurch, New Zealand. ©British Ornithologists’ Club 1983. The relationship of male Lesser Honeyguides Indicator minor with duetting barbet pairs by Lester L. Short and Jennifer F. M. Horne Received 28 June 1982 Our field studies of barbets (Capitonidae) in East Africa have been disrupted regularly by honeyguides (Indicatoridae) interacting with the barbets, and with each other. We particularly elicit approaches by honeyguides when we use our tape-recorder to play back barbet duets, the approaches being to us or to the barbets, which are also stimulated by our playback activities. We reported (Short & Horne 1979) on these responses by Indicator variegatus, I. minor and probably I. narokensis to various barbet species and to playback of the barbets’ voices. In that report we posed several questions relating to the honeyguide-barbet interactions. Further data now available allow us to narrow the quest for reasons underlying these honeyguide-barbet interactions. If we assume that, generally, the honeyguides coming to barbet vocal activities are females seeking a nest in which to lay an egg, since honeyguides [ Bull. Brit.Orn.Cl.19 83 103(F)] 26 are nest parasites especially of barbets (Friedmann 1955, 1968), the close approach of honeyguide females to singing, duetting barbets nevertheless would appear non-functional, the parasite being “interested” presumably in the hosts’ nest, not in the other activities of the barbets. We suggested (1979: 17) that honeyguide females, and perhaps males, might use those activities of barbets associated with breeding as “cues” that could trigger breeding readiness in honeyguides and even bring together prospective honeyguide mates. However, we remarked that such functions seemed both energetically wasteful and disadvantageous in that they arouse the barbets and facilitate their recognition of the honeyguides as harmful, particularly since barbet pairs often have helpers. The non-breeding helpers presumably could gain experience that would eventually increase the likelihood of successful breeding if they were to learn to attack and drive honeyguides from their vicinity. OBSERVATIONS In our garden outside Nairobi we can at any time of year elicit White- headed Barbet Lybius /eucocephalus tesponses to playback of its voice, the responses varying from chattering, aggressive overflights and close approach to the recordist, to excited calling and “greeting ceremonies” (see Short & Horne 1982). Between July and January we hear at intervals ageressive trills of Lesser Honeyguides Indicator minor in the garden. At those times, repeated playback of the barbets’ greeting ceremonies inevitably results in the appearance and approach of a Lesser Honeyguide which, when perceived by the barbets, is chased by one or more of them. The honeyguide often returns, and indeed may retaliate by attacking one or another of the barbets. Occasionally, two Lesser Honeyguides would simultaneously approach the playback recordist (this species is variable in plumage, but most individuals can be identified as I. minor by size and by the presence of a distinct moustachial stripe). In some 20 of such cases observed sporadically between 1979 and 1981, the honeyguides would attack one another, the pursuit taking precedence over honeyguide-barbet interactions (leaving the barbets perched, often “panting” from the exertion of chasing the speedier, more manoeuver- able honeyguide). We assume that the Lesser Honeyguide parasitizes L. leucocephalus, for we have reported (Short & Horne 1979) this honeyguide entering and being evicted from a nest of /eucocephalus. Our studies of the Black-collared Barbet Ly}ius torquatus (Short & Horne 1979, 1982) in coastal Kenya have provided over 2000 additional observations of honeyguide-barbet interactions. (Less frequent interactions of honeyguides with the barbets Lybius melanopterus, L. guifsobalito and L. rubrifacies recorded in our unpublished notes are not reported here.) Lybius torquatus is a frequent host of Indicator minor (Friedmann 1955, Ranger 1955). We have supplemented our observations by collection of 6 Lesser Honeyguides taken (after some minutes of observation) from beside duetting, displaying pairs of Black- collared Barbets south of the Nature Reserve in Arabuko-Sokoke Forest (Britton & Zimmerman 1979). Five of the 6 honeyguides collected proved to be males, much to our surprise, and 26 seemingly separate, aggressive, sustained honeyguide-honeyguide interactions suggest that many if not most of these also involve males. 27 [ Bull. Brit.Orn.Cl.19 83 103(1)| On the afternoon of 7 July 1979 we collected a Lesser Honeyguide that had been following a (playback-stimulated) frequently duetting pair of L. torquatus for some 15 minutes, moving from one duetting post to another, perching close to the barbets, interrupting them, being chased, and then returning to them. It was a male with enlarged (4 x 3 mm) testes. The next day we worked with another barbet pair, and at 08:00 spied a honeyguide following the pair, but at a greater distance than the previous day’s bird. This individual followed the barbets on 4 consecutive flights to singing sites (trees scattered about their territory) and watched them sing 3 duets. In most cases we found that a honeyguide attracted to a pair of duetting barbets approaches them closely, landing beside or even between them, thus dis- rupting any duet attempt, though the barbets may sing an interrupted duet or perform a greeting ceremony. In this instance, however, the honeyguide perched 3~5 m from the barbets and did not attempt to join them or fly directly to the sites they occupied. We collected the honeyguide from a perch 3 m from the barbets; it was a female in slight moult but with a somewhat enlarged ovary. In November 1981 we worked in the Arabuko-Sokoke Forest with a pair of Black-collared Barbets that were duetting regularly and excavating a cavity, whether for roosting or nesting is uncertain. Ato6:30 0n 19 November we observed one of the barbet pair chasing a honeyguide through the trees. The barbet then returned to its former perch and duetted with its mate. A honeyguide (uncertainly the same one) again appeared, flying to the barbet pair; all 3 flew off in a chase. We then saw what we thought was a barbet in flight fighting with the honeyguide, but both birds proved to be honeyguides and judged by voice were Lesser Honeyguides. They circled back and forth in pursuit of each other, tails fanned to exhibit the tail pattern, and called (trill calls); several times they grappled in the air, before they went off in a long pursuit flight. We stayed with the barbet pair and at 07:00 heard a honeyguide’s trill call to which we played back a Lesser Honeyguide call (one recorded by C. Chappuis in Malawi), and this brought a honeyguide to us, calling. It was chased about by one barbet as we recorded the honey- guide’s “ta-wee-wit”’ call —- a major vocalization of I. minor, song-like, but not the note given at a call site reported by Ranger (1955) — and its trill, and then we collected it from beside the 2 barbets. It was a male I. minor with testes 4 x 3 mm and 3 x 3 mm (left and right respectively). We heard 2 calls and had one sighting of a second honeyguide about the barbets during the next 40 minutes. Late that day we “lost” the barbet pair (they proved later to have gone north, presumably toward a roosting hole, between 16:00 and 17:00 hours), but as we repeatedly played back their duet a Lesser Honey- guide circled back and forth overhead, “searching” for the barbets. The honeyguide stayed with us for 10 minutes flying about us from perch to perch, then flew off. A short while later we played back the barbets’ duet and had 2 honeyguides circle overhead then go off in chase. Nearby, at 17:20 hours on the same day our barbet playback brought to us a single Lybius torquatus and 2 honeyguides, apparently Indicator minor. The honeyguides seemed to try to approach the barbet, but the barbet attacked one honeyguide as the other honeyguide also attacked it, and all 3 birds circled in a furious “dogfight”. We were unsuccessful in attracting a second barbet, but the one barbet was engaged with the 2 honeyguides, and they [ Bu//. Brit.Orn.Cl.19 83 103(1)| 28 with each other, for 50 minutes. The barbet chasing one honeyguide seemed to trigger an attack by the second honeyguide on the first, and this appeared to “confuse” the barbet, which shifted its attack to the other honeyguide. The barbet tired more readily and when it perched, the 2 honeyguides flew about in sweeping circles or directly off in a line, then back, still in pursuit of each other, to the barbet. We continued working with several barbets, seeing honeyguides daily as they came to our playback of the barbets or to the calling of the barbets themselves. Not only the barbet duets but even greeting ceremonies and the sounds accompanying their courtship feeding attracted honeyguides, causing them to approach. At 08:05 on 21 November 1981, we watched 2 honeyguides attracted by the playback of a Lybius torquatus greeting ceremony as they engaged in a chase near the excavating pair of barbets. The 2 honey- guides perched in a tree north of the excavation, gave low grating calls with tails fanned, bowing to each other and raising and lowering the (spread) tail. One of the honeyguides was notably smaller than the other and held its bill open, but gave the same displays, its call being a buzzy trill. The 2 flew in pursuit of each other, circled and came together grappling, floating downward toward the ground clutching each other, with tails spread, then breaking apart. The larger bird flew away, and the other followed. At 08:18 we played a duet of L. sorquatus, instantly bringing a moustached I. minor to us, then a second honeyguide; they attacked one another, then engaged in a fast chase, bursting through undergrowth and canopy, with tails spread, pecking and hitting each other in flight. They disappeared in a chase to the southwest. Again we brought the same 2 honeyguides back, this time with playback of the barbets’ greeting ceremony, and the honeyguides fought and chased round and round until, at 08:25, a barbet joined the fray, attacking one honeyguide; but before the chase had gone 20 m the same honeyguide was chasing the barbet, being much faster in flight. After a while they perched side by side, the barbet “‘panting’’, and then gave chase again. The second honeyguide, apparently watching, overflew and disappeared as the barbet and first honeyguide chased to and fro. At 07:43 on 22 November 1981 we employed honeyguide calls to bring the barbet pair to their excavation. After the barbets duetted near the excavation a honeyguide joined them. The barbets called and flew at the honeyguide, which zoomed upward in the air (the barbets dropped back down to a tree) and then without calling and with tail spread widely in a somewhat stilted flight, flew twice in a circle about zoo m in diameter centred over the area of the barbets’ excavation. A honeyguide was near this pair until o9:00 on that day, although we were unsuccessful in attracting honeyguides to this same pair of barbets later in the morning. In fact our rate of success in drawing honeyguides to barbet playback was greatest before 10:00 and after 16:00; the barbets responded at any time, but their response was more sustained, with more frequent duets and less rapid habituation, to playback early and late in the day. Next day (23 November) we worked close to the excavating pair of birbets. Two honeyguide trills were heard between 07:00 and 07:26, but we did not playback, preferring to watch the behaviour of the barbets. At 08:10 the barbet pair flew to a tall dead tree and duetted, a honeyguide instantly appearing and getting between the duetters. The “lead” barbet, namely the one which initiated movement 29 [ Bull. Brit.Orn.Cl.19 83 103(1)] to duetting sites and was presumably the male, viciously attacked the honeyguide. The barbet pair then flew and the honeyguide joined them and followed the lead barbet to another tree, where the 2 barbets performed their greeting ceremony. Again accompanied by the honeyguide, the barbets then returned to the dead tree and attempted a duet. At this point we fired at and missed the honeyguide; but one playback of the barbets’ duet instantly brought the honeyguide, trilling, and then the barbet pair, back to the dead tree. We collected the honeyguide from within 1 m of this barbet pair at 08:30. It proved to bea male I. minor with testes 3 x 3 mm. At 08:37, at the same site we played back the L. torquatus duet, and were rewarded by approach first to us, then to the excavating barbets, of another honeyguide. This honeyguide was chased by one barbet, but evaded it and flew back to the tree bearing the incompleted excavation. It landed near the excavation, then flew up to a perch and trilled. The barbets first gave chase, then the honeyguide chased one barbet off to the northwest. After a few minutes we played back a barbet duet and a barbet and honeyguide appeared together in a chase (apparently of the honeyguide by the barbet) before perching in the same dead tree from which we had just taken the male honeyguide as described above. A second barbet joined the first barbet and honeyguide and the 2 barbets attempted to duet, the honeyguide being perched only 1 m away when we collected the latter (at 08:57). This bird too was a male (testes 3.5 x 3.5 mmand 3 x 1.5 mm, left and right respectively). Just after 09:10, having glimpsed yet another honeyguide nearby, we played the barbets’ duet at a point between the tree bearing the excavation and the dead tree referred to above. The pair of barbets appeared im- mediately, with a honeyguide close behind them. A chase occurred, the lead barbet chasing the honeyguide, the latter reversing the pursuit, and over and over again. Another playback brought back the lead barbet, followed by the honeyguide, and then the second barbet; they all perched just west of the excavation site. L.L.S. went to the dead tree from which the 2 previous honeyguides had been collected, and played back the barbets’ duet, bringing in all 3 birds. The barbets managed to duet as the honeyguide flitted about them in a tight circle; but, when the duet ceased the honeyguide moved away from the barbets and was collected (at 09:25). This honeyguide, taken from the same tree as the previous 2, and from beside the same pair of barbets as the previous 2, as well as the male of 19 November, also was a male, with testes 3 x 2 mm and 4 x 3 mm (left and right respectively, a reversal from the usual left testis being longer). Thus, within one hour, 3 different male Lesser Honeyguides were collected as they interacted strongly with the same pair of barbets in the same tree. This barbet pair continued that morning to react, by duetting, to playback of their duet. One other dark-coloured honeyguide, of uncertain species, was seen to the west of this pair before we left the site and the area later that morning. The testes of the Lesser Honeyguides that we collected seem sufficiently enlarged to consider the birds as in breeding or pre-breeding condition (also fide R. Payne). However, we saw no copulations of honeyguides, nor did the Lesser Honeyguide males employ singing or sites from which to call (Ranger 1955) in order to attract females. The vocalizations of the honeyguides during their interactions with each other and with the barbets [ Bull. Brit.Orn.Cil.1983 103;(2)] 30 were usually trilling calls and squeak calls, which are those associated with aggression (Short & Horne 1979). It may be that the behaviour of the Kenyan honeyguides differs from that in more seasonally oriented popu- lations in southern Africa (note, for example, the large testes of both July and November Lesser Honeyguides). We ourselves have noted that gonads are somewhat enlarged in most barbets and honeyguides that we have collected (even in subadults of such barbets as Trachyphonus darnaudii) in Kenya, suggesting that irregularity of the rains or other perhaps associated factors demand a state of readiness to breed (or to defend resources necessary for breeding) all the year round. The above observations are summarized from our field and tape-recorded notes. We have noted many other honeyguide responses to calls and duets of various barbets (including, e.g. I. minor responses to L. melanopterus and L. guifsobalito), and also frequent honeyguide—interactions. The latter include some interspecific interactions (e.g. of I. minor to I. narokensis, of I. indicator to I. variegatus, of I. variegatus to I. narokensis, and of I. indicator to I. minor) as well as over a hundred instances of apparent I. minor intraspecific chases in the vicinity of barbet pairs. The behaviour of I. indicator (which is common in the areas worked) in regard to the barbets and our playback differed markedly from that of I. minor. Only rarely did a Greater Honeyguide appear when we played barbet duets, and it would either leave after a look at us, chase a Lesser Honeyguide if one was present, or (twice) commence guiding calls directed at us. Unfortunately not all of the honeyguides that we studied could be observed closely; some I. minor have very weak malar stripes that are not readily apparent (e.g. there is only a trace of the malar stripes in one of the 4 males just described, and the malar area varies in colour considerably in the other 3 birds) and small sized I. minor can be mistaken for I. narokensis or vice versa. Hence, identification of a honeyguide species (let alone deter- mination of its sex) in the field was not always possible. Nonetheless, it is apparent that many, if not most (possibly nearly all) the honeyguide— honeyguide and honeyguide—barbet interactions involved male honeyguides. DIscUssION We have established that Lesser Honeyguides of both sexes, including many males, are attracted to singing (duetting) pairs of certain barbets. There is ample evidence that 2 or perhaps more honeyguides are attracted simultaneously to duetting barbets (or to playback of their duets), and that when this occurs they engage in fights associated with the presence and location of the barbets or of the latters’ excavations. The collecting of 4 male Lesser Honeyguides, 3 within one hour in intimate association with the same duetting barbet pair clearly suggests that there is benefit to the males in such association, and the fighting we have described suggests that the honeyguides are exhibiting themselves with, and defending “ownership” of, particular barbet pairs against one another. The possibility also exists (Short & Horne 1979) that the honeyguides could utilize the duetting barbets as “cues” triggering or enhancing repro- ductive development in the honeyguides (of both sexes), helping to bring them reproductively into synchrony with their hosts. This could be accomplished by the honeyguides observing the barbets without actually approaching them and interfering with their displays. On the other hand, the interactions of the honeyguides with barbets may create a disturbance 31 [ Bull_Brit.Orn.Cl.1983 10x{1)] bse draws the attention of male and female honeyguides to the presence of territorial male. A female Lesser Honeyguide, by following a ladies pair, but not so closely as would a male, perhaps places herself in a favourable position from which to attract (through the barbets’ activities) the attention of a possible mate. We have no data on take-over of barbet holes by honeyguides. It is not even known where honeyguides roost, nor whether or not they require a cavity. They do not seem to usurp roosting holes from the barbets, for we have frequently watched barbets go to roost and in all cases in which honey- ides had been accompanying a barbet pair, the honeyguides disappeared well before the barbets roosted. We do not know the number of Lesser Honeyguides that may both approach a duetting pair of barbets already “claimed” by another honeyguide and then depart without attracting our attention. We have heard calls of honeyguides at a distance when observing one honeyguide following a barbet pair; but whenever 2 honeyguides were present with a pair of barbets there have been pursuits and apparent or actual conflicts (some conceivably male-female courtship chases), resulting in only one honeyguide being left with the barbets. There also have been 7 instances of a small honeyguide, ly I. narokensis (see Short & Horne 1979), present with a barbet pair and later replaced by a larger, moustached I. minor. Naturally, in some Lesser Honeyguide—barbet pursuits passing out of our view, the returning honeyguide may not have been the same individual. (We have circumstantial indications of this possibility from lengthy pursuits out of our sight accompanied by trilling bursts as if from 2 honeyguides.) Further corroboration of these results is desirable, including their extension to other species of Indicator. The employment of barbet pairs in territorial proclamation, or as an essential element, of a honeyguide territory is intriguing and unique in birds. Indeed, territoriality itself and the pair bond in honeyguides require investigation, especially in view of the lek-like mating of I. minor described by Ranger (1955) in southern Africa, and the fact that such behaviour contrasts strongly with that of host—parasite relations of the parasitic cuckoos (Cweu/us spp.) and cowbirds (e.g. Mal/othrus ater), which we have personally observed, and indigobirds (Payne 1973). Each female Lesser Honeyguide requires several host nests. Possibly a male, defending a territory containing several pairs of barbets, would thus try to ensure that only one egg (fertilised of course by the territorial male) is laid per host nest. (Note that young honeyguides kill other young in a nest, hence 2 honeyguide eggs in a mest would mean the death of one of the cone honeyguides.) Since honeyguides have hosts other than barbets and woodpeckers, one wonders if these other hosts’ nests are used to “dump” eggs when preferred hosts are unavailable. In any event, further data are needed to treat these possibilities. Acknowledgements: We are grateful to A. and P. Donnelly for assistance, to G. R. wan Someren for suggestions and help, and to the authorities of the Kenyan of Tourism and Wildlife for permission to obtain specimens. D. Amadon and R. B. Payne kindly read and commented constructively on the manuscript. The field studies Were supported by the L. C. Sanford Fund and the Ritter-Eisenmann Fund of The ets Micrium of Natural History. [ Bull. Brit.Orn.Cl.1983 103(1)] 32 References: Britton, P. L. & Zimmerman, D. A. 1979. The avifauna of Sokoke Forest, Kenya. Jour, East Afr. Nat. Hist. Soc. & Nat. Mus. No. 169. Friedmann, H. 1955. The honey-guides. U.S. Nat. Mus. Bull. 208. — 1968. Additional data on brood parasites in the honey-guides. Proc. U.S. Nat. Mus. 124: 1-8. Payne, R. B. 1973. Behavior, mimetic song and song dialects and relationships of the parasitic indigobirds (Vidua) of Africa. Orn. Mon. No. 11. Ranger, G. A. 1955. On three species of honey-guide; the Greater (Indicator indicator), the Lesser (/ndicator minor) and the Scaly-throated (Indicator variegatus). Ostrich 26: 70-87. Short, L. L. & Horne, J. F. M. 1979. Vocal display and some interactions of Kenyan honey- guides (Indicatoridae) with barbets (Capitonidae). Amer. Mus. Novit. No. 2684. — 1982. Vocal and other behaviour of Kenyan Black-throated Barbets Lybius torquatus. [bis 124: 27-43. Addresses: L. L. Short, American Museum of Natural History, New York, New York 10024, U.S.A. J. F. M. Horne, P.O. Box 24622, Karen, Nairobi, Kenya. ©British Ornithologists’ Club 1983. Books Received Plint, P. R. & Stewart, P. F. 1983. The Birds of Cyprus. Pp. 182. Maps, appendices. British Ornithologists’ Union, c/o The Zoological Society, Regent’s Park, London NW1 4RY. £12(£10 to members if ordered before end May ’83). This check-list, the sixth in the series of annotated avifaunal lists being published by the BOU, comprehensively covers an island of major importance to Palaearctic/African migrants crossing the eastern Mediterranean, but where even today several million birds are killed annually, mainly illegally. Besides the systematic list covering some 330 species and usefully providing a yardstick for estimating future population changes, there is discussion of past and present Cyprus ornithology, geography, geology, climate, vegetation, migration, breeding and conservation. The appendices include details of ringing recoveries, biometric data and sites of ornithological interest. Dunning, J. S. 1982. South American Land Birds. A photographic aid to identification. Pp. 364. Over 1000 coloured photographs. End paper maps. Harrowood Books, Newton Square, Pennsylvania, USA. $37.50 or$27.50 in paperback. The author, with the expert collaboration of R. S. Ridgely, has produced a book identifying over 2500 South American birds, 1112 of them in clear cut colour portrait photographs, 8-12 to a page, with useful brief notes and tiny distribution maps. It is a remarkable achievement. The technique, developed over 15 years, of trapping birds and then photographing them, when they have regained a relaxed posture, in what amounts to a mini portable studio in the field, has never been exploited on such a scale or so success- fully. The colours in some plates may appear slightly distorted, probably due to techniques and colour reproduction, but they are remarkably few and the resulting portrait gallery is as yet unique. The author’s main concern, in his capacity as a Director of the World Wildlife Fund, is to alert more people to the dangers menacing South American birds and he is generously donating all royalties to the WWF for purchase and protection of their threatened habitats. NOTICE TO CONTRIBUTORS Papers, whether by Club Members or by non-members, should be sent to the Editor, Dr. J. F. Monk, The Glebe Cottage, Goring, Reading RG8 9AP, and ate accepted on the understanding that they are offered solely for publication in the Bulletin. 'They should be typed on one side of the paper, with double- spacing and a wide margin, and submitted with @ duplicate copy on airmail aper. 3 A cientific nomenclature and the style and lay-out of papers and of Refer- ences should conform with usage in this or recent issues of the Ba//etin, unless a departure is explained and justified; but informants of unpublished observations (usually given as 7 Uitt. or pers. comm.) should be cited by initials and name only, e.g. “‘. . . catches wasps (B. Eater)”’, but “B.B.C. Gull informs me that .. .”. Photographic illustrations, although welcome, can only be accepted if the contributor is willing to pay for their reproduction. An author wishing to introduce a new name or describe a new form should append om., gen., sp. or subsp. nov., as appropriate, and set out the supporting evidence under the headings “Description”, “Distribution”, ‘“Type’’, “Measurements of Type” and “Material examined”, plus any others needed. A contributor is entitled to 10 free reprints of the pages of the Bu//etin in which his con- tribution, if one page or more in length, appears. Additional reprints or reprints of contri- butions of less than one page may be ordered when the manuscript is submitted and will be charged for. Authors may be charged for proof corrections for which they are responsible. BACK NUMBERS OF THE BULLETIN Available on application to Dr D. W. Snow, British Museum (Natural History), Tring, Herts HP23 6AP as follows: 1981-2(Vols. 101 & 102) £3 each issue, 1980 (Vol. 100) No. 1 £4, No. 2, 3 & 4 £2 each, 1973-9 (Vols. 93-99) £2 each issue (4 issues per year for Vol. 93 and after), 1969-72 (Vols. 89-92) £1.50 each issue (6 per year), 1929-68 (Vols. 50-88) £1 each issue (generally 9 per year), earlier than Vol. 50 £2 each issue (generally 9 per year); Indices Vol. 7o and after {1 each, Vols. 50-69 £2 each, Vol. 49 and before £4 each. Long runs (at least 10 years) for Vol. 50 and after are available at reduced rates on enquiry. Orders over £50 post free. MEMBERSHIP Only Members of the British Ornithologists’ Union are eligible to join the Club: applications should be sent to the Hon. Treasurer, Mrs. D. Bradley, 53 Osterley Road, Isleworth, Middlesex, together with the current year’s subscription. The remittance and all other payments to the Club should always be in ster/ing unless an addition of {1.00 is made to cover bank charges for exchange, etc. Payment of subscription entitles a Member to receive all Bulletins for the year. Changes of address and revised bankers’ orders or covenants (and any other correspondence concerning Membership) should be sent to the Hon. ‘Treasurer as promptly as possible. SUBSCRIPTION TO BULLETIN The Bulletin may be purchased by non-members on payment of an annual subscription of £15.00 (postage and index free). Send all orders and remit- tances in sterling, unless an addition of £1.00 is made to cover bank charges, to the Assistant Hon. Treasurer, Mr. S. A. H. Statham, 87 Gladstone Road, Watford, Hertfordshire WD1 2RA. Single issues may be obtained as back numbers (see above). CORRESPONDENCE Correspondence about Club meetings and other matters not mentioned above should go to the Hon. Secretary, R. E. F. Peal, 2 Chestnut Lane, Sevenoaks, Kent TN13 3AR. CONTENTS Page Crus Notes ss Si ie is en vee aa I S. N. Stuart. A Pyrenestes from northwestern Tanzania ... ay 3 S. L. Hirty & W. L. Brown. Range extensions of Colombian birds as indicated by the M. A. Carriker Jr. collection at the National Museum of Natural History, Smithsonian Institution 5 J. Fyetpsa. Geographic variation in the Andean Coot Fulica ardesiaca 18 Don Happen. A new species of Thicket Warbler C7ch/ornis (Sylviinae) from Bougainville Island, North Solomons Province, Papua New Guinea... $e a0 Jee as ae a 22 L. L. Shorr & J. F. M. Horne. The relationship of male Lesser Honeyguides Indicator minor with duetting barber pairs... 25 Booxs RECEIVED cs oa ae a a che ve ' 32 It is hoped to send the Bulletin by Bulk Air Mail to all European destinations outside the British Isles and by Accelerated Surface Post to almost every destination outside Europe commencing with Vol. 103 No. 1. This will only apply to copies despatched from the printers on publication. Those whose subscriptions have not been received by the beginning of a month of publication will have their copies despatched by surface mail, after their current subscription has been paid. Published by the BRITISH ORNITHOLOGISTS’ CLUB and printed by The Caxton and Holmesdale Press, 104 London Road Sevenoaks Kent. QL ISSN 0007 - 1595 Bulletin of the British Ornithologists’ Club Edited by Dr. J. F. MONK Volume 103 No. 2 June 1983 FORTHCOMING MEETINGS Tuesday, 5 July 1983, in the Senior Common Room, SHERFIELD BUILDING, Imperial College, S.W.7 at 6.30 p.m. for 7 p.m., Mr Paul Goriup will speak on Bustards. He will be speaking on bustards generally and, more particularly, on species that he has studied in the field, which include the Great Bustard Otis tarda and the Houbara Bustard Chlamydotis undulata. Those wishing to attend should send their acceptance with a cheque for £6.40 a person to reach the Hon. Secretary at 2 Chestnut Lane, Sevenoaks, Kent TN13 3AR (telephone Sevenoaks [0732] 450313) not later than first post on Thursday, 30 June. PLEASE NOTE THAT THIS 1S NOT OUR USUAL VENUES Tae SHERFIELD BUILDING IS ON THE WEST SIDE OF EXHIBITION ROAD IN THE MAIN BLOCK OF IMPERIAL COLLEGE AND LIES A SHORT DISTANCE NW OF THE CARILLON TOWER. Tuesday, 20 September 1983, at Imperial College in the Senior Common Room, SOUTH SIDE, PRINCE’S GARDENS, S.W.7. at 6.30 p.m. for 7 p.m., Mr George A. Smith, Jr., widely known for his work on parrots, will speak on Convergence and Radiation in Parrots. Those wishing to attend should send their acceptance with a cheque for £6.40 a person to reach the Hon. Secre- tary (address above) not later than first post on Thursday, 15 September. THIS IS OUR USUAL VENUE AND IS ON THE EAST SIDE OF EXHIBITION ROAD. Tuesday, 29 November 1983 — 750th MEETING OF THE CLUB. At the same time and place as the previous Meeting, the Survival Anglia film “‘Almost a Dodo” on the Shoebill will be shown. Tuesday, 10 January 1984. Mr M. K. Swales will speak on The Denstone College Expedition to Inaccessible Island (South Atlantic), with an intro- duction by Sir Hugh Elliott. Many copies of the Sw//e/zn must get thrown away annually by Members, copies which the Club would welcome. Please send all unwanted copies, and ask your Executors to do the same, to the Hon. Treasurer at 53 Osterley Road, Isleworth, Middlesex TW7 4PW at any time. Postage will be refunded if requested. COMMITTEE B. Gray (Chairman) Revd. G. K. McCulloch, O.B.E. (Vice-Chairman) R. E. F, Peal (Hon. Secretary) Mrs. D. M. Bradley (Hon. Treasurer) Dr. J. F. Monk (Editor) R. A. N. Croucher P, J. Conder, O.B.E. D. Griffin S. A. H. Statham ©British Ornithologists’ Club 1983. 33 [Bull Brit.Orn.Cl.19 83 103(2)] Bulletin of the BRITISH ORNITHOLOGISTS’ CLUB Vol. 103 No. 2 Published: 20 June 1983 ANNUAL GENERAL MEETING The Annual General Meeting of the British Ornithologists’ Club was held at Imperial College, London, $.W.7 on Tuesday, 17 May 1983 at 6 p.m. with Mr D. R. Calder in the Chair. Thirteen Members were present. The Minutes of the Annual General Meeting held on 18 May 1982 (Bull. Brit. Orn. Cl. 102:43 & xxiv) were approved and signed by the Chairman. The Report of the Committee for 1982 was presented; on the proposal of Mr P. J. Conder, seconded by the Revd. G. K. McCulloch it was approved unanimously that the Report be received and adopted. The Accounts for 1982 were presented by the Hon. Treasurer, who explained the salient points, including the increased excess of income over expenditure. She mentioned that the Investment Reserve of £20 in the previous year’s figures had become unnecessary upon the redemption at par in 1982 of the holding in 84% Treasury Loan 1980/82 and had been credited to Investment Income (General Fund) during 1982. In reply to an enquiry, it was explained that there was no entry in the Balance Sheet for the property “Clovelly” at Tring, as the Club was currently a tenant at a nominal rent but subject to the fulfilment of conditions prescribed in the Will of the late Herbert Stevens. On the proposal of Dr D. W. Snow, seconded by Captain Sir Thomas Barlow, the Accounts for 1982 were unanimously approved. The Editor reported that papers for the Bw//etin were mostly published within six to nine months of acceptance but that he hoped to reduce the waiting time by publication of larger issues as appropriate. There were fewer papers on Africa but more on the Far East and South America. There being no nominations additional to those of the Committee, the following were declared duly elected: Chairman: Mr B. Gray (vice Mr D. R. Calder, who retired on com- pletion of his term of office and was ineligible for re-election). Vice-Chairman: ‘The Revd. G. K. McCulloch, O.B.E. (vice Mr B. Gray, on his election as Chairman). Hon. Treasurer: | Mrs D. M. Bradley (re-elected). Hon. Secretary: Mr R. E. F. Peal (re-elected). Committee: Mr D. Griffin, M.A., and Mr S. A. H. Statham (vice The Revd. G. K. McCulloch on his election as Vice- Chairman and Mr J. G. Parker, who retired by rotation and was ineligible for re-election). The Editor proposed a vote of thanks to the retiring Chairman for all the work he had done for the Club, especially on legal matters concerning “Clovelly”; this was seconded by the Hon. Secretary and carried unan- imously. The Meeting closed at 6.20 p.m. £361 doy galt AddnsPIu4AT KATAVUA VNVIG uoMdtog) FAATVO “AX GIAVG *popzode3 useq DARY qnI) ou OF Sur wor sUOTIOUSULI} [Te FdP2[A\0U INO JO 3s9q 943 03 vy} WAYUOD puL aI”p IY} UO popud Ivak 943 IF JUNODY any pusdxy pur swosuy pur z7g61 soquiszaq] ISIE ye sv JOdYg doUL[Yg poyorae oy DAoTdde 94K sjuvIUNO IY poroyey ANVdWO7) GNV aTUVaS £961 dy yi y($ Sy xoso[ppryy ‘souUTg ‘suoprvy pIOJoIOPY “OsnoyZy pIojorsyzy *YAAoI04} DDU"pIOdI¥ UT or” AOY} IVY) AjIQIDD puL Sn 0} pawuasazd uOILWIOJU! puUe sIOYONOA ‘syOod ay} WOIZ dep IY UO popud 3A dy} JO} JUNODDV ainypusdxy pur aWOduy pue 7g61 soquiadeq ISI ye sv D9Yg adUL]eg PoyorIe oY3 posedosd savy aM ALVOMILYAD SLNVLNNODDY L6S*S¥ ozL‘olFf zLo‘s gSz‘g ———__"" su es 3 “* SLASSY LNYWUND LAN zS$g‘z bEz‘s 1gz‘1 Lzo%z : aS 4 ge ake ** sroupory Arpuns 6bz1 6£6 : sroqiosqng— zzE 89z oye ** siaqlusj\J—9oUvApE UT paAroooz suondyzosqns SHILITIGVIT LNANAND 3 SsaT bz65L o6P'11 L99‘$ 9gl‘L a srs yurg sduravg jeuoneny— be1‘z g09‘z vis Oy crn yunossy swsodoq— LL 989 ts ** juNOdDY OID DO IsOg— Sr 60+ ae J Oe yuNO.V JWorIND—syurg yw Ysvy I I og ‘i ** an[eA [eUIWION]—urjo][Ng JO 49035 SLASSY LNTWUND Shh S44 Py ** 4809 qv Z1/gooz yD0Ig Asnsvary, %$S ogg F INIWLSAANT GNA,J Lsnwy, 0g ——— 610'z oz —_ sas x sie PIEMIOF IYSnNoIQ oAIosoOy ¢ s5aT ool — nas ** 4809 qe zg/0g61 uvo7y Arnsvory, %$g oo1F — 61057 os 4s09 ye $661 y903g sonboyoxy %%o1 101‘z'F INIWLSTANT GNM, TVYANAS —: kq pojuosorday L655 ¥ ozL‘ory ool —_——._ 9S zz — Ool g$1°z a0 Bis Iwok Surlrnp sopes wo sposd.0rg * ppp — ool Te OC oo 1g61 Jaquiadeq ASE ye ooULTLg GNN,j Lsandag SNTAALS Str Shy oe is 20 1g61 Joquis.0q IsI£ We doUL[LG AOVOUT NOLONINUVG “yf “J—ANNY 1sawy, z$o8S ——— 610‘g aa Ivh‘z L96‘z pi ** ganjipusdxy J9AO dUIODUT JO ssooxXg ‘ppE 119‘ z$o%S ons * ie 1g61 Joquiadoq 3s1é ye DoUr[Eg GN, TWUAINaS aif Jk th ‘i 1g61 Zg61 ZQG6I ‘J9quUIDDNqj ISIE 3 Sv JODYS DOULT Eg oov G1¢ z61'F 1£g°¢ zes fob 19z zo1sy Lor‘1 90z 1g61 $60‘1 Lo6‘zF £9S$‘S o£ S*g (96) 17g Z1ze ogz*I 1Sz‘1 . SUNLIGNAdX FY AWAAO AWNOON] AO SsAOxe] ** ganjipusdxg snoourl[aost py om x “+ sodivyD yurg sadivyD Aduejuno.2y pur pny oe "* BuNI9,T JO sootONY 20 os er oourInsuy suoydsjay, AQUOTIeIG pur o8vIsog “BuNUIIgG a ** ginyipusdxg Ajorqng syoquinu yoeq uNey[ng Surjursd-sy ** NILATING dO saNssy LNAWAND HU SLSOT) NOILNGIULSIG GNV NOILVOIIAN UNIN[NG Jo WdvisOg $}sOD sIOYyINY /SsaT ee oe we soqeredog UNe[Ng Jo voHEoyTqng 29 Sunutsg aun LIGNad Xa ae ** oInjyIpuodxg “ a QWIOIUJ—SONI LAA SUMAWAN] HOVG—NILATINg AO sHIVS “M19 JUBUSAO’) JO sp29q GaaTAOOAY XVI, SNOONT ** 9o.uvinsuy pur suedoy :ssaT es Surry, Ayjaaoy>D Ayzodorg GHTAIZOAY LNAY pun,y isn, ai ve ‘+ pun, [eIouey AWOONT LNXALSHANT GAAIFOAY SNOILVNOG s i 3 sroqizosqns as suorjdryosqng sroquiayy CQHAIROAY SNOLLAWMOsSANS AWOONI TQ6I JOquIDDNq ISIE popud JeaA 9Y} JOJJUNOSDSY sINjIpusdxy_ pure swosuy 35 [Bull. Brit.Orn.Cl.19 83 103(2)] The seven hundred and forty-fifth Meeting of the Club was held in the Senior Common Room, South Side, Imperial College, London, S.W.7 on Tuesday, 18 January 1983 at 7 p.m. The attendance was 20 Members and 8 guests. Members ptesent were: B. GRAY (Chairman). Captain Sir THOMAS BARLOW, R.N., P. J. BELMAN, K. F. BETTON, Dr G. BEVEN, Mrs DIANA BRADLEY, P. J. CONDER, R. A. N. CROUCHER, J. H. ELGOOD, D. J. FISHER, A. GIBBS, R. H. KETTLE, J. KING, Dr A. G. KNOX, Revd. G. K. MCCULLOCH, D. G. MEDWAY, Dr J. F. MONK, R. E. F. PEAL, P. S. REDMAN and S. A. H. STATHAM. Guests present were: Mrs B. M. GIBBS, P. J. HAYMAN, Miss PATRICIA C. MED- WAY, Dr AMICIA MELLAND, Mrs DIANA C. MONK, Mtr and Mrs G. H. SEARLE and ROBIN W. WOODS. Mr Robin W. Woods gave a much appreciated address on “Some Birds of the Falkland Islands’’, metioning a number of species with which he had been particularly familiar and especially his ringing studies of Dolphin Gulls Leacophaeus scoresbit. The seven hundred and forty-sixth Meeting of the Club was held in the Senior Common Room, South Side, Imperial College, London, $.W.7 on Tuesday 8 March 1983 at 7 p.m. The attendance was 19 Members and 3 guests. Members present were: D. R. CALDER (Chairman), Major N. A. G. H. BEAL, P. J. BELMAN, K. F. BETTON, Mrs DIANA BRADLEY, P. J. CONDER, R. A. N. CROUCHER, J. H. ELGOOD, D. J. FISHER, B. GRAY, D. GRIFFIN, P. HOGG, Deter NOx, Revd. G. K. McCULLOCH, Dr J. F. MONK, R. E. F. PEAL, P. S. REDMAN, C. E. WHEELER and Lieut.-Col. T. C. WHITE. Guests present were: Dr C. J. CADBURY, Mrs I. MCCULLOCH and N. PICOZZI. Dr C. J. Cadbury gave a stimulating address on “The Restoration of Habitats for Birds” He dealt especially on the need to make the best use possible on land that is available and on what can be achieved in this respect. Bullfinches Pyrrhala pyrrbula and fruit crops The following is a resumé of the talk given to the Club by T. J. Seller on 22 June 1982: Bullfinches Pyrrhula pyrrhula are notorious pests of a wide range of fruit and soft fruit crops grown commercially, as well as of ornamental shrubs and fruit bushes in domestic gardens. They attack fruit buds between late November and April, and can decimate or destroy the potential crop. The timing and extent of damage varies markedly between yeats and some varieties of fruit seem to be more susceptible than others. Pear and apple trees have some degree of tolerance, but plum, gooseberry, black and redcurrant bushes have little and the effects of the damage accumulate. The cost of Bullfinch damage is difficult to estimate, but countrywide it could be in excess of a million pounds. We have been studying a 2 hectare pear orchard in Kent over a number of years, to analyse the course and extent of Bullfinch damage. During the winter of 1978, Bullfinches began taking pear buds in late December and the damage increased rapidly until by April no tree had escaped. Many of the trees had few, if any buds left and the crop picked the following autumn was small. Estimates indicated that in excess of 860,000 buds had been eaten, representing some 92% of those on the trees in mid-December. The financial loss was large, problably well over £2500. Succeeding years showed less dramatic losses of buds and these were not necessarily associated with a loss in the value of the crop picked. Detailed studies of the effects of Bullfinch damage showed it did not always reduce the crop; in some instances there was no effect and it could even be associated with a slight increase. Partly this was because a small proportion of the attacked buds produced some flowers; on heavily damaged trees these formed 60% of flower trusses. More important was that flowers (and buds) surviving on heavily damaged trees set more fruitlets than those on undamaged trees. Following flowering, usually in June when fruitlets were still small, there was typically a large natural drop of excess fruitlets. This was less in years when the number of buds, and consequently the number of fruitlets, was depleted. Thus on undamaged trees, many buds did not produce pears and Bullfinch damage increased the fruit-bearing potential of buds that escaped attack. Another factor was that heavily damaged trees produced larger pears, so a greater percentage of their fruit was of marketable size. Finally, the pear trees studied compensated for the loss of buds in one year by producing more the following year. As a result, their tolerance to attack was greatest in the third year, when a loss of 90% of the buds did not reduce their crop. [ Bull. Brit.Orn.Cl.1983 103(2)] 36 The activities of Bullfinches are not the only reason for crop reduction. The fruit is attacked later in the growing season by other birds such as tits, thrushes, Blackbirds and Starlings. Also, pears are made unfit for marketing by the activities of wasps and other insects, and by bad weather during the growing season; autumnal gales especially can cause significant losses. These factors result in an immediate and irretrievable loss of crop, in direct contrast to Bullfinch damage that may have little or no effect, even at high levels. Bullfinches ate a serious problem to tree and bush fruit growers. However, our work indicated that they may not have the exclusively bad influence that is often suggested. They ate active in orchards at a time when they are easily seen and blamed for crop losses, because there are no leaves on the trees. On the other hand, the damage they cause is only one of a number of factors that reduce the final harvest. Department of Pure and Applied Biology, T. J. Seller and Imperial College, London SW7 2BB. N. J. Matthews. First record of the Sooty Shearwater Puffinus griseus for Arabia by P. R. Colston and M. D. Gallagher Received 7 September 1982 The mostly intact skeletal remains of a medium sized Pufinus shearwater were discovered amongst debris on a tideline of a shelving beach near Azaiba, Batinah, Sultanate of Oman, on the Gulf of Oman, at 23°36’N, 58°20’E, on 23 June 1982 by Wg. Cdr. D. Foster. The specimen was passed to MDG who realised that it was unusual and took it to the British Museum (Natural History) (BMNH) where it is now lodged as a skeleton (BM S/1982-115-1) and where we identified it as a Sooty Shearwater Pufinus griseus. Although one wing was missing and the head detached, the rest of the corpse was apparently complete, still retaining the feathers of the tail and parts of the body. The dark blackish-brown wing showed the silvery-white under-wing pattern characteristic of griseus, and the long slender black bill matched other specimens in the BM. Measurements were: wing 298 mm; tail, strongly rounded with 12 tail feathers, 90 mm; bill (from skull) 52 mm; tarsus 55 mm; length of middle toe 62 mm. The primaries were abraded and the rest of the plumage also showed some degree of wear, so it was therefore probably a full grown adult. This migratory, cold-water species breeds in the sub-antartic around South America, New Zealand and Tasmania, departing between mid-March and May, mostly migrating rapidly northward across the equatorial Pacific and Atlantic Oceans to winter in the northern temperate zones — Bourne (1956) Sea Swallow 9:23-25; Phillips (1963) Ibis 105: 340-353; Cramp & Simmons (1977) Bzrds of the Western Palearctic 1:143-5. Sooty Shearwaters occur at sea south of Kerguelen I. in the southern Indian Ocean (Bourne 1956) and though there had been no records from further north, Bourne pointed out that there had been 2 records of the Short-tailed Shearwater P. tenuirostris accidentally migrating north in the “wrong” (Indian) ocean and that the Sooty Shearwater seemed equally likely to do the same thing (Bourne (1960) Sea Swallow 13:20; (1967) Ibis 109:152). A sighting of a total of 15 Sooty Shearwaters was subsequently reported from the east coast of Sti Lanka in November 1974 “when the size, colour, mode of flight and silvery wing linings allowed positive identification” (Sinclair 1977) /. Bomhay Nat. Hist. Soc. 74: 354). However it is surprising that Sinclair does 37 [ Bull, Brit.Orn.Cl.1983 103(2)] not appear to have identified the Wedge-tailed and Flesh-Footed Shearwaters Puffinus pacificus and P. carneipes which normally pass through Sri Lanka waters at that season, and it may be wondered whether there was some mistake. The present record appears to be the first from Arabia. The specimen’s condition and its position on the beach indicates arrival after the winter storms, in spring, at which time the strong, contrary, northeast monsoon winds of winter would have begun to decline, and when other species which breed in the southern hemisphere, such as Pale-footed Shearwater Puffinus carneipes and Wilson’s Storm Petrel Oceanites oceanicus, begin to move north- wards towards the cool waters of the upwelling off the Kuria Muria islands of Oman. Adresses: P. R. Colston, British Museum (Natural History), Tring, Herts. HP23 6AP. M. D. Gallagher, Oman Natural History Museum, P.O. Box 668, Muscat, Sultanate of Oman. ©British Ornithologists’ Club 1983. Notes on the birds of southwestern Banks Island, Northwest Territories, Canada by D. T. Holyoak Received 30 September 1982 The southwestern area of Banks Island, (c. 72°N, 125°W) arctic Canada, was visited from 30 June to 9 August 1981 with an expedition carrying out geological research. There were frequent opportunities to make ornitho- logical observations. Two of the bird species seen have not previously been recorded from Banks Island, 2 others are little known there and one had not previously been found nesting. This note records these, along with a list of the breeding birds found in the region around Sachs Harbour. PINTAIL Anas acuta On 1 July a female was flushed from a nest containing 5 eggs, on a tundra slope above a small marsh witha pool, c. 2 kmeast of Sachs Harbour. The only other record of Pintail was of 2 males and 5 females seen together near the Kellett River on 3 July. Although there are other summer observations of Pintail on Banks Island this is the first record of breeding; the only other breeding records from the Canadian Arctic Archipelago are from southern Victoria Island (A.O.U. 1957, Godfrey 1966). WHIMBREL Numenius phaeopus On 27 July one was seen flying east along the shore west of Sachs Harbour. There are a few other breeding season records from Banks Island but no proof of breeding (Manning ef a/. 1956, Godfrey 1966). BLACK-LEGGED KITTIWAKE Réssa tridactyla On 28 July a flock of about 65 (with c. 40 adults and c. 25 first-year birds) [ Bull. Brit.Orn.Cl.1983 103(2)| 38 was seen on Cape Kellett. On 3 August a flock of c. 60 (mostly adults) was seen resting on a sand bar in the estuary of the Sachs River at Sachs Harbour. The species has not previously been recorded from Banks Island (the breed- ing record given in A.O.U. 1957 is apparently erroneous). However, there are breeding colonies in northern Alaska and numerous sightings from the southwestern part of the Beaufort Sea (Gabrielson & Lincoln 1959, Frame 1973, Watson & Divoky 1974, Johnson ef a/. 1975) as well as sightings within 150 miles of the North Pole (Godfrey 1966). SABINE’S GULL Larus sabini Several sightings of single birds and groups of up to 6 were made around Sachs Harbour, Cape Kellett and Fish Lake. On 3 July a nest with 3 eggs was found in a shallow pool with emergent grasses and sedges c. 12 miles inland on a low terrace south of the Kellett River; 3 adult birds were present and 2 of them mobbed vigorously when we visited the nest. There are few other breeding records of this species from western Banks Island (Manning et al. 1956, Godfrey 1966). BARN SWALLOW Hirundo rustica One seen about Sachs Harbour repeatedly on 1 July; one seen along shore c. 4 miles west of Sachs Harbour on 9 July. These are the first records from Banks Island. The northern edge of the breeding range is well south of the arctic islands, but there are other records of stragglers from Cambridge Bay, Victoria Island (Godfrey 1966) and one of attempting to breed in northern Alaska (Childs & Maher 1960). OTHER SPECIES Other species recorded within 20 miles of Sachs Harbour (from the Kellett River south to Cape Currie) have all been reported before as breeding on Banks Island (Manning ef a/. 1956, Godfrey 1966). The full list excluding species noted above is as follows (names follow Voous 1973, 1977; breeding was confirmed for species marked*) :— *Yellow-billed Loon Gavia adamsii, *Arctic Loon G. arctica, *Red-throated Loon G. stellata, *Whistling Swan Cygnus columbianus, *Brent Goose Branta bernicla, *Snow Goose Anser caerulescens, Long-tailed Duck Clangula hyemalis, *Eider Somateria mollissima, King Eider S. spectabilis, *Rough-legged Buzzard Buteo lagopus, Gyrfalcon Falco rusticolus, Pere- gtine F. peregrinus, *Willow Grouse Lagopus lagopus, Rock Ptarmigan L. mutus, *Sandhill Crane Grus canadensis, *Semipalmated Plover Charadrius semipalmatus, *Lesser Golden Plover Pluvialis dominica, *Black-bellied Plover P. squatarola, Ruddy Tutnstone Arenaria interpres, *Pectoral Sandpiper Ca/lidris melanotos, *White-rumped Sandpiper C. fuscicollis, *Baird’s Sandpiper C. bairdii, Sanderling C. alba, Grey Phalarope Phalaropus fulicarius, *Pomarine Jaeger Stercorarius pomarinus, Parasitic Jaeger S. parasiticus, Long-tailed Jaeger S. longicaudus, *Glaucous Gull Larus hyperboreus, Thayet’s Gull L. thayeri, *Arctic Tern Sterna paradisaea, *Snowy Owl Nyctea scandiaca, *Horned Lark Eremophila alpestris, *Water Pipit Anthus spinoletta, Snow Bunting Plectrophenax nivalis, *Lapland Longspur Ca/carius lapponicus. Acknowledgements: My visit to Banks Island with an Expedition from the University of Reading was funded by N.E.R.C. and the Royal Society of London; the Polar Continental Shelf Project gave logistic support. Thanks are due to Dr. Peter Worsley for making many of the arrangements for this Expedition and for tolerating much extracurricular ornithology while we were there. References : A.O.U. 1957. Check-list of North American Birds. 5th ed. Baltimore, Maryland: A.O.U. Childs, H. E. Jr. & Maher, W. J. 1960. Nesting attempt by a pair of Barn Swallows in north Alaska. Condor 62: 141-142. 39 [ Bull. Brit.Orn.Cl.19 83 103(2)] Frame, G. W. 1973. Occurrence of birds in the Beaufort Sea, summer 1969. Auk go: 552-563. Gabrielson, I. N. & Lincoln, F. C. 1959. The Birds of Alaska. The Stackpole Co. & Wildlife Management Institute. Godfrey, W. E. 1966. The Birds of Canada. Ottawa: National Museum of Canada Bull. 203. Johnson, S. R., Adams, W. J. & Morrell, M. R. 1975. The birds of the Beaufort Sea. Part II. Observations of 1975 spring migration. Unpublished Rept. prepared under con- tract to the Canadian Wildlife Service, for the Beaufort Sea Project. Manning, T. H., Hohn, E. O. & Macpherson, A. H. 1956. The birds of Banks Island. Nain. Mus. Canada Bull, 143: 1-144. Voous, K. H. 1973, 1977. List of Recent Holarctic bird species. Non-Passerines. [bis 115: 612-638. Passerines. [bis 119: 223-250, 376-406. Watson, G. E. & Divoky, G. J. 1974. Pelagic bird and mammal observations in the western Beaufort Sea, late summer 1971 and 1972. U.S. Coast Guard Oceanogr. Rept. CG-373. Address: Dr. D. T. Holyoak, Department of Geography, University of Reading, 2 Earley Gate, Whiteknights Road, Reading RG6 2AU, England. ©British Ornithologists’ Club 1983. Mass spring migration of European Rollers Coracias garrulus in eastern Tanzania by C.. J. Peare Received 8 September 1982 It is well-known that vast numbers of European Rollers Coracias garrulus winter in the savannah regions of East Africa south of the Sahara (Moreau 1972). Ash & Miskell (1980) recorded a mass migration of this species in southern Somalia in spring 1979, and the obervations reported here indicate that, as expected, such migration is not an isolated event. During a visit to coastal Tanzania in spring 1982 I was able to record the period over which this mass migration—the evacuation of East Africa by Palaearctic migrants— occurred. On 22 March, large numbers of rollers were seen during a drive from Arusha to Muheza; and large numbers were present, especially in Sisal Agave stsalana plantations, on the following 4 days (23-26 March) in the general area between Muheza and Dar-es-Salaam. Sample counts indicated that the migrants outnumbered the indigenous Lilac-breasted Roller C. caudata by overt 50:1, as recorded by Moreau (1972). There was no evidence of mass migration on those days. At about 07.00 on 28 March, large numbers of European Rollers were flying northeast over Muheza. They flew high, at over 300 m, until about 10.00 when heavy rain brought them down to less than 100 m. The northeastward movement was then seen to include European Swallows Hirundo rustica, Striped Swallows H. abyssinica, a few Mosque Swallows H. senegalensis, White- rumped Swifts Apus caffer and a Peregrine Falco peregrinus. The species com- posing this migration were therefore similar to those described by Ash & Miskell (1980). Heavy rain showers continued up to 30 March and the movement of European Rollers continued uninterruptedly on 29 and 30 March; but at Tanga, on the coast, the direction of the movement was more northerly. On a journey from Muheza to Maramba on 31 March the number of European Rollers seen in Sisal plantations was much smaller than had been seen earlier [ Bull. Brit.Orn.Cl.1983 103(2)] 40 and the northeastward movement continued throughout the day. On 1 and 2 April my observations were on the coast, mainly around Pangani, south of Tanga. The rain had stopped and these days were pre- dominantly sunny with light southeast winds. A continuous stream of European Rollers migrated up the coast on both days, accompanied by European Swallows and small parties of unidentified falcons. Although the northeastward movement of rollers was seen inland, there was obviously a concentration of birds migrating up the coast. I did not attempt to estimate numbers, but on 1 and 2 April certainly tens of thousands coasted past and the numbers may have reached 6 figures. On 3 April European Rollers were comparatively scarce on the journey from Pangani to Korogwe with C. garrulus and C. caudata in about equal numbers. In the Korogwe area, about 100 km inland, migration of European Rollers was still observed on 3 and 4 April, but their numbers and concen- tration were much lower than on the coast. Furthermore, the direction of migration at Korogwe was east of northeastward, suggesting that the birds were heading for the coast, possibly to avoid flying over the Usambara mountains. On subsequent journeys in the Arusha, Dodoma and Manyoni areas very few European Rollers were seen. These observations suggest that the mass-migration reported by Ash & Miskell (1980) may be regular. In addition, it appears that the evacuation of the wintering area by European Rollers occurs over a relatively short period, possibly a fortnight or less. Acknowledgements: These observations were made during a visit to Tanzania financed by the Tropical Pesticide Research Institute, Arusha, the British Council, Overseas Development Administration and the Ministry of Agriculture, Fisheries and Food. References: Ash, J. S. & Miskell, J. E. 1980. A mass-migration of Rollers Coracias garrulus in Somalia. Bull. Brit. Orn. Cl. 100: 216-218. Moreau, R. E. 1972. The Palaearctic- African Bird Migration Systems. London: Academic Press. Address: Dr. C. J. Featre, MAFF Worplesdon Laboratory, Tangley Place, Worplesdon, Guildford, Surrey GU3 3LQ, UK. Crown © reserved. Description of the downy young of Lichtenstein’s Sandgrouse Péerocles lichtensteini and the significance of “unpatterned” downy young in the Pteroclididae by David H. Thomas and A. Paul Robin Received 8 September 1982 There does not appear to be any published description of the downy young of Lichtenstein’s Sandgrouse Pverocles lichtensteini (Harrison 1975), although it is hard to believe that such young have not been seen before by ornitho- logists. The following description is offered despite being based on a single individual and because the downy young’s appearance in this and some other sandgrouse species is unusual in the Pteroclididae. The possible significance of this will be discussed. Observations were made in the Moroccan Sahara (where Lichtenstein’s 4I [ Bull. Brit.Orn.Cl.1983 10 3(2)] Sandgrouse is at the northwestern limit of its range) during the course of a study of the sandgrouse there (for a full description of the location and general ecological conditions there see Thomas & Robin 1977). At 08.20 on 28 May 1974, a pair of Lichtenstein’s Sandgrouse with a single downy chick were found among boulders in stoney ground on the lower slopes of Jebel Bani (c. 30°N, 6°W), in the region of Foum Zguid. They were observed at 5-10 m range from a Landrover for about an hour, during which time the female and chick sheltered in the shade of a rock, while the male stood nearby, gular fluttering, in weak sunshine (air temperature 38°C and relative humidity 18°94 1 m above the ground in the shade of the Landrover at 08.55). When later all 3 birds moved off unhurriedly, the chick walked almost under the female or very close to her shaded side, hiding immediately in the shadow of a rock when we followed. The chick was photographed at close range, the following description being made from 2 diapositives (projected) and also from notes made at the time. Colour codes are in Munsell notation (Munsell Color 1975), determined by comparison with the pro- jected diapositive images, which may, of course, have shown some distortion from the true colour of the chick itself. Unfortunately no measurements were made, since we did not appreciate at the time that this plumage was undescribed. DESCRIPTION OF THE DOWNY YOUNG Age. Unknown, but the plumage was entirely downy and showed no out- ward sign of any later feather development. Since it had already left the nest site, the chick was presumably at least one day old, and may have been 2-3 days old. Sandgrouse normally have clutches of 2-3 eggs, which may hatch at one day intervals, and chicks do not normally leave the nest until the last chick is hatched and dried (Maclean 1976). Colouration. Crown and nape, throat, body, wings, thighs and the feathered anterior aspect of the tarsus all of a markedly uniform warm donkey brown, only slightly paler ventrally (7.5 YR 6/4) than dorsally (7.5YR 5/4). Lores, supercilium, ear coverts and below the eye a somewhat darker brown (light chocolate: 7.5YR 3.5/4), demarcated below by a short pale moustachial- stripe and above by a long pale lateral-crown stripe running from the bill to the back of the head, the pale stripes irregular in width and alignment. A pale spot just below and behind the eye was the same creamy buff (7.5 YR 8/4) as the other 2 pale strips (see Fig. 1). The texture of the down appeared very uniform, tather like dense velvet. The bare parts (bill, a narrow eye ring, toes and the posterior aspect of the tarsus) were light grey (5 YR 6/1). DISCUSSION The unusual feature of this downy plumage in comparison with that of many other sandgrouse species is the uniform colouration. Typically, downy young of sandgrouse are marked disruptively with mottled browns, buffs, black and white, more or less organised into symmetrically placed darker panels outlined and separated by paler lines, particularly on the head and _ dorsal aspect of the body. Specifically, this generalised description can be applied to the following species: Syrrhaptes paradoxus, S. tibetanus (Fjeldsaa 1976, 1977); Pterocles namaqua, P. alchata, P. quadricinctus (Fjeldsaa 1976); P. orientalis (Fjeldsaa 1977); P. [ Bull. Brit.Orn.Cl.1983 103(2)] 42 Figure 1. View of the downy young of Lichtenstein’s Sandgrouse Prerocles lichtensteini, showing the pattern of pale lateral-crown and moustachial stripes, the position of the pale spot below and behind the eye, the posterior margin of the darker facial colouring (dotted line between the pale stripes) and the otherwise unpatterned plumage. (Traced from a photograph; the bird was straddling a pebble, which obscured some plumage details.) exustus (Aldrich 1943, Harrison 1975); P. burchelli (DHT’s unpublished photographs of a chick collected by Mr. J. E. W. Dixon, Department of Zoology, University of Capetown, Rondebosch, S. Africa); P. decoratus, P. gutturalis, P. bicinctus (Mackworth-Praed & Grant 1952). This list includes all members of the family (see Hiie & Etchécopar 1957) except the Madagascar Sandgrouse P. personatus (for whose downy young no description was found), Lichtenstein’s Sandgrouse (in which the absence of patterning on the body has been noted already), and P. coronatus, P. sene- gallus and P. indicus, of which 3 species Fjeldsaa (1976: 213) comments that the downy young are “faintly marked . . . but the course of the light lines is nearly the same in all’. Of these last 3 species, P. indicus is probably closely related to P. /ichtensteini (and is possibly conspecific—Meinertzhagen 1954), and at least one description of the downy young of P. indicus (as “uniform earthy brown’”—Ali & Ripley 1969:93) seems to conform to the present one of P. “Uichtensteint. Colour photographs of P. coronatus downy young (George 1978: plate 32) shows them to be not unlike P. Achtensteini chicks in the comparative uniformity of general colouration and with a darker facial patch, while P. senegallus is slightly more patterned than coronatus or Lichten- steini (George 1978: plates 29-31, Mackworth-Praed & Grant 1952). The slight or absent patterning in downy young of Péerocles indicus, lichten- steini, coronatus and senegallus may represent convergent adaptations to extreme desert conditions rather than any phyletically close relationship. There has not been a recent evaluation of relationships within the sandgrouse family, but Bowen (1927) showed the existence of 2 distinct species groups in Pierocles (based on plumage characteristics and drinking behaviour): on these criteria, /ichtensteini seems close to indicus, bicinctus and quadricinctus, yet the latter 2 species have strongly patterned downy young. Similarly, coronatus and senegallus also seem otherwise phyletically close to species with more or less strongly patterned young (P. alchata, orientalis, namaqua, exustus, gutturalis and burchellii). Elsewhere we have shown that P. “ichtensteini, coronatus and senegallus show matked adaptations for extreme desert conditions (Thomas 43 (Bull. Brit.Orn.Cl.1983 103(2)] & Robin 1977), and zudicus also is found in arid areas (Ali & Ripley 1969). Thus, it seems that the loss of the “‘typical’’ family patterning of downy young and adoption of more uniform colouration has evolved at least twice in sandgrouse, in association with existence in extreme deserts: once in the P. lichtensteinilindicus species group and again in the P. coronatus|senegallus group. Interestingly, P. namaqua, bicinctus and burchellii have “typically” patterned downy young, yet survive well in the Namib and/or Kalahari deserts (Maclean 1968, Thomas & Maclean 1981). However, these 3 southern African species experience less extreme climatic conditions, and live to a greater extent in savanna (as well as desert) compared to the 4 Saharan/ Arabian/Indian species, coronatus, senegallus, lichtensteini and indicus (Thomas & Maclean 1981, Thomas ef a/. 1981), and so selection for “unpatterned”’ downy young may not have been so strong in the southern African species. Acknowledgements. Research was done with the permission of the Moroccan Government and with the help of the Moroccan Minister for Waters and Forests and of the Governor of the Region of Ouarzazate. DHT received grants from the Royal Society of London, the Science Research Council (B/RG/22939) and University College, Cardiff. References: Aldrich, H. C. 1943. Some notes on the Common Sandgrouse (Prerocles exustus Temminck) in Kaira District. J. Bombay Nat. Hist. Soc. 44: 123-125. Ali, S. & Ripley, S. D. 1969. Handbook of the Birds of India and Pakistan. Vol. 3. Oxford University Press: Bombay. Bowen, W. W. 1927. Remarks on the classification of the Pteroclididae. Am. Mus. Novit. 2732112. Fjeldsaa, J. 1976. The systematic affinities of sandgrouse, Pteroclididae. Widensk. Meddelr Dansk Naturhist. Foren. 139: 179-243. Fjeldsaa, J. 1977. Guide to the Young of European Precocial Birds. Skarv Nature Publications: Tisvildeleje, Denmark. George, U. 1978. In the Deserts of this Earth. Hamilton: London. Harrison, C. J. O. 1975. A Field Guide to the Nests, Eggs, and Nestlings of British and European Birds. Collins: London. Hie, F. & Etchécopar, R. D. 1957. Les Ptéroclididés. L’Oiseau et R.F.O. 27: 35-58. Mackworth-Praed, C. W. & Grant, C. H. B. 1952. Birds of Eastern and North Eastern Africa. Vol. 1. Longman: London. Maclean, G. L. 1968. Field studies on the sandgrouse of the Kalahari Desert. Living Bird. 7: 209-235. Maclean, G. L. 1976. Adaptations of sandgrouse for life in arid lands. Proc. Int. Orn. Congr. 16: 502-5 16. Meinertzhagen, R. 1954. Birds of Arabia, Oliver & Boyd: Edinburgh. Munsell Color. 1975. Munsell Soil Color Charts. Munsell Color: Baltimore. Thomas, D. H. & Maclean, G. L. 1981. Comparison of physiological and behavioural thermoregulation and osmoregulation in two sympatric sandgrouse species (Aves: Pteroclididae). J. Arid Environments 4: 335-358. Thomas, D. H., Maclean, G. L. & Clinning, C. F. 1981. Daily patterns of behaviour com- pared between two sandgrouse species (Aves: Pteroclididae) in captivity. Madoqua 12: 187-198. Thomas, D. H. & Robin, A. P. 1977. Comparative studies of thermoregulatory and osmo- regulatory behaviour and physiology of five species of sandgrouse (Aves: Ptero- clididae) in Morocco. J. Zool. Lond. 183: 229-249. Address: Dr. D. H. Thomas, Department of Zoology, University College (University of Wales), Cardiff CF1 1XL, U.K. Dr. A. P. Robin, Centre Hospitalier, Morlaix-Cedex, 29205 France. [Bull Brit.Orn.Cl.1983 103(2) | 44 Birds in the Brak and Sabha regions of central Libya, 1981-82 by P. J. Cowan Received 6 September 1982 In his check-list The Birds of Libya, Bundy (1976) collated separately records of birds seen in the Fezzan region of Libya (south of 30°N, west of 19°E) on the basis of the few published papers. I reported bird records for the Brak and Sabha regions (27°N, 14°E) of the Fezzan for September 1980 to June 1981 (Cowan 1982). The present paper reports my continued observations in the Brak and Sabha regions, from 28 August 1981 to 24 June 1982. Acquisition of a car considerably improved my exploration of the area and, for example, allowed discovery of a further pool at Ashkidah which pro- vided the Ashkidah breeding records in this paper. I had first visited the lake at Sabha in late March 1981. Subsequently I was able to visit the lake during autumn and winter also. Sites mentioned are shown on the map in Cowan (1982) though I have changed the spelling of Maharouga to Al Mahruqah (National Atlas 1978). Al Mahrugah, Agar, Brak and Ashkidah are settlements in the Wadi ash Shati. South of the Shati, across an arm of the Awbari Sandsea, is the town of Sabha. West of Sabha, in the Awbari Sandsea, is Gabroan village. Wetland areas are present in the vicinity of these settlements and include pools or small lakes. In the following list the sequence of species and checklist number of Bundy (1976) are used. 3. Podiceps nigricollis Black-necked Grebe. Noted at Sabha on 2 dates in Sep, one in Oct, one in Nov, 2 in Jan, one in Feb and one in Mar, 2-5 birds, but 10 in Feb. Also at Sabha, 6 grebes, presumably this species, once in Dec. First records of any grebe in the Fezzan. 11. Phalacrocorax carbo Cormorant. One at Brak, 19 Oct. First Fezzan record. 12. Ardea cinerea Grey Heron. Maximum seen c. 75 at Sabha on 1 Oct, but otherwise, and apparently in past (Bundy 1976), only seen in small numbers. 15. Botaurus stellaris Bittern. One Sabha 24 June. Previously recorded Apr and Oct (Bundy 1976). 18. Egretta garzetta Little Egret. Maximum seen, 28 at Sabha on 22 Sep. In autumn, only singles mentioned by Bundy (1976) although a similar maximum was recorded in autumn by Cowan (1982). 20. Nycticorax nycticorax Night Heron. Maximum, c. 25 at Sabha on 1 Oct. Apparently previously recorded only in small numbers in autumn. 23. Platalea leucorodia Spoonbill. Singles at Sabha, 21 and 22 Sep, 1 Oct, 8 Nov and 19 Apr and at Ashkidah 17, 23 and 27 Apr. Two previous Fezzan records. 24. Plegadis falcinellus Glossy Ibis. Records include 21 at Sabha on 21 Sep. Previously apparently only singles in autumn. 26. Anas acuta Pintail. Seen on 28 dates Sep-June, maximum c. 50 at Sabha, 23 Oct and at Ashkidah, 30 Oct. Maximum for the 5 dates Dec—Jan, c. 20 at Brak, 28 Jan. Bundy (1976) mentioned small numbers Oct—Nov, Feb—Apr. 29. Anas clypeata Shoveler. Recorded Sep—Dec on 6 dates, maximum 10 45 Bull. Brit.Orn.Cl.1983 103(2)] at Ashkidah on 16 Nov and Mar-Apr on 7 dates, maximum c. 15 at Ashkidah 5 Apr. Also one at Sabha, 24 June. Only one previous Fezzan record. 30. Anas crecca Teal. Recorded on 5 dates, Dec-Mar. Maximum, c. 25 at Sabha, 21 Jan. Previously recorded Sep-Nov only, apparently singles (Bundy 1976). 33. Anas querquedula Garganey. Recorded Sep—Nov on 4 dates and Feb- June on 14 dates. Maximum, 20 at Brak, 19 Sep. Bundy (1976) mentions records on 2 autumn dates, Cowan (1982) on 2 spring dates, involving small numbers. 37. Aythya ferina Pochard. One at Sabha, 23 Oct. First Fezzan record. 38. Aythya fuligula Tufted Duck. One at Sabha, 8 Nov. First Fezzan record. 39. Aythya nyroca Ferruginous Duck. Recorded Sep—June on 22 dates, maximum c. 100 at Sabha, 23 Oct. Jan maximum, 24 at Sabha, 6 Jan. June maximum, 28 (excluding downy young) at Ashkidah, 13 June. Previously only recorded Oct-Nov (Bundy 1976). Breeding records (at Ashkidah): several sightings in June of 2 females each rearing a single brood. When last seen brood size down to 8 and 5 downy young respectively. First Fezzan breeding records. 58. Circus aeruginosus Marsh Harrier. Recorded Sep—June on 31 dates, maximum 4 at Sabha, 8 Mar. Includes 2 dates in Dec, 3 in Jan and 2 in Feb. First January records for the Fezzan. 64. Hieraetus pennatus Booted Eagle. One Gabroan, 21 May. 3rd Fezzan record. 74. Falco concolor Sooty Falcon. One Sabha, 24 June. 4th Fezzan record. 81. Falco vespertinus Red-footed Falcon. One Sabha, 10 May. First Fezzan record. 85. Grus grus Crane. One at Ashkidah, 5 Dec. First Fezzan record. 87. Fulica atra Coot. Recorded Sep—June on 31 dates, maximum c. 140 at Sabha on 18 Feb. Autumn maximum, 72 at Sabha, 8 Nov. Includes 5 dates Dec—Jan, maximum 45 at Sabha, 6 Jan. Apr—June maximum excluding downy young juveniles, c. 60 at Ashkidah 17 Apr, 29 May and 7 June. Bundy (1976) mentioned only singles Oct-Nov, while Cowan (1982) had 2 spring dates involving only a few birds. Breeding records: a small number of broods reared at Ashkidah and apparently also at Sabha. Sightings of broods of downy young at Ashkidah late May to 19 June, maximum c. 4 broods on 29 May. Juveniles present Ashkidah in June, maximum of 6 seen on 7 June. 7 juveniles seen at Sabha 24 June. First Fezzan breeding records. 88. Gallinula chloropus Moorhen. Recorded Sep—June. Maximum adults seen 22, Ashkidah 19 June. Breeding records: at Ashkidah, one brood of downy young 29 May and one, apparently a different brood, 5 June. Juveniles at Ashkidah, Brak, Al Mahruqah and Sabha in June, maximum 7 Ashkidah 19 June. Breeding colony previously reported from Sabha (Bundy 1976). Probably a resident breeder at all suitable localities in west central Libya. 96. Charadrius alexandrinus Kentish Plover. At Sabha, singles on 15 Dec and 6 Jan. At Ashkidah seen Apr—June on 13 dates, maximum 4 (excluding downy young). Previously, singles on 5 dates in autumn (Bundy 1976). Breeding records (at Ashkidah): sightings involving at least 3 different broods. Observations of downy young Apr and June, including 2 broods of 2 downy young each on 19 June; one juvenile 2 May; nest with 3 eggs 5 June. First breeding records for the Fezzan. [ Ball. Brit.Orn.Cl.19 83 103(2)] 46 97. Charadrius dubius Little Ringed Plover. One on 22 Sep, 2 on 1 Oct and 2 on 21 Jan at Sabha. Also recorded Mar—Apr. First autumn and winter records for the Fezzan. One unseparated dubius|hiaticula at Sabha, 6 Jan. 98. Charadrius hiaticula Ringed Plover. 3 at Brak on 11 Sep, 3 at Sabha on 1 Oct and one at Ashkidah on 5 Nov and 14 May. First Fezzan records. 102. Pluvialis squatarola Grey Plover. One at Ashkidah, 23 Apr. First Fezzan record. 103. Vanellus vanellus Lapwing. One at Ashkidah, 5 Dec. Previously recorded by Cowan (1982) only. 105. Calidris alba Sanderling. At Ashkidah, one on 2 May and 2 on 24 May. Previously recorded on only 2 dates (Bundy 1976). 106. Calidris alpina Dunlin. 2 at Ashkidah, 5 Nov and one at Sabha, 8 Nov. First Fezzan records. 108. Calidris ferruginea Curlew Sandpiper. At Ashkidah, 3 on 27 Apr, one on 14 May and 2 on 24 May. 6 at Sabha 10 May and one at Gabroan 21 May. Only one previous Fezzan record. 110. Calidris minuta Little Stint. Recorded Sep—June including, at Sabha, c. 15 on 15 Dec, 7 on 6 Jan, c. 20 on 21 Jan and c. 15 on 18 Feb, the first winter records for the Fezzan. 111. Calidris temminckii Temminck’s Stint. One at Sabha, 22 Sep. First autumn record. 112. Gallinago gallinago Snipe. 2 on 22 Sep and one on 1 Oct at Sabha. Also recorded in Mar. An unseparated snipe sp. at Brak, 8 Dec. One previous autumn record. 116. Limosa limosa Black-tailed Godwit. Singles at Sabha on 21, 22 Sep and 1 Oct and at Ashkidah on 23, 27 Apr and 5 June. Two previous Fezzan records (Bundy 1976, Cowan 1982). 121. Philomachus pugnax Ruff. Recorded Sep-Oct and Mar—June. Also, one at Sabha on 15 Dec, the first Fezzan winter record. 123. Lringa erythropus Spotted Redshank. Recorded on 7 dates Sep—Nov, maximum 4 at Sabha on 1 Oct. Previously, records of 5 birds (Bundy 1976). 124. Tringa glareola Wood Sandpiper. One at Brak, 27 Sep. The second autumn record for Fezzan. Also recorded Apr and May. 126. Tringa nebularia Greenshank. One at Ashkidah on 30 Oct. Also recorded Apr, May. In autumn, previously, singles on one date only, at Sabha (Bundy 1976). 127. Iringa ochropus Green Sandpiper. Recorded Aug—Apr including, at Brak, singles on 8, 21, 24 Dec, 4, 14, 19 Jan and, at Sabha, singles on 15 Dec, 21 Jan. Winter records were first reported by Cowan (1982). 128. Tringa stagnatilis Marsh Sandpiper. One at Brak, 27 Sep. Also recorded Apr, May. Only one previous autumn record. 129. Tringa totanus Redshank. One at Sabha 22 Sep, one at Brak 27 Sep, 9 Oct. Also Mar, May. First autumn records for Fezzan. 130. Himantopus himantopus Black-winged Stilt. One at Sabha, 1 Oct. Recorded on 13 dates Mar—June, maximum 14 at Ashkidah 2 May. No previous autumn records. Previous maxima, 2 (Bundy 1976), 17 (Cowan 1982). “ 34. Cursorius cursor Cream-coloured Courser. One at Agar 22 Feb and 4 at Sabha 8 Mar. Bundy (1976) mentioned the uncertain Fezzan status of this species, previous records having been Oct-Nov and Feb—May. 47 [ Bull, Brit.Orn.Cl.1983 103(2)| 136. Glareola pratincola Collared Pratincole. 3 at Ashkidah, 24 May. In same flock were a further 3 pratincoles whose underwing colour was not de- termined. First Fezzan records of pratincoles. 143. Larus fuscus Lesser Black-backed Gull. One (L. f. fuscus) at Ashkidah on 24 and 29 May. No previous Fezzan records. 147. Larus ridibundus Black-headed Gull. At Sabha: one on 1, 23 Oct, 8 Nov; c. 50 on 15 Dec; 40, 6 Jan; c. 30, 21 Jan; c. 20, 18 Feb; 6, 8 Mar; one, 19 Apr. At Ashkidah, one on 28 Jan and 5 Apr. One at Brak, 19 Mar. 3 gulls, presumably this species, Ashkidah, 17 Apr. First recorded Cowan (1982), maximum 16. 148. Chlidonias hybrida Whiskered Tern. At Sabha, one on 1 Oct, 4 on 10 May. At Ashkidah, one on 24 May, 2 on 29 May. First Fezzan records. 149. Chlidonias leucopterus White-winged Black Tern. One at Sabha, 1 Oct. Second autumn record for the Fezzan. Also, at Ashkidah, 5 on 23 Apr, 3 on 27 Apr, 2 on 24 May and one on 13 June. 7 at Sabha, 10 May. 150. Chlidonias niger Black Tern. 4 at Ashkidah, 23 Apr. Previously, 5 birds recorded (Bundy 1976). 151. Gelochelidon nilotica Gull-billed Tern. One at Sabha, 1 Oct. Second Fezzan record. 164. Streptopelia senegalensis Palm Dove. Recorded Oct—June on 19 dates. Seen at Ashkidah, Brak, Sabha and Al Mahruqah. Probably a resident breeder. First recorded by Cowan (1982). 207. Riparia riparia Sand Martin. Recorded Sep—Oct on 8 dates. Also seen Mar—June. First recorded in autumn by Cowan (1982). 215. Motacilla cinerea Gtey Waytail. At Brak, one on 8 Dec and 27 Feb. Previously, singles on 4 dates (Bundy 1976, Cowan 1982). 251. Syia cantillans Subalpine Warbler. At Brak, one on 29 Aug, 23 Nov and 4 Jan. One at Al Mahruqah 29 Mar. Bundy (1976) gave one and Cowan (1982) 3 winter dates. 282. Saxicola rubetra Whinchat. Though recorded Apr—May as expected (Bundy 1976), no autumn records (cf. Cowan 1982). 283. Saxicola torquata Stonechat. One at Brak, 29 Nov. 3rd Fezzan record. 307. Passer hispaniolensis Spanish Sparrow. At Brak, 28 Oct-23 Mar with a maximum of c. 1000 on 31 Oct. Cowan (1982) recorded it early Nov to Mar though previously not recorded in Fezzan before Jan (Bundy 1976). I can add the following species to the list in Cowan (1982) of species seen by myself of which the Fezzan status is reasonably certain (Bundy 1976):— 21. Ciconia ciconia; 61. Circus pygargus; 79. Falco subbuteo; 161. Pterocles sene- gallus; 177/179. Apus apus|pallidus (anseparated); 192. Calendrella cinerea; 203. Hirundo daurica; 225. Ficedula albicollis (not semitorquata); 226. Ficedula hypoleuca; 252. Sylvia communis; 281. Phoenicurus phoenicurus. References: Bundy, G. 1976. The Birds of Libya. British Ornithologists’ Union: London. Cowan, P. J. 1982. Birds in west central Libya, 1980-81. Bu//. Brit. Orn. Cl. 102: 32-35. National Atlas of the Socialist People’s Libyan Arab Jamahiriya. 1978. Surveying Department, Secretariat of Planning, Tripoli. Arabic/English edition. Address: Dt. P. J. Cowan, Higher Institute of Technology, P.O. Box 68, Brak, Socialist People’s Libyan Atab Jamahiriya. © British Ornithologists’ Club 1983. [ Bull. Brit.Orn.Cl.19 83 103(2)] 48 On the nominate race of Césticola fulvicapilla (Vieillot), 1817 by P. A. Clancey Received 25 August 1982 Sylia fulvicapilla Vieillot, 1817, is based on the La Fauvette Rousse téte of Levaillant (1802), who encountered the species in the Cape to the south of the Orange R. and specifically near Old Camdeboo, close to the present town of Graaff-Reinet in the east of the Province at 32° 12’S, 24° 32’E, which is the accepted type-locality of the nominate race. In his major review of the genus Cisticola, Lynes (1930) gave the range of C. f. fulvicapilla as the Cape (except for ecologically unsuitable karoo country and the southwest) and Natal. Interestingly enough, Graaff-Reinet lies in the east of the Karoo and is well outside the currently established range of the species in the eastern Cape, the valley of the Great Fish R. being the western distributional limit in this sector. In my revision of C. fulvicapilla (Clancey 1971), I refined the range of the nominate race to read “‘southern and eastern Cape, east from about George, Lesotho, Natal and Zululand, and, probably, western Swaziland’—this being adopted in the recent S.A.O.S. Checklist (Clancey 1980). In this revision it was pointed out that the population occurring on the periphery of the massif of Lesotho differs from the norm of nominate fu/vicapilla in its greater size. Quickelberge (1972), in his report on 2 ornithological expedi- tions to the highlands of Lesotho, comments on the same 4 adult males from Moletsane (a remote outpost in northern Lesotho on the Maluti escarpment between Mamathes and Mapoteng at 6200 ft) which were commented on by Clancey in 1971, as demonstrating a redder, less brownish, pileum and lighter mantle when compared with the bulk of referred populations of nominate fulvicapilla. Two adult males taken in early August 1982 near the village of Rhodes (30° 48’S, 27° 28’E), and now in the Durban Museum, show similar characters to those established for Lesotho birds by Quickel- berge. While no specimens from anywhere near the type-locality of C. f. fulvicapilla exist in museum collections, material is available from the karoid country to the east of the mid- and upper valley of the Great Fish R., northeast to the versant of the seaward facing escarpment of the Drakensberg Range and Lesotho. This karoid element of the present nominate race of C. fu/vicapilla is clearly different from the populations of the southern and eastern Cape to Natal and western Zululand currently associated with them. No name in synonymy is available for such populations. Hartlaub’s Camaroptera natalensis of 1863 (proposed in Gurney, Ibis 1863: 323, pl. viii, Fig. 1), described on an Ayres skin from Pinetown, Natal, is pre-occupied in the genus Czsticola by Drymoica natalensis A. Smith, 1843: Durban, Natal. No name being available, the innominate populations may be known as Cisticola fulvicapilla dumicola subsp. nov. Type: 3, adult, Inanda, north of Durban, Natal, South Africa. Collected 8 May 1955 by Durban Museum personnel. In the collection of the Durban Museum, D. M. Reg. No. 2677. 49 [ Bull. Brit.Orn.Cl.1983 103(2)) Description: Differs from C. f. fulvicapilla of the karoid regions of the interior of the eastern Cape to the east of the mid- and upper valley of the Great Fish R., northeast to the Drakensberg escarpment and the Maluti of Lesotho, in having the vertex in moderately worn non-breeding dress (August) Argus Brown (Ridgway 1912) versus Dresden Brown, and with the dorsum deep Saccardo’s Umber, the red-brown of the head-top diffused as a wash over the mantle and scapulars, and the tail rather redder; in worn breeding dress (from November) with the vertex darker, more chestnut and less ochraceous, and with the back rather blacker. The ventral surface in both plumages is about the same. The size ranges smaller than nominate fulicapilla: wings in $3 48.5-51.5 (in nominate fu/vicapilla to 54) mm. Material examined: 50 specimens. Range: Moist coastal regions of the southern and southeastern Cape from the George/Knysna region to coastal Transkei, Griqualand East, Natal (except Drakensberg escarpment country of extreme west), and western Zululand. Measurements of the Type: Wing (flattened) 49.5, culmen from base 13, tail 40 mm. Remarks: The name chosen is from the Latin dumico/a: inhabiting thickets or coppices. With the description of C. f. dumicola, the range of nominate C. fulvicapilla will stand as outlined in the above “Description’’. References: Clancey, P. A. 1971. Miscellaneous taxonomic notes on African birds XXXII. Durban Mus. Novit. ix (5): 51-57 Clancey, P. A. (Ed.). 1980. 5.A.O.S. Checklist of Southern African Birds. p. 218. Southern African Ornithological Society: Johannesburg. Levaillant, F. 1802. Histoire Naturelle des Oiseaux d’ Afrique. Vol. iii: 98, plate 124. Paris. Lynes, H. 1930. Review of the genus Cisticola. Ibis Suppl. Ser 12 (6): 503-517. Quickelberge, C. D. 1972. Results of two ornithological expeditions to Lesotho. Durban Mus. Novit. 1x (17): 268. Ridgway, R. 1912. Color Standards and Color Nomenclature. The author: Washington, D.C. Address: Dt. P. A. Clancey, Durban Museum, P.O. Box 4085, Durban 4000, South Africa. ©British Ornithologists’ Club 1983 A new subspecies of the Usambara Weaver Ploceus nicolli by N.—E. Franzmann Received 14 August 1982 The affinity of the 2 East African weavers Ploceus olivaceiceps and P. nicolli has been disputed for some time. Both forms were originally described as separate species and were also treated as such by Mackworth-Praed & Grant (1960). It was apparently Moreau (1960: 465) who first suggested that olivaceiceps and nicolli were conspecific, repeating this in Peters’ (1962: 56), but later (1966: 88) changing his mind. Hall & Moreau (1970: 292) con- sidered nicol/i “at least an incipient species”. Britton (1980: 212) also treats nicolli as a colour form of o/waceiceps. [ Bull. Brit.Orn.Cl.1983 103(2)] 50 While working on a large collection of skins collected in Tanzania by Thorkild Andersen between 1947 and 1965, a specimen clearly belonging to this weaver complex, but different from the forms hitherto described, was discovered. For the reasons given below, this finding leads me to believe that P. o/ivaceiceps and P. nicolli are best treated as separate species. I propose the new form be considered a subspecies of P. nicolli, as follows: Ploceus nicolli anderseni subsp. nov. Type. Adult 3 collected 23 May 1952 by Thorkild Andersen in the Uluguru Mts., Tanzania, elevation 1500-1800 m. Collector’s No. 144. Type at the Zoological Museum, Copenhagen, Denmark. No. 10.5. 1982: 1. Description. The male differs from nominate xico//i in having forehead, crown and nape black with a sepia brown cast. There is no trace of yellow, not even on the forehead. The throat is very dark, almost black, as opposed to dark brown. This dark area is also more extensive in anderseni than in the nominate form, while the chestnut chest-band is narrower in anderseni. The feet are light brown, the bill black and the iris brown. The female anderseni differs from nominate females in having the throat darker brown. The crown of the female anderseni is very dark brown, almost black, whereas the crown colour of nominate females varies from dusky brown to very dark brown. Measurements of type (mm). Wing (chord) 88, tail 48, tarsus 23.3 and culmen from base 17.3. Range. The Uluguru Mountains (7° 10'S, 37° 40’E), Morogoro District, Morogoro Region, Tanzania above 1400 m. Sight records from the Mwani- hana Forest (7° 45'S, 36° 50’E), Uzungwa Mountains, Kilombero District, Morogoro Region can also be assigned to this subspecies. Specimens examined. Ploceus olivaceiceps (13). Mzimba, Malawi 1 g, 1 2 British Museum (Natural History) (BMNH); Kapiriuta, Malawi 1 9 (BMNH); Michiru Hill, Malawi 1 9 (BMNH); Unangu, Mozambique 1 ¢ (BMNH); Furancungo, Mozambique 3 gd, 3 929 (BMNH); Songea, Tanzania 1 g, 1 9 (Zoological Museum, Copenhagen) (ZMC). P. 2. nicolli (13). Amani, Tanzania 2 gg, 1 92 including type (BMNH); Lushoto, Tanzania 1 9? (BMNH); Shume, Manolo & Lushoto, West Usam- bara, Tanzania 3 $d, 2 99, 1 dimm, 3 99 imm (Peabody Museum of Natural History) (PMNH). P. n. anderseni. (2) Uluguru Mts, Tanzania 1 ¢ type (ZMC), 1 9 (PMNH). Remarks. The only 2 specimen of anderseni was collected in the Uluguru Mountains by G. H. Heinrich at 1600 m. on 10 December 1961 (Ripley & Heinrich 1966). I consider P. nicolli and P. olivaceiceps to be separate species for the following 2 reasons: first, although anderseni is much closer to olivaceiceps geogtaphically, it is even less similar to it than is nominate wicoli1. In the male o/ivaceiceps there is a distinct yellow forehead. To a varying degree this is also seen in nominate zico//i, but there is no trace of this character in anderseni. P. n. anderseni is generally darker around the head and throat than nominate nico/li, whereas olivaceiceps is much paler; secondly both nominate nicolli and anderseni ate inhabitants of the canopy of high mountain rain forest, whilst o/vaceiceps is a denizen of Brachystegia woodland (Stuart & van der Willigen 1978). I do not doubt the common ancestry of the 2 species, 51 (Bull. Brit.Orn.Cl.19 83 103(2)] but both have clearly diverged very considerably, both in plumage charac- teristics and habitat requirements. They are best regarded as forming a superspecies in the sense of Hall & Moreau (1970). Ploceus nicolli is a rather variable species. In the nominate subspecies, the forehead of the male is always dull yellow, but this may sometimes extend over the crown and nape. This was noted by Ripley & Heinrich (1966), who also pointed out the variability in the female head colour from dusky brown to black. With only 2 specimens of anderseni it is not possible to comment on such variability except to note that both Uluguru specimens have plumage characteristics well outside the limits of variation shown by nominate nicol/. Ploceus nicolli has recently been seen in the Mwanihana Forest, on the eastern scarp of the Uzungwa Mountains in eastern Tanzania. This is a new locality for the species. Very good field observations were obtained by S. N. Stuart and Miss T. A. van der Willigen at 1150-1200 m, both of whom noted the very dark heads and absence of yellow forehead. I therefore ascribe these birds provisionally to anderseni. In addition to the 2 specimens of anderseni collected in the Ulugurus, a field observation of one individual was recently made in Kinole Forest (6° 53’S, 37° 44’E) at 1350 m (Stuart & Jensen 1981) on the eastern side of the mountains. This subspecies is named after the collector, the late Thorkild Andersen, whose collection of birds from Tanzania is one of the most extensive ever made. Acknowledgements. 1 wish to thank the curators of the British Museum (Natural History), the Peabody Museum of Natural History, Yale, and the Zoological Museum, Copenhagen, Mr. I. C. J. Galbraith, Prof. C. G. Sibley and Dr. J. Fjeldsa respectively for their co-opera- tion. F. P. Jensen and S. N. Stuart gave valuable assistance and comments in the prepara- tion of this paper. S. N. Stuart also improved the English. I thank them all for their help. References: Britton, P. L. (ed.) 1980. Birds of East Africa. Nairobi: EANHS. Hail, B. P. & Moreau, R. E. 1970. An Aftlas of Speciation in African Passerine Birds. British Museum (Nat. Hist.): London. Mackworth-Praed, C. W. & Grant, C. H. B. 1960. African Handbook of Birds. Series 1, Vol. 2. Birds of Eastern and North Eastern Africa. 2nd Edition. London: Longmans, Green & Co. Moreau, R. E. 1960. oe and classification of the Ploceine weaver-birds. Part 2. Ibis 102: 443-47 — 1966. The Bird ees of Africa and its Islands. Academic Press: London. Peters, J. L. 1962. (Mayr & Greenway, Eds.) Check-list of Birds of the World. Vol. XV. Mus. Comp. Zool: Cambridge, Mass. Ripley, S. D. & Heinrich, G. H. 1966. Comments on the avifauna of Tanzania. Postilla 96: I-45. Sclater, W. L. 1931. Ploceus nicolli, sp. nov. Bull. Brit. Orn. Cl. 52: 26-27. Stuart, S. N. & van der Willigen, T. A. 1978. Report of the Cambridge Ecological Expedition to Tanzania. Cambridge. Stuart, S. N. & Jensen, F. P. 1981. Further range extensions and other notable records of forest birds from Tanzania. Scopus 5: 106-115. Address: N.-E. Franzmann, Game Biology Station, Kalo, DK- 8410 Ronde, Denmark. © British Ornithologists’ Club 1983. [ Bull. Brit.Orn.Cl.19 83 103(2)] 52 The Soft-plumaged Petrel, the Gon-gon and the Freira, Prerodroma mollis, P. feae and P. madeira by W. R. P. Bourne Received 21 July 1982 The soft-plumaged petrels allied to Pzerodroma mollis are a complex group of seabirds of wide distribution but uncertain affinities whose classification has caused repeated difficulty to systematists (Murphy & Mowbray 1951, Bourne 1957, 1966). In the past I have urged a cautious approach pending the acquisition of more information, a course adopted by Jouanin & Mougin (1979); but since the only important recent information indicates that a number of colonies may have been lost and the remainder include some of the rarest seabird populations in the North Atlantic (Bourne 1965, 1972, Bannerman 1965, 1968, Cramp & Simmons 1977), it seems time to reconsider their importance before we lose these as well. They are currently both the subject of debatable propositions about the difference between dark and light individuals of the same species (Clancey e¢ a/. 1981) and of intrinsic interest in their own right. DISCOVERY AND DESCRIPTION The first soft-plumaged petrel appears to have been collected at 6° 50’ N, 23° 46’ W off west Africa in October 1768 during Cook’s first voyage. It was named Procellaria crepidata by D. C. Solander in his notes and this was the name used in some accounts of the voyage, though unfortunately no descrip- tion was published at the time. An excellent drawing by Sydney Parkinson has since been reproduced by Lysaght (1959), from which it is recognisable as the form breeding locally in the Cape Verde Islands, where it is known as the Gon-gon, which was eventually named Oestre/ata feae by Salvadori (1899). This still appears to be the only record of the collection of an example of the North Atlantic populations at sea, and confirms that this form has a primarily tropical qietneenoe A number of specimens, including an almost uniformly dark grey bird, were next collected in the Southern Ocean by Gould (1860) on the way to Australia in 1838, and subsequently he obtained others from sailors. Gould (1844) published the first description of the main southern population under the name Procellaria mollis. He deduced that while the adults have white breasts, the young are grey (as in gulls), repeating this in his influential books on the birds of Australia. When it was eventually realised that while few petrels show much variation in appearance with age, many are polymorphic, P. mollis was therefore assumed to be one of them. Soon afterwards, in 1853, Frere obtained 2 more birds in Madeira (now at Cambridge—Benson in press). He presumably exhibited them as a new dis- covery, since shortly afterwards William Yarrell misidentified a stray speci- men, without data, of the rather similar nominate race of Fairy Prion Pachyptila turtur as a new species from Madeira for Gould (1855) to describe and name Procellaria brevirostris. Fortunately this name had already been applied to the Kerguelen Petrel by Lesson (Bourne & Elliott 1965), so that while this mistake has caused a good deal of speculation, it has had no permanent nomenclatural consequences. 53 (Bull. Brit.Orn.Cl.19 83 103(2)] The situation was eventually investigated thoroughly by Mathews (1924, 1932, 1934 4). First he appropriately named another stray specimen of doubt- ful origin, now in the American Museum of Natural History (AMNH) as Pterodroma dubius, and what appears to be Gould’s original grey “‘immature”’ in the British Museum (Natural History) (BMNH) as P. deceptornis. He followed this by describing a small population nesting in the mountains of Madeira, where they are known as the Freira, and large birds nesting on the offshore islands, as 2 new races, P.m. madeira and P.m. deserta. Finally in his definitive check-list (1934b) he synonymised all the southern birds under nominate P. mo//is but continued to accept 3 races from the North Atlantic, including large P.m. feae from the Cape Verde Islands, P. m. deserta and P. m. madeira. Most subsequent authors have considered P. m. deserta inseparable from P. m. feae (Bourne 1957, Jouanin ef a/. 1969, Cramp & Simmons 1977). Up to this time, apparently, it was assumed that the Soft-plumaged Petrels were most closely related to the larger members of the genus Pzerodroma in the Southern Ocean. Then Murphy & Mowbray (1951) reported the redis- covery of the long-lost Bermuda Petrel P. cahow which proved to be inter- mediate in size and appearance between P. mo//is and a group of large warm- water species including the Capped Petrel P. hasitata of the West Indies and the Dark-rumped and White-necked Petrels P. phaeopygia and P. externa of the Pacific (Bourne 7 Palmer 1962). I have already suggested that the full elucidation of the situation must depend on further investigation in the field, notably at another possible intermediate station, namely the Azores (Bourne 1966). : OBSERVATIONS IN THE FIELD The appearance and behaviour of live Soft-plumaged Petrels has been examined critically by Elliott (1954, 1957) in the Tristan/Gough group. He found that while most had white breasts, one distinct group nesting high on the main island were darker with more markings below, though otherwise quite similar, and concluded therefore that they were only a variety. Simul- taneously Rand (1954) reported what he took to be a uniformly dark form of P. mollis breeding commonly on Marion Island; so that the 2 authors, prior to publication, were consequently unable to to compare their observa- tions. This was unfortunate, since the specimens that Rand sent to the BMNH are in fact young Kerguelen Petrels (Bourne 1957). In the course of an investigation of the birds of the North Atlantic islands I examined nearly all soft-plumaged petrel specimens in northern museums (Bourne 1957, 1966). They showed little consistent geographical variation except that while North Atlantic birds are usually pale and white below, the southern ones are darker with a breast-band. The birds from the North Atlantic are also divisible into 2 groups of different sizes with different breeding-seasons. Thus the small, form madeira lacks much mottling and appears to breed in the early summer in the mountains of that island; and the large form feae, which tends to be streaked on the flanks, breeds in the autumn on the offshore islands of Madeira and in winter in the mountains of the Cape Verde Islands. About one in ten of all large series were darker and more heavily marked below and one in many hundreds of the southern population were dark all over. Further summer-breeding populations of P. mo//is which appear indistin- guishable from the nominate form have also been found in the Southern (Bull. Brit.Orn.Cl.1983 103(2)] 54 Ocean at the Crozets by Despin e¢ a/. (1972) and on Antipodes Island, south of New Zealand, by Warham & Bell (1979). There is also a specimen of P. mollis in the BMNH collected in the 1840s by Ross’ Antarctic Expedition and 2 others in the Melbourne Museum collected in February 1952 at Port Jeanne d’Arc from Kerguelen, from where Derenne ef a/. (1974) give other records. Individuals have also recently been reported from Macquarie and Chatham Islands (Jones 1980, Crockett 1981), where it now seems likely that bones formerly attributed on grounds of probability to the osteologically similar Mottled Petrel Pterodroma inexpectata, together with other bones from Amsterdam Island (Jouanin & Paulian 1960), really belong to further southern populations of P. mollis. While some of these populations may have been exterminated by introduced predators, it might still be worth searching for survivors (Bourne 1965, 1967, 1972, 1981). Further observations have now confirmed that while heavily-marked individuals can normally be found in variable proportions in most or all populations of Soft-plumaged Petrel, and may apparently predominate in the Prince Edward Islands (Clancey e¢ a/. 1981), uniformly dark individuals are rare. In addition to the bird obtained in the last century by Gould (1844, see Mathews 1932), about 12 have now been reported at sea (Sinclair 1978) and on Gough Island (Swales 1965) in the Southern Ocean, while on Marion Island one was found consorting in a hole with a white-breasted individual (Schramm 1982). At sea the paler birds showed contrasting darker markings of the type usual in Soft-plumaged Petrels, while the darker birds were darker than Kerguelen Petrels, with which they might have been confused, and which appear silvery at sea. Sinclair (1978) reports also that P. mollis can be separated by its characteristic shape and behaviour and the darker leading edge of the underwing. DISCUSSION Any attempt to classify the Soft-plumaged Petrels and their allies en- counters difficulties resulting from either gaps or overlaps in their distribu- tion, suggesting that either the missing populations are likely to have been lost or overlooked, as in the subantarctic islands of New Zealand or the Azores; or that the overlapping populations must either interbreed or repre- sent separate species (Mayr e¢ a/. 1953: 121). It may be useful to consider the problem under 3 headings. The medium-sized gadfly (or Soft-plumaged) petrels of the Southern Ocean Owing to the fact that the most important early authority, Gould (1844, 1860), apparently collected only one medium-sized grey gadfly petrel in the Southern Ocean, which by chance proved to be the rare dark form of P. mollis, there appears to have been persistent confusion over the relationship of this species to the more southerly grey Kerguelen Petrel P. brevirostris (Falla 1937, Rand 1954). Until I saw them together I also assumed that they must be closely related and might hybridise (Bourne 1966). In fact they appear to be rather distinct with numerous isolating mechanisms. Thus, in the first place, while the Soft-plumaged and Kerguelen petrels are rather similar in size and sometimes in appearance, they differ consider- ably in their structure and behaviour, the Kerguelen Petrel having a much larger eye, presumably an adaption for nocturnal vision in particular (Harper 1973). Secondly, they also have rather distinct annual cycles, the Kerguelen Petrel laying about 2 months earlier in the spring and completing its breeding 55 [ Bull. Brit.Orn.Cl.19 83 103(2)] cycle rapidly during the summer (Mougin 1969); the Soft-plumaged Petrel in contrast continues to feed its chick into the winter. Thirdly, I found that on Gough Island the 2 species use rather different nest-sites, the subtropical Soft-plumaged Petrels breeding in dry, sheltered burrows in the steep sides of the island, while the subantarctic Kerguelen Petrels preferred waterlogged holes in the bleak upland heaths. Similarly, while 1 formerly supposed that both species were originally replaced by the rather similar Mottled Petrel P. inexpectata in the New Zealand area until the sites were devastated by introduced predators (Bourne 1957), the presence of P. inexpectata on the outlying islands has never been proved, while P. wo//zs has in fact been found there instead. It consequently seems more likely that P. znexpectata was a specialised form which developed on and around the main islands of New Zealand, dispersing south in the summer and migrating into the northern hemisphere in the winter, whereas the sedentary P. wo//is continued to occupy the same niche on the outlying islands that it does in the Atlantic and Indian Oceans. (Incidentally I have seen P. mollis commonly at sea much further east in the Great Australian Bight than shown on the distribution map in Harper (1973). ) The medium-sized gadfly petrels of the North Atlantic While the Soft-plumaged Petrels of the Southern Ocean may vary in their appearance, they are all rather similar in their size and habits, nesting on the middle slopes of oceanic islands in the local summer. The situation is rather different in the North Atlantic, where, as already mentioned, in addition to a population of small-sized birds nesting in the summer on the cool, moist, heavily-vegetated upper slopes of Madeira, other groups of larger-sized birds breed in the autumn on its bare, arid outlying islands, and in the winter on the mountains of the Cape Verde Islands to the south and islets around Bermuda to the west. This situation is susceptible to 2 alternative explanations. All the birds may possibly be derived (together with P. cahow of Bermuda) from the large, winter-nesting Capped Petrel P. hasttata of the West Indies. This could have given rise when the climate deteriorated during the Pleistocene to a summer- nesting population on the mountains of Madeira, which subsequently colonised the Southern Ocean, where it is now replacing P. znexpectata in the New Zealand area following reduction of the latter by introduced predators. Alternatively the eastern North Atlantic populations may be derived from summer-nesting birds of the Southern Ocean, which colonised the North Atlantic during the Pleistocene, and since the climate became warmer have given rise to a winter-nesting population in the Cape Verde Islands, and have now colonised the lower levels of the Madeira group as well. The relationship between birds of different colours On the basis of a personal examination of specimens in northern collec- tions, including all the types, it is clear (see above) that while northern populations of the Soft-plumaged Petrels are normally pale with white breasts, and the southern ones darker with a band across the breast, occasional individuals in all populations are darker with more marking below, or even dark all over in the southern populations. The examination of many more specimens including recent series from the Antipodes and Crozet islands confirms this. [ Bull. Brit.Orn.Cl.19 83 103(2)] 56 Figure 1. Breeding distribution of Soft-plumaged and Capped Petrels. M-— Soft-plumaged Petrel Prterodroma mollis. - Petrel Mottled Petrel P. inexpectata. F— Gon-gon P. feae. X— Freira P. madeira. C- Bermuda Petrel P. cahow. H- Capped Petrel P. hasitata. D- Jamaica Petrel P.h. caribbaea. P- Dark-rumped Petrel P. phae- opygia. S— Hawaiian Petrel P.p. sandwichensis, E— Juan Fernandez Petrel P. externa. Y-— White-necked Petrel P. e. cervicalis. B— Barau’s Petrel P. barani. Populations circled are already known to be severely reduced in numbers, and those with small letters are only known from stray specimens and bones; the Jamaica Petrel (D) may be extinct. Clancey e¢ a/. (1981) have recently deduced instead, from South African specimens from a more limited series of sites, that the darker birds belong to a distinct race, P. w. dubia Mathews (1924), breeding in the southern Indian Ocean and expected to occur on Antipodes Island in the South Pacific. Regrettably, in fact, the type of P. m. dubia is so worn that it is difficult to be sure of its original appearance in the absence of information about its origin. Clancey e¢ a/. also report the occurrence of two small, pale petrel corpses resembling the North Atlantic race P. m~. madeira on South African beaches. In addition to the dark birds collected on Tristan by Elliott (1954), the BMNH also contains 2 small, pale birds, possibly immatures, which Elliott collected on nearby Inaccessible Island in May 1952. There is also a third, in spirit, collected on Marion Island in March 1961. Thus the full range of variation in the southern populations has now been collected in both the Tristan/Gough and Prince Edward groups and, as reported by Warham & Bell (1979), the Antipodes birds show a similar variation in appearance. It appears that while the soft-plumaged petrels allied to P. wo/lis become darker and more heavily marked from north to south, there is a wide range of variation and overlap in their appearance, so that plumage is not a very satisfactory basis for their classification. The populations in the Southern Ocean, which all appear to nest in the summer, are otherwise rather uniform in such characters as size and behaviour, and seem rather distinct from other southern petrels except possibly P. inexpectata, which differs in its voice (Warham 1979) and migratory habits; thus there seems little case for recog- nising racial variation there. The North Atlantic populations are more 57 [ Bull. Brit.Orn.Cl.19 83 103(2) | variable in size and annual cycle, intergrading through madeira, feae and cahow with the large, winter-nesting Capped Petrel P. hasitata of the West Indies (Fig. 1). While there is at present an important gap in the North Atlantic range of the group in the inadequately-explored Azores, where the birds may either have been overlooked or exterminated by introduced predators, Madeira appears to have been colonised twice. Presumably it was first occupied by the small, summer-nesting Freira madeira when it was cooler and wetter during the Pleistocene; but adeira has now become restricted to the upper slopes and the low ground has been colonised by the larger, winter-nesting Gon-gon feae, derived from the same stock but under the more arid condi- tions of the Cape Verde Islands since the climate became warmer in recent times. In consequence the Madeira area, which appears to have the best conditions for subtropical petrels in the North Atlantic, is now exploited by birds of a wider range of sizes breeding during much of the year. The most important conclusion arising from recent observations, for which I am indebted to C. Jouanin, R. de Naurois, G. Le Grand and D. Wingate, is that all the North Atlantic Pterodroma petrels are now rare and threatened, with cahow and madeira in particular already reduced to a few dozen birds. In this situation there has been a strange contrast between the assistance given to the Cahow, which has been regarded as a species and has received world-wide attention and special assistance (so that it is now recovering), and the Freira, which has been well-nigh ignored as a dubious race. The Gon-gon and the Freira are in fact sympatric forms of great interest which are both in grave danger on Madeira. Because it is difficult to say which of the two is closer to the Soft-plumaged Petrel, showing as the latter does an overlap in its variation in colour, it seems time to abandon taxonomic caution and I advocate treating all 3 binomially as distinct species, Pterodroma mollis, Pterodroma feae and Pterodroma madeira. References: Bannerman, D. A. 1965, 1968. Birds of the Atlantic Islands. Vols. 2 and 4. Oliver and Boyd: Edinburgh. Benson, C. V. in press. ype Species of Birds (skins) in the University Museum of Zoology, Cam- bridge, United Kingdom. Cambridge. Bourne, W. R. P. 1955. The birds of the Cape Verde Islands. [bis 97: 508-556. — 1957. Additional notes on the birds of the Cape Verde Islands. /bis 99: 182-190. — 1965. The missing petrels. Bu//. Brit. Orn. Cl. 85: 97-105. — 1966. Further notes on the birds of the Cape Verde Islands. /bis 108: 425-429. — 1967. Subfossil petrel bones from the Chatham Islands. /bis 109: 1-7. — 1972. General threats to seabirds. [CBP Bull. 11: 200-218. — 1981. The gadfly petrel skull and diving petrels from Macquarie Island. Nosornis 28: 142-143. Bourne, W. R. P. & Elliott, H. F. I. 1965. The correct scientific name for the Kerguelen Petrel. [bis 107: 548-550. Clancey, P. A., Brooke, R. K. & Sinclair, J. C. 1981. Variation in the current nominate subspecies of Pterodroma mollis (Gould) (Aves: Procellariidae). Durban Mus. Novit. 12: 203-213. . Cramp, S. & Simmons, K. E. L. 1977. The Birds of the Western Palearctic. Vol. 1. Oxford University Press. Crockett, D. 1981. 1980-81 Chatham Island Taiko Expedition. OSNZ News 18: 1. [ Bull. Brit.Orn.Cl.1983 103(2)] 58 Derenne, P., Lufbery, J. X. & Tollu, B. 1974. L’avifaune de l’archipel Kerguelen. CVFRA 33: 58-87. Despin, B., Mougin, J. L. & Segonzac, M. 1972. Oiseaux et mammiféres de I’Ile de l’Est, archipel Crozet (46° 25’S, 52° 12’E). CNFRA 1: 1-106. Elliott, H. F. I. 1954. On two new races and an undescribed variety from the Tristan da Cunha group. Bull. Brit. Orn. Cl. 74: 21-24. — 1957. A contribution to the ornithology of the Tristan da Cunha group. Jbis 99: 545-586. Falla, R. A. 1937. Birds. BANZ Antarctic Research Expedition 1929-1931. Reports (B) 11: 1-304. Gould, J. 1844. On the family Procellariidae with descriptions of ten new species. Amn. Mag. Nat. Hist. 13: 360-368. — 1855. Ona new species of the genus Prion. Proc. Zool. Soc. Lond. 1855: 87-88. — 1860. The Birds of Australia. London. Harper, P. C. 1973. The field identification and supplementary notes on the Soft-plumaged Petrel (Pterodroma mollis Gould 1844). Notornis 20: 193-201. Jones, E. 1980. A survey of burrow-nesting petrels at Macquarie Island based on remains left by predators. Notornis 27: 11-20. Jouanin, C. & Mougin, J.-L. 1979. Procellariiformes in Peters, J. L., Check-list of Birds of the World, Vol. 1, 2nd. ed.: 48-121. Jouanin, C. & Paulian, P. 1960. Recherches sur des ossements d’ oiseaux prevenant de Vile Nouvelle Amsterdam (Océan Indien). Proc. XII Int. Orn. Congr. 368-372. Jouanin, C., Roux, F. & Zino, A. 1969. Visites aux lieux de nidification de Pterodroma mollis **deserta’. Oiseau 39: 161-175. Lysaght, A. M. 1959. Some eighteenth century bird paintings in the library of Sir Joseph Banks. Bull. Brit. Mus. (Nat. Hist.) Hist. Ser. 1: 253-371. Mathews, G. M. 1924. (Description of Pterodroma dubius). Bull. Brit. Orn. Cl. 44: To. — 1932. The birds of Tristan da Cunha. Novit. Zool. 38: 13-48. — 1934a. (Remarks on the races of Prerodroma mollis). Bull. Brit. Orn. Cl. 54: 161, 178-179. — 1934b. A check-list of the order Procellariiformes. Novit. Zool. 39: 151-206. Mayr, E., Linsley, E. G. & Usinger, R. L. 1953. Methods and Principles of Systematic Zoology. McGraw-Hill: New York. Mougin, J.-L. 1969. Notes écologique sur la Petrel de Kerguelen Prerodroma brevirostris de I’Ile de la Possession. Ozseau 69 Spec. No. 58-81. Murphy, R. C. & Mowbray, L. S. 1951. New light on the Cahow. Pterodroma cahow. Auk 68: 266-280. Palmer, R. C., 1962. Handbook of North American Birds. Vol. 1. New Haven. Rand, R. W. 1954. Notes on the birds of Marion Island. [bis 96: 173-206. Salvadori, T. 1899. Collezioni ornitologische fatte nelle isole del Capo Verde de Leonardo Fea. Ann. Mus. Civ. Genova (2)20: 3-32. Schramm, M. 1982. Recent records of the dark form of the Soft-plumaged Ho Ptero- droma mollis from the subantarctic, Cormorant 10. Sinclair, J. C. 1978. The Kerguelen Petrel in South Africa and its comparison eth Soft- plumaged Petrel. Bokmakierie 30: 99-101. Swales, M. K. 1965. The sea-birds of Gough Island. /bis 107: 17-42, 215-229. Warham, J. 1979. The voice of the Soft-plumaged Petrel (Pterodroma mollis). Notornis 26: 357-360. Warham, J. & Bell, B. D. 1979. The birds of Antipodes Island, New Zealand. NNofornis 26: 121-169. Address. Dr. W. R. P. Bourne, Department of Zoology, Aberdeen University, Tillydrone Avenue, Aberdeen ABg 2TN, UK. © British Ornithologists’ Club 1983. 59 [ Bull. Brit.Orn.Cl.19 83 103(2)| Sexual size dimorphism in some montane forest passerines from south-central Africa by R. J. Dowsett Received 9 August 1982 The number of bird species in montane forests in Africa is notably less than that in lowland forest (Moreau 1966), and it has been suggested that montane species occupy broader niches. One way in which this might be accomplished is for there to be a significant difference in body size between the sexes of a species, enabling the species to exploit a greater range of food items. This paper examines mensural data from some African montane forest birds with this problem in mind. Forty-three bird species breed regularly in the evergreen forests of the southwestern Nyika Plateau in northern Malawi and adjacent Zambia (Dowsett-Lemaire in press), 24 of them comprising the understorey avifauna of my main study area on the Nyika. All 24 species were caught in mist nets and ringed during population studies, particularly between 1979 and 1982, mostly at an altitude of 2100-2200 m. Birds were weighed on each capture, and the wing-length of each indivi- dual measured at least once each year. Weights were taken to 0.1 g in the passerines considered in this paper, using Pesola balances whose accuracy was checked at intervals. Wing measurements (to 0.5 mm) were taken by the maximum chord method (Svensson 1975). Each bird was examined for moult, and those found to be moulting the longest primary feathers are excluded from the wing-length measurements analysed here. DETERMINATION OF SEX Sexual dimorphism in plumage exists in 11 of the 43 species in the Nyika forests, but only 4 of these are passerines resident in the understorey and thus able to be caught in any numbers: Cape Batis Batis capensis, Green- headed Sunbird Nectarinia verticalis, Eastern Double-collared Sunbird N. mediocris and Red-faced Crimson-wing Cryptospiza reichenovii (nomenclature follows Benson e¢ a/. (1971), amended by Dowsett & Dowsett-Lemaire (1980) ). Of these 4 dichromatic species, the Cape Batis has a yearling plum- age which for both sexes closely resembles the adult female dress. The remaining 20 species of the understorey are for all practical purposes mono- chromatic, although in the Starred Robin Pogonocichla stellata breeding males - can be separated from females by the blue-grey of their heads being slightly lossed. — 7 Because of these morphological similarities, therefore, I initially sexed birds of each species only by examination of the cloacal area of sexually active individuals (see Dowsett-Lemaire & Collette 1980). By this method, females could be recognized for at least one month during the breeding season and males for at least 3 months—in the Starred Robin, for example, males had greatly enlarged penile protuberances between mid-September and early January. All these species are strictly seasonal breeders on the Nyika (Dowsett & Dowsett-Lemaire in prep.), and as they were most easily mist- netted at that time, the great majority of individuals could be sexed with [ Ball. Brit.Orn.Cl_1g43 raz 2z)| 6a certainty. Other characters (e.g. song, presence of a brood patch, etc.) were not used, as they need not be the prerogative of one sex alone. Wing-lengths of the large number of birds sexed by cloacal examination were then analysed to see if there was a relationship between sex and wing- length. Thirteen species were sexed im numbers large enough for Student’s t-test to be used (those species im Table 1 with 2 sample of at least 10 of each sex). Eleven of these showed highly significant differences (Parx 9 z 76.93 3 75.94 ditto. Ig r 7G.32 z 76.50 ditto t. Number of individuals for which measurements are available im each of 2 years> year E (1979-40), 2 (1990-81), 3 (199 2—#z). z. Student’s t-test; ms. = not significant. Because of the confusing yearling plumage im the Cape Batis, only females sexed by cloacal examination are included im Table 1. This is the case with the White-chested Alethe Alethe fuellebarni, in which yearlings resemble adults but retain the shorter remiges of the juvenile, so that the wing-lengths of some first-year males are likely to fall within the ramge of adult females. In the Starred Robin, wing-lengths increase by 3-4 mm or 4-5% on moult 61 [ Bull Brit.Orn.Cl.1983 103(2)) into adult dress (n= 26), but in that species yearlings have a distinctive plumage quite unlike that of the monochromatic adults. Even when adult, a bird’s wing-length may vary from one annual moult to the next, but usually only to a small extent in passerines, and not suffi- ciently to influence any overall sexual differences. Table 2 shows that there were no significant changes in the mean wing-length of samples of individual Starred Robins measured at intervals of one and of 2 years. However, Thorne (1975) did show that in large samples of European Reed Warblers Arocephalus scirpacens there were small but significant increases in wing- length with age (of about 0.5; mm or 0.8% p.a.). Most other studies have shown no apparent increase with age (Flegg & Cox 1977, Stewart 1963). Wear through abrasion during the year also has only a small effect: the wing-lengths of 3 adult male Starred Robins were reduced by o.5—1.0 mm (0.6—1.2%) between the end of moult and the start of breeding, i.e. over a iod of about 6 months. Most birds in the present study were measured uring the breeding season, and so most data are strictly comparable as far as the degree of wear is concerned. SEXUAL DIMORPHISM IN WING-LENGTH In the present analyses, I include only a mean measurement for each individual which has been measured more than once, in order to avoid any bias from the frequent recapture of certain long-lived, highly territorial individuals. However, in determining overall ranges, I have taken all recorded measurements into account. For only 4 of the 17 species in Table 1 does there seem to be a complete dimorphism in size between the sexes. The samples for both Fiilleborn’s Black Boubou Laniaris sagan and the Green-headed Sunbird are too small for this to be confirmed. Although there is no overlap in measure- ments of known male and female Olive-breasted Mountain Bulbuls Axdro- padus tepbrolaemus, a large number of unsexed birds have wing-lengths of 95-97 mm, and so there probably is an overlap in fact. The sexual dimor- phism in sie dni in the Yellow-streaked Bulbul Phy/l/astrephus flavo- striatus is eled by differences in weights reported from several popula- tions—a feature of most species in this genus (Britton 1972). In those 13 species with some overlap in the ranges of wing-lengths for the sexes, the amount of overlap varies considerably. Two extremes are illustrated in Fig. 1: in the Yellow White-eye Zosterops senegalensis there is a very broad overlap, and some 92%, of all birds measured had wing-lengths of 57.5—62 mm inclusive. On the other hand, only 4% of all Starred Robins had wings of 80-80.; mm. Proportions overlapping in other species ranged from 6% to 96% (Table 1). It is of interest that 2 of the species that are strongly dimorphic in plumage (Cape Batis and Red-faced Crimson-wing) have an overlap in wing-length of 90% or more. No other measurement seems to be as suitable as wing-length for distin- guishing the sexes. Although tail-lengths of Starred Robin adults are also significantly longer in males, there is a greater degree of overlap between the sexes. This is also true of wing- and tail-lengths combined, which I have analysed, following the suggestion by Oatley (1982) that there was no over- lap between the sexes in this measurement in Natal, Dealing only with individual Starred Robins sexed by cloaca, tails overlapped between 61 and [ Bull. Brit.Ora.Cl.1983 103(2)] 62 Lassies YY ; n=96 N SOOO : srererers ; orererereretete® 10- LASSE . 6.6.06 OOOO ><) stetetete LSS Ses A RN oO O : © 0:0 0000666606 6666600060 0 of, 55 60 HTH 20 \owN “T 07 oee 80 85 mm Histograms showing sexual dimorphism in wing-lengths of 2 montane forest passerines on the Nyika Plateau: (a) Yellow White-eye Zosterops senegalensis; (b) Starred Robin Pogo- necichla stellata. The % is shown of each sample falling within each 1 mm of wing-length. 63.5 mm (comprising 16% of the population), and wings-+tails between 141 and 144.5 mm (9%). SEXUAL DIMORPHISM IN WEIGHTS The significant differences in wing-length between the sexes in several species are doubtless a reflection of differences in body size. However, it is difficult to establish a direct relationship between an individual bird’s wing- length and weight. In part this is because weight varies seasonally much more than does wing-length, but even when comparable data are available, there may still be no apparent relationship (Clark 1979, Snow & Snow 1963). A sample of 78 adult male Starred Robins measured in October and 63 [ Bull. Brit.Orn.Cl.1983 103(2)] November (lacking any apparent furcular fat deposit), when tested statis- tically gave a low value for the correlation of wing-length with weight (r=0.28). Similarly, there was no significant correlation between these parameters in 39 fat-free adult male Eastern Double-collared Sunbirds, measured when not moulting or breeding (r=o.38). Analyses could not be undertaken for other species of which large samples were measured because of the seasonal presence of visible fat deposits (Dowsett & Dowsett-Lemaire in prep.). TABLE 3 Sexual differences in weight (g) in some Malawi forest passerines Adult male Adult female n Range Mean n Range Mean Alcippe abyssinica 4 1833-2183 20.0 4 18,5—-21.0 19.4 Andropadus tephrolaemus 40 32.2-42.0 38.4 90 30.0—41.0 36.8 Phyllastrephus flavostriatus 11 28.1-35.8 31.8 14 22.0-32.4 252 Alethe fuelleborni 17 41.6-56.5 49.0 fe) 44.0-58.0 49.7 Cossypha anomala 28 22.6-27.4 25.0 39 20.6-27.4 24.1 C. caffra 35 26.2-34.0 28.9 23 25.4-32.0 28.0 Pogonocichla stellata 114 16.1-21.5 18.5 93 15.5—26.3 18.7 Turdus olivaceus bi 66.0-77.0 70.8 13 64.5-78.0 70.6 Chloropeta similis 8 10.9-12.7 11.8 5 II.I-14.0 12.6 Abpalis thoracica 20 10.9—13.3 T2.1 54 10.2—16.0 12.1 Batis capensis 36 LY.4—13.6 12.6 34 T1.3-15.6 13.3 Elminia albonotata 23 8.4-10.9 9.5 21 7.9-11.6 8.9 Laniarius fuelleborni 6 46.0-53.0 48. 4 43.0—-47.0 45.2 Nectarinia verticalis 5 12.9-15.5 14.2 4 11.8—14.8 13.5 NN. mediocris 48 7.2-10.7 8.8 44 6.3- 9.9 7.8 Zosterops senegalensis 96 9.3-12.3 10.7 97 8.9-14.1 Tie Cryptospiza reichenovii 23 I1.5—14.8 13.0 22 I1.9—-15.6 13.4 Not surprisingly, overall weights (Table 3) of the 17 species in Table 1 show very much more variation and much more overlap between the sexes than do wing-lengths. Even in species showing no overlap in wing-length, such as the Yellow-streaked Bulbul, there may be a large overlap in weights, the heaviest individuals usually being females containing eggs. Although such individuals can be recognised in the hand with experience, those with ovaries at an earlier stage of development usually cannot. Consequently, any analysis of geographical or sexual variation in weights would be biased if samples cannot be strictly comparable, e.g. sexed accurately, not breeding, fat-free and not moulting. Individual weights of adult forest birds on the Nyika are generally lowest when young are still dependent and also during the cold months (at least in insectivores), and highest just before breeding and afterwards, even during moult (Dowsett & Dowsett-Lemaire in prep.). Seasonal variation may be considerable: for example, female Starred Robins increase weight before saying to some 144% of their minimum weights, and even in males the maxima may often exceed 120% of minimum weights. Diurnal variation in this and species of similar size on the Nyika is in the region of 6.5% of body weight (Dowsett in press). [ Bull. Brit.Orn.Cl.1983 103{2)] 64 Discussion The significant differences in wing-length between the sexes of many of the montane forest birds studied on the Nyika suggest there may be signifi- cant differences in body size, despite the difficulty of demonstrating differ- ences in body mass. There may consequently be differences in the size of food eaten, but it is difficult to take this any further. Differences in wing-length may be paralleled by differences in bill-length: for example, bill-lengths of adult male Starred Robins are significantly greater than those of adult females (Student’s t-test, P $i31si OML® aiMila, eo@ VAIHSUVNVL @ e tas 1anin @ dNOUud 7a Spe NIMVAS ee Ch \ Nvans po LuOd OC we o “SI vuVvi — BILAAVW —ef} 100 YM yvINW ~~ Pd. HOW ost oO =.